A PRACTICAL TEXT-BOOK OF 
INFECTION, IMMUNITY 
AND BIOLOGIC THERAPY 
WITH SPECIAL REFERENCE TO IMMUNOLOGIC TECHNIC 
By 
JOHN A. KOLMER, M.D., Dr. P.H., D.Sc. (Hon.) 
Professor of Pathology and Bacteriology in the Graduate School of Medicine, 
University of Pennsylvania; and Member of the Research 
Institute of Cutaneous Medicine, Philadelphia. 
With an Introduction by 
ALLEN J. SMITH, M.D., Sc.D., LL.D. 
Professor of Pathology in the School of Medicine of the University of Pennsylvania 
CONTAINING 202 ORIGINAL ILLUSTRATIONS, 51 IN COLORS 
Drawn by ERWIN F. FABER 
Instructor of Medical Drawing, University of Pennsylvania 
THIRD EDITION, THOROUGHLY 
REVISED AND MOSTLY REWRITTEN 
Philadelphia and London 
W. B. SAUNDERS COMPANY 
1923 Copyright, 1915, by W. B. Saunders Company. Reprinted July, 1915. Revised, 
reprinted, and recopyrighted October, 1917. Reprinted January, 1920. 
Revised, entirely reset, reprinted, and recopyrighted October, 1923 
Copyright, 1923, by W. B. Saunders Company 
MADE IN U. S. A. 
PRESS OF 
W B. SAUNDERS COMPANY 
PHILADELPHIA DEDICATED TO MY WIFE 
B. C. H. 
WHO HAS ENABLED ME TO FIND THE 
TIME IN WHICH TO PREPARE THIS EDITION INTRODUCTION TO THE FIRST EDITION 
The last quarter of a century has witnessed an almost marvelous devel- 
opment of knowledge in the domain of medicine and the allied sciences, 
only a part, of course, of the extensive progress made in the field of general 
science. A striking portion of this advance has tended to broaden our 
knowledge of the principles and of the essential details of the processes of 
infection and immunity,* until these branches have today come to form 
almost a special science in themselves—an imperium in imperio. Aside 
from the personal factor, the writer’s immediate interest in the present 
volume, as originally projected, arose from the fact that he was desirous 
of having appear a series of exercises illustrative of the principles of im- 
munology—a class-book intended to set forth in permanent form the very 
excellent course of instruction that Dr. Kolmer has been giving during the 
past few years to selected groups of interested students and occasional post- 
graduate workers in the Medical School of the University of Pennsylvania. 
That it should have surpassed the original simple plan and grown into a 
volume of the present proportions is scarcely to be wondered at, if the 
temptation to elaborate the individual exercises by explanations and cog- 
nate considerations was in the slightest to be yielded to. This is due to the 
fact that in its growth the subject has acquired so much of undoubted im- 
portance in the form of isolated observed facts, and itself presents so many 
analogies and has led to so extensive a terminology, that the author who 
would attempt to link the observed facts into anything like logical sequence 
or to add in the least to the bare cook-book-like series of illustrative exer- 
cises any explanatory paragraphs, cannot avoid the fulness that Dr. Kolmer 
has found inevitable in presenting the subject. 
The branch of immunology, including primarily infection, and its 
ramifications into diagnosis and the actual treatment of disease, has brought 
to the parent subject of preventive medicine the greatest offering of the 
decades of its growth. Itself contributing to world expansion, it has nowhere 
found a greater stimulus than in the field of exotic pathology; and this last, 
in turn, has enriched internal medicine, even in its most common aspects. 
The first step in immunology may properly be ascribed to Jenner, with 
his bovine vaccine for smallpox, a step followed only after a long lapse of 
_years by Pasteur. On the heels of the latter there appeared at once, and 
has since then followed, an army of men whose names crowd the history of 
the subject, and which many of these are bound permanently to adorn. 
The old vague theories of infection have taken form, and to observed facts 
has been added productive theory. The great danger attending this 
luxurious development is that, temporarily at least, the simpler, and per- 
haps the more obvious, facts are likely to be neglected; and, also, that 
symbolization by theories elaborated to harmonize with discovered facts 
will be accepted too fully as explanatory when in reality it does not explain, 
and that, as a result, investigation will finally be hampered instead of aided. 
In the almost universal drift of experimental studies to internal stereo- 
chemical factors, are we not in danger of placing too little stress upon actual 
and possible physical factors? Is there no danger that, by failing to lay INTRODUCTION TO THE FIRST EDITION 
VI 
stress upon the obvious importance of the turbinate mechanism in the nose 
as a natural anatomic factor, our rhinologists may at least feel justified in 
sacrificing this mechanism too readily for what may be but trivial local 
reasons? Can we insist that every phenomenon described with facility in 
terms of the side-chain theory is really a manifestation of chemism, when 
perhaps, with added investigation along lines of physical absorption and 
the physical properties of colloids, an equally satisfying conception may be 
had, and possibly new facts be developed? Are we not blundering in rushing 
madly after matters of specificity as determined by antigen, when perhaps 
in reality we are confronted by potential and kinetic modifications due to 
peculiarities of diet or environmental circumstances? The verity of phago- 
cytosis is open to proof by observation, and its variations are likewise to be 
demonstrated. Is the explanation of opsonins so. convincing that merely 
the word itself is enough to satisfy the investigator? 
Infection and immunity constitute a definite chapter in pathologic 
science. The processes lack the dignity of a separate science only in that 
they present variations, and the fact that these are glossed over by brilliant 
theories and conceptions cannot prevent the deliberate recognition of serious 
incompleteness. Yet this criticism can be applied to the growth of every 
branch of scientific knowledge. It in no wise militates against the right and 
the need for setting forth the subject in the light that, for the time, is afforded 
it. The importance of the criticism lies only in its acknowledgment, lest the 
subject as at present understood be accepted as fixed. With this danger 
obviated, and with all theories accepted for the time only as working theories, 
and their adoption not urged to curtail investigations based on other views, 
their prosecution can be heartily applauded. 
This is the view that the writer believes that Dr. Kolmer has had in 
mind in his presentation of the subject as here set down. It is certainly 
true of the chapters that the present writer has had opportunity of exam- 
ining. In such a sense, therefore, the work is urged on the appreciation of 
the student, whether a laboratory worker or a mere seeker of knowledge. 
I have often been asked to what extent I believe it profitable to present 
the subject to the undergraduate student. I do not hesitate to answer that 
so far as the roster of the medical curriculum will permit, the laboratory 
demonstrations and exercises should form a part of the required course; and 
that, with all due caution to emphasize the fact that our present theory is 
not known to be final, and is offered merely tentatively, the verbal picture 
of the subject should be outlined before these beginners. To form some con- 
ception is necessary; and it is better, provided the mind be kept receptive, 
to follow a certain theory, even if it is unproved, than to do nothing at all 
or to work in confusion. Our American medical curriculum for under- 
graduates is so crowded with absolute essentials that the present subject is 
habitually neglected, save for a rapid lecture outline; this is an injustice 
to the student and to American medicine. I have tried to minimize this by 
providing, through Dr. Kolmer’s aid, a reasonable laboratory course in the 
essentials of the branch to volunteer classes at first, at hours that did not 
interfere with the regular curriculum—at present during periods open to 
election. Nevertheless, the subject, influencing as it does every branch of 
medical practice, must take its place with other commendable additions to 
the required schedule. That this can be done only by lengthening the 
course of study, either in the annual session or by adding a year to our 
present four-year course, is obvious, and to that end we are rapidly ap- 
proaching. 
Allen J. Smith. PREFACE TO THE THIRD EDITION 
To those whose continued appreciation and patronage have made neces- 
sary the preparation of this third edition, the author desires to extend his 
sincere thanks and to express the hope that it will receive the same generous 
recognition as its predecessors. More space has been devoted to the sub- 
jects of vaccine and serum therapy and the treatment of disease with non- 
specific protein substances. As explained later the subject of Chemotherapy 
has been omitted from this edition, being considered in a separate mono- 
graph now in course of preparation. For these reasons the title of this 
edition has been changed to Infection, Immunity, and Biologic Therapy. 
As stated in the preface of the first edition, the main purposes of this 
book are threefold, namely: 
1. To give to practitioners and students of medicine a connected and concise 
account of our present knowledge regarding the manner in which the body may 
become infected, and the method, in turn, by which the organism serves to pro- 
tect itself against infection, or strives to overcome the infection if it should occur, 
and also to present a practical application of this knowledge to the diagnosis, 
prevention, and treatment of disease. 
2. To give to physicians engaged in laboratory work and special workers 
in this field a book to serve as a guide to the various immunologic methods. 
3. To outline a laboratory course in experimental infection and immunity 
for students of medicine and those especially interested in these branches. 
During the six years elapsing since the second edition considerable 
advances have been made and especially in the fields of immunity and 
biologic therapy. I have found it advisable to largely rewrite the chapters 
dealing with these subjects with the introduction of a very large number 
of changes of varying importance. 
The bibliographies accompanying each chapter have been greatly en- 
riched in order to improve the book for reference purposes. Needless to 
state it has been impossible to include a complete bibliography on each 
subject, as this would have changed the essential character of the book and 
enlarged it too greatly. 
The descriptions of immunologic methods and technic for the admin- 
istration of sera, vaccines, etc., have been considerably amplified; I have 
endeavored to maintain the principle of describing methods with sufficient 
detail to make the descriptions worth while and especially helpful for the in- 
experienced. The chapters on precipitins, agglutinins, and complement fix- 
ation have been especially revised with this purpose in mind. References are 
made to the investigations of the author and his colleagues upon the standard- 
ization of the complement-fixation test in syphilis and a description of the 
new antigen and new method based upon these studies is included. Similar 
studies in complement fixation in various other bacterial and protozoon 
diseases and for the detection and differentiation of blood-stains, meats, and 
other protein substances have been completed and the results and methods 
are now being prepared for publication in a separate monograph in order 
not to enlarge the present volume too greatly. 
New chapters have been added on Hemagglutinins and especially in rela- 
tion to blood transfusion, and upon Serum Reactions in Syphilis Other 
Than Complement-fixation Reactions. 
The chapters on anaphylaxis, allergy, and hypersensitiveness have been 
almost entirely rewritten and new chapters included on Allergy in Relation 
to Infection and Immunity, Clinical Allergy, Allergic Skin Reactions, Treat- 
ment of Human Allergies, and the Schick Test for Immunity to Diphtheria. VIII 
PREFACE TO THE THIRD EDITION 
The chapters on vaccine and serum therapy have been largely rewritten 
and non-specific protein therapy included. New chapters have been pre- 
pared on the Principles of Active Immunization, Prophylactic Active Im- 
munization or Vaccination in diseases of human beings and the lower 
animals, Principles of Passive Immunization and the Use of Sera in the Pro- 
phylaxis of Disease and the Principles of Non-specific Protein Therapy. 
In the new chapter devoted to Vaccines, Sera, Blood and Non-specific 
Proteins in the Treatment of Disease, the administration and value of these 
are considered together under each disease instead of in separate chapters 
as is the usual custom. It is hoped that this plan will prove more helpful 
to practitioners by summarizing in one place under each disease what 
biologic therapy has to offer. 
A new chapter has been prepared on the Biologic Therapy of Tuber- 
culosis and also a new chapter on Blood Transfusion, with considerable 
attention to methods for transfusion, in view of the fact that the value of 
this form of biologic therapy has been greatly extended within recent years 
and the technic removed from the exclusive domain of surgery. 
The part devoted to Experimental Infection and Immunity embracing 
a system of teaching these subjects by means of a system of experiments 
introduced by me in the first edition in 1915, has been enlarged by the 
introduction of new experiments to keep abreast of advancements in our 
knowledge. Experience has proved to me the value of including this teaching 
section in this book instead of forming a separate book, as it enables the stu- 
dent doing more or less independent work to consult the text for exact 
descriptions of technic, discussions, and theory. This plan of teaching has 
continued to prove a valuable aid to the author in the Graduate School of 
Medicine of the University of Pennsylvania and, I believe, to other teachers 
as well in graduate and undergraduate medical schools. 
As previously stated, the subject of Chemotherapy has been omitted from 
this edition. So many advances have been made in our knowledge of this 
field of medical science that its inclusion would have enlarged the present 
volume too greatly. With the kind consent of the publishers I have con- 
sidered this subject, including the treatment of syphilis, in a separate mono- 
graph now in course of preparation. 
As stated in the preface of the first edition: “Since the larger portion 
of our knowledge of infection and immunity has been gained from studies 
upon the lower animals, it is not strange that these were early and directly 
benefited by a practical application of this knowledge to the prophylaxis, 
diagnosis, and treatment of many of the diseases to which these animals are 
subject. I have, therefore, included in this volume an account of those 
immunologic diagnostic reactions and applications of specific therapy that 
have a direct bearing upon veterinary medicine.” In the present edition 
more attention and space have been given these subjects and it is hoped 
that veterinarians will find them adequately and helpfully discussed. Of 
course the fundamental facts of infection and immunity and the technic 
of -various immunologic diagnostic methods are the same for the lower 
animals as for man. 
Many new illustrations have been included, and it is hoped that they will 
serve to elucidate the text and to teach, rather than merely to embellish. 
Finally, I beg again to express my deep appreciation of the unvarying 
efficiency and courtesy of the publishers. 
T a K 
Graduate School or Medicine, 
University of Pennsylvania. 
October, 1923. CONTENTS 
PAGE 
Introduction v 
PART I 
GENERAL IMMUNOLOGIC TECHNIC 
CHAPTER I 
General Technic 1 
CHAPTER II 
Methods of Obtaining Human and Animal Blood 11 
CHAPTER III 
Technic of Animal Inoculation 42 
CHAPTER IV 
The Preservation of Serums—Methods 53 
PART II 
PRINCIPLES OF INFECTION 
CHAPTER V 
Infection 58 
CHAPTER VI 
Infection (Continued) 84 
PART III 
PRINCIPLES OF IMMUNITY AND SPECIAL IMMUNOLOGIC 
TECHNIC 
CHAPTER VII 
Immunity—Theories of Immunity 118 
CHAPTER VIII 
Antigens and Antibodies 146 
CHAPTER IX 
The Various Types of Immunity 162 
CHAPTER X 
Opsonins 171 
CHAPTER XI 
Opsonic Index 176 
CHAPTER XII 
Preparation of Bacterial Vaccines 189 
CHAPTER XIII 
Antitoxins 203 X 
CONTENTS 
CHAPTER XIV page 
Ferments and Antiferments 227 
CHAPTER XV 
Bacterial Agglutinins 254 
CHAPTER XVI 
Hemagglutinins 291 
CHAPTER XVII 
Preclpitins 304 
CHAPTER XVIII 
Cytolysins 336 
CHAPTER XIX 
Bacteriolysins 357 
CHAPTER XX 
Hemolysins 381 
CHAPTER XXI 
Venom Hemolysis 414 
CHAPTER XXII 
Principles of the Bordet-Gengou Phenomenon of Complement Fixation 420 
CHAPTER XXIII 
Complement Fixation in Syphilis 442 
The Technic and Practical Value of the Wassermann and Other Complement- 
fixation Tests 442 
CHAPTER XXIV 
Precipitation Reactions in Syphilis 517 
CHAPTER XXV 
Complement Fixation in Bacterial Infections and for the Differentiation of 
Proteins 538 
CHAPTER XXVI 
Cytotoxins 566 
CHAPTER XXVII 
The Relation of Lipoids to Immunity 575 
CHAPTER XXVIII 
Allergy, Anaphylaxis, and Hypersensitiveness 594 
PART IV 
CLINICAL ALLERGY AND BIOLOGIC THERAPY 
CHAPTER XXIX 
Allergy in Relation to Infection and Immunity 643 
CHAPTER XXX 
Clinical Allergy 650 
CHAPTER XXXI 
Clinical Allergy (Continued) 671 CONTENTS 
XI 
CHAPTER XXXII PAGE 
Treatment of Human Allergies 721 
CHAPTER XXXIII 
The Schick Test for Immunity to Diphtheria 739 
CHAPTER XXXIV 
Principles of Active Immunization—Vaccines in the Prophylaxis and Treat- 
ment of Disease 748 
CHAPTER XXXV 
Prophylactic Active Immunization or Vaccination 766 
CHAPTER XXXVI 
Principles of Passive Immunization—Sera in the Prophylaxis and Treatment 
of Disease 835 
CHAPTER XXXVII 
Methods for the Administration of Serum in the Prophylaxis and Treatment 
of Disease 842 
CHAPTER XXXVIII 
Prophylactic Serum or Passive Immunization 860 
CHAPTER XXXIX 
The Principles of Non-specific Protein Therapy 879 
CHAPTER XL 
Vaccines, Sera, Blood, and Non-specific Proteins in the Treatment of Disease 888 
CHAPTER XLI 
Biologic Therapy of Tuberculosis 1042 
CHAPTER XLII 
Blood Transfusion 1077 
PART V 
LABORATORY COURSE IN EXPERIMENTAL INFECTION 
AND IMMUNITY 
Experimental Infection and Immunity. 1109 
Index 1159 LIST OF ILLUSTRATIONS 
FIG. PAGE 
1. An Electric Centrifuge 2 
2. Method of Making a Simple Capillary Pipet 3 
3. Method of Making a Looped Pipet 4 
4. Graduated Pipets 5 
5. Method of Making a Wright Blood Capsule 6 
6. Method of Making a Large Vaccine Ampule of a Test-tube 7 
7. A Satisfactory Syringe 9 
8. A Suction Pump 12 
9. Method of Pricking a Finger., 16 
10. Method of Securing a Small Amount of Human Blood 16 
11. Collecting Blood in a Wright Capsule 17 
12. Removing Serum from a Wright Capsule 17 
13. Method of Sealing a Wright Capsule 17 
14. Method of Obtaining Blood by Venipuncture for Serologic Tests 18 
15. Methods of Securing Blood by Puncture of Vein 19 
16. The Keidel Tube for Collecting Blood 20 
17. Parts of the Keidel Tube 20 
18. Method of Obtaining Blood from an Anterior Jugular Vein 21 
19. A Wet-cup for Securing Blood from Children (Blackfan) 21 
20. Method of Obtaining Blood by Cupping 22 
21. The Anterior Fontanel and Relations and Size of the Superior Longitudinal Sinus in 
a Newborn Infant 23 
22. A Cross-section of the Head of a Newborn Infant through the Posterior Angle of the 
Anterior Fontanel to Show the Size and Shape of the Superior Longitudinal Sinus 24 
23. Method of Obtaining Blood from the Superior Longitudinal Sinus and Injecting 
with a Syringe 24 
24. Method of Producing Local Anesthesia in Spinal Puncture 26 
25. Spinal Puncture in the Recumbent Posture and Measurement of Spinal Fluid 
Pressure with a Landon Mercurial Manometer 27 
26. Spinal Puncture in the Sitting Posture t 28 
27. Technic of Spinal Puncture 29 
28. Babcock Needles for Spinal Puncture 30 
29. An Improperly Constructed Needle with the Opening So Long and Beveled that 
Half of the Opening May Be Within the Dural Sac and Half Without, Allowing 
Escape of Fluid into the Epidural Space 30 
30. A Chart Used by the Author for Recording the Results of Examinations of Cerebro- 
spinal Fluid 31 
31. Method of Bleeding a Rabbit from the Ear 32 
32. A Dissection of the Neck of a Rabbit to Show Relation of the Carotid Artery. ... 33 
33. Method of Bleeding a Rabbit from the Carotid Artery 34 
34. Method of Bleeding a Rabbit from the Carotid Artery (Pasteur Institute Method) 35 
35. Method of Bleeding a Guinea-pig 36 
36. Method of Securing Blood from Heart of a Guinea-pig 37 
37. A Dissection of the Neck of a Sheep to Show the Relations of the External Jugular 
Vein 38 
38. Method of Bleeding a Sheep from the External Jugular Vein 39 
XIII XIV 
LIST OF ILLUSTRATIONS 
FIG. PAGE 
39. Method of Bleeding a Horse from the Jugular Vein 41 
40. Method of Subcutaneous Inoculation of a Guinea-pig 43 
41. Method of Intravenous Inoculation of a Rabbit 45 
42. A Wooden Box for Rabbits 45 
43. Roth’s Method of Injecting a Guinea-pig Intravenously 46 
44. A Dissection of the Neck of a Guinea-pig to Show the Relations of the External 
Jugular Vein 47 
45. Method of Intravenous Inoculation of a Guinea-pig 48 
46. Method of Intravenous Inoculation of a Rat 49 
47. Method of Intravenous Inoculation of a Horse 50 
48. Method of Intraperitoneal Inoculation of a Rabbit 51 
49. A Small Berkefeld Filter 54 
50. A Filter 55 
51. A Convenient Box for Drying Serums, Extracts, etc 57 
52. Abdominal Wall of Guinea-pig Showing Diphtheric Edema Facing 97 
53. Normal Adrenal Gland of a Guinea-pig Facing 98 
54. Adrenal Gland of a Guinea-pig After Fatal Diphtheric Intoxication Facing 98 
55. Phagocytosis (Macrophages) Facing 126 
56. Phagocytosis (Macrophages) Facing 126 
57. Positive Chemotaxis Facing 126 
58. Negative Chemotaxis Facing 129 
59. Ehrlich’s Side-chain Theory. Formation of Antitoxins 138 
60. Theoretic Structure of a Molecule of Toxin and Toxoid 139 
61. Ehrlich’s Side-chain Theory. Formation of Agglutinins and Precipitins 141 
62. Ehrlich’s Side-chain Theory. Formation of Cytolysins (Bacteriolysins, Hemol- 
ysins, etc.) 142 
63. General Scheme of Antigens and Their Antibodies 158 
64. Capillary Pipet for Opsonic Index Determination 179 
65. Mixing the Contents of a Capillary Pipet '. 180 
66. Method of Sealing a Capillary Pipet 181 
67. An Opsonic Incubator 181 
68. Method of Preparing a Blood Film 182 
69. Blood Films for Phagocytic Counts 182 
70. Tubercle Opsonic Index Facing 183 
71. An Unsatisfactory Film for Phagocytic Count Facing 183 
72. A Satisfactory Film for Phagocytic Count Facing 183 
73. An Opsonic Index Chart 188 
74. Preparation of a Bacterial Vaccine 191 
75. A Satisfactory Shaking Machine Mounted on a Concrete Block 191 
76. Capillary Pipet for Counting a Bacterial Vaccine 192 
77. A Satisfactory Preparation for Counting a Bacterial Vaccine Facing 193- 
78. An Unsatisfactory Preparation for Counting a Bacterial Vaccine Facing 193 
79. Counting a Bacterial Vaccine after the Method of Wright 193 
80. Instrument for the Standardization of Platinum Loops 195 
81. McFarland Nephelometer 196 
82. Hopkin’s Tube for Standardizing a Bacterial Vaccine 197 
83. A Stock Ampule of Vaccine (Large) 197 
84. Stock Bottle of Bacterial Vaccine 197 
85. A Small Vaccine Ampule 199 
86. Comer’s Automatic Pipet 199 
87. Theoretic Formation of Antitoxins 204 
88. A Flask of Diphtheria Culture 211 
89. A Large Toxin Filter 212 LIST OF ILLUSTRATIONS 
FIG. PAGE 
90. Separation of Blood-serum Facing 214 
91. A Hitchens Syringe 216 
92. A Battery of Hitchens’ Syringes 217 
93. A Dialyzing Cylinder for the Abderhalden Ferment Test 238 
94. The Ninhydrin Reaction (Abderhalden Ferment Test) Facing 239 
95. Theoretic Formation of Agglutinins 256 
96. Theoretic Structure of Agglutinin and Agglutinoid 257 
97. Diagrammatic Illustration of the Action of Agglutinins and Agglutinoids 257 
98. A Satisfactory Culture for the Microscopic Agglutination Reaction 274 
99. An Unsatisfactory Culture for the Microscopic Agglutination Reaction 274 
100. A Positive Agglutination (Widal) Reaction in Typhoid Fever 275 
101. Microscopic Agglutination Test with Dried Blood Facing 276 
102. Macroscopic Agglutination Reaction 280 
103. Macroscopic Agglutination Test. Pro-agglutination 280 
104. Agglutinoscope 281 
105. Macroscopic Hemagglutination 299 
106. Microscopic Hemagglutination 300 
107. Negative Hemagglutination Reaction 301 
108. False Clumping and Rouleaux Formation Due to Drying 302 
109. Hemin Crystals Facing 321 
110. Precipitin Test. Preparation of Extract of Blood-stains Facing 323 
111. A Rack for Precipitin and Agglutination Reactions 325 
112. Titration of a Precipitin (Serum) 326 
113. Method of Placing Immune Serum in Bottom of Test-tube by Means of a Pipet 
to Secure a Sharp Line of Contact in Precipitin Tests 326 
114. A Precipitin Reaction. Biologic Blood Test 327 
115. Theoretic Formation of Cytolysins (Hemolysins, Bacteriolysins, Cytotoxins) 337 
116. Theoretic Structure of a Polyceptor 339 
117. Scheme Showing Mechanism of Complement Fixation 352 
118. Theoretic Structure of a Bacteriolytic Amboceptor 362 
119. Method of Removing Exudate from the Peritoneal Cavity of a Guinea-pig 367 
120. Culture of Cholera Undergoing Bacteriolysis. A Positive Pfeiffer Reaction 368 
121. Culture of Cholera Before Bacteriolysis 368 
122. Stained Preparation of Cholera Undergoing Bacteriolysis Facing 368 
123. Stained Preparation of Cholera Before Bacteriolysis Facing 368 
124. Bactericidal Test (Looped Pipet Method of Wright) Facing 376 
125. Wright’s Many Stemmed Pipet 379 
126. Theoretic Structure of a Hemolytic Amboceptor 390 
127. Titration of Hemolytic Amboceptor Facing 409 
128. Venom Hemolysis Facing 417 
129. A Vial to Contain Blood for the Wassermann Reaction 445 
130. Titration of Hemolytic Complement Facing 469 
131. Wassermann Reaction (First Method) Facing 471 
132. Reading the Wassermann Reaction Facing 471 
133. Titration of Antigen for Anticomplementary Unit Facing 475 
134. Titration of Antigen for Antigenic Unit Facing 475 
135. Wassermann Reaction (Second Method) Facing 477 
136. A Wire Rack Used by the Author for Complement-fixation Tests; Carries 72 Tubes. 478 
137. A Larger Wire Rack Used by the Author for Complement-fixation Tests; Carries 
144 Tubes 478 
138. A Very Simple and Efficient Water-bath Used by the Author for the Inactivation 
of Sera and for Secondary Incubation. Size Indicated 479 
139. A Large Water-bath Used by the Author for the Secondary Incubation in Com- 
plement-fixation Tests. Size Indicated 480 
XV LIST OF ILLUSTRATIONS 
XVI 
FIG. PAGE 
140. A [Positive Reaction with the New Complement-fixation Technic. Shows 
Test-tubes Slightly Reduced in Size, the Volume in Each and Depth of Color. 
The Reaction is 4421 Facing 487 
141. Wassermann Reaction (Fourth Method) Facing 492 
142. Titration of Antihuman Hemolytic Amboceptor Facing 498 
143. Noguchi Modification of the Wassermann Reaction Facing 498 
144. The Noguchi Butyric Acid Test for Globulins 518 
145. A Sohxlet Extraction Apparatus with Vacuum , 524 
146. An Electrically Driven Mixer for Preparing Perethynol Suspension for Vernes’ 
Test 526 
147. Anticomplementary Titration of a Gonococcus Antigen Facing 541 
148. Gonococcus Complement-fixation Reaction Facing 543 
149. Colloidal Gold Reactions Facing 587 
150. Section of Anaphylactic Lung of Guinea-pig Showing Emphysema; also Infolding 
of Bronchial Mucosa 601 
151. Anaphylactic Contraction of Excised Sensitized Guinea-pig Uterus (Schultz- 
Dale Method) 602 
152. Urticarial Rash of Serum Sickness • Facing 659 
153. Multiform Rash of Serum Sickness Facing 660 
154. Method of Making Scarifications for Cutaneous Allergic Tests 675 
155. Method of Applying Allergens to Cutaneous Abrasions in Allergic Skin Tests 676 
156. Positive Allergic Reactions to Pollens Facing 678 
157. Method of Administering an Intradermal Injection 677 
158. Showing the Small Anemic Area Immediately After an Intracutaneous Injection. 678 
159. Local Serum Anaphylactic Reactions 680 
160. Method of Performing a von Pirquet Tuberculin Test • 697 
161. A Positive Cutaneous Tuberculin Reaction (von Pirquet) Facing 697 
162. A Positive Conjunctival Tuberculin Reaction (Wolff-Eisner-Calmette) Facing 698 
163. A Positive Percutaneous Tuberculin Reaction (Moro) Facing 699 
164. A Positive Luetin Reaction ..Facing 707 
165. The Schick Test for Immunity in Diphtheria Facing 743 
166. A Fading Schick Reaction Showing Brownish Discoloration and Fine Des- 
quamation Facing 744 
167. A Pseudopositive Schick Reaction Facing 744 
168. A Combined True and Pseudopositive Schick Reaction Facing 744 
169. Production of Cowpox Vaccine. 770 
170. Technic of Vaccination 772 
171. Technic of Vaccination 773 
172. Technic of Vaccination 773 
173. Technic of Vaccination 774 
174. Technic of Vaccination 774 
175. Technic of Vaccination Employing the von Pirquet Skin Borer for Scarifying the 
Skin 776 
176. Vaccinia (Seven-day Lesion) Facing 779 
177. Vaccinia (Nine-day Lesion) Facing 779 
178. Vaccinoid. A Vaccination Scar Facing 779 
179. Preparation of Rabies Vaccine 788 
180. Subcutaneous Injection of Serum 843 
181. Intramuscular Injection 844 
182. Intravenous Injection 845 
183. Method of Making Intravenous Injection by Means of a Syringe 846 
184. Method of Making Intravenous Injection by Means of a Syringe 847 
185. Method of Making Intravenous Injection by Gravity 848 
186. Apparatus for the Intravenous Injection of Serum (Rockefeller Hospital) 849 XVII 
FIG. PAGE 
187. Outfit for Intraspinal Injection of Antimeningitis Serum by Gravity 853 
188. Intraspinal Injection by Gravity 854 
189. Intraspinal Injection by Means of a Syringe 856 
190. Apparatus Employed by Author for Aseptic Collection of Small Amounts of Blood 858 
191. Collection of Blood 858 
192. Apparatus for Aseptic Collection of Large Amounts of Blood 859 
193. Preparation of Tuberculin 1053 
194. Taking Blood from the Donor (Lewishon’s Method) 1100 
195. Infusion of Citrated Blood into the Recipient (Lewisohn’s Method) 1101 
196. Apparatus Employed by the Author for the Collection, Citrating, and Infusion 
of Blood 1101 
197. Collection and Citration of Blood 1102 
198. Infusion of the Citrated Blood 1103 
199. Kimpton-Brown Tube 1104 
200. Injection of Blood with the Kimpton-Brown Tube 1105 
201. The Lindemann Cannula 1106 
202. Blood Transfusion with Lindemann Cannulas and Syringes 1106 
LIST OF ILLUSTRATIONS INFECTION, IMMUNITY, AND BIOLOGIC 
THERAPY 
Part I 
CHAPTER I 
GENERAL TECHNIC 
In this chapter simple methods are described for preparing capillary 
pipets and similar apparatus usually made in the laboratory, and a few 
general directions are given concerning the preparation of glassware and 
other material employed in the various methods described in succeeding 
chapters and in experimental work. 
It may be well here to utter a word of caution to the inexperienced against 
observing undue haste in performing the manipulations of immunologic technic. 
Careful and painstaking work is essential in order to secure reliable and suc- 
cessful results, and should never be sacrificed for speed, the latter being attained 
only by experience. 
1. A good centrifuge is one of the chief requisites of a laboratory equip- 
ment. While any good instrument will answer, preference should be given 
to the larger types, fitted for holding both 15 and 50 c.c. centrifuge 
tubes,. propelled by electricity, and mounted on a concrete block in the 
laboratory (Fig. 1). 
2. The machine must be well oiled. 
3. The counter tubes should be of the same weight—it is our custom 
to weigh the tubes on a small balance, and adjust the counter tubes until 
both are of equal weight. 
4. The centrifuge tubes should rest loosely upon a rubber disk or wad 
of cotton in the bottom of the metal tube or cup; otherwise centrifuge 
tubes are quite likely to be broken, especially if the machine is run at high 
speed. 
5. The machine should be started and stopped slowly, and unnecessary 
speed and long running time should be avoided. 
6. Never centrifugalize with cotton plugs in the centrifuge tubes. If 
the latter must be sealed, as when working aseptically, rubber stoppers 
should be used. However, if cotton plugs are large and fit tightly, they 
may be prevented from becoming displaced by passing through them two 
cross-pins in such manner that the ends will rest upon the edge of the tube. 
The plugs are thus prevented from being thrown to the bottom of the tube. 
7. If the centrifuge is out of order, however slightly, it should not be 
used, but repaired at once, or else it may be ruined. 
CENTRIFUGE 2 
GENERAL TECHNIC 
Fig. 1.—Electric Centrifuge. 
Mounted on a concrete block. The scales are for the purpose of weighing and counterbalancing 
the tubes. 
PIPETS 
1. Simple Capillary Pipets.—These, are made of soft glass tubing in 
the following way: 
Tubing having a caliber of 6 mm., with thin walls, that does not be- 
come opaque, brittle, or “run” on heating, and that does not contain lead, 
may be used. The question of alkalinity is also of importance in connec- 
tion with the tubing. Many of the cheaper grades undergo disintegrative 
changes, which are accompanied by the setting free of alkali, especially 
when the glass is heated. Glass of this kind should be discarded, as it may 
introduce an element of error into our experiments and observations. 
2. A convenient length of tubing—about 10 to 12 inches—is chosen; 
this will make two pipets. If a sufficient length of tubing for both sides 
is not available, one end may be heated and drawn out with forceps, or a 
handle may be added by fusing to this short end an odd piece of glass. 
It is convenient to have on hand a supply of tubes cut to correct lengths, 
plugged at each end with a ball of cotton, and sterilized in a hot-air sterilizer. 
They are then ready to be drawn out as needed, thus furnishing sterile 
pipets with cotton plugs that tend to prevent contamination. 
3. The flame must be so regulated as to play upon only so much of the 
tube as will suffice to furnish the glass required for drawdng out the tubing. PIPETS 
3 
If a Bunsen flame is used, the tip of the inner greenish flame should be 
applied. The margins of the flame are the hottest, and for this reason the 
tube must be shifted from side to side and be constantly rotated. 
4. In order to secure uniform heating and satisfactory pipets the tube 
must be kept constantly rotated from the moment it enters until it leaves 
the flame. The two ends of the tube are to rest upon the middle finger 
of each hand, while the thumb and forefinger hold the tube in position at 
either side and impart the rotatory movement. It is also necessary that 
the tube be displaced laterally from time to time, so as to bring each por- 
tion of the middle segment of the tube in turn into the edge of the flame 
(Fig. 2). If the latter precaution is omitted, we shall obtain a pipet with 
a central bulb or thicker segment and with thinner segments on each side 
corresponding to the portions of the tube which lie in the edges or hottest 
portion of the flame. 
5. The tube is heated in this manner until the glass is quite plastic. 
No attempt is made to draw out the tube until it has been entirely with- 
drawn from the flame, as otherwise a portion becomes unduly thin and 
plastic and divides, leaving a small, bent, and very poor pipet in each hand. 
The rapidity and force with wrhich the tube is drawn out determine the 
caliber of the capillary stem. By drawing rapidly a tapering capillary 
Fig. 2.—Method of Making a Simple Capillary Pipet. 
Shows manner of holding tubing in a flame and drawing into capillary tubes. A large portion has 
been removed from the center. 
tube is obtained; by drawing slowly a larger capillary tube of more uni- 
form caliber is obtained. Of course, the worker cannot take too much time, 
as the glass hardens quickly. With a little practice this part of the technic 
is soon mastered. Thorough and uniform heating and careful, steady 
pulling when the tube is sufficiently plastic are of primary importance. 
When, owing to an error in judgment in heating the tube, it is with- 
drawn before it is sufficiently plastic and begins to harden, the situation 
cannot be remedied by drawing out the tube quickly with a jerk. Similarly, 
when a tube has been partially drawn and hardens it cannot, as a rule, be 
reheated and drawn out to make a satisfactory pipet. 
6. After drawing out the pipets the hands should be held steady for a 
few seconds, i. e., until the glass has hardened; otherwise the tubes will 
bend and be less satisfactory. 
7. Capillary pipets are manipulated with rubber teats, which should be 
of the best soft vulcanized rubber, and should fit snugly upon the pipet, 
rendering it air-tight. 
2. Looped Pipets.—Looped pipets find their main application in the 
measurement of the bactericidal power of the blood, after the method of 
Sir A. Wright.1 
1 Technique of the Teat and Capillary Glass Tube, 1912. Constable & Company, London. 4 
GENERAL TECHNIC 
The essential features of these pipets are: (a) The capillary stem, which 
serves for measuring and mixing the bacterial emulsion and serum; (b) the 
portion that serves first as a chamber for the sterile nutrient broth and 
later as a cultivation chamber for determining whether the microbes that 
have been mixed with the serum have or have not been killed by it; (c) the 
glass loop, which acts as a trap, preventing extraneous contamination, and 
id) the handle, upon which the rubber teat can be fitted. With a little 
practice these pipets are readily made. 
1. Select glass tubing about 6 inches in length. 
2. Heat one portion about 2 inches from the end, and when sufficiently 
plastic, draw it out for about 2 or 3 inches or until it is long enough to give 
a spiral loop of the desired dimensions (Fig. 3). While the glass is still 
plastic hold the left hand steady, and with the right hand lower the tubing 
and make a spiral loop in such manner that the loop is closely applied, but 
does not touch the sides of the upper and lower segments of the tube. Actual 
contact with the sides must be avoided, for this would produce strain and 
predispose the tube to fracture. 
Fig. 3.—Method of Making a Looped Pipet. 
3. The longer portion of tubing is now heated and drawn out to a capil- 
lary pipet and broken through at the desired point. 
4. Instead of this method the capillary portion may be drawn before 
the loop is made. 
3. Graduated Pipets.—1. In this work 1 c.c. pipets graduated into 
c.c.; 5 and 10 c.c. pipets graduated into XV c.c. will render satisfactory ser- 
vice. The pipets should be calibrated to the tip, and should preferably 
be long, with a narrow lumen, rather than short with a wide lumen, as the 
latter renders the markings too close to one another. For pipeting small 
amounts, as in certain complement-fixation tests, a 0.2 c.c. pipet graduated 
to j-g-fj- c.c. will be found quite serviceable, permitting accurate measurement 
of small amounts of fluid. The entire length of these pipets is equal to the 
ordinary 1 c.c. pipet which renders the subdivisions far apart and quite 
easy to read. These pipets are made by competent dealers upon special 
request. 
2. These pipets should be perfectly clean and clear, sterilized, and have 
sharp, easily read markings. Pipets with broken tips are difficult to handle, 
and if calibrated to the tip are inaccurate. PIPETS 
Fig. 4.—Graduated Pipets. (American Jour. Syphilis, 1922, 6, 92.) 6 
GENERAL TECHNIC 
3. The worker should practice methods of making accurate measure- 
ments. The slightest slip may mean an inaccurate measurement and pro- 
duce untoward results. The mouth end and the pipeting finger should be 
dry, otherwise on' measuring small amounts the delivery will be jerky and 
usually unsatisfactory. 
4. After pipets have held infectious material they should be placed at 
once in a jar containing 1 per cent, formalin solution. After pipeting blood, 
milk, or serum the pipets should be rinsed or placed in a jar containing 
water or a weak lysol solution, as the formalin solution tends to harden 
these substances and renders cleaning quite difficult. 
5. The jar for holding soiled pipets should contain a layer of cotton, 
otherwise the tip may be broken off when the pipets are dropped in. 
Buck1 has recently devised a multiple pipet capable of delivery into 
12 test-tubes simultaneously; this pipet is especially serviceable in con- 
ducting large numbers of agglutination and complement-fixation tests. 
BLOOD CAPSULES 
Blood capsules were devised by Sir A. Wright for collecting small amounts 
of blood for examination. The essential features of a capsule are: (a) The 
upper straight limb which can be drawn out to serve as a needle for punc- 
Fig. 5.—Method of Making a Wright Blood Capsule. 
turing; (b) the recurved limb which makes it possible to fill the capsule by 
gravity without risk of the inflow being arrested by the blood running 
down and blocking the straight limb which provides an outlet for the air. 
These capsules are easily made and prove quite serviceable, especially 
for collecting small amounts of blood for making agglutination tests, opsonic 
measurements, etc. 
1. Take a piece of soft glass tubing about 10 or 12 cm. in length, and 
having an internal diameter of a least 5 mm. 
2. Draw one end out into a capillary stem, and break this at an appro- 
priate point (Fig. 5). 
3. Then reinsert the tube into the flame, and leaving a portion at least 
3 cm. in length to serve as the barrel of the capsule, draw out the tube into 
a capillary stem about 8 cm. in length, and bend it so as to form a stout 
recurved limb lying in the horizontal plane; now, before the glass has lost 
its plasticity, draw the capsule gently upward so that its long axis will be 
at an angle of about 30 degrees horizontally. Finally, separate the capsule 
1 Jour. Infect. Dis., 1916, 19, 267. CLEANING OF GLASSWARE 
7 
from the main tube by filing it across the capillary portion at the distance 
indicated in the accompanying illustration (Fig. 5). 
4. The straight limb may now be drawn to a sharp point and used as a 
needle. 
Test-tubes may be drawn out and converted into ampules for holding 
vaccines, serums, or other fluids. 
1. Thin-walled and sterilized test-tubes of appropriate size are chosen. 
2. The tube is heated at a point near the open end in the Bunsen flame 
in the same manner as the glass tubing, i. e., by keeping the tube constantly 
rotating with a lateral movement to insure uniform heating. 
3. When the glass has become plastic it is drawn out into a stout stem. 
4. After cooling, it is filed through at an appropriate place, being care- 
ful to leave a somewhat long stem (Fig. 6). 
Fig. 6.—Method of Making a Large Vaccine Ampule of a Test-tube, 
5. The open end may now serve as a funnel for filling the bulb. 
6. The bulb is now sealed by warming the air above the level of the 
fluid and then sealing the tip. With a long stem, in order to secure a por- 
tion of the contents, the sealed end may be broken off from time to time; 
it is readily resealed. 
7. Instead of this procedure the fluid may be placed in the test-tube at 
once, the upper end being heated in the usual manner and drawn out; the 
stem is broken through and the tip sealed. If the tube is small or the con- 
tents are such as will almost fill a tube, this method may not be successful, 
owing to the production of steam on heating the fluid, which either cracks 
the bulb or causes the tip to explode at the time of sealing. 
TEST-TUBES 
Different tests require test-tubes of varying sizes according to the nature 
of the work; as a general rule the width should be such as permits mixing 
the contents without capping the tube and inverting. The various sizes 
which I have found useful are described with the different methods. 
Test-tubes should be made of good glass with no lips and with round 
bottoms; they should be well annealed. 
1. All glassware used in immunologic tests should be physically and 
chemically clean; this is especially true of test-tubes, pipets, and flasks used 
CLEANING OF GLASSWARE 8 
GENERAL TECHNIC 
in complement-fixation work. As shown by Hektoen and Ruediger,1 Man- 
waring,2 Cumming,3 Brown, and myself4 traces of acids and alkalies may 
prove destructive for complement or hemolytic, and must be carefully 
avoided in all glassware and solutions. 
2. All glassware, including pipets, test-tubes, and flasks, should be bril- 
liantly clear and glistening, and the markings distinct and easily read. 
3. Unless infectious material has been used it is not generally neces- 
sary to boil the glassware; repeated boiling in soapy water tends to cloud 
the glass and render it unsightly. When this occurs it may be improved 
by immersion in 2 per cent, hydrochloric acid or the bichromate cleansing 
fluid (2 parts potassium bichromate, 3 parts commercial sulphuric acid, 
and 25 parts water) for twenty-four hours, followed by thorough rinsing in 
running water. 
4. After use tubes are emptied; rinsed in running tap-water; washed 
inside and outside in a pan or bucket of warm soapy water; thoroughly 
rinsed in running tap-water and placed upside down in metal baskets, and 
heated in a hot-air oven until a piece of fresh cotton placed in the oven has 
turned a light brown color. It is unnecessary to plug each tube with cotton, 
although it is advisable to plug the flasks because of their use during subse- 
quent work. In this connection mention may be made of the report of 
Langer,8 who found that cotton may contain antilytic substances, and to 
warn against permitting blood, complement serum, or other reagents to 
soak into cotton stoppers by reason of the possibility of thereby dissolving 
out anticOmplementary material. In some laboratories it is customary 
to boil the tubes in soapy water; 'rinse in tap-water; immerse briefly in 1 
per cent, hydrochloric acid to neutralize the alkali of the soap; rinse thor- 
oughly, and sterilize. I believe the boiling and immersion in acid are un- 
necessary as routine procedures, although it is occasionally necessary to 
do this with tubes which are spotted and particularly dirty; the rinsing 
after immersion in the acid bath must be particularly thorough in order to re- 
move all traces of acid. 
5. New test-tubes should not be used until thoroughly washed with 
soapy water, rinsed, and sterilized as described above. In some laboratories 
it is customary to give them an acid bath as described. 
6. While it is preferable to prepare the glassware on the day preceding 
the tests, the tubes are fit for use any time within a period of several days 
after preparation. 
7. Pipets are cleaned in the same way and placed in metal boxes or 
wrapped in newspaper and sterilized in the hot-air oven. The mouth ends 
may be plugged neatly and firmly with a bit of cotton. 
SYRINGES 
A good syringe is indispensable in performing bacteriologic and immuno- 
logic work. Various sizes should be at hand, but usually a graduated 5 
c.c. syringe answers most purposes. 
1. Many kinds of syringes are on the market. Those with rubber or 
leather plungers and packings are unsatisfactory, as they cannot be sterilized 
by boiling, and soon leak. Nothing is more exasperating than a leaking 
syringe, as with the leakage unknown quantities of inoculum are lost, not 
1 Jour. Infect. Dis., 1904, 1, 379. 
2 Ibid., 1904, 1, 112. 
3 Ibid., 1916, 18, 151. 
4 Amer. Jour. Syph., 1919, 38. 
5 Deutsch. med. Wchnschr., 1913, xl, 274. SOLUTIONS 
9 
to mention the possible dangers of contaminating the fingers, the animal, 
and the laboratory. 
Syringes with metal or glass plungers are to be preferred, as are also 
those upon which the needle may be fitted without screwing (Fig. 7). 
2. The old Koch syringe is fitted with a rubber bulb for filling and expell- 
ing the fluid. This arrangement is well adapted for making subcutaneous in- 
jections, but is somewhat dangerous for purposes 
of making intravenous injections on account of 
the danger of injecting air. 
3. Syringes may be sterilized by filling them 
with 1 per cent, formalin solution for a few min- 
utes, followed by several washings with sterile 
water or salt solution. This method is good for 
syringes having leather or rubber packings and 
plungers. It is not safe for blood-cultures, as 
spore-forming bacteria may escape the sterilizing 
process. 
4. With all glass or metal syringes it is best to 
boil the syringe, especially if a careful aseptic 
technic is to be employed. The plunger should 
be removed from the barrel, or else, whether it be 
of glass or metal, it will expand more rapidly than 
the accommodation of the barrel will permit. All 
parts should be placed in a pan or wrapped in 
gauze, warm water added, and boiling allowed to 
take place for a minute or so. After cooling the 
parts are adjusted. 
5. If infectious material has been used the 
syringe, after using, should be washed out and 
sterilized. The needles should be dried and wired, 
and a small amount of vaselin rubbed over to pre- 
vent rusting. The plunger may likewise be 
occasionally rubbed with a small quantity of 
vaselin. Needles may be kept in oil or in absolute 
alcohol; usually thorough drying and wiring pre- 
serves them in good condition. 
As a general rule, freshly distilled water should 
be employed, and particularly in those localities 
where limestone and other alkalies abound.1 
Physiologic Saline Solution.—Sodium chlorid 
(0.85 per cent.) in distilled water is best adapted 
for immunologic work. This solution may be pre- 
pared as follows: 
1. Keep C. P. sodium chlorid in a tightly stoppered bottle; if sufficient 
moisture has collected to render the salt somewhat lumpy, dry a portion 
in the hot-air oven for ten or fifteen minutes before weighing. 
2. Weigh out 8.5 grams and dissolve in 1000 c.c. of freshly distilled water 
in a chemically clean and dry flask furnished with a gauze-covered cotton 
stopper; filter through a good paper. 
3. Sterilize by heating in an Arnold for an hour; smaller bulks require 
SOLUTIONS 
Fig. 7.—A Satisfactory 
Syringe {Record). 
This syringe has a glass 
barrel and metal plunger. It 
is easily sterilized, durable, 
and works smoothly and ac- 
curately. 
1 Amer. Jour. Syph., 1919, 3, No. 1. 10 
GENERAL TECHNIC 
less heating. (Do not sterilize in an autoclave in order to avoid possible 
concentration of salt by loss of water in steam.) 
4. Before using in the Wassermann reaction it is well to test the tonicity 
by adding a drop of washed blood-cells to 5 c.c. of the solution in a test- 
tube; if there are no immediate signs of lysis or none after gentle mixing 
and standing aside for half to an hour, the solutions may be accepted. 
Sodium Citrate Solution.—This is prepared in 1 : 10 per cent, solution, 
using physiologic saline solution and not plain or distilled water. This 
solution is employed for the preventing of coagulation of blood and in- 
flammatory exudates. CHAPTER II 
METHODS OF OBTAINING HUMAN AND ANIMAL BLOOD 
As a general rule, when blood is withdrawn to obtain serum a careful 
aseptic technic should be employed. Similarly, when erythrocytes are to 
be obtained for purposes of immunization it is necessary to avoid contami- 
nation by proper cleansing of the parts, and the use of sterile needles, con- 
tainers, and solutions. In obtaining erythrocytes for making hemolytic 
tests it is not necessary that the blood be absolutely sterile, the ordinary 
precautions against gross contamination being sufficient. 
Blood may be withdrawn to obtain the corpuscles or serum, or both. 
OBTAINING ERYTHROCYTES 
Red blood-corpuscles are usually obtained and washed free of serum for 
the purpose of making complement-fixation and other hemolytic tests. 
For these purposes three methods are commonly employed, namely: (a) 
Bleeding into a suitable vessel, and defibrinating with glass beads or rods; 
(b) bleeding into an anticoagulating fluid, and (c) breaking up coagula of 
blood and thereby liberating erythrocytes. The latter method is frequently 
used for securing human cells from specimens of blood submitted for the 
Wassermann reaction. 
In a comparative study of these methods Brown and myself1 have found 
that the method of collection has but slight influence upon the erythro- 
cytes, provided the corpuscles are wTashed one or more times to remove all 
traces of liberated hemoglobin and serum. The following methods are 
satisfactory: 
1. Blood is collected in a sterilized flask or Mason jar containing glass 
beads or broken glass rod, and gently shaken for five or ten minutes; the 
blood is now filtered through a bit of cotton to remove particles of fibrin. 
2. Blood is collected in a flask containing 1 per cent, solution of sodium 
citrate in physiologic saline solution, or water, allowing at least 1 part of 
citrate solution to 4 parts of blood. Dog corpuscles are quite fragile, and 
Rous and Turner2 have found that the addition of | per cent, gelatin to the 
citrate-saline solution aids in protecting the cells. In collecting small 
amounts of blood it is better and more economical to use one of these anti- 
coagulating fluids. 
WASHING ERYTHROCYTES 
For purposes of immunization or in making hemolytic tests red blood- 
corpuscles are washed free of serum before being used for the following 
reasons: 
(1) Avoid possible precipitin reactions; (2) to avoid possible anti- 
complementary activity of the animal’s serum if the blood is more than 
a day old; (3) to avoid anaphylactic reactions if the cells are used for pur- 
pose of injecting rabbits at long intervals in the preparation of hemolysin, 
and (4) to avoid erroneous Wassermann reactions if human cells are being 
employed by carrying over serum with corpuscles. 
1 Amer. Jour. Syph., 1919, 3, 169. 
2 Jour. Exper. Med., 1916, 23, 219. 
11 12 
METHODS OF OBTAINING HUMAN AND ANIMAL BLOOD 
1. Place the citrated blood, which has previously been diluted with 
sufficient salt solution, in centrifuge tubes. Defibrinated blood may be 
placed in centrifuge tubes, and 5 to 10 volumes of sterile normal salt solu- 
tion added and thoroughly mixed. Tubes are then carefully balanced and 
centrifuged at moderate speed for five minutes. 
2. Remove the supernatant fluid down to the corpuscles with a sterile 
pipet. Add an equal volume of salt solution; mix the corpuscles and centri- 
Fig. 8.—A Suction Pump. 
By attaching a capillary pipet to the rubber tubing fluid may be removed without disturbing the 
sediment. 
fuge. This process should be repeated once more in order to insure thor- 
ough washing of the corpuscles to remove all traces of serum. For remov- 
ing supernatant fluids which are to be discarded the suction pump shown 
in Fig. 8 will be found very useful. It is well to fit the rubber tubing with 
a capillary pipet which permits the supernatant fluid flush with the sedi- 
ment to be removed. 
3. After the last washing the supernatant salt solution should be care- 
fully removed when the corpuscles are ready for use. PRESERVATION OF ERYTHROCYTES 
13 
For the sake of necessity, economy, or convenience it may be neces- 
sary in certain laboratories to attempt the preservation of the blood used 
in the preparation of hemolysins and suspensions for complement-fixation 
tests. Two methods have been tried, although the great majority of workers 
prefer the use of fresh blood: the formalin method of Bernstein and 
Kaliski,1 also suggested by Armand-Delille and Launoy,2 consisting in 
adding 0.5 c.c. of formalin (40 per cent.) to 400 c.c. of defibrinated blood 
(approximately 1 : 800 dilution of formalin), and the method of Rous 
and Turner,3 consisting in collecting blood in Locke’s solution containing 
1 per cent, sodium citrate in the proportion of 1 part of blood to 4 parts 
of solution, washing three times in Locke’s solution containing 0.25 per 
cent, gelatin, and preserving in Locke’s solution +2.8 per cent, saccharose. 
Bernstein and Kaliski found that defibrinated sheep blood preserved 
with 1 : 800 formalin proved satisfactory for two months, whereas plain 
defibrinated blood was generally unfit for use after seven days; also that 
blood collected in a 1 per cent, solution of ammonium oxalate as an anti- 
coagulant, followed by the addition of formalin to 1 : 800, kept as well as 
defibrinated blood with the same amount of formalin. These investigators 
found that formalin in dilution as low as 1 : 200 was without effect upon 
the Wassermann reaction, and that formalized cells could be used for the 
immunization of rabbits in the production of antisheep hemolysin. Experi- 
ments with citrated human blood with the addition of formalin to 1 : 400 
dilution yielded satisfactory results over a period of four weeks. Rein- 
mann,4 in a study of both methods with sheep blood, found that formalized 
cells (Bernstein-Kaliski) were satisfactory for four weeks, and that sac- 
charose-preserved cells (Rous-Turner) were satisfactory from twenty-one 
to twenty-five days when kept in sealed ampules. 
After a comparative study of these methods Brown and myself5 found 
the following methods satisfactory: 
1. Plain, defibrinated blood kept at a low temperature; a suitable jar 
supplied with glass beads or small pieces of glass rod and a cover is sterilized 
with dry heat, and used for the collection of blood at an abattoir or by 
venupuncture of a sheep. After defibrinating by shaking the whole is kept 
at 2° to 4° C., leaving the defibrinated blood with the fibrin clot. 
2. Formalized blood after the method of Bernstein and Kaliski kept 
at a low temperature. Defibrinated blcod collected as described is filtered 
through a small wad of absorbent cotton into a sterile flask, and 0.1 c.c. 
of pure formalin (39 to 40 per cent.) added to each 80 c.c. of blood; or the 
blood may be collected in 1 per cent, sodium citrate in physiologic saline, 
or Locke’s solution, in the proportion of 1 part blood to 4 of solution, and 
formalized in the same manner. For small amounts of blood a 1 : 10 dilu- 
tion of formalin in physiologic saline solution (1 c.c. of formalin and 9 c.c. 
saline solution) may be used in amount of 0.1 c.c. for each 8 c.c. of blood. 
This is the simplest and most practical method for the preservation of 
cells over a period of two to four weeks. 
Preserved blood tends to become dark in color and the cells increasingly 
fragile; it should not be used unless the following two conditions at least 
are fulfilled: 
PRESERVATION OF ERYTHROCYTES 
1 Ztsch. f. Immunitatsf., 1912, 13, 490. 
2 Ann. d. l’Inst. Pasteur, 1911, 25, 222. 
3 Jour. Exper. Med., 1916, 23, 219. 
4 Jour. Lab. and Clin. Med., 1916, 2, 200. 
5 Amer. Jour. Syph., 1919, 3, 169. METHODS OF OBTAINING HUMAN AND ANIMAL BLOOD 
14 
1. When washed with 3 or more volumes of physiologic saline solution 
by centrifugalization there should be no discoloration of the supernatant 
fluid'after the second washing. 
2. The blood should become of a brighter and normal red color. 
OBTAINING BLOOD-SERUM 
If serum is desired at once, blood should be drawn into sterile centri- 
fuge tubes, and the tube immersed in cold water for from five to ten minutes; 
this facilitates clotting. The clot is then broken up with a sterile platinum 
wire or glass rod, and the serum secured by rapid centrifugalization. Or 
blood may be drawn into sterile cylinders, Petri dishes, or centrifuge tubes, 
and allowed to stand at room temperature for a few hours, after which they 
should be placed in a refrigerator until the serum separates. Blood never 
should be drawn into Erlenmeyer flasks because of the difficulty of draw- 
ing off serum without disturbing the clot. When drawn into Petri dishes, 
care should be taken that the layer of blood is not too thin, otherwise dry- 
ing will occur with poor separation of the serum. As a rule, the best results 
are secured by placing blood in centrifuge tubes, for if separation is poor 
or does not occur at all, the clot may be broken up and serum secured by 
centrifugalization. So far as possible, avoid drawing blood from an animal 
immediately after feeding, as under these circumstances the serum is likely 
to be milky or opalescent. 
OBTAINING CORPUSCLES AND SERUM 
For certain purposes it may be desirable to obtain both serum and 
corpuscles; these may be secured in the following way: 
1. Place blood in a large centrifuge tube or cylinder, and defibrinate 
with rods or glass beads. 
2. Centrifuge thoroughly. 
3. Remove the serum, which is slightly discolored on account of defibri- 
nation, wfith capillary tube and rubber teat. 
4. Filter the corpuscles into a centrifuge tube through a wisp of cotton 
in a funnel to remove small particles of fibrin. 
5. Add normal salt solution, and proceed with the washing process. 
OBTAINING BLOOD PLASMA 
In obtaining blood plasma it is necessary to avoid coagulation of blood 
by securing and handling the blood with the least amount of trauma to 
leukocytes (paraffined tubes), and centrifuging rapidly and at once. 
In the paraffin method devised by Freund, centrifuge tubes are coated 
with paraffin and chilled to a low temperature in order to prevent coagula- 
tion and the collection of plasma. In my experience this method has 
proved unsatisfactory, inasmuch as the plasma regularly coagulated when 
brought to body temperature. 
Meeker has devised a method consisting in marking an appropriate 
centrifuge tube at 30 c.c. and placing within 0.75 c.c. of a 2 per cent, solu- 
tion of sodium oxalate, followed by drying in a gas flame with even dis- 
tribution of the oxalate up to the fiduciary mark, but avoiding boiling. He 
found that 0.0005 gram sodium oxalate was sufficient to decalcify 1 c.c. 
of human blood. Blood was then collected from a congested vein up to 
the 30 c.c. mark followed by admixing with the powdered oxalate adhering 
to the walls of the centrifuge tube, and immediate centrifugalization. 
With this method it was not always possible to avoid having blood OBTAINING SMALL AMOUNTS OF HUMAN BLOOD 
15 
come in contact with the unprotected glass above the mark, or to secure 
even mixture of blood and oxalate; for these reasons partial coagulation 
not infrequently occurred with the production of serum. 
These faults were corrected in the following method devised by Wata- 
nabe1 in my laboratory, which is a combination of the paraffin tube and 
Meeker methods: 
1. A centrifuge tube is marked to indicate a volume of 20 c.c., and 1 
c.c. of a 2 per cent, solution of sodium oxalate added; this amount of oxalate 
is double that used by Meeker and effectually prevents coagulation. 
2. The solution of oxalate is then dried as evenly as possible in the test- 
tube held over a gas flame up to, and a little beyond, the 20 c.c. mark. 
3. The upper portion of the tube is now coated with a thin layer of 
molden paraffin by means of a small soft brush so that there is no unpro- 
tected glass. 
4. Blood is collected by means of a short, wide bored needle into the 
prepared tube which is gently rotated during and immediately after the 
collection of 20 c.c. and immediately centrifuged for thirty minutes at high 
speed. The supernatant plasma is then pipeted to a second plain centri- 
fuge tube and centrifuged at high speed for two to two and a half hours. 
The resulting plasma is free of leukocytes, and almost or entirely free of 
blood-platelets. 
OBTAINING SMALL AMOUNTS OF HUMAN BLOOD 
For obtaining small amounts of blood—up to 2 or 3 c.c.—for the Widal 
reaction, complement-fixation, and other tests the following method is 
satisfactory: 
1. Wash the last joint of the middle finger with alcohol. If the hand 
is cold, it should be warmed by immersing it in hot water. Before punc- 
turing compress the finger and squeeze in such a manner as to drive the 
blood toward the end of the finger. 
2. Prick deeply with a broad blood lancet, Hagedorn needle, or scalpel 
(Fig. 9). 
3. Collect the blood in a small test-tube—about 8 by 1 cm.—such 
as is used in performing the Noguchi reaction for the serum diagnosis of 
syphilis (Fig. 10). 
4. By squeezing the finger sufficient blood can usually be obtained 
from one puncture practically to fill a tube of the size mentioned. One to 
2 c.c. of serum are easily obtained in this manner, and this is sufficient for 
conducting the ordinary serum reactions. When the treatment of syphilis 
is being guided by the Wassermann reaction, frequent tests are necessary, 
and a patient may object to submitting to repeated venipuncture. The 
method for securing blood just described is so simple and efficient that 
objections to it are never made. 
5. Blood may also be drawn in a Wright capsule, made by drawing out 
ordinary thin glass tubing in the Bunsen burner (see p. 4). After sufficient 
blood has been collected (Fig. 11), the straight empty end is sealed with a 
flame and then cooled (Fig. 13). The blood is then shaken into this sealed 
end and the bent end, in turn, sealed with the flame. Care should be taken 
not to heat the blood. When the serum has separated the tube is opened 
by filing at a point above the clot and breaking, protecting the hands with 
a towel. The serum is carefully removed with a capillary pipet and nipple 
(Fig. 12). 
1 Jour. Immunology, 1919, 477. 16 
METHODS OF OBTAINING HUMAN AND ANIMAL BLOOD 
The patient’s finger is grasped firmly and lanced with a Daland lancet across the folds of skin. When 
lanced parallel with the skin-folds the wound is likely to close before sufficient blood is secured. 
Fig. 9.—Method of Pricking a Finger. 
By pricking the finger deeply across the lines of the skin with a broad lancet two or more cubic 
centimeters of blood are easily collected in a small test-tube. Do not use a large tube, as blood may 
be lost on the sides of the tube. 
Fig. 10.—Method or Securing %. Small Amount of Human Blood. OBTAINING SMALL ,1 MOUNTS OF HUMAN BLOOD 
17 
Fig. 11.—Collecting Blood in a Wright Capsule. 
Fig. 12.—Removing Serum prom 
a Wright Capsule. 
Fig. 13.—Method of Sealing 
a Wright Capsule. 
6. To obtain blood from infants and small children the large toe may 
be punctured, but, as a rule, better results are obtained by wet cupping or 
by puncturing a vein. 18 
METHODS OF OBTAINING HUMAN AND ANIMAL BLOOD 
OBTAINING LARGE AMOUNTS OF HUMAN BLOOD 
Larger quantities of human blood may be required for making comple- 
ment-fixation reactions, the Abderhalden ferment test, etc. 
(a) Phlebotomy.—1. In adults a prominent vein at the elbow, such 
as the median basilic, is usually chosen. In children less than a year old 
Fig. 14.—Method of Obtaining Blood by Venipuncture for Serologic Tests. 
(Keen’s Surgery.) 
A tourniquet of garter elastic is being employed with a slip knot; above, a plain No. 16 needle 
grasped with a hemostat has been entered into a vein; below, a Keidel tube is being used, with a hemo- 
stat in position to crush the stem of the ampule. 
this vein is not suitable, better results being obtained when the external 
jugular or a temporal vein is used (Fig. 18). 
2. Place a rubber tourniquet or a few firm turns of a wide muslin ban- 
dage above the elbow. I can recommend the use of ordinary garter elastic 
adjusted with a slip knot as shown in Fig. 14. 
3. Apply tincture of iodin to the skin over the vein. The vein may be OBTAINING LARGE AMOUNTS OF HUMAN BLOOD 
19 
rendered more prominent by directing the patient to open and close the 
hand several times. 
4. Steady the skin over the vein, and insert the needle in the direction 
of the blood-current (Fig. 15). It is more awkward and of no practical 
Fig. 15.—Methods for Securing Blood by Puncture of a Vein. 
The middle figure shows distention of the veins of the arm about the elbow. The needle is entered 
by a quick upward thrust. Practically any prominent and firm vein may be used. The upper left- 
hand figure shows collection of blood in a test-tube. Usually 10 c.c. or more are easily collected before 
dotting occurs. To secure large amounts use a larger needle with a smooth bore (preferably a platinum- 
iridium needle). The lower right-hand figure shows collection of blood in a Keidel tube. 
advantage to puncture in a downward direction toward the hand. The 
needle should be sharp and of a size midway between the ordinary hypo- 
dermic and a large antitoxin needle, as the former is too small and the latter 
is unnecessarily large; gage No. 18, shown in Fig. 15, is quite satisfactory. 20 
METHODS OF OBTAINING ANIMAL AND HUMAN BLOOD 
The blood is then allowed to drop into a sterile tube. It is not necessary 
to attach a syringe, although 5 to 10 c.c. of blood are obtained more quickly 
by this means on account of the possible gentle suction. Needle and syringe 
should be sterilized by boiling. When larger quantities of human serum 
are required, as in autoserum therapy, a platinum-iridium needle should 
be used, as coagulation in the needle is less likely to occur; besides, these 
needles are readily sterilized by heating in the flame. 
Fig. 16.—The Keidfl Tube for Collecting 
Blood. 
Fig. 17.—Parts of the Keidel Tube. 
E is the vacuum bulb which is attached to 
the needle by a piece of rubber tubing (D); the 
glass tube (B) covers the needle and the whole 
is sterilized. 
5. Loosen the tourniquet, withdraw the needle quickly, and seal the 
wound with a touch of flexible collodion. 
6. Instead of a syringe the 5 c.c. vacuum bulb devised by Keidel has 
proved quite satisfactory (Fig. 16). This apparatus consists of a 5 c.c. 
ampule with arm drawn out to a capillary tip and sealed after a vacuum 
has been created by heating (Fig. 17, B). A short piece of rubber tubing OBTAINING LARGE AMOUNTS OF HUMAN BLOOD 
21 
Fig. 18.—Method of Obtaining Blood from an Anterior Jugular Vein. (Keen’s Surgery.) 
The patient is a child six years of age; shows the position of the patient and manner of distending 
the vein by pressure above the clavicle. A 5 c.c. Record syringe fitted with a No. 20 needle is being 
used for the withdrawal of blood. The distended vein is painted with tincture of iodin to indicate 
the position. 
Fig. 19.—A Wet Cup for Securing Blood erom Children. (Devised by Blackfan.) 
The cup is held in this position over a scarified area; air is exhausted by means of a pump at- 
tached to the rubber tubing; blood collects in the small test-tube. (Made by Hynson, Westcott & 
Dunning, Baltimore, Md.) 
connects the needle and the capillary portion of the ampule. A needle of 
No. 25 gage is fitted tightly into the free end of the rubber tubing. A slender 22 
METHODS OF OBTAINING HUMAN AND ANIMAL BLOOD 
glass tube closed at one end and flaring slightly at the other serves as a 
protection for the needle, which it covers when the apparatus is sterilized. 
The apparatus is sterilized in a hot-air oven at 150° C. for one hour. To 
obtain a specimen of blood the needle is inserted into a vein, and the capil- 
lary end of the ampule crushed with a hemostat through the rubber tubing, 
blood flowing into the ampule and replacing the vacuum (Figs. 14 and 15). 
The protecting glass tubing is then replaced. 
Not infrequently, especially in children and in obese adults, one fails to 
enter a vein. Several attempts to do so may result in ruining one or more of 
the tubes. The late Dr. Alfred Reginald Allen devised a useful modification 
in the technic of using this handy tube; this consisting in detaching the 
Fig. 20.—Method of Obtaining Blood by Cupping. (Keen’s Surgery.) 
The child is seven years of age; the Blackfan apparatus is being used and blood collected from a scarified 
area into a sterile test-tube. 
bulb from the rubber tubing and needle, inserting the latter into the vein, 
and when the blood appears, quickly attaching the bulb and breaking the 
neck with'a hemostat, in the usual manner. By this method the bulb is 
not broken until one is sure he has entered a vein and secured a specimen 
of blood. 
(b) Wet Cupping.—1. This method is particularly applicable for secur- 
ing blood from infants. 
2. Cleanse an area over the back just below the angle of the scapula. 
3. Scarify with a few superficial linear incisions or with a special scarifier. 
4. Apply a cup and exhaust the air wdth special syringe. The vacuum 
produces marked congestion of the skin with a ready flowT of blood. 
5. Carefully release the cup and pour blood into a tube. OBTAINING LARGE AMOUNTS OF HUMAN BLOOD 
23 
6. The apparatus devised by Blackfan, and shown in the accompanying 
illustrations (Figs. 19 and 20), is quite satisfactory and collects blood in a 
sterile tube. 
(c) Placental Blood.—For purposes of immunization corpuscles may be 
obtained by collecting placental blood. 
1. After tying and cutting the cord the placental end is placed care- 
fully in a 150 c.c. flask or bottle containing from 25 to 50 c.c. of sterile 2 
per cent, sodium citrate in physiologic salt solution. To avoid contamina- 
tion the cord may be lightly sponged with 1 per cent, formalin solution 
and severed with sterile scissors. 
Fig. 21.—The Anterior Fontanel and Relations and Size of the Superior Longitudinal 
Sinus in a Newborn Infant. (Keen’s Surgery.) 
2. By exerting pressure on the uterus blood may be squeezed out of the 
placenta. The flask is then sealed with a sterile cotton plug and gently 
shaken. 
3. The corpuscles are obtained by centrifugalization or sedimentation. 
(d) From Superior Longitudinal Sinus.—In infants fifteen months or 
less in age blood is readily obtained from the superior longitudinal sinus 
if the anterior fontanel is still open (Figs. 21 and 22). The latter opera- 
tion, first conducted by Tobler,1 and since highly recommended by many 
pediatrists, is conducted as follows: 
The skin is carefully cleansed and sterilized with tincture of iodin; the 
puncture is best made in the median line of the posterior angle. The needle 
attached to a 5 c.c. syringe (Luer or Record) should be about gage No. 18 
1 Monatschr. f. Kinderhl., 1915, 13, 384. 24 
METHODS OF OBTAINING HUMAN AND ANIMAL BLOOD 
with a short sharp bevel, and is passed inward at a right angle for a dis- 
tance of about 4 mm. and suction made (Fig. 23); if blood does not flow 
the needle should be passed about 2 mm. further, which suffices for the 
majority of children up to fifteen months of age. The needle and syringe 
Fig. 22.—A Cross-section of the Head of a Newborn Infant Through the Posterior Angle 
of the Anterior Fontanel to Show the Size and Shape of the Superior Longitudinal 
Sinus. (Keen’s Surgery.) 
should be carefully sterilized by boiling and the whole operation conducted 
in an aseptic manner. With proper care the operation may be done as a 
safe, quick, and efficient means for obtaining small amounts of blood from 
infants. Goldbloom1 has devised a special needle-holder for the purpose 
Fig. 23.—Method of Obtaining Blood from the Superior Longitudinal Sinus and Injecting 
with a Syringe. (Keen’s Surgery.) 
The child is six months of age; the shape and size of the anterior fontanel have been outlined; 
a 5 c.c. Record syringe fitted with No. 18 needle is being employed, the needle having been entered for 
about 6 mm. in the median line at the posterior angle, and perpendicular to the sinus. 
of guarding against passing the needle too far and transfixing the sinus; 
while it is handy and convenient, the physician accustomed to conducting 
venipuncture will probably find it unnecessary and prefer to pass a needle 
1 Amer. Jour. Dis. Child., 1918, 16, 388. METHOD OF SECURING CEREBROSPINAL FLUID 
25 
attached to a syringe slowly and carefully, making gentle suction with the 
piston at intervals to determine when the sinus has been entered. 
The transfusion of blood, serum, salt solution, and neo-arsphenamin is 
easily conducted by this technic, the injections being given with a syringe; 
in injecting arsphenamin and neo-arsphenamin great care must be exercised, 
as disastrous results have followed faulty technic due to the injection of 
these irritating substances into the brain. 
METHOD OF SECURING CEREBROSPINAL FLUID (RACHICENTESIS) 
The chief purpose in making spinal puncture is to obtain and examine 
cerebrospinal fluid as an aid to the diagnosis of cerebrospinal diseases. It 
is mainly of value in neurologic and psychiatric practice, for the purpose of 
securing fluid for making the Wassermann reaction, for a study of cytologic 
changes, alterations in protein content, and the like. Not infrequently the 
procedure is required as an aid to establishing a diagnosis of meningeal 
diseases in children, particularly tuberculous meningitis, epidemic cerebro- 
spinal meningitis, meningeal irritation, “serous meningitis,” etc. 
Contraindications.—Ordinarily, when skilfully performed, spinal punc- 
ture is a harmless procedure. Unless the necessity for obtaining fluid is 
very urgent the operation should not be done on persons in poor physical 
condition. Kaplan has cautioned against making lumbar puncture in the 
presence of tumors of the posterior fossa, particularly of the cerebellum. 
When it is highly desirable to study the fluid of such cases 2 c.c. may be 
withdrawn, and immediately replaced with sterile normal salt solution, 
or if no immediate effects are observed, the patient may be kept in bed 
for the next twenty-four hours. 
The rapid withdrawal of fluid, and especially rapid withdrawal with 
the patient in an upright position, may create sufficient negative pressure 
in the brain stem to produce hyperemia, hemorrhage, and foraminal hernia, 
the engagement of the brain stem in the foramen magnum; hyperemia and 
reflex disturbances of the choroid plexus may be followed by hypersecre- 
tion of fluid with increased intracranial pressure which are probably re- 
sponsible for the headache, vertigo, and vomiting sometimes following the 
operation rather than the leakage of fluid into the epidural tissues follow- 
ing withdrawal of the needle. In view of the great number of times lumbar 
puncture is performed the percentage of accidents and complications, how- 
ever, are comparatively small and the operation may be done with con- 
siderable safety if certain precautions are observed as follows: (1) Have the 
patient lying on the side rather than sitting; (2) evacuate slowly, preferably 
measuring the pressure after the escape of every 1 to 2 c.c. when drawdng 
fluid from persons with choked disk, and suspected as suffering with brain 
tumor; (3) never remove more than 5 c.c. of fluid for diagnostic purposes 
unless the pressure is high due to increased volume of fluid in meningitis; 
(4) stop if the patient complains of headache; (5) keep the patient in bed 
for eighteen to twenty-four hours. In common with many others I have 
frequently permitted patients to be up and about after withdrawal of spinal 
fluid without any ill effects, but believe that this practice has been respon- 
sible for several instances of lumbar puncture headache, and consider ad- 
visable the precaution of routinely keeping the patient in bed for at least 
eighteen hours. 
Preparation of Patient.—In bed-fast patients the puncture may be 
made at any time; with ambulatory patients, however, the most suitable 
time is late in the afternoon, so as to permit the patient to rest overnight. 26 
METHODS OF OBTAINING HUMAN AND ANIMAL BLOOD 
The ordinary preparations consist in scrubbing the skin of the lumbar 
region with green soap and hot water, using gauze sponges, followed by 
washing with alcohol and ether. The area is then covered with sterile 
gauze, and just before the puncture is made an application of 10 per cent, 
tincture of iodin is made; or the preliminary cleansing may be omitted, 
two or three coats of the iodin being sufficient. After the fluid has been 
secured the iodin may be removed with alcohol and gauze. The operator’s 
hands should be cleansed carefully and washed in alcohol and bichlorid 
solution or weak formalin, or he may put on sterilized rubber gloves before 
handling the needle and performing the operation itself. 
Anesthesia.—In the majority of instances an anesthetic is not necessary. 
In tabes dorsalis and general paralysis (two conditions most frequently 
Fig. 24.—Method of Producing Local Anesthesia in Spinal Puncture. (Keen’s Surgery.) 
The patient is an obese adult male; the crest of the right ilium is outlined and the “soft spot” 
between the third and fourth lumbar vertebrae located. An intracutamous injection of sterile 1 per 
cent, novocain is being made with a 1 c.c. Record syringe fitted with a No. 26 needle. The position 
of the needle will then be changed to the perpendicular and the subcutaneous tissues infiltrated as far 
as the needle will reach. 
requiring spinal puncture) the operation is peculiarly painless. Sick chil- 
dren are apparently not greatly disturbed, but in adults it may be necessary 
to infiltrate the skin about the site of puncture with 1 per cent, eucain 
(sterile) or cocain solution. The skin over the “soft spot” is infiltrated with 
a fine needle (No. 26), and 1 c.c. syringe, a whitish elevated patch about 
the size of a dime being produced (Fig. 24); the needle is now slowly passed 
vertically, and the deeper tissues anesthesized to the depth of the hypo- 
dermic needle. After withdrawal the spinal puncture needle is introduced 
in the same opening. Ethyl chlorid is much less satisfactory except for 
the mental effect it has upon the patient. Children may receive a few drops 
of ether. With nervous patients it is good practice to obviate nervous shock 
by adopting a few simple precautions against causing unnecessary pain. 
Measuring the Pressure of Spinal Fluid.—The Landon mercurial man- METHOD OF SECURING CEREBROSPINAL FLUID 
27 
ometer is quite useful; finer and more accurate readings, however, are pos- 
sible with an air or a water manometer, the fluctuations of the column of 
fluid being greater. ' 
In taking the pressure it is imperative that a uniform technic be employed; 
for example, the patient should be in a quiet horizontal position. Upright 
posture and coughing increase the pressure. 
The pressure is read before fluid is allowed to escape; the position of the 
instrument is shown in Fig. 25. 
The normal pressure of cerebrospinal fluid varies greatly according to 
technic, and readings are of value only when a uniform method is employed. 
The normal pressure for adults in a quiet horizontal position with mercury 
manometer of Landon is from 6 to 10 mm., with an average of 8 mm.; 
Fig. 25.—Spinal Puncture in the Recumbent Posture and Measurement of Spinal Fluid 
Pressure with a Landon Mercurial Manometer. (From Frazier’s Surgery of the Spine and 
Spinal Cord, D. Appleton & Co., New York.) 
The special needle is in position and the manometer adjusted for reading of the pressure before 
spinal fluid is collected; the reading in this instance was 20 mm. of mercury. The instrument is made 
by the Harvey E. Pierce Co. of Philadelphia. 
with the manometer of Levinson, from 130 to 150 mm. The pressure of 
children is about one-third less, being 45 to 90 mm. of water and 0 to 4 mm. 
of mercury. Pressure in millimeters of water may be expressed in milli- 
meters of mercury by dividing by 13. 
Cerebrospinal fluid pressure may be increased in a variety of conditions: 
(1) In certain cases of congenital or acquired internal and external hydro- 
cephalus with hypersecretion and normal or impaired absorption; (2) in 
acute and chronic meningitis; (3) in acute meningeal congestion—the so- 
called “serous meningitis”; (4) in hemorrhage, and (5) in various space- 
restricting lesions such as tumors and fragments of bone and localized 
chronic inflammatory changes. 
Technic of Lumbar Puncture.—The patient may either sit in a chair 28 
METHODS OF OBTAINING HUMAN AND ANIMAL BLOOD 
and bend forward, or lie on the left side on the edge of a bed or table (Fig. 
25). In the case of sick persons, particularly children, the latter position 
is necessary; it is also advisable with nervous patients, as they are likely 
to bend backward suddenly or jump up when the needle is inserted, and I 
have known the needle to be broken off at such a time. The back should 
be arched backward, the patient bending forward, and the knees being 
drawn up over the abdomen. 
With the sitting posture, however, lumbar puncture is an easier opera- 
tion; the patient may either sit straddling a chair or on the edge of a table 
or bed with the body well bent forward, and the arms folded over a pillow 
in the pit of the abdomen (Figs. 26 and 27). The patient should be com- 
fortable, relaxed, and the back well arched to widen the interarticular spaces. 
The needle used for lumbar puncture should be selected with care and 
be neither too large nor too small. In my experience only two sizes are 
Fig. 26.—Spinai. Puncture in the Sitting Posture. (Keen’s Surgery.) 
The patient is an obese adult male, bent well forward over a folded pillow; the crests of the ilia 
are outlined. The needle has been passed between the third and fourth lumbar vertebrae in the median 
line; spinal fluid is being collected in sterile graduated centrifuge tubes. 
required—one with an outside diameter of 1 mm. (gage No. 19) for the 
practically painless puncture of persons, and particularly adults, in whom 
there are no symptoms of infectious meningitis, and the spinal fluids of 
normal consistency, and a second of larger caliber (gage No. 15) for use 
when suppurative meningitis is suspected, in which case the spinal fluid 
may be denser and flow less easily (Fig. 28). For infants both needles may 
be cut in half and properly pointed, although these smaller sizes are not 
absolutely necessary. Needles are available made of platinum and steel; 
the former bend very easily when brought in contact with bone and thus 
protect the patient, but likewise are easily bent in the operation, and de- 
flected from the proper course by muscular movements, and for these reasons 
I prefer the latter. It is important for the needle to have a short bevele'd 
tip and not a very sharp point. 
The needle should be sterilized by boiling in water for several minutes. METHOD OF SECURING CEREBROSPINAL FLUID 
29 
The operator now selects a “soft spot” for puncture. By running the 
finger along the spines of the vertebrae this will be found to be between 
the third and fourth lumbar spinous processes, about on a level with the 
posterior superior spines of the ilia. The needle is grasped firmly and in- 
serted with a sudden thrust, exactly in the median line, and straight forward. 
The thrust should be sufficient to push the needle through the skin and 
muscles into the spinous ligaments; it may then be inserted more slowly, 
a sudden “give way” indicating that the canal has been entered (Fig. 29). 
This route is better than the lateral route, as there is less danger of striking 
vertebral processes or other obstructions. The stilet is now withdrawn. 
The patient is sitting on the edge of a chair and is bent forward; the crests of the ilia are indicated 
by black lines, and are on a level with the spinous process of the fourth lumbar vertebra; the “soft 
spot” is found just above. The needle is shown in Fig. 28. The first tube receives the first few drops 
of fluid, which are usually blood tinged. 
Fig. 27.—Technic of Spinal Puncture. 
Usually the first fluid to appear is stained with blood and should be col- 
lected in a separate tube. From 5 to 10 c.c. of fluid are then collected in a 
second sterile tube, the needle is quickly withdrawn, and the puncture 
wound sealed with collodion and cotton or with adhesive plaster. 
It sometimes happens that, on withdrawing the stilet, no fluid issues 
forth. In this case the patient is instructed to take a deep breath, and if 
fluid does not appear now, the stilet may be inserted gently to dislodge 
any material that may be occluding the needle, or the needle may be with- 
drawn a trifle if it has been inserted too far, or may be advanced a little 
if it has not entered the canal. If, however, the tap proves a dry one, or if 
only a few drops of blood are obtained, it is not advisable to make another 30 
METHODS OF OBTAINING HUMAN AND ANIMAL BLOOD 
puncture, as the second attempt is likely to prove as unsuccessful as the 
first. 
After-treatment of the Patient.—Occasionally the needle may strike a 
nerve filament, which occurrence is followed by more or less pain along the 
course of its distribution; puncture of the bone is likely to be followed by 
pain for several hours. The majority of patients are so little affected by 
lumbar puncture that no precautions as regards the after-treatment are 
necessary. As previously stated, it is advisable for the patient to rest over- 
night. Sudden release of pressure 
or the nervous shock may give rise 
to severe headache of one or of 
several days’ duration; persons of 
hysteric temperament may, in addi- 
tion, suffer from diarrhea and vomit- 
ing. Rest in bed, the application 
of ice-bags, and the administration 
of sedatives are usually sufficient to 
relieve these after-effects. 
Fig. 28.—Babcock Needles for Spinal Punc- 
ture. (Keen’s Surgery.) 
The needle on the left is gage No. 19 with an 
inside diameter of 1 mm., and recommended for 
routine spinal puncture and intraspinal injec- 
tions when the spinal fluid is of normal con- 
sistency. The needle on the right is gage No. 15 
with an outside diameter of 1.75 mm., and recom- 
mended for spinal .puncture and intraspinal in- 
jections in suppurative meningitis with thickened 
spinal fluid. Both needles fit the various sizes of 
Record syringes. 
Fig. 29.—An Improperly Constructed Needle 
with the Opening So Long and Beveled 
that Half of the Opening May Be Within 
the Dural Sac and Half Without, Allow- 
ing Escape of Fluid Into the Epidural 
Space. (From Frazier’s Surgery of the Spine 
and Spinal Cord, D. Appleton & Co., New 
York.) 
Disposal of the Fluid.—As a general rule, the fluid should be sent at 
once to a laboratory, as total cell counts and bacteriologic cultures are best 
made with fresh fluid. For the Wassermann reaction it is not advisable 
or necessary to add a preservative, as the fluid will keep for several days 
in a good refrigerator; if, however, the fluid is to be kept for longer periods 
of time, 0.1 c.c. of a 1 per cent, solution of phenol may be added to each METHOD OF SECURING CEREBROSPINAL FLUID 
31 
cubic centimeter of fluid. The chart shown in Fig. 30 has been found quite 
useful for recording the results of examination. 
CEREBROSPINAL FLUID EXAMINATION 
Name: Age: Ward: Date: 
Clinical Diagnosis: Physician: 
Pressure 
Amount 
Physical 
Properties 
Cells per 
c. inm. 
Differential Cell Count 
Protein Tests 
Wassermann Reaction 
Lymph. 
Polys. 
Endothel. 
Pandy 
Noguchi 
Kaplan 
Blood 
Spinal Fluid 
Colloidal Gold Reaction 
Registry 
No. 
Color Reactions 
Dilutions of Spinal Fluid 
Remarks 
1 
1 :10 
2 
1 :20 
3 
1 :40 
4 
1 :80 
5 
1 : ICO 
6 
1 :320 
7 
1 :640 
8 
1 : 1280 
9 
1 : 2560 
10 
1 :5120 
5 
4 
3 
2 
1 
0 
Colorlc 
Pale bl 
Lilac o 
Rcd-bl 
Red— 
Bacteriological Examination 
Smears: Culture: Animal Inoculation: 
Additional Chemical Examinations 
Quantitative Dextrose: Quantitative Protein: Quantitative Chloride: 
Qualitative Dextrose: Examined by: 
Fie. 30.—A ChartJUsed by the Author for Recordingjthe Results of Examinations of Cerebrospinal Fluid. (Keen’s_Surgery.) 32 
METHODS OF OBTAINING HUMAN AND ANIMAL BLOOD 
Rabbit.—1. Flip an ear vigorously with the hand, and rub with xylol 
and alcohol. The xylol produces marked congestion and afterward should 
be carefully removed with alcohol and water, as it produces a low-grade 
inflammatory reaction. 
2. Puncture a marginal vein with a large needle. The blood will flow 
quickly in drops and practically any amount up to 10 c.c. or even more 
may be collected in a centrifuge or test-tube (Fig. 31). For making pre- 
liminary tests of serum during immunization 2 c.c. of blood is usually suffi- 
cient. Bleeding may be checked by making firm pressure over the puncture. 
Guinea-pig.—1. Blood may readily be removed directly from the heart 
by anesthetizing the animal with ether, and inserting a sterile needle into 
the heart at the point of maximum pulsation. A syringe for aspiration 
OBTAINING SMALL AMOUNTS OF ANIMAL BLOOD 
Fig. 31.—Method of Bleeding a Rabbit from the Ear. 
may be attached, but better results are secured by adjusting a suction 
apparatus. By means of a short piece of rubber tubing the needle may be 
connected to a test-tube so arranged that a partial vacuum is created by 
attaching to a water suction pump. As soon as the heart has been entered, 
blood is seen to flow into the tube and the constant suction prevents clot 
formation in the needle. In this manner 5 c.c. of blood may readily be 
obtained. 
2. Blood may also be secured by aspirating the external jugular vein. 
The vein is exposed by making a small incision, as in giving intravenous 
injections. 
3. Sufficient blood to make many complement-fixation tests may be 
secured from a large pig by rubbing the ear vigorously with xylol and mak- 
ing a small incision in the margin. Bleeding is facilitated by attaching a OBTAINING LARGE AMOUNTS OF ANIMAL BLOOD 
33 
small test-tube with a side arm to a suction pump. When the proper tube 
is held firmly over the ear 5 c.c. of blood may be obtained by this method. 
Sheep.—Small amounts of blood may be obtained by puncturing one 
of the ear veins. 
Rabbit.—After immunization of a rabbit has been completed the animal 
is usually bled to death, the object being to secure the maximum quantity 
of serum in a sterile condition. Various methods may be used. The animal 
OBTAINING LARGE AMOUNTS OF ANIMAL BLOOD 
Fig. 32.—A Dissection of the Neck of a Rabbit to Show the Relations of the Carotid Artery 
T, trachea; A, carotid artery; V, internal jugular vein; V.N., vagus nerve. 
should be anesthetized by ether or high rectal injection of a gram of chloral 
hydrate in 10 c.c. of water, deep sleep being induced by the latter in from 
five to ten minutes, and lasting for several hours, during which time opera- 
tive procedures produce no pain. 
First Method (Nuttall).—The animal is fastened to an operating board 
or, preferably, held by an assistant, and the hair over the neck and thorax 
is moistened with a 1 per cent, lysol solution. By means of a sterile knife 
the skin is cut longitudinally and the neck muscles exposed for a consider- 
able distance. The animal is then held upright by the assistant over a 
sterile dish or a large sterile funnel, emptying into a cylinder or 50 c.c. 34 
METHODS OF OBTAINING HUMAN AND ANIMAL BLOOD 
centrifuge tube. The operator stretches the neck by carrying the head 
backward, and severs the large vessels on one or both sides of the neck 
with a sharp sterile scalpel or razor, avoiding opening the trachea and esopha- 
gus. After bleeding, the dish is covered or the tube plugged and set aside 
for the serum to separate. This method is quite simple, may be employed 
by the inexperienced, and usually yields a large amount of sterile serum. 
Second Method.—The animal is fastened to the operating board and 
the neck is stretched by placing a roller beneath it. The hair over the neck 
is clipped close, and the skin moistened with alcohol and 1 per cent, lysol 
solution. The carotid artery of one side is exposed by making a straight 
incision through the skin over the trachea and skinning well to one side, 
Fig. 33.—Method of Bleeding a Rabbit from the Carotid Artery, Second Method 
exposing the sternohyoid muscles and external jugular vein. The carotid 
artery, internal jugular vein, and pneumogastric nerve are to be found at 
the outer border of the sternohyoid muscles (Fig. 32). By means of blunt 
dissection the artery is exposed and carefully isolated. Two small spring 
clamps or hemostats are then applied close together at the distal end, and 
the artery divided between them. The proximal end is then held with 
forceps within the mouth of a sterile cylinder or large centrifuge tube. 
The wall of the artery is incised with fine scissors proximal to the forceps, 
and the blood is allowed to flow into the vessel. The yield of blood may be 
increased somewhat by exerting pressure on the animal’s abdomen and 
thorax. OBTAINING LARGE AMOUNTS OF ANIMAL BLOOD 
35 
To avoid the risk of contamination in the foregoing method the apparatus 
shown in Fig. 33 may be used. The whole apparatus is sterilized in the 
autoclave before using. After the artery has been exposed and isolated 
a temporary clamp is applied to the proximal end. A small incision is made 
in the wall of the artery, and the cannula inserted and fastened with a 
ligature. The clamp is then removed and blood collected in a large tube. 
Third Method.—The following method, employed at the Pasteur In- 
stitute at Paris, has been found very useful. The animal—a rabbit or a 
guinea-pig—is anesthetized and secured to an operating-table. The carotid 
artery is carefully and aseptically exposed, and separated from the tissues 
for a distance of at least 1 inch; a ligature is now tied securely about the 
artery at the distal end of exposure; a second ligature is placed in posi- 
tion and looped loosely, ready to tie about the proximal end (Fig. 34). 
Fig. 34.—Method oe Bleeding a Rabbit from the Carotid Artery, Third Method. 
The bottom of a large sterile test-tube is heated and drawn out to a fine 
point, as shown in the illustration (Fig. 34), and the tip is broken off. The 
operator now places his moistened forefinger under the artery, elevating 
and rendering it taut; the tip of the tube is then passed through the 
wall into the interior of the vessel toward the heart. The moment the vessel 
is entered the blood-pressure drives the blood into the tube, so that 20 c.c. 
are soon secured. An assistant now ties the ligature below the site of punc- 
ture; the tube is withdrawn, and the tip sealed in a flame. The ends of the 
ligatures are cut short and the wTound is stitched. Healing usually occurs 
at once, and if subsequent study of the blood is required, the other carotid 
and the femorals can be used similarly for securing it. 
Fourth Method.—The animal is fastened to the operating board, and 36 
METHODS OF OBTAINING HUMAN AND ANIMAL BLOOD 
the hair over the neck and thorax moistened with alcohol or lysol solution. 
The right thorax is then incised and held open by an assistant. The right 
lung is seized with sterile forceps and quickly severed at the base with 
Fig. 35.—Method or Bleeding a Guinea-pig. 
sterile scalpel or scissors. The heart is then punctured, and the blood is 
quickly removed from the thoracic cavity with a sterile 25 c.c. pipet with 
a large opening. Unless the lung is removed it tends to float and block the OBTAINING LARGE AMOUNTS OF ANIMAL BLOOD 
37 
end of the pipet. Everything must be in readiness, as otherwise blood will 
be lost, flooding the thoracic cavity. 
Guinea-pig.—Pig serum is usually secured to furnish complement in 
hemolytic tests, and should be used within twenty-four or forty-eight hours 
after bleeding. Precautions to insure sterility are not, therefore, usually 
necessary. 
1. The animal is anesthetized with ether and the large vessels of the 
neck on one side are exposed by a longitudinal incision. These are severed, 
and the blood is collected in a Petri dish or in a centrifuge tube by means 
of a funnel (Fig. 35). 
2. By means of a sharp-pointed scissors the vessels on one or both sides 
of the neck may be incised transversely at one cut, inserting the blade 
deeply and close to, but avoiding, the trachea and esophagus. 
3. Small amounts of blood may be obtained by aspiration from the 
heart of the living animal. A 5 c.c. Record syringe fitted with a No. 20 or 
Fig. 36.—Method of Securing Blood from Heart of a Guinea-pig. 
22 needle is employed. The animal is fastened to a board or held by an 
assistant and lightly anesthetized. The point of maximum pulsation is 
determined and the needle slowly entered into the right chambers of the 
heart. As a general rule, 2 to 5 c.c. of blood may be obtained by gentle 
suction, the amount depending upon the size of the animal. Large male 
animals are recommended and may be used repeatedly. After withdrawal 
of the needle the animal rapidly recovers, although occasionally bleeding 
may follow into the pericardial sac (Fig. 36). 
Rats.—1. Small quantities of blood may be obtained by snipping off 
the tip of the tail of the animal and milking blood into an appropriate 
sterilized tube containing glass beads, or 2 per cent, sodium citrate solu- 
tion. In this manner one or more cubic centimeters of blood are easily 
obtained, and at once defibrinated and injected into the peritoneal cavities 
of other animals, as in inoculating trypanosomes, etc. 38 
METHODS OF OBTAINING HUMAN AND ANIMAL BLOOD 
2. Large quantities of blood are obtained by severing the large vessels 
of the neck under anesthesia. 
Fig. 37.—A Dissection or the Neck of a Sheep to Show the Relations of the External 
Jugular Vein. 
T, trachea; O.M., omohyoid muscle; E.J.V., external jugular vein; S.C.M., sternocleidomastoid 
muscle. This dissection was made soon after natural death and shows the position and size of the 
vein with the head held backward as it is when blood is removed according to the technic described 
in the text. When distended the vein is even larger than shown; it is quite superficial and is usually 
palpable when pressure is made over the vein at the base of the neck. 
Sheep.—Blood may easily be obtained from a freshly killed animal. 
The first flow of blood is discarded, and a portion of the remainder is col- 
lected in a large, sterile, thick-walled flask containing glass beads. By 
shaking vigorously the blood is defibrinated if one desires to obtain cor- OBTAINING LARGE AMOUNTS OF ANIMAL BLOOD 
39 
puscles, or the blood may be collected in a cylinder and defibrinated by 
whipping with glass rods. 
It is usual, however, in large laboratories to keep a sheep and remove 
the blood as it may be required. Small amounts may be obtained from the 
ear vein, larger quantities being secured from an external jugular vein in 
the following manner: 
1. One may do the bleeding alone, although the aid of an assistant is 
usually necessary, especially if the animal is large and vicious. 
2. The sheep is thrown on its back, and the head is held on the knees 
of an assistant seated on a low box or stool. 
Fig. 38.—Method of Bleeding a Sheep from the External Jugular Vein. 
The operator is distending the vein by pressure over the base of the neck with the left hand. 
When distended the vein can usually be felt beneath the skin. The needle here shown is reduced to 
a little more than half the actual size. 
3. The operator may straddle the animal to hold down the fore feet, 
although this is not necessary unless the animal is vicious. 
4. The wool on the left side of the neck is clipped closely with scissors 
and alcohol applied. 
5. The operator then grasps the neck low down with the left hand, and 
by means of the thumb exerts pressure over the base of the neck. The 
external jugular vein will be found in a groove between the omohyoid and 
sternomastoid muscles (Fig. 37). Firm pressure over the base of the neck 
usually distends the vein, which may be seen or easily felt. After locating 
the vein the pressure should be released for an instant, when the disten- 40 
METHODS OF OBTAINING HUMAN AND ANIMAL BLOOD 
tion will disappear. In this way the operator may be more certain that he 
has located the vein. 
6. A sterile stout needle, at least 2 inches in length and provided with 
a trocar and special shank for firm grasping, is passed quickly into the 
distended vein in an upward and inward direction (Fig. 38). It is essential 
that the needle be sharp, otherwise it will be turned aside by the wall of 
the vein. The end of the needle must not have too long a bevel, or the 
point will pierce the opposite wall before the body of the needle is well 
within the vein. The trocar is now removed, and blood collected in a flask 
or bottle and defibrinated with glass beads and rods. A short piece of 
rubber tubing may be attached to the needle. A suction apparatus is not 
needed because the flow of blood is good so long as pressure is preserved 
over the vein at the base of the neck. 
7. When the required amount of blood has been secured, pressure is 
released and the needle quickly withdrawn. Bleeding ceases at once, and 
the neck is then washed with alcohol. 
8. By this method the same vein may be used over and over again for several 
years. I have never known infection to occur, although the gradual formation 
of scar tissue about the site of puncture may interfere with the operation. 
Hog .—Blood may be secured from hogs by clipping off a small portion 
of the tail with a sharp razor or scissors, beginning at the tip. Bleeding 
is usually quite free, but is easily controlled by a tourniquet and bandage. 
The serum of hogs immunized against hog cholera is secured in this manner. 
Monkey.—1. Small quantities of blood—up to 10 or 20 c.c.—may 
readily be obtained from a small vein just beneath the skin which crosses 
over the inner malleolus at the ankle. When a tourniquet is applied the 
vein becomes prominent; the hair is’clipped, and tincture of iodin applied 
over the skin; a small needle is passed into the vein, and the blood collected 
in a centrifuge tube. 
2. Large quantities of blood may be obtained from the femoral or ex- 
ternal jugular vein under light ether anesthesia. 
Dog.—1. Small quantities of blood may be obtained in the following 
manner: Apply a tourniquet just above the knee; clip the hair over, the 
anterior surface of the leg, and cleanse with tincture of iodin and alcohol; 
make a small incision in the long axis, exactly in the median line; a fairly 
large vein appears at once just beneath the skin; by inserting an appro- 
priately sized needle several cubic centimeters of blood are quickly and 
easily secured. The wound should be very small, and usually requires no 
treatment other than an application of collodion and cotton. 
2. Large quantities of blood are obtained from the external jugular 
vein with or without ether anesthesia; the neck is shaved and cleansed; 
the skin is incised over the vein, which is just beneath the skin, and blood 
removed with a sterile needle and syringe. Pressure over the base of the 
neck renders the vein more prominent. In the case of large dogs incision 
is not necessary, as it is easy to enter the vein directly through the skin, 
as in bleeding the sheep from the external jugular vein or the human from 
a vein at the elbow. Blood may also be secured from the femoral vein 
under ether anesthesia. 
Horse.—1. Small quantities of blood for making agglutination and 
complement-fixation tests may readily be secured from a superficial vein 
about the leg. The hair is clipped over the selected area, and the skin 
sterilized with tincture of iodin. A tourniquet is applied to render the vein 
prominent, the vessel is steadied between forefinger and thumb, and a 
needle quickly inserted. OBTAINING LARGE AMOUNTS OF ANIMAL BLOOD 
41 
2. Larger quantities of blood are secured from the external jugular vein. 
This operation is easily conducted in an aseptic manner and blood collected 
in sterile jars. The neck about the region of the vein, usually on the left 
side, is clipped, and a large area washed with hot lysol solution. A sterile 
sheet may be thrown about the shoulders. The animal is held or placed 
in specially constructed stalls that prevent him from backing away or 
causing mischief. In large antitoxin laboratories bleeding is conducted in 
special rooms, where a careful aseptic operating-room technic may be ob- 
served. 
The external jugular vein is rendered prominent by exerting pressure 
at the base of the neck by the application of a special tourniquet or by the 
thumb and fingers of the left hand, the thumb being placed just above the 
vein. A small incision is made through the skin directly above the vessel. 
A large needle is passed under the skin for a distance of an inch or two 
and then thrust into the vein. Direct puncture into the vein is avoided, 
as the needle track under the skin closes after the needle is withdrawn 
and serves to seal the puncture. The needle is attached to sterile rubber 
tubing that conducts the blood into special jars (Fig. 39). In this manner 
from 6 to 12 liters of blood are easily obtained. 
Fig. 39.—Bleeding a Horse from the Jugular Vein. CHAPTER III 
TECHNIC OF ANIMAL INOCULATION 
This is a highly important part of immunologic work, as both for serum 
diagnosis and for serum therapy the serum must be secured from animals 
that have been artificially immunized. Successful inoculation requires un- 
remitting care and thoroughness, as the toxic effects of the proteins in 
general may kill an animal before immunization has been completed. No 
hard-and-fast rules can be laid down; the weight of an animal and the 
reaction to an injection should decide the frequency and the size of subse- 
quent injections. It is better to proceed slowly and gradually than to give 
too large a dose at once and at too frequent intervals. 
1. Select an appropriately sized syringe that does not leak upon being 
tested with water. As has been stated elsewhere, nothing is more unsatis- 
factory than a leaking syringe, for not only may the hand become soiled, 
but an unknown quantity of inoculum is lost. 
2. The inoculum should be sterile. This is especially desirable when 
giving intravenous and intraperitoneal injections. When living cultures of 
bacteria are to be injected the syringe and the needle should be sterilized 
in order to avoid the introduction of contaminating organisms. 
3. Remove the plunger from the barrel, and sterilize all the parts by 
boiling for at least one minute. As previously stated, an all-glass syringe 
or a glass barrel and metal plunger is the most satisfactory (see Fig. 7). The 
old-fashioned syringe with washers and rubber-tipped plunger should find 
no place in a laboratory. 
4. After cooling, expel the water and load the syringe. This may be 
done by drawing the fluid directly into the syringe and measuring the dose 
by its markings or by pipeting the exact dose into a sterile Petri dish or 
capsule and drawing up in the syringe. 
5. The animal should be fastened or held firmly and in an easy posi- 
tion. Everything should be in readiness, so that the injections may be 
given thoroughly and carefully. 
6. In preparing the inoculum care should be exercised that no solid 
particles enter the syringe. Aside from possibly blocking the needle and 
interfering with the injection, the subcutaneous injection of small frag- 
ments may do no particular harm, but in intravenous inoculation they 
may cause fatal embolism. To obviate this danger the inoculum should, 
if possible, be filtered through sterile filter-paper before the syringe is filled. 
7. Air-bubbles should be removed. The injection of small bubbles of 
air into subcutaneous tissues may cause no harm, but when injected into 
veins they may cause serious disturbances or immediate death. To avoid 
this the syringe, after being filled, should be held vertically, with the needle 
uppermost. The needle should be wrapped in cotton soaked in alcohol, 
and the piston of the syringe pressed upward until all the air is expelled 
from the barrel and the needle. If a drop of inoculum is forced out, it will 
be collected on the cotton, which should immediately be burned. 
8. Injections should be given slowly. 
GENERAL RULES 
42 METHOD OF PERFORMING SUBCUTANEOUS INOCULATION 
43 
9. The animal is then tagged or marked, or its coloring recorded. In 
the case of rabbits, the metal ear tag is best. All data, e. g., the date, size 
of dose, preparation and kind of inoculum, etc., should be recorded in writ- 
ing. 
10. When it is necessary to incise the skin in order to reach a vein an 
anesthetic may be given. With superficial veins, and in subcutaneous in- 
oculations, the injections may be given so readily and easily that no more 
pain can be felt than that which accompanies similar injections in human 
beings. 
Animals may be actively immunized in a variety of ways and in differ- 
ent locations in the animal body. For a particular antibody a certain 
method may be found especially efficacious, and this is dealt with in a sub- 
sequent chapter. In serologic work immunization may be performed by 
subcutaneous, intramuscular, intravenous, intracardial, and intraperitoneal 
injections. 
METHOD OF PERFORMING SUBCUTANEOUS INOCULATION 
Fluid Inoculum.—1. Injections are usually given in the median line of 
the abdominal wall. 
Fig. 40.—Subcutaneous Inoculation of a Guinea-pig. 
A fold of skin is pinched up and the needle entered into the fold. The skin is then released, and the 
injection slowly given. A swelling indicates that the injection is subcutaneous. 
2. Have the animal (a rabbit or a guinea-pig) held firmly by an assistant 
or secured to the operating-table. 44 
TECHNIC OF ANIMAL INOCULATION 
3. Clip the hair where injection is to be made—it is not always neces- 
sary to shave the area. Apply a 2 per cent, iodin in alcohol solution. 
4. Pinch up a fold of skin between the forefinger and the thumb of the 
left hand; hold the syringe in the right hand, and insert the needle into 
the ridge of skin between the finger and thumb, and push it steadily on- 
ward until the needle has been inserted about an inch (Fig. 40). Care 
must be exercised not to enter the peritoneal cavity. Relax the grasp of 
the left hand and slowdy inject the fluid. If the skin is raised, this showrs 
that the injection is subcutaneous. If it is not, the needle should be slightly 
withdrawn and inserted. 
5. Withdraw the needle, and at the same time cover the puncture with 
a wad of cotton wet with alcohol. A touch of flexible collodion over the 
puncture completes the operation. 
Solid Inoculum.—Steps 1 to 3 are the same as in the preceding. 
4. Raise a small fold of skin with a pair of forceps, and make a tiny 
incision through the skin with a pair of sharp-pointed scissors. 
5. With a probe, separate the skin from the underlying muscles to 
form a funnel-shaped pocket. 
6. By means of a fine-pointed forceps or a glass tube syringe introduce 
the inoculum into this pocket and deposit it as far as possible from the 
point of entrance of the instrument. 
7. Close the wound with collodion and cotton. A single stitch with 
fine thread may be necessary. 
METHOD OF MAKING INTRAMUSCULAR INOCULATION 
1. These injections are usually made into the posterior muscles of the 
thigh or into the lateral thoracic or abdominal muscles. 
2. Clip away the hair over the selected area, cleanse, etc., as for sub- 
cutaneous injection. 
3. Steady the skin over the selected muscles with the slightly separated 
left forefinger and thumb. 
4. Thrust the needle of the syringe quickly into the muscular tissue 
and slowly inject the fluid. 
METHOD OF MAKING INTRAVENOUS INOCULATION 
Rabbit.—1. The posterior auricular vein along the outer margin of the 
ear is better adapted than a median vein for this purpose. 
2. If a number of injections are to be made commence as near the tip 
of the ear as possible, as the vein may become occluded with thrombi, and 
subsequent inoculations may then be given nearer and nearer the root of 
the ear. 
3. The animal should be held firmly, as the slightest movement may 
result in piercing the vein through and through and require reinsertion of 
the needle. This is accomplished satisfactorily by placing the rabbit upon 
the edge of the table and holding it firmly there by grasping the neck and 
front quarters, the assistant at the same time compressing the root of the 
ear with the thumb and forefinger. 
4. If the hair is long, clip it. 
5. The ear is struck gently with the fingers and washed with alcohol 
and xylol; the friction will render the vein more conspicuous. 
6. The ear is grasped at its tip and stretched toward the operator, or 
the vein may be steadied by rolling the ear gently over the left index-finger 
and holding it between the finger and thumb. METHOD OF MAKING INTRAVENOUS INOCULATION 
45 
7. The inoculum should be free from solid particles and all the air 
excluded from the syringe. As a general rule, the amount injected should 
be as small as possible, and the temperature of the inoculum be near that 
of the body. If the syringe is filled shortly after sterilization, when it has 
cooled enough to be comfortably hot to the touch, the heat will warm the 
injection fluid and not be hot enough to cause coagulation. 
Fig. 41.—Intravenous Inoculation of a Rabbit. 
8. Hold the syringe as one would hold a pen, and thrust the point of 
the needle through the skin and the wall of the vein until it enters the lumen 
of the vein (Fig. 41). 
The wooden box shown in Fig. 42 is very convenient for holding rabbits 
for intravenous injection or for bleeding from the ears. 
Fig. 42.—A Wooden Box for Rabbits. 
A convenient means for holding the animal for intravenous injection and for bleeding from an ear vein. 
9. Direct the assistant to release the pressure at the root of the ear, 
and slowly inject the inoculum. If the fluid is being forced into the sub- 
cutaneous tissue, which will be evident at once by the swelling which occurs, 
the injection must cease and another attempt be made. 
10. The needle is quickly withdrawn, a small piece of cotton moistened 
with alcohol placed upon the puncture wound, and firm compression applied. 46 
TECHNIC OF ANIMAL INOCULATION 
Wash the ear thoroughly with alcohol and water to remove xylol, otherwise a 
low-grade inflammation will follow, which will render subsequent injections 
more difficult. 
Guinea-pig.—1. Since the superficial veins are quite small, it is neces- 
sary to make the injection into the external jugular vein. Rous,1 however, 
has devised a method for injecting by means of ear veins, which is especially 
useful for large light skinned animals when repeated injections are to be 
given. Roth2 has described a method utilizing a large superficial vein 
running diagonally across the back of the hind leg (Fig. 43). 
2. The animal is tied to the operating-table and the hair clipped aw7ay 
about the neck and shoulder on the right side, and 2 per cent, iodin in alcohol 
applied. 
3. A small roll is placed under the neck of the animal to render the 
operative area tenser and more easily accessible. 
4. A few drops of ether may be given by an assistant, although one soon 
learns to expose the vein quickly and there is practically no pain after the 
Fig. 43.—Roth’s Method of Injecting a Guinea-pig Intravenously. 
skin has been incised. If anesthesia is employed it should be just sufficient 
to overcome the struggles of the animal. 
5. The assistant is directed to hold the head backward in the median 
line. 
6. Pick up the skin just above and in the middle of the space between 
the shoulder and the tip of the upper end of the sternum—just above and 
about in the center of the area where a clavicle in the human would be 
situated. With small sharp scissors incise the skin for about § inch. Sepa- 
rate the subcutaneous tissue gently with forceps; a large vein at once comes 
into view (Fig. 44). Gently dissect it free for about j inch. 
7. Pick up the vein with a pair of fine forceps, insert the needle of the 
syringe gently in the long axis of the vein, and slowly inject the fluid (Fig. 45). 
8. Withdraw the needle and apply firm pressure with a wad of clean 
1 Jour. Exper. Med., 1918, 27, 459. 
2 Jour. Bacteriology, 1921, 6, 249. METHOD OF MAKING INTRAVENOUS INOCULATION 
47 
gauze or cotton. It is not necessary to tie off the vein. A stitch may be 
inserted to close the skin wound and flexible collodion applied. 
Mice and Rats.—1. Mice and rats may be injected through a caudal 
vein of the tail. These veins are quite small, and the injection requires 
a fine needle and some experience in the manipulations. 
Fig. 44.—A Dissection of the Neck of a Guinea-pig to Show the Relations of the External 
Jugular Vein. 
The skin has been turned aside and the superficial fascia and fat removed; the position of the vein 
is well shown and is readily exposed by a small and superficial incision. 
2. Fasten the mouse in a special trap so that the tail alone will be ex- 
posed. Grasp the tip between the left thumb and index-finger and hold 
the tail fully extended. 
3. A caudal vein is rendered prominent by the gentle application of 
heat in the form of hot water, or by vigorous rubbing with xylol or alcohol. 
The superficial cells become softened, and may be scraped off with a sharp 
scalpel, exposing a vein on each side of the middle line of the tail. 
4. It is usually advisable to have an assistant steady the tail while the 48 
TECHNIC OF ANIMAL INOCULATION 
inoculation is being given; a fine needle is essential. Inoculation should 
begin as near the tip of the tail as possible, and in subsequent inoculations 
gradually approach the root (Fig. 46). 
5. Rats may also be injected through the external jugular vein in exactly 
the same manner as a guinea-pig is inoculated (Fig. 45). The animal is 
fastened to a small operating board and an assistant holds the head to 
the left, which stretches the tissues of the right shoulder and side of the 
neck. A small incision is made midway between the middle line of the 
neck and the tip of the fore-shoulder. With superficial dissection a promi- 
Fig. 45.—Intravenous Inoculation of a Guinea-pig. 
The vein is steadied by a pair of fine forceps and the injection given through a small needle. The 
incision here shown is larger than actually required in practice; the vein is also smaller than normal, 
as the animal was dead for a few hours prior to making the illustration. 
nent vein appears; this vein is picked up with fine forceps and the injec- 
tion is readily given through a fine needle. 
Horse.—1. Horses are usually injected in the external jugular vein. 
2. The hair of the neck in the region of the site of inoculation should 
be clipped and thoroughly scrubbed with a hot solution of lysol. 
3. The horse should be held by an assistant; if the animal is vicious 
the injections should be given in a specially constructed stall. 
4. The vein is distended by the operator, who grasps the region with 
his left hand, the thumb being directly over the vein (Fig. 47). METHOD OF MAKING INTRAVENOUS INOCULATION 
49 
5. The needle is inserted beneath the skin and passed upward for a 
short distance, and then thrust into the vein. 
6. After the injection has been given either with a syringe or, when 
the inoculum is large in amount, by gravity from a large jar the needle 
is quickly withdrawn. Bleeding ceases as soon as pressure over the vein 
is removed. 
Sheep and Goats.—In sheep and goats the intravenous injection is 
made in the external jugular vein directly through the skin. The hair 
Fig. 46.—Method of Intravenous Inoculation of a Rat. 
The hairs and superficial layers of the skin have been scraped away with a scalpel. The vein on 
each side of the middle line appears as a bluish line in the subcutaneous tissues. 
is clipped and the part shaved and disinfected. Compression by the finger 
at the root of the neck renders the vein more prominent. Injections are 
also readily given through a popliteal or a femoral vein. If necessary, a 
small incision may be made through the skin in order to expose the vein 
chosen for the injection. 
Dog.—Dogs may be injected through the external jugular or popliteal 
veins. The animal should be fastened to the operating-table. 50 
TECHNIC OF ANIMAL INOCULATION 
2. There is a small vein just beneath the skin in the median line, along 
the anterior surface of the leg, which is readily accessible. Clip away the 
hair and disinfect with iodin and alcohol. Direct the assistant to grasp 
the thigh just above the knee to distend the vein and prevent movement, 
and make a small incision directly in the median line. A small vein is seen 
at once. Dissect free or pick up gently with fine forceps and insert a small 
sharp needle. The injection can thus be readily given. Withdraw the 
Fig. 47.—Intravenous Inoculation of Horse. 
The operator causes the vein to distend and become prominent by pressure with the left hand. 
The needle is entered beneath the skin and is pushed upward for an inch or more before the vein is 
entered. When withdrawn, this tunneled passage closes and prevents bleeding. Larger injections 
may be given in the same manner by gravity. 
needle, apply firm pressure, and insert a single stitch. Bind the wound 
with a few turns of a gauze bandage or seal with collodion and cotton. 
METHOD OF MAKING INTRACARDIAL INOCULATION 
1. Guinea-pigs may be injected by the intracardial route instead of in- 
travenously. The technic is not, as a rule, more difficult, and no ill effects 
are noticed. Not infrequently, however, attempts to inject in the heart 
fail, and frequent trials are not permissible on account of the danger of 
injuring the organ. 
2. The animal is tied to the operating board, or held firmly by an 
assistant; an anesthetic may be given 
3. Determine the point of maximum pulsation to the left of the sternum 
by palpation, and quickly insert a thin, sharp needle at the selected area. METHOD OF MAKING I.NTRACARDIAL INOCULATION 
51 
Fig. 48.—Method of Performing Intraperitoneal Inoculation of a Rabbit. 
The head is held downward; the intestines gravitate toward the diaphragm (note distention); 
this leaves an area between the umbilicus and pelvis relatively free of intestines and lessens the danger 
of puncturing the intestines. 
A flow of blood indicates that the needle has entered the heart. Attach 
the previously filled syringe and slowly inject the contents. 
4. Detach the syringe in order to make sure that the injection was 52 
TECHNIC OF ANIMAL INOCULATION 
intracardial as intended, which is indicated by a flow of blood; then quickly 
withdraw the needle. The puncture wound may be sealed with collodion. 
METHOD OF MAKING INTRAPERITONEAL INOCULATION 
Rabbit.—1. Clip the hair and shave an area about 2 inches in diameter 
in the median abdominal line just below the umbilicus. Apply 2 per cent, 
iodin in alcohol. 
2. Direct an assistant to hold the animal firmly, head down. With the 
animal in this position the loops of intestine tend to sink toward the dia- 
phragm, leaving an area above the bladder which is sometimes free of 
intestines (Fig. 48). 
3. The syringe is grasped firmly and the needle inserted beneath the 
skin for a short distance in the direction of the head in the long axis of the 
animal, when the hand is raised and the needle forced forward through the 
peritoneum. When the peritoneum has been entered this is evidenced by 
a relaxation of the abdominal muscles. The needle is then withdrawn 
slightly and the injection made. 
Guinea-pig.—1. Direct an assistant to hold the animal firmly upon its 
back. This is better than fastening it to an operating-table, for it permits 
relaxation of the abdominal wall when the injection is to be made. 
2. Clip the hair close to the skin in the median abdominal line. A small 
area may be shaved, although this is not necessary. Disinfect with an 
application of iodin in alcohol. 
3. With the left forefinger and thumb pinch up the entire thickness of 
the abdominal parietes in a triangular fold, and slip the peritoneal sur- 
faces over each other to ascertain that no coils of intestine are included. 
4. Grasp the syringe in the right hand and insert the needle into the 
fold near its base. 
5. Release the fold and inject the fluid. If a swelling forms, this shows 
that the needle is in the subcutaneous tissues, and another attempt should 
be made to enter the peritoneum. 
6. It may be difficult to pinch up the parietes without including the 
intestine. In such case straighten out the animal and stretch the skin be- 
tween the left forefinger and thumb. Insert the needle obliquely until it 
is beneath ths skin. A slight thrust suffices to pierce the peritoneum, when 
the abdominal muscles will be felt to relax. Withdraw the needle slightly 
and inject the fluid. 
7. Seal the wound with a touch of collodion. CHAPTER IV 
THE PRESERVATION OF SERUMS—METHODS 
It is well to remember that serum collected shortly after a meal is likely 
to be cloudy or opalescent; it is therefore advisable that blood be collected 
several hours after eating or during a period of fasting. 
After securing a specimen of blood the container should be set aside 
and kept at room temperature until the serum separates. If the serum 
is to be used at once blood may be collected in centrifuge tubes, allowed 
to coagulate, and then broken up, as gently as possible, with a sterile glass 
rod and thoroughly centrifuged. On account of the mechanical rupture of 
erythrocytes such serums are usually tinged with hemoglobin. After serum 
has separated from the clot it should be transferred to another tube or, 
if this is not immediately possible, the container should be placed in the 
refrigerator to retard hemolysis, which may soon occur and render the 
serum unfit for many purposes. 
Small amounts of serum are best removed from the clot with a capil- 
lary pipet and teat or with an ordinary graduated pipet with rubber tub- 
ing and mouth-piece, in order that one may see exactly what he is doing 
and not disturb the clot. As a perfectly clear serum is always to be desired, 
serums mixed with corpuscles should be centrifuged. 
It may be stated, as a general rule, that all normal and immune serums 
should be collected as aseptically as possible and handled in a careful and 
aseptic manner, so as to insure a clear and sterile product. Notwithstand- 
ing the method of preservation all serums should be kept in a refrigerator 
or ice-chest at a low temperature. 
Normal serums that are to be used for purposes of immunization are 
best preserved in small amounts in separate ampules, or in a large stock 
bottle holding from 100 to 200 c.c. and well stoppered. In the production 
of precipitin serum, for example, sufficient serum of an animal may be 
obtained at a single sitting for the whole course of injections, and this serum 
is best preserved in separate ampules. Each ampule should contain suffi- 
cient serum for one injection, and be sealed and marked. In this manner 
the risk of contaminating a stock bottle is obviated. In the preservation 
of normal serum or the serum of luetics to be used as controls for the Wasser- 
mann reaction, it is better to store them in small amounts in sterile ampules. 
As a rule, it is best not to add a preservative to serums that are to be 
used for purposes of immunization, for, if the dose of serum is large, enough 
preservative may be injected to place the health of the animal in jeopardy. 
However, chloroform may be added in proportion of 1 : 10 or 1 : 20, pro- 
vided the serum is placed in the incubator or heated in the water-bath 
at 40° C. for fifteen minutes in order to drive off the chloroform previous 
to injection. 
PRESERVATION OF NORMAL SERUMS 
PRESERVATION OF IMMUNE SERUMS 
Immune serums may be preserved either in the fluid or in the dry form. 
Preservation in Fluid Form with Antiseptics.—Practically any immune 
serum may be preserved in the fluid state by adding a suitable preservative 
53 54 
THE PRESERVATION OF SERUMS—METHODS 
in the proper dose without exerting any deleterious influence on the anti- 
body content. The exceptions to this general rule are the precipitin serums, 
because these should be crystal clear, and a preservative may render the 
serum slightly cloudy. According to Uhlenhuth, Weidanz, and Wede- 
mann, such serums should be filtered through a sterile Berkefeld filter and 
then stored without adding an antiseptic. 
Various antiseptics have been advocated for the preservation of serums. 
Hemolytic serum is well preserved by adding an equal amount of chemically 
Fig. 49.—A Small Berkefeld Filter. 
The fluid to be filtered is poured into the glass cylinder surrounding the earthen or porcelain 
■“candle.” Negative pressure within the candle is produced by the water pump, which exhausts the 
air from the flask. The filtrate is collected in the test-tube within the filter flask. All parts are readily 
sterilized in an Arnold sterilizer or autoclave. The sterile cotton plug prevents air contamination. 
pure glycerin to the serum after it has been inactivated by heating at 55° C. 
for a half-hour in a water-bath. The addition of 0.1 c.c. of a 5 per cent, 
solution of phenol in salt solution to each cubic centimeter of immune serum 
usually suffices to keep the fluid free from contamination, and produces only 
very slight, if any, clouding. Likewise, the addition of 2 per cent, formalin 
in a 5 per cent, solution of glycerin in normal salt solution, in the propor- 
tion of 1 : 10, makes a very useful antiseptic. Neither lysol nor trikresol 
should be used in the preservation of a serum, as they are more likely to 
produce clouding than does phenol. In order to avoid the formation of precipitates when cresol is added to 
serum, Krumwiede and Banzhaf1 have recently advocated the use of a mix- 
ture of equal parts of ether and cresol. On the addition of such a mixture 
to serum the solution floats on the surface, causing a slight haze at the point 
of contact. If shaken immediately no precipitate forms or, at most, very 
little. The addition of ether to serum for therapeutic purposes does not 
appear to be harmful and the mixture is recommended for the preservation 
of antitoxins and other sera. 
PRESERVATION OF IMMUNE SERUMS 
55 
The cotton plug in the “candle” is removed and the fluid poured within the candle (hollow). 
The water is then turned on and the stop-cocks are opened; a vacuum is produced within the flasks, 
which draws the fluid through the candle. The filter is readily cleansed and sterilized (autoclave) 
and is quite efficient. 
Fig. 50.—A Filter. 
Preservation in Fluid Form by Bacteria-free Filtration.—If serums are 
to be preserved in fluid form without the addition of an antiseptic, special 
precautions in bleeding, collecting, and separating should be observed. If 
contamination has probably occurred, the serum should be filtered through 
a sterile Berkefeld filter (Fig. 49). If the serum proves to be sterile it is trans- 
ferred, with the aid of a sterile pipet, into ampules of 1 c.c. capacity. These 
are sealed hermetically and kept in the refrigerator. 
Small amounts of serum may be lost in a large filter and a smaller 
1 Jour. Infect. Dis., 1921, 28, 367. 56 
THE PRESERVATION OF SERVMS—METHODS 
apparatus should therefore be used. The filter shown in the accompanying 
illustration (Fig. 50) is quite serviceable, as the flask and earthen candle- 
filter may be wrapped in a towel and sterilized in the autoclave. The appa- 
ratus may carefully be attached to a suction pump and the serum pipeted 
off into the hollow of the candle and filtered, the filtrate being removed 
at the completion of the process by another sterile pipet. 
Method of Cleaning Filter.—Care should be exercised regarding the re- 
action of Berkefeld filters and particularly new filters. Before use they 
should be boiled in distilled water at least three times for five minutes each 
time, and scrubbed with a small soft brush after each boiling. After the 
filter is set up hot, neutral, distilled water should stand in it for about 
five minutes and then wTashed through under gentle pressure until the fluid 
is clear and neutral to phenolphthalein, when the filter is ready for use. 
After use the filter should be boiled in distilled water, scrubbed, and dried 
in the air. 
Preservation in Fluid Form by Freezing.—Freezing a serum often ren- 
ders it cloudy or causes a precipitate to be deposited, and interferes with 
the usefulness of a serum that should be absolutely clear. Freezing is the 
only practicable method so far devised for the preservation of thermolabile 
substances, such as complement. A small apparatus, named the “Frigo,” 
has been devised for this purpose by Morgenroth. A satisfactory appa- 
ratus may be made by constructing a wTooden box with a smaller sheet- 
metal-covered inner compartment, the space between them being well 
packed with sawdust. This inner box is then filled with crushed ice, and 
the wThole is covered with a lid lined with several layers of felt. 
Preservation in Powder Form.—When serum is poured out in thin 
layers and dried, it forms yellowish, amorphous masses, that may be col- 
lected and ground into a powder, which keeps well and forms an excellent 
medium for the preservation of many immune serums, especially those of 
the agglutinating type. Various toxins, such as tetanus toxin and cobra 
venom, may also be preserved in this form. 
The serum or toxin may be spread out in thin layers on large glass plates, 
or placed in shallow dishes and dried in the incubator. After a few hours 
the dried serum, which adheres only slightly to the dish, can be removed 
with a spatula and placed in a mortar, and ground and stored in sealed 
tubes. 
The drying process is better carried out in vacuo, and the large serum 
institutes are provided with these special drying apparatus. A simple form 
may be prepared after the method of Taeze as follows: Place a large glass 
bell-jar with a ground base and a large opening at the top on a polished 
iron plate. Set this on a large tripod, as this will facilitate heating with a 
Bunsen burner. The serum is placed within the jar in a shallow dish, and 
the jar fastened to the iron plate with hot paraffin or wax. The opening at 
the top is closed with a three-holed rubber stopper: one hole carries a 
thermometer; a second is connected with a manometer (not absolutely 
necessary), and the third carries a bent glass tube which is connected, by 
means of thick-walled rubber tubing, to a suction pump. A low flame is 
kept burning so as to keep the temperature at about 35° C. The degree of 
vacuum secured makes little difference, and usually that obtained with an 
ordinary water-suction pump, allowing for leaks in the tubing, is sufficient, 
rendering manometric measurements unnecessary. 
I have secured equally good results in evaporating serums and tissue ex- 
tracts with an ordinary electric fan enclosed in a properly sized oblong 
wooden box to concentrate the air current and exclude dust (Fig. 51). PRESERVATION OF IMMUNE SERUMS 
57 
The dried serum should be dissolved in sterile normal salt solution be- 
fore it is used. 
Preservation in Dried Paper Form.—This is a very serviceable method 
for preserving hemolysins and, to a lesser extent, agglutinins. In the preser- 
vation of hemolytic amboceptor Noguchi advises the use of Schleich and 
Schull’s paper No. 597. The paper is cut into squares about 10 by 10 cm., 
and saturated with the serum which, after preliminary titration, has been 
found satisfactory. Sufficient serum is added to wet the sheets evenly, 
any excess of serum being absorbed wTith other sheets of paper. Each 
square is placed separately upon a clean sheet of unbleached muslin and 
dried at room temperature. When thoroughly dry the squares are care- 
Fig. 51.—A Convenient Box for Drying Serums, Extracts, Etc. 
fully ruled off with a hard pencil into widths of about 5 mm. and cut into 
strips. The paper is then standardized and preserved in dark glass vials 
in a cool, dark place. 
Preservation in the Living Animal.—In the living animal an immune 
serum may be preserved by removing a small amount of blood from time 
to time as needed, the titer being preserved or raised by occasional injec- 
tions. This method, however, may be unsatisfactory and expensive, espe- 
cially with the smaller animals, as they frequently show a marked tendency 
to sicken and die, or may succumb to anaphylaxis. After a time, too, they 
fail to respond to injections with the formation of antibodies, a condition 
ascribed to atrophy of the cell-receptors (receptoric atrophy). Part II 
CHAPTER V 
INFECTION 
Parasitic Causes of Disease.—Broadly considered, the causes of all 
diseases may be grouped in four classes as follows: (1) Vegetable and animal 
parasites, or the products of their activity; (2) chemical agents of non- 
parasitic origin, as the inhalation and swallowing of poisonous gases and 
dust in the industrial diseases; (3) physical agencies, as trauma and heat, 
and (4) those of unknown etiology. 
The subject of infection concerns all diseases caused directly or indirectly 
by vegetable and animal parasites; directly, as in the infectious diseases, 
and indirectly, as for example, by the poisons believed to be produced by 
the bacterial flora of the intestines under certain abnormal conditions. 
Infection is not to be confused with mere surface contamination, and 
many factors are to be considered in relation to both the parasite and the 
resistance of the host; this chapter deals with the subject of the mechanism 
of infection by vegetable and animal parasites, and the succeeding chapter 
with the mechanism of the production of disease by these parasites. 
Disease may be caused, therefore, by both vegetable and animal para- 
sites as follows: (1) By the pathogenic bacteria; (2) by the pathogenic 
higher plants or microfungi; (3) by pathogenic microparasites of animal 
origin as some of the protozoa, etc.; (4) by some of the larger animal para- 
sites, as the worms, and (5) by unknown filterable viruses of plant or animal 
origin. 
Our knowledge of the mechanism of infection is largely based upon 
studies with the pathogenic bacteria; much of this information is applicable 
to the mechanism of infection with microfungi and protozoa and to infesta- 
tion with the larger animal parasites. 
Throughout this chapter I have used the word microparasite as in- 
cluding the bacteria, microfungi, animal parasites of microscopic size with 
particular reference to the pathogenic protozoa, and the filterable viruses; 
the word parasite has been used to include these and likewise animal parasites 
of macroscopic size. 
Invasion and Contamination.—The skin and adjacent mucous mem- 
branes contain numerous microparasites, and under normal conditions 
these may invade the tissues, but they are usually quickly destroyed and 
unable to proliferate, so that mere invasion does not necessarily constitute 
infection. 
Unfortunately, custom has sanctioned the use of the term “infection” 
as synonymous with contamination. The bacteriologist may speak of the 
air, water, or his culture-medium as being infected when they contain micro- 
organisms, or in other words, are not sterile; similarly the surgeon may 
speak of a knife or splinter of wood as being infected, whereas, while these 
may be infective or capable of producing infection, it is more accurate to 
speak of them as being contaminated. In the early days of bacteriology, 
the mere presence of micro-organisms in or on the skin and mucous mem- 
branes was regarded as equivalent to infection. It is now well known that 
58 DEFINITIONS 
59 
a person may harbor various micro-organisms, such as staphylococci, strepto- 
cocci, and pneumococci, without apparent injury to the host, and this 
surface contamination, or even occasional invasion of the tissues, does not 
necessarily indicate that the host has been, is, or will be ill. 
Definitions.—When, however, microparasites have passed the normal 
harriers of the skin or mucous membranes and have invaded and proliferated 
in the deeper tissues, the process is spoken of as an infection. 
By common consent, the term infestation, or infestment, is being applied 
in a similar manner to the presence and growth of animal parasites of macro- 
scopic size, as the intestinal worms; thus, the intestine may be infected with 
Bacillus typhosus, and infested by Taenia saginata. 
A microparasite may be intimately associated with and have its norma 
habitat in a certain part of the body and do no harm until special condi- 
tions arise, when it may rapidly invade the tissues and produce infection; 
this condition has been described by Adami as a subinfection, and is illus- 
trated by the constant presence of staphylococci and streptococci in the 
tonsils of most persons, usually harmless, but capable, under special condi- 
tions, of producing severe and even fatal infection. 
The abnormal state resulting from the deleterious local and general inter- 
action between a host and an invading parasite, with consequent tissue changes 
and symptoms, constitutes an infectious disease. 
As has previously been stated, not every invasion of the deeper tissues 
by microparasites results in injury or disease. A certain number of bacteria 
and protozoa are constantly gaining admission to the deeper tissues of the 
alimentary and the respiratory tracts without producing apparent injury 
to the host, as they tend to be destroyed very soon after they gain entrance. 
Furthermore, bacteriologic studies of lymphatic glands and other tissues 
removed during life or soon after death at autopsy, not infrequently show 
the presence of diphtheroid bacilli and other micro-organisms possessing 
feeble or no demonstrable pathogenic powers and indicating that various 
bacteria may gain access to the deeper tissues without producing a true 
infection. The terms invasion and infection are not, therefore, synonymous. 
Every true infection is accompanied by local changes, although these may 
be so slight as to escape notice; an infectious disease is practically made up of 
similar phenomena, but these are of an exaggerated or marked degree. 
The hygienist distinguishes between: (1) Sporadic, or isolated, cases of 
infection; (2) endemic, in which a certain microbic disease affects the in- 
habitants of a given area year after year, and (3) epidemic, in which a dis- 
ease appears suddenly and affects a large number of inhabitants, the num- 
ber of cases rapidly increasing and decreasing. Among the lower animals 
equivalent terms for the types just described are sporadic, enzootic, and 
epizootic. A pandemic disease is one that is epidemic over a large territory. 
In all infections there are two inseparable factors to be considered: 
1. The offensive forces of the infecting agent, dependent upon its patho- 
genic or disease-producing nature and its power of defending itself against 
the antagonistic forces of the host and of thriving under these conditions. 
2. The resistance offered by the host and mainly dependent upon cer- 
tain physical or non-specific local factors or specific antibodies, which con- 
stitute the defensive mechanism, or immunologic factors. 
The former is concerned with the general subject of infection and the 
latter with that of immunity. 
Parasites and host may live together in apparent harmony, owing to the 
ability of the host to restrain the activity of the parasite and neutralize its 
injurious effects or to an absence of infectivity on the part of the parasite 60 
INFECTION 
until the vital resistance of the host is diminished or the pathogenicity of the 
parasite is increased, when the neutral relations are disturbed and infection 
occurs. 
RELATION OF INFECTION TO IMMUNITY 
From what has been said it is apparent that the subject of infection 
forms the basis for the study of immunology, for, paradoxic as it would at first 
appear to be, infection must usually have occurred in order that immunity 
may be acquired. This relation is not always apparent; for instance, man 
and some of the lower animals may possess a natural immunity to a cer- 
tain parasite because of the presence of various physical or non-specific 
defensive factors, or to specific antibodies produced as the result of an 
earlier and unrecognized infection, or even one that has been inherited; 
under any circumstances, however, natural immunity is usually relative 
and seldom absolute. In passive immunity the same conditions are gener- 
ally operative, and the antibodies present in the serum used to confer a 
passive immunity are produced in some other animal as the result of an 
active infection. 
It may be stated,-therefore, that specific antibodies are produced only 
by stimulation of the body cells, and that this stimulation is furnished by 
the infecting agent either in living, disease-producing form, or in a modified 
and attenuated state, i. e., in the form of a vaccine; thus it will be seen that 
infection and immunity are intimately associated, and that, generally 
speaking, there can be no pronounced protection unless infection has taken 
place. 
SOURCES OF INFECTION 
Parasites, and particularly microparasites, are to be found everywhere. 
For general purposes they may be roughly divided into two classes—sapro- 
phytes and pathogens. The saprophytes are those parasites which thrive 
best in dead organic matter and perform the very important function of 
reducing, by their physiologic activities, highly organized material into 
those simple chemical substances that may again be utilized by the plants 
in their constructive processes, and in this manner maintain the important 
chemical relation between the animal and the plant kingdom. Pathogens, 
on the other hand, find the most favorable conditions for growth and activity 
upon the living tissues of higher forms of animal life. They include most 
of the pathogenic, or disease-producing, bacteria and animal parasites. 
No marked separation between these two divisions can be made, as 
numerous species occupy a transition point between the two. The terms 
are merely relative, and parasites ordinarily saprophytic may develop 
pathogenic powers when the resistance of the host is sufficiently reduced 
by another infection, fatigue, exposure, or other deleterious influence. In 
other words, a pathogen or pathogenic microparasite is one that can grow 
in the living tissues because the immunologic defenses of the host are not 
sufficiently strong to resist it; in most cases, however, as will be pointed 
out further on, a higher degree of immunity can be produced artificially, 
rendering the microparasite in question relatively harmless for that particular 
animal. Similarly, under certain circumstances, the resistance of the body, 
or of a part of it, may be broken down to such an extent that micro-organisms 
ordinarily regarded as saprophytes may gain access to the deeper tissues, 
flourish, and produce disease. 
Accordingly, no fundamental distinction between pathogenic and non- 
pathogenic parasites can be made. Any apparent differences are due not 61 
EXOGENOUS AND ENDOGENOUS INFECTIONS 
only to various degrees of pathogenicity possessed by the parasite, but 
also to the different degrees of resistance against their attacks, since a 
microparasite that is highly pathogenic toward one animal may be quite 
harmless to another. 
CONTAGIOUS AND INFECTIOUS DISEASES 
Just as all pathogenic parasites do not possess the same habits of growth, 
so, likewise, they vary in their vitality and in their ability to proliferate 
under various conditions when removed from the animal body. Some are 
able to grow only at body temperature or, indeed, only in the human body 
itself; when removed from these conditions they may retain their vitality 
for a short period of time, but are Unable to proliferate; from this it follows 
that communication of these parasites and their disease must be direct 
or immediate, i. e., from person to person, or almost direct, by the con- 
veyance of the infecting agent in the form of fomites, such as dust, epidermal 
scales, or discharges, or as the result of bites of suctorial insects. This 
form of infection, which requires such direct means of transmission, and of 
which gonorrhea is an example, constitutes what are known as contagious 
diseases. 
Other parasites may be able to preserve their pathogenic powers and pro- 
liferate outside of the body at ordinary temperatures, and may even with- 
stand great extremes of heat or cold and various nutritional deficiencies; they 
may exist thus for weeks, and carry the disease to a second individual through 
contaminated material. Infections, the result of indirect transmission, are 
known as infectious diseases. 
There are no hard-and-fast rules that can be set down in classifying 
parasitic infections; parasites that are commonly transmitted by one means 
may, under slightly altered conditions, be transmitted by another. The 
usual classification by which certain diseases are classified as contagious and 
others as infectious should be abolished, and all should be grouped under the 
term “infectious,” there being a definite understanding of those cultural 
characteristics that render infection more likely to occur by direct and 
immediate contact, and those that may occur in an indirect or roundabout 
manner. It may, therefore, be stated that all parasitic diseases are infectious; 
the term “contagious’’ may be reserved for those spread or contracted as the 
result of direct contact. 
Infection may occur as the result of the admission of microparasites to 
the tissues from sources entirely apart from the individual infected (exo- 
genous infection), or from the admission of some of those microparasites 
living normally and harmlessly on the skin and adjacent mucous mem- 
branes, and which, under special conditions, have assumed pathogenic 
properties (endogenous infections). 
Exogenous infections are the more usual form, and result from contact 
with infective material outside the body. 
1. Microparasites, such as typhoid and cholera bacilli, and pathogenic 
amebas, which can live for varying periods of time in water and foods, are 
particularly likely to gain entrance through the gastro-intestinal tract. 
Micro-organisms may be present in milk derived directly from diseased 
animals or tissues, and when ingested, may produce disease. Thus, for 
example, the bacillus of tuberculosis may be conveyed in either milk or 
flesh, young children being particularly exposed to this method of infection. 
EXOGENOUS AND ENDOGENOUS INFECTIONS 62 
INFECTION 
2. The atmosphere may be laden with micro-organisms, which, whether 
or not capable of proliferating outside of the body, are prone to gain en- 
trance through the respiratory tract, especially through the upper air- 
passages, the pharynx and tonsils being often the seat of the infection. 
3. Micro-organisms capable of existing on the skin may gain entrance 
to the deeper tissues as the result of wounds. Under these conditions of 
lowered vitality of the local tissues micro-organisms that would otherwise 
be harmless may become pathogenic and morbidly affect the host, either 
locally or generally. As the skin is brought so freely in contact with ex- 
ternal objects, various microparasites, and particularly the pathogenic 
cocci, may gain entrance to the dermis. Wounds may be infected by the 
teeth and secretions of animals, or by various weapons and implements con- 
taminated with infective material, as, e. g., the virus in the saliva of rabid 
dogs, or the spores of the tetanus bacillus on rusty nails. 
Contact with unclean objects of various kinds—eating utensils, catheters, 
syringes, dental instruments, etc.—may serve to transfer pathogenic para- 
sites from one person to another. This is especially likely to occur if the 
skin or mucous membrane is abraded, the infecting parasites thus gaining 
ready access to the deeper tissues. In some infections, however, even this 
local injury is unnecessary, as the parasite may be able to proliferate and 
produce lesions on an intact surface, as, for instance, the diphtheria bacillus 
in the pharynx, and various fungi, such as Achorion, Trichophyton, etc., 
on the scalp and skin in general. 
Microparasites affecting the genital organs are likely to be conveyed 
directly from one sex to the other in conjugation, or to the child during 
parturition. 
4. Suctorial insects may serve as the medium by which microparasites 
are transmitted from person to person. In most instances the transmis- 
sion is a purely mechanical process, as witness the transmissions of the 
plague bacillus in the intestinal contents of the rat flea; in the case of malaria, 
on the other hand, the interposition of the mosquito is essential to complete 
the life cycle of the protozoon. 
5. Micro-organisms infecting the placenta may pass to the fetus by way 
of the umbilical vein. 
Endogenous infections arise as the result of the activity of microparasites 
having their normal or customary habitat in the body. Such infections do 
not represent so much an assumption of pathogenic power on the part 
of the microparasite, as they do a disturbance of the defensive mechanism 
of the host, whereby the normal relations are disturbed, and microparasites 
that normally are harmless, become infective and disease-producing. While 
the disturbance of the defensive mechanism may be general, it is far more 
likely to be local; an example is that of appendicitis the result of Bacillus 
coli infection following passive congestion due to fecal impaction of the 
colon. 
AVENUES OF INFECTION 
Local infection may occur in any portion of the body, and any part 
may prove the point of entrance of parasites to the body fluids, the result 
being a general infection. Owing, however, to the peculiar pathogenic 
properties of different bacteria and their affinity for the cells of certain 
tissues, coupled with a peculiar tissue susceptibility for certain bacteria 
or their products, we find that many diseases have regular avenues of in- 
fection, and, indeed, in a few instances infection of the human body may 
be possible only through a particular and definite route. Infections of the AVENUES OF INFECTION 
63 
gastro-intestinal, respiratory, and genito-urinary tracts and various sinuses 
with external openings must be considered as being potentially surface in- 
fections. The outer layers do not consist merely of the skin and adjacent 
mucous membranes, but are made up of all layers covering surfaces and 
channels, which, however, indirectly communicate with the exterior. In 
the higher animals there is only one direct channel of communication be- 
tween the actual interior and the exterior of the body, this being through 
the fallopian tube of the female, which normally has so fine a lumen and 
is so well protected that to all intents and purposes it may be regarded as 
closed. In certain inflammatory conditions of the genital organs, and par- 
ticularly after parturition, the fallopian tube may be open, and afford a 
direct route for the transmission of infection from the external parts to the 
peritoneal cavity. 
Living in and on the actual and potential external surfaces are count- 
less micro-organisms, which are for the most part harmless, a few being, 
however, actually or potentially dangerous. 
1. The skin and adjacent mucous membranes, particularly in those por- 
tions where warmth and moisture abound, are well adapted to bacterial 
growth, and their contact with surrounding objects causes a large variety 
of micro-organisms to adhere to them. 
As a result, the bacteriology of the skin is quite complex, since it may 
lodge micro-organisms from the air, from water, and from soil. A group 
of cocci and diplococci, particularly the Staphylococcus epidermidis albus 
of Welch, and the various pseudodiphtheria bacilli, are habitually present 
upon the human skin. When local injury occurs, they may produce minor 
suppurative lesions, and may be concerned in the production of certain 
skin diseases, such as eczema, impetigo contagiosa, the pustules of variola, 
etc. 
Other micro-organisms may find temporary lodgment upon the skin, 
and are in no sense regular inhabitants. For example, the fingers and hands 
may become contaminated with colon, typhoid, and tubercle bacilli, pneu- 
mococci, etc. 
The skin forms a very important barrier against the entrance of para- 
sites into the deeper tissues. The greater number of local surgical infec- 
tions result from the entrance of bacteria into lesions of the skin, although 
these lesions may be so small as to escape notice. 
Certain parasites are capable of producing direct action on the skin 
without previous existing injury, and especially upon the mucous mem- 
branes, where moisture and higher temperature are more favorable to 
growth. For example, a few of the higher fungi, such as Microsporon, 
Achorion, and Trichophyton, seem able to establish themselves in the 
superficial cells and invade the deeper tissues through the hair-follicles; 
staphylococci may reach the roots of hair-follicles and sweat-glands and 
set up suppurative conditions; diphtheria bacilli may lodge directly on the 
intact mucosa of the upper air-passages and cause local necrosis and general 
intoxication; cholera bacilli may have a similar effect upon the intestinal 
mucosa; the Koch-Weeks’ bacillus and the gonococcus may produce severe 
inflammation of an intact conjunctiva, etc. 
2. The respiratory organs commonly afford admission to certain parasites. 
The nose may be the seat of local infection with Bacillus influenzae, Micro- 
coccus catarrhalis, Bacillus diphtheriae, and other bacteria; it maybe the en- 
trance point for meningococci and the virus of anterior poliomyelitis. Simi- 
larly, the entrance of such unknown infectious agents as those of scarlet 
fever, measles, and smallpox can best be accounted for by assuming that 64 
INFECTION 
they were inhaled and later entered the blood; there is much clinical evidence 
to support the belief that the contagium of scarlet fever is present in the 
discharges of the upper air-passages of persons suffering from that infection. 
Whether or not tuberculosis of the lungs is the result of the inhalation 
of tubercle bacilli is a much disputed point, but it cannot be denied that 
this theory most readily accounts for the far greater frequency with which 
tuberculosis affects the lungs than it does other organs of the body. 
Pneumonia, caused by the pneumococcus of Weichselbaum, probably 
results from the direct inhalation of one of the various types of pneumo- 
cocci, and bronchopneumonia of children is certainly chiefly inspiratory in 
origin. 
3. The digestive tract may be the portal of entrance of many infections. 
The mouth usually harbors various fungi and bacteria, which may produce 
local infections, and either directly or indirectly cause caries of the teeth. 
The putrefactive changes they may produce is being generally recognized 
as having an important bearing on the causation and symptomatology of 
several infections, and a carious tooth has been found the portal of entry 
of micro-organisms causing a general infection. The tonsils are well known 
to be the breeding and lodging place of various microparasites causing 
many general infections, such as acute rheumatic fever, tuberculosis, and 
possibly typhoid fever. The pharynx may harbor the micro-organisms of 
diphtheria, pneumococcus angina, etc. 
Normally, except for the presence of a few sarcinae, the stomach is prac- 
tically sterile. Under special conditions, however, typhoid, dysentery, 
cholera, tubercle, and other infectious bacteria may escape the germicidal 
effects of the hydrochloric acid, and, reaching the alkaline intestinal con- 
tents, which are rich in soluble proteins and carbohydrates, are rendered 
capable of producing their respective infections. 
Although these conditions are primarily of the nature of local infection, 
there is much experimental evidence to show that bacilli, and particularly 
tubercle bacilli, may pass through a practically intact intestinal wall and 
find their way to the lymph-glands or to the blood-stream itself. 
Aside from these direct and specific infections, various other micro- 
organisms, by fermentative action, may alter the intestinal contents and 
produce toxic products capable of exciting acute and severe toxemias. 
Some authorities—e. g., Metchnikoff—regard the various types of colon 
bacilli as producing toxic products responsible for chronic degenerative 
lesions of the cardiovascular and other organs. The digestive tract is there- 
fore regarded by some pathologists as a constant menace to health, in that 
it permits bacteria to enter the lymphatic and blood-streams, or to pro- 
duce toxic substances detrimental to health and longevity. Adami has 
drawn particular attention to a condition which he terms subinfection and 
which is dependent upon the constant entrance of colon bacilli into the blood, 
whence they enter the liver, where their final dissolution takes place, appear- 
ing as fine, dumbbell-like granules inclosed in the cells. 
4. The genital organs are the seat of various local infections that may 
become wide-spread and general. Normally, the urethra may contain a 
few cocci which lodge about the meatus; the acid secretions of the vagina 
are generally inimical to bacterial growth, and the uterus and bladder are 
usually sterile. But three micro-organisms—the gonococcus, Treponema 
pallidum, and the bacillus of Ducrey—here find favorable conditions for 
growth, and are usually transmitted from person to person by means of 
sexual congress. The local gonococcal lesion may be the portal of entry 
of gonococci into the blood-stream, resulting in wide-spread metastases in NORMAL DEFENSES AGAINST INVASION 
65 
the heart valves and joints. The local syphilitic lesion is quickly followed 
by general infection. Chancroids alone remain localized, although the 
initial lesion frequently spreads quite rapidly by continuity of tissues. In 
rarer instances other micro-organisms, such as the ordinary pyogenic cocci, 
tubercle bacillus, and diphtheria bacillus, may infect these organs and be 
transmitted by sexual conjugation. t 
5. There is considerable controversy of opinion regarding the suscep- 
tibility of the placenta and the filtering properties it possesses for various 
infectious agents. A study of this subject by Neelow1 would indicate that 
the non-pathogenic bacteria do not pass from the mother through the 
placenta to the fetus. Other pathogenic agents may, however, pass through 
quite readily; for example, pregnant women suffering from smallpox may 
be delivered of infants showing active lesions of prenatal infection, and 
syphilitic infection of the fetus is a well-known condition. Most contro- 
versy centers around congenital tuberculosis, and directly opposing views 
for and against prenatal infections are held by several authorities. Baum- 
garten is of the opinion that many children are subject to antenatal infec- 
tion, though the disease infrequently develops in a few of them. 
The general subject of antenatal infection and pathology is a field re- 
quiring considerable investigation and research. 
NORMAL DEFENSES AGAINST INVASION 
When the large area of the body that is subject to traumatic injury and 
accidental infection is considered, it is remarkable that, considering the 
enormous numbers of various bacteria, infection does not occur more fre- 
quently. 
Bacterial invasion of the tissues is of frequent occurrence, but in health 
they do not usually cause infection and tend to be destroyed very soon 
after they enter the tissues. 
It may be well to discuss at this point the factors tending to prevent 
invasion, and leave the consideration of the defensive mechanism whereby 
the body destroys bacteria after successful invasion and thus prevents 
infection for the chapter on Natural Immunity. 
Of the factors preventing invasion with bacteria and animal parasites, 
the following are recognized: 
1. The structure of the surface layer of epithelium. The epidermal 
cells offer a mechanical obstacle to invasion. This resistance is naturally 
more complete where the cells are thickened and most compact. In the 
depths of glands and in mucous membranes where numerous glands are 
present, and where the layers are thinner and moisture exists, the barrier 
is less complete. 
2. Surface discharges are potent factors in preventing invasions by: 
(1) Washing away the microparasites mechanically; (2) by germicidal 
activity through the presence of various chemical agents, such as acids, 
which they may contain, and (3) by antiseptic and even bactericidal sub- 
stances that may be present in the form of antibodies. 
The saliva, with its antiseptic and germicidal properties, is potent in 
preventing infections of the mouth and upper air-passages; when this secre- 
tion is diminished, as during the course of high fever, bacterial activity is 
enhanced, which is evidenced by the development of fetid sordes about 
the teeth and on the lips. 
The acidity of the gastric juice and its germicidal powers are well known 
1 Centralb. f. Bakt., 1902, orig., xxxi, 691. 66 
INFECTION 
and appreciated; similarly, the urine, the milk, and to a slight extent, the 
bile, have been demonstrated by Adami to exert a distinct antiseptic effect 
upon certain bacteria, such as the Bacillus coli. 
Surface moisture and discharges about the nose and throat are also 
potent factors in mechanically removing bacteria from inspired air and 
no doubt frequently prevent bacterial invasion of the lower respiratory 
tract, where more mischief may be done. 
3. The cells of certain excreting glands may possess bactericidal and 
excretory powers of value in preventing bacterial invasion (Adami). 
MECHANISM OF INVASION 
We will now consider the method by which invasion, the first step of 
what may be an infection, is brought about. In brief, one or all of the 
normal defenses just described must be overcome; in some instances the 
parasites, by their inherent disease-producing powers, may accomplish this 
unaided; in other instances the resistance is overcome by a general lowering 
of the vitality of the body defenses. 
1. Traumatic solution of the surface layers of epithelial cells is a very 
important factor in the production of infection, as the invading micro- 
parasites are thus given easier access to the deeper and less resistant tissues. 
The pathologist or surgeon may, in the course of his work, contaminate 
his hands with secretions containing virulent micro-organisms, and may 
yet escape infection unless a small break in the surface epithelium, in the 
form of a scratch or a needle-prick, is present. 
2. As has been previously stated, certain bacteria, notably the diph- 
theria bacillus, by concentrating at one point, may lower the vitality and 
cause necrosis of superficial cells of the mucosa lining the upper air-passages, 
and in this manner induce a local break in the continuity of the epithelial 
covering. Staphylococci may exert a similar action in the depths of sweat 
and sebaceous glands, and, indeed, certain fungi, such as the Trichophyton, 
Microsporon, and Achorion, may attack the intact skin. While, therefore, 
solution of the surface coverings is a very important source of many infec- 
tions, it is not essential for the production of all. 
3. Alterations of the surface discharges, either in quantity or in quality, 
may permit bacteria to proliferate freely and produce sufficient toxic matter 
to affect the surface cells, lower their vitality, and destroy them, with the 
result that they may gain entrance to the deeper tissues. When the secre- 
tions are diminished or altered, as, for example, the saliva during a fever, 
unless the mouth is carefully and frequently cleansed, it becomes putrescent 
with bacterial growth. Similarly, catarrhal gastritis, or any other factor 
tending to lower acidity of the gastric juice, favors infection by this route. 
4. Not infrequently bacteria may gain access to the deeper tissues or 
to an internal organ, and infection may occur without any recognizable 
solution of continuity of the surface epithelium. In these hidden or “crypto- 
genic infections” the entrance point of the parasites may be healed over 
or the infecting micro-organisms may have been carried to the circulating 
body fluids by the wandering cells. 
Not infrequently, in cases of tuberculosis of the cervical and mesen- 
teric glands in children, there may be no signs whatever of local irritation 
in the fauces or in the intestine to explain the source of infection. The 
tonsils are now strongly suspected and, indeed, known to be the source of 
entry of bacteria causing several acute and chronic infections. 
The leukocytes, in their phagocytic activities, no doubt play an im- MECHANISM OF INVASION 
67 
portant role in the production of cryptogenic infections, especially when 
an excessive number of pathogenic bacteria have congregated at one point, 
and congestion, increased leukocytic infiltration, and a lowered vitality of 
the tissues have occurred prior to the invasions of micro-organisms. Wan- 
dering cells are commonly found on mucous membranes, gathering up 
various bacterial and cellular debris. They may carry a virulent micro- 
organism into the deeper tissues and, although this may not produce an 
infection, a large number of bacteria so transported may be able to resist 
destruction and prove capable of causing infection. 
5. Aside from the question of local conditions in the process of infec- 
tion other factors may exert an influence. The temperature of the host 
may be unsuitable for the growth of a certain parasite, even though it has 
gained entrance to the deeper tissues; a particular route for the introduc- 
tion of the infecting agents may be necessary, as in typhoid fever and cholera, 
which are probably always intestinal infections, and, finally, even after the 
infecting agent has reached the deeper tissues, extension is prevented by a 
local inflammatory reaction. In many such instances the question of natural 
immunity is brought into intimate relation with the subject of infection. 
After invasion has occurred some bacteria can best sustain themselves 
against the defenses of the host at the local point of entry. Such micro- 
organisms may, however, possess unusual vitality and indirectly, through 
the lymphatics, find their way to the blood-stream, producing a bacteremia. 
This is a morbid condition characterized by the presence of micro-organisms 
in the circulating blood. 
Some micro-organisms may gain entrance to the general circulation 
more readily than others, and their mode and route of entry vary in the 
different infections. It is essential that they possess an unusual degree of 
invasive power, and be capable of protecting themselves against the mani- 
fold defensive factors contained in the blood. Kruse believes that in local 
infections the high pressure of an inflammatory exudate may force bacteria 
into the adjacent vessels; that they may sometimes be carried into the 
deeper tissues, and even into the blood-stream, by leukocytes is not to be 
denied. 
When bacteria have entered the circulation they may act as emboli 
in the finer capillaries or, being unable to remain in the circulation, may 
collect in the capillaries of less resistant tissues, proliferating and produc- 
ing local metastatic lesions, usually purulent in character. The condition 
thus produced is known as pyemia. 
Saprophytic bacteria or pathogenic bacteria of feeble invasive powers 
may be able to grow in diseased tissues, such as gangrenous areas, and may 
assist in effecting morbid changes, producing toxic products of decomposi- 
tion which, when absorbed into the body, give rise to a series of toxic phe- 
nomena, such as fever, rapid pulse, malaise, etc. This condition is known 
as sapremia, a term that has also been applied to the decomposition of 
relatively sterile organic material and absorption of the toxic products, as 
when portions of placenta or fetal membranes are retained in the uterus 
after childbirth. 
The term toxemia is employed rather loosely to mean the presence of any 
toxic material. Its use should be limited to the condition resulting from the 
absorption of the poisonous substances produced by the non-invasive bacteria them- 
selves, as in diphtheria and tetanus. Septicemia is the term applied to the presence 
in the body fluids of toxic products generated by the pyogenic micro-organisms. 68 
INFECTION 
MECHANISM OF INFECTION 
Since bacterial invasion is of frequent occurrence the question naturally 
arises, Why are not infections, both local and general, more frequent? Thus, 
abrasions of the surface epithelium are not uncommon in the presence of 
active micro-organisms; tubercle bacilli may be inspired, and typhoid bacilli 
may be swallowed, the altered local conditions affording opportunity for 
producing infection, and yet the host may escape. 
Bacterial invasion, therefore, does not necessarily mean infection, and it 
may be stated that infection can only take place when— 
1. The micro-organisms are sufficiently virulent. 
2. When they invade the body by appropriate avenues and reach sus- 
ceptible tissues. 
3. When they are present in sufficient numbers. 
4. When the host is generally susceptible to their action. 
5. When the micro-organisms are able to resist the defensive forces of 
the host through special agencies aside from their offensive forces. 
Not all these factors must necessarily be present before infection may 
occur. A micro-organism may be particularly virulent, so that numbers 
are relatively unimportant; a host or a portion of the host may be so sus- 
ceptible or vulnerable to infection that a micro-organism of low virulence, 
which, under normal conditions, would be totally unable to produce infec- 
tion, may now prove pathogenic. 
VIRULENCE 
Virulence refers to the disease-producing power of a micro parasite, and is 
dependent upon two variable factors: (1) Toxicity, and (2) aggressiveness, or 
the invasive power of the microparasites. In most infections usually both 
factors are operative. 
Toxicity is the term applied to the kind and amount of poison or toxin 
produced. This poison may be readily soluble, or exogenous, diffusing into 
the surrounding tissues and being readily absorbable; or it may be endog- 
enous, and contained chiefly within the micro-organisms, and be liberated 
only upon the dissolution of the cell. 
Aggressiveness is a term applied to the invasive powers of a micro- 
organism to enter, live, and multiply in the body-fluids, or, in other words, 
to the aggressive or progressive forces of the micro-organism in its new 
environment. 
Toxicity is generally confused with aggressiveness, a highly toxic micro- 
organism being regarded as an aggressive one. For example, the bacillus 
of tetanus is highly toxic because of the production of a potent soluble 
poison which gives rise to the symptoms of tetanus, although it is only 
slightly aggressive, being almost unable to multiply in the tissues. The 
anthrax bacillus, on the other hand, is highly aggressive, owing to the fact 
that it usually multiplies to such an extent that it can be found in each 
drop of blood and in every organ of an infected animal; nevertheless it is 
but slightly toxic, the animals frequently showing few or no symptoms 
until shortly before death. The toxicity of a micro-organism should, there- 
fore, be regarded separate from its aggressiveness, although in many in- 
fections both factors are so intimately concerned that the term “virulence” 
may be used to express the degree of pathogenicity or the total disease- 
producing power. 
The virulence of a micro-organism is more or less specific, i. e., the toxin 
produced by one species is different from that produced by another in the VIRULENCE 
69 
kind of disease produced and the species of animal infected. Some toxins 
are active for certain animals only and not for others. Micro-organisms of 
one group may possess general and common pathogenic properties differing 
only in degree; those of different morphologic, and cultural characters may 
possess totally different powers. 
The virulence of a given species is subject to great variation. A few 
bacteria almost constantly retain their virulence, even when kept for years 
under artificial conditions; as an example may be mentioned the diphtheria 
bacillus; others quickly lose their virulence as soon as they are grown artifi- 
cially, as, e. g., the influenza bacillus; in others—and probably the larger 
class—the virulence may be raised or lowered according to the experimental 
manipulations to which they may be subjected. Variations may also be 
observed among members of the same group of micro-organisms, and even 
among individual micro-organisms of the same strain. 
Decrease of virulence of a micro-organism may be brought about artifici- 
ally by repeated growth in or upon culture-media, especially when transfers 
to fresh media are made at prolonged intervals. This decrease probably 
depends upon an actual decrease in virulence, and particularly upon the 
selection, in artificial growth, of the less virulent or vegetative forms which 
grow actively and soon exceed in number their more pathogenic fellows. 
Each time the culture is transplanted more of the vegetative and fewer of 
the pathogenic micro-organisms are carried over, until finally the patho- 
genic bacteria are entirely eliminated, or their virulence totally destroyed, 
and the entire culture is composed only of vegetative or harmless forms of 
bacteria. 
Various other agencies lead to artificial lessening of virulence, such as 
exposure, for short periods of time, to a temperature just under the thermal 
death-point; exposure to sunlight; exposure to small quantities of anti- 
septic or germicidal substances; the action of desiccation; subjection to in- 
creased atmospheric pressure, etc., these methods being commonly em- 
ployed in the preparations of vaccines to be used for purposes of active 
immunization. 
The passage of a micro-organism or virus through animals usually in- 
creases its virulence, but may modify or attenuate it, as in the case of the 
passage of smallpox virus through the calf, when it loses forever its power of 
producing smallpox. 
Increase in virulence can best be secured by passing the micro-organism 
through animals. It is practically impossible, by any means, to make a 
known non-virulent micro-organism virulent, although it is comparatively 
easy to increase the virulence of a culture that has become well-nigh non- 
virulent on account of prolonged artificial cultivation. This fact is worthy 
of emphasis, and is well illustrated by the large amount of work that has 
been done in fruitless attempts to render non-virulent, diphtheria-like 
bacilli virulent by passage through various animals or growth on special 
culture-media. 
In cases where the virulence is slight or absent, experimental manipula- 
tions of the culture are directed toward gradual immunization of the micro- 
organisms to the defensive mechanism of the body of the animal for which 
the organism is to be made virulent. This is well explained according to 
the hypothesis of Welch, and will be referred to again in the latter part 
of this chapter. A number of methods are made use of for this purpose: 
(a) Passage through animals, which enables the micro-organisms gradu- 
ally to immunize themselves or adopt certain morphologic and biologic 
changes enabling them best to resist the defensive forces of the host. Since 70 
INFECTION 
these defensive forces vary with different animals, and, indeed, with the 
various organs of the same animals, it is usual to find that virulence raised 
by animal passage affects only the animal or the particular organ of a cer- 
tain animal, and not all animals in general. Thus, in general, the passage 
of bacteria through rabbits increases their virulence for rabbits and not for 
mice, dogs, pigeons, etc.; passage through mice may increase their virulence 
for mice, but not for rabbits, guinea-pigs, etc. 
(b) The use of collodion sacs for increasing virulence has been advocated, 
especially by French investigators. When micro-organisms are inclosed in 
a collodion capsule of the proper thickness and placed within the abdominal 
cavity of a suitable animal, the slightly modified body juices are able to 
transfuse through the sac, impeding the development of such micro-organ- 
isms as are unable to immunize themselves or withstand the injurious in- 
fluences. In this manner a race of virulent bacteria are artificially selected 
which can endure the defensive agencies of those juices with which they 
have come in contact. 
(c) The addition of animal fluids to the culture-medium may enable the 
bacteriologist to maintain or even to increase the virulence of a micro- 
organism according to the principles of artificial selection. The fluid, either 
a serum or whole blood, is secured in a sterile manner and added to the 
medium in a raw or unheated condition. In this manner the micro-organisms 
are exposed to some of the defensive agencies contained in the juices under 
these conditions, and this tends to destroy the less resistant bacteria, en- 
courage the more resistant, and at least maintain, for a longer or a shorter 
time, the virulence of a culture freshly isolated from a lesion or cultivated 
by animal passage. 
THE AVENUE OF INFECTION AND TISSUE SUSCEPTIBILITY 
Successful infection of the body by certain parasites can be accomplished 
only when invasion takes place through appropriate avenues. Thus typhoid, 
cholera, and dysentery infection seems to take place through the gastro- 
intestinal tract, and doubtfully by inhalation, and not at all through the 
skin or urogenital system; gonococci usually enter the body through the 
genital organs or the eye, and not through the respiratory apparatus or 
through the skin. The route of infection is less important with micro- 
organisms characterized by great aggressiveness and producing general, 
rather than local, infections. For example, in most animals anthrax is a 
general bacteremia, regardless of the route of invasion; plague rapidly be- 
comes a bacteremia, whether the bacilli are inhaled, rubbed into the skin, 
or reach the lymphatics through superficial abrasions; similarly, local staph- 
ylococcus and streptococcus infection may become general, regardless of the 
route of invasion or the location of the local lesion. 
The avenue of invasion is also of importance in determining the form, 
nature, and virulence of an infection. Thus virulent pneumococci lodging 
in the pharynx may produce a pseudomembranous angina; in the eye, a 
severe conjunctivitis; and in the lungs, a pneumonia. When tubercle bacilli 
gain admission through the skin, they may produce lupus, or a low-grade 
inflammatory disease rarely terminating fatally. When inhaled, they may 
produce tuberculosis of the lungs; in the throat they may reach the tonsils 
and later the local lymphatic glands, etc. When swallowed, they may pro- 
duce ulceration of the intestines, or pass through the intestinal walls and 
involve the mesenteric glands, and later the lungs or other organs. 
Just as general susceptibility of the host renders infection more likely THE AVENUE OF INFECTION AND TISSUE SUSCEPTIBILITY 
71 
to occur, so local susceptibility may be induced by injury and fundamental 
disorders. These changes may not only furnish pabulum for the invading 
bacteria, but more especially reduce the local resistance of the body defenses. 
Even more important, however, is the predisposition of some pathogenic 
micro-organisms to attack certain tissues or organs, and the fact that these 
tissues are particularly weak in defensive power so that the bacteria natur- 
ally lodge where conditions are most favorable for their growth. 
Selective Tissue Affinity.—While the primary focus of infection is de- 
termined largely by the route of invasion, the selective affinity of micro- 
organisms or their toxins for certain tissues and the inherent tissue sus- 
ceptibility to the toxins or bacteria are best in evidence in the location of 
secondary foci or localization of the infection in general bacteremias. Thus, 
the seat of the principal local lesions in pneumonia is the lungs, and in 
typhoid fever the lymphoid tissues, especially that of the spleen, and Peyer’s 
patches in the intestine. It is true that mechanical factors may aid in this 
selection, as, e. g., the occlusion by emboli of micro-organisms caught in 
the capillaries of organs; but, in general, we must conclude that either 
(1) micro-organisms tend to be destroyed in every tissue or organ except 
those that are poor in defensive forces and are susceptible, or (2) that micro- 
organisms or their products circulate passively through a tissue and do not 
lodge because they possess no affinity for these cells. In many infections 
both processes are probably operative, and at least we are led to the very 
important conclusion, laid down by Adami, that “in infections the body is 
never involved as a whole. Coincidentally with the growth of the specific 
germs in individual organs there tends to be a reaction to, and destruction 
of, the same in other parts.” 
Nickols and Hough1 and Reasoner2 have isolated strains of Treponema 
pallidum from the nervous tissues that appeared to have selective affinities 
for the cornea, choroid, and retina of the eyes of rabbits; Noguchi3 has 
noticed that various types produced lesions in rabbits of certain distinct 
characters and with considerable constancy. Further and similar studies 
may show that various strains of Treponema pallidum may possess selec- 
tive tissue affinities and thereby explain the early development of lesions 
in the central nervous, cardiovascular, and cutaneous systems of different 
persons. In fact, Levaditi and Marie4 has recently reported experiments 
showing the existence of two distinct strains of Treponema pallidum, one 
producing a large primary and well-defined cutaneous lesion, designated as the 
dermotropic strain, and a second producing a small and ill-defined primary 
and Secondary lesion, but possessing marked affinity for the tissues of the cen- 
tral nervous system, designated as the neurotropic strain. According to the 
investigations of Rosenow5 various bacteria, and particularly streptococci, 
exhibit extreme degrees of tissue affinity and produce various constant and 
distinct lesions in rabbits after inoculation by various routes. This subject 
is further discussed in the section devoted to Focal Infection. 
The numeric relationship of parasites to infection is very important, 
and the number alone may determine whether or not it shall occur. Usually 
the normal defensive factors of the body are sufficient to overwhelm one or 
a few bacteria unless they are especially virulent. When an intercurrent 
1 Jour. Araer. Med. Assoc., 1913, 60, 108. 
2 Jour. Amer. Med. Assoc., 1916, 66, 1917; ibid., 1916, 67, 1799. 
3 Jour. Lab. and Clin. Med., 1917, 2, 472. 
4 Bull, de l’lnst. Pasteur, Nov., 1919, 741. 
5 Jour. Infect. Dis., 1915, 16, 240; ibid., 1915, 16, 367; ibid., 1915, 17, 219; ibid., 1915, 
17, 403; ibid., 1916, 16, 501; ibid., 1916, 18, 383; ibid., 1916, 19, 333; Jour. Amer. Med. Assoc., 
1914, 63, 1835; ibid., 1915, 64, 1968. 72 
INFECTION 
or chronic disease, malnutrition, or injury renders the host more susceptible 
than normal, fewer bacteria than would otherwise be required may suc- 
cessfully infect the body. Also with those microparasites with well-marked 
aggressiveness, such as the anthrax bacillus, a few may be sufficient, if they 
reach the circulating fluids, to produce infection. Thus, Webb, Williams, 
and Barbor1 have found that one anthrax bacillus was sufficient to infect a 
white mouse, and as few as 20 tubercle bacilli were sufficient in one instance 
to infect a guinea-pig. 
Likewise Kolmer, Schamberg, and Raiziss2 have found in experimental 
trypanosomiasis that the infection of wThite rats by intraperitoneal injec- 
tion with varying numbers of trypanosomes, counted after the method of 
Kolmer,3 greatly modifies the time of appearance of trypanosomes in the 
peripheral blood and the duration of life, and has an important bearing 
upon studies in the chemotherapy of experimental trypanosomiasis. 
Park has directed attention to the fact that when bacteria are trans- 
planted from culture to culture, under supposedly favorable conditions, 
many of them die; it is highly probable that when they are transplanted 
to an environment that is likely to be unfavorable, as are the body tissues 
with various defensive mechanisms, many more must die. This is an im- 
portant point to bear in mind in attempting to correlate experimental re- 
sults with the natural cause of an infectious disease. In the laboratory we 
reproduce disease experimentally by the immediate injection of millions 
of bacteria, whereas in nature there is rarely any such immediate over- 
whelming of the tissues. For example, pneumonia may be produced experi- 
mentally in dogs by the injection of a large number of virulent pneumococci 
directly and at once into the bronchi, yielding a positive result with a micro- 
organism which, under natural conditions and in smaller numbers, would 
be relatively innocuous for the animal under observation. 
d’Herelle’s Intestinal Bacteriophage in Relation to Infection.—Accord- 
ing to d’Herelle4 there may be an additional defensive agency which micro- 
parasites must overcome before infection can occur, and particularly infec- 
tions of the gastro-intestinal tract. He has found in filtrates of broth 
cultures of the feces of some normal human beings and lower animals an 
agent capable of attacking and dissolving living bacteria and particularly 
dysentery bacilli, which is regarded as a normal agency of defense. In the 
course of bacillary dysentery, typhoid and paratyphoid fevers, cholera, and 
other bacterial infections of man and the lower animals this bacteriolytic 
agent has been found greatly increased, and is regarded by d’Herelle and 
others as playing an important role in recovery from disease. 
This substance is regarded by d’Herelle as a living organism of ultra- 
microscopic size which lives only at the expense of living bacteria, entering 
them and secreting a diastatic ferment which kills the bacterium and brings 
about its dissolution by a digestive process; for it he has proposed the name 
bacteriophagum intestinale. Others regard this “bacteriophage” as an enzyme 
produced by bacteria themselves rather than a living ultramicroscopic 
parasite. The varied theories concerning its nature and a more complete 
discussion of the subject is given in the chapter on Ferments and Anti- 
ferments. Certainly the presence of an agent of this kind in filtrates of 
cultures of the feces of some normal healthy human beings and the lowrer 
1 Transactions Sixth International Congress on Tuberculosis, 1908, p. 194. 
2 Jour. Infect. Dis., 1917, 20, 10; ibid., 1917, 20, 35. 
3 Jour. Tnfect. Dis., 1915, 16, 311; ibid., 1915, 17, 79. 
4 Le bacteriophage—son role dans 1’immunite, Masson et Cie, Paris, 1921. This book 
has recently been translated into English by Smith and published by Williams & Wilkins 
Company, Baltimore, Md. GENERAL SUSCEPTIBILITY IN RELATION TO INFECTION 
73 
animals, and especially in human beings with bacillary dysentery, may be 
regarded as proved, although the nature of the bacteriolytic agent is not 
definitely known. d’Herelle believes that there is but one species of “bac- 
teriophage,” but an infinite number of strains, each possessing the power of 
attacking a certain number of bacteria. While a normal inhabitant of the 
intestine of all living beings, d’Herelle believes that it may be taken into the 
circulation and exert its protective and curative action at any point in the 
body. 
It is also claimed by d’Herelle that the virulence of the bacteriophage 
for bacteria may be increased; certainly it may be enhanced quantitatively 
as is true of enzymes in general under proper conditions for production. 
Bacteria, on the other hand, are regarded as capable of acquiring a resistance 
and protecting themselves against this agent, accompanied by changes in 
morphology. The “bacilli take a coccoid aspect and become surrounded 
by a capsule. They become inagglutinable. They resist phagocytosis. 
They are endowed with a very great vitality and a very high virulence. 
Loss in resistance is accompanied by a return to normal form and properties.” 
In relation to infection it may be stated that the work of d’Herelle has 
shown that in the test-tube at least filtrates of cultures of the feces may 
digest certain bacteria and that filtrates of these digested bacteria in turn 
will digest cultures of various bacteria in succession. The lytic agent is 
filterable; whether it is a living parasite or an enzyme is not definitely known. 
Whether it exerts an appreciable or important role in natural resistance to 
bacterial infection in general and intestinal infections in particular cannot 
be stated. Probably it exerts some protective activity and when present 
must be overcome in cases of infection occurring by way of the gastro- 
intestinal tract; it may also exert a more important role in recovery from 
intestinal infections and particularly dysentery, although, according to 
Davison,1 the oral and subcutaneous administration of the filtrates have not 
proved curative in the treatment of bacillary dysentery of children. 
GENERAL SUSCEPTIBILITY IN RELATION TO INFECTION 
Under normal conditions the body cells of a host will invariably offer 
some resistance to invasion and infection by pathogenic microparasites. 
When, however, any condition that depresses or diminishes general physio- 
logic activity and vitality exists, the host may be unable to master these 
defensive forces, and accordingly becomes predisposed or more susceptible to 
infection. 
Predisposition may be inherited or acquired. 
Inherited predisposition may be: (a) Specific, or species susceptibility, 
as, e. g., dogs to distemper; cattle to contagious pleuropneumonia; hogs 
to hog cholera; man to gonorrhea; chancroids, acute exanthemata, typhoid 
fever, etc. (b) Racial, as Eskimos to measles and syphilis, ordinary sheep 
to anthrax, whereas Algerian sheep are immune, etc. Racial susceptibility 
is frequently but a lack of acquired immunity; for instance, measles, syphilis, 
gonorrhea, and other diseases brought by settlers to foreign peoples among 
whom these diseases were previously unknown, find them peculiarly sus- 
ceptible and the diseases unusually virulent, (c) Familial, i. e., members 
of a family may, through generations, be unusually susceptible to scarlet 
fever, tuberculosis, rheumatism, rheumatoid arthritis, metabolic disturb- 
ances, etc. (d) Individual predisposition, which depends principally upon 
sex, age, and peculiar tissue susceptibility. Thus, infants are especially 
1 Arner. Jour. Dis. Child., 1922, 23, 531. 74 
INFECTION 
prone to contract certain infections on account of the immature develop- 
ment of the body cells, and this susceptibility to infection is further in- 
fluenced by acquired factors, chiefly malnutrition. On the other hand, 
very young children enjoy an immunity to several infections, such as typhoid 
fever, scarlet fever, and even diphtheria, probably due, as Abbott has 
suggested, to the fact that pathogenic substances that may set up molecular 
and destructive disturbances in the poorly developed cell have but little 
effect upon the more inert protoplasm of the immature cell, and that if 
certain bacteria gain admission to the tissues the cells may destroy them, 
their toxins not combining with the molecular side-chains, and, as a conse- 
quence, not injuring or interfering with the cell functions. 
Acquired susceptibility bears a more important relation to infection, 
and may be due to various factors, most of which lead to a state of reduced 
vitality, normal physiologic processes being impaired to a greater or less 
degree. 
(a) Overwork or overstrain leads to general or local predisposition to 
disease. Those engaged in hard labor, mental or physical, which involves 
late hours and inadequate periods of rest and recreation, frequently asso- 
ciated with inadequate nutrition and foul air, are likely to succumb to 
tuberculosis, typhoid fever, pneumonia, etc. 
The influence of overstrain on acute infections has been shown experi- 
mentally by Charrin and Roger,1 who found that white rats naturally 
immune to anthrax became quite susceptible after being compelled to turn 
a revolving wheel until exhausted before they were inoculated; similarly 
of 4 guinea-pigs who were placed in a cage so constructed that they were 
forced to keep moving for one or two days 3 died in from two to nine days 
after the experiment. Smears and cultures made from the livers, spleens, 
and blood gave positive results. 
(ib) Previous infection with the same or another infectious disease may 
predispose the individual to renewed infection. Thus, some infections, 
such as erysipelas, furunculosis, acute rheumatism, pneumonia, and in- 
fluenza, not only fail to leave the body-cells immune, but actually pre- 
dispose to second attacks. Whether the micro-organisms of these diseases 
are not all destroyed, but are retained in the system and become active 
when the general vitality is lowered, or whether a new infection occurs, is 
not definitely known, and probably either may occur. 
One attack of an infectious disease may weaken the tissues and render 
them susceptible to an infection of a different nature. Thus, the acute 
exanthemata may follow one another, and tuberculosis may supervene 
upon any of them. 
(c) Malnutrition exerts some effect on the resistance to infection. Thus, 
the terrible epidemics of plague, cholera, typhus fever, and typhoid fever 
which have followed in the wake of famines in Europe and Asia during the 
past centuries are examples of the influence of malnutrition as a factor in 
predisposing to disease. The tendency of marasmatic infants to develop 
enterocolitis, thrush, bronchopneumonia, and other infections, and of 
scorbutics to local infections of the mouth, illustrates the influence of in- 
sufficient food in decreasing the resistance to disease. Here may also be 
included local malnutrition, such as loss of nerve or blood supply, predis- 
posing to local infection, especially with pyogenic micro-organisms. 
(d) Diet produces some variation in the resisting powers to infection. 
For example, the ordinary wild rat is said not to be susceptible to anthrax 
unless it is fed for a week or more on coarse dry food, when it becomes 
1 Compt. rend. Soc. de Biol, de Paris, January 24, 1890. DEFENSIVE MECHANISM OF THE MICRO-ORGANISM 
75 
susceptible. Here, of course, malnutrition may come in intimate relation- 
ship with diet, as an inefficient diet may greatly lower the general resist- 
ance. The influence of diet is particularly noticeable from the fact that 
the diseases of carnivorous animals are not the same as those that affect 
herbivorous animals, and that each class is frequently immune to some of 
the diseases that attack the other. 
(e) Intoxications of various kinds predispose to infections. Thus, it is 
a common clinical observation that excessive indulgence in alcoholic bever- 
ages predisposes to infections, notably pneumonia. Abbott1 has demon- 
strated experimentally that the daily administration to rabbits, of 5 to 
10 c.c. of alcohol introduced into the stomach by a tube, renders these 
animals more susceptible to infection with Streptococcus pyogenes and 
Bacillus coli. Wagner, Leo, and Platania have also found animals that 
under the influence of chloral, phloridzin, alcohol, and curare are more 
susceptible to infection. 
(/) Exposure to cold and wet frequently lowers the resistance of man 
and other warm-blooded animals to infection. The influence of these fac- 
tors, well illustrated in the etiology of “colds” and pneumonia, is not with- 
out experimental foundation. Thus Pasteur found that fowls, which are 
naturally immune to anthrax, are readily infected if they are inoculated 
after their body temperature has been reduced by a cold bath. Conversely, 
Gibier2 has shown that frogs, wffiich are also naturally immune to anthrax, 
are readily infected if their temperature is previously elevated and main- 
tained at 37° C. 
(g) Trauma and morbid conditions in general may predispose to infec- 
tion. Thus injuries reduce the local resistance and facilitate local infec- 
tions that vary with the severity and extent of the trauma. The increased 
susceptibility of injured joints and pneumonic lungs to tuberculosis, the 
frequent and oftentimes extensive streptococcus infection accompanying 
scarlet fever and smallpox, the increased susceptibility of diabetics to 
furunculosis and local gangrenous lesions of the skin—all show the increased 
susceptibility of individuals already injured or diseased to infection. 
THE DEFENSIVE MECHANISM OF THE MICRO-ORGANISM IN RELATION 
TO INFECTION 
After invasion has occurred, the question of whether or not the microf 
organism can overcome the defensive forces of the host and prove patho- 
genic may depend to some extent upon the peculiar defensive factors o- 
the invading microparasite against the offensive mechanism of the host, 
aside from their toxins or other distinctly offensive forces. 
Morphologic and Physiologic Changes of the Micro-organisms.—For 
example, capsule formation or thickening of the ectoplasm of certain bac- 
teria is evidence of their increased powers of resistance against the oppos- 
ing forces of the host. The capsule may be quickly lost when the micro- 
organism is cultivated on artificial media, and its virulence be correspond- 
ingly lowered, but by repeated animal inoculations a race of capsulated 
organisms with increased virulence is produced, explaining in a way the 
mechanism of animal passage in raising the virulence of a given organism. 
This, however, is not invariable, and indeed, may act in a contrary manner, 
as the passage of smallpox virus through heifers attenuates and modifies 
instead of increasing its virulence. 
Aggressins.—The micro-organism may actively secrete a material that 
1 Jour. Exper. Med., 1896, 1, 447. 
2 Compt. rend. Acad, de Sci. de Paris, 1882, xcix, 1605. 76 
INFECTION 
overwhelms the defensive forces of the host. This phase of the subject 
has been studied exclusively by Bail, who sought to prove that the ques- 
tion of pathogenicity of a micro-organism is dependent upon its ability to 
secrete substances that are able to paralyze the protective forces of the 
host, especially the leukocytes. These substances .are called “aggressins,” 
and they were distinguished by the fact that they were formed by living 
bacteria and only in the living body. In support of this theory Bail was 
able to show that substances are present in the exudates of fatal infections, 
which, when injected in small quantities into another animal with sub- 
lethal doses of the micro-organism, would cause a rapidly fatal infection. 
Later Wassermann and Citron showed that “artificial aggressins” could be 
prepared by autolyzing bacteria in water or serum. While the subject of 
aggressins is still unsettled, there is strong evidence to show that they may 
be the endotoxins liberated by the breaking up of the micro-organism. 
The well-known statement of Metchnikoff, that a particular virulent 
micro-organism is not so readily taken up by leukocytes as is an avirulent 
strain, may be explained by the fact that the micro-organism, in its virulent 
state, secretes substances that repel the phagocytes, neutralize the opsonins, 
or form actual leukocytic toxins. This action may be due to liberated 
endotoxins or, as Bail claims, to specific secretory substances of the bac- 
terium—the aggressins—specifically formed and liberated by the micro- 
organism for protection against the host. It is probable that the lysin 
produced by the “bacteriophage” of d’Herelle is a substance of this kind. 
As previously stated, d’Herelle regards his bacteriophage as an organism of 
ultramicroscopic size entering bacteria and secreting a diastatic ferment 
capable of preventing phagocytosis and producing bacteriolysis. The sub- 
ject is discussed at greater length in the chapter on Ferments and Anti- 
ferments. 
Hypothesis of Welch.—Not entirely foreign to this subject is the very 
interesting hypothesis of Welch. A bacterium may not only produce sub- 
stances directly inimical to the defensive forces of the host, but it may 
actually immunize itself against these defensive powers. “Looked at from 
the point of view of the bacterium, as well as from that of the animal host, 
according to the hypothesis advanced, the struggle between the bacteria 
and the body-cells in infections may be concerned as an immunizing con- 
test in which each participant is stimulated by its opponent to the produc- 
tion of cytotoxins hostile to each other, and thereby endeavors to make 
itself immune against its antagonist.” 
It is well known that, when freshly isolated from a patient having ty- 
phoid fever, the typhoid bacillus resists agglutination, whereas it becomes 
easily agglutinable after a period of artificial cultivation. It may be as- 
sumed that, when active, the bacillus as an infecting agent gradually be- 
came more resistant against the agglutinating properties of the patient’s 
serum, and that, when grown on artificial media, it loses this resistance 
by being removed from the stimulating influence of the infected body. 
This hypothesis, however, would go a step further in assuming the 
possibility of the receptors of the invading bacteria anchoring certain con- 
stituents of our body-fluids, and being stimulated to the production of 
various cytotoxins, which attack the leukocytes, erythrocytes, nerve-cells, 
liver, kidney, etc. In other words, each bacterium may be conceived as 
being composed of a central atom group with numerous side chains, just 
as Ehrlich conceived the hypothetic structure of body cells, and that these 
side chains, primarily present for the purpose of anchoring food material, 
may likewise anchor various pathogenic animal substances, with the pro- MIXED INFECTION 
77 
duction of substances acting as antibodies to the opposing forces of the 
host. Welch assumed that these bodies were of the nature of amboceptors, 
which may become complemented by bacterial complement or by endo- 
complements of the tissue-cells; this is of secondary importance, and there 
is no reason why they may not be of different structure, and similar to all 
three orders of antibodies produced by body cells according to Ehrlich’s 
side-chain theory of immunity. Of further interest in this connection are 
the investigations of d’Herelle referred to above, who described morphologic 
changes in bacteria offering resistance to destruction by his “intestinal bac- 
teriophage,” an organism of ultramicroscopic size or an enzyme produced 
by bacteria and preying upon pathogenic and non-pathogenic bacteria them- 
selves. These investigations have emphasized anew the fact that bacteria 
possess intricate mechanisms of defense against destruction by various 
chemical and physical agents. As stated by d’Herelle, “although the bac- 
teriophage is capable of acquiring a virulence for the bacterium, the bac- 
terium on its side is capable of acquiring a resistance to the bacteriophage. 
The virulence of the one and the resistance of the other are not fixed, but are 
essentially variables, being enhanced or attenuated according to the in- 
herited properties of each of the two germs, and according to the circum- 
stances of the moment which favor the one or the other of the two antag- 
onists.” 
This hypothesis may also possibly explain certain instances of so-called 
species and organ virulence, whereby the virulence of an organism artificially 
increased by repeated passage through animals of the same species, does 
not manifest this increased virulence for animals of different species. If, 
for example, the virulence of the chicken cholera bacillus is increased by 
repeated passage through the chicken, the increase of virulence affects this 
animal, but does not affect the guinea-pig. Certain organs may likewise 
be subject to a similar selective virulence if the increase in virulence has 
been induced by the specific intervention of those organs and this selec- 
tive virulence shows itself, irrespective of the manner in which the infection 
was produced. 
That virulence of this order is playing an important role in the proc- 
esses of infection is a theory supported by the discovery that different 
strains of the same species of bacteria are found to produce characteristic 
lesions, and whule this affinity for a certain organ may be natural and in- 
herent, there can be no doubt that it may also be experimentally induced 
and acquired. For example, according to Rosenow, a certain strain of 
streptococcus will produce arthritis; another, endocarditis; another, gastric 
ulcer, etc. 
This remarkable species and organ specificity may be due to the fact 
that the bacteria of a particular culture have been immunized against 
defensive forces of a particular animal host or a certain organ of the host, 
so that, when introduced, they thrive as a result of their special and ac- 
quired offensive forces. On the other hand, the specificity may be due to 
the fact that the bacteria have been accustomed to a certain nutriment 
furnished by a particular species or organ, and that they cannot thrive 
unless they receive this special nutriment, and, as a result, the species or 
organ fulfilling this requirement will become the special seat of infection 
(Simon). 
MIXED INFECTION 
Several different micro-organisms may produce infection at the same 
time, or one may follow the other or others and produce secondary infec- 78 
INFECTION 
tion. A disease, as amebic dysentery, may be due to the combined activity 
of an animal parasite and a bacterium or several varieties of bacteria. The 
combined effects, upon the tissues of the host, of the products and action 
of two or more varieties of pathogenic bacteria, and also of the influence 
of these different forms on each other, are of great importance in the pro- 
duction of disease. The metabolic products of one bacteria may neutralize 
or accelerate the action of an associated species, or combine to form a new 
substance entirely different from its antecedents. 
Thus, pyogenic cocci affect anthrax bacilli in an injurious manner; on 
the other hand, aerobic bacteria accelerate or make possible the growth of 
anaerobes by absorbing uncombined oxygen. Tetanus bacilli will not 
grow outside of the body in the presence of oxygen unless aerobic bacteria 
are associated with them; not infrequently tetanus bacilli and their spores 
would not develop in wounds were it not for the presence of the aerobic 
bacteria introduced with them; this factor is of much importance, espe- 
cially in tetanus produced by cowpox vaccine, where, through careless 
treatment of the lesion, both tetanus bacilli and pyogenic cocci are admitted 
to the wound. 
Again, it may be found that one micro-organism increases the viru- 
lence of another; thus, the scarlet-fever virus is favorable to the develop- 
ment of streptococci. 
Generally all infections of mucous membranes are mixed infections. 
Numerous bacteria are present upon the mucosa of the air-passages and 
gastro-intestinal tract; these are usually harmless unless the resistance 
of the host is lowered in some manner, in which case not only one but several 
varieties of these bacteria invade the tissues and cause infection. When 
one pathogenic micro-organism, such as the typhoid bacillus, has caused 
the primary infection, because of the local and general conditions of lowered 
vitality of the tissues, these otherwise saprophytic bacilli tend to intensify 
the infection. Blood infections, on the other hand, are usually due to one 
form of bacteria, and even when two or more varieties are introduced, 
only one, as a rule, is capable of surviving and developing. The products 
of certain bacteria, on the other hand, may immunize the host against in- 
fection with other bacteria, for, as shown by Pasteur, attenuated chicken- 
cholera cultures may produce immunity against anthrax. In the intestine 
harmless varieties of bacteria may be made to crowd out more dangerous 
ones; this is exemplified by the ingestion of soured milk which contains 
lactic acid bacteria, as advocated by Metchnikoff. 
FOCAL INFECTION 
In a general manner bacterial infections may be divided into two main 
groups. In one group the effects of infection are soon manifest by symp- 
toms of disease and this group includes the acute infectious diseases and a 
host of local infections as those following trauma, surgical manipulations, 
or occurring without demonstrable predisposing factors other than ex- 
posure. 
Focal Infection.—In a second group micro-organisms gain access to the 
tissues at a certain place and produce a localized and confined infection 
without symptoms, or such slight symptoms as are commonly disregarded; 
from these primary foci or localized areas of infection, however, the micro- 
organisms or their products may gain access to the lymph or blood streams 
and produce more serious secondary foci or metastatic infections in neigh- 
boring tissues or distant and unrelated organs. This is called focal infec- FOCAL INFECTION 
79 
tion, the primary focus being defined by Billings1 as “circumscribed area of 
tissue infected with pathogenic micro-organisms,” and the metastatic 
lesions as secondary foci. 
The subject has deservedly attracted considerable attention in recent 
years in both the medical and dental professions; unquestionably focal in- 
fection is responsible for many infections and diseased states of hitherto 
indefinite or unknown etiology. The detection of the primary focus or 
foci and treatment of primary and secondary foci, are subjects of consider- 
able importance and worthy of closer clinical and laboratory investigation. 
Primary Foci.—Billings states that “primary foci of infection may be 
located anywdiere in the body. Infection of the teeth and jaws, with the 
especial development of pyorrhea dentalis and alveolar abscess, infection 
of the faucial and nasopharyngeal tonsils and of the mastoid, the maxil- 
lary and other accessory sinuses are the most common forms of focal in- 
fection. Submucous and subcutaneous abscesses including the finger and 
toe nails are occasional foci. Chronic infection of the gastro-intestinal 
ulcers and intestinal stasis due to morbid anatomic conditions; chronic 
infection of the genito-urinary tract, including metritis, salpingitis, vesicu- 
litis, seminalis, prostatitis, cystitis, and pyelitis, are not uncommon forms. 
Infected lymph-nodes, which are secondary to the primary foci named, 
become additional depots of local infection. The secondary lymph-node 
infection may persist after the etiologic, distal, primary focus has been 
removed or has spontaneously disappeared. Other secondary foci may 
appear in various tissues as a part of the general or local disease which 
results from a primary focus. The tissues so infected may constitute new 
foci, which in part explains the chronicity of many local and general in- 
fections.” Primary foci are especially apt to develop along the upper air- 
passages, bacteria from the mouth and nose gaining access to crypts in the 
tonsils and adenoid tissue, to the apices of teeth, through necrotic dentin and 
pulp canals, and to the nasal accessory sinuses. In general terms the tonsils 
and adenoid tissue should first be suspected in searching for primary foci 
in children and adolescents; the gums, apices of teeth, and bronchi in adults 
and particularly those over forty years of age. Of course, not all individuals 
with abscesses at the roots of one or more teeth detected by roentgenologic 
ray study or with severe gingivitis (oral sepsis) show the effects of focal 
infection; there are many factors governing focal infection which are not 
as yet understood and, unquestionably, natural immunity and virulence of 
the micro-organisms are important factors influencing the development of 
secondary foci. 
Etiology.—As is to be expected a variety of micro-organisms may be 
responsible for the primary foci, including streptococci, pneumococci, staph- 
ylococci, and micrococcus catarrhalis. Of all bacteria so far identified, how- 
ever, streptococci are certainly the most important, by reason of the re- 
searches of Rosenow and others wrho have found streptococci in primary 
and secondary foci more frequently than other micro-organisms. Of con- 
siderable importance in this connection is the possibility of streptococci 
growing in primary foci acquiring specific pathogenicity for certain tissues 
in the nature of tissue tropism or elective tissue affinity. Rosenow,2 using 
special methods, has isolated streptococci from the lesions in acute rheu- 
matic arthritis, cholecystitis, appendicitis, gastric and duodenal ulcer, 
herpes zoster, and erythema nodosum, which were low in virulence and 
similar to each other in morphologic and cultural characteristics, but when 
1 The Lane Medical Lectures on Focal Infection, D. Appleton & Co., 1916. 
2 Jour. Amer. Med. Assoc., 1914, 63, 903; ibid., 1835; ibid., 1915, 65, 1687. 80 
INFECTION 
injected into animals, each strain tended to localize electively in the tissue 
from which it was isolated. In most instances some joint involvement 
also occurred in the experimental infections, indicating that the vascular 
anatomy of the joints is an important factor in the production of infective 
arthritis. It is probable that the micro-organisms in the foci may be affected 
by changing biochemical properties of the tissues and become accustomed 
to a special chemical environment which, in addition to trauma and viru- 
lence, determines the tendency of a micro-organism to show elective tissue 
tropism. This subject of selective tissue affinity has been previously dis- 
cussed on page 71; Rosenow’s observations require more general con- 
firmation, but are of great interest in connection with this very important 
subject of focal infection. The investigations of Davis1 and Means2 tend 
to show that evidences of apparent selective tissue affinity are to be ex- 
plained rather on the basis of anatomic structure, susceptibility, and trauma 
of the tissues. 
As previously stated, bacteria or their toxins from a primary focus, 
are believed to produce secondary foci of infection by direct extension or 
by distribution by the lymphatic and blood streams. Cultural studies 
indicate that the bacterium rather than its toxins, are actually transported 
and responsible for the secondary foci. 
Secondary Foci.—Many different tissues and organs may become the 
seat of secondary foci; probably the joints are involved more frequently 
than any other tissue due, in part to the vascular anatomy of these tissues 
favoring the occurrence of bacterial embolism. Iritis, and particularly rheu- 
matic iritis, may possibly be due to focal infection; Rosenow3 and Irons, 
Brown and Nadler4 have successfully produced iritis experimentally in 
rabbits by injecting streptococci intravenously, and it is probable that 
many cases of iritis regarded as idiopathic are due to bacterial embolism 
from some primary focus of infection. Certain affections of the nervous 
system have been ascribed to focal infection by Rosenow5 and Hall.6 Jud- 
son Daland7 and Lewellys Barker8 believe that any of the following may 
be caused by oral sepsis and focal infection: infectious arthritis, hyper- 
trophic osteo-arthritis, local osteomyelitis, myositis, acute infections, endo- 
carditis, secondary anemias, multiple neuritis, and various lesions in the 
gastro-intestinal, urogenital and endocrinic systems. Ravitch9 believes 
that certain dermatoses may be ascribed to focal infection; Grulee and 
Gaarde10 have described cases of acute hemorrhagic nephritis occurring in 
children probably secondary to a preceding acute tonsillitis. These refer- 
ences serve to show the importance of focal infection in relation to the 
etiology of a wide variety of diseases demanding most careful study and 
especially from the standpoints of diagnosis and treatment. 
INFECTION WITH FUNGI 
A large number of parasitic fungi are known to produce disease in man 
and the lower animals, the subject of mycology being almost as extensive 
1 Jour. Amer. Med. Assoc., 1912, 58, 1283. 
2 Archiv. Int. Med., 1918, 22, 617. 
3 Jour. Infect. Dis., 1915, 17, 403. 
4 Jour. Infect. Dis., 1916, 18, 315. 
6 Jour. Amer. Med., Assoc., 1916, 67, 662. 
6 Jour. Amer. Med. Assoc., 1917, 69, 689. 
7 Dental Cosmos, May, 1916; Canada Lancet, August, 1917 
8 Jour. Dental Research, 1920, 2, 43. 
9 Jour. Amer. Med. Assoc., 1916, 67, 430. 
10 Jour. Amer. Med. Assoc., 1915, 65, 312. INFECTION WITH ANIMAL PARASITES 
81 
as that of bacteriology, although less important, because so few of these 
infections are of a dangerous and fatal character. According to Castellani 
and Chambers1 the parasitic fungi for man are practically all found among 
the Phycomycetes, the Ascomycetes, and the Fungi imperfecti. 
The majority of these pathogenic microfungi produce infections of the 
skin and adjoining mucous membranes, as, for example, ring-worms of the 
hairy and non-hairy portions of the skin (Microsporon audouini; Tricho- 
phyton tonsurans, etc.), favus (Achorion schoenleini), Tinea versicolor 
(Microsporon furfur), thrush (Monilia or O'idium albicans), sporotrichosis 
(Sporotrichum beuermanni), and many tropical diseases. Others may in- 
fect internal organs, as well as Aspergillus fumigatus (aspergillosis of the 
lungs, eye, ear, nose, wounds, and skin), Nocardia bovis, and other varieties 
(actinomycosis of nasal mucosa and lungs and actinomycotic mycetomas), 
and other fungi producing disease primarily or gaining access to the tissues 
as secondary invaders. 
Practically all that has been presented in this chapter on the mechanism 
of infection with bacteria is applicable to infection by the pathogenic fungi. 
Curiously, investigators in the field of mycology have devoted their efforts 
almost exclusively to morphologic and cultural characteristics of the various 
fungi without giving much attention to the mechanism of infection. 
They are usually transmitted by direct contact from man to man, man 
to lower animal, or, more commonly, by a lower animal to man. Infection 
is apparently governed by the same principles as is infections by bacteria, 
among which virulence, avenue of infection and tissue susceptibility, in- 
timacy of contract as influencing numeric infection, invasiveness and natural 
immunity of the skin in different parts of the body, among races of people, 
and even between the child and adult of the same race, are important factors. 
In this connection mention may also be made of the phytotoxins, or 
toxic substances, derived from certain plants closely resembling the bac- 
terial toxins and possessing lytic and agglutinative properties for the cor- 
puscles of many animals. These will be discussed in more detail in the 
following chapter. 
INFECTION WITH ANIMAL PARASITES 
As previously stated many animal parasites of microscopic and macro- 
scopic size are known to infect or infest man and the lower animals. A 
great number of saprophytic and pathogenic animal parasites have been 
studied and classified, although the mechanism of infection by those of 
microscopic size has not received much attention, most effort being devoted 
to a study of the morphology, life history, and host or hosts of the respec- 
tive organisms. 
Among the pathogenic animal parasites the Protozoa and worms are 
the most important, particularly the former. Various spirochetes and 
treponemata (syphilis, yaws, relapsing fever, infectious jaundice, and rat- 
bite fever), Leishman bodies (kala-azar, oriental sore, etc.), trypanosomes 
(sleeping sickness, Chagas’ disease), amebae (dysentery, Craigiasis), malaria, 
intestinal flagellates and ciliates, and other sporozoa, flukes, tapeworms, 
hookworms, filarise, and trichina worms are of importance in connection 
with the subject of infection and investation with animal parasites. 
Infection with animal parasites is similar in many respects to infection 
with bacteria. Owing to the difficulty of isolating and cultivating these 
parasites in vitro, our knowledge of their toxic properties is somewhat 
1 Manual of Tropical Medicine, 3d ed., 1920, William Wood & Co. 82 
INFECTION 
meager. Most attention has been given to a study of their life history 
and the modes of transmission. 
Primary infection with animal parasites is often facilitated by, or in 
some instances only rendered possible through, the intervention of special 
carriers, usually various species of the Arthropoda. Thus, we now know 
that malaria is transmitted through the bite of infected mosquitos; African 
relapsing fever and Texas cattle fever, through the bite of certain infected 
ticks; trypanosomiasis, through biting flies. The ova of various intestinal 
parasites may require residence in certain of the lower animals before they 
can infect man. 
Infection may occur along the same routes as bacterial infection, and 
is governed in general by the same factors of local selection, tissue suscep- 
tibility, etc. Biting insects usually deposit the parasite directly in the sub- 
cutaneous tissues or in the circulatory fluids. Abrasion of the epithelium 
may be necessary in order to produce infection with Treponema pallidum, 
as in the majority of the bacterial infections. The ova or larva of other 
parasites may be swallowed or find lodgment in the upper or lower air- 
passages or accessory sinuses. 
It would appear that our natural defenses against infection with animal 
parasites are much weaker than those against bacteria. This is probably 
due to the greater resistance offered by animal parasites to such physical 
destructive influences of the host, as the acidity and germicidal activity 
of the secretions, temperature, etc., as well as to a general lack of natural 
antibodies in the body fluids of the host, and inability of leukocytes and 
other phagocytic cells to deal successfully with the invaders. That natural 
immunity against infection with certain animal parasites may exist is shown 
by the prevalence of certain infections among man, and their absence among 
lower animals, or vice versa. 
The aggressiveness of animal parasites is in general probably even greater 
than that of most bacteria, and a more or less extensive infection apparently 
occurs in all cases in wdiich the parasite had made successful invasion, some 
multiplying in the blood-stream (malaria, relapsing fever, trypanosomiasis, 
Texas fever, filariasis), others in the lymph stream (filariasis), and others 
in the tissues (syphilis, trichiniasis, amebiasis), without much opposition 
on the part of the host. Whether these factors are due to the aggressive 
forces of the parasites wdiich neutralize the defenses of the host, or wdiether 
they are due to the hardiness of the parasites and a lack of defense on the 
part of the host, is not known, but probably the latter is generally the case. 
As with bacteria, animal parasites show a wrell-marked selective affinity 
for certain tissues, as the malarial plasmodium for red blood-corpuscles, 
trypanosomes and spirochetes for blood plasma, trichina for voluntary 
muscle, various parasites for the intestinal canal and even for certain por- 
tions of the intestinal tract, others for the lung, etc. 
SUMMARY 
From what has been said it is clear that infection differs from mere 
surface contamination, and cannot be said to occur until the invading para- 
sites have reached the deeper tissues, or a point where they may grow and 
multiply. The surface epithelium and various secretions offer the most 
potent local obstacles to infection, but even when these barriers are broken 
down the invaders may not survive the onslaughts of various protective 
agencies of the host. In order to withstand and overcome these attacks 
the bacterium may undergo certain morphologic and physiologic changes SUMMARY 
83 
and actively secrete a substance that is inimical to the defensive forces of 
the host, or immunize itself against these forces. Thus, a certain species 
of bacteria may become selectively fortified or immunized against a certain 
host or organ of that host, and show a specific affinity for producing infec- 
tion of a certain animal or a particular organ. When the bacterium has 
overcome the defensive forces of a host, it may, by the formation and action 
of exogenous and endogenous toxins, bacterial proteins, mechanical block- 
ing of vessels, or formation of ptomains, produce disease. These various 
factors will be considered in greater detail in the following chapter. CHAPTER VI 
INFECTION (Continued) 
PRODUCTION OF DISEASE 
When pathogenic parasites have reached the deeper tissues and mul- 
tiplied, infection has occurred, but, as previously stated, tissue changes of 
sufficient extent to produce definite lesions and symptoms of disease may 
or may not result, depending upon whether or not the defensive forces of 
the host are able to overcome the invaders or are overcome by them. If 
the latter has occurred, and the invading parasite is firmly established in 
its host, the question of how the parasite and its products cause disease, 
that is, the mechanism of the production of an infectious disease, arises for 
consideration. 
The agencies by which bacteria successfully batter down the defenses 
of the body in the production of infection and the lesions and symptoms 
of an infectious disease are manifold and complex. Inflammatory changes 
are usually produced in a certain tissue, organ, or system of organs, and the 
resulting symptoms of disease depend largely upon: (1) the portal of entry, 
the number, virulence, and rate of multiplication of the bacterium which 
were discussed in the preceding chapter; (2) the production of poisonous 
products by the bacteria; (3) the production of toxic protein cleavage sub- 
stances by proteolytic enzymes of bacteria, leukocytes, and fixed tissue 
cells, and (4) the degree of disturbance of physiologic function of the par- 
ticular organs directly infected and those secondarily involved by circu- 
lating toxic substances. 
Exotoxins and Toxinemia.—One of the most important agencies con- 
tributed by bacteria in the production of infection and disease are the 
soluble or exogenous toxins which they make and secrete. With some 
bacteria, as the diphtheria, tetanus, dysentery, and botulism bacilli, these 
toxins constitute the chief offensive weapon, being secreted in the infected 
tissues and producing local inflammation and tissue necrosis; they are also 
absorbed into the lymph and blood constituting a condition of toxinemia, 
and may selectively localize or affect tissues and organs remotely situated 
from the primary site of infection, as occurs in tetanus and diphtheria. 
In addition to these typical “toxin producers,” it is highly probable 
that the majority of bacteria, including the pyogenic cocci (staphylococci, 
streptococci, pneumococci, gonococci, and meningococci), the pathogenic 
bacilli not already mentioned (typhoid, colon, cholera, influenza, etc.), 
and even many saprophytic bacteria are capable of producing some of 
these exotoxins which exert influence in the pathogenesis of disease. It 
appears that these toxins possess certain properties in common, although 
quantitatively they vary within extreme limits, and hence in relative im- 
portance in the causation of disease. 
Endotoxins.—In addition to the exotoxins it is highly probable that 
some of the pathogenic bacteria contain preformed intracellular toxins or 
endotoxins, which are liberated and become operative upon disintegration 
of the bacterial cells. Within recent years the existence of these endotoxins 
has been questioned and the toxicity of disintegrated bacteria assigned to 
Production of Disease by Bacteria 
84 PRODUCTION OF DISEASE BY BACTERIA 
85 
the presence of protein cleavage products resulting from enzymic activity 
or the lytic processes of amboceptors and complement. It is highly prob- 
able, however, that preformed toxins are to be found within some patho- 
genic bacteria and notably the pneumococcus, bound in some manner to 
the protoplasm which, being liberated upon dissolution of the cell, are 
capable of exerting an influence in the production of disease. 
One possible effect of these endotoxins is in the reduction of the phago- 
cytic activities of leukocytes; these effects have been especially studied by 
Bail who believes that they are caused by separate products of bacteria 
called aggressins. Whether or not they are endotoxins or separate sub- 
stances cannot be definitely stated, but at any rate poisons possessing this 
leukotoxic action are to be found in seme inflammatory exudates; doubt- 
less they contribute a factor in the pathogenesis of disease by combating 
phagocytosis, which is one of the most important defenses against bacterial 
infection. 
Bacterial Protein.—Aside from the effects produced by exotoxins and 
endotoxins it would appear that bacterial protein and particularly their 
nucleoproteins, possess pyogenic properties and may aid in the production 
of local inflammation and suppuration. Furthermore, toxic protein cleav- 
age substances are probably produced by the digestion of dead and de- 
vitalized bacteria by their own proteolytic enzymes or those derived from 
the leukocytes, fixed tissue cells, and fluids of inflammatory exudates. In 
the majority of diseases the amounts of toxic split proteins derived from 
bacterial substrate must be small, but nevertheless are to be considered 
as additional factors in the pathogenesis of disease. 
Furthermore, recent investigations have indicated that bacterial pro- 
teins may bring about certain obscure and unexplained physical and chemical 
changes in the plasma resulting in the removal of antienzymes followed by 
digestion of blood constituents by proteolytic enzymes with the produc- 
tion of protein poisons. 
Toxic Exudates and Ptomains.—Additional toxic protein substances are 
probably produced during some infections by digestion of constituents of 
inflammatory exudates by the liberated proteases of leukocytes and fixed 
tissue cells. These protein cleavage products are known to be highly toxic 
for experimental animals and wrhen absorbed from inflammatory foci 
doubtless contribute in an important manner to the production of fever 
and other phenomena of infection. 
In addition to these toxic products resulting primarily or secondarily 
from bacterial infection it is probable that the mechanical action of bacteria 
may sometimes be a factor of importance and especially in bacteremias of 
pyogenic cocci during which metastatic abscesses develop in the kidneys 
and other organs. In certain protozoan infections, as malaria and trypano- 
somiasis, embolism may play a more important role by blocking small, 
but physiologically important, vessels with masses of parasites. 
Bacterial Toxemia.—The sum total of the effects of those different 
soluble substances produced during bacterial infection may be said to con- 
stitute bacterial toxemia. The term “toxemia’’ is frequently used for desig- 
nating the effects of an exotoxin, as in tetanus, but, as mentioned above, 
I believe the term toxinemia is better adapted for designating the effects 
of an exotoxin. 
Toxemia is scarcely ever entirely absent in bacterial infections, but 
varies considerably in degree, being severe in such diseases as typhoid fever, 
influenza, pneumonia, meningitis, etc., and relatively feeble in minor pyo- 
genic surgical infections. The general symptoms, however, are strikingly 86 
INFECTION 
similar whatever the variety of infection, embracing fever, changes in the 
pulse, quantitative and qualitative changes in the leukocytes, malaise, 
anorexia, muscular weakness and pains, headache, and depression. 
Bacterial toxemia strictly refers to the presence in the blood and lymph 
of toxic products of bacterial activity. As stated above, these may be almost 
solely exotoxins in diphtheria and tetanus, but in other infections are due to 
exotoxins, liberated preformed endotoxins, and various soluble and toxic 
split proteins resulting from the cleavage of destroyed bacteria and, more 
importantly, of destroyed fixed tissue cells, and the fluid and cellular ele- 
ments of inflammatory exudates. 
Aside from the primary local effects produced by infection varying 
according to the part or organ involved, these toxic substances may pro- 
duce important secondary effects in organs remotely situated, and particularly 
the nervous system, heart, spleen, and kidneys. 
The production of bacterial diseases is, therefore, an exceedingly com- 
plex process in which many factors are concerned and for which simple 
explanations are not possible. It is true that a large amount of experi- 
mental data has accumulated, but so many technical and unexpected bio- 
logic factors enter into experiments that the subject still demands a great 
deal of study and investigation for the elucidation of many problems con- 
cerning infection and the production of disease. 
Production of Disease by Higher Plants 
Little is definitely known of the agencies by which the microfungi pro- 
duce disease. Curiously, the subject has not received much attention from 
mycologists, most effort being devoted to morphologic and cultural char- 
acteristics. 
Some fungi, as those causing ring-worm, produce but slight inflammation, 
while others produce extensive suppuration, as in actinomycosis and blasto- 
mycosis. It is probable that some fungi produce exogenous toxins and that 
endotoxins may be liberated upon breaking up the fungi; these, in addition 
to toxic split protein products from fungi and body cells, are likely respon- 
sible for the local inflammatory changes and also for the symptoms of 
toxemia accompanying severe infections. Infections with fungi may be 
mixed with bacterial infection, the latter adding elements in the production 
of local and general reactions. A study of the toxins of microfungi and 
other agencies by which they produce disease offers an interesting and 
extensive field of investigation in mycology. 
Production of Disease by Animal Parasites 
Animal parasites may produce deleterious effects in five principle ways: 
1. By abstracting food material from the intestine which has not yet 
been assimilated. This is probably of minor importance, but may be a 
pathogenic factor in infestations with some of the tapeworms and especially 
Taenia saginata, which may be many yards in length. 
2. By abstracting blood as by the hookworms, flukes, leeches, and blood- 
sucking arthropods. 
3. By mechanical injury to tissues and organs. In this connection may 
be mentioned the blocking of blood-vessels by blood flukes, trypanosomes, 
and sub tertian malaria protozoa; the partial or complete blocking of lymph 
vessels by filaria and of bile and pancreatic ducts by liver flukes. 
Under this heading mention may also be made of certain parasites de- 
structive for tissue cells, as the malarial protozoon for erythrocytes; the THE COURSE OF INFECTIOUS DISEASE 
87 
pathogenic ameba which may bore into the intestinal mucosa, the lung 
flukes, fly maggots and trichinella, guinea-worm and itch mites producing 
injury by migratory and boring activities. 
4. By carrying pathogenic bacteria into the tissues. This refers chiefly 
to Amoeba histolytica boring into and under the intestinal mucosa carrying 
Bacillus coli and other bacteria which aid in the inflammatory processes, 
ulceration, and toxemia; also Amoeba gingivalis boring deeper and deeper 
into the tissues of the gums, and carrying various bacteria into deeper and 
healthy tissue in the production of pyorrhea gingivalis. 
5. By the formation of toxic substances. Comparatively little is known 
of this phase of the problem. Some, as, e. g., Treponema pallidum and the 
spirochete of relapsing fever, probably cause disease largely through the 
production of toxins, especially of the intracellular variety. The chill, 
fever, and sweat of malaria suggest the liberation of toxic products co- 
incident, or nearly so, with segmentation and rupture of plasmodia. The 
late symptoms of sleeping sickness, and the whole course of relapsing fever 
are strikingly similar to the bacterial toxemias. Gastel1 has shown the 
presence of a toxemia in infestations with Trichinella spiralis and the pro- 
duction of toxins by other worms is possible. 
Since the chief pathologic effect of some of the parasitic worms seems 
to be an anemia, the possible production of an hemolysin is to be considered 
as a partial cause of anemia in addition to hemorrhages from the bowel. 
Hemolysins have been repeatedly extracted from certain types of intestinal 
parasites. Faust and Tallqvist2 believe that the hemolytic substance from 
an intestinal cestode (Bothriocephalus) is in the nature of oleic acid esters, 
and perhaps other fatty acids. The more recent researches of Schwartz 
of the Bureau of Animal Industry,3 who obtained hemolysins from Ascaris, 
Ancylostoma, Trichuris, and several other forms, seem to show, however, 
that different agents must be involved and that the toxic symptoms of 
helminthiasis are probably caused by the secretion of toxic substances. 
Furthermore, Beumer4 has recently shown that feeding oleic acid to animals 
is not accompanied by permanently untoward effects, which disproves the 
possible harm of oleic acid as a hemolytic cause of anemia. 
Nevertheless, we know comparatively little of the offensive factors, and 
still less of the immunologic defensive factors, operative during the course 
of infestations with animal parasites. With the development of a technic 
for the cultivation of animal parasites in vitro, similar to that devised for 
the ameba, certain trypanosomes, spirochetes, and malarial plasmodia, 
we will be enabled to study the products of their growth or of disintegration, 
and the immunologic agencies concerned in infection and recovery; this 
offers a very important and fruitful field for research. 
THE COURSE OF INFECTIOUS DISEASE 
In conclusion, we may briefly consider the results of infection or the 
general symptoms following bacterial growth and the manner in which 
these are produced. 
The Stages of Infection.—Practically all infections pass through the 
following stages: 
1. The period of incubation, which begins at the time of infection and 
ends with the development of the earliest general symptoms, during which 
1 Centralbl. f. Bakteriol., orig., 1914, 74, 254. 
2 Arch. f. exper. Path. u. Pharmakol., 1907, 57, 367. 
3 Jour. Amer. Med. Assoc., 1920, 75, 1786. 
4 Biochem. Ztschr., 1919, 95, 239. 88 
INFECTION 
time the invading parasites are multiplying in the tissues of the host. Dur- 
ing this stage no symptoms, or only those of a purely local nature, are 
present. This period varies considerably in different infections, and to a 
lesser extent in different individuals having the same infection. Some 
parasites may be so virulent as to overwhelm the body cells, thus making 
the period of incubation very short or entirely unobservable. On the other 
hand, as, e. g., in rabies, the period may be of several weeks’ and, indeed, 
of several months’ duration. In tuberculosis there is usually a primary 
local growth, which develops so gradually and the toxins are so slowly 
diffused that it is difficult or, indeed, impossible, to estimate the length 
of the period of incubation. 
According to Vaughan, during the period of incubation the bacteria or 
their toxins or the viruses are actively engaged in changing the natural 
body proteins into new and specific bacterial proteins, and since this stage 
is constructive, there are no symptoms and the host is not ill. Even with 
the experimental administration of the most poisonous of toxins a definite 
period of incubation is usually to be observed, which cannot be reduced 
below a certain minimum, independent of the size of the dose injected; in 
general, however, a large dose of bacteria or of toxin is likely to be followed 
by a shorter period of incubation than if a smaller dose were administered. 
Similar views have been advanced by von Pirquet. In studying serum 
sickness, an anaphylactic phenomenon frequently observed in man follow- 
ing the administration of horse-serum, von Pirquet argued that the period 
of from eight to ten days usually following the injection before the appear- 
ance of symptoms was the time required for the production of the anti- 
body, which then reacted upon the serum still remaining in the body cells 
and fluids, and that the products of this interaction caused the lesions and 
symptoms of serum sickness. It was then but a short step to apply these 
principles to infectious diseases. This “period of incubation” was formerly 
regarded as representing a stage during which the infecting micro-organisms 
multiply in the body of the infected individual, to that point at which they 
could give rise to symptoms of disease through the agency of their toxins 
or through interference with the metabolism of the host in other ways. 
He and Vaughan wrould have us believe that during this period antibody 
formation is taking place, and that an antibody-antigen reaction will occur 
with the development of pathologic changes and symptoms just as soon 
as these changes have progressed to a certain point. The period of incuba- 
tion will vary not only in point of time of reaction but also qualitatively 
and quantitatively, and using this as a basis von Pirquet recognizes three 
main groups, depending upon whether the antibody is present in our body 
fluids as the result of a previously acquired infection (accidental or by 
vaccination), or whether it must first be developed. 
Group I.—Reaction appears after eight to twelve days, as in measles, 
smallpox, whooping-cough, chickenpox, and other infectious diseases in 
which the antibodies must be developed before the symptoms are produced. 
If at this time the antigen, i. e., either the albumins of the horse-serum, if 
we are dealing with serum injections, or the bacteria, in case of an infection, 
has disappeared from the body, no symptom will, of course, result; if, 
however, some of the material is still present, a reaction occurs, during 
which the protein poison (anaphylatoxin) is produced, and to which, in 
turn, the symptoms that then develop may logically be attributed. 
Group II.—The reaction appears after three to seven days. If, on the 
other hand, the secondary infection, as, e. g., pneumonia, erysipelas, etc., 
is acquired after a lapse of months or several years, or if the second injec- THE COURSE OF INFECTIOUS DISEASE 
89 
tion of serum is given after this time, i. e., at a time when the antibodies 
called forth by the primary infection or first injection have disappeared, 
a certain interval of time will elapse before symptoms of sickness develop, 
as in the case of the first group. This interval, however, instead of being 
from eight to twelve days, is now from three to seven, a fact readily ex- 
plained on the basis that a cell that has once been stimulated to active 
antibody formation will subsequently respond to the same stimulus with 
increased activity. This has been called by von Pirquet the accelerated 
reaction. 
Group III.—The reaction appears immediately. If the first injection 
of horse-serum or infection is followed by actual disease or vaccination, the 
reinjection or reinfection is acquired at a time when the antibodies are 
present in the circulation in considerable amount, a reaction will occur 
either immediately or within the first twenty-four hours. This reaction 
may be quite virulent in intensity, although it is shorter in duration than 
when it occurs in the first group, von Pirquet speaks of this as the immediate 
reaction. It is to be observed in cases of serum sickness where the symptoms 
develop almost immediately following an injection of serum months and 
even years after a previous injection; it also occurs in cowpox vaccination, 
where a local reaction takes place very quickly and soon disappears after a 
previous attack of smallpox or vaccination. 
If the antibodies are present in lesser amounts, the reaction may occur 
in from the second to the fourth day; this is called the torpid early reaction. 
At the time of the second injection of serum or reinfection with bacteria 
a small amount of antibody may still be present; this will give an immediate 
though mild reaction, and is not enough to neutralize the total amount of 
foreign protein introduced. A portion of the latter, therefore, will result 
in the production of an additional amount of antibody, which occurs in an 
accelerated manner, and coming in contact with some of the free antigen, 
gives rise to the accelerated reaction. Hence, we may have an immediate, 
followed by an accelerated, reaction. 
To illustrate these principles, von Pirquiet names vaccinia as an ex- 
ample of an acute infection in which the processes may be observed on the 
skin. As the result of vaccination a colony of micro-organisms is formed 
on the skin. For the first two days the local response is evidently traumatic 
in character. After the third or fourth day the specific reaction sets in, 
in the form of a small papular elevation surrounded by a small areola due 
to the local action of toxins or protein poison from disintegrated micro- 
organisms. By the eighth day a vesicle has formed, and from its contents 
new colonies can be grown on thousands of other arms. But one or two 
days later the ferment-like antibody appears. The colony is attacked, its 
contents are digested, a toxic substance is formed that diffuses into the 
neighboring tissues, and the intense local inflammation which we call the 
areola appears. In addition, the toxin enters the general circulation and 
fever sets in. Simultaneously the micro-organisms are destroyed, and we 
may no longer be able to vaccinate with the contents of the now yellow 
pustule. After two or three days the real struggle is ended, although the 
local lesion may be aggravated by secondary infection, and the body con- 
tains the new antibody for a long time. 
If we now revaccinate, the antibody present will at once attack and 
digest the micro-organisms introduced into the scarification, and, as these 
do not have time to multiply, only an extremely small amount of toxin is 
formed which gives the “immediate or early reaction” in vaccinia. If a 
number of years have elapsed between the first and the second vaccination, 90 
INFECTION 
antibodies may be absent or present in only small amount, but the body 
cells have been “keyed up” by the first vaccination, and hence react more 
quickly to the second. The antibodies are produced in from three to five 
days, and attack the micro-organisms before they have had time to multiply 
in sufficient numbers; the relatively small amount of digestion product 
produces a comparatively mild local inflammation and practically no general 
symptoms. This is sometimes known as the “immunity reaction,” or 
vaccinoid, and is illustrated in Fig. 178. 
In a given case the period of incubation may be determined by several 
factors: 
(a) The number of parasites gaining entrance, and especially their 
toxicity and aggressiveness. The primary factors are the degree of toxicity 
and the amount of toxic substances produced and absorbed. 
(b) Upon the site of infection. Thus, the introduction of rabies virus 
or of tetanus bacilli into the tissues of the face or into a deep wound is likely 
to be followed by a shorter period of incubation than when these are intro- 
duced into the foot or in superficial wounds. 
(c) Upon the degree of resistance offered by the host. For instance, 
one individual may contain more antitoxin or bacteriolysin for a certain 
bacterium than another, and consequently a longer period of incubation 
is required, during which these substances are neutralized and an excess 
of toxic bacterial substance is produced. In fact, these may offer such 
resistance to the bacterium that the process of infection is inhibited, or 
but slight and evanescent disturbances appear. 
(.d) Upon the general susceptibility of the host and the route of invasion. 
2. The period of prodromal symptoms, characterized by systemic dis- 
turbances of a relatively mild type, due to diffusion of the bacteria and 
their products into the general circulation and their wide-spread effect 
upon the body cells in general. If the parasites select a special tissue or 
organ for attack, as the typhoid bacillus for lymphoid tissue, and pneu- 
mococci for the lungs, definite symptoms develop later, their nature depend- 
ing on the special tissue or organ involved. The prodromata, however, 
are more marked, and indicate a wide-spread but mild action upon the body 
cells in general. Vaughan believes that these symptoms mark the time 
when sufficient proteolytic ferments have been generated by the body cells 
against the new bacterial protein of the invading bacteria to attack the 
latter, splitting the molecule and liberating a toxic moiety responsible for 
the general symptoms of intoxication. 
3. The period of fastigium, or of high fever, during which the disease is 
at the height of its severity. Special and distinctive symptoms and lesions, 
according to the organ or organs especially involved, are present; the struggle 
between the offensive and defensive forces of parasite and host is at its 
height, with remissions or exacerbations dependent upon the supremacy 
of any one of these, and the general stability of the body cells in withstand- 
ing the wear and tear. During this time the protective proteolytic ferments 
of Vaughan are most active in disrupting the newly formed bacterial pro- 
tein, with the liberation of the toxic portion. This process may be so active 
as to overwhelm the host with the toxic split product, or lead to grave 
secondary lesions, such as extensive necrosis, perforation of a viscus, or 
hemorrhage. 
4. The period of decline, during which the patient is gradually over- 
coming the infection, and amelioration of the symptoms takes place. 
5. The period of convalescence is now ushered in, during which the host 
gradually overcomes the effects of disease and returns to health. THE COURSE OF INFECTIOUS DISEASE 
91 
During this entire time the emaciation and tissue exhaustion leave the 
patient quite weak, and undue exertion, errors in diet, or reinfection may 
lead to a relapse, or a reactivation of the disease. Certain sequelae or morbid 
conditions may follow a disease, and are due to the same original cause; 
e. g., in typhoid fever the development of cholecystitis; at any time during 
the disease complications, or morbid conditions due to some other micro- 
parasite, as the development of pneumonia during the course of typhoid 
fever, may seriously jeopardize the life of the patient. 
Grades of Infection.—According to the manner in which a parasite and 
its products act upon the cell of a host and the power of the host to neu- 
tralize or overcome these the following various grades and types of infection 
are encountered: 
(a) Malignant or fulminating infection, during which there is no fever, 
but, on the contrary, a subnormal temperature, with rapid prostration of 
the patient and death within a brief period. The cells of the body are over- 
whelmed and paralyzed by the toxic substances; metabolism is arrested, 
and the heat centers are exhausted with the fall of the temperature, an 
indication of the intense and overwhelming intoxication. 
(&) Acute infection, which is the ordinary type of an infectious disease 
as previously described, and having a definite incubation period, prodromal 
symptoms, fastigium, defervescence, and convalescence. 
(c) Chronic infection, or a prolonged process characterized by insidious 
onset and symptoms of relatively mild or moderate severity, and termi- 
nating either in death, after months or years, or in gradual recovery. A 
chronic infection may be remittent, as may be observed in the rheumatic 
group of disorders; during the remission with defervescence the infecting 
bacterium is not totally destroyed, and subsequently lights up, producing 
an acute exacerbation of the disease. 
In chronic infections it would appear that the parasites develop and 
produce their toxins slowly, or that these are slowly and imperfectly ab- 
sorbed on account of the sluggish local circulation and the presence of scar 
tissue. The body cells become accustomed, as it were, to these toxic prod- 
ucts, and produce only sufficient antibodies to effect their immediate neu- 
tralization. The bacteria themselves become distinctly resistant to the 
action of the tissues and the defensive forces, and there is neither the same 
degree of intoxication nor reaction as are seen in acute infections. Grad- 
ually, however, the body cells become exhausted, and unless the cells are 
aroused and stimulated by judicious administration of bacterial vaccines 
to produce an oversupply of antibodies, the host shows progressive emacia- 
tion and weakness. 
The Systemic Reaction to Infection.—It is not within the scope of this 
book to discuss the various symptoms of infection, and we will limit our- 
selves to a brief discussion of the most important, namely, the febrile re- 
action. 
According to Vaughan, the fever of infection is due mainly to the toxic 
split protein resulting from the action of the protective proteolytic fer- 
ments upon the new bacterial protein. This observer and his associates 
were able, by the injection of multiple doses of protein derived not only 
from the typhoid bacillus but from various vegetable and animal proteins, 
to reproduce experimentally in rabbits a febrile reaction known as protein 
fever, and which is not unlike typhoid fever. This induced fever may con- 
tinue for weeks, and is accompanied by increased nitrogen elimination and 
gradual wasting; it is followed by immunity, and the serum of immunized 
animals digests the homologous protein in vitro. As has repeatedly been 92 
INFECTION 
stated, Vaughan regards the split toxic product as the cause of the general 
symptoms of infection, the special and characteristic symptoms and lesions 
of the different diseases depending upon the site where the bacterial pro 
teins have been deposited, and where they are, in large part at least, 
digested. 
In addition to this toxic action of split protein, fever may be due: (a) 
to the unusual activity of the cells supplying the proteolytic enzymes, and 
(b) to the cleavage of the foreign bacterial protein by these ferments. 
The fever of infection, therefore, is caused by the toxic action of patho- 
genic parasites, both bacterial and animal forms, upon the body cells and 
heat-regulating centers. It must be regarded by itself as a beneficent 
phenomenon, inasmuch as it marks a reaction of the body cells to toxic 
agents, for the purpose of neutralizing these and, by the development of 
antibodies, ridding the body of foreign substances. 
BACTERIAL TOXINS 
Nomenclature.—Of all the various means whereby bacteria produce 
disease, none possesses so much importance as the poisonous substances 
knowm as toxins, elaborated by the metabolic activities of the micro-organ- 
isms. A few classes of bacteria secrete this poisonous principle directly 
into the tissues or artificial culture-media in which they are growing, and 
hence are known as soluble, exogenous, extracellular, or true toxins. Other 
bacteria retain most of their toxins within the bacterial cell, and for this 
reason are called endotoxins, or intracellular toxins; these are liberated 
upon the disintegration of the bacteria by various mechanical, physical, 
or chemical means. 
By common consent the term “toxin” is applied to the soluble or true 
toxins, such as those of diphtheria and tetanus, and hence the term, when 
used without further qualifications, may be considered to refer to toxins 
of this class. 
Aside from bacterial toxins, characteristic poisons are also produced by 
certain of the higher plants (phy to toxins) and animals (zootoxins), and 
although few are of medical interest, their study has thrown considerable 
light on the phenomena of toxin-antitoxin immunity. 
EXTRACELLULAR BACTERIAL TOXINS 
Definition.—Bacterial toxins may be defined as poisonous products pro- 
duced by bacteria in both living tissues and artificial culture-media. The 
symptoms resulting from their activity appear after a certain period of 
incubation, and all are capable of stimulating the production of specific 
antitoxins. They represent the chief poisonous product of bacteria, and 
are mainly responsible for the symptoms of infection caused by the specific 
bacteria that have produced them. The first to be studied was diphtheria 
toxin by Roux and Yersin1 in 1888 and 1889. The methods adopted by 
these investigators enabled others to discover analogous toxins of several 
other bacteria. Faber2 and Brieger and Frankel3 soon succeeded in sepa- 
rating the toxin from the tetanus bacillus, a toxin capable of producing in 
animals tetanic contractions as typical as those obtained with cultures of 
the tetanus bacillus. 
1 Ann. d. l’Inst. Pasteur, 1888, 2, 629; ibid., 1889, 3, 273. 
2 Berl. klin. Wchnschr., 1890, 717. 
3 Berl. klin. Wchnschr., 1890, No. 11. EXTRACELLULAR BACTERIAL TOXINS 
93 
The true toxins causing infection in man are chiefly: 
1. Diphtheria toxin. 
2. Tetanus toxin. 
3. Botulism toxin (a form of meat poisoning). 
4. Dysentery toxin (Kruse-Shiga). 
5. Staphylotoxin, streptotoxin, Bacillus welchii toxin, and other bacterial 
toxins. 
General Properties of Soluble Toxins.—Many of the true toxins are ex- 
tremely labile, and susceptible to the action of heat, light, age, etc.; conse- 
quently an absolutely pure toxin is practically unknown. Most bacterial 
toxins are apparently destroyed by heating at 58° to 65° C.; Landsteiner 
and von Rauchenbichler,1 Dreyer and Blake,2 however, believe that at 
this temperature the toxins of staphylococci and Bacillus megaterium are 
masked by union with other substances, but not actually destroyed inas- 
much as additional exposure to 100° C. for ten minutes or more restores 
toxicity. Oxygen, even as it occurs in the air, is harmful; all oxidizing 
agents, including the oxidizing enzymes, quickly destroy them, and Pitini3 
has ascribed the harmful effects of toxins to their power of reducing the 
oxidizing capacity of the tissues. Some substances seem to attack only the 
toxophore portion of the toxin molecule, e. g., iodin and carbon disulphid 
(Ehrlich). In the preparation of antitoxin the first doses of toxin are 
frequently modified by adding a chemical of this nature. According to 
Gerhartz4 ac-rays tend to weaken the toxins. 
Because of their great lability the toxins do not lend themselves to 
accurate chemical analysis. Our knowledge of them has been gained largely 
through a study of the lesions and symptoms produced by injecting the 
toxins into susceptible animals. 
They are, so far as known, uncrystallizable and thereby differ from 
ptomains; they are soluble in water and dialyzable through thin but not 
thick membranes. They are precipitated along with peptones by alcohol 
and also by ammonium sulphate. 
The toxins are all poisonous, but in order to exert their toxic effect 
they must enter into chemical combination with cells; hence there is a 
necessary period of incubation before symptoms of their activity appear. 
Most bacterial toxins are not absorbed from the intestine (botulinus toxin 
excepted), and when introduced into the gastro-intestinal tract they are 
usually unable to produce symptoms and are quickly destroyed. 
An essential property of a toxin lies in the fact that we can immunize a 
subject against it, and are able to demonstrate the presence of antitoxin within 
the serum of the immunized animal. 
Chemical Properties of Exotoxins.—As has just been stated, the exact 
chemical nature of toxins is unknown. This is due principally to the fact 
that pure toxins of bacteria are rarely obtainable except in conjunction 
wfith their associated products, such as lysins, pigments, acids, etc., as 
well as to the great lability of the toxins. A summary of the results of re- 
searches into the chemical nature of toxins would indicate that they are 
toxalbumins, albumoses, or allied to the albumoses. Certain investigators 
have reported that very active toxins obtained by purification processes 
did not give the protein reactions, yet toxins are digested by proteolytic 
ferments, and, like proteins, are precipitated by nucleic acid (Kossel). Ac- 
1 Ztschr. f. Immunitatsf., orig., 1908, 1, 439. 
2 Lancet, 1904, 2, 409. 
3 Biochem. Zeit., 1910, 25, 257. 
4 Berl. klin. Woch., 1909, 46, 1800. 94 
INFECTION 
cording to Field and Teague,1 the toxins act like electropositive colloids, 
but diffuse faster than do proteins. Our present knowledge of the chemistry 
of the true toxins has been expressed thus by Oppenheimer: “We must 
be contented to assume that they are large molecular complexes, probably 
related to the proteins, corresponding to them in certain properties, but 
standing even nearer to the equally mysterious enzymes with whose proper- 
ties they show the most extended analogies both in their reactions and in 
their activities.” Warden, Connell, and Holly,2 in a recent study of the 
nature of the toxins and antigenic activities of Bacillus diphtherige and B. 
megaterium, concluded that the lysins and toxins of these two micro-organ- 
isms wTere the same substances, being respectively the specific fat antigens 
existing in definite and particular colloidal states. 
Precipitation of the Exotoxins.—After the toxin has been secured by 
filtration, crystals of ammonium sulphate are added in large excess over 
the saturation point, and the whole kept at 37° C. for eighteen hours. The 
toxin is precipitated and rises to the surface along with the albumoses and 
peptones. This is skimmed off and quickly dried with an electric fan and 
cold air. The residue is ground into a fine powder and stored in vacuum 
tubes kept at a low temperature and in a dark place. During this process 
there may be considerable deterioration and especially with tetanus toxin. 
Banzhaf has obtained highly potent and dried diphtheria toxin by slightly 
acidulating the toxin broth and adding absolute ethyl alcohol up to 65 per 
cent.; after an hour or two the slight precipitate is filtered off, quickly dried, 
and kept in ampules. 
Structure of Exotoxins.—According to Ehrlich, the toxin molecule con- 
sists of a main central atom or radical, with a large number of organic side 
chains grouped, as in other organic compounds, about this main radical. 
Each of the side or lateral arms is composed of two portions—one, the 
haptophore group, which has a chemical affinity for certain chemical con- 
stituents of the tissues of susceptible animals, and the other, the injury- 
producing portion, called the toxophore group (Fig. 40). An animal is 
susceptible to a toxin only when its cells contain substances that possess 
a chemical affinity for the haptophore group of the toxin, and also sub- 
stances susceptible to the toxic action of the toxophore group. 
The toxophore group is far more unstable and susceptible to deleterious 
influences than is the haptophore portion. When the molecule has lost 
the toxophore radical it is known as a toxoid, which is still capable of unit- 
ing with the side arms of cells, but is devoid of toxic action. 
Nature of Exotoxins.—It has been abundantly demonstrated that toxins 
are colloids, and in many respects bear a close resemblance to enzymes. 
The toxins are synthetic products of bacterial activity. The}' are of abso- 
lutely specific nature, and in this manner differ from ptomains, which are 
cleavage products from the medium upon which the bacteria have been 
grown. Furthermore, ptomains of similar properties may be produced by 
several different, kinds of bacteria, and accordingly are non-specific in nature. 
Toxins, like ferments, can give rise to antibodies, whereas ptomains cannot 
produce them. 
The extracellular or soluble toxins differ from the intracellular toxins 
in that they are more easily diffused throughout the animal juices, and 
that their diffusion occurs independently of the invasiveness of the bacteria, 
so that comparatively few micro-organisms growing at some unimportant 
focus and causing but slight local lesions may be able to give rise to pro- 
1 Jour. Exper. Med., 1907, 9, 86. 
2 Jour. Bacteriology, 1921, 6, 103. EXTRACELLULAR BACTERIAL TOXINS 
95 
found general intoxication. This is well illustrated in diphtheria, where 
the local lesion in the throat may be quite small, and in tetanus, where it 
may indeed be undiscoverable—yet either, through the action of their 
toxins on special tissues, may cause profound intoxication and death. 
Similarity Between Toxins and Ferments.—The toxins bear a well- 
recognized and close resemblance to the organic ferments; these points of 
resemblance may be summarized as follows: 
1. Both toxins and ferments are products of the metabolism of living 
animal and vegetable cells, and may be extracellular (free enzymes and 
soluble toxins) or intracellular (intracellular enzymes and endotoxins). 
2. Both are colloids; both pass through porcelain filters and are largely 
held back by dialyzing membranes. 
3. Both are usually affected by temperatures above 70° C., the toxins 
slightly more, however, than the ferments. Most toxins and ferments are 
destroyed at 80° C. In solutions both toxins and ferments deteriorate with 
the production of toxoids and fermentoids. 
4. The activities of both toxins and ferments seem to depend largely 
upon the temperature to which they are exposed. 
5. Both exhibit a latent period before manifesting their individual 
activities; in general, the effect of each is more rapid the larger the amount 
present. Both are poisonous for animals and when injected produce anti- 
bodies. 
6. Both substances represent a method or means by which the organism 
attempts to modify its environment and render the surroundings suitable 
for nutrition and growth. 
7. Both show a strong affinity of their substratum, and first manifest 
their activity by combining with it. For example, fibrin placed in gastric 
juice at 0° C., and then repeatedly washed in cold water to remove all 
traces of pepsin will undergo digestion when raised to body temperature. 
Similarly, if red corpuscles are placed in fresh tetanus toxin at 0° C. for 
an hour, washed repeatedly with cold normal saline solution, and then 
raised to 37° C., hemolysis will take place, indicating the primary union 
of the bacterial hemolysin or tetanolysin with the corpuscles. In a similar 
manner toxins probably unite chemically with tissue cells, as the toxin 
quickly disappears from the blood following its injection and but a small 
fraction can be recovered from the excretions. Furthermore, the injection 
of an emulsion of these cells into other animals may be followed by specific 
symptoms of intoxication. 
8. The one great difference, however, between toxins and enzymes is 
the greater activity of the latter, even very minute amounts of an enzyme 
having the power to split up or decompose large quantities of complex 
organic compounds. An enzyme attaches itself to a substance and absorbs 
water; the molecule breaks down, the enzyme is liberated, and then attacks 
another molecule, this process being repeated until large amounts of fer- 
mentable substances have been attacked. When, however, a toxin has 
united with a substance it loses its identity and in this manner it follows 
the law of multiple proportions. This has been discussed as it relates to 
the soluble toxins of diphtheria and tetanus, and is likewise easily demon- 
strable in the action of tetanolysin upon erythrocytes of the rabbit. It is 
true that a toxin may become dissociated and attack another molecule, 
but this action is different from that of an enzyme, because the molecule 
first attacked is not injured. However, as Adami points out, the toxins 
may be equally active in the body until arrested by antitoxins, although 
experiments in vitro clearly demonstrate the greater activity of the ferments. 96 
INFECTION 
Von Liebermann1 denies the identity of toxins and ferments on the 
basis of experiments with ricin, because this substance appears to be “used 
up” in the agglutination of erythrocytes, is not appreciably affected by 
hydrocyanic acid, and is more thermostabile than toxins. Coca,2 on the 
other hand, believes that the toxin of cobra venom is a ferment in the nature 
of a lipase and subscribes to the view that toxins are ferments. Both of 
these investigators, however, worked with non-bacterial toxins and abso- 
lutely conclusive evidence that toxins are ferments has not been produced, 
largely because it is practically impossible to isolate either in an absolutely 
pure form for experiments. As bearing upon the relation of bacterial toxins 
to ferments mention may be made of the experiments of Dr. Moshage and 
myself,3 showing that toxin production by diphtheria bacilli is not abso- 
lutely parallel with the production of carbohydrate-splitting ferments, 
although these ferments were produced most frequently and vigorously 
by the best toxin-producing bacilli. 
Selective Action of Exotoxins.—Extensive studies of the toxins of diph- 
theria and tetanus and of cobra venom have shown that they are quite 
complex, and are usually composed of two or more distinct and separate 
toxins possessing different pathogenic properties, although one of these 
may predominate in producing symptoms. 
All infections with the group of true toxic-producing bacteria manifest 
certain non-specific symptoms of general intoxication, namely, fever, head- 
ache, malaise, prostration, etc.; but the typical symptoms of these diseases 
are due to the remarkable selective action of the toxins upon certain cells 
or organs, dependent upon the ability, chemical, physical, or both, of the 
toxin to combine with these specific cells. For example, tetanus toxin con- 
tains tetanospasmin, that has a special affinity for nervous tissue; and tetanol- 
ysin, a poison that has a selective affinity for erythrocytes and is hemotoxic. 
Ehrlich has shown that these are really different toxins, and not one toxin 
with a twofold function, even the antitoxins of the two being different. 
Similarly, the general symptoms and necroses of diphtheria are attributed 
to the main toxin of the bacillus, and the nerve lesions and paralyses to a 
secondary but distinct secretory product known as toxon. This latter view 
of Ehrlich’s, howrever, is much disputed, many investigators believing that 
toxon represents a degenerated or modified form of the one toxin. 
Wadsworth and Vories4 have recently shown that neither the leuko- 
cytes of the dog nor those of the guinea-pig neutralize or combine with 
diphtheria or tetanus toxin. Although tissue combines with and neutralizes 
tetanus toxin it has no action on diphtheria toxin. 
The special affinities of toxins for certain tissues have analogies among 
the poisons of higher plant life, as, for example, strychnin has a similar 
selective affinity and is said to be specific in its action upon the motor 
cells. 
The venom of various serpents, especially that of the cobra, has specific 
action; the erythrocytes of various animals are readily attacked by it, 
and the cells of the respiratory center are apparently profoundly affected. 
Aside from the special effects of the toxins upon certain cells and tissues, 
it must be remembered that toxins may involve the body cells in general, 
and particularly those of the parenchymatous organs, such as the kidneys, 
heart, and liver, causing coagulation of the protoplasm (cloudy swelling) 
1 Deutsch. med. Wchnschr., 1905, 31, 1301. 
2 Jour. Infect. Dis., 1915, 17, 351. 
3 Jour. Infect. Dis., 1916, 19, 28. 
4 Jour. Immunology, 1921, 6, 413. Fig. 52.—Abdominal Wall of Guinea-pig Showing Diphtheric Edema. 
Shows abdominal wall of a guinea-pig forty-eight hours after subcutaneous injection with 2 c.c. 
of a seventy-two-hour bouillon culture of a diphtheria bacillus isolated from the throat of a diphtheria 
convalescent. SPECIAL PROPERTIES OF THE PRINCIPAL TOXINS 
and final dissolution. The harm brought about by the toxins or toxic 
products of the pyogenic group of micro-organisms, for instance, acts mainly 
in this manner. 
97 
SPECIAL PROPERTIES OF THE PRINCIPAL TOXINS 
1. Diphtheria Toxin.—Diphtheria bacilli vary considerably, both in 
tissues and in artificial culture-media, in the quantity of toxin secreted; 
thus, in bouillon large amounts are seldom found in less than from seven 
to fourteen days. 
The action of the toxin is dependent upon the dosage, and a certain 
period of time must always elapse before the symptoms appear, the mini- 
mum being about one day. Large doses may shorten this period of incuba- 
tion, but cannot diminish it below a certain limit. 
The lesion of diphtheria is practically always local, and is usually situ- 
ated on the mucous membrane of the upper air-passages. It is characterized 
by the formation of a pearly white membrane that is adherent to the under- 
lying edematous tissues. The toxin produces necrosis of the surface epi- 
thelium, and the product, together with fibrin and leukocytes, constitutes 
the membranous exudate. From this focus toxin is absorbed by the lym- 
phatics and blood-stream, and distributed throughout the body, the bacilli 
being rarely found in the blood or internal organs. Later, the effects of 
toxin intoxication are shown by paralyses of certain motor nerves and 
ganglia, particularly those of the palate and heart. 
When a guinea-pig receives a subcutaneous inoculation with diphtheria 
toxin, a typical hemorrhagic gelatinous edema develops at the site of inocu- 
lation (Fig. 52). Upon opening the abdominal cavity one finds but little 
peritoneal exudate, but the vessels of the mesentery are injected and the 
adrenal glands show characteristic acute hyperemia (Figs. 53 and 54). 
Bloody pericardial and pleural exudates will be found in the thorax and 
solidified areas in the lungs. Guinea-pigs surviving a dose of toxin may, 
after two or four weeks, begin to show paralysis of the hind and then of 
the fore extremities, a condition analogous to the postdiphtheric paralysis 
occurring in man and ascribed to the effects of toxon. 
Method of Testing the Virulence and Toxicity of Diphtheria Bacilli.— 
Young guinea-pigs weighing from 250 to 300 grams are quite susceptible to 
diphtheria toxin, and are used in determining the strength of a toxin and 
in standardizing antitoxin. The test may be of great value in the manage- 
ment of convalescent and “carrier” cases of diphtheria, harboring bacilli 
in the upper air-passages, in determining whether the micro-organisms are 
dangerous or merely harmless non-pathogenic saprophytes. It is prac- 
tically impossible, from the morphology of the organism alone, to decide 
whether or not a given culture is dangerous, and prolonged quarantine may 
not only be irksome and inconvenient, but, if the organisms are proved to 
be harmless, it is unnecessary as well. 
To be reliable, however, such a test must be carried out very carefully. 
In the case of a highly virulent culture the mere introduction of a few 
organisms beneath the skin will suffice to demonstrate their dangerous 
character, but with cultures only slightly virulent more care is necessary, 
for although the patient may show no ill effects as a result of the presence 
of the bacilli, in the throat of another and less immune individual they may 
be highly dangerous. 
The following method has been used by the author in many hundreds 
of such tests, and has proved of distinct value: 98 
INFECTION 
1. Make a culture of the part harboring the bacilli on a tube of Loffler serum medium. 
Incubate at 35° C. for from eighteen to twenty-four hours; prepare a smear and stain with 
Loffler’s methylene-blue. If diphtheria bacilli are present they must be isolated in pure 
culture. Never attempt a guinea-pig test with an impure culture! 
2. Isolate by the “streak” method on plates of blood-serum. 
3. Inoculate a tube of 1 per cent, glucose bouillon, which is neutral or slightly alkaline, 
with several dijferent colonies. 
4. Incubate at 35° C. for three days, keeping the tube in a slanted position in order to 
give the culture as much oxygen as possible. If a good growth does not appear in twenty- 
four hours, transplant to another tube of bouillon until the bacilli have been “educated” 
to grow on the medium. 
5. Examine for purity. Select a 250- to 300-gram guinea-pig and inject 2 c.c. of the 
unfiltered culture in the median abdominal line. Animals over the weight specified are 
more resistant and less reliable for the test. The unfiltered culture is used, since toxin is 
but one element of the disease-producing power of diphtheria bacilli, and toxin production 
in bouillon may not be a true index of the toxin production in mucous membranes. 
6. Carefully observe the animal for at least four days. Even slight toxemia, especially 
if accompanied by edema at the site of injection, should be regarded as a positive result 
(Fig. 52). 
7. After death perform a careful autopsy. Make cultures of the edematous area, perit- 
oneum, and heart blood. Diphtheria bacilli may be found in the edematous fluid, but will 
rarely lie found in the peritoneum or in the blood. Observe whether acute hyperemia of the 
suprarenal glands is present (Figs. 53 and 54). 
8. Not infrequently animals showing mild or even an absence of the symptoms of toxemia 
develop paralysis of the hindquarters two or three weeks later. According to Ehrlich, this 
paralysis is due to the action of “toxon,” a toxic substance secreted by the bacillus or, as 
believed by others, a modified form of toxin. 
9. To prove that diphtheria was the cause of the toxemia or death mix 2 c.c. of the 
culture in a test-tube with 1 c.c. of diphtheria antitoxin (500 units). After standing aside 
for an hour at room temperature, inject the mixture subcutaneously in the median abdominal 
line of a 250- to 300-gram guinea-pig. Symptoms of toxemia do not develop. 
In a comparative study of the above and other methods Kolmer and 
Moshage1 found that washing off a pure culture from a slant of Loffler’s 
blood-serum media with 10 c.c. of sterile salt solution, emulsifying and 
injecting 4 c.c. subcutaneously in the median abdominal line of a 250- 
to 300-gram guinea-pig, yielded equally delicate results. The intracutaneous 
method of Neisser, which has also been advocated by Zingher and Soletsky,2 
while being more economical, in that 2 pigs suffice for 4 or even. 6 tests, was 
found to yield somewhat indefinite reactions with bacilli of low virulence. 
Standardizing Diphtheria Toxin.—The strength of a diphtheria toxin is 
estimated by injecting subcutaneously a series of guinea-pigs weighing 
approximately 250 grams, with decreasing amounts of toxin. How many 
dilutions will be necessary it is impossible to state; for exact results several 
pigs of the same weight should be inoculated with the same dose, and the 
effects should show various gradations, dependent upon the size of the 
successive doses. In order to obtain a uniform method for estimating the 
strength of a diphtheria toxin and thus obtain comparative values, a standard 
unit has been adopted, consisting of the smallest amount of toxin that will kill 
a healthy guinea-pig weighing about 250 grams in from four to five days. This 
is known as the minimum lethal dose, or dosis lethalis minimus. The technic 
used for determining this dose is given in the chapter on Antitoxins. 
A quick and accurate method for estimating the amount of diphtheria 
toxin present in the body fluids of a diphtheric patient would be of value 
in controlling the antitoxin treatment of this infection. At present the 
amount of antitoxin administered is regulated according to the clinical 
condition of the patient. Uffenheimer has used a method for determining 
the presence of toxin, consisting in injecting intraperitoneally a 250-gram 
guinea-pig with 0.1 to 0.4 c.c. of the patient’s serum, diluted with 2 to 4 
1 Jour. Infect. Dis.,1,1916, 19, 1. 
2 Jour. Infect. Dis., 1915, 17, 454. Fig. 53.—Normal Adrenal Gland of a Guinea-pig. 
Fig. 54. -Adrenal Gland of a Guinea-pig After Fatal Diphtheric Intoxication. SPECIAL PROPERTIES OF THE PRINCIPAL TOXINS 
c.c. of salt solution. The presence of a distinct doughy edema of the ab- 
dominal cavity after seventeen to twenty-four hours indicates the presence 
of diphtheria toxin, an observation that may be confirmed by making an 
autopsy at the end of forty-eight hours. The diagnostic value of this method 
has not been adequately established; it is doubtful if it yields any informa- 
tion other than is more readily gained by making a good cultural examina- 
tion of the patient, and it does not aid in the estimation of the quantity 
of toxin, which is the result most desired. 
Diphtheria toxins have been classified into three groups, depending upon 
the degree of avidity for antitoxin they display, viz., pro to toxin, deute- 
rotoxin, and tritotoxin. Each of these toxin groups may, in whole or in 
part, be converted into toxoids. The prototoxin has a greater affinity for 
the antitoxin than has the deuterotoxin, and the deuterotoxin has a greater 
affinity for the antitoxin than has the tritotoxin. The same relation is 
apparent with the three toxoids, which are not poisonous, but which have 
the same power of combining with antitoxin as have the toxins from which 
they take their origin. 
In standardizing antitoxin, it is found in general that with a perfectly 
fresh toxin a certain amount of antitoxin will just neutralize a definite 
amount of toxin. If older toxin is used, it is found that the toxin has lost 
about one-half its toxic power, but retains its initial power for neutralizing 
antitoxin. Ehrlich explained this by showing that the diphtheria toxin 
molecule is composed of two groups—one the carrier of the toxic qualities, 
the toxophore group, which is quite labile; the other uniting the whole mole- 
cule with antitoxin, being capable of neutralizing it, and characterized by 
its stability. The toxophore group being destroyed as in old toxin, the 
poison loses its toxic qualities, but retains its power to bind antitoxin. 
This modified toxin or non-poisonous diphtheria toxin has been designated 
by Ehrlich diphtheria toxoid. 
2. Tetanus Toxin.—Of all bacteria classed as true toxin producers, 
none possesses greater toxicity than does the tetanus bacillus. The number 
of organisms producing sufficient toxin to cause a fatal infection may be 
so small that careful anaerobic cultures made from the local lesion of in- 
fection, together with injection of the wound secretions into white mice, 
may fail' to disclose the presence of tetanus bacilli. 
According to Ehrlich, tetanus toxin is composed of two separate and 
distinct substances: (1) Tetanospasmin, a neurotoxin, wdiich is very labile 
and responsible for the severe symptoms of the infection; (2) tetanolysin, 
a hemotoxin, which is more stable and destructive for erythrocytes. 
Tetanus toxin is prepared by cultivating the bacillus in bouillon under 
strict anaerobic conditions. Since tetanospasmin is so susceptible to the 
influence of heat, age, and even light, the toxin is best preserved in a dry 
form. The standard of tetanus toxin consists of 100 minimal lethal doses 
of a precipitated and dried toxin, preserved at the Hygienic Laboratory 
of the Public Health and Marine Hospital Service. 
If susceptible animals, such as mice or guinea-pigs, are injected sub- 
cutaneously or intravenously with tetanus toxin, they begin to manifest 
symptoms after a certain period; these are due to the action of tetano- 
spasmin upon motor nerve-cells, and are characterized by hypersensitive- 
ness, clonic convulsions, and rigidity of the muscles. In man the symptoms 
of tetanus are similar to those in the animal, the spasm starting quite regu- 
larly in the muscles of the lower jaw. 
Experiments by Wassermann and Takaki have demonstrated that an 
especially close affinity exists between tetanus toxin and certain struc- 
99 100 
INFECTION 
tures, particularly that of the central nervous system. Most writers agree 
that the toxin reaches these tissues largely by way of the nerve paths. 
Teale1 believes that some toxin may ascend to the central nervous sys- 
tem by way of the axis-cylinders of the nerves, also, and to a greater extent, 
along the perineural lymphatics. According to his experiments the toxin 
does not pass through the choroid plexus into the cerebrospinal fluid. Peter- 
son2 has found that the toxin is absorbed not only by nerve-cells but also 
by the leukocytes and fixed tissue cells of several organs of different animals 
including the dog, rabbit, and guinea-pig. 
3. Botulism Toxin.—This poison is generated by the Bacillus botulinus, 
first isolated by Van Ermengem in 1896 from a ham during an epidemic 
of meat poisoning. It is the cause of a type of meat and sausage poisoning 
called botulism, more frequent in those countries where raw meat is eaten, 
and frequently confused with “ptomain poisoning.” 
The bacillus is a motile, spore-forming, anaerobic bacterium, which 
grows at room temperature and causes marked gas formation in glucose 
media. 
The toxin is readily produced in anaerobic alkaline bouillon cultures. 
It is quite labile. As shown by Edmondson, Giltner, and Thom3 different 
cultures vary in toxin production and both bacilli and toxin-free spores 
are virulent for guinea-pigs; according to Shippen4 the bacilli will grow and 
produce toxin in aerobic cultures in symbiosis with staphylococci. 
Symptoms of botulism appear only after a definite period of incubation, 
which varies from twenty-four to forty-eight hours. In contradistinction 
to the meat poisonings produced by other organisms, those due to Bacillus 
botulinus may show few or no symptoms directly referable to the intestinal 
tract, the chief symptoms being due to toxic interference with the cranial 
nerves: loss of accommodation, ptosis, dilated pupils, aphonia, dysphagia, 
and hypersecretion of mucus from the mouth and nose. 
Sporadic toxemic-like disease in cattle sometimes designated as “forage 
poisoning” has occurred with varying severity throughout the Middle 
Western states; Graham and Schwarze5 have observed outbreaks in equines 
which were quite definitely related to the consumption of feed containing 
Bacillus botulinus toxin, an anerobic bacillus biologically resembling B. 
botulinus (Type B) being isolated from a corn silage. 
Guinea-pigs are quite susceptible, and may be infected by way of the 
mouth. The symptoms of intoxication usually follow in twenty-four hours, 
and are characterized by motor paralysis, dyspnea, and hypersecretion of 
mucus from the nose and mouth. 
4. Dysentery Toxin.—The distinct types of dysentery bacilli vary ex- 
ceedingly in their powers to produce toxins, the strongest poisons being 
produced with bacilli of the Shiga-Kruse variety, less regularly active 
ones, with bacilli of the Flexner type. 
Investigations have shown quite conclusively that dysentery itself is a 
true toxemia, its symptoms being referable to the absorption of the toxins 
of the bacillus from the intestine. Flexner, who has studied this subject 
with great care, believes it probable that most of the pathologic lesions 
occurring in the intestinal canal are referable to the excretion of dysentery 
toxin rather than to the direct local action of the bacilli. The action of 
1 Jour. Path, and Bact., 1919, 23, 50. 
2Ztsch. f. Immunitatsf., orig., 1910, 8, 498. 
3 Archiv. Int. Med., 1920, 26, 357. 
4 Archiv. Int. Med., 1919, 23, 346. 
5 Jour. Bacteriology, 1921, 6, 69. SPECIAL PROPERTIES OF THE PRINCIPAL TOXINS 
101 
the dysentery toxin upon animals is very characteristic, and throws much 
light upon the disease in man. Intravenous injection of the toxin in rabbits 
is followed by marked diarrhea, rapid fall in temperature, respiratory em- 
barrassment, and terminal paralysis. Upon autopsy the intestinal mucosa, 
especially that of the cecum and colon, shows marked inflammatory involve- 
ment, supporting Flexner’s observation of the necrotic action of excreted 
toxin. 
Dysentery bacilli also produce an endotoxin, and poisonous substances 
are easily obtained by extracting the bacilli themselves or by filtration of 
properly prepared bouillon cultures. The toxin is fairly stable, and well 
preserved under toluol in the refrigerator. 
As shown by Olitsky and Kligler1 the Shiga dysentery bacillus produces 
a true exotoxin possessing a marked affinity for nervous tissue and pro- 
ducing muscular weakness and paralysis of rabbits, and an endotoxin, re- 
sponsible for the production of intestinal lesions. Thjdtta2 has recently 
grouped dysentery bacilli into three groups, the Shiga bacillus being toxic 
and belonging to Group I; the atoxic bacilli, as the Flexner, Strong, and 
Hiss Y strains, were placed in Group II, while Group III embraces a second 
toxic strain similar to Group I. 
5. Staphylococcus Toxins.—Two definite toxins have been isolated from 
cultures of Staphylococcus pyogenes aureus and albus, one of which exerts 
a destructive action on erythrocytes (hemotoxin), and the other on leuko- 
cytes (leukocidin). 
An antihemotoxin that counteracts the effects of the toxin may be 
produced experimentally, and in human staphylococcus infections the 
demonstration of such antihemotoxic substances in the blood-serum may 
be of aid in making the diagnosis of staphylococcus infections. This anti- 
staphylolysin may be found normally in small amounts in the serum of 
man and horse, and when antihemotoxic tests with human serum are made 
a normal control should always be included. Antileukocidins have also 
been produced, but are not of practical importance. 
The hemotoxin is readily formed in cultures of staphylococci; roughly, 
the amount produced depends upon the virulence of the culture. In human 
cases of staphylococcus infections this toxin produces hemolysis in vivo, 
and is partly responsible for the grave anemia that is frequently present. 
Orcutt and Howe3 have recently identified a staphylolysin as a fatty acid 
or soap formed by the action of a lipase elaborated by the cocci when culti- 
vated in the presence of fat. 
6. Streptococcus Toxins.—The grave systemic symptoms that so fre- 
quently accompany slight streptococcus lesions are strong indications that 
these micro-organisms produce a powerful diffusible poison, although exten- 
sive researches into the nature of these poisons have not given us any clear 
understanding of the subject. 
Streptococci may yield soluble toxins that, wdien administered to guinea- 
pigs, produce rapid collapse and death. While these toxins are not com- 
parable in potency to the soluble toxins of diphtheria and tetanus, they 
have, nevertheless, been differentiated from the endotoxins contained within 
the cell-bodies, and have been found to possess less toxicity. 
Beside these toxins, some streptococci produce a hemolysin first de- 
scribed by Bordet4 wdiich may be conveniently observed by cultivation of 
1 Jour. Exper. Med., 1920, 31, 19. 
2 Jour. Bacteriology, 1919, 4, 355; ibid., 1921, 6, 501. 
3 Jour. Exper. Med., 1922, 35, 409. 
4 Ann. d. l’lnst. Pasteur, 1901, 15, 880. 102 
INFECTION 
the organisms upon blood-agar plates. This hemotoxin is partly responsible 
for the sanguineous character of a streptococcus exudate. Nakayama,1 
in a recent study of these streptococcus poisons, identified one as a leuko- 
cidin, an apparently different poison from the hemolytic toxin or strepto- 
lysin. Ruediger2 failed to produce specific serum antilysins for the hemo- 
toxin. 
Gas Bacillus and Other Bacterial Toxins.—During the World War 
many wounds showed the presence of gas-producing and spore-forming 
anaerobes wdiich received special attention to determine their role in wTound 
infections and from the standpoint of development of specific serum therapy. 
Three of these spore-forming bacilli were found especially important, namely, 
Bacillus tetani, B. wTelchii (B. aerogenes capsulatus, B. perfringens), and 
B. edematis maligni (bacillus of malignant edema, vibrion septique). 
Bull and Pritchett,3 De Kruif and Bollman4 have studied the toxin of 
Bacillus welchii with particular care, finding that most strains produce small 
amounts of a true toxin in special media which is toxic for the pigeon, pro- 
ducing necrosis of muscle. The presence of necrotic tissue in wounds favors 
the growth of these bacilli and toxin production; Bull and Pritchett suc- 
ceeded in manufacturing an antitoxin for the toxin of B. welchii which 
possessed prophylactic and curative properties in experimental infections. 
Further reference to the results observed in the prophylaxis and treatment 
of wounds is given in the chapter on Serum Therapy. 
Exogenous or soluble toxins have also been found by Haslam and Lumb5 
to be produced by Bacillus chauveaui (blackleg); Parker has described a 
toxin for B. influenza, and Zinsser6 believes that many pathogenic and 
even non-pathogenic Gram-positive and Gram-negative micro-organisms, as 
streptococci, influenza bacilli, typhoid colon, and dysentery bacilli, B. 
prodigiosus, Staphylococcus aureus and meningococci, may, under favor- 
able conditions, produce small amounts of what appears to be true soluble 
toxins, although these have not been definitely proved not to be endotoxins. 
Bacterial Hemagglutinins and Hemolysins or Hemotoxins.—Aside from 
the exogenous poisons mentioned above, mention has been frequently made 
of hemolytic poisons produced by some pathogenic and saprophytic bacteria, 
notably Bacillus tetani, B. welchii, staphylococci, and streptococci., Other 
micro-organisms may contain an endohemotoxin, as the pneumococcus, 
capable of converting hemoglobin into methemoglobin. 
All of these hemotoxins possess an affinity for erythrocytes and act 
upon them directly; antisera have been produced for some of them capable 
of neutralizing their effects, but it is not certain that this may be partly 
due to serum cholesterol. 
Some bacteria may produce agglutinins for human erythrocytes, among 
them being the staphylococcus, Bacillus typhosus, and B. pyocyaneus; 
these have not been studied as thoroughly as the hemotoxins. 
These hemolytic poisons are probably responsible in whole or part 
for the secondary anemias occurring in the infectious diseases and particu- 
larly those accompanied by septicemia; they may also produce hemo- 
globinuria. Connell and Holly7 believe that they consist of bacterial fats 
in definite colloidal states and further reference to them will be made in 
Chapter XX. 
1 Jour. Infect. Dis., 1920, 27, 86. 
2 Jour. Infect. Dis., 1907, 4, 277. 
3 Jour. Exper. Med., 1917, 26, 119. 
4 Jour. Infect. Dis., 1917, 21, 588. 
5 Jour. Infect. Dis., 1919, 24, 362. 
6 Jour. Immunology, 1920, 5, 265. 
7 Jour. Bacteriology, 1921, 6, 89. TOXINS OF MICROFUNGI AND HIGHER PLANTS 
TOXINS OF MICROFUNGI AND HIGHER PLANTS (PHYTOTOXINS) 
103 
As previously mentioned, the power of forming toxins is not confined 
to bacteria alone. It is well known that certain plants produce soluble 
toxins which may be operative in the production of disease, and it is highly 
probable that some of the pathogenic microfungi may do likewise. 
For example, the poisons of Rhus toxicodendron (poison ivy and oak) 
and Rhus venenata (sumac and dogwood) are well known to produce a 
form of dermatitis designated as dermatitis venenata. Some persons are 
extremely susceptible to these poisons suggesting hypersensitiveness or an 
anaphylaxis to the proteins of the poisons; this subject is discussed at greater 
length in Chapter XXVIII. 
The best known phy to toxins are as follows: Ricin (from the castor bean, 
Ricinus communis)-, crotin (from the seeds of Croton tiglium)-; abrin (from the 
seeds of Abrus precatorius); robin (from the leaves and bark of Robinia 
pseudoacacia); curcin (from the seeds of Jatropha curcus), and the pollens 
of a large variety of grasses and plants (ragweed, goldenrod, etc.). A large 
number of poisonous mushrooms are also known to contain poisons—phallin 
(from Amanita phalloides), having been especially studied by Robert and 
Ford and Abel. 
Relation to Infection and Immunity.—Ehrlich and his colleagues con- 
ducted a large part of the pioneer investigations in immunity with abrin, 
ricin, and crotin because they were more stable than bacterial toxins, pro- 
duced well-defined test-tube reactions of hemolysis and hemagglutination 
and proved antigenic for animals, enabling the production of immune sera 
which neutralized the poisons in a manner analogous to the neutralization 
of diphtheria toxin by antitoxin. These investigations focused attention 
upon these plant poisons; Ehrlich believed that they possess the same 
theoretic structure as bacterial exotoxins, namely, the presence of toxo- 
phore and haptophore groups. Since specific antitoxins may be prepared 
for some of them, they are regarded as true toxins similar to the exotoxins 
of bacteria. 
Of course, not all of the poisons to be obtained from plants can be classed 
as phytotoxins in this meaning of the term; some, like the saponin substances 
and various alkaloids, do not produce antitoxins when injected into animals. 
True phytotoxins, similar to bacterial exotoxins, are probably of protein 
nature and serve as antigens with the production of specific antitoxins when 
injected into animals subcutaneously or intravenously, or even by feeding. 
Action of Phytotoxins.—Just what role these plant toxins play in the 
production of disease is difficult to state. It is highly probable that some 
microfungi produce sufficient exotoxin to produce necrosis of epithelial 
cells and an inflammatory reaction. 
Whether or not the injurious effects of poison ivy and the pollens of 
certain plants and grasses are due to preformed toxins or to hypersensitive- 
ness to the proteins of these plants, is difficult to state; all available data 
indicates the latter, and only certain persons appear to suffer from contact 
with them. Dunbar1 claims to have secured toxins from many different 
pollens for which antitoxins could be prepared by the immunization of 
animals. 
Many of the phytotoxins produce agglutination and hemolysis in vitro 
and in vivo. According to Field2 the agglutinating function of ricin is 
separate and distinct from the toxic function. 
1 For full review of this subject see Glegg, Jour. Hygiene, 1904,4; Lefman, Zt. f. Hygiene 
1904, 47, 153; Wolff-Eisner, Deut. med. Wchnschr., 1906, 32, 138. 
2 Jour. Exper. Med., 1910, 12, 551. INFECTION 
104 
Flexner1 found that ricin and abrin produce histologic changes in animals 
similar to those caused by diphtheria toxin. Ricin apparently carries a 
toxin-destroying endothelial cell with the production of hemorrhages. 
Bunting2 has described severe changes in the bone-marrow. 
All of this indicates the close relationship of these toxins to pharma- 
cology. Bacteria which are plants are known to contain exotoxins for 
which antitoxins can be prepared, and endotoxins for which antitoxins 
cannot be prepared. The higher plants apparently may contain toxins 
of a protein nature similar to the bacterial exotoxins and other toxins or 
poisons similar to the endotoxins, which are frequently glucosids. 
Nature of Phytotoxins.—Some, and especially those from which anti- 
toxins may be prepared, resemble proteins and have long been referred to 
as “toxalbumins.” Jacoby was able to secure preparations of ricin and 
abrin that did not give protein reactions, and he regarded them as large 
molecular colloids, closely resembling the proteins with which they are 
associated, but still not giving the usual protein reactions. More recent 
work by Osborne, Mendel, and Harris,3 however, indicates that these toxins 
are proteins; at any rate, the toxic properties of ricin was found inseparably 
connected with the coaguble albumin of castor beans, and destructible by 
tryptic digestion. 
Other so-called phytotoxins have been largely identified as glucosids. 
Ford4 found the hemotoxin of Amanita phalloides, the phallin of Robert,5 
to be a glucosid, but yet capable of acting as an antigen with the produc- 
tion of an antiserum; the thermostabile Amanita toxin, on the other hand, 
is probably an alkaloid, and gives no reactions for either glucosids or pro- 
teins. Other mushroom poisons (Amanita muscaria and Helvella esculenta) 
and the poisons of Rhus toxicodendron and Rhus diversiloba are either 
glucosids or poisons classed as alkaloids. 
Toxins of Protozoa and Higher Animals (Zootoxins) 
While many Protozoa are known to produce disease our knowledge of 
the mechanisms involved is very incomplete, and largely because of the 
technical difficulties involved in studies bearing upon the production of 
toxic substances. 
It is probable that pathogenic Treponemata (syphilis; yaws) produce 
exotoxins; clinically, certain symptoms in acute syphilis are suggestive of a 
toxinemia. Toxic substances are also produced in amebiasis, malaria, and 
other protozoon infections, although the exact nature of the “toxins” is 
unknown. 
Whether or not the worms produce “toxins” is not definitely known; 
in infestations with tapeworms there is evidence of antibody production 
suggesting the absorption of antigenic substances. It is not improbable 
that a toxic substance is produced by Trichinella and that some of the 
symptoms in helminthiasis are due to toxic products including those of a 
hemolytic nature. Whether or not these substances are deserving of the 
designation “toxins” cannot be stated. 
The most important animal toxins (zootoxins) are those of the toad, 
spider, snake, scorpion, and bee. The most striking characteristic of these 
1 Jour. Exper. Med., 1897, 2, 197. 
2 Jour. Exper. Med., 1906, 8, 625. 
3 Amer. Jour. Physiol., 1905, 14, 259. 
4 Jour. Infect. Dis., 1906, 3, 191; ibid., 1906, 4, 434. 
5 St. Petersb. Med. Wchn., 1891, 16, 463 and 471. TOXINS OF PROTOZOA AND HIGHER ANIMALS 
105 
toxins is that an immunity against them can be established; in this respect 
they resemble true toxins. All are quite complex in structure and properties, 
and all are more or less hemotoxic. 
Snake Venoms.1—Medically, these are of particular interest. They 
were first thoroughly investigated by S. Weir Mitchell (1860) and Mitchell 
and Reichert (1883), and have aroused considerable attention because of 
their similarity to bacterial toxins and the aid their study has been in the 
elucidation of immunologic problems. 
Properties of Venom.—In 1883 Mitchell and Reichert described two 
poisonous proteins, constituents of venom, one of which seemed to be a 
globulin and the other a proteose or “peptone.” Faust2 believes that the 
poisons are not proteins, but glocosids free from nitrogen, and that they 
belong to the saponin group of hemotoxic agents. It may be that these 
glucosids are bound to proteins, and can be removed with the globulin in 
fractional separation, or that they may come down, at least in part, with 
the albumoses of the venom. 
Various enzymes have been found in venoms; e. g., proteases (Flexner 
and Noguchi) and lipases (Noguchi); the latter probably have a definite 
relation to many of the effects of venom intoxication, especially hemolysis 
and fatty degeneration of the tissues. 
The poisons, as a rule, produce both local and severe general disturb- 
ances, the rapidity of the onset of the symptoms and the prognosis in a 
given case depending largely on the situation of the bite. Most of these 
poisons exert their effect primarily upon the nervous and vascular systems, 
besides exhibiting other toxic properties. 
Nature of Venoms.—All snake venoms possess a hemolytic power, and 
venom hemolysis is one of the most interesting of biologic phenomena. 
Flexner and Noguchi3 have distinguished and classified the various ele- 
ments as hemotoxins, hemagglutinins, neurotoxins, leukotoxins, and endo- 
theliotoxins (hemorrhagin). The endotheliolytic action of the toxins is 
shown in the glomerular capillaries, where it causes hemorrhage and hema- 
turia (Pearce4). 
Cobra hemotoxin is especially characterized by its power of dissolving 
the corpuscles of certain species (man, dog, guinea-pig, rabbit) without 
the presence of serum. The explanation of this interesting phenomenon 
has excited extensive discussion. It is probable that the hemotoxin is in 
the nature of an amboceptor (Flexner and Noguchi), which is activated, 
in the absence of serum, by complementing substances (chiefly lecithin) 
present in the red cells, and in this manner producing hemolysis of these 
cells. In syphilis the quantity of red-cell lecithin is probably diminished 
after the primary stage, so that when using definite dilutions of venom that 
are known to hemolyze a certain quantity of normal erythrocytes, an ab- 
sence of hemolysis of the red corpuscles of a given patient would infer a 
decrease in complementing lecithin in these corpuscles and indicate the 
presence of syphilis. The technic of this reaction and its value as a diag- 
nostic procedure will be discussed further on under the head of Venom 
Hemolysis. 
1 See Faust: Die tierischen Gifte, Braunschweig, 1906; Noguchi: Carnegie Institution 
Publications, 1909, No. Ill; Calmette: Les venins, etc., Paris, Masson, 1907; Wells, Chem- 
ical Pathology, W. B. Saunders Co., 1920. 
2 Arch. exp. Path. u. Pharm., 1907, 56, 236; 1911, 64, 244. 
3 Jour. Exp. Med., 1903, 9, 257; Univ. of Penna. Med. Bull., 1902, 15, 345. 
3 Jour. Exp. Med., 1909, 11, 532. 106 
INFECTION 
ENDOTOXINS 
It is well known that many pathogenic bacteria produce such small 
amounts of exotoxin that their toxicity is not to be explained on this basis; 
this group includes the staphylococci, streptococci, pneumococci, meningo- 
cocci, gonococci, typhoid-colon group, and numerous other pathogenic 
and saprophytic micro-organisms. Pfeiffer originally taught that the 
toxicity of these bacteria was due to a preformed toxin which he called 
“endotoxin.” 
It is probable that microfungi and the pathogenic Protozoa, notably 
the Spirochetes and Treponemata, contain similar poisons. 
Definition.—Endotoxins are preformed toxic substances or toxins re- 
tained in the bodies of micro parasites until released by disintegrative processes. 
This definition is essentially that given by Pfeiffer, although many in- 
vestigators doubt the existence of preformed intracellular toxins, and much 
new light has been thrown upon the probable mechanism of disintegration 
of bacteria in the course of which these poisons are produced. 
Methods for Obtaining Endotoxins.—Endotoxins are obtained from 
bacteria by thorough disintegration of their bodies. A variety of methods 
have been proposed, some of which are objectionable because they permit 
the formation of protein digestive products which are commonly mistaken 
for preformed endotoxins. Experimentally endotoxins should be obtained 
with a minimum of heat and as rapidly as possible; the following methods 
have been used: 
1. By cultivating the micro-organism on solid or in liquid media, rapidly 
washing young cultures with cold isotonic saline solution by centrifuging, and 
suspending in saline at 37° C. for brief periods of time for total or partial 
autolysis. This is essentially one of the methods employed by Cole1 for 
demonstrating an endotoxin in pneumococci. Cole also produced these 
endotoxins by suspending pneumococci in dilute solutions of bile salts at 
37° C. for ten minutes, or for a half hour on ice. 
2. By suspending washed micro-organisms in saline solution or distilled 
water and alternately freezing and thawing. This is essentially the method 
employed by Rosenow2 for securing an endotoxin from penumococci. 
3. By freezing and grinding, as in the process of securing typhoid endo- 
toxin employed by Rowland and Macfayden,3 or by the grinding of dried 
bacteria as employed by Howlett.4 
Autolytic digestion by prolonged cultivation in broth or non-nutrient 
fluids cannot be recommended as time is thereby permitted for the activity 
of proteolytic enzymes liberated from dead or devitalized bacterial bodies. 
True or preformed endotoxins should be so designated only when these 
toxic substances are secured by rapid disintegration of washed bacterial cells, 
as in the methods employed by Cole; furthermore, the fluids should be thor- 
oughly centrifuged to remove bacterial bodies. 
The methods advocated by Friedberger and Vallardi,5 Neufeld and Dold,6 
and others, consisting of treating bacteria with immune serum and comple- 
ment outside of the body, is objectionable because it permits of the produc- 
tion of protein split products from the protoplasm of the bacterial cells; 
the same objection holds for the alcoholic potash method of Vaughan and 
]Jour. Exper. Med., 1912, 16, 644. 
2 Jour. Infect. Dis., 1912, 11, 235. 
3 Centralb. f. Bakt., orig., 1903, 34, 618. 
4 Proc. Roy. Soc. Med. (Path. Sect.), 1911. 
6 Ztschr. f. Immunitatsf., orig., 1910, 7, 94. 
8 Berl. klin. Wchn., 1911, xlviii, 55. ENDOTOXINS 
107 
Wheeler,1 and the watery soda method of Schmittenhelm and Weichardt. 
These methods have greatly complicated the subject and are discussed in 
a succeeding section of this chapter in relation to the role of proteins and 
protein split or digestive products in relation to disease. 
Properties and Nature of Endotoxins.—Their chemical nature and 
structure are unknown because of the impossibility of securing endotoxins 
in pure form and free from other bacterial substances and products. Koch’s 
old tuberculin has long been regarded as an endotoxin liberated during 
prolonged cultivation of the tubercle bacillus in broth. It appears to be a 
polypeptid, giving no biuret reaction, but being destroyed by pepsin and 
trypsin (Laevenstein and Pick). Pick regards tuberculin as a secretory 
toxin closely related to the true exotoxins. It is probable that some toxin 
is actually secreted into the culture medium and that the major portion, 
which is of a somewhat different nature, is intimately related to the proto- 
plasm of the bacterial cells. 
Endotoxins are generally more resistant to heat than exotoxins and 
do not deteriorate as rapidly upon standing. Rosenow found his pneumo- 
coccus endotoxin soluble in ether and destructible by heating at 60° C. 
for twenty minutes and by weak hydrochloric acid. He has suggested 
that the toxic substance is probably a base containing amino groups of 
nitrogen. 
Satisfactory antitoxins for endotoxins have not been produced, and this is 
an important point in differentiating between an exotoxin and an endotoxin 
of any particular micro-organism. Animals immunized against endotoxin 
develop substances in their serum that are bactericidal, bacteriotropic, 
and agglutinative to the bacteria from which the poisons were derived, 
but the serum itself is not antitoxic for the endotoxins. Therapeutic serums 
for use against infections caused by the endotoxin class of bacteria are 
largely bacteriolytic and bacteriotropic in action. The endotoxins of some 
bacteria, and particularly those of streptococci, seem to repel the leuko- 
cytes, or exert a negative chemotactic influence, which may effectually 
retard or entirely prevent phagocytosis; in this respect they resemble the 
aggressins of Bail. Immune serums owe a portion, at least, of their thera- 
peutic value to the power they possess of overcoming this influence and 
facilitating phagocytosis. These serums, however, have not proved of as 
much value as have the diphtheria and tetanus antitoxins in the treatment 
of the respective infections mentioned, and have proved a check to the 
progress of serum therapy. 
Action of Endotoxins.—The toxicity of endotoxins has been largely 
determined by intravenous injection in guinea-pigs and rabbits. Most 
observers have reported the development of symptoms and lesions resemb- 
ling acute or delayed anaphylaxis as dyspnea and coughing, partial emphys- 
ema and focal hemorrhages in the lungs, stomach, intestines, and epi- 
cardium. Gradual wasting with fever and death not infrequently follow 
during immunization of rabbits, goats, and horses. 
Aside from these effects it would appear that endotoxins are capable 
of arresting the phagocytic activity of leukocytes and other cells; in this 
regard they appear to be antiopsonic. This phase of the subject has been 
especially studied by Bail, who believes that these effects are due to separate 
products of bacterial activity designated as “aggressins.” Most investi- 
gators, however, regard these “aggressins” as endotoxins or toxic products 
of protein digestion or disintegration. A further discussion of these sub- 
stances follows: 
1 Jour. Infect. Dis., 1907, 4, 476. 108 
INFECTION 
In an attempt to explain certain observations of Koch to the effect that 
when a tuberculous animal is injected intraperitoneally with a fresh culture 
of tubercle bacilli it succumbs quickly to an acute attack of the disease, 
the resulting exudate being composed almost exclusively of lymphocytes, 
Bail1 has advanced the hypothesis that bacteria may secrete aggressins, 
or substances that aim to protect the micro-organism by either neutralizing the 
action of opsonins or directly repelling the body cells and preventing phago- 
cytosis. Bail found that if he removed a tuberculous exudate, sterilized it, 
and injected it into healthy animals, it had practically no effect. If tubercle 
bacilli were injected alone, lesions would develop in the usual number of 
weeks; but if sterile exudate and tubercle bacilli were injected together, 
death would follow in about twenty-four hours, indicating that the exudate 
contained a substance that acutely paralyzed the defensive forces of the 
animal, and thus greatly increased the virulence of the bacilli. That this 
effect was not the summation of endotoxins in the exudate plus living micro- 
organisms was shown by Bail, who found that when large quantities of 
exudate alone were injected no untoward effects resulted, whereas the in- 
jection of a small amount of exudate, plus a sublethal dose of bacteria, 
would regularly produce acute infection and death. Bail, therefore, con- 
cluded that the exudate contained a substance that allowed the bacilli to 
become more aggressive, and for this reason he called this hypothetic sub- 
stance “aggressin.” He assumes that in a tuberculous animal the tissues 
are permeated with the aggressin, and that when fluid collects in the body 
cavities after the injection of tubercle bacilli, this fluid contains large quan- 
tities of aggressin. This prevents migration and collection of polynuclear 
leukocytes, but not of lymphocytes, and hence allows the bacilli to develop 
rapidly, producing acute symptoms. On the other hand, when tubercle 
bacilli are injected into the peritoneal cavity of a healthy guinea-pig, poly- 
nuclear leukocytes which engulf the bacilli are attracted, thus inhibiting 
their rapid development, there being no aggressin to prevent phagocytosis. 
Similar results were obtained with other micro-organisms. Bail in- 
oculated cholera and typhoid bacilli into the pleural and peritoneal cavities 
of animals, and an acute local infection occurred. From the exudates so 
produced he removed the bacteria by centrifugalization, and completed 
the sterilization with antiseptics or with heat at 44° C. The clear fluid 
obtained was found to possess but mild toxic properties, and large amounts 
could be injected into animals of the same species without producing any 
marked effects; when, however, it was injected into an animal together 
with a sublethal dose of the particular micro-organism, an acute and fatal 
infection followed. Similar results were secured with the bacilli of dysentery, 
chicken cholera, pneumonia, and other diseases. Burgers and Hosch2 have 
described “aggressins” for dysentery and typhoid bacilli and staphylococci. 
Bail’s Classification of Bacteria.—Bail found that bacteria differed in 
their power of forming aggressins; he therefore used this principle in mak- 
ing a division of bacteria into three classes, according to their disease- 
producing power, as dependent largely upon whether or not the micro- 
organism can produce an aggressin that is active against the protective 
forces of the host, particularly against opsonins and leukocytes. 
1. Saprophytes, or those bacteria that, when injected even in large 
doses, do not produce any characteristic disease. 
Aggressins 
1 Wien. klin. Woch., 1905, 8, 14, 16, and 17; Berl. klin. Woch., 1905, 15; Zeit. f. Hyg., 
1905, i, 3; Arch. f. Hyg., 1905, 52, 272, and 411. 
2 Ztschr. f. Immunitatsf., orig., 1909, 2, 31. AGGRESSINS 
109 
2. True parasites, or those bacteria that, when injected even in the 
smallest amounts, will produce disease and death. These are truly virulent, 
and the number of bacteria increase so rapidly as to be demonstrable in 
every drop of blood and in all the organs. Examples of true parasites are 
the bacilli of anthrax and of chicken cholera, the tubercle bacillus for guinea- 
pigs, and the bacilli of the group of hemorrhagic septicemia for rabbits. 
3. Half or partial parasites are those bacteria the infectious nature of 
which depends upon the number of bacteria injected. The smaller the 
number, the milder the symptoms, until a dose is reached below which no 
disturbances are produced. Organisms of this class possess some virulence 
and toxicity, examples being the Bacillus typhosus and the Spirillum cholerae. 
It is to be remembered, however, that these effects are but relative, and 
dependent upon the organism, the species of animal, and the mode of in- 
fection. For example, the bacillus of anthrax is saprophytic for the frog 
and hen unless the temperature of these animals is brought to the body 
temperature of the human; a bacillus of the group of hemorrhagic septi- 
cemia of rabbits is saprophytic for human beings, a half parasite for the 
guinea-pig if injected subcutaneously, and a true parasite for the same 
animal if injected intraperitoneally. 
Nature of Aggressins.—The aggressins in inflammatory exudates are 
presumably substances capable of paralyzing the protective agencies of 
the body. Bail regards the aggressins as of the nature of endotoxins liber- 
ated from the bacteria as a result of bacteriolysis, and believes that they 
act by paralyzing the polynuclear leukocytes, thereby preventing phago- 
cytosis. In general, the production of these aggressins goes on more actively 
the greater the resistance to the bacteria; they are produced in greater 
quantities during the struggle between the bacteria and the body cells, 
although they may be produced artificially in the test-tube with large 
numbers of bacteria and a non-poisonous agent (serum or distilled water) 
which can disintegrate the cells. In this manner Wassermann and Citron 
have produced “artificial aggressins,” which act in the same general manner 
as the u natural aggressins” of Bail. 
By many the aggressins are regarded as endotoxins, and while they 
may possess the nature of endotoxic substances, it is to be remembered 
that there is no definite relation between the poisonous qualities of the 
aggressins and their power to increase the virulence of an infection. It 
is probable, as has been showrn by Wassermann and Citron, that patho- 
genic bacteria contain small amounts of natural aggressin. This aggressin 
may be regarded as a normal antibody of the bacterium against the defen- 
sive forces of the body cells of a host. During infection these aggressins 
or antibodies are naturally greatly increased, as the bacteria require more 
and more protection. Being contained to some extent within the bacterial 
cells, the antibodies are somewhat similar to endotoxins: while endotoxins 
may be regarded as offensive agents of bacteria, aggressins may be their 
defensive agents. The belief is in keeping with the hypothesis of Welch1 
and also of Walker,2 according to which it may be presumed that bacteria, 
as living cells, wdien so placed that they are exposed to the defensive forces 
of their host, are, under favorable conditions stimulated to produce reciprocal 
antibodies for their protection, and to generate them in increasing amounts 
as may be necessary. 
Bail regards the aggressins as new substances; as already stated others 
regard them as simple endotoxins; still others believe them to be free bac- 
1 Brit. Med. Jour., 1902, 2, 1105. 
2 Jour, of Path., 1902, 8, 34. 110 
INFECTION 
terial receptors, and that these receptors may combine with bacteriolytic 
amboceptors, producing, as it were, a deflection of the amboceptors, so 
that the bacteria themselves are not attacked, and thus continue to pro- 
liferate. The action of aggressins is not dependent upon the toxicity of 
the endotoxins, for the fluid containing them is devoid of toxic effects; 
at most, therefore, if they are of the nature of receptors, they possess no 
toxophorous portion. Gal1 believes that bacterial filtrates, extracts, and 
aggressins are only end-products of proteolysis; Pokschischewsky2 likewise 
regards the so-called aggressins as toxic protein substances (toxopeptids). 
d’Herelle3 has observed that the substance produced by bacteria in resistance 
to the dissolving action of a diastatic enzyme produced by an ultramicrobe 
or intestinal bacteriophage, described above, acts like an aggressin by 
paralyzing the phagocytic activities of leukocytes. 
Anti-aggressins may be produced experimentally by gradually im- 
munizing animals with sterile exudates, and this immunity may be trans- 
ferred passively from one animal to the other by inoculation of its immune 
serum. These antiaggressins are quite specific, and neutralize the aggres- 
sins in an exudate. Numokawa4 found the sera of rabbits immunized with 
“aggressins” largely bacteriotropic in activity; Haslam5 has successfully 
immunized guinea-pigs and calves with “blackleg aggressin” (edematous 
fluid of calves dying in one to three days after inoculation with a pure cul- 
ture of Bacillus chauveaui), and recommends immunization of calves with 
this fluid for the prevention of blackleg. 
BACTERIAL PROTEINS 
While in tetanus and diphtheria, exotoxins are the chief and almost 
sole pathogenic agents, in practically all other bacterial infections the 
proteins of the micro-organisms are to be considered in addition to exo- 
toxins and endotoxins, as contributing in an important manner to the pro- 
duction of local inflammation and toxemia. The same is probably also 
true in infections with microfungi, Protozoa, and other animal parasites. 
Nature of Bacterial Proteins.—Vaughan6 and his co-workers who have 
studied bacterial proteins quite extensively, regard bacteria as essentially 
particulate, specific proteins. They have not been able to demonstrate 
the presence of cellulose and carbohydrates; fats and waxes that may be 
present are somewhat secondary and less essential constituents or stored 
food material. The sum total of the work of these observers would indicate 
that the greater part of bacteria are made up of true proteins, especially 
nucleoproteins or glyconucleoproteins, and although they may be simple 
in structure, they are chemically complex—quite as much so as many of 
the tissues of the higher plants and animals. 
When bacterial cellular substances are split up with mineral acids or 
alkalies they yield ammonia, mono-amino- and diamino-nitrogen, one or 
more carbohydrate groups, and humin substances. These protein sub- 
stances are the same as those obtained by the hydrolysis of vegetable and 
animal proteins. 
By digestion with dilute acids or alkalies, especially the latter, in the 
form of a 2 per cent, solution of sodium hydroxid in absolute alcohol, a sol- 
1 Ztschr. f. Immunitatsf., orig., 1912, 14, 685. 
2Ztschr. f. Immunitatsf., orig., 1912, 15, 186. 
3 Compt. rend. Soc. d. biol., 1920, 83, 97; ibid., 1921, 84, 339, 538. 
4 Ztschr. f. Immunitatsf., orig., 1909, 3, 172. 
5 Jour. Immunology, 1920, 5, 539. 
6 Protein Split Products, Lea & Febiger, 1913. BACTERIAL PROTEINS 
111 
uble split product is obtained that resembles in some respects the prota- 
mins, although they do not all give a satisfactory biuret reaction. This 
product is highly toxic, but shows no specificity in its action, being the 
same whether derived from pathogenic or from non-pathogenic bacteria, 
or from egg albumen or other protein substance. All that is definitely known 
regarding it is that it is toxic, protein in nature, but simpler in structure 
than the complex proteins of the bacterial cells, themselves. 
Massive cultures of colon, typhoid, pneumonia, and diphtheria micro-organisms are 
grown in special large tanks containing agar; anthrax is grown in Roux flasks, and tubercle 
bacilli in glycerin beef-tea cultures. After removal of the growths the bacterial cellular sub- 
stances are washed once or twice with sterile salt solution by decantation, and then repeatedly 
washed with alcohol, beginning with 50 per cent, and increasing the strength to 95 per cent. 
The substance is then placed in large Soxhlet flasks and extracted first for one or two 
days with absolute alcohol, and then for three or four days with ether. These extractions 
should be thorough in order to remove all traces of fats and wraxes. 
After extraction the cellular substance is ground, first in porcelain, then in agate mortars, 
and passed through the finest meshed sieves to remove bits of agar. The person grinding 
the cellular substance should wear a mask in order to protect himself against poisoning. 
Vaughan reports that, despite this precaution, several workers have been acutely poisoned, 
especially with the typhoid bacillus. Of course, there is no danger of infection, as the bac- 
teria are killed during the treatment. If the finely ground cellular substance, in the form of 
an impalpable powder, is kept in wide-mouthed bottles in a dark place, it will retain its 
toxicity for years. This powder constitutes the bacterial protein substance, which may be 
split up by various means. Vaughan found digestion with 2 per cent, caustic soda in absolute 
alcohol especially satisfactory for extracting the poisonous group from bacterial or any other 
protein. 
A weighed portion of the protein, prepared as above, is placed in a flask, covered with 
from fifteen to twenty-five times its weight of absolute alcohol in which 2 per cent, of sodium 
hydroxid has been dissolved. The flask, fitted with a reflux condenser, is heated on the water- 
bath for one hour, where it is allowed to cool and the insoluble portion collected on a filter. 
After thorough draining the insoluble part is returned to the flask and the extraction re- 
peated. It has been found that three extractions are necessary in order to split off all the 
poisonous group. The temperature of these extractions is 78° C., the temperature of boiling 
absolute alcohol. By this method the protein is split into two portions, one of which is 
soluble in absolute alcohol and is poisonous, while the other is insoluble in absolute alcohol 
and is not poisonous (Vaughan). 
The insoluble and non-poisonous portion of the cellular proteins shows 
most of the color reactions for proteins, and contains all the carbohydrate 
of the unsplit molecule and most of the phosphorus. 
Action 'of Bacterial Proteins.—The effects produced by bacterial pro- 
teins are not specific; the protein substance of non-pathogenic bacteria 
and, indeed, many proteins derived from vegetable and animal sources 
have equally marked pyogenic properties. All foreign proteins introduced 
into the circulation of animals are more or less toxic, and the toxic effects 
of all bacterial proteins are, in general, quite similar and non-specific. 
In practically all bacterial bodies after removal of toxins and endo- 
toxins a certain protein residue remains which, when injected into animals, 
is able to produce various grades of inflammatory reaction leading to tissue 
necrosis and abscess formation. This substance was first thoroughly studied 
by Buchner, who named it bacterial protein, and regarded it as identical in 
all bacteria and having no specific toxic action, but characterized in general 
by its power of exerting a positive chemotactic influence on leukocytes, 
and thereby favoring the formation of pus. For example, in the develop- 
ment of an ordinary staphylococcus abscess it is probable that the proteins 
of the cocci, aside from their toxins, aid in producing tissue necrosis and 
in attracting leukocytes to the infected area. Similarly, an extract of dead 
tubercle bacilli may produce a tuberculoma or the tissue changes incident 
to tuberculosis, differing, however, from true tubercle in that they do not 112 
INFECTION 
contain living bacilli and consequently are not infectious. When cultures 
of diphtheria bacilli are filtered and the residue washed, it is found that 
extracts of the bacterial substances or the bodies of the dead bacilli them- 
selves are quite free from the typical toxin, but the bacterial substances 
or the proteins isolated from them, when injected into the subcutaneous 
tissues of animals, are found to produce a strong inflammatory reaction 
and necrosis of the tissue cells. Bacterial proteins are apparently much 
more toxic than ordinary proteins, as shown by Schittenhelm and Weichardt,1 
and few produce as much inflammatory reaction. 
Bacterial protein substances may be responsible for certain minor ana- 
phylactic reactions, as has been observed occasionally in the administration 
of ordinary bacterial vaccines. They may bear an important relation to 
the development of the state of hypersensitiveness of a tuberculous person 
in the course of a series of tuberculin injections. 
In addition to these effects produced by the proteins of bacterial cells, 
it is highly probable that split cleavage proteins are produced through the 
agency of proteolytic enzymes from devitalized bacteria, leukocytes, and 
other body cells. Vaughan believes that “ferments” are produced during 
infection which produce this digestion or cleavage of bacterial protein and 
that a toxic fraction results similar to that secured with alcoholic potash 
described above, responsible for the local inflammation, fever, and injury 
to the nervous system: 
Theory of Vaughan.—According to Vaughan and his co-workers, all 
true proteins contain a common and non-specific poisonous group. This 
group may be regarded as the central or keystone portion of every protein 
molecule, with secondary and possibly tertiary subgroups, in which the 
specific property of different proteins is inherent. When the main or pri- 
mary group is detached from its subsidiary group, it manifests its poison- 
ous action by the avidity with which it attacks the secondary group of 
other proteins. These are detached from their normal positions, and conse- 
quently deprive the living protein of its power of functionating normally. 
When proteins are split, the chemical nucleus or non-specific toxic portion 
is more or less completely set free, and its toxicity varies according to the 
thoroughness with which the secondary groups have been removed. 
The pathogenicity of a bacterium is determined not by its capability of 
forming a poison, but by the ability it possesses to grow and multiply in 
the animal body. When, during an infection, a pathogenic micro-organism 
reaches the deeper tissues, it is not immediately killed by the defensive 
ferments of the host, but continues to grow and multiply, throwing out a 
ferment that feeds upon the native proteins of the body cells, tearing them 
down and building up a specific bacterial protein that may select a certain 
point of predilection in which it is most prone to accumulate. Thus, the 
typhoid bacillus accumulates in the adenoid tissue of Peyer’s patches on 
the intestine, the spleen, and the mesenteric glands; the pneumococcus 
tends to lodge in the lungs; the smallpox virus selects the skin, etc. 
The bacterial toxins and viruses, as, e. g., diphtheria toxin and the virus 
of smallpox, are regarded as ferments of protein nature, capable of attack- 
ing native body protein and building up a specific foreign protein. This 
foreign bacterial protein is formed during the period of incubation of dis- 
ease when there is no effective resistance on the part of the body cells to 
its growth and multiplication. During this time the infected person is 
not ill, so that the foreign protein in itself cannot be toxic, and the body 
■cells are busy preparing and elaborating a new and specific ferment that 
1 Miinch. med. Wchn., 1911, 58, 841. BACTERIAL PROTEINS 
113 
will digest and destroy the foreign protein. When this new ferment be- 
comes active, the first symptoms of disease appear, and the active stage 
of the disease marks the period over which the parenteral digestion of the 
foreign protein extends. These specific ferments split up the foreign protein 
and liberate the toxic portion or the protein poison; this poison is not a 
toxin and is not specific, but occurs commonly in all proteins. 
The characteristic symptoms and lesions caused by the various infec- 
tious processes are determined largely by the location of the foreign protein. 
The poison elaborated is the same in all infectious diseases, and it is the 
location of the infection, rather than the exact nature of the infecting agent, 
which gives rise to the more or less characteristic symptoms and lesions 
of the several infectious diseases. 
Death may be produced by the too rapid breaking-up of the foreign 
protein, and the consequent liberation of a fatal dose of the protein poison- 
or it may result from a lesion induced by the products of this disruption, 
such as perforation of the intestine and hemorrhage in typhoid fever, or 
it may follow from chronic intoxication and consequent exhaustion. If 
recovery takes place, the individual enjoys an immunity of variable dura- 
tion, owing to the presence of specific ferments capable of destroying the 
particular substrata if infection should occur. 
It is this power of body cells, when permeated by a foreign protein, to 
elaborate a specific antiferment by which the protein is destroyed that, 
in the opinion of Vaughan, forms the basis of a correct understanding of 
infection and immunity. 
Friedberger’s Theory.—A similar theory has been advanced by Fried- 
berger and his colleagues1 to explain the production of toxic substances 
from bacteria, which they believe are similar to endotoxins and responsible 
for anaphylaxis. According to these investigators bacteria are broken 
down by means of bacteriolytic amboceptors and complement with the 
production of a poison called “anaphylatoxin,” held responsible for the 
general reaction of local inflammation, fever, injury to the central nervous 
system, etc. All bacteria are supposed to furnish the matrix for such a 
poison; only the antibodies are regarded as specific. Whenever antibody, 
its antigen and complement meet cleavage occurs and the poison is pro- 
duced; as proteolysis continues the split products become non-toxic. This 
theory is essentially similar to that previously described by Vaughan; 
Friedberger regards the digestive processes due to antibodies and comple- 
ment, while Vaughan calls the same substances “ferments.” 
Similar views are held by Thiele and Embleton,2 who believe that endo- 
toxins are not preformed, but produced in the body by protein cleavage 
processes by complement and antibodies or autolytic ferments. 
These investigations by Vaughan, Friedberger, Thiele, and Embleton 
resulted in the construction of a very simple explanation for the mechanism 
of the production of diseases by bacteria. Further investigations, however, 
have considerably weakened their original position and attractiveness. 
For example, Keysser and Wassermann,3 Bordet,4 Nathan,5 Jobling and 
Peterson,6 and others have shown that the protein poison of Vaughan, 
corresponding to the “anaphylotoxins” of Friedberger and the “endotoxins” 
1 Ztschr. f. Immunitatsf., orig., 1910, 6, 179, 299; ibid., 1910, 7, 94, 665, 748. 
2 Ztschr. f. Immunitatsf., orig., 1913, 19, 643, 666. 
3 Ztschr. f. Hyg., 1911, 68, 535. 
4 Compt. rend. d. 1. Soc. d. Biol., 1913, 74, 1213. 
5 Ztschr. f. Immunitatsf., orig., 1913, 74, 225. 
6 Jour. Exper. Med., 1914, 19, 485. 114 
INFECTION 
of Thiele and Embleton, may be produced in vitro without the presence 
of bacterial protein by substituting such substances as kaolin, agar, and 
chloroform. From these investigations it would appear that the protein 
poison is not derived from cleavage of the bacteria, but from the blood 
constituents. Investigations of Novy and DeKtuiff1 have confirmed this 
view and have shown that the important factor in the agar experiments 
is the physical state of the agar and that ferment action is probably not at 
all involved. 
As shown by Moldovan,2 Doerr,3 Slatineau and Ciuca,4 and others, 
and confirmed by DeKruiff, even blood and serum can be rendered toxic 
in a manner similar to the toxicity of blood treated with agar and peptone, 
by withdrawal and transfusion in the preclot period or after defibrination, 
poison production, and fibrin formation occurring in the preclot stage. 
These studies have reduced the importance of bacterial protein as a 
source of toxic substances; the specific “ferments” mentioned by Vaughan 
have not been satisfactorily demonstrated and there is no conclusive evi- 
dence that bacterial proteins are digested by amboceptors and complement 
as described by Friedberger, Thiele, and Embleton. On the contrary, the 
researches mentioned above indicate a different role for bacterial protein, 
namely, that toxic substances are derived from the blood constituents, 
the changes being brought about by physical or chemical changes in which 
bacterial protein may play a role which DeKruiff has spoken of as analogous 
to a catalyst, but otherwise inert. 
Toxicity of Digested Bacterial Protein; Autolysis.—However, it is highly 
probable that during infection the protein of dead bacteria may undergo 
a process of digestion by means of their own proteolytic enzymes (auto- 
lysis), or by enzymes derived from leukocytes, fixed tissue cells, and plasma. 
It is also probable that digestion of bacteria and other microparasites may 
release preformed endotoxins. This subject is presented in greater detail 
in the chapter on Ferments and Antiferments. 
Unquestionably these cleavage bacterial proteins are toxic, reducing 
the coagulation time of the blood, producing fever, leukocytosis, and other 
symptoms described by Vaughan as “protein fever”; the amount produced 
from bacteria, however, must be small and constitutes an objection to accept- 
ing them as the sole or, at least, very important factors in the production 
of toxemia as maintained by Vaughan, Friedberger, Thiele, and Embleton. 
Briefly summarizing the important relation of bacterial protein to in- 
fection and toxemia it may be stated: (1) Whole bacteria and particularly 
their nucleo proteins are apparently pyogenic; (2) digestion of dead bac- 
terial protein by proteolytic enzymes from bacteria, leukocytes, fixed tissue 
cells, and plasma results in the production of small amounts of toxic cleavage 
products, and (3) bacteria or their protein constituents may bring about 
certain physical or chemical changes resulting in the digestion of blood 
constituents and the production of a protein poison capable of exciting 
local inflammation and general toxemia. 
It is highly probable that similar changes occur in infections with micro- 
fungi and Protozoa; all of the investigations referred to above were con- 
ducted with bacteria. 
1 Jour. Infect. Dis., 1917, 20, 449. 
2 Deut. med. Wchnschr., 1910, 36, 2422. 
3 Wien. klin. Wchnschr., 1912, 25, 331, 339. 
4 Compt. rend. d. 1. Soc. d. Biol., 1913, 74, 631. TOXIC EXUDATES AND PTOMAINS 
115 
TOXIC EXUDATES AND PTOMAINS 
In addition to the effects produced by bacterial proteins, and especially 
nucleoprotein, in relation to the local inflammatory reactions and toxemia 
of bacterial infections as discussed in the preceding paragraphs, it is highly 
probable that proteolytic enzymes from bacteria, leukocytes, fixed tissue 
cells, and plasma become active in inflammatory exudates under certain 
conditions, and produce toxic protein cleavage products by digestion of the 
dead or devitalized constituents of inflammatory exudates and fixed tissue 
cells. 
It is hardly necessary for me to review here the numerous studies that 
have been made with the enzymes of bacteria, leukocytes, and fixed body 
cells1; proteases have been found in all which, under certain circumstances, 
become active when the restraining activity of antienzymes is checked. 
Substrates for these enzymes are furnished by dead bacteria as previously 
discussed, and more importantly, by dead constituents of inflammatory 
exudates including dead or devitalized fixed body cells of the part infected. 
Proteoses and peptone appear in the early stages to be followed by leucin, 
tyrosin, tryptophan, phenols, skatole, indole, aromatic oxy-acids, mercaptan, 
etc. 
These protein cleavage products are known to be extremely toxic for 
experimental animals and it is reasonable to believe that in certain infec- 
tions some may be absorbed and contribute important elements to the 
production of toxemia. This would appear to be especially true in exten- 
sive pyogenic infections with large collections of pus as in carbuncle, em- 
pyema, suppurative meningitis, burns, extensive suppurating wounds, etc. 
Furthermore, these toxic substances may actually aid in the production 
of local inflammatory changes although, fortunately, they also appear to 
possess bactericidal properties and may aid in the sterilization of pus and 
localized infections in general, thereby aiding in breaking up a vicious cycle 
of events. 
Ptomains.—It was at one time believed that the symptoms of many 
diseases were due to the absorption of soluble basic nitrogenous substances 
produced by bacterial action upon various albumins, these toxic, alkaloid- 
like substances being known as ptomains. It was soon found, however, 
that the ptomains produced by pathogenic bacteria were insufficient of 
themselves to cause the symptoms and lesions characteristic of the respec- 
tive micro-organisms; that they were in general less toxic than the cultures 
themselves; that the majority of ptomains are not very poisonous; and 
that they are not specific, since equally potent ptomains are produced by 
non-pathogenic bacteria. This lack of specificity is in sharp contrast to 
the toxins. No matter upon what medium a true toxin producer is grown, 
the toxin is qualitatively the same, whereas the nature and toxicity of 
ptomains depend upon the micro-organisms, the culture-medium used, the 
duration of growth, and the quantity of oxygen furnished. The same micro- 
organism, when grown on different media or under different conditions, 
may produce totally different ptomains. 
Ptomains may, however, produce disease, and even death, when they 
are ingested with food that has undergone bacterial decomposition. In 
most instances of meat poisoning, however, which are frequently ascribed 
to the presence of ptomains, a specific micro-organism, the Bacillus botulinus, 
or a member of the B. enteritidis group of Gartner, is usually responsible. 
1 For an excellent review see Wells: Chemical Pathology, W. B. Saunders Co., 4th ed., 
1920, 48-127. 116 
INFECTION 
The commonest sources of ptomain poisoning are improperly preserved 
meats, fish, sausages, cheese, ice-cream, and milk. This subject received 
full consideration in Vaughan and Novy’s “Cellular Toxins.” 
A number of ptomains are known, and of some the exact chemical con- 
stitution has been established. Brieger has separated one from decompos- 
ing flesh and cholera cultures, called cadaverin, which Ladenburg has 
shown to be pentamethylendiamin, and prepared synthetically. Muscarin, 
isolated by Schmiedberg and Brieger, and tyrotoxicon, isolated by Vaughan, 
are also well known. 
In the isolation of bacterial ptomains Breiger’s method is generally 
employed, which consists of acidulating large amounts of culture with 
hydrochloric acid, boiling, filtering, evaporating the filtrate to a syrupy 
consistency, dissolving in 96 per cent, alcohol, and precipitating and puri- 
fying by means of an alcoholic solution of mercuric chlorid. 
Besides occurring in food poisoning, ptomains may be formed as the 
result of putrefactive processes going on in abscesses, gangrenous areas, 
and within the gastro-intestinal canal, and enough of these may be absorbed 
to produce symptoms of intoxication. Under these conditions it is pos- 
sible for bacteria to produce ptomains that may be absorbed and produce 
symptoms of intoxication without the bacteria themselves actually gain- 
ing entrance to the tissues and, therefore, not constituting, according to 
our definition, a true infection. Pernicious anemia, chlorosis, and allied 
conditions have been ascribed to the absorption of such ptomains from the 
intestinal canal. Obviously it is difficult or impossible to always differ- 
entiate between bacterial toxins and bacterial ptomains, or the products 
of protein decomposition dependent upon bacterial activity, and we can 
but admit the possibility of the production and absorption of both bacterial 
toxins and ptomains under certain pathologic conditions. Most ptomains 
probably are produced as the result of decomposition of the dead protein medium 
upon which the bacteria grow, and to a lesser extent by the destruction of the 
bacterial cells themselves. It is extremely doubtful if ptomains are produced 
in important quantities by pathogenic bacteria infecting living tissues. 
In former years the theory as to the influence of mechanical blocking 
of vessels with masses of bacteria was regarded with much favor in the 
etiology of certain infections, particularly anthrax. At the present time 
this factor has not the same importance, for while it is true that bacterial 
emboli may occasion harm by blocking important vessels, further researches 
have shown that it is doubtful if any pathogenic micro-organisms are totally 
devoid of toxic action, and that their toxins are largely responsible for the 
tissue changes and symptoms of infections. 
It can readily be understood that emboli of micro-organisms may pro- 
duce metastatic lesions; thus, when staphylococci are injected into the ear 
vein of a rabbit they produce abscesses in the kidney and heart, and masses 
of bacteria from an ulcerative endocarditis when carried to different por- 
tions of the body will cause abscess formation; but the question in hand, 
however, deals with the effects of mechanical blocking itself. 
Investigations with anthrax bacilli have shown that they are remark- 
ably free from soluble toxins and endotoxins, although the local lesion 
develops so rapidly and becomes so quickly necrotic as to suggest very 
strongly the action of some local toxic substance. Cases of human anthrax 
seldom prove fatal if the lesion is removed and the blood-stream remains 
MECHANICAL ACTION OF BACTERIA MECHANICAL ACTION OF BACTERIA 
117 
free from the bacilli. Vaughan has shown that anthrax protein possesses 
toxic qualities, and since the majority of fatal cases of anthrax show enor- 
mous numbers of bacilli in the blood-stream and internal organs, it may 
be that this bacteremia produces an accumulation of toxins which is greatly 
augmented when the body cells of the host have produced an antiferment 
that splits up the protein of the bacilli, the combined toxic substances 
being responsible for the severe symptoms and death. 
With protozoan disease, the possibility of serious symptoms following 
blocking of vessels is far greater and, indeed, the cerebral symptoms of 
malignant malaria and sleeping sickness may be due in part to the block- 
ing of small, but physiologically important, vessels with masses of plas- 
modia and Trypanosoma gambiensi, together with the absorption of toxic 
agents and the products of disintegration. Thus Bass, who has successfully 
cultivated the malarial plasmodium outside of the body, believes that the 
parasites, after attaining sufficient size, lodge in the capillaries of the body, 
especially where the blood-current is weakest, and where slight obstruction 
occurs as the result of the protruding inward of nuclei of the endothelial 
cells. Here they remain and develop until segmentation takes place. In 
the meantime other red corpuscles are forced against them, and if the open- 
ing in the infected cell is in a favorable location, one or more merozoites 
pass directly into another cell; if it is not, the merozoites are discharged 
into the blood-stream and are speedily killed. Part III 
CHAPTER VII 
IMMUNITY.—THEORIES OF IMMUNITY 
In the preceding chapter on the mechanism of infection and the produc- 
tion of an infectious disease the statement was frequently made that the 
microparasites of disease are required to overcome the defensive forces 
of a host which are ever on guard to protect the organism against parasitic 
invasion and infection. Certain of these defenses are natural to the host 
and in a great majority of instances suffice to protect the body against 
invasion and infection with bacteria, animal parasites, and various in- 
animate and injurious substances. When, however, these natural defenses 
are broken down and infection has occurred, the body cells are not usually 
rendered powerless and completely overcome, for the products of infection 
serve as a stimulus to the body cells, calling forth renewed cellular activity 
and the production of various specific defensive weapons, termed anti- 
bodies, which maintain an incessant struggle against the invading patho- 
genic agents in an effort to rid the body of them and to neutralize their 
products. 
Just as microparasites have various offensive weapons, consisting chiefly 
of their toxins, so, in like manner, the defensive forces of the host are numer- 
ous and even more complex. If the toxin of a micro-organism is its chief 
pathogenic weapon, as, e. g., the soluble and extracellular toxin of the diph- 
theria or the tetanus bacillus, then the body cells produce an antitoxin 
as their chief defensive force; if the offensive weapon is largely in the nature 
of an endotoxin, as, for example, the endotoxins of the typhoid or the cholera 
bacillus, then a chief antibody is in the nature of a bacteriolysin. In other 
infections, especially those due to the pyogenic cocci, certain of the body 
cells, and chiefly the polynuclear leukocytes, are observed in the tissues 
to have engulfed the invaders bodily (phagocytes) in an endeavor to digest 
them and neutralize their products. In addition to these chief antibodies, 
there are others that appear to aid them in their work. 
That one attack of many of the infectious diseases may protect the in- 
dividual against subsequent attacks, or at least render subsequent attacks 
mild and harmless, is well known. In India and the East for centuries 
practical advantage has been taken of this observation in the management 
of smallpox. In order to protect persons against a severe attack of variola 
they were deliberately brought in contact with a person suffering with a 
particularly mild form, in the hope that, by inducing a mild attack of short 
duration, they would thus obtain protection against the severe, disfiguring, 
and fatal form of the disease. 
The practice, however, was not without danger to the individual and to 
the public at large, as the induced disease would at times become malig- 
nant, and constitute a focus of infection for an entire community. When 
Edward Jenner discovered that inoculation with cowpox virus could not 
produce smallpox, but would, nevertheless, stimulate the production of 
118 DEFINITION 
119 
specific antibodies and confer immunity against it, an enormous forward 
stride was taken that has since proved a priceless boon in helping to rid 
the world of the dreadful scourge of smallpox. 
The object of all these procedures has been to secure a resistance or im- 
munity to smallpox, either by inducing a mild form of the disease or by 
protecting the individual by means of inoculation with a virus that has 
been so changed in its passage through a cow as to render it unable to pro- 
duce smallpox, but yet is capable of stimulating the body cells to produce 
antibodies that will neutralize the effect of the true virus. This induced 
resistance to a given infection constitutes immunity or resistance, and since 
the body was purposely inoculated and the body cells rendered active in 
producing the antibodies, this form of resistance is known as active acquired 
immunity. 
Many persons recover from an infection that may have been unusually 
severe, not because the infecting agent became exhausted or died for want 
of pabulum, but because it had been gradually worsted in the battle with 
the defensive forces of the host. In many such instances the host is now 
immune to this infection for a longer or shorter time, because the body 
cells have been so profoundly impressed that they continue generating 
defensive weapons or antibodies for some time after the last vestige of the 
infecting agent has disappeared. Or, on the other hand, the quantity of 
antibodies may be so great that they may persist for varying periods of 
time, even for the remainder of life, ever on guard, and ready to overwhelm 
their specific enemy should it ever again gain access to the tissues. 
Here, then, arises the question concerning the mechanism of recovery 
from an infection, and since this is so intimately concerned with the general 
subject of resistance to disease, it is considered under the general head of 
immunity. * 
Even superficial observation shows that not all persons are equally 
susceptible to a given disease, and during the course of epidemics it will be 
seen that some individuals, although freely exposed, escape infection. Cer- 
tain species of animals may likewise display a uniform resistance to an 
infection that will readily enough attack another species of the same general 
family. It has been 4emonstrated experimentally that a certain patho- 
genic bacterium will produce a severe infection in one species of animals 
and not in another. It may frequently be noticed that even though an 
infection occurs, it is readily overcome by the natural resources of the host, 
the latter escaping with slight or no symptoms of disease. In other words, 
certain persons and animals apparently possess a natural resistance or im- 
munity to disease, wrhich may be general, non-specific, or due to specific 
antibodies, this type of immunity being frequently relative and seldom 
absolute. 
Definition.—Immunity, therefore, in a broad sense, is the effective re- 
sistance of the organism against any deleterious influence; in the usual and 
more restricted meaning the term is applied to resistance against infection 
with vegetable and animal parasites and their products, which are pathogenic 
for other animals of the same or of different species. 
It should be remembered that immunity means not only the ability 
to resist an infection or successful invasion of the tissues by micropara- 
sites, but also the continual resistance offered as long as the infection lasts; 
that is, immunity implies not only resistance to the onset of infection, 
but also to the course and progress of the resulting infectious disease. The 
science of immunity has, therefore, for its object the study of the mechan- 
ism of resistance to and recovery from an infection. 120 
IMMUNITY—THEORIES OF IMMUNITY 
HISTORIC 
The development of the science of immunity forms one of the most in- 
teresting chapters in the history of medicine. Even in ancient history we 
can trace the conception of our modern ideas on immunization. 
Hippocrates taught that the factor that causes a disease is also capable 
of curing it—practically the same theory as the more modern homeo- 
pathic doctrine of “similia similibus curantur.” Pliny the Elder recom- 
mended the livers of mad dogs as a cure for hydrophobia, thus coming 
very near to the basis of the Pasteur discovery. As was pointed out by 
Elizabeth Fraser, the same idea is expounded in the mythologic tale of 
Telephus, who cured his wound by applying rust from the swrord which 
inflicted it, and in the story of Mithridates, King of Pontus (B. C. 120), 
who immunized himself against poisons by drinking the blood of ducks 
that had been treated with the corresponding toxic substances. 
Immunization against various venoms has been practised by many of 
the savage tribes of Africa since earliest times. Mention has previously 
been made of the method of preventive inoculation against smallpox prac- 
tised in Asia and other Oriental countries for several centuries by exposing 
the subjects to mild cases of the disease. 
A very definite step in progress must ever be associated with the name 
of Edward Jenner, who first demonstrated experimentally, and in a scientific 
manner, that cowpox conveyed to man protected him against smallpox. 
Jenner was not the first person to make this observation, as many of the 
Gloucestershire farmers knew that cowpox protected them against small- 
pox; nor was he the first deliberately to inoculate persons with cowpox 
virus, as this method had been practised sporadically before his time. Jenner 
was, however, the first medical man to give the matter serious thought and 
consideration, and to test the method as thoroughly and scientifically as 
it was possible to do at that period. Thus, he inoculated with smallpox 
virus those whom he had previously vaccinated with cowpox virus, and 
found them immune to smallpox. These experiments courageously 
repeated, until a great truth was established, which has resulted in almost 
completely eradicating the disease from those countries or communities 
where vaccination is thoroughly carried out. As was to be expected, Jenner 
met with considerable opposition, and this is readily understood when it is 
remembered that even today—one hundred and eighteen years later— 
there are those who refuse to accept, or are unable to realize, the great 
benefits of this pioneer work. Jenner could not explain his results; he 
maintained that he was dealing with a modified form of smallpox. We of 
today have no better means of establishing the truth of the efficiency of 
cowpox vaccination, nor have we improved any on his method. The specific 
germ of smallpox is still undiscovered, and we must agree with Jenner that 
cowpox is probably a modified form of smallpox and practically harmless, 
the virus of cowpox being the virus of smallpox modified, attenuated, and 
rendered practically innocuous by passage through a lower animal. 
Nothing further of importance was accomplished during the following 
eighty years, until the next and even greater epoch ushered in the dis- 
coveries in bacteriology and the first immunization by Pasteur based on 
scientific reasoning. The chickens around Paris were being destroyed by 
a virulent intestinal infection, and Pasteur first isolated the causative 
micro-organism, a minute bacillus, which he found was capable of produc- 
ing the disease experimentally in healthy chickens. Quite by accident, 
so it seemed, he discovered that cultures of this bacillus could by pro- HISTORIC 
121 
longed cultivation be attenuated, for when these cultures were inoculated 
into chickens, the fowls did not die or suffer any ill consequences; further, 
and what was of the utmost importance, when these same chickens were 
inoculated with virulent cultures, they were found to be immune to chicken 
cholera. Here, then, was the key to active immunization in the prevention 
of disease, and Pasteur possessed the genius to realize the full significance 
of his discovery. 
Armed with this knowledge Pasteur and his assistants, Roux and 
Chamberland, next studied anthrax, an infectious disease of cattle that 
was causing a great annual loss to the farmers of France, and the bacillus 
of which was among the first pathogenic micro-organisms to be discovered. 
It was found that prolonged cultivation of this bacillus was insufficient 
to attenuate the cultures, as the spores were highly resistant and retained 
their pathogenicity under extreme circumstances and over prolonged periods 
of time. 
In 1880 Touissant published a method of attenuating the bacilli by 
heating the blood of an infected animal to 55° C. for a few minutes; later, 
Chauveau secured similar results by heating fresh cultures for a few minutes 
at 80° C. Both methods were uncertain, and neither safe nor practical. 
After prolonged experimentation Pasteur found that cultivating the bacilli 
at the relatively high temperature of from 42° to 43° C. resulted in gradual 
attentuation, and if this cultivation was continued, the cultures were robbed 
entirely of their disease-producing power. Further, subcultures of these 
growths when kept at 37° C. did not regain their original virulence, but 
maintained for generations the grade of attenuation reached in the original 
culture the result of cultivation for a certain number of days at the higher 
temperature. In this manner Pasteur was able to control to some extent 
the degree of attenuation, and by inoculating first a highly attenuated 
and later a less markedly attenuated culture he could immunize animals 
against anthrax. This discovery was soon amply verified. The original 
method is practically the one employed today, and is proving of consider- 
able economic value. 
Pasteur’s next great experiment was undertaken for the relief of rabies, 
a condition in which for the first time he came to deal with a disease that 
not infrequently affects man. His success and the results of his discovery 
of an effective means of immunization against hydrophobia were even 
greater than in previous experiments. Here he was dealing with a disease 
of unknown etiology, the causative agent of which he could not cultivate 
artificially, but which he sought to attenuate by a new process—that of 
drying. 
Pasteur first established that the virus of rabies is contained within 
the tissues of the brain and spinal cords of infected animals. 
He then invented a method of inoculating animals by making subdural 
injections of an emulsion of these tissues. By repeated passage of a virus 
through a number of rabbits a virus of fixed pathogenic power (virus fixe) 
was obtained. By inoculating rabbits with this virus and removing their 
spinal cords immediately after death and drying these over a desiccating 
agent at room temperature, he found that he could modify the virulence 
of the virus at will, depending on the length of the period of drying. By 
emulsifying small portions of attenuated spinal cord in salt solutions and 
injecting these he was able gradually to immunize animals against rabies, 
and finally he applied the treatment successfully to the prevention of rabies 
in the human being. 
Antirabic vaccination is largely responsible for extending our knowl- 122 
IMMUNITY .—THEORIES OF IMMUNITY 
edge of the possibility of securing immunization. Pasteur has taught us 
at least three different methods for modifying a virus in the preparation 
of a vaccine, and that each disease, being itself a special entity, having 
its own characteristics, must be dealt with along special lines. 
These discoveries were largely empirical, and the explanations of their 
mechanism are now only of historic interest. It was not until 1883, when 
Metchnikoff shed light upon the problems of immunity by making a series 
of remarkable studies on the role played by certain of the body cells in 
overcoming infection, and the part they played in the processes of immunity 
in general, that the world was given a glimpse into the dark problems of 
immunity. These observations were soon followed by investigations show- 
ing the importance of the body fluids, and since that time a great deal of 
work has been done upon these subjects. As a consequence, a large amount 
of data of a wholly new order has accumulated, accompanied by the intro- 
duction of a host of new terms expressing diverse views and theories ad- 
vanced by individual workers. Of the many theories advanced from time 
to time to explain the phenomenon of immunity, two have claimed the 
most attention: one ascribes protection and cure to the activity of cer- 
tain body cells; this is known as the cellular theory; and the other attributes 
these qualities to the body fluids—the humoral theory. The chief exponent 
of the former is Metchnikoff, with his theory of phagocytosis, whereas Ehrlich 
is the father and leader of the latter, with that marvelous invention of 
human ingenuity, the side-chain theory. 
OBSOLETE THEORIES OF IMMUNITY 
The earlier hypotheses advanced by various investigators are now only 
of historic interest, as in the light of subsequent discoveries and observa- 
tions they have failed to offer adequate explanations. 
Pasteur’s own theory and explanation of the mechanism of acquired 
immunity sought to show that the micro-organisms living in the infected 
animal used up some substance essential to their existence, so that, for 
lack of proper nourishment, the micro-organisms were eventually forced 
to retire, the soil being unfit for further occupation. Pasteur1 expressed 
himself as follows on results of experiments with the bacillus of fowl cholera: 
“The muscle which has been much affected has, even after healing and 
repair, become in some way incapable of supporting the growth of the 
micro-organism, as if the latter, by a previous culture, had eliminated from 
the muscle some principle that life does not bring back and whose absence 
prevents the development of the small organism. There is no doubt that 
this explanation, to which the plainest facts at the moment lead us, will 
become general and applicable to all the virulent disease.” This was known 
as the “exhaustion theory.” 
Chauveau considered it more probable that the micro-organisms, after 
having lived in the body of an infected animal, produced substances that, 
accumulating in the blood, had an inhibitory action on the bacteria and 
were inimical to their further existence. This was known as the “retention 
theory” and in some particulars was just the opposite of the exhaustion 
theory. Mention may also be made of the theory of Buchner, which as- 
cribed immunity to the property of the animal organism to reinforce local 
resistance of the organs by means of an inflammatory reaction which pro- 
tected these tissues against reinfection by the same germ for indefinite 
periods. 
Compt. rend. Acad. d. sc., Paris, 1880, xc, 247. THEORY OF PHAGOCYTOSIS AND CELLULAR RESISTANCE 
123 
THEORY OF PHAGOCYTOSIS AND CELLULAR RESISTANCE AND IMMUNITY 
CHEMOTAXIS 
Historic.—As Lord Lister1 stated in 1896, “If ever there was a romantic 
chapter in pathology, it has surely been that of the story of phagocytosis.” 
The author of this “story” is Elie Metchnikoff.2 His early researches on 
phagocytosis in the lowly organized forms of life constituted the starting 
point for an entirely new series of researches on the subject of Immunity, 
and his treatise on the “Comparative Pathology of Inflammation” must 
ever remain a most entertaining work and a medical classic. 
Several observers before Metchnikoff had considered that leukocytes 
might assist in bringing about the destruction of microparasites. Panum3 
(1874) suggested that the bacilli of putrefaction might, by making their 
way into the blood-corpuscles and being carried off to the lymph-glands, 
spleen, etc., thus disappear from the body fluids. Carl Roser (1881) had 
also observed the ability of certain “contractile cells” to ingest the enemy 
that enters the animal body. These statements, however, were poorly 
supported by scientific data, and the subject was not followed in subse- 
quent research. As the result of zoologic studies, Metchnikoff was led to 
discover the part played by body cells in the processes of immunity. He 
observed that when a food particle arrives in the vicinity of a simple uni- 
cellular organism as an ameba, the latter, by reason of its “irritability,” 
moves forward and sends out processes of its protoplasm (pseudopodia) 
that flow around the particle and finally gather it into the interior of the 
cell. The particle then undergoes a process of intracellular digestion, losing 
its sharp outline and clear appearance, becoming granular, and disappear- 
ing within the protoplasm of its host. Similar studies were made of the 
processes of nutrition in many unicellular animals, then through actininas, 
sponges, and similar animals of transparent and simple organization. 
Metchnikoff was primarily a biologist, and up to this time was mainly 
interested in the processes of nutrition in the simple forms of animal life. 
At this stage, however, he became greatly impressed by Cohnheim’s des- 
cription of the phenomena of inflammation. The diapedesis of leuko- 
cytes through the walls of the blood-vessels in an inflammatory reaction 
impressed him as significant and similar to the movements of the ameba 
in its process of feeding, and led to comparative studies in inflammation 
in the lower forms of life, where processes were much simpler to watch 
and may indicate what occurs in the higher vertebrates. 
He found that when daphnias are invaded by the spores of a yeast— 
the monospora—they may multiply in the body of the host and bring about 
its destruction. When, however, a few spores gained access, he found that 
the daphnia’s leukocytes approached them, formed a wall around them, 
and finally brought about their destruction by a process of digestion. He 
also observed that if rose prickles were stuck into large bipinnaria larvae 
of star fish, these were soon enveloped in a mass of ameboid cells. From 
this he concluded that, in inflammation, the exudation of leukocytes may 
be regarded as a reaction against any sort of injury, whether mechanical 
or due to bacterial invasion. 
Metchnikoff traced this defensive reaction against an invading micro- 
organism from small invertebrates up to man, and showed that instead 
1 Reports, Brit. Assoc. Adv. Sci., London, 1896, p. 26. 
2 For a complete review of the work of Metchnikoff and his students read Metchnikoff’s 
Immunity in Infective Diseases; English translation by Binnie, published by the Macmillan 
Company of New York, 1905. 
3 Virchow’s Archiv., 1874, lx, 347. 124 
IMMUNITY.—THEORIES OF IMMUNITY 
of bacteria attacking leukocytes and forcing a passage into them, as was 
then believed, they were indeed pursued and engulfed by the leukocytes. 
Connecting his various discoveries, he was able to formulate the idea that 
“the intracellular digestion of unicellular organisms and of many inverte- 
brates had been hereditarily transmitted to the higher animals, and re- 
tained in them by the ameboid cells of mesodermic origin. These cells, 
being capable of ingesting and digesting all kinds of histologic elements, 
may apply the same power to the destruction of micro-organisms.” To 
a cell and especially to a leukocyte possessing this activity and power he 
applied the name phagocyte, because of its ability to act as a scavenger, 
gathering up the living and dead material. 
The Original Theory of Phagocytosis.—The theory as originally ad- 
duced by Metchnikoff regarded the leukocytes and certain other cells as 
specific fighting cells, able to engulf and consume living as well as dead 
bacteria and cellular debris. The outcome of any infection would depend 
upon the success or failure of the phagocytes to overcome the invaders: 
if they were successful in ingesting all the bacteria, their victory meant 
recovery; if, on the other hand, they were destroyed by the invaders, the 
host was overcome by the infection. 
Phagocytes may ingest not only living and dead bacteria, but also red 
corpuscles, cellular debris, inorganic particles, such as coal-dust, and even 
soluble substances, such as bacterial toxins. 
Subsequent discoveries have shown that many other factors are present 
that considerably modify the workings of so simple a process. Metchni- 
koff has, therefore, modified his theory from time to time as new discoveries 
were made, but has always preserved the primary importance of the phago- 
cyte itself. 
Before stating the revised theory as it stands today, we will describe 
the kind of cells that may act as phagocytes, and consider the methods 
and reasons why these cells assume the functions of phagocytes. 
Varieties of Phagocytes.—Not only leukocytes, but other body cells, 
have been found active in the processes of phagocytosis. Metchnikoff has 
divided the phagocytes into two great classes: 
1. Microphages—principally the polymorphonuclear neutrophilic leuko- 
cytes (see Fig. 57). The eosinophilic leukocytes are also included in this 
group, but are of doubtful importance and weak in phagocytic powers. 
The small lymphocyte may be included in this class, although it is usually 
considered with the second group. 
2. Macrophages, principally the large mononuclear leukocytes; ameboid 
cells of the spleen and lymphatic glands; alveolar cells of the lung; endo- 
thelial cells of the serous cavities and lvmph-spaces; bone-corpuscles and 
giant-cells of bone-marrow and embryonic connective-tissue cells (Figs. 55 
and 56). As shown experimentally by Jones and Rous the phagocytic 
powers of fibroblasts appear to be very feeble.1 These investigators have 
shown that phagocytosis of blood-pigment, bacteria, etc., which takes 
place in granulation tissue is probably carried on wholly by endothelial 
and wandering cells. 
The most important are the leukocytes, especially the polymorpho- 
nuclear leukocytes, and the large lymphocytes of the blood. All the leuko- 
cytes, however, have phagocytic powers, as is well seen in opsonic determi- 
nations. Eosinophils are seldom known to ingest bacteria, but in infections 
with animal parasites, or after the injection of extracts of animal parasites, 
both a local and a general increase in eosinophilous forms may be observed. 
1 Jour. Exper. Med., 1917, 25, 189. THEORY OF PHAGOCYSTOIS AND CELLULAR RESISTANCE 
125 
Small lymphocytes are much less active than the large, presumably be- 
cause they contain less of the mobile cytoplasm, and consist chiefly of the 
structurally fixed nuclear substance, and while they take up but a small 
number of bacteria, they may be observed to contain various other cells, 
such as red corpuscles and cellular debris. 
Besides the leukocytes, some of the tissue cells which are free or have 
the power of becoming so are actively phagocytic. Endothelial cells of the 
lymph-spaces and serous cavities are especially active, not only in the 
phagocytosis of other cells and cellular debris, but also of various bacteria. 
The experiments of Keyes1 have been particularly significant in this con- 
nection. He has shown that the high immunity of the pigeon to virulent 
pneumococci injected into the blood-stream is apparently due, in part 
at least, to their rapid phagocytosis by the hemophagic cells in the liver 
and spleen. A similar destruction of pneumococci in the liver and spleen 
has been shown to occur by Berry and Melick,2 after intraperitoneal injec- 
tion. 
In exceptional instances epithelial cells may act as phagocytes. In 
the presence of an irritant these cells may become detached and act as 
phagocytes; this is exemplified in the case of chronic passive congestion 
of the lungs, in which the alveolar cells of the lung ingest the hemosiderin 
formed and deposited (heart failure cells). 
Relation of the Cell Types to Infection and Phagocytosis.—The kind of 
cells that take part in phagocytosis is determined to some extent by the 
nature of the irritant. Thus, in acute pyogenic infections the polymorpho- 
nuclear cell is found to be most active (see Fig. 57). It is extremely rare 
to find these cells containing bacilli in the tissues, although they will take 
them up readily enough under the artificial conditions of an opsonic deter- 
mination (see Fig. 70). In chronic bacterial infection, such as tubercu- 
losis and syphilis, and in infections with various fungi, the small lympho- 
cyte and macrophages are the types most concerned. 
Experimental evidence regarding lymphocytic activity is quite con- 
tradictory. Undoubtedly many of the cells in the lymphocytic accumula- 
tions seen in such conditions as tuberculosis and syphilis are not really 
lymphocytes from the blood, but are newly formed cells of the tissues. 
There is no direct means of inducing experimentally a local accumulation 
of lymphocytes similar to that induced by most any irritant, resulting 
in an outpouring of polymorphonuclear cells. Long-continued intoxica- 
tion of animals may result in increasing the numbers of lymphocytes, but 
the local introduction of the toxin leads to an accumulation of polynuclear 
cells, rather than lymphocytes. Reckzeli3 found that in lymphatic leukemia, 
in which the lymphocytes greatly exceed the polynuclears, the pus from 
an acute lesion or the fluid from the vesicles produced by cantharides, con- 
tained practically no lymphocytes, but was composed of the usual poly- 
nuclear cell forms. Wlassow and Sepp4 state that lymphocytes are not 
capable of ameboid movement or phagocytosis at ordinary body tempera- 
ture; Wolff,5 on the other hand, claims that tetanus and diphtheria toxins 
produce lymphocytosis in experimental animals, and Zieler6 claims that 
in the skin of rabbits exposed to the Finsen light active migration of lympho- 
cytes takes place during the reaction. General lymphocytosis may be 
1 Jour. Infect. Dis., 1916, 18, 272. 
2 Jour. Immunology, 1916, 1, 119. 
3 Zeit. f. klin. Med., 1903, 50, 51. 
4 Virchow’s Archives, 1904, 176, 185. 
6 Berl. klin. Woch., 1904, 41, 1273. 
6 Centralbl. f. Pathol., 1907, 18, 289. 126 
IMMUNITY.—THEORIES OF IMMUNITY 
produced experimentally by the injection of pilocarpin and muscarin, but 
these bear no relation to the vital process of phagocytosis, as they are appar- 
ently extruded from the lymphoid organs by contraction of the smooth 
muscles (Harvey1). 
As previously stated, the eosinophils undergo a marked increase during 
infections with various animal parasites. 
The typical macrophages, such as the endothelial cells of serous cavities 
(Figs. 55 and 56) and the lymph-spaces, are mostly concerned in the phago- 
cytosis of other cells and inorganic material. Brodie2 considers the phago- 
cytosis of leukocytes and red corpuscles by the endothelial cells of the 
lymph-glands and the spleen as the normal end of these cells. It is a mis- 
take to believe, however, that they do not ingest bacteria, since endothelial 
cells are extremely active phagocytically for bacteria; for example, Bartlett 
and Ozaki3 have shown that the fixed cells of the spleen and liver exert an 
important role in the phagocytosis of staphylococci in vivo. On the other 
hand, polymorphonuclear leukocytes may be observed to contain red cor- 
puscles, especially when aided by a suitable opsonin. 
Variations in the Phagocytic Activities of Leukocytes.—Studies in vitro, 
and especially in connection with investigations bearing upon the opsonins, 
have shown that the leukocytes may vary in phagocytic activity in both 
health and disease. 
Tunnicliff4 found the leukocytes at birth somewhat less active than 
in the adult to streptococci, pneumococci, and staphylococci. Their activity 
diminished considerably during the first month of life and does not reach 
that of the adult leukocytes until about the third year. This observer 
has also found that the leukocytes in recent exudates as those produced by 
intrapleural injections of aleuronaut, are more phagocytic for streptococci, 
pneumococci, and tubercle bacilli than the leukocytes of the blood of the 
same animal.5 
In pneumonia, scarlet fever, erysipelas, and other diseases in which 
there is acute leukocytosis, the phagocytic power of the leukocytes may be 
greater than normal, as shown by the experiments of Rosenow,6 Potter and 
Krumwiede,7 Tunnicliff,8 Boughton,9 and others. In leukocythemia, on 
the other hand, Ledingham,10 Bushnell,11 Keutzler,12 and others found that 
the polymorphonuclear leukocytes, and especially the abnormal ones, had 
deficient phagocytic power. 
Glynn and Cox13 have shown that during immunization of donkeys 
with anthrax bacilli, the leukocytes acquire enhanced phagocytic activities 
for these bacilli as compared with the leukocytes from normal donkey 
controls. It is probable that the differences are due to the presence of 
specific opsonins within or upon the leukocytes of the immune animals 
liberated after washing but, nevertheless, a specific difference between 
normal and immune cells has been demonstrated. 
1 Jour. Physiol., 1906, 35, 115. 
2 Jour. Anat. and Physiol., 1901, 35, 142. 
3 Jour. Med. Res., 1913, 35, 465; ibid., 1917, 37, 139. 
4 Jour. Infect. Dis., 1910, 698. 
5 Trans. Chic. Path. Soc., 1911, 8, 208. 
6 Jour. Infect. Dis., 1906, 3, 683. 
7 Jour. Infect. Dis., 1907, 4, 601. 
8 Jour. Infect. Dis., 1911, 8, 302. 
9 Jour. Infect. Dis., 1910, 7, 111. 
10 Lancet, London, 1906, February 10, 368. 
11 Brit. Med. Jour., 1907, 2, 142. 
12Ztsch. f. klin. Med., 1909, Ixvii, 131. 
13 Jour. Path, and Bacterid., 1911, 16, 535. Fig. 55.—Phagocytosis—Macrophages. 
A smear of peritoneal exudate from a guinea-pig twenty-four hours after injection with 3 c.c. of 
a 5 per cent, suspension of pigeon’s blood. Note that the corpuscles (nucleated) are being ingested 
by the large endothelial cells of the peritoneum. 
A smear of peritoneal exudate from the same guinea-pig forty-eight hours after injection with 
pigeon’s blood. Note the large numbers of corpuscles ingested by the endothelial cells. In most in- 
stances the corpuscles have undergone digestion, the nuclei being more resistant. These nuclei are 
shrunken, and in some instances are broken up. The extracellular red blood-corpuscles are swollen 
and stain lightly. 
Fig. 56.—Phagocytosis—Macrophages. Fig. 57. Positive Chemotaxis. Phagocytosis of Staphylococci CHEMOTAXIS 
127 
In certain chronic infections, notably chronic pneumococcus endo- 
carditis and chronic erysipelas, the phagocytic activity of the leukocytes 
may be either above or below normal. Hektoen1 has concluded that the 
plasma independent of its opsonic function may directly influence not 
only the phagocytic activity of leukocytes, but intraleukocytic digestion 
as well, the mechanism and significance of these changes being obscure. 
These changes in the phagocytic activity of leukocytes renders un- 
tenable the principle of Wright that in studying the opsonins the leuko- 
cytes may be taken as a constant; in acute as well as in some chronic in- 
fections a real grasp of phagocytic activity is obtainable only by the de- 
termination of the combined phagocytic powers of the leukocytes and 
serum of a given patient, and preferably with the infecting rather than 
a stock strain of the micro-organism. Further reference to this subject 
will be made in the chapter on Opsonins, because of the influence of the 
source of leukocytes upon the results of determining the opsonic index in 
disease. 
CHEMOTAXIS 
An important question in the study of the phenomena of phagocytosis 
is the manner in which the various leukocytes and other body cells are 
attracted to a focus of infection and brought into contact with the micro- 
parasites or other foreign substances. It must be assumed that some means 
of communication must exist between this point and the leukocytes in the 
circulating blood. Since there is no direct communication by way of the 
nervous system or other structural route, it wTould appear that the only 
mode of communication is through the body fluids. Chemical agencies, 
produced either directly by the bacteria or other foreign substance, or indirectly 
by their action upon cells at the site of residence in the tissues, are regarded as 
furnishing the attractive forces that are transmitted through the body fluids 
and exert what has been called by Bordet,2 chemotaxis. 
The movement of a cell in response to a chemical stimulus is a phe- 
nomenon that is displayed by almost all motile and unicellular organisms, 
whether animal or vegetable, and by the leukocytes and other unfixed 
cells of the higher animals. As a rule, chemical stimuli serve to attract 
cells to the site of infection, thus constituting what is known as positive 
chemotaxis; on the other hand, the stimuli may fail to attract or actually 
repel the cells, or be so powerful as to paralyze them en route, this con- 
stituting negative chemotaxis. 
Positive Chemotaxis.—That leukocytes reach the site of an infection 
because of chemical substances produced by bacteria at this point 
first clearly demonstrated by Leber3 in 1879. This writer observed that 
in keratitis leukocytes invaded the avascular cornea from the distant 
vessels, not in an irregular manner, but direct to the point of infection, 
where they accumulated. As dead cultures of staphylococci produced a 
similar although a less pronounced accumulation of leukocytes, Leber 
sought the chemotactic substance in their bodies and isolated a crystalline, 
heat-resisting substance—phogosin—which attracted leukocytes in the 
tissues. 
Since these fundamental studies were made many other investigations, 
with various chemical substances of many different origins, have been under- 
taken upon leukocytes, amebse, ciliata, and plasmodial forms, indicating 
1 Jour. Amer. Med. Assoc., 1911, lvii, 1579. 
2 Ann. de l’Inst. Pasteur, 1896, x, 104. 
3 Fortschritte der Med., 1888, 6, 460. 128 
IMMUNITY.—THEORIES OF IMMUNITY 
that chemical substances are mainly concerned in exerting either a positive 
or a negative chemotactic influence. 
Experimental evidence tends to show that cells respond to stimuli of 
various kinds chiefly through the effect of these stimuli upon surface tension: 
if they decrease the surface tension, the cell goes forward; if they increase 
the tension, the cell recedes. 
The behavior of leukocytes in inflammation may be explained on these 
purely physical grounds. At the site of cell injury or infection chemical 
substances are produced that tend to lower the surface tension of leuko- 
cytes and thus exert a positive chemotactic influence. These chemical 
stimuli are transmitted by the body fluids to the nearest capillaries, where 
they enter through the vessel wall and come in relation with the slowly 
moving peripheral leukocytes. The leukocytes will be brought into touch 
by the chemotactic substances most largely on the side from which the 
substances diffuse; accordingly, the surface tension being least nearest 
the stomata in the capillary wall, this results in the formation of pseudo- 
podia, and motion in this direction, dragging the nucleus along in an appar- 
ently passive manner. Those cells, therefore, containing most of the mobile 
cytoplasm, such as the polynuclear leukocytes, are chiefly affected in these 
processes; those containing little cytoplasm and a relatively large and 
dense nucleus, such as the lymphocytes, are affected primarily to a much 
less extent. 
Once outside of the vessel wall, the leukocytes tend to move toward 
the focus from which the chemotactic substance comes. If the leukocyte 
meets a substance that greatly lowers its surface tension, it will flow around 
the object and inclose it, this constituting phagocytosis. The toxins of 
the ingested bacteria may kill the cell, or so equalize surface tension that 
movement ceases. Otherwise the leukocytes tend to move forward until 
checked by any one of several influences, as pointed out by Wells: (1) 
Until the chemotactic substance has been used up or removed, or from 
any of the causes that terminate inflammation; (2) the leukocytes may 
reach a point where the chemical stimulant is so generally diffused that 
surface tension is decreased equally in all directions and motility stops; 
(3) the leukocytes may reach a place where toxins or other chemical sub- 
stances coagulate their cytoplasm or ferments cause their solution; (4) they 
may be blocked by a dense wall of leukocytes and other cells while being 
held fixed by the chemical attraction that diffuses through this wall. These 
factors would explain the formation of the wall of leukocytes about an 
area of infection. When, for example, the abscess has ruptured or has been 
incised, with removal of the chemotactic substances, there may be less 
chemotactic substances in the center of the inflamed area than there is 
further out; hence, the leukocytes will move away from the center toward 
the periphery, following the chemotactic substances back into the blood- 
vessel and lymph-stream. This would explain the dispersion of living 
leukocytes at the close of an inflammatory process. 
General leukocytosis can be explained equally well by assuming that 
the chemotactic substances from the area of inflammation, reaching the 
blood-stream, pass through the bone-marrow, lowering surface tension 
and attracting leukocytes into the blood-stream as long as it contains more 
chemotactic substances than the marrow. 
The exact chemical nature of chemotactic substances is unknown. In 
bacterial infection the toxins, and especially the protein of dead micro-organisms, 
are regarded as mainly responsible for the occurrence of positive chemotaxis. 
Chemotaxis and phagocytosis of chemically inert particles, such as coal-dust, Fig. 58.—Negative Chemotaxis. 
A smear of exudate from the peritoneal cavity of a guinea-pig twenty-four hours after injection 
with virulent streptococci. The exudate was thin, serous, and tinged with hemoglobin. Note the 
large numbers of streptococci and relatively few leukocytes. CHEMOTAXIS 
129 
stone-dust, and pigments, are more difficult to explain on this physical 
basis of alteration in surface tension. It is probable that the death of tissue 
cells, brought about by these materials, may produce the chemical stimulant 
responsible for a mild but definite chemotactic influence. Although the 
movement of amebse and similar higher animals cannot be fully explained 
on this physical basis, the surface tension theory best explains leukocytic 
movement. Although the ameba may possess some special property that 
endows it with the power of selecting and engulfing a food particle, it would 
appear to be entirely unreasonable to assume that a simple, undifferentiated, 
and naked leukocyte possesses similar powers. The physical theory, there- 
fore, appears to be the most reasonable offered in explanation of the ameboid 
movements of these simple cells. 
Negative Chemotaxis.—In nearly all infections we find that leukocytes 
are attracted in large numbers into the involved area, i. e., nearly all bacteria 
give off substances that are positively chemotactic. In certain infections, 
however, we may find the tissues poor in leukocytes, as exemplified in 
infections due to the presence of virulent streptococci. This negative chemo- 
taxis is more difficult of explanation. Kantlack doubts the existence of 
really negative chemotactic action upon leukocytes. Verigo1 also considers 
that as yet no actual negative chemotactic substances have been satisfac- 
torily demonstrated; certainly no marked example of negative chemotaxis 
has been shown since methods involving the study of phagocytosis in vitro 
have been devised. It is true that virulent bacteria appear to repel the 
leukocytes but, as Kantlack has pointed out, these are not necessarily 
examples of negative chemotaxis, and it is probable that the paucity in 
numbers of the leukocytes about such an area of inflammation is due to 
their overstimulation or paralysis and destruction of the powerful ferments 
that are given off by the bacteria. Thus, Metchnikoff has asserted that 
leukocytes might, after a time, be attracted toward substances that would 
kill them. Therefore, while leukocytes will migrate freely toward sub- 
stances that would kill them, they may be destroyed before they reach the 
inflammatory area, or, having reached there, are promptly destroyed and 
pass into solution (Fig. 58). 
While it is doubtful, therefore, whether substances are produced by 
bacteria that actually repel leukocytes, the point has not been definitely 
settled. If such substances exist, it would appear that they are closely 
identified with either the endotoxins or the aggressins, the latter being definite 
secretory products of bacteria that neutralize opsonins and retard phago- 
cytosis. In many instances it is probable that the same substances that 
exert a positive chemotaxis are, when concentrated, negatively chemotactic, 
through overstimulation and paralysis of the leukocytes. With diminution 
in the numbers or vitality of bacteria and dilution of their chemotactic 
substances, this inhibiting influence is removed and the leukocytes are 
attracted to the focus of infection, thus explaining in a way those instances 
in which positive chemotaxis is observed to follow a primary period of nega- 
tive chemotaxis. This subject is further discussed below in the section 
dealing with the influence of bacterial products on phagocytosis. 
The Effect of Drugs Upon Phagocytosis.—Cantacuzene,2 working in 
Metchnikoff’s laboratory, early showed that phagocytosis of cholera vibrios 
in the peritoneal cavity of guinea-pigs was retarded during opium narcosis. 
Since then a great deal of work has been devoted to the influence of chemical 
substances upon phagocytosis and antibody production in relation to the 
treatment of disease. 
1 Arch. d. med. Exper., 1901, 13, 585. 
Ann. d. Inst. Pasteur, 1898, 12, 288. 130 
IMMUNITY—THEORIES OF IMMUNITY 
Rubin1 found that alcohol, ether, chloroform, and narcotics in general 
reduce resistance to infection by affecting the leukocytes or substances 
derived from them; Hektoen and Ruediger2 have shown that various salts 
are barium chlorid, calcium chlorid, magnesium chlorid, etc., retard phago- 
cytosis in the test-tube presumably by some influence upon opsonins. Ham- 
burger,3 however, has found that very small quantities of calcium salts 
increase phagocytic power to a considerable extent not only in the test- 
tube, but also the body. Similarly, numerous other substances—iodoform, 
benzene, camphor, turpentine, alcohol, chloral, fatty acids, and balsam 
of Peru—all applied in very small doses, showed a stimulating effect on 
phagocytosis, but paralyzed when given in greater concentration; inasmuch 
as these substances are soluble in fats, Hamburger believes that slight 
quantities of them may dissolve the outer layer of phagocytic cells and 
thereby increase their plasticity. 
In the investigations of Arkin4 it appears that the action of chemical 
substances on phagocytosis varies with their composition and pharmaco- 
logic action. Substances which have an inhibitory effect on oxidation, 
such as chloroform, ether, morphin, potassium cyanid, and alcohol, all 
depress phagocytosis. On the other hand, substances like iodoxybenzoate 
which owe their pharmacologic and germicidal action to the presence of 
physiologically active oxygen and colloidal metals which stimulate oxida- 
tive processes, have a stimulating effect on the phagocytosis of streptococci 
and staphylococci in vitro. Substances like caffein and antipyrin which 
have little if any effect upon oxidation were found to have slight or no 
influence upon phagocytosis, whereas calcium, magnesium, and mercuric 
chlorids, quinin, strychnin, and arsenic compounds were believed to stimu- 
late phagocytosis both in vitro and in vivo. Tunnicliff,5 on the other hand, 
found that dilute solutions of calcium chlorid, sodium salicylate, lactic 
acid, and magnesium chlorid increased phagocytosis of diphtheria bacilli 
in vitro, but not in vivo. Otsubo,6 however, found M/8 solutions of certain 
salts as magnesium sulphate, sodium carbonate, calcium chlorid, and others 
to diminished phagocytosis of streptococci in mice; higher dilutions did 
not appear to stimulate phagocytosis either in vitro or in vivo. 
Graham7 has found ether depressive for phagocytosis; Bartlett and 
Ozaki8 found phagocytosis of staphylococci in vivo reduced in phosphorous 
poisoning and after chloral anesthesia. Dewey and Nuzum9 found the 
injection of colloidal suspensions of cholesterin in rabbits depressive for 
phagocytosis and antibody formation. Klecki10 has reported the stimulat- 
ing effect of small doses of radium emanations upon phagocytosis of staphyl- 
ococci and colon bacilli; Irala11 found that ultraviolet rays promoted phago- 
cytosis of staphylococci, but with prolonged exposure phagocytosis was 
arrested, due to injury of the leukocytes. 
1 Jour. Infect. Dis., 1904, 1, 425. 
2 Jour. Infect. Dis., 1905, 2, 128. 
3 Physikalisch-Chemische Untersuchungen fiber Phagozyten, etc., Verlag. von J. F. 
Bergmann, Wiesbaden, 1912. 
4 Jour. Infect. Dis., 1912, 11, 427; ibid., 1913, 13, 408. 
8 Jour. Infect. Dis., 1916, 19, 97. 
8 Jour. Infect. Dis., 1921, 28, 18. 
7 Jour. Infect. Dis., 1911, 8, 147. 
8 Jour. Med. Research, 1913, 35, 465; ibid., 1917, 37, 139. 
9 Jour. Infect. Dis., 1914, 15, 472. 
10 Ztschr. f. Immunitatsf., orig., 1912, 13, 589 
u Ann. d’lgrene, 1920, 30, 28. RESULTS OF PHAGOCYTOSIS 
131 
THE EFFECT OF BACTERIAL PRODUCTS ON PHAGOCYTOSIS 
As recently shown by Wadsworth and Hoppe1 bacterial products de- 
press the phagocytic activities of leukocytes in vitro. Diphtheria toxin 
causes immediate depression when placed in mixtures of leukocytes and 
bacteria. This depressor action could not be neutralized by antitoxin, 
heat, or light, and variations in the constitution of the culture broths, which 
greatly affected true toxin production, caused no variation in the production 
of the depressing substance. The depressing action of young culture broths 
was found to be less marked than that of older cultures. It was also found 
that digestion with proteolytic enzymes either wholly or partially destroyed 
the depressing element. The substance could be isolated by absorbing 
it with leukocytes and then washing it from them with salt solution. After 
removal of the substance the leukocytes regained their phagocytic activity. 
As discussed more completely in the chapter on Ferments and Anti- 
ferments the digestion products of enzymes from bacteria and tissue cells 
may exert a negative chemotactic influence. d’Herelle states that the 
diastatic ferment produced by his “bacteriophage” and regarded as the 
active bacteriolytic agent repels phagocytosis. 
These and other experiments indicate that phagocytosis in vivo may be 
readily influenced during infection by bacterial products exerting negative 
chemotaxis. 
RESULTS OF PHAGOCYTOSIS 
Cytases and Endolysins.—After phagocytosis has been accomplished 
the fate of the engulfed objects depends upon their nature. In general, 
they undergo a process of digestion. The ameba, for example, is able to 
kill and digest engulfed material through an intracellular ferment regarded 
as a form of trypsin, demonstrated by Mouton2 and called amebadiastase. 
According to Metchnikoff, the digestion of erythrocytes and tissue frag- 
ments is accomplished through an enzyme of the macrophages called macro- 
cytase; that of bacteria or other substances engulfed by microphages by a 
similar enzyme called microcytase. 
Following the general law that living protoplasm cannot be digested, 
we are confronted with the very important question as to whether living 
bacteria are engulfed by phagocytes or whether they are first destroyed 
by extracellular agencies before they undergo phagocytosis. 
It seems positively established at the present time that leukocytes 
do take up living bacteria, which may either grow inside the leukocyte 
or be destroyed by intracellular substances called endolysins. On the other 
hand, leukocytes do not take up extremely virulent bacteria, hence the 
question arises as to the importance of substances in the body fluids which 
neutralize the repelling substances of bacteria and facilitate phagocytosis. 
This subject, which has considerably modified Metchnikoff’s views of 
phagocytosis, will be considered in a succeeding chapter. It will suffice 
here to state that leukocytes may engulf living bacteria possessing some 
virulence, for not infrequently an infection may be spread by bacteria 
transported into deeper tissues by phagocytes, when they resist the germi- 
cidal activity of the endolysins, bring about the death of the phagocyte, 
and are thus liberated into new tissues. 
Death of the engulfed bacteria is, therefore, brought about by endo- 
lysins3 that are probably different from the digestive enzymes, or cytases 
1 Jour. Immunology, 1921, 6, 399. 
2 Compt. rend, de l’Acad. Sci. de Paris, 1901, cxxxiii, 244. 
3 For general review, see Kling, Ztschr. f. Immunitatsf., 1910, 7, 1. IMMUNITY.—THEORIES OF IMMUNITY 
132 
which bring about the digestion of dead cells. These endolysins are strongly 
bactericidal and have a complex structure resembling bacteriolysins. 
According to Weil1 they are not specific. They are resistant to 65° C. or 
even higher, do not readily pass through porcelain filters, are precipitated 
by saturation with ammonium sulphate, and resemble the enzymes in 
many respects (Manwaring2). It is probable that the endolysins act not 
only upon bacteria that have been phagocytosed, but also upon free bacteria 
when liberated through disintegration of the leukocytes. In this manner 
the endolysins would closely resemble the bacteriolysins and support the 
contention of Metchnikoff that these important substances contained in 
the body fluids are derived primarily from the cells which he has classed 
as phagocytes. According to Schneider3 lymphocytes and macrophages 
seem to contain little or no endolysin, and these cells are not so active in 
the phagocytosis of virulent bacteria as are the microphages. 
It is possible, however, that in certain instances cells not only fail to 
kill the microparasites they ingest, but actually protect them from circu- 
lating antibodies; apparently the micro-organisms of leprosy, tuberculosis, 
gonorrhea, and leishmaniosis may live more or less habitually within tissue 
•cells, and, as demonstrated by Roux and Jones,4 living phagocytes are 
able to protect ingested bacteria from the destructive substances in the 
surrounding fluid, and even from a potent homologous antiserum. 
Indigestible substances, if chemically inert, may remain in cells, par- 
ticularly in fixed tissue cells, for variable periods of time. The leukocytes 
seem to transfer such particles to other tissues, particularly to the lymph- 
glands. It is probable that these phagocytes are in turn engulfed by the 
endothelial cells. Macrophages of the lymph-sinuses or the leukocytes 
may be destroyed in the glands, and their contents rephagocyted by these 
cells. In just what manner these insoluble particles reach the gland stroma 
or perilymphangeal tissues is unknown; it is probable that they are liberated 
from the lining endothelial cells, and are again seized by the young connec- 
tive-tissue cells. 
THE MECHANISM OF PHAGOCYTOSIS 
Aside from the question of whether fixed and wandering cells may en- 
gulf virulent microparasites and the influence of substances in the body 
fluids upon this phase of phagocytosis, we have for consideration the mechan- 
ism whereby these cells engulf bacteria and other microparasites, various 
cells, and inorganic particles. Based upon the direct observations of Metch- 
nikoff and his pupils the phase of engulfing is accomplished by means of 
pseudopods and the ameboid movement of the phagocyting cell, whereby 
the particles become adherent and are finally rolled or passed into the 
protoplasm of the phagocyte. Recent investigations by Barikine5 and Kite 
and Wherry6 would indicate, however, that in so far as leukocytes are con- 
cerned these processes are of minor importance, and that phagocytosis 
depends upon the physical conditions of the surface of the phagocytic 
cells, a “stickiness” of the leukocytes, whereby bacteria and other par- 
ticles become adherent, the engulfing being a purely passive process which 
depends upon protoplasmic streaming within the cells. They believe that 
1 Arch. f. Hyg., 1911, 74, 289. 
2 Jour. Exper. Med., 1912, 16, 250. 
3 Arch. f. Hyg., 1909, 70, 40. 
4 Jour. Exper. Med., 1916, 23, 601. 
6 Ztschr. f. Immunitatsf., orig., 1910, 8, 72. 
6 Jour. Infect. Dis., 1915, 16, 109. THE RELATION OF THE BODY FLUIDS TO PHAGOCYTOSIS 
133 
such substances as opsonins act in increasing phagocytosis merely because 
they increase the “stickiness” of the cells, and that phagocytosis depends 
essentially upon the relative stickiness of phagocytes and bacteria; in- 
creased phagocytosis in the presence of serum and particularly unheated 
serum is ascribed to an increased stickiness of the leukocytes. Mechanical 
contact as well as certain variations in chemical reactions may, therefore, 
result in the production of those changes in contour and protoplasmic 
streaming responsible for rolling the particle within the protoplasm of the 
cell. 
The researches of Lawson1 add emphasis to these observations. Con- 
trary to current opinion, she believes that the malarial parasite is not intra- 
cellular, but extracellular and possibly pericellular. They become attached 
to the erythrocytes at “mounds” which the parasites tend to surround with 
pseudopods. 
These researches indicate, therefore, a purely physical base in explana- 
tion of phagocytosis tending to render untenable the older conceptions of 
the mechanism involved. 
THE RELATION OF THE BODY FLUIDS TO PHAGOCYTOSIS 
Important and far-reaching as were Metchnikoff’s researches and con- 
clusions, they were not allowed to pass unchallenged, especially by the 
adherents of the humoral school, who were able to show the potent influences 
of the body fluids in the mechanism of recovery from infections where phago- 
cytosis was little in evidence, or, indeed, where phagocytosis was impos- 
sible. It was shown that Metchnikoff’s original theory was untenable, 
and that the leukocyte is almost impotent if removed from the influence 
of the body fluids. 
As demonstrated by Denys, Leclef, Fliigge, Nuttall, Pfeiffer, and others, 
bacteria may be killed, i. e., may undergo a process of lysis or disintegration, 
by means of substances in the blood-serum entirely independent of phago- 
cytosis. Later researches by Wright, Neufeld, and their co-workers demon- 
strated most clearly that even in those infections in which phagocytosis 
was observed and known to be of great importance the bacteria are first 
prepared for phagocytosis by substances in the body fluids, and that with- 
out this preliminary preparation of the bacteria phagocytosis was slight 
and of little consequence. 
Metchnikoff corroborated most of these discoveries, and modified his 
theory from time to time to meet the developments and keep them within 
the limits of the phagocytic theory. For example, when bacteriolysis was 
shown by Bordet to be due to two separate substances in the body fluids, 
which he called substance sensibilisatrice and alexin (later renamed by 
Ehrlich amboceptor and complement respectively), Metchnikoff claimed 
that this phenomenon was extracellular digestion, similar to the intra- 
cellular digestion that occurs within the phagocyte, and brought about by 
ferments secreted and liberated from leukocytes or other cells classed as 
phagocytes. He regards alexin as a cytase secreted by leukocytes, or liber- 
ated upon their disintegration; similarly the substance sensibilisatrice 
is regarded as a free ferment (fixateur), derived principally from leuko- 
cytes, and concerned in preparing the bacterial or other cell for the digestive- 
like action of the cytase. 
Aside from these free ferments that are capable of producing extra- 
cellular lysis, Metchnikoff has long known that other substances that aid 
1 Jour. Exper. Med., 1915, 21, 584. 134 
IMMUNITY.—THEORIES OF IMMUNITY 
phagocytosis itself may be present in the body fluids; he regards these as 
of the nature of stimulins, or substances that stimulate leukocytes to be- 
come more actively phagocytic. On the other hand, Leishman, Wright 
and Douglas, Neufeld and Rimpau, Hektoen, and others have clearly 
demonstrated that they facilitate phagocytosis not by stimulating the 
leukocytes, but rather by lowering the resistance of bacteria or in some 
way rendering them more vulnerable to phagocytosis (opsonins, bacterio- 
tro pins). 
Thus the gap between the original cellular theory, which ascribed pro- 
tection and cure to phagocytosis pure and simple, and the humoral theory 
(finally summed up by Ehrlich in his side-chain theory), which ascribed 
the chief and primary roles to substances in the body fluids, and relegated 
phagocytes to a position of secondary importance, regarding them only 
as scavengers that remove dead or disabled micro-organisms, has been 
filled with discoveries correlating both processes. 
The vitality of the leukocyte is to be regarded as important in the con- 
sideration of phagocytosis as a means of defense. While the body fluids 
are acting upon the invaders, the leukocytes themselves are probably under- 
going quantitative and qualitative changes. They are increasing in numbers, 
and as Rosenow1 has shown, are undergoing more specific changes. Thus, 
for instance, the leukocytes from a pneumonia patient were found more 
vigorous against invasion of the pneumococcus than are those from a normal 
person, regardless of the influence of serum. 
When a microparasite is ingested the process has only begun. Unless 
suitable endolysins are present and the endotoxin is absorbed or otherwise 
dealt with, and unless suitable digestive enzymes are secreted and the 
bacterium is dissolved, the process is useless, or indeed, if viable bacteria 
are transported to other parts of the body, it may be dangerous. 
THE REVISED THEORY AND ROLE OF PHAGOCYTOSIS IN IMMUNITY 
As previously stated, Metchnikoff has revised his theory from time to 
time, as these discoveries were made on the influence of substances in the 
body fluids, not only upon phagocytosis itself but also upon the processes 
of immunity in general. He would regard extracellular cytolysis (bacterio- 
lysis, hemolysis, etc.) as due to the same ferments that bring about the 
destruction and solution of the ingested bacterium or other cell within 
the phagocyte, and further, these extracellular ferments are derived from 
the cells that are classed as phagocytes. By this method of reasoning he 
would preserve the importance of the phagocytic theory. 
In local infections phagocytosis is usually well marked, and no doubt 
plays an important part in resistance to and recovery from these condi- 
tions. Recent investigations by Bull2 have shown that following aggluti- 
nation of various bacteria in vivo phagocytosis of the micro-organisms 
frequently follows. In infections due to the various pathogenic micrococci, 
as staphylococci, pneumococci, and streptococci, phagocytosis appears to 
be most active and an important means of overcoming the infection. In 
other infections, as those due to the typhoid bacillus and allied bacilli, it 
is probable that extracellular substances or antibodies are chiefly operative 
in affording protection or in overcoming infection, although certain of 
these, as the agglutinins and opsonins, facilitate phagocytosis. 
The question, then, of the relative importance of the cellular and humoral 
theories of immunity resolves itself to a consideration of the origin of the 
1 Jour. Infect. Dis., 1906, 3, 683. 
2 Jour. Infect. Dis., 1915, 16,109. HUMORAL THEORY OF IMMUNITY 
135 
substances in the body fluids so potent in both processes. If they are de- 
rived solely from the cells known to act as phagocytes, then the cellular 
theory of phagocytosis, in its broader meaning, is the one explanation of 
the processes of immunity as they are now understood. This, however, 
has never been proved, and it is entirely likely that these substances are 
products of a general, rather than of a more restricted, cellular activity, 
so that ultimately all immunologic processes are cellular in origin. For 
this reason we prefer to speak of the phagocytic cell in its relation to immunity 
when dealing with the relation and activity of microphages and macro- 
phages in a limited sense in the process of phagocytosis. 
HUMORAL THEORY OF IMMUNITY 
The humoral theory of immunity, which would ascribe the power to re- 
sist infection to the body fluids, may be said to have had its origin in 1896, 
when Fodor1 discovered that the blood of the rabbit will kill anthrax bacilli 
in the test-tube independent of cells and phagocytosis. Under the inspira- 
tion of Fliigge,2 Nuttall3 confirmed Fodor’s results and went further, show- 
ing that the bactericidal power of the blood and other fluids is due to a 
substance of undetermined nature which is destroyed by heating to 55° C. 
for one hour. Fliigge now based a theory of immunity on the presence of 
bactericidal substances in the body fluids, which was adopted and devel- 
oped by Bouchard4 and his school. Later Buchner5 adopted this theory, 
and sought to explain the bactericidal action of blood-serum as dependent 
upon a special constituent which he called alexin (protective substance). 
With the discovery, in 1890, of antitoxins by von Behring and Kitasato 
the theory received fresh support, and while an effort was made to demon- 
strate that antitoxins were of paramount importance in acquired immunity, 
evidence soon accumulated to show that this antitoxic power is operative 
only in a few diseases, chiefly in diphtheria and tetanus. 
Fresh support to the “humoral” as against the “cellular” explanation 
of immunity was given by Pfeiffer6 in 1894, with the discovery that cholera 
vibrios introduced into the peritoneal cavity of a guinea-pig previously 
immunized against cholera became transformed into granules, and ulti- 
mately passed into complete solution (bacteriolysis), apparently without 
the aid of cells. Bordet7 then showed that this phenomenon was due to 
two distinct substances—one, the “sensitizing substance,” which is specific 
and exists only in the immune serum, acting only on the bacteria against 
which the animal was immunized, and the other a non-specific substance, 
found in the fresh serum of practically all animals, and to which he gave 
the name “alexin,” and which was later renamed by Ehrlich and called 
“complement.” 
Of the various theories offered in explanation of these observations, the 
suggestive, fascinating, though highly hypothetic theory of Ehrlich,8 known 
as the side-chain theory, has been most widely accepted and adopted to 
explain new discoveries as they were made. The theory has, indeed, aided 
1 Deutsch. med. Wchn., 1886, 617; ibid., 1887, 745. 
2 Zt. f. Hyg., 1888, 4, 208. 
3 Zt. f. Hyg., 1888, 4, 353. 
4 Les microbes pathogenes, Paris, 1892. 
6 Archiv. f. Hyg., 1890, 10, 84; Centralb. f. Bakteriol., 1889, 5, 817; ibid., 1889, 6, 1, 561; 
ibid., 1889, 8, 65. 
8Zt. f. Hyg., 1894, 18, 1. 
7 Ann. d. l’Inst. Pasteur, 1898, 12, 688. 
8 Klin. Jahrb., 1897.6, 299; “Croonian Lecture,” Proc. Roy. Soc., London, 1900, lxvi, 424. 136 
IMMUNITY.—THEORIES OF IMMUNITY 
investigators in making new discoveries. Nevertheless the contention of 
Bordet, that its too ready acceptance without sufficient convincing proof 
has retarded investigation, should not be ignored. 
The basis of this theory, as originally proposed, bore no relation to the 
subject of immunity, but was advanced in 1885 to explain the processes of 
nutrition. 
Ehrlich asserts that a cell has two important functions: The first is the 
special physiologic function, as that of a nerve-cell to conduct; of a gland- 
cell, to secrete, etc. The second function is that of nutrition, and presides 
over the processes of waste and repair. Furthermore, each of the mole- 
cules composing the complex cell is believed to possess these two functions, 
i. e., one is concerned with the special function of the molecule, and the 
other, the more important functional portion, is concerned in the nourish- 
ment of the molecule. 
The second portion, or that concerned with nutrition, is of more im- 
portance in relation to the problems of immunity. Ehrlich conceives this 
as consisting of a special executive center or main portion (“Leistenkern”), 
in connection with which there are nutritive side chains, receptors, or hap- 
tines (“Leitenketter”), which “seize,” or rather enter into chemical com- 
bination with, suitable food atoms, which is followed by a sort of digestive 
or absorptive process, whereby the food material is incorporated in the 
molecule. 
The function of “seizing” molecules of food from the surrounding tissues 
implies a selective action or chemical affinity between food atoms and the 
portion of a cell or side arm for which it has a chemical affinity, for we 
cannot conceive that all atoms that circulate in the blood and lymph are 
suitable for all cells at all times. 
The food molecule in the fluid surrounding the cell is conceived as posses- 
sing a special or haptophore portion for union with the side arm of a cell 
molecule, and when brought into relation with one of the side arms or 
receptors of the cell molecule, the two are “anchored,” or unite, just as a 
key fits a lock. The second stage involves a process that may be compared 
to digestion, by which the food material is prepared and absorbed, in whole 
or in part, into the molecule of protoplasm. 
These processes, therefore, are conceived as being chemical rather than 
physical, and our diagrammatic representations of them have no necessary 
or actual morphologic basis. One is quite likely to regard the main central 
portion as the nucleus of a cell, and the side arms as small morphologic 
projections resembling the prickles of certain epidermal cells. These proc- 
esses are concerned with each molecule of a cell, the main portion, or “Lei- 
stungskern,” being conceived as diffusing through the nutritive part of the 
molecule, and the side-arm receptors, or “Leitenketter,” as numerous 
atoms or groups of atoms, each of which has a chemical affinity for some 
particular food substance circulating in the body fluids, and necessary 
for the life of the molecule in question. 
Later this theory was amplified by Ehrlich to explain the action of 
toxins and the production of antitoxins. It assumes that the side arms 
to a cell molecule are exceedingly numerous, not only because nutritive 
substances are varied, but because special cells also possess different and 
special side chains, which anchor pathologic material. When infection 
occurs, and in addition to toxins the physiologically normal substances 
are brought to the cells, they likewise find suitable receptors in practically 
all or certain cell groups, and become anchored, causing more or less damage 
to the cells. HUMORAL THEORY OF IMMUNITY 
137 
Having combined with the side arms or receptors of a cell, the toxin 
may be sufficiently potent to kill the cell, and if a large number of cells are 
so injured, symptoms of disease present themselves and death of the in- 
fected host may follow. On the other hand, although the cell has lost one 
or more of its side arms, it may not be dead, and it proceeds at once to 
repair the damage done. According to Weigert’s “overproduction theory,” 
nature is lavish in its processes of repair, and the cell not only replaces the 
lost receptors, but produces them in numbers; the excess receptors, having 
no space for attachment to the cell, are thrown off into the blood-stream. 
Each of these cast-off receptors or haptines possesses the same structure as 
the original receptor. These free receptors, then, are capable of com- 
bining chemically with their antigen, neutralizing the antigen, and rendering 
it innocuous. In diphtheria and tetanus the antigen is largely the soluble 
toxin of the bacilli, and the antitoxins are these cast-off receptors produced 
as a result of the stimulating action of the toxins upon the cells. This 
excess of receptors is made by repeatedly injecting a horse with increasing 
doses of diphtheria toxin. By injecting this receptor-laden (antitoxin) 
serum into one suffering from diphtheria the receptors unite with free 
diphtheria toxin and thus protect the body cells. 
For the production of these receptors, or antibodies, as they are now 
called, it is necessary, as previously stated, that the antigen enter into 
chemical combination with the cell, so that the usual illustrations showing 
the theoretic union of antigen and side arm by physical contact alone 
probably do not correctly portray what actually occurs. As Adami points 
out, the antigen probably enters into intimate relationship with the cell, 
and the continued stimulation of its presence is responsible for the pro- 
duction of an excess of receptors, in addition to the overproductive tendencies 
of nature’s repair. 
It is also necessary that the antigen possess sufficient toxic power at least 
to stimulate the cell, for otherwise antibodies may not be produced. Food 
material, for instance, being physiologic, is assimilated by the cells with- 
out stimulating the production of antibodies, as otherwise the food would 
be attacked by cast-off receptors and rendered useless before it reaches 
cells, the process ending in starvation and death. 
The host in whom certain cells with special receptors for a given poison 
are present will make use of these, no matter how the pathologic agent 
is introduced. This affinity is well illustrated in tetanus, where the effects 
produced are dependent to a large extent upon the selective affinity of the 
toxin for nerve-cells. 
The Three Orders or Receptors and Corresponding Antibodies 
First Order: Antitoxin and Simple Antiferments.—The simplest re- 
ceptor of the cell molecule is composed of a single arm or haptophore, for 
union with the haptophore portion of a food molecule. As previously stated, 
a molecule of toxin is conceived as being composed of two portions—one, 
the haptophore, for union with the side arm or receptor of a cell molecule, 
and the second, the toxophore, in which its toxic action resides. 
The first stage of intoxication of a cell produced by a true toxin con- 
sists in the union of the haptophore portion of the toxin molecule to a re- 
ceptor or side arm of the cell molecule, this receptor being one that fits the 
toxin molecule “like a key fits a lock.” Each molecule of the body cell 
has innumerable receptors, of which only a certain number are suitable for 
a particular toxin. The toxin molecule is now anchored to the living cell, 138 
IMMUNITY.—THEORIES OF IMMUNITY 
and, as animal experiments with a great number of toxins show, this union 
is a firm and enduring one (Fig. 59). 
So long as the union lasts the side chain involved cannot exercise its 
normal nutritive physiologic function—the taking up of food-stuffs. Fur- 
thermore, the toxophore group of the toxin molecule may now exert an 
injurious, enzyme-like action on the protoplasm of the cell, with the result 
that the protoplasm is poisoned. If only a few of the cell receptors are 
united with toxin molecules, or if the toxin is of low toxicity, the effects on 
the cell may be slight. If more are joined to the molecule or the toxin is 
highly poisonous, the whole molecule, and finally the cell itself, may be 
greatly disturbed, and produce marked symptoms, or the host may be 
destroyed. 
Fig. 59.—Theoretic Formation of Antitoxins. 
The central white area represents a molecule of a cell; the shaded portion represents the cell 
itself; the surrounding area represents the body fluids about the cell. 
r, A receptor of the molecule {first order)-, A, overproduction of receptors, which are being cast 
off; A1, a cast-off receptor free in the body fluids—now an antitoxin; A3, a molecule of antitoxin combi- 
nation with a toxin molecule T3. A*, a cast-off receptor still within the parent cell; T, a toxin mole- 
cule in combination with the receptor of a cell molecule; T2, a toxin molecule free in the body fluids; 
T3, a toxin molecule in combination with antitoxin; T*, a molecule of toxoid (toxophore group lost). 
Since the receptors joined to the toxin molecules are incapacitated or 
destroyed, the damage is repaired by the regeneration of new receptors. 
According to the reparative principles worked out by Weigert, the repair 
is not a simple replacement of the defect—the compensation proceeds far 
beyond the necessary limit; indeed, overcompensation is the rule, and this 
forms the basis of Ehrlich’s theory. If, after repair has taken place, new 
quantities of toxin are administered at proper intervals and in suitable 
quantities, the side chains that have been produced by the regenerative 
process are taken up anew in combination with the toxin, and so again the 
process of regeneration gives rise to the formation of fresh side chains. 
“The lasting and ever-increasing regeneration must finally reach a stage at HUMORAL THEORY OF IMMUNITY 
139 
which such an excess of side chains is produced that, to use a trivial ex- 
pression, the side chains are present in too great a quantity for the cell to 
carry, and are, after the manner of a secretion, handed over as a needless 
ballast to the blood. Regarded in accordance with this conception, the 
antitoxins represent nothing more than side chains reproduced in excess during 
regeneration, and therefore pushed off from the protoplasm and so coming to 
exist in the free state” (Ehrlich). 
This theory explains the specificity of the antitoxins for a given toxin; 
thus the latter causes specific chemical stimulation of the cell, which in- 
duces the formation of specific side chains—the cast-off receptors—which 
are capable of uniting with the toxin molecules free in the body fluids and 
thus neutralizing them; they are, therefore, called antitoxins. 
This theory also explains why a minute quantity of toxin is capable of 
stimulating the production of a large amount of antitoxin, and why the 
production of antitoxin persists for some time. The toxin molecule must 
be conceived as entering into the protoplasm of a body molecule and re- 
siding there for some time, acting as a stimulus to the 
cell, with consequent production of antitoxin. Dia- 
grammatic representations of this process would seem 
to show that a physical union exists between toxin and 
cell receptors, resulting in the destruction of receptor, 
which drops off and is replaced by a number of recep- 
tors that, for lack of space for attachment to the cell, 
are thrown off into the blood-stream. In reality, by 
the act of immunization certain cells of the body be- 
come converted into cells that secrete specific antitoxin, 
and, as shown by Salmonson and Madsen, the admin- 
istration of pilocarpin, which augments the secretion of 
most glands, also produces in immunized animals a 
rapid increase in the antitoxin content of the serum. 
The formation of antitoxin is constanly going on, and 
so throughout a long period the antitoxin content of 
the serum remains nearly constant. 
In the production of antitoxin the haptophore group 
is the essential and important portion of the toxin molecule. Even though the 
toxophore group is lost—and when this occurs the toxin is called toxoid 
(Fig. 60)—the haptophore group is capable of uniting with receptors and 
stimulates the production of antitoxin. In fact, in effecting immunization 
with powerful toxins it may be necessary, in the first few injections that are 
given, to convert artificially all or a portion of the toxin into toxoid, so 
that antitoxins will be produced that will protect the animal against sub- 
sequent overdoses of toxin. 
The production of antitoxins must, in keeping with this theory, be 
regarded as a function of the haptophore group of the toxin. It is easy, 
therefore, to understand why, out of the great number of alkaloids, none 
is in a position to cause the production of antitoxins, for alkaloids possess 
no haptophore group that anchors them to the cells of organs. As has been 
stated, in the formation of antitoxin the haptophore group of the toxin 
molecule is the essential portion; the toxophore group is much less impor- 
tant, and during immunization the symptoms of illness due to the action of 
the latter group are not essential to and play no part in the production of 
antitoxin. It must be said, however, that a toxin molecule with an intact 
toxophore group is more stimulating than a toxoid in which this group is 
absent; therefore, in artificially immunizing horses for the production of 
Fig. 60.—Theoretic 
Structure of a 
Molecule of Tox- 
in and Toxoid. 
1, Toxin: H, hapto- 
phore group for union 
with the receptors of 
cells or antitoxin; T, 
toxophore group. 
2, Toxoid: Same 
Structure as toxin mole- 
cule except that the 
toxophore group is 
lost. 140 
IMMUNITY.—THEORIES OF IMMUNITY 
antitoxin after the first few injections increasing amounts of toxin are 
administered. 
Antibodies of the Second Order (Agglutinins and Precipitins).—As 
new discoveries were made, Ehrlich amplified his theory of the formation 
of antibodies, but always upon the original and basic conceptions as just 
set forth. 
We have seen that the simplest molecules of food substances, toxins, and 
ferments, substances really in solution, are anchored to molecules of cell 
protoplasm by means of the simple side arms of the latter. When this 
chemical union has taken place the food or toxin may be assimilated with- 
out undergoing any further change. With more complex food substances, 
however, some preparatory treatment is necessary before they become 
available for final assimilation. The large molecule may readily enough be 
anchored to the molecule of the cell, but it probably requires some prepara- 
tion before it becomes available for the nutrition of the cell. 
Accordingly, Ehrlich assumed that the body cells are furnished with 
another order of side chains or receptors composed of two portions; one 
part or group for union with the food substance, and called the haptophore 
group; and the second portion, called the toxophore or zymophore group, 
in which the special function of the receptor resides. 
Similarly, certain pathogenic agents that are more complex than soluble 
toxins or ferments combine with receptors of this kind. One arm, the 
haptophore group of the receptor, combines with the haptophore portion 
of the pathogenic molecule, and then the second or toxophore portion of 
the receptor exerts some special action upon the attached molecule. Re- 
ceptors or haptines of this nature are known as receptors of the second 
order; antibodies of the same structure, produced and cast off into the 
blood-stream as the result of toxic injury and stimulation of body cells, are 
known as antibodies of the second order. 
Two such antibodies are well known. In one we find that the toxophore 
group of the antibody causes clumping or agglutination of its antigen, or 
the agent that caused its production, and hence this antibody is called an 
agglutinin. In typhoid fever, for example, the bacillus or one of its more 
complex products causes the production of an antibody of this nature, so 
that when the serum of a typhoid fever patient is mixed with the bacilli, 
the latter lose their motility and form clumps or agglutinated masses. This 
phenomenon was first observed by Gruber and Durham, and was applied 
in a practical way to the diagnosis of typhoid fever by Widal and Griin- 
baum. The second antibody of this class, the precipitins, resemble the 
agglutinins quite closely (Fig. 61). 
Kraus discovered that if a bouillon culture of the typhoid bacillus is 
filtered through porcelain, and a few drops of serum from a typhoid fever 
patient or from an animal immunized by injections of typhoid bacilli are 
added to a small quantity of the bacilli-free filtrate, a faint cloud will appear 
resembling in some respects that observed at the line of contact between 
nitric acid and urine that contains a trace of albumin. The toxophore por- 
tion of this antibody, therefore, appears to coagulate or precipitate soluble 
substances, and, accordingly, the antibody is known as a precipitin. As 
will be pointed out later, various protein substances, such as blood-serum, 
milk, egg-albumen, etc., may cause the production of specific precipitins. 
Antibodies of the Third Order (Hemolysins, Bacteriolysins, Cyto- 
toxins).—Still more complex molecules of food material require conversion 
into simpler substances before they may be assimilated by the molecules 
of the cell. It is essential that they undergo a sort of digestion, and ac- HUMORAL THEORY OF IMMUNITY 
141 
cordingly Ehrlich has conceived that special side arms or receptors exist 
for this purpose, these being composed of two grasping portions, or hapto- 
phore groups, one for union with the complex food molecule, the second 
for union with a special, ferment-like substance present in the blood and 
called complement. The receptor, therefore, acts simply as a connecting 
link or interbody between food molecule and complement, bringing the two 
into relation with each other when the food molecule is rendered soluble, 
i. e., undergoes lysis. 
With highly organized cell material, such as red blood-corpuscles or 
bacteria, it is found that receptors of this nature bring about their de- 
struction by lysis by attaching them to a suitable complement. During 
infections with various bacteria, therefore, we find that numerous anti- 
The central white area represents a molecule of a cell; the shaded portion represents the cell 
itself; the surrounding area represents the body fluids about the cell. 
R, Receptor of the molecule (second order)-, R2, overproduction of receptors, which are being cast 
off; A, a cast-off receptor which now constitutes the antibody; A, A-, agglutinins in combination with 
the antigen (bacilli). 
Fig. 61.—Formation of Agglutinins and Precipitins. 
bodies are produced. If the bacteria produce soluble toxins, specific anti- 
toxins are produced to counteract the effects of these; other products 
stimulate the production of agglutinins and precipitins; still other products 
or the whole cell cause the production of antibodies, which are not in them- 
selves destructive, but which have the specific power of combining with the 
cell and bringing about its lysis or destruction by bringing it into relation 
with the ferment-like complement. It is only by means of a special anti- 
body of this nature that a complement may be united with the pathogenic 
agent, i. e., the complement itself cannot act directly upon the cell, but 
must be united by means of the antibody. 
Ehrlich has termed an antibody of this nature an amboceptor, or inter- 
body. In structure, amboceptors are believed to possess two combining 142 
IMMUNITY.—THEORIES OF IMMUNITY 
or grasping portions: one, the haptophore or antigenophore group, for 
union with the cell; the other the complementophile group, for union with 
a complement (Fig. 62). 
The lysins (bacteriolysins, hemolysins, and other cytolysins) are anti- 
bodies of this order. If, for example, the erythrocytes of one animal are 
injected into an animal of a different species, hemolysins will be produced, 
the hemolysin being a specific hemolytic amboceptor that will unite corpus- 
cles of the animal used in the injection, and only these cells, with a com- 
plement, and thus bring about their solution or lysis. If certain bacteria 
(e. g., the cholera bacillus) are injected into an animal, specific bacterio- 
lysins (bacteriolytic amboceptors) will be produced. Similarly, specific 
Fig. 62.—Formation of Cytolysins (Hemolysins, Bacteriolysins, Cytotoxins). 
The central white area represents a molecule of a cell; the shaded portion represents the cell 
itself; the surrounding area represents the body fluids about the cell. 
R, Receptor of the molecule (third order)-, R2, overproduction of receptors, which are being cast 
off; A, a cast-off receptor which now constitutes the antibody of amboceptor; C, molecule of comple- 
ment free in the body cells and body fluids; A2, A4, amboceptors in combination with molecules of a 
cell (antigen), and a complement; A3, an amboceptor in combination with a molecule of a cell. The 
cell (antigen) is now said to be sensitized. Lysis does not occur because a complement is not united. 
amboceptors are produced during the course of infections with typhoid ba- 
cilli, and are largely instrumental in combating and overcoming this infec- 
tion. It is important to remember, however, that although these ambo- 
ceptors probably prepare their antigens for lysis, or, in the meaning of 
Bordet, “sensitize” them, they are not in themselves lytic, final solution 
of the antigen being accomplished by the ferment-like substance—the 
complement. 
COMPATIBILITY OF THE PHAGOCYTIC AND HUMORAL THEORIES 
When we seek to compare the theory of Metchnikoff with that of 
Ehrlich, we find that they differ only in minor details, the fundamental COMPATIBILITY OF PHAGOCYTIC AND HUMORAL THEORIES 
143 
principles not being contradictory; they may, rather, be regarded as one 
set of phenomena viewed from different aspects. 
Since its original announcement Metchnikoff has, on different occa- 
sions, enlarged upon his theory to meet certain discoveries, made chiefly 
by adherents of Ehrlich’s theory, showing the presence of substances in 
the blood-serum and other body fluids that are potent in the processes of 
immunity independent of cells. Metchnikoff claims, however, that these 
antibodies are derived from the group of cells classified as phagocytes, and 
thus would preserve the primary importance of his theory. Ehrlich, on the 
other hand, while not denying that these cells may be a source of their 
formation, points out that they are not necessarily the sole or supreme 
source, but may be formed by the general body cells or by special groups 
of cells possessing a selective affinity for the pathogenic agent. 
The theory of Ehrlich is essentially a chemical one, and maintains that 
the union of food or pathologic material with cells is a chemical union; 
his views, therefore, possess that degree of definiteness necessary to con- 
stitute a plausible chemical theory. The theory of Metchnikoff would 
explain processes of nutrition and immunity as largely founded on a physical 
basis, and is, therefore, necessarily more general, being largely biologic and 
vitalistic. 
The two theories differ in two more or less hypothetic points: (1) In 
the manner by which material enters into relation with cells, and (2) the 
relative importance of certain cells in the formation of antibodies. Other- 
wise both are intimately related, in that phagocytosis is unimportant if 
removed from the influence of antibodies in the body fluids, and these same 
antibodies, although probably formed according to Ehrlich’s theory, are 
derived in part from Metchnikoff’s phagocytes. 
Phagocytosis, whether by leukocytes, endothelial cells, or by newly 
developed connective-tissue cells, is very common, and is obviously a 
most important factor in the destruction of pathogenic bacteria and in the 
cure of infectious disease. In virulent infections, however, phagocytosis 
may not be apparent; the leukocytes are not attracted, and those in the 
vicinity undergo dissolution. Later in these infections, however, phagocy- 
tosis may become apparent, due, according to Metchnikoff, to the “adapta- 
tion” of the cells to the products of the invading micro-organism, whereby 
the weak or negative chemotaxis is converted into an active positive chemo- 
taxis with vigorous digestion. This, however, is not primarily due to in- 
creased digestive capacity of the phagocytes, but to an increase of opsonins 
in the body fluids; these opsonins prepare the bacteria for digestion. 
The original phagocytic theory did not explain the destruction of bac- 
teria within the living tissues without the intervention of leukocytes, and, 
what is even more striking, a similar destruction occurring in vitro by serum 
and other body fluids totally devoid of cells. Bacteriolysis has been shown 
to be due to two different substances—one, a thermolabile, ferment-like 
body called “cytase” by Metchnikoff and “complement” by Ehrlich, and 
the other a more specific thermostabile body, called “fixateur” by Metchnikoff 
and “amboceptor” by Ehrlich. These substances appear to play an im- 
portant role in certain infections, as, for example, in typhoid fever and 
cholera, and were studied mainly by the adherents of the side-chain theory. 
Metchnikoff recognized their existence and significance, but endeavored to 
preserve the primary importance of the phagocytic theory by claiming that 
they are products of the group of cells classified as phagocytes. Ehrlich, 
however, wffiile not denying that these cells may be one source, holds that 
they are not necessarily the sole source, but that they are products of gen- 144 
IMMUNITY.—THEORIES OF IMMUNITY 
eral cellular activity or of special groups of cells that have shown a com- 
bining affinity for the antigens. 
For example, Metchnikoff holds that there are but two complements 
—macrocytase and microcytase—and that these are formed by destruction 
or solution of macrophages and microphages. Ehrlich maintains that 
there are many complements, and that these are the excretory products of 
leukocytes, and probably of other cells as well. Ehrlich teaches also that 
specific amboceptors or fixateurs may be products of various body cells 
other than those classified as phagocytes, and Metchnikoff recognizes their 
existence, but holds that they are formed and discharged solely by the 
leukocytes or other phagocytic cells. Ehrlich has shown the manner in 
which complement and amboceptor produce bacteriolysis, and Metchnikoff 
has amplified his theory to meet these observations, to the extent that 
destruction of bacteria is recognized as being brought about either intra- 
cellularly, by the digestive action of the leukocytes, or extracellularly, by 
the enzyme-like action of the cytase, or complement, working through the 
intermediation of the fixateur or amboceptor, and that cells that are poten- 
tially phagocytic give origin to these antibodies. 
Regarding the structure of toxins and the action of antitoxins the two 
theories are divergent, and whereas Metchnikoff is inconclusive, Ehrlich 
presents definite conceptions that are well supported by experimental data. 
Metchnikoff maintains that it is the cells that absorb the “toxin” that 
furnish the antitoxin. In other words, the enzymes, as microcytase and 
macrocytase, exert their action not only upon the more complex molecules 
of micro-organisms but also upon their simpler toxins, fixing or otherwise 
altering them until they can finally be destroyed. This explanation wrould 
lead us to conclude that the nerve-cells which bind the tetanotoxin are 
capable of furnishing antitoxin, whereas experimental observations are 
absolutely opposed to this narrower view. Metchnikoff also maintains 
that antitoxin acts by stimulating the leukocytes to absorb and destroy 
toxin, whereas Ehrlich has clearly shown that antitoxin, by combining 
chemically with the toxin, neutralizes it, a process that may be shown in 
vitro entirely independent of cells. 
From what has been said it will be seen that the twTo theories are not 
essentially divergent, and that we are unwarranted in clinging to one view 
to the absolute exclusion of the other. The question rests largely on which 
of the body cells are most active in forming antibodies, and also on a recog- 
nition of the role played by phagocytosis in certain infections, such as 
staphylococcus, streptococcus, and pneumococcus infections. Ehrlich has 
attempted an explanation of the method by which body cells form anti- 
bodies, and the manner in which these antibodies overcome their antigens; 
he has placed both processes upon a chemical basis, involving no one par- 
ticular group or class of cells. Metchnikoff, on the other hand, has shown 
the important role played by phagocytosis in many infections, and claims 
that the antibodies in the circulating fluids are the products of these phago- 
cytes; he places immunity more largely upon a physical basis. 
The various phenomena of immunity cannot be ascribed either to the 
activity of the body cells or to the body fluids alone, to the total exclusion 
of the other—both are intimately concerned in the various phases of im- 
munity. 
It is, moreover, becoming more obvious that too little attention has 
been paid to the influence of the micro-organism in the phenomena of im- 
munity reactions. It is important to recognize that some bacteria are 
apparently able to immunize themselves against the combative forces of COMPATIBILITY OF PHAGOCYTIC AND HUMORAL THEORIES 
145 
their hosts, as is demonstrated by the manner in which streptococci and 
pneumococci protect themselves with a capsule and resist phagocytosis. 
Virulent strains and “resistant races” may be evolved in this manner. 
This has been demonstrated by Ehrlich with regard to the action of various 
arsenical compounds on protozoa, work that finally culminated in the 
brilliant discovery of salvarsan. Thus atoxyl may not kill all the trypano- 
somes in an infected animal, those escaping acquiring a new power of re- 
sistance to the poison and become atoxyl-resistant. The production of 
“resistant races,” not only among the protozoa but also in the class of bac- 
teria, complicates enormously the practical problems of immunity. CHAPTER VIII 
ANTIGENS AND ANTIBODIES 
ANTIGENS 
Briefly defined, antigens are substances that can cause the formation and 
appearance of antibodies in the body fluids. 
The number of substances capable of acting as antigens is very large; 
practically any foreign protein substance introduced parenterally will act 
as an antigen. For example, living bacteria, fungi and animal parasites and 
their soluble products act as antigens during infection even when the latter 
are present in the nature of an intestinal infestation; dead micro-organisms 
in the form of vaccines may also act as antigens. Any cell or body fluid 
of one animal may prove an antigen when injected into an animal of a 
different species. Indeed, the cells of a certain organ may act as an antigen 
in the same animal when displaced and set free into the circulation for trans- 
mission to other parts of the body. 
Specific and Non-specific Antigens.—When antigenic substances are 
injected into heterologous animals the antibodies or protective principles 
engendered are highly specific for the antigen as a whole or for its con- 
stituents. It is now known, however, that the injection of some unrelated 
and non-specific agent may act as a general stimulant and induce the body 
cells to throw off small amounts of antibody unrelated to the non-specific 
agent injected. This has an important bearing upon non-specific “protein 
therapy” and is discussed more completely in Chapter XXXIX. 
Nature of Antigens.—So far as is now known, antigens are colloids, and 
are usually protein in nature. Every known soluble protein may in some 
degree act as an antigen, and recent investigations would seem to show, 
although they do not definitely prove, that toxic glucosids and various 
lipoids may to some extent act in this same capacity. The protein antigens 
may be quite varied: thus antibodies are produced not only by the injec- 
tion of bacteria or their toxins, but also by erythrocytes, serums of different 
animals, egg-albumen, milk, etc. 
Of the cleavage products of proteins, it is certain that none of the amino- 
acids and simple polypeptids can act as antigens; there is, however, some 
evidence to show that the proteoses possess antigenic properties. It has 
been shown by Gay and Robertson1 that if the antigenic cleavage products 
of casein are resynthesized by the reverse action of pepsin into a protein 
resembling paranuclein, this synthetic protein is capable of acting as an 
antigen. Protamins and globin were found to be non-antigenic, although 
globin combined with casein formed a compound of antigenic power in that 
it produced an antibody yielding complement-fixation reactions with 
globin. 
Whether the entire protein molecule, or only groups thereof, deter- 
mine the characteristics of the antigen and the antibody is not definitely 
known. Wells and Osborne2 have recently submitted evidence showing 
that a single protein molecule can act as an antigen and produce more 
than one antibody. 
1 Jour. Biol. Chem., 1912, 12, 233; Jour. Exp. Med., 1912, 16, 479; 1913, 17, 535. 
2 Jour. Infect. Dis., 1913, 12, 341. 
146 ANTIGENS 
147 
Non-protein Antigens.—Ford1 was able to immunize rabbits by in- 
jecting a toxic glucosid contained in extracts of Amanita phalloides, pro- 
ducing an antibody antihemolytic for the hemolysin of Amanita when 
diluted 1 ; 1000. Abderhalden and others have found that specific enzy- 
motic substances appear in the blood of animals injected with carbohy- 
drates and fats. Recent developments in immunologic research would 
indicate, therefore, that a complex toxic glucosid that can be hydrolyzed 
by enzymes may act as an antigen. 
The intimate relationship of lipoids to complement-fixation reactions, 
especially in syphilis, has naturally led to investigations regarding the 
possibility of lipoids acting as true antigens. In testing for the Wasser- 
mann reaction the use of lipoids in the form of tissue extracts to serve as 
an antigen does not mean that it is a true antigen; in fact, experimental 
work indicates quite strongly that these lipoidal substances are incapable 
of producing antibodies when injected into animals. 
Much and others have worked with lipoids secured from a streptothrix, 
and which is called “nastin,” and they assert that this substance may be 
used in immunizing animals with the production of an antibody yielding 
complement fixation, with nastin as the antigen. Similar results have 
been described for the fatty substances from tubercle bacilli (“tuberculo- 
nastin”). Kleinschmidt2 accepts the antigenic nature of nastin in reactions, 
but was unable to secure antibodies by immunizing rabbits with this sub- 
stance. 
Ritchie and Miller3 could demonstrate no antigenic activity in the 
lipoids of serum or corpuscles. Thiele4 calls attention to the fact that 
lipoids possess no specificity, and that they cannot act as antigens with 
the production of antibodies. On the other hand, Meyer5 has reported 
the production of specific complement-fixation antibodies by immuniz- 
ing rabbits with acetone-insoluble lipoidal substances obtained from various 
teniae. He has also found the acetone-insoluble fraction of tubercle bacilli 
to serve as antigens in complement-fixation reactions with antibodies of the 
tubercle bacillus, and much more effectively than with the protein residue 
of the bacilli. Beigel6 has observed that after injecting lecithin in rabbits 
an increase occurs in the lipase content of the blood and tissues, writh the 
presence of complement-fixing antibodies, and Jobling and Bull7 have 
found an increase in serum lipase after immunizing with red corpuscles. 
It will be noticed, therefore, that the results of various investigations 
regarding the true antigenic properties of lipoids are not in accord. It 
should be emphasized that the complement-fixation reaction does not con- 
stitute a reliable index to the study of this problem, as so little is understood 
of the actual nature of this reaction itself. That lipoids serve a very im- 
portant purpose in the absorption or fixation of complement in vitro, as is 
so wrell demonstrated in Wassermann’s reaction for syphilis, is undoubtedly 
true, but this does not indicate that the antibody in the blood-serum of 
syphilitics is in the nature of a true lipoid antibody, and, indeed, investiga- 
tion on this subject wrould seem to indicate that it is not.8 
1 Jour. Infect. Dis., 1907, 4, 541. 
2 Berl. klin. Wchnschr., 1910, 47, 57. 
3 Jour. Path, and Bact., 1913, 17, 427. 
4Ztschr. f. Immunitatsf., 1913, 16, 160. 
5Ztschr. f. Immunitatsf., 1910, 7, 732; 1911, 9, 530; 1912, 16, 355. 
6 Deut. Arch. f. klin. Med., 1912, 106, 47. 
7 Jour. Exper. Med., 1912, 16, 483. 
8 Further discussion on the question of lipoids acting as antigens will be found in Chapter 
XXVII in a consideration of anaphylactogens. 148 
ANTIGENS AND ANTIBODIES 
It will be understood, therefore, that the question of substances other 
than proteins acting as true antigens must be regarded as an open one, 
requiring further investigation. The relation of proteins, however, to the 
production of antibodies has been fully established, and is at present receiv- 
ing renewed attention through the researches of Vaughan and his co-workers 
and Abderhalden. As has been stated in a previous chapter, Vaughan re- 
gards the protein constituents of bacterial and other calls as the main 
antigenic principle capable of causing the production of specific proteolytic 
ferments, which split the new bacterial protein, releasing a toxic product 
responsible for the symptoms and lesions of the infection. Abderhalden 
has also demonstrated the presence of proteolytic ferments in the blood- 
serum after experimental immunization with proteins, and in the serum 
of pregnant women, due to the antigenic stimulation of syncytial cells, 
capable of splitting their substrata in vitro into amino-acids and other 
simple cleavage products. These investigations serve to show the intimate 
relation that proteins bear to the problems of infection and immunity, 
and demonstrate that antibodies may be largely in the nature of ferments, 
and that immunologic reactions, both in the living tissues and in the test- 
tube, are largely in the nature of disintegrative enzymic processes. 
Drugs as Antigens.—A large amount of work has been devoted to the 
subject of formation of protective substances to morphin, cocain, strychnin, 
and other alkaloids in explanation of tolerance to these drugs. Gioffredi1 
and Hirschlaff2 claimed to have produced antimorphin sera of protective 
value, but Morgenroth,3 in a particularly thorough set of experiments, showed 
that such sera could not be produced. More recently Pellini and Green- 
field4 have reached the conclusion that no substance is formed in the blood- 
serum of a human being who has acquired a high tolerance to morphin, which 
is capable of conferring any degree of immunity to the toxic action of mor- 
phin on an animal into which it is injected. Likewise they have shown 
that the blood of a tolerant animal does not contain any protective sub- 
stance against morphin. 
ANTIBODIES 
The term “antibody” is used to designate the specific bodies produced by the 
cells of a host in reaction against an antigen, as an infecting microparasite 
and its products or other foreign protein. 
Various kinds of antibodies may be produced by the same antigen and 
by different antigens. Some neutralize the soluble toxin of the antigen 
(antitoxin); others agglutinate or precipitate their antigens (agglutinins and 
precipitins); still others cause complete dissolution of the antigen (hemo- 
lysins, bacteriolysins, etc.), and others again may so alter the antigen and 
lower its resistance as to render it more easily phagocytable by the body 
cells (opsonins or bacteriotropins). 
Tissues Concerned in the Production of Antibodies.—A large amount 
of experimental work has been conducted in the study of the problem of 
where in the body the antibodies are formed that develop in response to 
immunization. The recent investigations of Hektoen and Curtis,5 Gates,6 
and Launoy,7 who studied the effect on antibody production of the removal 
1 Archiv. ital. d. biol., 1897, 28, 402; ibid., 1899, 31, 398. 
2 Berl. klin. Wchn., 1902, 39, 1149, 1177. 
3 Berl. klin. Wchn.. 1903, 40, 471. 
4 Arch. Int. Med., 1920, 26, 279. 
5 Jour. Infect. Dis., 1915, 17, 409. 
6 jour. Exper. Med., 1917, 27, 725. 
7 Ann. de i’Inst. Pasteur, 1915, 29, 213. ANTIBODIES 
149 
of various organs, and Hektoen1 and Simonds and Jones,2 on the influence 
of exposure to rr-rays, and of Simonds and Jones,3 on the effect of injections 
of benzol, indicate that the mechanisms concerned for the production of 
antibodies are quite secure from certain disturbances and are principally 
located in the leukocytes and blood-forming organs, as the spleen, lymphatic 
tissues, and bone-marrow. 
Carrel and Ingebrigsten4 have observed the formation of hemolysin for 
goat erythrocytes in cultures of guinea-pig’s bone-marrow and lymph- 
gland in vitro. 
Gowan,5 however, does not believe that the spleen, leukocytes, lymphatic 
apparatus, thyroid, or kidney are principally concerned in antibody pro- 
duction; extirpation experiments did not appear to delay the production of 
antibody. For various reasons, including the well known function of the 
liver for removing foreign material from the circulation and its great meta- 
bolic activity, Gowan believes that this organ is deserving of more attention 
from the possibility of playing an important role in the formation of anti- 
bodies. 
In leukemia the presence of very large numbers of leukocytes and hy- 
peractivity of the leukocytogenic tissues would suggest the possibility of 
increased antibody production and enhanced resistance to infection. Just 
the reverse, however, is known to occur. In regard to antibody production 
both Rotky6 and Howell7 have found that the leukemic patient produces 
antibodies poorly or not at all, probably due to excessive proliferative 
changes in the hematopoiectic tissues. 
Of further interest in this connection is the question of local production 
of antibodies, that is, by the tissues at the site of infection. Romer8 found 
that the tissues of the conjunctiva could apparently produce antiabrin, that 
is, developed a tolerance for abrin ascribed to the production of an anti- 
toxin for this substance. Hektoen,9 however, could find no evidences of 
local production of hemolysins for goat and rat erythrocytes following the 
injection of these cells into the anterior chamber of the eye of the dog or into 
the tissues of the pleura. Recently Klauder and myself10 observed the com- 
plement-fixing antibody of syphilis in secretions from chancres in sufficient 
amounts to give strongly positive Wassermann reactions before the blood- 
serum reacted positively, indicating a local production of antibody in 
syphilis which may possess some practical diagnostic value. It is highly 
probable that in some infections and particularly those of a chronic char- 
acter, as syphilis and tuberculosis, there is some production of antibodies 
in the local tissues about the lesions. 
Distribution of Antibodies in the Body Fluids.—The concentration of 
both natural and immune antibodies varies considerably in the different 
body fluids, largest amounts being found in the blood. 
As between the plasma and serum of the blood, the investigations of 
Dreyer and Walker11 have indicated that in the rabbit larger amounts of 
immune agglutinin for typhoid bacilli occur in the plasma than in the serum. 
I Jour. Infect. Dis., 1915, 17, 415. 
;Jour. Med. Research, 1915, 33, 183. 
3 Jour. Med. Research, 1915, 33, 197. 
4 Jour. Amer. Med. Assoc., 1912, 58, 477. 
5 Jour. Path, and Bacteriol., 1911, 15, 262. 
6 Zentralbl. f. Innere. Med., 1913, 35, 953. 
7 Archiv. Int. Med., 1920, 26, 706. 
8 Arch. f. opthal., 1901, 52, 72. 
9 Jour. Infect. Dis., 1911, 9, 103. 
10 Arch. Dermat. and Syph., 1922, 5, 566. 
II Jour. Path, and Bacteriol., 1909, 14, 39. 150 
ANTIGENS AND ANTIBODIES 
In a thorough and painstaking study of this subject by Watanabe,1 in my 
laboratory, it was found that both natural and immune antibodies exist in 
plasma and serum in equal degree. Watanabe also devoted much time and 
study to the question of preformed complement in the plasma, the results 
of his studies indicating that complement exists in the plasma in practically 
the same concentration as in the corresponding serum. 
Investigations by Becht and Greer,2 Hektoen and Carlston3 have shown 
that natural and immune antibodies of various kinds in the dog are present 
in largest amounts in the blood followed in order by the thoracic lymph, 
neck lymph, pericardial fluid, cerebrospinal fluid, and aqueous humor of the 
eye. Immunization increased the concentration of antibodies in these 
fluids in the same order. 
Antibodies are commonly found in inflammatory exudates including 
blister fluid. In the secretions, as urine, saliva, tears, and milk, they are 
generally absent unless traces are found during periods of unusual activity, 
as the syphilis antibody in the milk of a syphilitic woman during the first 
few days of lactation. Antibodies are likewise usually absent from trans- 
udates, as pleural and ascites fluid unless there are relatively large amounts 
in the blood at the same time. The cerebrospinal fluid is usually free of 
complement and antibodies during health, and in this respect resembles 
other transudates; in syphilis, however, the complement-fixing and other 
antibodies are commonly present and, indeed, may be found in this fluid 
when absent from the blood, indicating the possibility of local production 
of these antibodies in the tissues of the central nervous system. 
Chemical Nature of Antibodies.—Owing to the impossibility of obtain- 
ing antibody in absolutely isolated and pure form accurate chemical anal- 
yses have not been possible. Furthermore, they are comparatively unstable 
and easily altered by strong chemical reagents. Consequently, our knowl- 
edge of the chemistry of antibodies is largely based upon indirect chemical 
and biologic analyses, and upon reactions analogous to those of known 
chemical substances. 
It has long been known that antibody and, notably, diphtheria anti- 
toxin is apparently associated with certain fractions of serum and par- 
ticularly the globulins, obtained by salting out methods. These methods, 
however, are inadequate for determining the true nature of antibodies owing 
to the absorption of the latter by serum proteins. 
The studies of Huntoon, Masucci, and Hannum,4 on the nature of anti- 
body contained in antipneumococcus serum, have proved unusually inter- 
esting and valuable; these investigators do not believe that this antibody 
belongs to the serum proteins, and have drawn the following conclusions 
on the nature of antibody from the results of their experiments: 
1. The antibody molecules are of large size, not been dialysable, and 
indicating their colloidal nature. 
2. Antibodies are not affected by trypsin over considerable "periods, 
indicating either that they are not protein in nature, or have been racemized 
by the dilute alkali used, or belong to the peptid group having a carboxyl- 
amino linkage. 
3. Antibodies are not precipitated by solutions containing little or no 
electrolytic content, indicating that they are not of an euglobulin nature. 
4. Antibodies are not soluble in ether; therefore they are not of the lipin 
group. 
5. Antibodies free from any gross amount of globulins are not pre- 
1 Jour. Immunology, 1919, 4, 77. 
2 Jour. Infect. Dis., 1910, 7, 127. 
3 Jour. Infect. Dis., 1910, 7, 319. 
4 Jour. Immunology, 1921, 6, 185. ANTIBODIES 
151 
cipitated or affected by a short exposure to 30 per cent, sodium chlorid 
solution, indicating that they are not of a pseudoglobulin nature. 
6. Antibodies are not injured by certain dilute alkalies or acids. 
7. Antibodies are not affected by temperature up to 60° C. Higher 
temperatures progressively destroy or alter their nature. 
The Antigen-antibody Reaction; Antigen-antibody Equilibrium; Dis- 
sociation of Antigen and Antibody.—The laws governing the reactions be- 
tween antigen and antibody have never been satisfactorily worked out. 
Numerous investigations have established certain facts bearing upon the 
interaction of antigen and antibody, indicating that the mechanism is one 
of physical chemistry and particularly of colloidal reactions. 
When antigen and antibody are brought together in solution under 
proper conditions, there is an immediate tendency for union which may 
bring about demonstrable physical changes, as in the precipitin and agglu- 
tination reactions. In other words, antigen and antibody cannot coexist 
in vitro without uniting, although the union may not destroy the identity 
of either. For example, apparently neutral mixtures of diphtheria toxin 
and antitoxin when injected into animals may result in the dissociation of 
toxin capable of exercising antigenic functions, which is the basis of von 
Behring’s method of active immunization in diphtheria. Weil1 has shown 
that a precipitate formed by the interaction of serum antigen and pre- 
cipitin may be dissociated by various means in vitro with the recovery of 
both antigen and antibody. Similar studies by Bordet,2 Morgenroth,3 
Muir,4 and Landsteiner5 have shown that hemolysins and agglutinins already 
in union with corpuscles may act upon fresh corpuscles, indicating that 
fixed antibody may attack fresh antigen. 
Such studies indicate, therefore, that antigen and antibody do not 
exist together in the blood in a free state; however, they may coexist in some 
form of loose combination easily dissociated both in vivo and in vitro without 
destroying the identity of either. Whether or not they may coexist without 
union in body cells is an open question upon which there are no established 
facts to warrant discussion. 
As previously stated, the union of antibody with antigen may be broken 
up in vitro by certain procedures with the recovery of antigen or antibody, 
or both, in apparently unchanged condition. Weil has experimented very 
extensively upon this question of equilibrium and dissociation of antigen 
and antibody, employing the precipitin and anaphylactic reactions. He has 
shown that antigen and antibody may be recovered from precipitate by 
extractions with salt solution, solutions of sodium carbonate, and with 
trypsin and leukocytes. Gay and Chickering6 and Chickering7 have util- 
ized similar principles for concentrating antipneumococcus serum by carry- 
ing down the antibodies in a precipitate for the recovery of antibody. 
More recently Huntoon and Etris8 have studied the question of dissocia- 
tion of antigen and antibody in vitro very extensively for the double pur- 
pose of securing antibody in solution as pure as possible free of serum con- 
stituents for chemical studies and for the preparation of antibody for thera- 
peutic purposes in more concentrated form and free of the sensitizing action 
1 Jour. Immunology, 1916, 1, 35. 
2 Ann. de l’Inst. Pasteur, 1900, 14, 257. 
3 Miinch med. Wchn., 1903, 1, 61. 
4 Studies in Immunity, London, 1909, 13. 
6 Miinch. med. Wchn., 1902, xlix, 1905. 
'’’Jour. Exper. Med., 1915, 21, 389. 
7 Jour. Exper. Med., 1915, 22, 248. 
8 Jour. Immunology, 1921, 6, 123. 152 
ANTIGENS AND ANTIBODIES 
of serum. They have worked with various antigens and antibodies with 
special reference to the pneumococcus, and have found that various substances 
may dissociate antigen and antibody, as treatment with 10 per cent, sac- 
charose solutions, 5.6 per cent, dextrose solutions, distilled water, normal 
salt solutions, ammonium carbonate solutions, and salt solution containing 
a small amount of sodium carbonate. They doubt that antipneumococcus 
antibody can be dissociated absolutely free; it would appear that the anti- 
body is attached to bacterial fragments still present in the solutions, as 
filtration removes some antibody, although in solutions containing a small 
amount of alkali, as sodium carbonate, the antibody may be free, inasmuch 
as these may be filtered without loss of antibody. As a result of these 
studies Huntoon and his associates have been able to prepare solutions of 
protective and curative antipneumococcus antibody that contained so 
little protein as to yield negative or indefinite protein reactions and which 
failed to sensitize guinea-pigs or did so very irregularly. Furnhata,1 working 
on the chemical nature of hemagglutinins, has concluded that they are 
colloidal substanes closely associated with protein substances in serum, but 
are not to be considered as belonging to proteins in the ordinary sense. 
These results are full of promise for the improvement of serum therapy, and 
particularly in relation to elimination of sensitization and anaphylactic 
reactions by serum proteins. 
Transmission of Antibodies and Immunity.—Transmission in Colostrum. 
—The transmission of antibodies has always been a subject of much interest 
which has attracted considerable attention, but with conflicting statements 
based upon the results of experimental work. Authors have not always 
differentiated between the transmission of immunity and antibodies which 
are not entirely analogous. 
Transmission of antibodies from mother to fetus is only possible through 
the placental circulation; after birth transmission may take place through 
the ingestion of colostrum and milk. 
The greater number of investigations have been conducted with the 
agglutinins for bacteria and erythrocytes, and the results have generally 
been negative, that is, these antibodies have not been found in the serum of 
the fetus or newborn animal or have been found in much smaller amounts 
than in the serum or milk of the mother; this conclusion was reached by 
Gruenbaum,2 Halban,3 Schenk,4 and others for human beings, and by Kraus 
and Loew,5 Park,6 Luedka,7 v. Eisler and Sohma8 for the lowrer animals. 
Fellenberg and Doell9 found that the serum of a child sometimes shows a 
stronger, sometimes a weaker, reaction than that of the mother, and they 
decline to believe in any constant ratio between them. Tunnicliff10 showed 
that the opsonic power of serum for various bacteria was less at birth than 
during adult life, and that opsonins decreased during the first month of life. 
Reymann,11 in a study of bacterial and hemoagglutinins in kids, found 
that they were absent from the serum at birth, although normally present in 
the blood of the mother goats. As a general rule they were present in large 
1 Japan Med. World, 1921, 1, 1. 
2 Miinch. med. Wchn., 1897, 44, 330. 
3 Wien. klin. Wchn., 1900, 13, 545. 
4 Monat. f. Geburtsh., 1904, 19, 159, 344, 568. 
8 Wien. klin. Wchn., 1899, 12, 95. 
6 Proc. Soc. Exper. Biol, and Med., 1903, 1, 19. 
7 Centralbl. f. Bakteriol., orig., 1904, 37, 288, 418. 
8 Wien. klin. Wchn., 1908, 21, 684. 
9 Ztschr. f. Geburtsh., 1913, 75, 285. 
10 Jour. Infect. Dis., 1910, 7, 698. 
11 Jour. Immunology, 1920, 5, 227. ANTIBODIES 
153 
amounts in the colostrum, from which it would appear possible for them to 
be transmitted to the young by nursing. Howell and Eby1 found that the 
offspring of immunized rabbits, as a rule, have antibodies in their serum 
which persist in appreciable but decreasing amounts for four to six weeks. 
From these investigations it would appear, therefore, that normal or 
natural antibodies are not usually transmitted from mother to offspring 
during intra-uterine life; immune antibodies produced in the mother by an 
attack of disease or by vaccines may be transmitted, although this is the 
exception rather than the rule. After birth antibodies may be trans- 
mitted by colostrum and milk, although they may rapidly disappear from the 
blood of the young. It is well known that a mother cannot transmit im- 
munity to smallpox, although a child may be borne immune by reason of 
an intra-uterine attack of this disease. Furthermore, the apparent im- 
munity of a child of a syphilitic mother to syphilis is now believed due to 
infection of the child with this disease. It would appear, therefore, that 
placental transmission of natural and immune antibodies is a rare occur- 
rence; when antibodies are found in the blood of the fetus or the newborn 
it is more likely that there has been a placental transmission of antigen 
with stimulation of the fetal tissues independent of those of the mother. 
In diphtheria, however, there is apparently a natural transmission of 
antitoxin, inasmuch as the great majority of infants are immune to this dis- 
ease and yield negative Schick reactions. The majority of adults are like- 
wise immune, and presumably the mother may transmit an immunity good 
for at least one or two years. After that time the majority of children 
become susceptible to diphtheria and some later again acquire an immunity. 
Whether the child of a Schick positive mother is susceptible to diphtheria 
on the basis of the Schick reaction cannot be stated; studies to determine 
this point have not been made. 
No doubt the colostrum is of considerable importance in relation to 
acquired resistance of the newborn. Little and Orcutt2 have recently 
shown that the blood of newborn calves did not contain agglutinins for 
Bacillus abortus until after colostrum had been taken. Howe3 and Orcutt 
and Howe4 have further shown that certain of the blood proteins are missing 
in calves until they have ingested colostrum containing these globulins. 
This tends to show that both protective and nutritive principles in colostrum 
are absorbed, and probably colostrum and the mother’s milk are of con- 
siderable importance in this relation not only as foods, but as transmitters 
of resistance against infection and of vitamins concerned in nutrition. 
Smith and Little5 have further shown that the calf deprived of colostrum 
lacks something that prevents intestinal bacteria from invading the body 
and multiplying in the various organs, especially bacilli of the colon group 
regarded as producing diarrhea (“scours”) and multiple arthritis. Smith 
and Little have found these infections more common among calves deprived 
of colostrum, and they conclude that the function of the colostrum is 
essentially protective against miscellaneous bacteria that are harmless later 
on when the protective functions of the calf have become more effective. 
Immunity may be transmitted: (1) By the placental passage of anti- 
bodies; (2) by the passage of antibodies in colostrum or milk; (3) by the 
active immunization of both mother and offspring by the same immunizing 
1 Jour. Infect. Dis., 1920, 27, 550. 
1 Proc. Soc. Biol, and Med., 1922, 19, 331. 
3 Jour. Biol. Chem., 1921, 49, 115. 
4 Jour. Exper. Med., 1922, 36, 181. 
5 Jour. Exper. Med., 1922, 36, 181. 154 
ANTIGENS AND ANTIBODIES 
agent, or (4) by direct transmission in the germ plasma of the parents. The 
latter is probably the mechanism concerned in the transmission of the 
natural immunity of races and species to certain diseases, and of which we 
know little or nothing of the underlying principles. 
The Influence of Nutrition, Drugs, and x-Rays Upon Antibodies and 
Immunity.—The influence of nutrition upon immunity is particularly well 
seen among the young, the child being nourished with mother’s milk gen- 
erally escaping the numerous infections to which the undernourished bottle- 
fed baby is subject. Errors in diet among artificially fed children may 
readily favor infection not only of the intestinal tract, but of the skin and 
other tissues as well, although the function of antibody production may 
not be interfered with as indicated by the results of experiments conducted 
by Hektoen,1 who found that rats fed on pure vegetable proteins (the stunt- 
ing food of Osborne and Mendel) reacted to antibody production about as 
well as fully nourished rats. 
Previous reference has been made to the influence of drugs upon phago- 
cytosis (see page 129), 
Numerous investigations within recent years have indicated that cer- 
tain drugs may induce a state of temporary immunity to trypanosome in- 
fections by stimulating the antibody-producing tissues, the leukocytic 
mechanism, or both, or combine with antibodies and render the latter 
more active. 
Ehrlich and Shiga2 have shown that mice infected with caderas and 
treated with one or more injections of trypan-red, developed a temporary 
immunity which could not be ascribed to an antibody response following 
infections with the parasites alone or to the presence of unexcreted dye, 
but rather to the presence of antibodies in response to the stimulating 
influence of the drug; later Ehrlich3 demonstrated the same phenomenon 
with T. brucei and Halberstaedter4 in similar studies found the immunity 
highly specific, that is, mice infected with dourine and treated with trypan- 
red developed an immunity to dourine alone and not to other trypanosomes, 
as T. brucei or vice versa. Corroborative evidence of the apparent effect 
of this and other drugs upon antibody production was given later by the 
extensive work of Terry,5 who found that a strong immunity against surra 
of India was obtained by injecting mice with dyes either alone or in com- 
bination with acetylatoxyl. That the action of the drugs is indirect rather 
than wholly trypanocidal, was seemingly shown by the fact that large 
intraperitoneal injections of surra and caderas were capable of infecting 
mice when introduced as early as twenty-four hours after the drug and 
before the latter had been wholly excreted. 
Further indications of the possible important relation of drugs to im- 
munity is shown in the reports of several homeopathic physicians, as in 
that Watters,6 who claimed that the administration of calcium sulphid 
increased the opsonic index to staphylococci; of Mellon,7 who found that 
the administration of baptisia influences favorably the production of group 
agglutinins for typhoid and other closely related bacteria and that veratrum 
viride increased the opsonic index to pneumococci; of Wheeler,8 who claims 
1 Jour. Infect. Dis., 1914, 15, 245. 
2 Berl. klin. Wchn., 1904, 41, 329, 362. 
3 Berl. klin. Wchn., 1907, 44, 233, 280, 310, 341. 
4 Centralbl. f. Bakt., orig., 1915, 38, 525. 
5 Monograph No. 3, Rockefeller Inst. 
6 North Amer. Jour. Homeop., 1909, 24, 460. 
7 Med. Century, 1913, 20, 261. 
8 Brit. Homeopath. Jour., 1914, 4, 243. ANTIBODIES 
155 
that phosphorus increases the opsonic index of human serum to the tubercle 
bacillus; of Wesselhoeft,1 whose experiments were interpreted as indicating 
the curative effects of quinin in malaria, could not be ascribed entirely to 
its parasiticidal activity, but probably in part to a favorable influence upon 
the production of antiplasmodial antibodies; and of Hooker,2 who showed 
that the administration of phosphoric acid, arsenious anhydrid, and mer- 
curic chlorid homeopathically to normal persons, resulted in the elaboration 
of agglutinins and complement-fixing antibodies for Bacillus typhosus, B. 
paratyphosus A and B, and B. dysenteriae. In several of these investigations 
the drugs alone were administered to healthy persons, and the appearance 
of an increase of certain group antibodies in the blood-serum was inter- 
preted as an increase of normal or natural antibody, and an indication of the 
possible stimulating influence of these drugs upon antibody-producing tis- 
sues and a means of their curative value in certain diseases. Probably 
pilocarpin has received more attention than any other drug in this connec- 
tion and particularly in relation to the production of diphtheria antitoxin. 
It is now generally accepted, however, that pilocarpin and other drugs 
have little or no influence upon the production of antibodies. Hajos and 
Sternberg3 have recently shown that pilocarpin, adrenalin, atropin, strophan- 
thus, sodium salicylate, antipyrin, potassium chlorate, calcium chlorid, 
etc., have no influence upon the production of typhoid agglutinins and 
hemolysins in the rabbit; Joachmoglu and Wada4 have likewise found that 
atropin and pilocarpin have no influence upon the production of typhoid 
agglutinins by the rabbit. 
Following the introduction and encouraging results of arsenical com- 
pounds in the experimental chemotherapeusis of protozoan infections, 
several investigators have studied their possible influence upon antibody 
production and particularly the influence of dioxydiamidoarsenobenzol 
(salvarsan), with the result that a general impression exists that part of the 
curative influence of dioxydiamidoarsenobenzol in spirochetic and trypan- 
osome infections is to be ascribed to the influence of the drug in stimulating 
the production of protective and curative antibodies in addition to its 
powerful parasiticidal activity. Aggazzi5 found that arsenious acid, atoxyl, 
and arsenophenylglycin increased the output of typhoid agglutinin; Fried- 
berger and Masuda6 claim that salvarsan increases the content of normal 
agglutinins and hemolysins in the serum; Boehncke7 found that the admin- 
istration of salvarsan may be followed by an increase of diphtheria anti- 
toxin and of various bacteriolysins, opsonins, and precipitins, but not of 
complement-binding substances; Weisbach8 also claims that the adminis- 
tration of salvarsan results in an increase of agglutinin and hemolysin, while 
Reiter9 was unable to note any such influence, his experiments indicating 
that large doses of the drug lowers resistance to various bacteria. 
As further indications of the probable important relation of certain 
drugs to immunity are several reports indicating that their administration 
may be followed by an increase of complement in the serum. Weil and 
Duport10 have reported that the intravenous administration of sodium 
1 New England Med. Gaz., 1913, 48, 64, 637. 
2 New England Med. Gaz., August, 1914. 
3 Ztschr. f. Immunitatsf., 1922, 34, 218. 
4 Arch. f. Exper. Path, and Pharmakol., 1922, 93, 269. 
5 Ztschr. f. Immunitatsf., orig., 1909, 1, 736. 
6 Therap. Monatschr., 1911, 25, 288. 
7 Ztschr. f. Chemotherap., 1912, orig., 136. 
8 Ztschr. f. Immunitatsf., orig., 1914, 21, 187. 
9 Ztschr. f. Immunitatsf., orig., 1912, 15, 116. 10 Compt. rend. Soc. Biol., 1913, 74, 802. 156 
ANTIGENS AND ANTIBODIES 
bicarbonate to rabbits resulted in an increase of serum complement; Fenv- 
vessy and Freund1 claim similar results with the intravenous administration 
of calcium chlorid, and Ciuca2 found that the injection of tartar emetic and 
salvarsan was followed by an increase of serum complement in normal and 
trypanosome-infected animals, wThile the administration of atoxyl caused 
a decrease of complement in the serum of normal animals and in a propor- 
tion of trypanosome-infected animals. 
In the experiments by Toyama and myself,3 while massive doses of ar- 
sphenamin and mercuric chlorid tended to suppress antibody production 
and cause a decrease in complement, smaller doses tended to increase the 
production of agglutinins and augment the complement content after a 
primary decrease. 
Perhaps no drug has attracted as much attention as alcohol in its effects 
upon resistance to disease. The investigations of Muller,4 Wirgin,5 Laitinen,6 
and others indicate that alcohol in mildly intoxicating amounts for several 
days after the injection of an antigen restrains the formation of antibodies. 
More recently Reich7 has observed that chronic alcoholism tends to lower 
the bactericidal and phagocytic activity of the blood for typhoid bacilli 
and to diminish the resistance of red blood-corpuscles to hypotonic salt 
solution. Data of this kind shows, therefore, that the prolonged adminis- 
tration of alcohol tends to reduce antibody production and to lower bodily 
resistance to disease. 
As previously stated (page 148) drugs, as morphin, cocain, atropin, arsenic, 
and strychnin, do not produce antibodies. While it is common clinical ex- 
perience to find addicts highly tolerant to the effects of these alkaloids, stud- 
ies with their sera have generally shown that protective substances capable 
of neutralizing the effects of the drug are not to be found. 
Since x-rays may have a direct and destructive action on lymphocytes, 
the lymphoid, and myeloid tissues, several investigators have studied the 
influence of these rays upon antibody production. Benjamin and Sluka8 
found that in rabbits exposure to the x-ray before the injection of beef-serum 
diminished very much the production of precipitins; Lawen9 also observed 
that the formation of bacterial agglutinins and lysins was retarded by 
exposure to x-rays. Murphy and Ellis10 have found that after exposure to 
the x-ray mice (normal and splenectomized) became more susceptible to 
bovine tuberculosis than normal animals. Murphy and Taylor11 found that 
mice exposed to the Roentgen rays become more susceptible to tumor 
transplants, apparently due to the action of the rays upon lymphocytes. 
Hektoen12 found that prolonged exposure of the white rat to the Roentgen 
ray markedly reduced the production of hemolysin for sheep corpuscles, 
due, it was assumed, to the destructive action on the lymphatic tissues, 
the spleen, and the bone-marrow. Similar experiments with the dog and 
rabbit yielded the same results. Simonds and Jones13 found the formation 
1 Ztschr. f. Immunitatsf., orig., 1913, 18, 666. 
2 Bull. d. 1. Soc. Path. Exot., 1914, 7, 626. 
3 Jour. Immunology, 1918, 3, 301. 
4 Archiv. f. Hyg., 1904, li, 368. 
5 Centralbl. f. Bakt., 1905, 38, 200. 
6 Ztsch. f. Hyg., 1907-1908, lviii, 139. 
7 Arch. f. Hyg., 1916, lxxxiv, 337. 
8 Wien. klin. Wchn., 1908, 21, 311. 
9 Mitt. a. d. Grenzgeb. d. Med. u. chir., 1909, 19, 141. 
10 Jour. Exper. Med., 1914, 20, 397. 
11 Jour. Exper. Med., 1918, 28, 1. 
12 Jour. Infect. Dis., 1915, 17, 415. 
13 Jour. Med. Research, 1915, 33, 183. ANTIBODIES 
157 
of agglutinins appreciably lowered after exposure to x-rays, although bac- 
teriolysins, opsonins, and complement-fixing antibodies were less affected or 
not at all. 
These observations harmonize with the view that antibodies are largely 
produced in the spleen, lymphoid tissues, and bone-marrow, as these struc- 
tures suffer most directly from the action of the Roentgen ray when applied 
in large amounts. After these primary effects have passed away the power 
to produce antibodies may actually increase, due, as Hektoen suggests,1 
to regenerative changes in the spleen and lymphatic glands. Particular 
interest is attached to the influence of x-rays upon tuberculosis; as pre- 
viously stated, Murphy found heavy exposures to increase the susceptibility 
of guinea-pigs, similar results being reported by Morton.2 Kellert3 and 
Corper,4 however, failed to note any injurious effects from single exposures, 
so that the effects of the x-rays are probably in accordance to dosage and 
exposure. Small amounts of these rays may be beneficial, as evidenced by 
their favorable influence upon the treatment of tuberculosis of lymph- 
glands, furunculosis, and other bacterial infections, even though the rays 
are not bactericidal. 
Influence of Temperature Upon Antibody Formation.—While exposure 
to cold is generally believed to be an important etiologic factor in the pro- 
duction of certain diseases, and especially those of the respiratory tract, only 
a small amount of work has been devoted to influence of temperature upon 
antibody production and leukocytic activity. 
Graziani5 and Fukuhara6 found that chilling rabbits was usually followed 
by increased antibody production, while Trommsdorff7 and Lissauer8 re- 
ported just the opposite results, namely, that chilling reduced antibody pro- 
duction. Roily and Meltzer9 and Ludke10 found that typhoid agglutinins 
and lysins are produced more rapidly and abundantly in rabbits that are 
kept overheated than in those which are kept cool. Foord11 has found that 
chilling rabbits twice a day during a period of immunization for seven to 
ten minutes at 8° C. did not influence hemolysin production, although chill- 
ing was accompanied by slight increase in agglutinin production for typhoid 
bacilli. 
Specificity of Antibodies.—Antibodies are usually specific for their an- 
tigen, and it is upon this general law that the reactions of immunity are 
based. It should be remembered, however, that not all antibodies are 
protective; the agglutinins, for instance, apparently do not injure their 
antigen. On the other hand, an animal may enjoy an immunity without 
demonstrating the presence of any antibody in the body fluids, and another 
animal may show antibodies generally considered as possessing protective 
powers, as, for example, the bacteriolysins, without necessarily being 
immune. 
Upon what does the specificity of antibodies and immunologic reactions 
depend? Specificity was at first believed to depend solely upon some 
3Jour. Infect. Dis., 1920, 27, 23. 
2 Jour. Exper. Med., 1916, 24, 419. 
3 Jour. Med. Research, 1918, 39, 93. 
4 Am. Rev. of Tuberculosis, 1918, 2, 587. 
5 Centralbl. f. Bakteriol., orig., 1906, 42, 633. 
6 Arch. f. Hyg., 1908, 65, 275. 
7 Arch. f. Hyg., 1908, lviii, 1. 
8 Arch. f. Hyg., 1907, 63, 332. 
9 Deutsch. Arch. f. klin. Med., 1909, xciv, 385. 
10 Deutsch. Arch. f. klin. Med., 1909, xcv, 424. 
11 Jour. Infect. Dis., 198, 23, 159. 158 
ANTIGENS AND ANTIBODIES 
peculiar biologic relationship of the antigens, for it was found comparatively 
easy to differentiate the serum of animals of dissimilar nature by means of 
the precipitin and other reactions, and, as serum proteins, which seemed to 
be quite similar chemically, but which were obtained from unrelated species, 
were sharply differentiated by the biologic reactions, it was considered that 
the specificity must be dependent upon some principle quite apart from the 
ordinary chemical substances. 
With the use of proteins other than serums, and especially when more 
or less purified proteins were employed, it has been quite firmly established 
that specificity depends upon chemical composition, and that differences in 
species, as exhibited by their biologic reactions, depend upon distinct differences 
in the chemistry of their proteins (Wells). 
Pick and his co-workers have shown that two kinds of specificity exist 
in each protein molecule: (1) One of these is easily changed by various 
Fig. 63.—General Scheme oe Antigens and Antibodies. 
Antitoxins and antiferments: R, Receptor of a molecule of a cell; T, a toxin molecule; t, toxo- 
phore group of the toxin molecule; h, haptophore group of the toxin molecule; A, cast-ofi receptor and 
constitutes antitoxin. 
Agglutinins and precipitins: A.R, Receptor of cell with antigen attached; B, a bacterial mole- 
cule (antigen) attached to a receptor; A or P, an agglutinin or precipitin; h, haptophore group of the 
antibody; a, agglutinophore group of an agglutinin. 
Hemolysins, etc.: A, Cast-off amboceptor (hemolysins, bacteriolysin, etc.); h, haptophore group 
of amboceptor; c, complementophil group; C, molecule of complement. 
physical agents, such as heat, cold, and partial coagulation. When an 
antigen is altered by heat it produces an antibody that reacts best with 
the heated antigen; heating does not, however, destroy the characteristics 
of the antigen of this species, as its antibody will not react with the heated 
antigen of another species. (2) The second alteration involves a profound 
chemical change of the antigen, whereby it is so altered that it loses the 
characteristics peculiar to the species, and produces an antibody that will 
react with the altered antigen, but not with the unaltered antigen, even 
from the same animal. For example, it is possible so to alter the serum 
protein of a rabbit by treatment with nitric acid that the nitroprotein 
injected back into the same rabbit will produce an antibody specific for the 
nitroprotein, but which does not react with the unchanged serum protein. 
These changes are apparently closely related to the aromatic radicals of the 
protein antigen, for they are effected by introducing into the protein mole- 
cules substances that are known to combine with the benzine ring, e. g., iodin, 
diazo- and nitro groups. Pick, appreciating the fact that the number of different aromatic radicals in the protein molecule are limited, interprets 
the significance of these radicals as depending upon their arrangement, 
rather than upon their number, in the protein molecule. Granting the 
number of possible variations in the arrangement of the amino-acids in a 
protein molecule which the great number of these radicals provides, there 
is no difficulty in understanding the possibility of an almost limitless num- 
ber of specific distinctions between proteins. 
It may be stated, however, in general, that immunologic reactions, 
such as that of anaphylaxis, are as delicate in distinguishing between proteins 
as are chemical analyses. Distinctions may be made by these reactions 
with quantities too small for making accurate chemical determinations. 
It may be useful here to draw up in tabular form a list of the various 
antigens and antibodies with which we are mainly interested in that por- 
tion of immunity involving infection with vegetable or animal parasites, 
and the products of their metabolism or degeneration (Fig. 63). 
ANTIBODIES 
159 
Antigens 
Toxins: 
1. Soluble bacterial toxins (diphtheria and 
tetanus toxins, etc.). 
2. Phyto- (vegetable) toxins (ricin, abrin, 
etc.). 
3. Simple zoo- (animal) toxins (snake, 
spider, toad venoms). 
4. Complex zootoxins, as snake venom, re- 
quiring complement for action. 
Enzymes or ferments (rennin, lipase, etc.). 
Precipitogenous substances (soluble animal 
and vegetable proteins). 
Agglutinogenous substances (bacteria, ery- 
throcytes, etc.). 
Opsonigenous substances (bacterial endo- 
toxins or aggressins?). 
Cytoligneous substances: 
1. Vegetable cells (bacteria). 
2. Animal cells (erythrocytes, spermato- 
zoa, kidney tissue, etc.). 
Antibodies 
Antitoxins: 
1. Antitoxins (diphtheria and tetanus anti- 
toxins, etc.). 
2. Anti- (phyto-) toxins (antiricin, anti- 
abrin, etc.). 
3. Anti- (zoo) toxins (antivenins). 
4. Antihemolysins, etc. 
Antienzymes (antirennin, antilipase, etc.). 
Precipitins. 
Agglutinins. 
Opsonins (acting singly or with complement). 
Cytolysins: 
1. Bacteriolysins. 
2. Hemolysins, spermatolysins, nephro- 
lysins, etc. 
Non-specific Immunity.—While diagnostic immunologic reactions with 
the body fluids are largely based upon the specificity of antibodies and the 
specific nature of the antigen-antibody reaction, resistance or immunity to 
disease may call into play certain non-specific factors which may possess 
considerable practical importance. Of these agencies, proteolytic ferments, 
leukocytosis, and the febrile reaction appear to be of most importance and 
especially in relation to the treatment of disease with bacterial and other 
proteins. Langer1 has recently stated that such a simple procedure as the 
daily removal of small amounts of blood from a rabbit results in the in- 
creased production of antibodies. Olsen,2 however, states that there is no 
actual increase to be ascribed to the effects of venesection alone, but simply 
the usual daily fluctuations in serum antibodies. This subject of non- 
specific immunity is discussed more completely in Chapter XXXIX. 
Heterophile Antigen and Antibody.—The presence of antibodies in the 
blood is commonly ascribed to the effects of immunization with specific 
antigens. This is especially true of antibodies developing during disease 
or as a result of artificial immunization. The presence of natural anti- 
bodies in the blood, however, is much more difficult of explanation; for 
1 Ztschr. f. Immunitatsf., 1921, 31, 290. 
2Ztschr. f. Immunitatsf., 1921, 31, 284. 160 
ANTIGENS AND ANTIBODIES 
example, the presence of hemolysin for sheep corpuscles in the sera of the 
majority of human beings, rabbits, and other of the lower animals. Like- 
wise the presence of small amounts of agglutinins for typhoid and other 
bacilli in the sera of some human beings who have never had typhoid fever 
or vaccine, etc. It has been commonly stated that occult immunization was 
probably responsible—that the antigens gained access to our tissues in hidden 
and unknown ways. 
The work of Forssman,1 however, indicates another possibility. He has 
shown that the injection of rabbits with emulsions of the liver and kidney 
of the guinea-pig results in the production of antisheep hemolysin. He has 
designated these non-specific antigens as heterologous and the hemolysin as 
heterologous antibody in contradistinction to homologous antibody en- 
gendered by the immunization of rabbits with sheep corpuscles. Friede- 
mann has suggested the terms heterophile antigen and heterophile antibody, 
and these terms have become more generally adopted because it is not that 
the antibody is generated by a different kind of antigen, but that it has an 
affinity for the receptors of a species other than those in response to which 
it was developed. 
Since the work of Forssman and his associates numerous other investi- 
gators, as Orudschiew,2 Rothacker,3 Doerr and Peck,4 Amako,5 Friedberger 
and Schiff,6 Sachs and Nathan,7 and others have discovered other hetero- 
phile antigens for antisheep hemolysin in rabbits, the subject being sum- 
marized as follows: 
(a) Organs (except blood-corpuscles) of guinea-pig, horse, dog, cat, mouse, 
fowl, tortoise, several kinds of fish, horse urine, some bacteria, etc., contain 
heterophile antigens for antisheep hemolysin in rabbits and are said to be 
of the “guinea-pig type.” 
(b) Organs of ox, rabbit, pig, man, rat, goose, pigeon, frog and eel lack 
this property and are called by Bail and Margulies8 animals of the “rabbit 
type.” Forssman originally believed that the heterophile antisheep he- 
molysin engendered by the injection of guinea-pig liver into rabbits was not 
fixed or absorbed by the emulsions of guinea-pig liver cells. Orudschiew, 
however, showed that absorption occurs, but that there is a relatively low 
affinity between the two. This view is now generally accepted and the 
production of heterophile hemolysin by the organs and substances mentioned 
above is ascribed to the wide distribution of the same or common antigenic 
substance for sheep hemolysin. Teniguchi9 and others have found that 
the substance in organs capable of fixing or absorbing heterophile hemolysin 
resides in the lipoids of these tissues, especially in those which are soluble 
in alcohol and ether and insoluble in acetone (the so-called lecithin fraction). 
The addition of cholesterol increases fixing capacity. The heterophile anti- 
body not only combines with these lipoids but also fixes complement in their 
presence and forms precipitates with them. These alcohol-soluble lipoids, 
however, are not capable of engendering the heterophile hemolysin; appar- 
ently whole extracts of the tissues are required. For this reason Taniguchi 
states that the antigenic activity of the organs of the “guinea-pig type” 
1 Biochem. Ztsch., 1911, 37, 78; 1912, 44, 336; 1914, 51, 6. 
2Ztschr. f. Imrrunitatsf., orig., 1913, 16, 268. 
3Ztschr. f. Immunitatsf., orig., 1913, 16, 491. 
4 Biochem. Ztsch. 1913, 40, 129; Ztsch. f. Immunitatsf., orig., 1913, 19, 251. 
6Ztschr. f. Immunitatsf., orig., 1914. 22, 641. 
6 Berl. klin. Wchn., 1913, 1557, 2328; Ztsch. f. Immunitatsf., 1919,28, 217, 237. 
7Ztschr. f. Immunitatsf., orig., 1913, 19, 235. 
8 Ztschr. f. Immunitatsf., orig., 1913, 19, 185. 
9 Jour. Pathology and Bacteriology, 1921, 24, 217, 241, 456. ANTIBODIES 
161 
resides in some lipoid-protein complex, whereas combining affinity resides in 
the lipoids. 
It is to be noted that all experiments have been conducted with rabbits 
and not all investigators have been sufficiently careful in controlling the 
factor of the natural antisheep hemolysin in rabbit-serum. Unless great 
care is exercised it is easy to fall into the error of regarding natural he- 
molysin the product of stimulation by heterophile antigens as emulsions of 
guinea-pig liver and kidney cells. I have found that immunization of 
rabbits with these substances sometimes results in the production of anti- 
sheep hemolysin, but only in rabbits known to contain natural antisheep 
hemolysin by preliminary titrations. With rabbits known to be free of 
natural hemolysin, the injection of guinea-pig liver and kidney have not 
resulted in the production of antisheep hemolysin. In my opinion these 
heterophile antigens are capable of stimulating an increased production of 
natural antisheep hemolysin by non-specific stimulation of the tissues con- 
cerned in the production of this substance. This possibility is strengthened 
by the observation that rabbits immunized with typhoid bacilli may pro- 
duce typhoid agglutinin after a period of rest by stimulation with other 
substances, as the injection of staphylococci and various non-specific pro- 
teins. In fact, the treatment of disease with various protein substances is 
based in part upon the observation that various substances may stimulate 
the production of antibodies against bacteria in a purely non-specific manner, 
providing the antibody-producing tissues are previously “sensitized” or 
tuned to the production of these antibodies. This subject is discussed in 
more detail in the chapters dealing wdth the mechanism of non-specific pro- 
tein therapy and the non-specific activities of vaccines. 
Normal and Immune Antibodies.—Antibodies, as antitoxins, opsonins, 
agglutinins, bacteriolysins, hemolysins, and other cytolysins, are frequently 
found in the sera of persons and in the lower animals. These are designated 
as normal or natural antibodies in contradistinction to those produced dur- 
ing disease or by artificial immunization called immune antibodies. The 
presence of these natural antibodies has never been adequately explained, 
although since some are absent at birth, to be acquired later in life, infec- 
tions may be responsible, and that what are regarded as natural antibodies 
are rather sometimes antibodies of the immune variety. 
Antiantibodies.—As shown by Ehrlich and Morgenroth1 it is possible 
by injecting hemolysins to produce antihemolysins which are capable of 
counteracting the effect of hemolysins or of neutralizing them. Bordet2 
found that these antihemolysins may be produced not only by immuniza- 
tion with hemolytic immune serum but also with normal serum of the 
same species, even though this normal serum contains no corresponding 
amboceptors. Ehrlich has explained this phenomenon by showing that 
antiamboceptors act against the complementophil groups of all amboceptors 
and that a normal serum used for immunization may contain these groups. 
Similar data bearing upon the non-specificity of the antiamboceptors was 
obtained by Pfeiffer and Friedberger,3 who found that antiamboceptors 
obtained by immunizing with cholera serum acted also against typhoid 
serum. Walker4 has found that animals may be immunized against an im- 
mune serum, finding that they are then less capable of being protected by 
that serum, but with no increased susceptibility to infection. 
1 Berl. klin. Wchn., 1899, No. 22, 481; ibid., 1900, No. 21, 453. 
2 Ann. d. l’Inst. Pasteur, 1904, No. 10. 
3 Centralbl. f. Bakteriol., 1903, 34, 70; ibid., 1904, 37, 138. 
4 Jour. Path, and Bacteriol., 1903, 8, 34. CHAPTER IX 
THE VARIOUS TYPES OF IMMUNITY 
Kinds of Immunity.—As has been stated in the preceding chapters, it 
is generally agreed that various antibodies and other protective agencies 
exist, although opinions differ as to the source and relative importance of 
these to resistance to and recovery from various infections. Whether or 
not a particular antibody is derived from a certain group of cells is largely 
a matter of individual opinion because of the difficulty of deciding the 
point by actual experimental evidence. Of far more importance is a knowl- 
edge of the properties of antibodies and of the role they may play in the 
processes of immunity. It is seldom that resistance to, or recovery from, 
an infection is dependent upon one defensive factor: usually several agencies 
are operative, although one factor may predominate. For example, anti- 
toxins are known to neutralize their respective toxins, and are of most 
value in combating the toxemias, such as diphtheria and tetanus; bacterio- 
lysins cause the death of and may totally destroy their antigens, and play 
an important part in the recovery from infections with bacilli of the typhoid- 
colon and cholera groups; phagocytosis in itself is of importance in staphylo- 
coccus infections, and is of primary importance, in conjunction with the op- 
sonins, in recovery from pyogenic infections in general; agglutinins and 
precipitins do not appear to have a direct inimical influence on their an- 
tigens, but are probably secondary factors, and contribute in some manner 
toward their destruction. Along with important non-specific factors these 
various antibodies are responsible for the different forms of immunity, 
which may now be considered in their more general aspects. 
There are two kinds of immunity—natural and acquired. Acquired 
immunity is divided into two subvarieties: active and passive. 
Antiblastic Immunity.—While immunity to disease is believed to be 
dependent in part upon phagocytic activity and antibodies as just described, 
Ascoli1 believes that immunity may sometimes be due to forces antagonistic 
to the growth of the micro-organism in the body. From his studies in 
anthrax immunity he ascribed to antianthrax serum as antiblastic action 
directed against the metabolic activities of this organism. Dochez and 
Avery2 have adopted this term “antiblastic immunity” to describe the 
retardation of growth of pneumococcus by antipneumococcus serum, which 
they believe is, in part at least, dependent upon inhibition of metabolic 
function, particularly the proteolytic and glycolytic functions, of pneu- 
mococci resulting in a retardation of nutritional processes and consequent 
inhibition of growth. 
Depression or Infection Immunity.—It has been for years a matter of 
clinical experience that during one disease a second infection may show 
retrogressive changes. For example, Zupnik, Muller, and Leiner3 have 
shown that the malarial paroxysm in the typhoid fever patient frequently 
brings about either a temporary or permanent detoxication and improve- 
ment; the effect of erysipelas on tumors and of pregnancy on tumors are 
related phenomena. 
Morgenroth, Biberstein, and Schnitzer4 have recently shown that mice 
infected with streptococci develop in four to twenty-four hours a temporary 
1 Centralbl. f. Bakertiol., orig., 1908, xlvi, 178. 
2 Jour. Exper. Med., 1916, 23, 61. 
3 Wien. klin. Wchn., 1916, 29, 64. 
4 Deut. med. Wchn., 1920, xlvi, 337. 
162 NATURAL IMMUNITY 
163 
and relative resistance to a second injection of the same or other strepto- 
cocci in doses that prove fatal for control animals; in other words, that dur- 
ing the existence of general streptococcus infection the animals are tem- 
porarily refractory or resistant to added injections of streptococci fatal for 
normal animals. Wiegand1 and Berliner and Citron2 have confirmed these 
results, working with chicken cholera infections of guinea-pigs. 
The same phenomenon is seen in syphilis. It is generally believed that 
the uncured syphilitic is immune to reinfection with T. pallidum, although 
the patient is susceptible to exacerbations of his own infection. In other 
words, it would appear that the syphilitic develops an immunity to all other 
strains of T. pallidum but his own. This subject will be discussed in more 
detail in Chapter XXXIV, but is mentioned here as an example of this so- 
called depression immunity. 
The mechanism of this resistance is not known. In the experimental 
work mentioned above it has been shown that the streptococci or chicken 
cholera organisms used in the superimposed infection are not killed, but 
may be found in the animals. The animal is simply refractory or immune 
for a time to its effects in a manner analogous to the production of the 
state of antianaphylaxis or desensitization to non-infectious protein agents. 
The immunity is not absolute, but only relative, inasmuch as superinfection 
with organisms of specially high virulence or in extra large doses may over- 
come the condition. 
This immunity has been designated as “depression immunity” because 
the reaction of the body to superinfection is depressed or the activity of 
one disease depressed by the development of a second infection. For this 
condition of resistance or immunity to superinfection I believe the term 
“infection immunity” more appropriate. The benefit to be noted upon one 
disease by the development of a second, as the influence of malaria upon 
typhoid fever, erysipelas, and pregnancy upon tumors, etc., probably in- 
volves a different mechanism introducing the curative activity of non- 
specific agents, as fever, leukocytosis, and the stimulation of ferment and 
antibody production, discussed in more detail in the chapter on Non-specific 
Protein Therapy. 
NATURAL IMMUNITY 
Natural immunity is the resistance to infection normally possessed, usually 
as the result of inheritance, by certain individuals or species under natural 
conditions. 
The mechanism of this type of immunity is very complex, and bears 
an intimate relation to the subject of infection, both local and general, 
the nature of the infecting parasite, and the presence or absence of specific 
antibodies in the body fluids. In many instances this type of immunity 
is dependent upon non-specific causes—is frequently relative and seldom 
absolute. For example, fowls are immune to what may be called an ordi- 
nary dose of tetanus toxin, but succumb readily to larger doses; rats are 
highly immune to diphtheria toxin, and readily withstand the effects of an 
amount equaling 1000 lethal doses for a guinea-pig, but still larger doses may 
prove fatal; hedgehogs possess complete or almost complete immunity for 
the amount of snake venom deposited in an ordinary strike, but if the 
venoms of several snakes are collected and injected at one time the result 
is fatal. 
Species immunity is a type of natural immunity, best illustrated by 
1 Inaug. Dissertation, Marburg, 1920. 
2 Deut. Med. Wchn., 1920, xlvi, 997. 164 
THE VARIOUS TYPES OF IMMUNITY 
the immunity of man to certain diseases of the lower animals, such as fowl 
cholera, swine-plague, distemper, Texas cattle fever, mouse septicemia, etc.; 
and, conversely, by the immunity of animals to diseases common to man, 
such as measles, cholera, typhoid fever, scarlet fever, chickenpox, etc. 
Although the close relation of man to the domestic animals furnishes ample 
opportunity for infection, yet a complete immunity is frequently observed. 
Racial immunity is that type of natural immunity existing among mem- 
bers of the same species. For example, negroes are believed to enjoy im- 
munity to yellow fever and Mongolians to scarlet fever. As a matter of 
fact, well-marked examples of racial immunity are extremely rare, as not 
infrequently the disease in question may have been acquired in early infancy 
in a clinically unrecognized form. 
Similarly, close biologic relationship is no guarantee that animals will 
behave alike toward infection. For example, the white mouse is immune 
to glanders, the house mouse is somewhat susceptible, and the field mouse 
is highly susceptible. The rabbit, guinea-pig, and rat are rodents, but 
though the rabbit and the guinea-pig are susceptible to anthrax, the rat is 
largely immune. Mosquitoes, though closely related, behave differently 
toward the malarial parasite. The Culex does not carry the parasite at all, 
and of the Anopheles, one species, Anopheles maculipennis, is quite suscep- 
tible and well recognized as a carrier of the parasite, whereas Anopheles 
punctipennis, though closely related, is not susceptible to it. 
Examples of individual immunity, while not infrequent, are not con- 
stant and seldom absolute. Certain persons appear to possess a defi- 
nite immunity to scarlet fever and diphtheria, although they may be freely 
exposed; others may pass through various epidemics of other infectious dis- 
eases, such as measles, pertussis, etc., without becoming infected. I have 
noticed, on several occasions, that resident physicians, on service in scarlet 
fever wards for many months or years, having escaped infection though 
brought in intimate contact with severe forms of the disease, finally con- 
tracted the disease upon returning from a short vacation. 
Mechanism of Natural Immunity.—Natural immunity may be due to 
the following causes: 
1. Various non-specific factors may prevent infection; among these may 
be mentioned: (a) The integrity of the epithelium of the skin and mucous 
membranes; (b) activity of enzymes in the skin (see page 167), and (c) the 
chemical and physical action of various secretions, such as the gastric fluid, 
the intestinal juices, and the saliva. 
2. A particular route for the introduction of infecting micro parasites may 
he necessary. For example, intestinal diseases, such as typhoid fever and 
cholera, are usually due directly to swallowing of the infecting micro-organ- 
isms, infection in this type of disease seldom, if ever, occurring through the 
skin. This is probably due in part to the lowered vitality of the intestinal 
mucosa, together with a peculiar selective affinity of the bacteria for the 
cells of these tissues, aided by the biologic nature of the invading bac- 
terium, which grows best under the more favorable cultural conditions of 
the intestinal canal. This selective action is further illustrated by the ten- 
dency of dysentery toxin to attack the intestinal mucosa when the bacilli 
or toxin is administered intravenously. 
3. Certain tissues appear to possess a marked local immunity to certain 
bacteria. In considering examples of local immunity, various factors, such 
as the question of exposure, the thickness of the epidermis, and the kind 
and quantity of the local secretions, must be borne in mind. For example, 
Trichina spiralis affects the muscles, never the bones, and but rarely any NATURAL IMMUNITY 
165 
other tissue. Likewise, although diphtheria in the throat may spread in 
many directions, it seldom passes down the esophagus. 
Some differences are known to exist in regard to local immunity as 
observed in the child and in the adult. For example, ringworm of the 
scalp is practically unknown among adults, whereas children under seven 
years of age are quite susceptible to the disease. These differences may be 
due to the greater susceptibility in general of young tissues to infection, and 
the local immunity constitutes but an index to the general rise in resisting 
power accompanying improvement in strength and vitality. In some cases 
this may be due perhaps to an actual strengthening of local tissues, as in the 
case of the adult vaginal mucosa, which is immune to the gonococcus, whereas 
the thin and immature infantile membrane is peculiarly susceptible. 
In general, our knowledge of local immunity is quite incomplete. The 
subject is a difficult one, hence most attention has been given to the study 
of general immunity. A striking example of acquired local immunity may 
be seen in a patch of psoriasis, where the center is observed to be largely 
free from scales, whereas the margins are quite active. 
The question of local immunity may be largely determined by various 
local non-specific factors, such as loss of blood-supply due to traumatism, 
thrombosis, tight bandaging, etc., and the action of severe irritants, tending 
to produce necrosis of the tissues. 
4. The importance of phagocytosis in natural immunity must be empha- 
sized. Micro-organisms are constantly gaining entrance to the tissues 
through numerous small abrasions of the skin and along the intestinal and 
respiratory tracts, and investigations have shown how important the 
wandering cells are in preventing infection, being ever on guard and ready 
to pick up and dispose of any injurious material. Even after mild infection 
has occurred, the local inflammatory reaction in which the phagocyte is a 
prominent factor may be so prompt in overcoming the invaders that the 
host will escape serious infection. 
The natural immunity of the frog to anthrax has been shown to be 
partly dependent upon the activity of the leukocytes in engulfing and dis- 
posing the bacilli. 
Similarly, a mild irritant may produce hyperemia and exudation or 
local accumulation of leukocytes, which aid in establishing a local im- 
munity largely dependent upon phagocytosis. In this manner the intra- 
peritoneal injection of sterile bouillon or even of salt solution may produce 
exudation and increase the immunity to infection. 
5. It may be that even after the introduction of a micro-organism or its 
toxin no harm results because of a lack of suitable receptors on the part of the 
body cells of the host for union with the pathogenic agent. For example, tetanus 
toxin, being unbound by the cells, produces no effect on the turtle, and anti- 
toxin is not produced. On the other hand, suitable receptors may be 
present that will bind the toxin, but produce no harmful effects because the 
body cells are not susceptible to the action of the microparasite or its prod- 
ucts. Thus it is asserted that tetanus toxin has no effect upon the alligator, 
although the toxin is bound and antitoxin is produced by its body cells. 
In other instances, a host may escape infection owing to the fact that 
there is a lowered affinity between a pathogenic agent and the body cells, 
so that but a small amount of harmful substances are bound to the body 
cells, and no particular harm results, whereas a larger dose, uniting with a 
greater number of cells, is capable of producing some disturbance. 
6. A natural antitoxin immunity may exist, as the immunity of the 
alligator to tetanus toxin, just mentioned. Similarly, natural diphtheria 166 
THE VARIOUS TYPES OF IMMUNITY 
antitoxin may prevent infection, especially in those persons known to har- 
bor virulent bacilli in the fauces. In such instances, however, it is difficult 
to exclude entirely the possibility that a previous minor infection has oc- 
curred, as natural antitoxin immunity persists much longer than the passive 
immunity resulting from the administration of an antitoxin serum. 
Otto, who has recently investigated the content of diphtheria antitoxin 
in the blood of normal persons, found more than unit of antitoxin 
in each cubic centimeter of the blood of those who had been in close con- 
tact with cases of diphtheria without having been sick themselves; others 
usually had much less. Observations would tend to show that this quan- 
tity of antitoxin is generally sufficient to confer immunity to diphtheria, and 
the object of von Behring’s method of active immunization is to induce 
the production of at least that much antitoxin by the body itself. Otto1 
found that diphtheria carriers, both those who had had the disease and 
those who had not, contained more antitoxin in their blood than did pa- 
tients who had just recovered from an attack. This shows that the mere 
presence of bacilli in the throat is sufficient to stimulate the production of 
antitoxin, on which the immunity of the carrier himself would seem to 
depend. Undoubtedly physicians and nurses who are freely exposed to 
diphtheria and yet escape infection owe their safety rather to an acquired 
immunity the result of repeated contact with the bacilli than to a natural 
antitoxin immunity. 
More recently Burrows and Suzuki2 have studied the natural immunity 
of the rat to diphtheria toxin and the chicken to tetanus toxin by means of 
cultures of tissues; these investigators found that certain cells possessed a 
peculiar resistance and that antitoxins may be present in the plasma. 
Coca, Russell, and Baughman3 found that the white rat can survive the 
injection of 1000 times the minimal lethal dose of toxin for the guinea-pig; 
this natural immunity was found not to be due to the presence of normal 
or natural antitoxin in the blood, but to the property of the cells of the rat 
of preventing the toxin from entering them or attaching itself to them. 
7. In some instances a natural immunity may be dependent, at least in 
part, upon antibacterial activity, due to the presence of bacteriolysins and 
bacteriotropins in the body fluids, as, for example, that shown by the dog 
and the rat to anthrax. In other instances, however, a similar immunity 
may be observed that cannot be ascribed to the presence of antitoxins or 
bacteriolysins. In this type of immunity microparasites are apparently 
unable to sustain themselves, and proliferate in one animal, although 
aggressive enough in another of the same species. 
8. An immunity to infection, especially with such micro-organisms as 
the anthrax bacillus, which is markedly aggressive and but slightly toxic, may 
be due to the presence or production of antiaggressins. This immunity would 
seem to depend not upon the bactericidal properties of the serum or leuko- 
cytes, nor upon the antitoxins, but on the presence of substances that pre- 
vent the micro-organisms from exercising their special aggressive forces. 
9. Finally, an immunity may exist because the parasite or other foreign 
cell does not obtain suitable nutrition in a host and thus fails to grow. This 
condition of athrepsia is responsible for what has been called athreptic im- 
munity. It has been more recently studied by Ehrlich, who found that 
upon transferring mouse cancer to the rat, the tumor grewT for a short time 
only, or presumably until the special nutriment carried over with the tumor 
1 Deutsch. med. Wchn., March 12, 1914, 542. 
2 Jour. Immunology, 1917, 3, 219, 233. 
3 Jour. Immunology, 1921, 6, 387. ACQUIRED IMMUNITY 
167 
was consumed. While there is no experimental basis for assuming that a 
similar condition may be present in bacterial life, yet such a cause may be 
operative and should be kept in mind. 
The Skin in Relation to Natural Immunity; “Exophylaxis.”—Mention 
has been made that natural immunity to some diseases may be due to the 
resistance offered to microbic invasion by the intact epithelial cells of the 
skin and mucous membranes; also to the bactericidal action of the secretions. 
In addition to these protective agencies the skin is known to contain 
different proteolytic enzymes and especially the skin of adult human beings; 
these enzymes are probably protective by reason of their destructive effects 
upon microbes. 
Hoffmann1 has called particular attention to the part played by the 
skin in natural immunity, its activity in protecting against the entrance 
of pathogenic microbes—an exophylaxis—as well as the possibility of it 
being a source of some internal secretion or the seat of production of en- 
zymes and antibodies playing an important role in recovery from disease 
of the internal organs. Hoffmann has made the epigrammatic statement 
that “the skin is the grave of the parasites” and leans to the view that in the 
acute exanthematous diseases the body endeavors to rid itself of the toxic 
substances through the skin by a process of digestion in which leukocytes 
and enzymes are concerned. 
Bloch2 has also expressed the conviction that the skin possesses an 
important biologic function by means of which the internal organs are 
protected against microbic infection. Both he and Hoffmann have drawn 
attention to the fact that an intoxication ensues when the function of the 
skin is destroyed, as by varnishing or burning; that the histologic structure 
greatly favors the absorption of secretions, and that in the exanthematous 
diseases the internal organs are spared to the degree that the eruption is 
manifest, with the therapeutic experience that anything increasing the 
eruption influences the patient in a favorable manner. The clinical fact 
that measles is least dangerous when the eruption is prompt and profuse and 
that in syphilis, involvement of the central nervous system is less likely 
when the cutaneous and mucous membrane lesions are well marked, are also 
to be mentioned in this connection. In syphilis, however, it may be that 
there are different strains of pallida, one, the dermotropic strain, producing 
marked cutaneous lesions, and a second or neutropic strain, producing early 
involvement of the tissues of the central nervous system. 
It is apparent, however, that the skin plays an important role in natural 
immunity to infection, and it may well be that it is an organ playing an 
important part in the mechanism of recovery of disease of the other organs 
as well. 
Acquired immunity occurs in two distinct forms: (1) Active and (2) 
passive. A mixed form may exist, brought about by a combination of factors 
necessary for the development of the other two. 
Active Acquired Immunity.—Active acquired immunity is that form of re- 
sistance to infection brought about by the activity of the cells of a person or animal 
as a result of having had the actual disease in question, or as a residt of artificial 
inoculation with a modified or attenuated form of the caustive micro parasite. 
The essential feature of this immunity is that the cells and tissues of 
persons or animals should, by their own efforts, and as a result of their 
ACQUIRED IMMUNITY 
1 Deut. Med. Schn., 1919, xlv, 1233. 
2 Cor.-Bl. f. schweiz. Aertze, 1914, xliv, 1377; ibid., 1917, xlvii, 1259. 168 
THE VARIOUS TYPES OF IMMUNITY 
own active struggle against the action of a microparasite or its products, 
overcome these and become less susceptible to them than they were before. 
This form of immunity is gained, therefore, only as the result of an 
active struggle between body cells and infecting agent, and this battle may 
be of any degree of severity, ranging from an attack of the disease itself 
that may threaten life, down to the most transitory and trivial reaction 
due to the injection of a minute dose of a mild vaccine. 
Active acquired immunity may be gained: (1) By accidental injection, 
which is the most familiar form of acquired immunity, and follows an 
attack of an infectious disease, such as scarlet fever, measles, varicella, 
variola, or typhus fever; (2) by inducing an attack of the disease by artificial 
inoculation. This latter method of producing an active acquired immunity 
was illustrated by the ancient, obsolete, and discarded practice of smallpox 
inoculation, by which healthy persons were inoculated with the virus of a 
mild case of smallpox at a time when no epidemics existed and the person 
was in good general health and able to secure proper attention from the 
outset. 
This process of immunization is used much more extensively in vet- 
erinary practice, where an occasional untoward or fatal result is of com- 
paratively little importance if by its means an outbreak can be controlled 
or the great majority of the animals saved. As a rule, an attempt is made 
to render the induced disease as mild as possible by (a) using a small 
amount of infective material; (b) by inoculating it through an unusual ave- 
nue; (c) by inoculating it at a time when the animals are naturally less 
susceptible, or (d) by a combination of these methods. For example, Texas 
cattle fever, which is due to a protozoan (Piroplasma bigeminum) conveyed 
by the bites of infected ticks, may be combated by exposing calves while 
still milk fed to the bites of a few infected ticks. Another method consists 
in injecting a small amount of blood from an infected animal directly into 
the jugular vein. The object is to induce a mild attack of the disease. 
Occasionally a severe or fatal reaction occurs, but the number of these un- 
toward results is much lower than the mortality among untreated animals. 
(3) Active immunity may also be gained by vaccination, i. e., by inocula- 
tion with a virus or microparasite or its products, modified and attenuated 
by passage through a lower animal (Jennerian vaccination) or by various 
other means, as age, unfavorable cultural conditions, heat, germicides, etc. 
(Pasteurian vaccination or bacterination). These subjects are considered 
more fully in the chapter on Active Immunization. 
Active immunity, whether induced accidentally or artificially, may be 
antitoxic, as after recovery from diphtheria or as the result of active im- 
munization with diphtheria toxin, as by von Behring’s method; or anti- 
bacterial, as the immunity following typhoid fever or induced by typhoid 
vaccination, and largely dependent upon the presence of bacteriolvsins in 
the circulating fluids. 
During the process of active immunization an animal not infrequently 
fails to react to relatively large doses of toxin, and at the same time the 
quantity of antibody in the body fluid may decrease. This phenomenon 
has been explained as being due to atrophy of the receptors of the body 
cells (receptoric atrophy), whereby the toxin fails to exert its deleterious 
influence because it fails to unite with the body cells. It is curious, however, 
that the toxin is innocuous when present in a free state within the body 
fluids, even though unbound to the body cells; this condition is not well 
understood, and may be dependent upon other factors. A rest may restore 
the activity of the receptors and cells, a fact that is well recognized in the ACQUIRED IMMUNITY 
169 
immunization of horses for the preparation of antitoxin. Not infrequently 
a rabbit fails to produce hemolytic amboceptor if the injections of erythro- 
cytes are too frequent. After a rest, however, the animal may react promptly 
with much smaller doses. 
Passive Acquired Immunity.—As the name indicates, this is a form of 
immunity that depends upon defensive factors not originating in the person 
or animal protected, hut is passively acquired by the injection of serum from 
one that has acquired an active immunity to the disease in question. 
This is a sort of secondary immunity, acquired by virtue of having 
received antibodies actively formed by another animal that has had to 
resist the infecting agent in order to produce them. Two well-known 
examples of this type of serums are the diphtheria and tetanus antitoxins. 
These are produced by injecting horses with successive doses of the re- 
spective toxins. The horses are required to combat the effects of the toxins, 
and acquire an active immunity of increasing grade due to the production 
of antitoxins. When the animals are bled the antitoxin-laden serum, 
separated from the corpuscular elements, may be used for conferring an 
immunity in a person or another animal simply by injecting the serum, the 
latter receiving and enjoying an immunity in a passive manner. 
Passive immunity is specific, that is, the serum of an animal immunized 
against one micro-organism will protect a second animal against that and 
against no other. This type of immunity is acquired just as soon as the 
immune serum has become mixed with the blood of the person or animal 
injected, and there is no negative phase. Hence in severe infections our 
hopes of specific therapy rest on the production of passive immunity. No 
matter how sick the recipient may be, under ordinary circumstances the 
immune serum produces no further disturbance than would be expected 
from the injection of a normal serum. The recipient’s body cells have no 
additional burdens, or very slight ones only, to bear, and these are more than 
counterbalanced by the release from combat with toxic substances neutralized 
by the antibodies in the immune serum. Unfortunately, this field of therapy 
is limited, although recent discoveries are indicating the reasons for failure, 
and when these are eliminated, the field of usefulness will be much extended. 
Passive immunity is of shorter duration than active immunity, and the 
former is especially indicated in prophylaxis for warding off an acute infec- 
tion that has a relatively short incubation period. The degree of passive 
immunity is also seldom equal to that of an active immunity. The anti- 
bodies produced by our own cells are more lasting and possess higher pro- 
tective value. This is an important factor in von Behring’s method of 
immunization in diphtheria, when a small amount of toxin loosely bound 
to antitoxin is injected in the belief that the toxin becomes dissociated and 
serves to stimulate our body cells into producing our own antitoxin. 
Passive acquired immunity is usually antitoxic, as, for example, that in- 
duced by the administration to man of diphtheria antitoxin prepared by the 
body cells of the horse. Antibacterial serums may likewise induce a passive 
immunity, as, for instance, that used in immunization against plague. 
It is evident, therefore, that the processes whereby infections are over- 
come and immunity is conferred, and the general reactions that follow 
the introduction into the body of modified antigens in the practice of im- 
munization, are complex processes, and in none is one antibody produced 
or solely responsible for the resulting immunity. The properties and action 
of the known antibodies are considered in subsequent chapters, particular 
attention being given to methods for determining their presence in the 
body fluids, which serve as an aid to the diagnosis of infection as based 170 
THE VARIOUS TYPES OF IMMUNITY 
upon the general law that the antibody is specific for its antigen, and so, 
when the presence of an antibody is demonstrated, it may be assumed that 
the antigen is or has been present. 
Nothing is known concerning the nature of the immunity that is ac- 
quired against several infections, such as scarlet fever, measles, smallpox, 
etc., nor will much be known until the causes that give rise to these con- 
ditions have been discovered. 
Theory of Vaughan.—According to Vaughan, the inability of a bacterial 
cell to grow in the animal body either because it cannot feed upon the 
protein of the body or because it is itself destroyed by the ferments elabo- 
rated by the body cells explains all forms of bacterial immunity, either natural 
or acquired. Thus in antitoxin immunity the toxin is regarded as a fer- 
ment that splits up the proteins of the body cells, setting the protein poison 
free. The body cells react with the formation of an antiferment or anti- 
toxin, which neutralizes the toxin and prevents cleavage. The toxin itself 
is regarded as harmful only in so far as it is able to set free the protein poison 
responsible for the symptoms of the infection. 
Natural immunity to any infection, according to Vaughan’s theory, is 
explained as being due to an inability of the infecting agent to grow in 
the animal body. 
Acquired immunity, due to recovery from an infection or occurring as 
a result of vaccination, is regarded as the outcome of the development in 
the body, during the course of the infective process, of a specific ferment 
that, on renewed exposure, immediately destroys the infection. The 
vaccine is the same protein that causes the disease, so modified that it will 
not produce the disease, but yet so little altered that it will stimulate the 
body cells to form a specific ferment that will promptly and quickly destroy 
the infecting agent on exposure. 
Summary of Kinds of Immunity.—The various kinds of immunity and 
the factors probably concerned in their production, may be summarized as 
follows: 
1. Due to non-specific factors: 
1. Barrier of epithelium. 
2. Various secretions. 
3. A particular route of infection may be neces- 
sary, aided by the biologic nature of the in- 
vading bacterium. 
2. Due to local tissue immunity and selective action 
of micro-parasites for certain tissues. 
3. Due to phagocytosis. 
4. Due to lack of suitable receptors of body cells for a 
particular bacterium. 
5. Due to natural antitoxins. 
6. Due to natural bacteriolvsins. 
7. Due to anti aggressins. 
8. Due to lack of suitable food material—athrepsia. 
Natural Immunity 
Active (accidentally or 
artificially acquired) 
1. Antitoxic. 
2. Antibacterial. 
Acquired Immunity 
1. Anti oxic. 
2. Antibacterial. 
Passive CHAPTER X 
OPSONINS 
Historic.—Although there can be no doubt as to the importance of 
phagocytosis in the mechanism of recovery from infection, yet it was shown 
by Metchnikoff, as early as 1893, that the body fluids contained substances 
that greatly facilitated the phagocytic process, and that leukocytes removed 
from this influence were practically powerless to engulf and destroy the 
invading bacterium. In other words, if leukocytes and bacteria are washed 
free from all traces of serum and then mixed, very few of the leukocytes will 
be found capable of phagocytizing the bacteria, which means that spon- 
taneous phagocytosis is feeble and hence of slight importance. When, how- 
ever, fresh serum is added, especially the serum of an animal immunized 
against the micro-organism used in the experiment, phagocytosis is marked, 
and, indeed, most impressive. Metchnikoff attributed this difference to the 
influence of a substance in the serum that stimulated (stimulins) the leuko- 
cytes to become phagocytes, but later researches have shown that this is prob- 
ably erroneous, and that the serum facilitates phagocytosis not by exerting a 
stimulating influence upon the leukocytes, but by preparing the bacteria for 
the process by making them, as it were, more attractive to the leukocytes. 
Denys and Leclef,1 in 1895, were among the first to demonstrate the 
effect of serum on bacteria in the process of phagocytosis, and the fact 
that the active substance was not bactericidal in action, but in the nature 
of a new antibody. Since Metchnikoff had shown that freshly isolated or 
virulent strains of bacteria were not readily phagocytized, but seemed to 
resist or repel the leukocytes, it was natural for these observers to suggest 
that the action of this substance in serum was to neutralize the exotoxins 
and endotoxins of micro-organisms that were regarded as responsible for 
negative chemotactic influences, and thus, by robbing them of at least two 
defensive weapons, prepare them for phagocytosis. 
The subject remained in an uncertain state until 1903, when Wright,2 
and later Wright and Douglas, demonstrated more clearly this action of 
serum upon bacteria in aiding phagocytosis. Using their own modifica- 
tion of the technic devised by Leishman for measuring the phagocytic 
power of the blood, these observers first determined the direct depend- 
ence of phagocytosis upon some ingredient of the blood-serum, and fur- 
ther proved that this substance acts directly upon bacteria, is bound by 
the bacteria, and renders them more easily ingested by the leukocytes, 
i. e., more readily phagocytable. To this substance they gave the name 
opsonin (from opsono, I prepare food for). At the same time, and in- 
dependently of Wright, Neufeld and Rimpau conducted similar inves- 
tigations with immune serum and reached similar conclusions, but called 
the substance bacteriotropin. Since then both terms have been used— 
the former more frequently in English literature—and this is permissible, 
providing that it is understood that both are practically the same antibody, 
and not distinct and separate from each other. 
> As will readily be understood, the bacterial opsonins have been studied 
most extensively, but opsinins may be present in normal and immune serums 
1 La Cellule, 1895, xi, 175, Centralbl. f. Bakteriol., Abt., 1898, 24, 685. 
2 Proc. Roy. Soc., 1904, lxiii, 128. 
171 172 
OPSONINS 
for other cells, such*as erythrocytes, and these hemopsonins are regarded as 
separate antibodies, distinct from hemagglutinins and hemolysins. 
Definition.—Opsonins are substances in normal and immune serums 
which act upon bacteria and other cells in such a manner as to prepare them 
for more ready ingestion by the phagocytes. 
Properties and Nature of Opsonins.—There is considerable difference 
of opinion regarding the identity and probable structure of opsonins in 
normal and immune serums. Just as agglutinin for a bacterium, such as 
Bacillus typhosus, may be found in varying amounts in normal serum, so 
various opsonins for different bacteria may be found in normal serums. 
These normal opsonins appear more or less specific for a given bacterium, 
and in immune serum the specific opsonic substance for the particular 
bacterium or cell with which immunization has been produced is developed 
to a high degree. Both owe their full effect to the interaction of two sub- 
stances. One of these, the common substance, is thermolabile, and de- 
stroyed by heating the serum to from 56° to 58° C. for half an hour, whereas 
the other more specific substance remains unaffected. The latter, in both 
normal and immune serums, is opsonic by itself, although in the absence of 
the common thermolabile substance to a less degree, and is produced anew 
and specifically by artificial immunization or as the outcome of spontaneous 
infections. 
Before the exact interaction of serum and cells in phagocytosis had been 
made clear Metchnikoff and his students attributed phagocytosis by im- 
mune serum to the so-called “fixateurs” or “substance sensibilatrice,” which 
in general are regarded as identical with Ehrlich’s amboceptors. Dean1 
expressed the view that amboceptors may exercise the functions of op- 
sonins, which consequently cannot be regarded as independent substances. 
Neufeld and Topfer2 and Barratt,3 however, early showed that a serum may 
contain opsonin for erythrocytes without being hemolytic. 
In a series of investigations Hektoen4,5 has clearly shown that bacterio- 
opsonins and hemopsonins are distinct antibodies and may be differentiated 
from amboceptors by variation in resistance to heat and by absorption. 
This view is now generally maintained. 
The true nature of opsonins is difficult to understand. They have 
been compared by Hektoen and Rudiger6 to receptors of the second order, 
with a haptophore and a toxophore or opsoniferous group. Receptors of 
this order, however, are active and independent of the presence or absence 
of complement, whereas the opsonins, although active to some extent in 
the absence of complement, are far more so if a complement is present as 
showTn by Dean,7 Cowie and Chapin,8 Eggers,9 Browning,10 and others. 
Kolmer, Toyama, and Matsunami11 have shown that the addition of 
guinea-pig serum (complement) to commercial antimeningococcus serum in 
quantities that by themselves have little opsonic effect decidedly increases 
opsonic activity; these results have been confirmed by Hektoen and Tunni- 
cliffe,12 and similar findings have been reported by Meyer,13 working with 
antipneumococcus serum. These and earlier observations indicate that 
opsonic sera, normal as well as immune, owe their full activity to a ther- 
1 Proc. Roy. Soo., B, 1905, 76, 506. 4 Jour. Infect. Dis., 1906, 3, 434. 
2 Centralbl. f. Bakteriol., orig., 1905, 38, 456. 6 Jour. Infect. Dis., 1909, 6, 78. 
3 Proc. Roy. Soc., B, 1905, 7$, 524. 6 Jour. Infect. Dis., 1905, 2, 128. 
7 Proc. Roy. Soc., 1907, 79, 399; ibid., 1905, 76, 506. 
3 Jour. Med. Res., 1907, 17, 95, 213. 11 Jour. Immunology, 1918, 3, 156. 
9 Jour. Infect. Dis., 1908, 5, 263. 12 Jour. Infect. Dis., 1921, 29, 553. 
10 Jour. Med. Res., 1908, 19, 201. 13 jour. Infect. Dis., 1920, 27, 82. SPECIFICITY OF NATURAL AND IMMUNE OP SON INS 
173 
mostable opsonin and a thermolable complement-like substance which 
greatly promotes the action of the first substance. 
In this respect they resemble amboceptors, or receptors of the third 
order, opsonins in normal and immune serums representing respectively 
normal and immune bacterial amboceptors. One objection to this view of 
their structure is their activity, however slight, when the thermolabile 
substance is removed by heating, unless the amboceptors are complemented 
by an endocomplement, as from the bacteria themselves. 
At the present time, therefore, not a few observers doubt that opso- 
nins exist as true and separate antibodies, and are inclined to regard ther- 
molabile opsonin (largely the opsonin in fresh normal serum) as a com- 
plement, and thermostabile opsonin (largely immune opsonin or bacterio- 
tropin) as an amboceptor; it would appear that either alone, but more 
especially the latter, may facilitate phagocytosis to some extent. This 
process is, however, much more marked when both substances are acting in 
unison. While it is true that the bacteriolysin and opsonin content of a 
serum do not run parallel, our methods for measuring these are not entirely 
satisfactory; both intracellular and extracellular lysis may be mere differ- 
ences in degree, depending upon the nature of the bacterium or the con- 
centration of the antibodies rather than upon separate and distinct anti- 
bodies. 
Specificity of Natural and Immune Opsonins.—Normal serum usually 
contains opsonin for many different kinds of bacteria and erythrocytes. 
The question whether this wide range of opsonic action is dependent on a 
common opsonin or on several more or less specific opsonins has been an- 
swered differently by different investigators. 
The investigations of Bulloch and Western,1 MacDonald,2 Rosenow,3 
and Hektoen4 indicate that normal human serum contains several more or 
less distinctly specific opsonins for various bacteria and for alien red blood- 
corpuscles; on the other hand; Simon,5 York and Smith,6 Russel,7 Axamit 
and Tsuda,8 Muir and Martin,9 Levaditi and Inmann,10 and Klien11 have 
maintained on the basis of absorption tests that opsonin in normal serum 
for bacteria is a common opsonin and that thorough absorption of a serum 
with one bacterium will remove all of the normal opsonins for other bacteria. 
As previously stated, heating serum reduces opsonic activity, and it 
would appear that this thermolabile complement-like opsonin is non-specific 
and removable in part or whole by heating or by absorption not only with 
bacteria, but likewise with charcoal, chalk, yeast, cellular debris, and other 
substances, as shown by Simon, Neufeld, and Hune, Levaditi and Inmann, 
Muir and Martin, and others. Normal serum may, however, contain 
various thermostabile opsonins for bacteria and erythrocytes, and these 
appear to be specific and removable only by absorption with specific antigen. 
Immune opsonins developing during the course of infection or after vac- 
cination are thermostabile and highly specific; the specificity of these has 
never been seriously questioned. 
1 Lancet, 1905, 2, 1603; Proc. Roy. Soc., B, 1906, 77, 531. 
2 Aberdeen University Studies, 1906, 21, 323. 
3 Jour. Infect. Dis., 1907, 4, 285. 
4 Jour. Infect. Dis., 1908, 5, 249. 
5 Jour. Exper. Med., 1906, 8, 651; ibid., 1907, 9, 487. 
6Biodiem. Jour., 1906, 2, 74. 
7 Johns Hopkins Hosp. Bull., 1907, 28, 252. 
8 Wien. klin. Wchn., 1907, 20, 1045. 
9 Proc. Roy Soc., 1907, 79, 187. 
10 Compt. rend. Soc. de Biol., 1907, 62, 683. 
11 Johns Hopkins Hosp. Bull., 1907, 18, 245. 174 
OPSONINS 
Properties of Natural and Immune Opsonins.—Natural opsonins or those 
found in normal serum are largely thermolabile, that is, easily destroyed or 
inactivated by heating. According to Wright and Douglas,1 Hamilton,2 
and others they are practically destroyed by heating for thirty minutes at 
60° C. These opsonins also deteriorate quickly and disappear after three 
or four days’ exposure to room temperature. In other words, the general 
properties of normal or natural opsonins are closely similar to the com- 
plements. 
As shown by Levaditi,3 Neufeld,4 Muir and Martin,5 and others, immune 
opsonins, on the other hand, are highly thermostabile, resist drying, and are 
easily preserved. They closely resemble the amboceptors in these general 
properties. 
According to Noguchi6 opsonins reveal their maximum action in a 
medium of neutral reaction. In this respect, as well as their high resistance 
in the dry state to high temperatures, leads Noguchi to emphasize that 
opsonins have certain properties characteristic of the ferments. 
Source of Opsonins.—Little is definitely known regarding the source 
of opsonin. Thermostabile opsonin—that which is increased by artificial 
immunization or during disease, and is largely in the nature of an ambo- 
ceptor—is probably a product of general cellular activity, and especially of 
the local cells at the site of infection. Thermolabile opsonin—largely the 
opsonin occurring in normal serum, and in the nature of a complement— 
is probably a product of the leukocytes and other cells as well, as it has never 
been proved that the leukocytes are the sole source of the complements, as 
Metchnikoff would have us believe. 
They may be absent from inflammatory exudates, as shown by Opie,7 
due probably to absorption by bacteria or cellular debris. Woodhead and 
Mitchell8 have found opsonins in cows’ milk and whey in slightly less than 
the concentration in the corresponding sera. 
Susceptibility to Opsonification.—As previously stated in the chapter 
on Infection, not all bacteria are equally susceptible to opsonification. 
As a general rule, recently isolated and virulent micro-organisms resist the 
influence of opsonins until they have undergone culture several times. This 
resistance may be due to capsule formation, thickening of the ectoplasm, 
actual self-immunization of the bacterium, or the influence of endotoxins 
as a protective means against the antibodies of a host, all of these being 
weakened or lost upon artificial culture-media. 
Effect of Opsonins on Bacteria.—We know nothing definite regarding 
the manner in which opsonins prepare bacteria for phagocytosis except that 
opsonification in itself apparently does not impair the vitality of the bac- 
terium, in so far, at least, as its viability is concerned. 
Role of Opsonins in Immunity.—Although the exact identity of normal 
and immune opsonins and their relation to other antibodies is as yet un- 
settled, the important relation they bear to processes of immunity is gen- 
erally recognized, especially their ability in aiding resistance to infection 
by facilitating phagocytosis. That phagocytosis is an important factor in 
resistance to infection and recovery from disease cannot be denied, and the 
1 Proc. Roy. Soc., 1903, 72, 357; ibid., 1904, 73, 128. 
2 Jour. Infect. Dis., 1908, 5, 570. 
3 Ann. d. l’lnst. Pasteur, 1901, 15, 904. 
4 Centralbl. f. Bakteriol., 1907, 38, 456. 
6 Proc. Roy. Soc., 1907, 79, 187. 
6 Jour. Exper. Med., 1907, 9, 455. 
7 Jour. Exper. Med., 1907, 9, 515. 
8 Jour. Path, and Bacteriol., 1906, 11, 408. PRODUCTION OF IMMUNE OPSONINS 
175 
importance of opsonins in the processes of immunity are in direct relation. 
Evidences of phagocytosis by the circulating leukocytes are only occa- 
sionally encountered, but phagocytosis by the polymorphonuclear leukocytes 
in the tissues and by fixed tissue-cells is a common phenomenon. Opsonins 
greatly facilitate phagocytosis by polymorphonuclear leukocytes and by 
endothelial cells as well, as shown by Manwaring.1 They are operative in 
some infections more than in others, and they are especially active in those 
conditions in which phagocytosis is recognized as the chief defensive force, 
as, for example, in pyogenic infections. In these conditions their presence 
has been taken as a measure {opsonic index of the resistance of the host) 
and, largely through the researches of Wright and Douglas, a technic for 
detecting their presence, kind, and quantity in the body fluids has been 
devised, the method and information it yields being of value under certain 
limitations and in some infections. (See next chapter.) 
If experiments in vitro may be taken as an example of what occurs 
in vivo, it must be true that leukocytes are capable of consuming an enor- 
mous number of bacteria. Experiments with washed leukocytes—those re- 
moved from the influence of serum—show that spontaneous phagocytosis is 
very slight. Metchnikoff declared these experiments to be untrustworthy 
for the reason that the various manipulations of washing injures the vitality 
of the leukocytes. When, however, bacteria are opsonized, that is, are 
exposed to a serum containing opsonins, and then are thoroughly washed, 
it is found that the washed leukocytes engulf enormous numbers of bac- 
teria, showing that Metchnikoff’s objection to these experiments is un- 
warranted. Granting, then, that what we call opsonins are substances that 
facilitate phagocytosis, and that phagocytosis is a process of great im- 
portance, especially in certain infections, we must conclude that opsonins 
play a very important role in immunity; in fact, they constitute the very 
basis of the phenomenon of phagocytosis in the broader meaning of the 
term. 
Production of Immune Opsonins.—1. These may be produced in the 
same manner as the agglutinating serums, immune opsonins being readily 
demonstrated in the same serums. For actual diagnostic work artificial 
immune opsonins are seldom required, but to secure an immune serum for 
experimental studies on opsonins a culture of Staphylococcus pyogenes 
aureus may be used in immunizing a guinea-pig as follows: 
First dose: 1 loopful of twenty-four-hour agar culture in 2 c.c. 
NaCl solution heated for one-half hour at 58° C. and given sub- 
cutaneously. 
Second dose: 1 loopful in 2 c.c. NaCl, heated; intraperitoneally. 
Third dose: 2 loopfuls in 2 c.c. NaCl, heated; intraperitoneally. 
Fourth dose: 3 loopfuls in 2 c.c. NaCl, heated; intraperitoneally. 
Fifth dose: 6 loopfuls in 2 c.c. NaCl, heated; intraperitoneally. 
Sixth dose: 1 agar slant in 4 c.c. NaCl, heated; intraperitoneally. 
2. Bleed the animals one week after the last injection has been made. 
3. Owing to its large size, Bacillus anthracis may be substituted. This 
is a spore-forming organism, and since it is dangerous unless scrupulous care 
in handling is exercised, it is not usually wise to employ it in experimental 
work. 
1 Jour. Immunology, 1916, 1, 401. CHAPTER XI 
OPSONIC INDEX 
Whether opsonins are regarded as separate antibodies or as being 
identical with complements and amboceptors, a measure of their quantity 
and power may be of aid in formulating a diagnosis, as a guide to active 
immunization, and as one means of determining the potency of various 
immune serums used for therapeutic purposes, such as antimeningococcus 
and antipneumococcus serums. We are mainly indebted to Leishman, 
Wright and Douglas, Neufeld and Rimpau, and their co-workers for devising 
a technic that, however imperfect it may be according to the results ob- 
tained, has opened a new and important field for the study of immunologic 
processes. 
Principle.—This is based upon the method devised by Wright and 
Douglas, whereby it was sought to determine the amount and kind of 
opsonin in a patient’s serum by comparing the degree of phagocytosis 
with that occurring wrhen normal serum was used. 
Definition.—The opsonic index is the ratio of the number of bacteria ingested 
by a given number of phagocytes in the presence of a patient’s serum, to the 
number ingested by the same number of phagocytes in the presence of normal 
serum. 
“An equal volume of the patient’s serum, measured in a capillary pipet, 
is mixed with an equal volume of a suspension of washed leukocytes derived 
from a normal blood. After this ‘phagocytic mixture’ has been digested 
for a suitable period at 37° C., film preparations are made and stained. 
“A ‘phagocytic count’ is then undertaken, i. e., the average bacterial 
ingest of the leukocytes in the phagocytic mixture is determined, and this 
is compared with the average ingest of the leukocytes in a phagocytic mix- 
ture made wfith normal blood. 
“The expression thus obtained, 
Average ingest of the individual phagocyte in the mixture containing the 
patient’s serum. 
Average ingest of the individual phagocyte in the mixture containing nor- 
mal serum is .denoted the opsonic index” (Wright). 
Purpose of the Method.—The opsonic index aims to serve as a guide: 
1. In diagnosing the presence of bacterial infection, or rather in dis- 
covering whether the natural protective powers of the patient’s blood 
have been diminished or increased as the result of the immunizing influence 
of the infection. 
2. In connection with vaccine therapy, to guard against diminishing 
the opsonin content of the patient’s blood; to assure ourselves that our 
efforts to increase them have been successful, and occasionally to ascertain 
how long the store of opsonin that has been obtained for the patient remains 
in the blood. 
Limitation of the Method.—In ascertaining the opsonic index of a pa- 
tient’s serum, we must take it for granted—although it has not been proved: 
176 PRECAUTIONS IN TECHNIC 
177 
1. That the bacteria act the same in the body as they do in the test-tube. 
This is known not to be the case, for virulent organisms resist phagocytosis, 
whereas a non-virulent strain of the same bacterium is easily phagocyted. 
If, therefore, a laboratory culture of attenuated organisms is used in mak- 
ing the opsonic index, the result can hardly be accepted as a criterion of 
the power of the patient to overcome the “resistant” or more virulent 
organism as it occurs in the body. This source of error can be overcome 
in a manner if the micro-organism is isolated and used at once before attenua- 
tion occurs. 
2. That the leukocytes are a constant factor, and need not be taken 
into account. Investigation has shown that, as a result of infection, the 
leukocytes probably undergo qualitative changes and it is hardly fair to 
accept phagocytosis by normal leukocytes as a criterion of phagocytosis 
with the patient’s own leukocytes, as it occurs in the body during the in- 
fection. 
3. The method assumes that phagocytosis by the polynuclear leuko- 
cyte plays a large part in overcoming the infection. In many cases, how- 
ever, this is by no means proved. For example, in tuberculosis it is not 
this form of leukocyte, but the mononuclear form or the lymphocyte, which 
seems to be more important, and hence it is difficult to understand how the 
index with the polynuclear leukocyte can aid the question of diagnosis or 
treatment. 
4. The chances for error are considerable. To be of any value, the work 
requires experience and painstaking care. The results obtained by compe- 
tent workers with the same blood may show variation, but it must be said 
that, with strict attention to technic and insistence upon perfect prepara- 
tions, the worker may usually obtain valuable results. An index taken 
at one time by one person and later by another, and so on, will not be of 
as much value as when all are taken by the same worker, who brings prac- 
tice, skill, and conscientious care to his aid. 
Precautions in Technic.—1. Proper controls should be used. When 
dealing with the tubercle bacillus, the staphylococcus, or any other sapro- 
phyte of the external surfaces, or with any pathogenic organism with which 
we have normally no relations, the serum of a normal individual or the 
mixed serum of a number of normal persons will furnish the standard of 
comparison. When, on the other hand, we are dealing with intestinal 
bacteria or with a saprophyte of the mucous membrane, where, as a rule, 
any relation with them will be denied, it is difficult to establish a standard 
of health. Pooled serum is, therefore, necessary, and will furnish a standard 
for comparison for the purpose of measuring the fluctuations that may 
occur in the patient’s blood. 
2. A reasonable degree of phagocytosis should occur in the control 
serum. This is one of the main drawbacks to the value of the method for 
certain pathogenic organisms, as the pneumococcus, meningococcus, strepto- 
coccus, etc., may resist phagocytosis in normal serum, and thereby show 
abnormally high indices with immune serum. 
3. Efforts at spontaneous phagocytosis should be suppressed in order to 
measure more accurately the opsonin, as shown by the degree of phago- 
cytosis independent of the inherent activities of the cell itself. Spontaneous 
phagocytosis can largely be overcome by using 1 per cent, solution of sodium 
citrate such as is used for the collection of leukocytes. 
4. The ingest of a sufficiently large number of phagocytes should be 
counted. As a general rule, 100 cells should represent the minimum. 178 
OPSONIC INDEX 
The necessary constituents for making the test are as follows: 
1. The patient’s serum and normal serums to serve for the control. 
2. A bacteria] emulsion. 
3. A suspension of washed human leukocytes in normal salt solution. 
Collection of Patient’s and Control Serum.—1. The blood is collected in 
a Wright capsule, as described in Chapter II. 
2. Care must be taken not to heat the blood when sealing the tube. 
In drawdng off the serum avoid an admixture of corpuscles, as these may 
lower the opsonic index. 
3. If coagulation is incomplete or the serum has not been well separated, 
the clot may be broken up gently with a platinum wire and the tubes centri- 
fuged. Slight discoloration of the serum from mechanically breaking up 
a few erythrocytes will not interfere with the test. 
4. If gross contamination is avoided, blood may be kept for one or two 
days in a dark place without much deterioration of its opsonin content. 
5. It is always well to collect the control bloods at the same time the 
patient’s blood is taken, or, if this cannot be done, to place them in an ice- 
chest as soon after collection as possible. When conducting the test the 
control serums are pooled and mixed in a clean watch-glass. 
The Bacterial Emulsion.—1. This must be perfectly uniform, free from 
clumps, and must not undergo agglutination, either spontaneous or with 
the serum to be tested. 
With many bacteria, especially the motile ones, such as Bacillus coli 
and B. typhosus, it is comparatively easy to secure a uniform emulsion. 
Staphylococci, streptococci, and pneumococci, as a rule, present no difficul- 
ties. After growing the culture for from eighteen to twenty-four hours on 
slants of a suitable medium, add 3 c.c. of sterile 1 per cent, salt solution, 
and gently remove the bacterial growth with a platinum loop. The mix- 
ture is then pipeted into a separate flask or thick glass test-tube containing 
glass beads, and shaken by machine or by hand until it is thoroughly emul- 
sified. If necessary, the emulsion may be centrifugalized to remove clumps 
and is then ready for use. 
2. It must consist of bacteria that stain evenly and well. 
Only young cultures should be used; for example, an eighteen- to twenty- 
four-hour culture of freely growing organisms and a seven days’ growth of 
tubercle bacillus. 
3. It must be of such strength as to give a leukocytic ingest that will 
enable adequate differentiation to be made of the opsonin content of the 
various serums. 
An emulsion that does not yield a count of at least 250 bacteria within 
100 leukocytes is too weak to yield a satisfactory differential count. If 
the emulsion is too thick, bacteria overlie the leukocytes and introduce 
error. As a general rule, a suspension containing 500,000,000 bacteria 
per cubic centimeter is satisfactory. Experience will teach the right density 
to be used, and frequent trials may be necessary before the right one is 
secured. 
4. The bacteria must be such as wTill not prove seriously dangerous. In 
order to obviate any danger attending work with such organisms as the 
glanders and the tubercle bacillus, first kill the culture by pouring on a 
10 to 40 per cent, solution of formalin, mix the culture, shake, transfer 
to a centrifuge tube, and centrifugalize until the bacteria have been carried 
to the bottom of the tube. Pipet off the supernatant formalin, wash in 
TECHNIC TECHNIC 
179 
normal salt solution, centrifuge, pipet off again, and 
finally mix the sediment in sufficient salt solution to 
make a satisfactory suspension. 
The Washed Leukocytes.—These should consist 
mainly of the polynuclear leukocytes of a healthy per- 
son washed free from any admixture with serum. As 
usually obtained, the leukocytes are mixed with red cor- 
puscles. It is necessary to collect blood for the leuko- 
cytic mixture from a person whose corpuscles are known 
to be insensitive to agglutination, as otherwise there is 
an undue lowering of the opsonic effect. Owing to the 
fact that the leukocytes are likely to be altered in disease it 
would appear better practice to use the leukocytes from the 
patient, instead of those from a normal person, as de- 
scribed above and advised by Wright. 
1. Place 4 c.c. of sterile 2 per cent, solution of sodium 
citrate in distilled water in sterile 10 c.c. centrifuge tubes. 
As shown by Evans1 some of the sodium citrate on 
the market is of an acid reaction and, therefore, not 
suitable for use since leukocytes absorb H-ions from 
weakly acid solutions with injury to their phagocytic 
activities. The leukocytes may be protected if neces- 
sary by using a buffered saline solution. 
2. Prick the finger and add 1 c.c. of blood. Agitate 
well to insure thorough mixing. 
3. Centrifuge at a sufficiently high speed to mix the 
red corpuscles and leukocytes at the bottom of the tube 
and avoid clumping of the leukocytes. 
4. Draw off the supernatant fluid, add 5 c.c. of 1 per 
cent, salt solution, mix, and centrifuge. 
5. Wash once more. Draw off the supernatant fluid. 
6. Add sufficient salt solution to bring the total 
volume up to that of the blood originally taken, i. e., 
1 c.c. Mix well. 
7. The leukocytes should be used at once; as shown 
by Glynn, they tend to lose in phagocytic activity after 
standing two hours or longer. Leukocytes may also 
be obtained from a' rabbit or guinea-pig by injecting 3 
to 6 c.c. of sterile distilled water, peptone solution, 
bouillon, or suspension of aleuronaut into a pleural 
cavity and collecting the exudate after twenty-four hours. 
The Test.—1. Prepare capillary pipets of approxi- 
mately the same caliber. These are made by taking 
6-inch lengths of soft clean glass tubing having an ex- 
ternal diameter of inch, heating them in the middle 
in the tip of the blowpipe or the Bunsen flame until about 
| inch length of tubing is quite soft. Remove from the 
flame, and by rapidly separating the two hands draw 
out the molten glass to a length of from 18 to 20 
inches. After cooling, the capillary thread is cut across 
with a small file, so that from 6 to 8 inches is left at- 
tached to each piece of tubing. The ends must be cut 
square, as ragged and uneven ends are difficult to handle. By means of a 
Fig. 64.—Capillary 
Pipet for Opsonic 
Index Determina- 
tion. 
1 Jour. Immunology, 1922, 7, 271. 180 
OPSONIC INDEX 
wax-pencil make a fine mark at a point about an inch from the free end 
of each capillary thread. This indicates the unit volume (Fig. 64). 
2. Adjust a well-fitting rubber teat, and draw up the unit volume of 
blood-cells. A tiny bubble of air is now allowed to enter the thread, and 
then 1 volume of the bacterial emulsion is added; another air-bubble is 
allowed to enter, and finally one volume of serum, so that wre have named 
in their order in the capillary tube from above downward, one volume of 
blood-cells, an air-bubble, one volume of bacterial emulsion, an air-bubble, 
and one volume of serum (Fig. 64). 
Fig. 65.—Mixing the Contents of a Capillary Pipet. 
Due precautions must be exercised to avoid the formation of bubbles. 
3. By making gentle pressure on the teat these are then blown out on 
the surface of a clean glass slide, and perfect mixture effected by making 
alternate aspiration and expulsion from the capillary tube at least six times 
(Fig. 65). 
4. Carefully reaspirate into the capillary thread, so that the mixture 
occupies about the middle, and seal the tip in a low Bunsen flame (Fig. 66). 
5. Remove the teat, and with the wax-pencil mark the tube with the 
name or number of the serum. 
6. A similar preparation is prepared with the pooled serum (control). 181 
TECHNIC 
7. The phagocytic mixtures are then placed in an incubator at 37° C. 
for fifteen minutes, except in the case of such bacteria as the Bacillus typhosus 
Fig. 66.—Method of Sealing a Capillary Pipet. 
The tip of the pipet is placed in the edge of a flame. The teat is held in the same position until the tip 
has been sealed, when it may be removed without disturbing the contents of the pipet. 
Fig. 67.—An Opsonic Incubator. 
and the B. coli, as lysin and agglutinin may be present in the serums of such 
bacteria when the period is reduced to ten minutes. The special opsonic 182 
OPSONIC INDEX 
incubators built to accommodate individual pipets are particularly service- 
able (Fig. 67). 
8. The tubes are then removed from the incubator, the teats readjusted, 
the tip of the capillary threads scratched with a file, and evenly broken off. 
The phagocytic mixture is carefully expelled on a clean glass slide, and a 
The slide is laid on a flat surface; the drop of blood is placed near one end; the spreader is held 
between the thumb and middle finger of the left hand, at an angle of about 30 degrees, and quickly 
pushed to the opposite end of the slide. 
Fig. 68.—Method of Preparing a Blood Film. 
perfect mixture made by alternate aspiration and expulsion. Avoid air- 
bubbles. The whole is then reaspirated, and a small drop of the mixture 
placed on each of two clean slides that have been roughened with emery 
paper about f inch from one extremity. With the edge of a spreader slide 
held at an angle of about 30 degrees, and with moderate pressure, the drop 
Fig. 69.—Blood Films for Phagocytic Counts. 
The first slide (on the extreme left) is too thick and honeycombed, due to a greasy slide and large 
drop of blood; the second is likewise thick and uneven; the third is too thin, and was spread with too 
small an amount of blood and with the spreader held too upright; the fourth (extreme right) is a satis- 
factory film; it was spread on a dean slide, is even, smooth, and of the proper thickness. 
is distributed evenly along about 1§ inches of the surface of the slide (Fig. 
68). The smears are made in duplicate, because one may be more nearly 
perfect (Fig. 69) than the other, or one may be spoiled in the staining, when 
the second may be utilized. Each slide is then labeled at one end. 
9. After drying in the air, the slides must be fixed and stained. Fig. 70.—Tubercle Opsonic Index. 
A smear, stained after the method given in the text. Case IX, C. M., aged twenty-two years; active 
pulmonary tuberculosis; opsonic index, +1.6. 
Fig. 71.—An Unsatisfactory Film for a Phagocytic Count. 
Note that the leukocytes are collected in masses of erythrocytes. The slide was greasy and the smear 
too thick. 
Fig. 72.—A Satisfactory Film for a Phagocytic Count. TECHNIC 
183 
(а) For Ordinary Bacteria.—1. Fix by covering the slide with a satu- 
rated aqueous solution of mercuric chlorid for one minute. Wash in water. 
2. Cover with carbolthionin and stain for one or two minutes. Wash 
in water. 
3. Dry thoroughly. 
(б) For Acid-fast Bacilli.—1. Fix by inverting the films for thirty seconds 
■over a watch-crystal or jar containing formalin, being careful that there 
are no drops of formalin on the edge of the vessel that might come in con- 
tact with the preparation. The films may be fixed also with a saturated 
solution of mercuric chlorid or with pure methyl alcohol for two minutes. 
Wash in water and dry. 
2. Heat a portion of carbolfuchsin almost to boiling in a test-tube, and 
pour the hot stain over the films. Allow to remain for at least fifteen minutes. 
Wash under the tap and dry. 
3. Cover with a 5 per cent, solution of nitric or sulphuric acid for half 
a minute or longer if necessary, until decolorization is complete. Wash 
thoroughly under the tap. 
4. Cover with 4 per cent, aqueous solution of acetic acid for one to two 
minutes to remove the hemoglobin from the red cells. Wash and blot 
lightly. 
5. Cover with Lbffler’s methylene-blue for two minutes. Wash in water 
and dry thoroughly (Fig. 70). 
10. Examination of the stained films with the oil-immersion objective 
of the microscope will show that polynuclear leukocytes have collected 
more toward the edges and the end at which the spreading was completed. 
The individual leukocytes, however, should be separated from one another 
(Figs. 71 and 72). 
11. The edge of the film is examined, and the number of bacteria found 
in each series of five consecutive phagocytes is noted. If the technic has 
been satisfactory, no great divergence should be found in the count of each 
set of five cells. 
12. If the films are satisfactory, divide 100 phagocytes into groups of 
20. The average ingest of each group should not show a difference of over 
10 per cent., otherwise the technic has been faulty and it is necessary to 
count 250 phagocytes or to repeat the test. If divergence is due to the 
fact that every now and then one cell has a considerable higher ingest than 
others and the bacteria are well separated, hyperactivity of the cell is prob- 
ably the cause. If the bacteria are all clumped together it must be assumed 
that there has been a lack of care in preparing the bacterial emulsion or 
that agglutinin is present in the serum, and the test must be repeated with 
fresh precautions. 
13. In opsonic work the question as to how a certain element ought to 
be counted becomes quite evident and important. The proper method of 
procedure is to determine definitely how they may best be dealt with, and 
then to follow the rule adopted consistently. If an organism rests on the 
border of a cell, it will be better to consider it as within the cell and count it. 
Diplococci or division forms may be counted as one or as two, provided 
we are consistent in our method. Individual organisms, as distinguished 
from zoogleic masses, which may be lying on the top of the cell, are counted 
as if they were within the cell; for we have no means of determining definitely 
whether or not our suspicions are well grounded. In the case of a beaded 
■or vacuolated bacillus it is always better to count the whole element as a 
.unit. 
14. The phagocytic index is the average number of bacteria or other 184 
OPSONIC INDEX 
cells ingested per leukocyte after counting at least from 50 to 100 cells. 
The total number of bacteria ingested is divided by the total number of 
phagocytes, the result being the average number of bacteria ingested per 
leukocyte, i. e., the phagocytic index. 
15. The opsonic index is then given by the ratio: 
Patient’s phagocytic index 
Control phagocytic index 
For example, with patient’s serum, 100 phagocytes contain 300 bacteria, 
the phagocytic index being fD-§ = 3. With the control serum, 100 phago- 
cytes contain 500 bacteria, the phagocytic index being = 5. The opsonic 
index is: 
3 Patient’s phagocytic index _ 
5 Control phagocytic index 
16. Simon and Lamar have suggested a modification of Wright’s method 
that has been adopted by many laboratories. It consists in diluting the 
patient’s and control serums up to 1 : 10 or 1 : 100, and preparing mixtures 
of various dilutions with leukocytes and thinner bacterial emulsions. The 
percentage of phagocytic cells in the mixtures containing the serum to be 
tested is compared with the mixtures containing normal serum. It is, 
therefore, a method of comparative phagocytic indices. 
Veitch Method.—A much simpler technic, which has the distinct ad- 
vantage of using the patient’s own leukocytes, has been described by Veitch.1 
1. The ordinary capillary pipet described is employed fitted with a 
rubber teat or better with a piece of rubber tubing and mouthpiece. The 
pipet is marked about | inch from the free end in the usual manner. 
2. Blood is obtained from a finger; two volumes being drawn into the 
pipet followed by a bubble of air, two volumes of 1.5 per cent, solution of 
sodium citrate, bubble of air, and one volume of bacterial suspension. The 
whole is immediately blown out on a clean glass slide, thoroughly mixed, 
and finally drawn up into the pipet the end of which is sealed in a flame. 
3. The pipet is incubated for fifteen minutes, when smears are made and 
stained in the usual way. 
4. A control is set up at the same time and in the same manner, using 
the blood of a healthy person, the opsonic index for the patient being cal- 
culated as described. 
McCampbell Method. — Similar methods have been described by 
McCampbell2 and Crane,3 employing the ordinary white corpuscle pipet. 
McCampbell prepares the bacterial suspension by washing off young cul- 
tures from solid medium with small amounts of sterile 0.85 per cent, sodium 
chlorid, and 0.8 per cent, sodium citrate solution. The bacterial emulsion 
is first drawn to the mark 0.5, followed by an equal volume of patient’s 
blood. These are mixed, redrawn into the stem of the pipet, the ends being 
sealed with a rubber band, and incubated for fifteen minutes, when smears 
are made and stained in the usual manner. The blood of a healthy person 
is used in the same manner as a control for the purpose of expressing an 
opsonic index. 
QUANTITATIVE ESTIMATION OF BACTERIOTROPINS (NEUFELD) 
Of the various methods for standardizing an immune serum, particularly 
antimeningococcus serum, and of obtaining some idea of its potency, that 
1 Jour. Path, and Bacteriol., 1908, 12, 353. 
2 Amer. Jour. Med. Sci., 1911, 141, 724. 
3 Jour. Amer. Med. Assoc., 1909, 52. QUANTITATIVE ESTIMATION OF BACTERIOTROPINS 
185 
of determining the bacteriotropic or opsonic index of the serum is in most 
general use. 
Neufeld’s1 technic is that generally employed, and is similar to the 
serum dilution method employed by Simon. It varies from the technic of 
Wright in several particulars: 
1. The immune serum is free from complement (thermolabile opsonin). 
2. The actual number of bacteria within the leukocytes are not counted. 
Various dilutions of serum are used, and the highest dilution in which the 
bacteria are ingested in great numbers is compared with a normal serum 
in similar dilution as a control. The highest dilution that still favors phago- 
cytosis is then taken as the bacteriotropic titer of the serum. 
Serum.—The serum is inactivated by heating it to 55° C. for one-half 
hour. In old or carbolized serums this may be omitted, as they are usually 
free from complement. Tuberculous serums also should not be heated, 
as their bacteriotropins are very susceptible to heat. 
Normal serum from an animal of the same species as was used in the 
preparation of the immune serum should be used as the normal control. 
An exactly parallel series of dilutions wdth normal salt solution are made 
of the immune serum and pooled normal serums in a series of small test- 
tubes. At least 0.5 c.c. of each dilution should be available for the test; 
the following dilutions may be used: 1 : 10, 1 : 20, 1 : 50, 1 : 100, 1 : 200, 
1 : 400, 1 : 600, 1 : 800, 1 : 1000, 1 : 2000, etc. After working for some 
time with normal serums one soon learns the dilution in which the normal 
bacteriotropins are attenuated. It may not be necessary, therefore, to 
use the whole series of dilutions with the normal serum. 
Leukocytes may be obtained in several different ways: (1) By injecting 
a guinea-pig intraperitoneally sixteen to twenty-four hours previously 
with from 5 to 10 c.c. of sterile aleuronat solution. Pipet the peritoneal 
exudate into about 20 c.c. of sterile 1 per cent, sodium citrate in normal 
salt solution in centrifuge tubes. Centrifugalize, and wash the leukocytes 
again three times with sterile normal salt solution. The sodium citrate 
solution prevents the coagulation and formation of clumps of leukocytes. 
2. Instead of aleuronat a sterile 25 per cent, solution of peptone may be 
injected in the same amount. 
3. If rabbit’s leukocytes are preferred, 3 to 6 c.c. destilled water or 10 
c.c. of aleuronat should be injected into each pleural sac or 20 c.c. intra- 
peritoneally. For mice, an injection of 1 c.c. of aleuronat intraperitoneally 
is sufficient; human leukocytes may be obtained after the method of Wright. 
4. After the final washing the leukocytes are suspended in sufficient 
normal salt solution until an opacity equal to a 0.3 per cent, lecithin emul- 
sion in salt solution is attained. 
Culture.—Cultures should be selected with great care in order to avoid 
using one that displays a well-marked tendency to undergo “spontaneous 
phagocytosis,” or, on the other hand, one unduly resistant to phagocytosis. 
Usually an old strain of meningococci is serviceable; it is generally neces- 
sary to try out a number of strains and select the best. 
Meningococci are grown for twenty-four hours on slants of glucose 
agar. To each slant add 0.5 c.c. each of bouillon and of normal salt solu- 
tion, and emulsify the growth. Or the bacteria may be employed in the form 
of a sixteen- to twenty-four-hour homogeneous broth culture. Tubercle 
bacilli may either be triturated in an agate mortar wdth 1.5 per cent, salt 
solution added slowly drop by drop, or the tubercle powder of Koch may 
be employed in an emulsion prepared in the same manner. The resultant 
1 Arb. a. d. k. Gesundheitsamte, 1907, 25, 165. 186 
OPSONIC INDEX 
emulsion should be freed from coarser clumps by brief centrifugalization, 
but, as a general rule, it is very difficult to secure a uniform emulsion of 
tubercle bacilli by any method. 
The Test.—1. The mixtures are made preferably in a series of small 
test-tubes about 5 cm. long and 1 cm. wide. 
2. Mix 0.1 c.c. of each dilution of immune serum with 0.1 c.c. of bac- 
terial emulsion. Plug each tube with cotton. 
3. Mix and incubate for one hour. 
4. Add 0.1 c.c. of leukocytic emulsion to each tube. Double this quantity 
may be used if the emulsion is weak. 
5. Mix and incubate for from one-quarter to two hours, depending 
upon the variety of micro-organism. 
6. At the end of the second incubation the leukocytes will have settled 
to the bottom of the tubes. Carefully remove the supernatant fluid from 
each tube; mix the sediment well with a loop, and make smears on slides. 
Label each slide carefully to correspond to its serum dilution. 
7. Dry the smears in the air, fix with methyl alcohol, and stain with 
carbolthionin, as previously described. 
The Controls.—1. A series of the lower dilutions of pooled normal 
serums are set up in exactly the same manner. 
2. A tube containing bacteria and leukocytes without serum—to detect 
the extent of spontaneous phagocytosis. 
Readings.—A great number of fields are examined microscopically, and 
a note is made of the weakest dilution that still favors phagocytosis. No 
attempt is made to count the phagocyted bacteria. The relative amount 
of phagocytosis with the immune serum in various dilutions is compared 
with the normal controls, and the result is expressed as the bacteriotropic titer. 
Simon’s Method.—This method of counting the number of empty 
leukocytes with a given dilution of serum is followed; a similar count is 
made of the normal serum control in the same dilution; thus, if in the con- 
trol film 25 per cent, of leukocytes were empty, and in the patient’s film, 
50 per cent., the index would be -§-§-=0.5. A study of the results obtained 
by this method, and by careful counting after Wright’s method, shows that 
they are fairly comparable, and the method may be used where it is only 
necessary to determine whether the index is high or low. 
Hygienic Laboratory Method.—The method employed by Evans in the 
Hygienic Laboratory for estimating the bacteriotropins in antimeningococcus 
serum is described in Chapter XL. 
Precautions.—1. If phagocytosis is entirely absent one should not con- 
clude that bacteriotropins are not present. The leukocytes may have been 
injured, especially if heterologous leukocytes have been present; control 
examinations with homologous leukocytes (i. e., from the same animal), 
as the serum should result in phagocytosis. 
2. The time during which the tubes were in the incubator may have 
been too short or too long. Most micro-organisms require from one-half 
to two hours—meningococci require about one-half hour; pneumococci 
usually need at least two hours; typhoid and cholera bacilli about fifteen 
minutes to thirty minutes, as they undergo extracellular or even intra- 
cellular lysis quite readily. 
3. If the control of bacteria and leukocytes alone shows well-marked 
phagocytosis the test should be repeated with another strain. 
The method employed by Evans in the Hygienic Laboratory for esti- 
mating the tropins in antimeningococcus serum is described in Chapter 
XL. PRACTICAL VALUE OF THE OPSONIC INDEX 
187 
Practical Value of the Opsonic Index 
1. In competent hands the opsonic index of normal persons to most 
pathogenic organisms has been found to vary from 0.8 to 1.2. As previously 
mentioned, it is difficult to find a perfectly normal serum for such micro- 
organisms as the Bacillus coli, the staphylococci, B. tuberculosis, etc., as it is 
unlikely that any individual can altogether escape active infection at some 
period of his life. As menstruation approaches, even wider fluctuations occur, 
the normal index being re-established by the second or third day. During 
the first year of life the opsonic index varies to such a degree that it has little 
or no practical value. 
2. Although a large amount of work has been done with the opsonins 
in disease, it is the consensus of opinion that the determination of the opsonic 
index has less practical significance than was originally believed. One point 
is clear, however, that the work of Wright and others has broadened the 
field of vaccine therapy and placed it upon a firmer foundation. Aside 
from the 10 per cent, of chances of technical error in making an opsonin 
measurement, other factors may be present that are entirely beyond con- 
trol and cannot be measured by the immunisator, and that may seriously 
affect the value of the opsonic index. 
3. As a diagnostic procedure, Wright has shown that the opsonins possess 
a certain specificity, and that in a given infection a low index for a certain 
micro-organism indicates that this organism is probably the etiologic factor. 
This possibility is strengthened if the opsonic index for this micro-organism 
is increased by careful manipulation or exercise of the diseased part, when 
auto-inoculation occurs, with consequent increase of opsonin. 
4. In prognosis the opsonic index may have some value in deciding 
whether an infection has been entirely overcome or is still active. An 
attempt is made to induce auto-inoculation, as by gentle massage of a 
knee-joint or a hip; active exercise; deep breathing, etc., and the index is 
made before, and at frequent intervals after, such attempts. If the index 
remains unchanged within the normal limits, the assumption is that the 
infection has been overcome; if, on the other hand, an increase in opsonin 
occurs, this indicates that an active focus remains. 
5. Most value was placed by Wright upon the opsonic index as a guide 
to the size and frequency of doses of bacterial vaccines in the treatment of 
disease. A large number of careful determinations showed that an injec- 
tion of vaccine is followed by a decrease of the opsonins (negative phase), 
which is of variable degree and duration, according to the amount injected 
(Fig. 73). This is followed by an increase (positive phase), and coincidentally 
there is a corresponding improvement in the patient’s condition. This 
subject is discussed more fully in the chapter on Active Immunization. 
The purpose of proper vaccination, therefore, is so to gage and time the 
different doses that a pronounced or prolonged negative phase is prevented 
as far as possible, and a high positive phase secured and maintained. It is 
obvious that the technic of opsonic measurement consumes much time, 
and that the immunisator cannot mark the index at the time a dose of vac- 
cine is given. However, the determination of the opsonic index at proper 
intervals after the first dose of vaccine may give valuable information as 
regards the reaction of the patient, and serve as a guide to the size and 
frequency of subsequent doses. 
As a routine measure the opsonic index has fallen into disuse, vaccine 
therapy being largely guided by the clinical evidences of reaction and the 
condition of the patient. That it has distinct value, particularly in scientific 188 
OPSONIC INDEX 
investigation, is generally admitted, and it is well to remember that in the 
early years following Wright’s investigations the practice of vaccine therapy 
was limited to those skilled in determining the index, preparing the vaccine, 
and carefully guiding and guarding its administration. It is to be regretted 
that the wholesale and indiscriminate manufacture and use of vaccines 
have brought this valuable field of therapy inevitably into disrepute. This 
Fig. 73.—An Opsonic Index Chart. 
is being realized more and more, and the effort is being made to restore 
the value of this form of therapy. This effort consists in recognizing the 
possibilities and limitations of the method, and confining its practice to 
those who possess at least sufficient knowledge of bacteriology to prepare 
a vaccine and make an opsonic measurement, the best results being secured 
by co-operation between bacteriologist and clinician. CHAPTER XII 
PREPARATION OF BACTERIAL VACCINES 
In this chapter a method for preparing bacterial vaccines will be de- 
scribed, the general discussion of vaccine therapy, with the special technic 
for preparing cowpox vaccine, rabies vaccine, tuberculin, and other special 
vaccines, being taken up in the chapter dealing with Vaccines in the 
Prophylaxis of Disease. 
Definition.—Bacterial vaccines are “sterilized and enumerated suspensions 
of bacteria which furnish, when they dissolve in the body, substances which 
stimulate the healthy tissues to a production of specific bacteriotropic substances 
which fasten upon and directly or indirectly contribute to the destruction of the 
corresponding bacteria'1'1 (Wright). 
TECHNIC FOR PREPARING BACTERIAL VACCINES 
Bacterial vaccines are made: (1) By procuring the infected material; 
(2) by preparing pure cultures of the bacteria that are to be attacked; (3) 
by making suspensions of these in saline solution, adding a preservative, 
and placing in proper containers. 
1. Procuring Infected Material.—Various precautions, according to 
existing circumstances, should be taken to avoid contamination and to 
secure material that is truly representative of the focal secretions. For 
instance, pus should be collected from an abscess cavity or sinus after the 
surrounding tissues have been cleansed with dilute tincture of iodin, for 
if we secured a culture of the relatively harmless Staphylococcus epidermidis 
albus from the skin instead of the Staphylococcus aureus, which may be 
the cause of infection, our vaccine will have little or no value. 
Nasal secretion may be secured after cleansing the nasal orifice with 
soap and warm water, passing a sterile cotton swab through a nasal speculum, 
and rubbing the surfaces of the lower turbinates and septum lightly. 
An ear should be cleansed, the excess of secretions removed with sterile 
swabs, and the culture be made of pus from the infected tissues. Various 
saprophytes quickly gain admission and grow in the necrotic pus, whereas 
the infecting bacterium is more likely to be found in the tissues. 
In the collection of sputum special care is required: the patient should 
be instructed to brush the teeth with a sterilized brush, rinse the mouth 
several times with boiled water, and after swallowing several mouthfuls of 
wat.er to cough and expectorate into a wide-mouthed sterilized bottle. 
The sputum may be plated at once, or washed several times in sterile Petri 
dishes with sterile water and then cultured. 
Lung puncture may occasionally be required in infective lung conditions 
in which sputum is not obtainable or is too badly contaminated. A 5 to 
10 c.c. all-glass syringe with a strong needle is sterilized by boiling, and 
2 or 3 c.c. of peptone broth introduced. The skin of the chest wall over 
the site of infection, as shown by clinical evidence, is sterilized with tincture 
of iodin, and puncture made into the pulmonary tissues. When the de- 
sired depth has been reached, 1 c.c. of the broth is injected gently into the 
tissues, and after the lapse of a few seconds reaspirated as far as possible 
into the syringe. During the operation the patient should refrain from 
189 190 
PREPARATION OF BACTERIAL VACCINES 
respiratory movements in order to minimize any risk of lacerating the 
pulmonary tissues (Allen). 
Urine should always be withdrawn with a sterile catheter after thor- 
oughly cleansing the meatus. This last is especially important, for the in- 
fection may be due to a certain strain of Bacillus coli, and unless we are 
successful in obtaining a culture of this particular strain the vaccine will 
have little value. 
Blood specimens are taken with a sterile syringe from a prominent vein 
at the elbow after the skin has been cleansed and sterilized with tincture 
of iodin, and cultured in large amounts on proper culture-media. 
2. Preparing Pure Cultures.—This is frequently the most difficult step 
in the whole technic, for some micro-organisms, as, for example, the gono- 
coccus and Bacillus influenzas, grow slowly, require special culture-media, 
and their colonies may easily be overlooked. To secure pure cultures, and 
especially to select one or at most two organisms that may be the chief 
offenders, considerable bacteriologic knowledge is necessary, and no simple 
rules or directions can be given in the limited space of this volume. 
1. Stained smears of the secretions of a lesion may indicate the nature 
of the infection and the best culture-medium and technic to use for pur- 
poses of isolation. 
2. Cultures of the lesion may be made upon solid media, and isolation 
carried out after a primary growth has been secured. With proper care 
primary cultures may be grown, such as staphylococci from the pus of a 
freshly incised abscess, or the micro-organism of a case of cystitis or pyelitis 
by securing urine with the aid of a sterilized catheter. If slowly growing 
organisms, such as Bacillus influenzas, gonococcus, pneumococcus, etc., are 
to be cultured, “streak” plates are usually satisfactory, and as a routine 
the best culture-media are, as a rule, those containing serum or blood. 
3. The cultures that are to be worked up into a vaccine are usually best 
made on solid media. If the cultures show the presence of a single micro- 
organism the preparation of a vaccine is relatively simple; if more than 
one bacterium is present it is advisable to plate out the mixture and secure 
each germ in pure culture. The practice of preparing mixed vaccines by 
washing off the mixed growths is not to be recommended because there is 
no way of insuring the presence of sufficient numbers of the different bac- 
teria, and the most important one may be present in few numbers by reason 
of being overgrown by the more hardy bacteria. Saprophytic bacteria 
should, of course, be eliminated and separate vaccines prepared of the 
other bacteria decided upon for incorporation in the vaccine. After each 
vaccine has been prepared a mixture is made in such a way that each cubic 
centimeter of the vaccine carries the desired number of each bacterium. 
4. In making a bacterial vaccine of a freely growing micro-organism for 
an individual patient it will suffice to plant two agar slant tubes; when 
dealing with bacteria that grow much less luxuriantly, such as streptococci 
and pneumococci, four to six tubes should be used. 
5. Cultures are usually grown for twenty-four hours at 37° C., but in 
the case of rapidly growing organisms a shorter period in the incubator 
will suffice. 
6. When the cultures are ready a smear of each growth is made and 
stained in order to see that pure cultures of the right micro-organism were 
made. 
7. Inasmuch as the immunizing power of a vaccine is in most cases a 
factor of the virulence of the organism, this being especially true of such 
organisms as the pneumococcus, streptococcus, and influenza bacillus, it TECHNIC FOR PREPARING BACTERIAL VACCINES 
191 
is well, whenever possible, to employ the first pure subculture for the prepara 
tion of the vaccine. 
3. Preparation of the Emulsion.—Carefully observing aseptic precau- 
tion throughout, pour a portion of a test-tube of a sterile normal salt solu- 
tion over the surface of the first culture, shaking the fluid in such a way 
Removing the culture of bacteria by gently rubbing over the medium (agar-agar) with a sterilized 
platinum loop. 
Fig. 74.—Preparation of a Bacterial Vaccine. 
as to bring the micro-organisms into suspension. If the culture is not easily 
washed from the medium a sterile platinum loop may be used to remove 
the growth, care being taken not to cut into the medium and mix the frag- 
ments with the bacterial suspension (Fig. 74). 
Fig. 75.—A Satisfactory Shaking Machine Mounted on a Concrete Block 
(International Centrifuge Co.) 
The bacterial suspension thus obtained is poured on the surface of the 
second culture, bringing this into suspension, and repeating the process 
until the whole series of cultures have been suspended, adding more salt 
solution if necessary. 192 
PREPARATION OF BACTERIAL VACCINES 
The final suspension is transferred to a sterile, thick-walled flask con- 
taining glass beads, and shaken by hand or in a mechanical shaker until 
the bacterial masses are broken up (Fig. 75). This may 
be especially difficult with diphtheria bacilli and strepto- 
cocci. Unless the emulsion is perfectly homogeneous, the 
larger particles may be removed by brief centrifugalization 
or, better, by filtering through a sterile filter. 
There is evidence to show that bacteria grown on cul- 
ture-media containing peptone may produce objectionable 
toxic substances capable of producing anaphylactic phe- 
nomena (Reichel and Harkins1). In addition, when, in the 
preparation of a vaccine, bacteria grown on a serum medium 
are washed off with normal salt solution, a portion of the 
serum may be removed and this may be capable of produc- 
ing disagreeable local and general reactions. For these 
reasons it is advisable to wash2 all suspensions by repeated 
centrifugalization until the supernatant fluid reacts nega- 
tively to the biuret or ninhydrin reaction (Willard 
Stone). 
4. Counting of the Bacterial Suspension.—Standardiza- 
tion, best accomplished by counting the bacterial elements 
contained in a unit volume of the suspension, is necessary 
in order to adjust our initial dose as experience will dictate 
and for guidance in making subsequent injections. 
In dealing with a vaccine we have to count both the 
dead and the living bacteria, making no distinction, for 
both furnish the chemical agent that calls forth the elabo- 
ration of bacteriotropic substances. Inasmuch as sharp 
definition and the staining properties of bacteria may be 
lost in the process of sterilization by heat, the specimen of 
vaccine to be examined should be secured before steriliza- 
tion is undertaken. 
The counting or standardization may be done in several 
ways: (a) By mixing equal portions of normal blood and 
bacterial emulsion and counting the proportion of cor- 
puscles to bacteria in our mixture (Wright); (b) by dilut- 
ing, staining, and counting with the hemocytometer, as 
in the enumeration of red blood-corpuscles; (c) for stand- 
ardizing large quantities of bacterial vaccine the method 
of Kolle or (d) that of Hopkins may be used. Nephelo- 
metric methods are also quite simple, quick, and fairly 
accurate. 
Of these four methods probably counting by means of the 
hemocytometer is the most accurate and to be preferred. In 
the Wright method the counts are apt to be too low and 
the bacteria irregularly distributed in the film. Glynn, 
Powell, Rees and Cox,3 and Fitch4 have found the counting 
chamber method best, and especially a chamber having a 
depth of 0.02 mm. used for counting platelets, although 
the usual chamber having a depth of 0.1 mm. may be 
used satisfactorily. 
Fig. 76.—A Capillary Pipex for Counting a Bacterial Vaccine. 
This illustration shows the pipet loaded with three equal volumes of sodium citrate solution, blood, and bacterial emulsion in proper order. 
1 Centralbl. f. Bakteriolog., etc., 1913, 69, 142. 
2 Jour. Amer. Med. Assoc., 1914, 63, 1011. 
3 Jour. Path, and Pact., 1913, 18, 379. 4 Jour. Amer. Med. Assoc., 1915, 64, 893. The micro-organisms are well separated and evenly distributed among the corpuscles; the spread is 
even and regular, and the corpuscles are not gathered in rouleau formation or irregular clumps. 
Fig. 77.—A Satisfactory Preparation for Counting a Bacterial Vaccine. 
Fig. 78.—An Unsatisfactory Preparation for Counting a Bacterial Vaccine. 
The micro-organisms are mostly gathered in irregular masses instead of being separated and 
evenly distributed; the corpuscles are not separated and evenly distributed, but show a tendency to 
gather in clumps; both factors render a count difficult, inaccurate, and unsatisfactory. TECHNIC FOR PREPARING BACTERIAL VACCINES 
193 
(a) Method of Wright.—Prepare a simple capillary pipet, making a 
mark on its stem about an inch from the tip, and fit a teat to its barrel 
(Fig. 59). Cleanse and prick the finger, press out a drop of blood, take 
up the pipet and draw up into it first one volume of sodium citrate solu- 
tion, one of blood, and then either one volume of bacterial suspension or 
two or more volumes, if it appears on inspection to contain much fewer 
than 500,000,000 of bacteria to the cubic centimeter. To guard against 
crimping of the corpuscles in drying the films Wright advocates aspirating 
one or two volumes of distilled wrater after the blood and bacterial suspen- 
sion. 
Now expel from the pipet first only the distilled water and bacterial 
emulsion, and mix these, so that there may be no danger of the red cor- 
puscles becoming hemolyzed, and then proceed to mix together the whole 
contents of the pipet, aspirating and re-expelling these a dozen times. Then 
make two or three microscopic films from the mixture, spreading these 
out on slides that have been roughened with emery. 
The films are dried in the air, fixed by immersing them for two minutes 
in a saturated solution of corrosive sublimate, washed thoroughly, and 
stained for a minute with carbolfuchsin diluted 1 : 10 or carbolthionin for 
two to five minutes, and then washed and dried. 
The films are now given a preliminary examination. If red corpuscles 
and bacteria are found in approximately the same numbers and the sus- 
pension is free from bacterial aggregates, the count may be made (Fig. 77). 
If either the bacteria or the corpuscles are largely in excess, new mixtures 
and new films must be made. In case the bacteria are gathered in clumps 
the suspension should be shaken again and new films prepared (Fig. 78). 
When satisfactory films have been obtained the actual counting may 
be done. This is carried out with an oil-immersion lens, and in order to 
secure accuracy it is necessary to restrict or divide the field by a small 
square diaphragm made of paper or cardboard, or by inscribing cross lines 
on a small clean cover-glass and dropping them on the diaphragm of the 
eye-piece. 
A field is now chosen at random, and the corpuscles and bacteria are 
counted, the results being jotted down on a sheet of paper, keeping each 
enumeration separate and writing the numbers in two columns (Fig. 79). 
Proceed at random from field to field, traversing every part of the slide. 
Establish a rule for counting corpuscles that transgress or touch the edge 
of the field. Eliminate from consideration any parts of the films in which 
Fig. 79.—Counting a Bacterial Vaccine After the Method of Wright. Special Ocular. 194 
PREPARATION OF BACTERIAL VACCINES 
the preparation is unsatisfactory as regards the staining, or with respect 
to the integrity of the red corpuscles. The examination is continued until 
at least 500 corpuscles have been counted, half of the count being made 
from the second slide. The number of micro-organisms counted is now 
totaled, and the approximate number per cubic centimeter estimated. 
Let us assume, for example, that 600 red cells and 1200 bacteria have been 
counted. Now, a cubic millimeter of blood contains 5,500,000 red cor- 
puscles, and equal volumes of blood and emulsion were taken. A cubic 
, , . 5,500,000 X 1200 
millimeter of the emulsion, therefore, contains = 11,000,000 
organisms per cubic millimeter, or 11,000,000,000 per cubic centimeter. 
(,b) The second method of counting is precisely similar to that employed 
for the enumeration of blood-corpuscles, the diluting and staining fluid 
being made by adding to a 1 per cent, solution of sodium chlorid in dis- 
tilled water sufficient formalin to make 2 per cent., and alcoholic gentian- 
violet, 5 per cent. The emulsion is drawn up in a white corpuscle pipet 
to the mark 0.5, and with diluting fluid to the mark 11. The contents 
are then mixed thoroughly for several minutes, several drops expelled, 
a drop placed in the counting chamber, and properly covered with a special 
thin cover-glass. The bacteria are allowed to remain in the counting cell 
for at least half an hour prior to enumeration. A large number of small 
squares are counted, and the average of one square obtained. By multiply- 
ing this figure by 4000 and then by 20, the number of bacteria per cubic 
millimeter is obtained, and 1000 times this figure gives the number of bac- 
teria contained in 1 c.c. of the vaccine. If the emulsion is highly concen- 
trated the red cell pipet may be used and the calculations made accordingly. 
Callison has advised the use of the following method: 
Using the red corpuscle pipet, the bacterial emulsion is drawn to 0.5 and diluting fluid 
to 101. Shake vigorously for three minutes, allow two or three drops of fluid to escape and 
load the hemocytometer chamber. After standing one-half hour the bacteria in a small square 
are counted or several small squares are counted and the average taken for one square. Mul- 
tiplying this figure by 8 and adding eight ciphers gives the number of bacteria per cubic 
centimeter of vaccine. 
The diluting fluid is prepared as follows: 
Hydrochloric acid, 2 c.c. 
Bichlorid of mercury (1 : 500), 100 “ 
Acid fuchsin, 1 per cent, aqueous solution, q. s. to deep cherry red. 
Filter. 
(c) Method of Kolle.—A platform loop adjusted to fit No. 2 of a Lauten- 
schlager wire gage (Fig. 80) measures about 4 mm., and holds approxi- 
mately 2 mg., or about 2,500,000,000 organisms. By growing an organism 
on slants of agar and emulsifying a certain number of loopfuls in a measured 
quantity of saline solution an approximate method of standardization is 
obtained. According to Kolle, ordinary slants of agar will hold about 
15 loopfuls of staphylococci, Bacillus typhosus, or B. coli, and about 5 loopfuls 
of streptococci and gonococci. 
(d) Method of Hopkins}—This is based upon concentrating a bacterial 
suspension by centrifugalization and the preparation of standard emul- 
sions from the sediment. The emulsion is filtered through a small cotton 
filter to remove larger clump of bacteria and particles of agar, and is then 
placed in specially constructed centrifuge tubes (International Centrifuge 
Company, see Fig. 82), covered with rubber caps, and centrifugalized for 
1 Jour. Amer. Med. Assoc., 1913, x], 1615. TECHNIC FOR PREPARING BACTERIAL VACCINES 
195 
half an hour at a speed of approximately 2800 revolutions a minute. The 
salt solution and bacteria above the 0.05 mark are then removed, and 5 c.c. 
saline solution is measured into the tube, so as to make a 1 per cent, emul- 
sion. If the sediment does not reach the 0.05 mark, its volume is read on 
the scale, and a corresponding quantity of saline is added to make the emul- 
Fig. 80.—Instrument for the Standardization of Platinum Loops. 
The 2-mm. loop holds approximately 2,500,000,000 bacteria (as Bacillus typhosus). 
sion 1 per cent, in strength. The bacteria are forced into suspension, the 
vaccine transferred to a sterile tube, and the micro-organisms killed in the 
usual manner. 
Estimations of carefully counted suspensions obtained by centrifugali- 
zation in the foregoing manner gave the following results: 
Per 
Cent. 
Billion 
Per c.c. 
Staphylococcus aureus and albus 
1 
10.0 
Streptococcus hasmolyticus 
1 
8.0 
Gonococcus 
1 
8.0 
Pneumococcus 
1 
2.5 
Bacillus typhosus 
1 
8.0 
Bacillus coli 
1 
4.0 
(e) Nephelometer Methods.—McFarland1 has devised a useful instru- 
ment utilizing precipitates of barium sulphate for the purpose of preparing 
bacterial suspensions of proper density for opsonic work and in the prepara- 
tion of vaccines: 
Two solutions, one a 1 per cent, solution of chemically pure sulphuric 
acid, the other a 1 per cent, solution of chemically pure barium chlorid, 
are prepared, then the one added to the other so that ten standards, con- 
taining 9.9 c.c. of the sulphuric acid and 0.1 c.c. of the barium chlorid, 9.8 
1 Jour. Amer. Med. Assoc., 1907, 49, 1176. 196 
PREPARATION OF BACTERIAL VACCINES 
c.c. of the sulphuric acid, and 0.2 c.c. of the barium chlorid, 9.7 c.c. of the 
sulphuric acid and 0.3 c.c. of the barium chlorid, etc., are prepared. Accord- 
ing to the number of cubic centimeters of barium chlorid added, these are 
denominated 1, 2, 3, 4, etc. This prepares 10 c.c. of each standard solution 
which are sealed in small test-tubes or ampules, care being taken to use 
tubes of uniform diameter, and also to provide similar tubes for mixing the 
bacteria to be compared with the standard (Fig. 81). 
With this nephelometer a vaccine of density corresponding to Tube 
No. 1 contains approximately 300,000,000 per cubic centimeter; density 
corresponding to Tube No. 3 contains about 1,000,000,000 per cubic centi- 
meter, and density corresponding to Tube No. 5 about 1,800,000,000 per 
cubic centimeter, etc. 
Fig. 81.—Nephelometer. 
Recently Dunham1 has advised the use of the Kober nephelometer for 
the standardization of vaccines. 
Vaccines of Fungi.—In preparing vaccines of fungi as for the treatment 
of ringworm, actinomycosis, and sporotrichosis, the fungus is grown on 
appropriate solid media, scraped off in such manner as to avoid removal 
of the medium, and ground in a sterile mortar with weighed amounts of 
crystals of sodium chlorid. Grinding should be continued until the threads 
are well broken up. Sterile water is gradually added until the resulting 
emulsion is isotonic and of a density corresponding to about Tube No. 5 
of McFarland’s nephelometer. The vaccine may be briefly centrifuged 
to remove clumps, but cannot be filtered. It is sterilized with heat and 
preserved with 0.3 per cent, tricresol in the usual manner. 
5. Sterilizing the Vaccine and Testing its Sterility.—When, after the 
preliminary examination, the films for counting have been found satis- 
1 Jour. Immunology, 1920, 5, 337. TECHNIC FOR PREPARING BACTERIAL VACCINES 
197 
factory, a pause is made to start the process of sterilization, which may 
continue wdiile the count is being made. Either heat or a germicide may 
be used for sterilizing vaccine, preferably the former. 
The vaccine may be placed in a test-tube, which is then sealed (Fig. 
83), and the whole immersed in the water-bath; a simpler method, and 
one just as good, is to place the flask or tube of vaccine in the bath, observ- 
ing special care to see that the water is above the level of the vaccine. 
Efficient sterilization is dependent upon permitting the process to continue 
at the minimum temperature for the minimum length of time. With the water- 
bath at 56° to 60° C. sterilization is nearly always complete in an hour. 
Boiling is likely to reduce the antigenic activity of a vaccine.1 
Fig. 82.—Hopkins’ Tube 
for Standardizing a 
Bacterial Vaccine. 
Fig. 83.—A Stock 
Ampule of Vac- 
cine. 
Fig. 84.—Stock Bottle of Bacterial 
Vaccine. 
The vaccine should be now cultured to test its sterility. At least a 
dozen platinum loopfuls are transferred, under strict aseptic precautions, 
to a slant of a suitable culture-medium, such as Loffler’s blood-serum or 
blood-agar; this is incubated at least twenty-four hours, or longer if the 
organism is a slowly growing one. It is then examined, and if found sterile, 
the preparation of the vaccine may be completed. If not, the vaccine is 
heated for another hour, or, preferably, a new vaccine is prepared. 
6. Dosage of Bacterial Vaccines.—As a general rule it is a good practice 
to so prepare a vaccine that each cubic centimeter contains the maximum 
number desired; in administering the vaccine the first dose may be 0.1 or 
1 Lancet, London, 1915, 2, 150. 198 
PREPARATION OF BACTERIAL VACCINES 
0.2 c.c. gradually increased until the maximum amount of 1 c.c. is being 
given at one time. Further discussion on dosage, method of administration, 
and reactions is reserved for the chapter on Vaccine Therapy. 
In preparing vaccines of one micro-organism each cubic centimeter may 
tain the following numbers of various bacteria: 
Staphylococci 
1.000.000,000 
Streptococci 
500,000,000 
Pneumococci 
500,000,000 
Gonococci 
500,000,000 
Micrococcus catarrhalis 
500,000,000 
Bacillus typhosus 
1,000,000,000 
Bacillus para typhosus 
1,000.000,000 
Bacillus coli 
1.000,000.000 
Bacillus pyocyaneus 
1,000,000,000 
Bacillus influenza 
1.000,000.000 
In preparing mixed vaccines the amounts of each micro-organism must 
be smaller than indicated above; as a general rule it suffices to add equal 
numbers of each so that the total will be approximately 1,000,000,000 per 
cubic centimeter. 
7. Diluting the Vaccine and Adding a Preservative.—Having made the 
count and sterilized the vaccine, it is next diluted with sterile saline solu- 
tion so that each cubic centimeter contains the dose decided upon. If 
the treatment is likely to be prolonged, a sufficient number of doses should 
be provided for. It is a good plan not to dilute all the vaccine, but to pre- 
serve the remainder undiluted in case larger doses are subsequently needed. 
If, for instance, a vaccine of Staphylococcus aureus contains 1,500,000,000 
organisms per cubic centimeter and the dose decided upon is 500,000,000 
per cubic centimeter, sufficient vaccine for 30 doses is prepared by with- 
drawing 10 c.c. of vaccine in a sterile container and adding 18.8 c.c. of 
sterile salt solution and 1.2 c.c. of a 5 per cent, solution of tricresol or phenol 
as a preservative. This mixture is agitated to insure thorough mixing, 
and 0.2 c.c. of 5 per cent, tricresol is added to each 5 c.c. of vaccine as a 
preservative against chance contamination. Thus, in the foregoing ex- 
ample, 1.2 c.c. of the diluted tricresol would be added. The amount of stock 
vaccine is estimated or measured, 0.5 per cent, phenol or tricresol is added, 
and the vaccine stored in a sterile container in the refrigerator, being first 
properly labeled with the patient’s name, the date, and the number of 
bacteria per cubic centimeter. 
The vaccine may now be placed in a sterile vaccine bottle, fitted with 
a sterile rubber cap, and properly labeled (Fig. 84). When it is to be ad- 
ministered, the cap is touched with tincture of iodin, the needle plunged 
through the cap, and a dose withdrawn with a sterile syringe. The punc- 
ture in the cap is then sealed with a drop of flexible collodion. This method 
is inexpensive, and with proper care is quite satisfactory, especially for 
stock vaccines. 
Another and probably better method, especially for autogenous vac- 
cines, consists in tubing each dose in separate sterile ampules (Fig. 85), 
which are then sealed in the flame. When the vaccine is to be administered, 
the ampule is well shaken, the neck broken in a towel, and the contents 
aspirated into a sterile syringe. These ampules may be purchased ready 
for use or be made in the laboratory, using 6 mm. soft glass tubing. For 
pipeting a vaccine into ampules the special automatic pipet shown in the 
illustration (Fig. 86) is quite convenient. As a rule, vaccines should be PREPARATION OF SENSITIZED BACTERIAL VACCINES 
199 
preserved in a cool place, such as a refrigerator. As shown by Hitchens 
and Hansen1 freezing has slight or no deleterious effects. 
Deterioration of Bacterial Vaccines.—As a general rule it is highly 
probable that freshly prepared vaccines possess a higher immunizing power 
than vaccines preserved for some 
time; this is especially true of per- 
tussis vaccines. 
The deterioration of vaccines 
bears a very important relation to 
temperature. For example, McCoy 
and Bengston,2 in a study of the 
deterioration of typhoid vaccine 
conducted over a period covering 
two and a half years, found that 
vaccines stored at 37° and 20° C. 
(room temperature) showed marked 
deterioration in six months. Vac- 
cines kept at 15° C. and lower did 
not show any appreciable reduction 
in agglutinin-producing properties 
for fifteen to eighteen months. After 
twenty-four months, however, a 
noticeable deterioration had taken 
place and this was still more evi- 
dent after storage for thirty months. 
Fig. 86.—Comer’s Automatic Pipet. (Steele 
Glass Co., Philadelphia.) 
The inner tube to the tip of the pipet holds 
exactly 1 c.c. If the rubber teat raises too much 
fluid, the excess is received in the glass reservoir; 
when too much fluid accumulates in this, it may 
be emptied by turning the point of the inner 
tube downward and ejecting the fluid by pressure 
on the teat. 
Fig. 85.—A Small Vaccine Ampule. 
Capacity 1 c.c. 
It would appear, therefore, that ordinary saline vaccines should be kept as far 
as possible at a temperature below 15° C. and used within six to twelve months. 
According to some investigations vaccines preserved in glycerol undergo 
much less deterioration. 
PREPARATION OF SENSITIZED BACTERIAL VACCINES 
A highly immune serum is prepared by immunizing a series of rabbits 
or a goat with the micro-organism to be used in preparing the vaccine. 
1 Amer. Jour. Pub. Health, 1913, 3. 
2 Hygienic Laboratory Bull., No. 122. PREPARATION OF BACTERIAL VACCINES 
The first injections consist of heat-killed emulsions, administered sub- 
cutaneously. After the first or second dose the period of heating is gradu- 
ally reduced, and the dose increased, until finally the injections may be 
given intravenously and with living micro-organisms. From time to time 
a small amount of serum should be examined for immune bodies: with the 
typhoid-cholera group, by testing for bacteriolysin and agglutinins; with 
staphylococci and streptococci, by agglutination, bacteriotropic, and com- 
plement-fixation tests; with pneumococci, gonococci, and meningococci, 
by bacteriolytic, agglutination, and bacteriotropic tests. When a highly 
immune serum is secured the animal is bled, the serum isolated, heated 
to 56° C. for half an hour, and stored in a strictly aseptic manner. 
To sensitize the bacteria, thick, even emulsions of young cultures in 
normal salt solution are treated with one-half to an equal bulk of inactivated 
immune serum, and the mixture gently agitated at room temperature for 
from six to twelve hours. The emulsion is then thoroughly centrifuged, 
and the residue of bacteria wrashed three times with sterile salt solution, 
after the manner in which the red corpuscles are washed. After the final 
washing the bacteria are resuspended in salt solution, shaken for a. time 
to insure breaking up of agglutinated clumps, counted, heated at 60° C. 
for an hour, cultured as a test for sterility, and then diluted so that the 
emulsion will gontain slightly larger doses than a corresponding dose of 
ordinary vaccine prepared for administration. 
Sensitization probably consists in the union of bacteriolytic ambo- 
ceptor with its antigen, and when injected serves, with the patient’s com- 
plement, to hasten solution or lysis of the bacteria (antigen), thereby liberat- 
ing quickly the chemical substances required for the stimulation of anti- 
bodies. 
Metchnikoff and Besredka are using sensitized living bacteria, and their 
work is being followed with much interest. In this country strict legal 
restrictions and regulations exist regarding the sending of living cultures 
through the mails. If, therefore, the method should fulfil the high claims 
and expectations made for it, there may be considerable difficulty in bringing 
it into general use. 
200 
PREPARATION OF LIPOVACCINES 
These vaccines contain the bacterium suspended in oil instead of water 
or saline solution. They were first mentioned by Warden1 and later advo- 
cated by LeMoignic and Pinoy,2 and Whitmore, Fennel and Petersen3 for 
prophylaxis against typhoid fever. Their chief advantage consists in the 
slow absorption of the oil permitting the administration of a single large 
dose of bacteria instead of the three doses of saline vaccine. Antibody 
production results from the administration of these lipovaccines, but, as 
shown by Bengston4 and others, not as promptly and scarcely to the same 
degree as follows the administration of equal numbers of bacteria in saline 
suspension. Whitmore has prepared these vaccines of typhoid and paraty- 
phoid bacilli, pneumococci, meningococci, and other bacteria for prophylactic 
immunization in the army; technical difficulties, however, and especially ster- 
ilization, together with rather severe reactions, have limited the usefulness 
of lipovaccines, so that they have been largely abandoned in favor of the 
saline vaccines. 
1 Jour. Amer. Med. Assoc., 1915, 65, 2080. 
2 Compt. rend. Soc. de biol., 1916, 79, 209. 
3 Jour. Amer. Med. Assoc., 1918, 70, 427; ibid., 902. 
4 Hygienic Laboratory Bull., No. 122. PREPARATION OF LIPOVACCINES 
201 
Whitmore and Fennel have prepared these vaccines as follows: 
1. The bacteria are grown on agar in Kolle flasks and removed with a 
vacuum scraper. Or the bacterium (as the pneumococcus) may be grown 
in dextrose broth and secured by centrifuging. 
2. The bacterial mass is poured in sterile Petri dishes, dried in an oven 
at 53° C. or frozen and dried in vacuo, a flaky mass resulting which crumbles 
into a fine powder ready for incorporation into a sterile oil. 
3. The bacterial mass is now weighed under sterile precuations and 
ground in a ball mill for at least forty-eight hours. A few cubic centimeters 
of chloroform-ether may be added to the mass at the beginning to insure 
final sterilization. 
4. Lanolin sterilized by autoclaving at 15 pounds for thirty minutes 
is warmed and sufficient added to make 10 per cent, in the completed vac- 
cine; this mixture is ground for another twenty-four hours, when sterile 
olive or cottonseed oils (autoclaved) are added and grinding continued 
for twenty-four hours. 
5. The final oil suspension is now heated at 53° C. for one hour on a 
water-bath and cultured for sterility. Lewis and Dodge1 advise heating 
the pneumococcus vaccine to 130° C. for three hours wdiich does not appear 
to injure the immunizing properties, although heating typhoid lipovaccine 
in this manner greatly injured their immunizing properties. 
Pneumococcus lipovaccine has been so prepared that each cubic centi- 
meter contained the equivalent of about 6,000,000,000 of each of the three 
fixed types (I, II, and III). Meningococcus vaccine has been prepared 
containing 10,000,000,000 of each type in each cubic centimeter. Dysentery 
vaccine has been prepared so that each cubic centimeter contained about 
2,000,000,000 of each of the three principal types (Shiga, Flexner, and 
Y types). Typhoid-paratyphoid vaccine usually contained about 2,500,- 
000,000 of each of the three strains of bacilli. These examples serve to 
show the approximate strength of the various lipovaccines. 
Rosenow and Osterberg2 have described a method for preparing lipo- 
vaccines in which the germs are suspended in water, sterilized with 2 per 
cent, cresol for twelve hours plus heating to 60° C. for one-half hour, fol- 
lowed by the addition of sterile lanolin and olive oil. The water is then 
removed by vacuum distillation at about 65° C. and the resulting suspen- 
sion in oil so diluted with sterile oil that 50,000,000,000 of bacteria are 
contained in 1 c.c. The authors have also given the following method for 
the preparation of small amounts of lipovaccines and particularly autog- 
enous vaccines: 
“The common 6-ounce nursing bottle has been found useful for the 
preparation of autogenous lipovaccines. It serves admirably as a culture 
flask, centrifuge tube, and vacuum flask. The bacteria are grown in tall 
columns of glucose broth (150 c.c. per bottle) for twenty-four hours, centrif- 
ugalized, the supernatant clear broth decanted, and the sediment suspended 
in 10 c.c. of a 1.5 per cent, solution of purified cresol in water or salt solu- 
tion. This is thoroughly mixed and placed at 36° C. for from two to fifteen 
hours, when cultures are made. Streptococci and pneumococci are usually 
killed in from two to twenty-four hours. As soon as the suspension is found 
to be sterile it is centrifugalized; the supernatant fluid is decanted, and 
6 c.c. of cottonseed oil containing 2 per cent, anhydrous lanolin and a number 
of sterile glass beads or steel shot are added. The mixture is emulsified by 
being shaken for a short time. The small amount of water from this water- 
1 Jour. Exper. Med., 1920, 31, 169. 
2 Jour. Amer. Med. Assoc., 1919, 73, 87. 202 
PREPARATION OF BACTERIAL VACCINES 
bacterial-oil suspension is now removed by applying the vacuum and 
immersing the bottom of the bottle in water heated to 60° C. By means 
of vigorous shaking at intervals the removal of the wrater is hastened. The 
vacuum and the heat are applied until bubbling ceases and the mixture 
becomes clear. The time required depends on the completeness of the 
vacuum and the amount of water to be removed, but the clearing usually 
takes place in from twenty minutes to one hour. If larger amounts of 
bacteria are required the water or salt solution suspensions of a number 
of bottles are placed in one, and a correspondingly larger amount of oil is 
added. By the use of Y-tubes the water from a series of suspensions may 
be removed at one time. 
“If, for example, bacteria such as influenza bacilli, gonococci, and meningo- 
cocci, which grow better on solid mediums, are to be used, the growth 
should be scraped together and washed off with salt solution so that the 
final suspension is roughly equivalent to that containing the bacteria from 
the broth culture. Sterilization and the further steps are carried out as 
above. In the case of the more resistant bacteria, such as staphylococci 
and paratyphoid bacilli, heating the cresolized suspension to 60° C. for 
one hour hastens the sterilization. The final bacterial content of the lipo- 
vaccine is calculated on the basis of counts made of the bacteria suspended 
in salt solution or on the basis of the total number of bacteria per cubic 
centimeter of broth culture. In the broth used in our laboratories the 
amount of growth of pneumococci and streptococci is usually about 2,000,- 
000,000, and the vaccine is made to contain approximately 50,000,000,000 
of these organisms per cubic centimeter. The number of bacteria in the 
oil may be increased ten or twenty fold without interfering materially with 
the evaporation of the water or with the even distribution of the bacteria.” CHAPTER XIII 
ANTITOXINS 
For general purposes the antibodies produced during infection may be 
divided into two groups, the first consisting of those antibodies that are 
truly antagonistic to the bacterium or its products responsible for their 
production, and the second those that are not in themselves destructive, 
but that probably prepare the bacterium for the action of a more powerful 
antibody of the first group. 
To the first group belong the antitoxins, which neutralize the toxins of 
a bacterium without being directly destructive to the micro-organism itself; 
and the bacteriolysins, which are truly destructive, causing the bacterium 
to break up and finally disappear. 
To the second group belong the opsonins, which, as we have seen, pre- 
pare the bacterium for phagocytosis; and the agglutinins and precipitins, 
which, while not in themselves destructive, probably in some manner pre- 
pare their antigen for the action of bacteriolysins, just as opsonins prepare 
them for phagocytosis. 
Definition.—Antitoxins are antibodies in the blood that are capable of 
directly and specifically neutralizing the dissolved toxins that caused their 
production. 
Historic.—Bacteriolysins were discovered before antitoxins. Their dis- 
covery is due to the researches of Nuttall, Fodor, Buchner, and others, 
who showed that normal serum, and especially the serum of animals arti- 
ficially immunized against a certain bacterium, was able to exert a destruc- 
tive action on the micro-organism, causing its dissolution and final dis- 
appearance. This property of the blood-serum was found to diminish with 
age, and to disappear completely when the serum was heated to 56° C. 
Buchner laid greatest stress upon the importance of the thermolabile sub- 
stance which he called alexin, but later researches have shown that the 
main factors are the specific bacteriolysins, which, however, are practically 
powerless to destroy their antigen without the co-operation of alexin (later 
renamed “complement” by Ehrlich). 
While these studies were being made, in the hope of thus explaining all 
phases of immunity, Behring discovered that in diphtheria infections in- 
duced experimentally, while the animals became more and more immune, 
virulent bacilli may, nevertheless, be present at the site of injection. Here, 
then, was an example of immunity that could not be explained on the basis 
of bacteriolysis. Later, in 1890 and 1892, Behring,1 in collaboration with 
Kitasato and Wernicke, made further important discoveries, showing that 
the blood-serum of animals actively immunized against diphtheria and 
tetanus would protect normal animals against these diseases, and, further- 
more, that the blood-serum of the immune animals did not possess bac- 
tericidal properties. These observers also demonstrated that such serum 
could be used therapeutically for the cure of an infection already in progress. 
Soon after these discoveries Ehrlich2 showed that specific antitoxins 
(antiricin, antiabrin, etc.) could also be produced against the poisons of 
some plants, and Thisalix and Bertrand,3 and Calmette4 produced a similar 
1 Deutsch. med. Wchn., 1890, 1145, 1245. 
2 Ibid., 1891, 976, 1218. 
3 Compt. rend. Soc. de biol., 1894, 111. 
4 Ibid., 1894, 120, 204. 
203 ANTITOXINS 
204 
antitoxin (antivenin) against snake poison. Other observers since then have 
increased the list of poisons against which antitoxins can be produced; as, 
for example, Kempner has produced an antitoxin against the poison of 
Bacillus botulinus, and Wassermann one against that of B. pyocyaneus. 
Formation of Antitoxins.—It was formerly believed that there was a 
direct conversion of toxin into antitoxin, but this certainly is not the case, 
for the amount of antitoxin produced is altogether out of proportion to the 
amount of toxin injected. 
Antitoxins are formed by those cells that anchor the toxins. In order 
to produce them it is necessary that the toxin enter into direct union with 
the cells and exert a stimulating influence on them, for where a loose union 
occurs, as between cells and alkaloids, antibodies are not formed. 
Fig. 87.—Theoretic Formation of Antitoxins. 
The central white area represents a molecule of a cell; the shaded portion represents the cell itself; 
the surrounding area represents the body fluids about the cell. 
r, A receptor of the molecule {first order); A, overproduction of receptors, which are being cast 
off; A2, a cast-off receptor free in the body fluids—now an antitoxin; A3, a molecule of antitoxin combi- 
nation with a toxic molecule T3; A3, a cast-off receptor still within the parent cell; T, a toxin mole- 
cule in combination with the receptor of a cell molecule; T2, a toxin molecule free in the body fluids; 
T3, a toxin molecule in combination with antitoxin; T\ a molecule of toxoid (toxophore group lost). 
Having entered into chemical union with the side arms of cells, a toxin 
may destroy the entire cell, and if a sufficient number of these are destroyed, 
the host will show symptoms of infection and may succumb. If, however, 
the cell itself is not destroyed, but only one or more of the side arms in- 
jured, the damage is repaired by the cell forming new side arms that have 
a specific affinity for the toxin responsible for their production. According 
to Weigert’s overproduction theory, a cell once stimulated to produce these 
side arms or receptors continues to produce them for some time, even after 
the stimulus has been removed. In this manner the specific receptors are 
produced in excess, and since all cannot remain attached to the parent cell, 
the excess is discharged into the blood-stream. Each of these cast-off re- 
ceptors is capable of uniting with toxin, thus neutralizing the poisonous PROPERTIES OF ANTITOXINS 
205 
principles of the toxin and rendering it practically harmless. Antitoxins, 
therefore, are nothing more than these cast-off receptors, which have a specific 
affinity for their toxins (Fig. 87). 
As Adami has pointed out, it is probable that the toxins exist for 3ome 
time within the cell, not as part and parcel of the cell, but as a stimulating 
agent that causes the cell to develop the habit of producing the specific 
receptors. The mere union of toxin with a receptor, causing it to fall off, 
and being followed by nature’s mode of repair, with the formation of an 
excess of receptors and no further stimulation, is hardly sufficient to explain 
the enormous activity of the cells. 
That antitoxins may be produced locally was illustrated by the experi- 
ment of Romer with abrin. This substance has a peculiarly powerful effect 
upon the conjunctiva. By gradually immunizing the right conjunctiva of 
a rabbit with increasing doses, it was shown that, after killing the animal 
and triturating the conjunctiva with a fatal dose of abrin, an injection of 
the emulsion of the right or immunized conjunctiva was without effect, 
whereas the emulsion from the left proved fatal. Thus it will clearly be 
seen that the cells that had absorbed the abrin had developed and con- 
tained antiabrin in sufficient amounts to neutralize the poison. 
A single attack of diphtheria does not cause the production of large 
amounts of antitoxin. For this, repeated, possibly minute infections are 
more effectual, wdiich pass over with minimal or misinterpreted symptoms. 
Otto1 found that diphtheria carriers, both those who had had the disease 
and those who had not, contained more antitoxin in their blood than patients 
who had just recovered from an attack. This indicates that the mere presence 
of bacilli in the throat is sufficient to stimulate the production of antitoxin 
on which the immunity of the carrier himself would seem to depend. 
While leukocytes, such as Metchnikoff’s macrophages, are likewise active 
in the formation of antitoxins, it is certain that they are not the only cells 
involved. Metchnikoff claims that antitoxins are merely toxins altered by 
leukocytic activity, rather than constituents of tissue cells; this explana- 
tion is, however, inadequate, and it has been shown experimentally that 
the quantity of antitoxin produced is so far in excess of the amount of toxin 
injected as to render this view untenable. 
Structure of Antitoxins.—According to the side-chain theory, antitoxins 
are the simplest of antibodies, being composed of a single arm or hapto- 
phore group for union with the toxin, and called receptors of the first order. 
While illustrations of this theoretic structure will convey the impression 
of mere physical contact or union with toxin, it is to be remembered that 
experimental data indicate that the union and consequent neutralization 
of the toxin are chemical processes. 
Properties of Antitoxins.—While chemical analyses to determine the 
nature of antitoxin serums were made as early as 1897, little is known re- 
garding it because it is impossible to secure the antitoxic element free from 
serum and serum constituents. Brodie2 was among the first to show that 
diphtheria antitoxin was completely precipitated from a solution by any 
means which removed the globulins, and his observations wrere extended 
by Belfanti and Carbone,3 and by Seng.4 But the first notable advance 
in this study was made by Hiss and Atkinson,5 who estimated the globulin 
1Deutsch. med. Wchn., 1914, March 12, 542. 
2 Jour. Path, and Bacteriol., 1896, 4, 460. 
3 Abstr. Centralbl. f. Bakteriol., 1, 1898, 23, 906. 
4 Ztschr. f. Hyg. u. Infektionskrankh., 1899, 31, 513« 
5 Jour. Exper. Med., 1900, 5, 47. 206 
ANTITOXINS 
content of the serum of a large number of horses at different stages of im- 
munization against diphtheria toxin. As a result of these experiments 
these workers arrived at the conservative conclusion that a low potency 
coincided with a low globulin content, but that it was not possible to regard 
the absolute amount of globulin as an index of the antitoxin content of the 
serum. Similar studies were made by Ledingham1 and by Gibson and by 
Banzhaf.2 The former concludes from the data furnished by the immuni- 
zation of a horse and a goat with diphtheria toxin that there would seem to 
be “some intimate relation between the amount of antitoxin developed and 
the quantity of the globulins.” Gibson and Banshaf are still more guarded 
in their conclusions. These workers state that while the greatest rise in 
the serum globulin was usually coincident with maximum antitoxic potency, 
the extent of this increase was practically independent of the antitoxin 
potency when the resulis on more than one horse were contrasted. They, 
therefore, are inclined to the view that there may be no relation between 
the absolute or percentage increase of the serum globulin and the antitoxic 
potency in the plasma of different horses; indeed, they have shown that the 
increase in the serum globulin of refractory horses may surpass that in the 
plasma of some of those yielding a high antitoxin. Similar deductions were 
made by Banzhaf and Famulener3 in the immunization of goats. They 
found that the total protein content and the protein partition may be normal 
at a time when the animal shows the maximum number of antitoxic units. 
Meyer, Hurwitz, and Taussig4,5 subscribe to the view that the globulin 
change is not a necessary concomitant of the elaboration of antibodies, 
although it is now well established that this protein fraction may increase 
strikingly during the process of immunization. Their conclusions have 
been derived from parallel studies of the serum protein fractions and of 
the antitoxin content of a number of different animals immunized not 
only with diphtheria toxin but also with the soluble toxins of the bacilli 
of tetanus and of botulism. In general, the results obtained in the different 
animals and with the different soluble toxins point in the same direction 
and lend support to the view that in animals immunized with bacteria and 
their toxins the curves of serum globulin increase and the development of 
antitoxin potency do not run parallel. Gibson and Banzhaf showed that 
the portions of the globulin precipitate soluble in saturated sodium chlorid 
solution carried most of the antitoxin, and with this discovery a practical 
method of eliminating much of the non-antitoxic portion of the serum wras 
perfected. 
The relation of antitoxins to proteins has also been studied, digestive 
ferments being permitted to act on antitoxic serum. It has been shown 
that antitoxin resists tryptic digestion to a well-marked degree; in this 
respect it resembles the serum globulin. All the evidence obtained indi- 
cates that a closer relation of antitoxins to proteins exists than has been 
shown for the toxins, although all attempts to separate antitoxins from 
proteins have thus far failed. 
At the Bureau of Animal Industry in Washington, Berg and Kelser® 
have attempted anew to secure a protein-free antitoxin preparation on the 
one hand, and, on the other, to determine whether antitoxin can be de- 
stroyed by procedures that leave the protein intact. The outcome of these 
1 Jour. Hyg., 1907, 7, 65. 
2 Jour. Exper. Med., 1910, 12, 411. 
3 Coll. Stud. Research Lab., Dept. Health, New York, 1915, 8, 208. 
4 Jour. Exper. Med., 1916, 24, 515. 
5 Jour. Infect. Dis., 1918, 22, 1. 
6 Proc. Nat. Acad. Sci., 1918, 4, 174. NATURAL ANTITOXINS 
207 
experiments was to show that antitoxin destruction may take place with 
or without cleavage of protein and the authors suspected that tetanus anti- 
toxin, for example, may be a substance of non-protein nature. The evidence, 
however, is not conclusive so long as a protein-free antitoxin is not obtained. 
Of great interest in this connection are the investigations of Huntoon, 
Masucci and Hannum,1 who have succeeded in preparing solutions carry- 
ing antipneumococcus antibodies which give but feeble chemical reactions 
for the proteins and apparently fail to sensitize guinea-pigs. 
Antitoxins are fairly resistant bodies, and a properly prepared anti- 
toxic serum, when kept in a cool place and protected from light and air, 
may be preserved for a year or more with very little deterioration in strength. 
At times, however, for unknown reasons antitoxins gradually deteriorate, 
losing about 2 per cent, in strength a month. Anderson2 has calculated 
the yearly loss in potency at about 20 per cent., although occasionally it 
may go as high as 25 per cent, when the serum is kept at room temperature. 
At 15° C. the yearly loss was about 10 per cent, and at 5° C. about 6 per 
cent. Old sera, unit for unit, were found just as potent as fresh sera. But 
little difference was noted in the keeping qualities of whole serum and solu- 
tions of the globulins. Manufacturers have endeavored to calculate this 
loss in strength, and have placed a label on each package of antitoxin, 
bearing a date beyond which the serum is not guaranteed to contain the 
amount of antitoxin present at the time it was put up. 
The antitoxins, with few exceptions, are far more stable than the toxins, 
resisting heating up to 62° C., but gradually deteriorating with higher 
temperatures. Boiling destroys them completely. They are readily pre- 
served with small amounts of chloroform, phenol, tricresol, etc., although 
strong solutions of these produce destructive changes. Putrefaction of the 
serum destroys the antitoxin content. Ehrlich has devised the best method 
for their preservation, which consists in drying the serum in vacuo and pre- 
serving it in the dark, at a low temperature, in the presence of anhydrous 
phosphoric acid. So preserved, antitoxin retains its strength for prolonged 
periods and is used in standardizing toxins. 
Natural Antitoxins.—The appearance of so-called natural antitoxins can 
be explained on the basis of Ehrlich’s theory. Since the antitoxin is com- 
posed of receptors that are not new bodies, but simply normal receptors 
produced in excess, it is reasonable to assume that a few may be thrown 
off occasionally, constituting the natural antitoxin. 
Small amounts of natural diphtheria antitoxin may be found in certain 
individuals. Since the diphtheria bacillus is so wide-spread in its distribu- 
tion, it is possible that minor subinfections may be responsible for anti- 
toxin production, and this is probably always the case when large amounts 
are found. 
There are numerous examples of natural antitoxin immunity among the 
lower animals, the most notable being the resistance of the rat to diphtheria 
toxin and the chicken to tetanus toxin. The investigations of Burrows and 
Suzuki3 and Suzuki,4 employing cultures of tissues, have indicated that this 
resistance is due to antitoxins in the plasma and to special resistance of 
certain cells of these animals. 
Information regarding natural antitoxins for other members of the toxin- 
producing group of micro-organisms is less complete, although it is highly 
probable that natural antitoxins for these exist. 
1 Jour. Immunology, 1921, 6, 185. 
‘Jour. Infect. Dis., 1910, 7, 481. 
3 Jour. Immunology, 1918, 3, 219. 
4 Jour. Immunology, 1918, 3, 233. 208 
ANTITOXINS 
The Schick Test for Natural Diphtheria Antitoxin.—Schick1 has worked 
out a simple and practical skin test which apparently has proved satisfactory 
and trustworthy and of distinct value for detecting natural immunity to 
diphtheria among human beings. 
This test consists in the intradermic injection of a minute dose of diph- 
theria toxin. If the individual possesses an amount of antitoxin equal to at 
least one-thirtieth of a unit in each cubic centimeter of blood-serum the 
injected toxin is neutralized and no reaction follows; if, however, the indi- 
vidual does not have antitoxin in the body fluids the injected toxin acts as 
an irritant to the skin, producing in twenty-four to forty-eight hours a small 
area of redness and edema. A positive reaction indicates that the individual 
does not possess natural antitoxin in his blood and, therefore, that he is 
susceptible to diphtheria; a negative reaction indicates that natural anti- 
toxin is present and that he is, in all probability, immune to diphtheria. 
In the presence of exposure to diphtheria persons reacting positively to 
the toxin skin test should receive a prophylactic dose of antitoxin, while 
those reacting negatively may wdth safety be spared the injection. A de- 
tailed description of this test is given in a later chapter. 
Specificity of Antitoxins.—Antitoxins well illustrate the law of specificity 
that exists between antigen and antibody, since they are strictly specific 
for their toxins. Diphtheria antitoxin will neutralize only diphtheria toxin; 
tetanus antitoxin, only tetanus toxin, and so on through the list. This 
specificity is not confined to the particular toxin-producing organism that 
generates the antitoxin; for example, there are various kinds of diphtheria 
bacilli, differing as regards morphology and toxicity, although one anti- 
toxin appears to act the same with their various toxins. 
Types of Toxic-producing Bacilli; Monovalent vs. Polyvalent Antitoxic 
Sera.—Of considerable interest and practical importance is the question 
whether different types of diphtheria, tetanus, and other toxin-producing 
bacilli produce toxins corresponding to type and whether ‘these are neu- 
tralizable by monovalent antitoxins for each of the respective species of 
bacilli. 
Several types of tetanus bacilli are recognized according to their re- 
sponse to agglutinins, but the toxins of all are neutralized by a monovalent 
antitoxin. Durand2 and Havens3 have found different types of diphtheria 
bacilli on the basis of agglutination tests, but the toxins of all of these are 
neutralized by a monovalent (No. 8) antitoxin according to the studies of 
Park, Williams, and Mann.4 Practical experience with tetanus and diph- 
theria antitoxins in the prophylaxis and treatment of tetanus and diph- 
theria has abundantly proved the efficacy of the monovalent antitoxins 
for these two diseases as discussed in Chapter XL. However, Bacillus 
botulinus shows at least two different types producing toxins which respond 
only to homologous antitoxins. 
Nature of the Toxin-antitoxin Reaction.—While the injection of toxin- 
antitoxin mixtures into the lower animals is the only practical method of 
testing and standardizing the curative and prophylactic powers of their 
serums, this method does not throw much light upon the nature of the 
toxin-antitoxin reaction, or show how antitoxin overcomes the toxin. 
Antitoxin is protective and curative, in that it actually destroys the 
toxin in a manner similar to the dissolution of a bacterium caused by a 
1 Miinch. med. Woch., 1913, lx, 2608. 
2 Jour. Infect. Dis., 1920, 26, 388. 
3 Compt. rend. Soc. Biol., 1918, 81, 1011 and 1920; ibid., 1919, 83, 611 and 613. 
4 Jour. Immunology, 1922, 7, 243. NATURE OF THE TOXIN-ANTITOXIN REACTION 
209 
specific bacteriolysin; or it may influence the tissue cells in some way and 
render them more resistant to the toxins, a view that was held by Roux, 
and particularly by Buchner; or the antitoxin may form a direct chemical 
union with the toxin, similar to the chemical neutralization of an acid by 
a base—an opinion early held by Behring and elaborated later by Ehrlich. 
Experimental data support the view of chemical union with the toxin. 
In the test-tube some time is required for the union of toxin and antitoxin 
to occur; this union is hastened by heat and retarded by cold; it is more 
rapid in concentrated than in dilute solutions, and in general takes place 
in accordance with the law of multiple proportions—all of which tends 
to show the close similarity of the toxin-antitoxin reaction to a chemical 
process. 
It is generally conceded that antitoxin does not directly destroy the 
toxin, for when neutral mixtures of toxin and antitoxin are injected into 
animals, portions of toxin may become dissociated and unite with tissue 
cells possessing greater affinity for the toxin, and symptoms of infection 
may result. It is probable that toxin and antitoxin form a distinct com- 
pound, and this action requires time for its consummation. For example, 
Martin and Cherry, by filtering mixtures of toxin and antitoxin through 
fine filters that wrould permit the toxin molecule to pass through but re- 
strain the larger antitoxin molecule, found that, if filtered immediately, 
all the toxin in the mixtures was extruded, but that, as the interval be- 
tween mixing and filtration was prolonged, less and less toxin appeared in 
the filtrate, until finally, twTo hours after mixing, no toxin whatever passed 
through the filter. 
This element of time in support of the chemical nature of the reaction 
is further strengthened by the experiments of Calmette with snake venom 
and antivenin, and likewise serves to demonstrate that the antitoxin ap- 
parently does not directly destroy the toxin. Although most toxins are 
thermolabile, Calmette found that snake venom is rendered inert by heating 
to 68° C., whereas the antivenin remains uninfluenced by a temperature of 
80° C. When neutral mixtures of venom-antivenin were heated to 70° C., 
they were found to become toxic again, presumably on account of the de- 
struction of the antivenin, the venom itself not being destroyed. Similar 
experiments were carried out by Wassermann with mixtures of pyocyaneus 
toxin-antitoxin, with similar results. In both instances, however, as devel- 
oped later, if the mixtures had been allowed to stand longer, these results 
would not have been secured. Although performed originally to show that 
an antitoxin does not act by actually destroying its toxin, these experi- 
ments simply demonstrate the importance of the element of time in the 
reaction, without throwing any real light upon the nature of the new toxin- 
antitoxin compound, if such exists. 
That toxin is counteracted by antitoxin, independent of the participa- 
tion of living tissue cells, has been quite conclusively proved by experi- 
ments in vitro. Ehrlich showed that the agglutinating qualities of ricin— 
a vegetable toxin—may be overcome in the test-tube by adding antiricin, 
the corresponding antitoxin. Similar results were obtained by Ehrlich with 
tetanolysin and tetanus antitoxin, and by Stephens and Myers with cobra 
venom and its antivenin. 
It is probable that antitoxin has a similar action when injected for thera- 
peutic purposes, as for curing an infection. The longer the interval that 
has elapsed between the time of infection and the administration of anti- 
toxin the less satisfactory will be the result, as antitoxin becomes less 
powerful when toxins have formed a firm union with the body cells. This ANTITOXINS 
210 
is especially true in tetanus, where even very large doses of antitoxin may 
be incapable of dissociating the toxin molecule from the nerve cells, the 
serum, therefore, being of greatest value in prophylaxis. In diphtheria, 
however, the union between toxin and cells is less firm, and the antitoxin 
is probably capable of neutralizing the toxin already present in the cells, 
and especially any toxin that may become dissociated from the cell or is 
freshly prepared by the diphtheria bacillus at the site of infection. The 
indication, therefore, in giving antitoxin, is to give a dose large enough to 
neutralize all free and loosely bound toxin, with an excess to neutralize 
dissociated toxin and that prepared by the bacillus during the course of 
the infection. 
The introduction of the test-tube experiment into the investigation of 
these reactions permitted more exact observations to be made, and the evi- 
dence secured by this means, as well as by carefully graded quantitative 
animal experiments, would seem to indicate that we should accept, for the 
present at least, the conception of the chemical nature of the process. 
PRODUCTION OF ANTITOXINS FOR THERAPEUTIC PURPOSES 
Diphtheria and tetanus antitoxins are manufactured on a large scale 
and are used extensively in the prevention and cure of these infections. 
They are prepared by immunizing horses with carefully graded and in- 
creasing doses of the respective toxins until the serum of the animals shows 
a sufficiently high antitoxin content, after preliminary trials, to warrant 
more extensive bleeding. Large quantities of blood are then collected 
aseptically by puncturing the jugular vein. The serum is carefully separated 
and standardized according to an accepted technic in order to determine 
the antitoxin content in units. A small amount of preservative is added, 
and the serum is finally dispensed in special containers or syringes ready 
for administration. In some laboratories it is customary to precipitate 
the globulin fraction of the serum with magnesium or ammonium sulphate, 
and redissolve the portion containing most of the antitoxin in saturated 
sodium chlorid solution. The bulk of the serum is thus greatly decreased 
and objectionable constituents largely eliminated, to the obvious advantage 
of the preparation for therapeutic purposes. 
Antitoxins have also been prepared for other bacterial toxins, as those 
of the dysentery bacillus (Kruse-Shiga) and Bacillus botulinus, for the 
vegetable toxins in pollen and for the animal toxins in snake venoms. 
There are other serums for the treatment of certain infections, which 
depend for their effects chiefly upon the presence of bacteriolvsins and 
immune opsonins, and these are described in a subsequent chapter. 
The Production of Diphtheria Antitoxin 
The following, taken largely from Park,1 is a widely used and accepted 
technic for the production of diphtheria antitoxin: 
Production of the Diphtheria Toxin.—A strong diphtheria toxin should 
be obtained by growing a virulent culture in a 2 per cent, nutrient peptone 
bouillon made from “bob” veal, of an alkalinity that should be about 9 c.c. 
of normal soda solution per liter above the neutral point to litmus, and 
prepared from a suitable peptone (Witte). The broth should be poured 
into large-necked Erlenmeyer flasks in comparatively shallow layers so as 
to allow of the free access of air, and maintained at a temperature of about 
35° to 36° C. (Fig. 88). 
1 Park and Williams, Pathogenic Micro-organisms, Lea & Fehiger, 1920, 181. THE PRODUCTION OF DIPHTHERIA ANTITOXIN 
211 
In the Hygienic Laboratory of the Public Health and Marine Hospital 
Service “Smith’s bouillon” is used for preparing the toxin. This is made 
of fresh lean beef, after the muscle sugar and all other sugars have been 
removed by fermentation with a good culture of Bacillus coli. The reaction 
is adjusted until 0.5 per cent, acid to phenolphthalein, that is, still distinctly 
alkaline to litmus, and 1 per cent, peptone, 0.5 per cent, sodium chlorid, 
and 0.1 per cent, dextrose are added. The reaction is again noted and 
adjusted to + 0.5 per cent. The broth is then filtered through filter-paper 
into flasks and test-tubes and sterilized in the autoclave at a temperature 
of 120° C. for twenty minutes. 
After incubating for from seven to ten days the culture is removed, and 
its purity having been tested by microscopic and cultural methods, it is 
rendered sterile by the addition of 10 per cent, of a 5 per cent, solution 
of carbolic acid. After forty-eight hours the dead bacilli have settled on 
Fig. 88.—A Flask of Diphtheria Culture. 
The bacilli grow on the surface and form a scum. As the culture grows older the bacilli die and 
sink to the bottom of the flask. A flask of this shape affords a large surface of culture-medium in con- 
tact with oxygen and facilitates toxin production. 
the bottom of the jar, and the clear fluid above is siphoned off, filtered, 
and stored in full bottles in a cold place until needed (Fig. 89). 
The relation of reaction (alkalinity) of the broth has been studied by 
Hitchens1 and many others. Davis2 has recently advocated the use of a 
bouillon with an initial reaction falling within a certain zone of alkalinity 
varying in hydrogen-ion concentration from 7.0 X 10~8 to about 5.0 X 10—9. 
He has found that diphtheria bacilli cultivated in plain broth produces 
an initial increase in hydrogen-ion concentration, followed by a steady de- 
crease until a limited alkalinity is obtained. No direct relationship was 
found between hvdrogen-ion concentration of the medium and toxicity. 
Davis3 has cautioned against the presence of any fat in the broth, as it inter- 
feres with pellicle formation, and found beef infusion broth with 2 per cent, 
peptone and 0.5 per cent, salt to be most satisfactory (pH = 8.2). Large 
1 Jour. Med. Research, 1904, 13, 523. 
2 Jour. Lab. and Clin. Med., 1917, 3, 358. 
3 Jour. Bacteriology, 1920, 5, 477. 212 
ANTITOXINS 
flasks of this broth inoculated with twenty-four-hour cultures in “starter 
flasks” cultivated at 36° to 38° C. for ten to twelve days was found to give 
the largest yields of toxin. 
Fig. 89.—A Large Toxin Filter. 
The culture is contained in the large bottle on the shelf and drains into the flask, which, in turn, 
empties into the earthen “candle.” By means of a vacuum the culture is filtered through the “candle” 
and collects in the large bottle at the base of the stand. 
As shown by Robinson and Rettger1 diphtheria toxin cannot be pro- 
duced unless complex nitrogenous bodies are present, as some of the pro- 
teoses; in protein-free media there is very little toxin production. During 
1 Jour. Med. Research, 1917, 36, 357. THE PRODUCTION OF DIPHTHERIA ANTITOXIN 
213 
recent years considerable trouble was occasioned by lack of sufficient quan- 
tities of Witte’s peptone, but American brands of peptone are now yielding 
satisfactory results. 
Testing the Toxin.—The strength of the toxin is then tested by inject- 
ing a series of guinea-pigs with carefully measured amounts. When in- 
jected hypodermically, less than 0.005 c.c. should kill a 250-gram guinea- 
pig, and a toxin requiring more than 0.01 c.c. to kill a pig of this weight 
is too weak for present purposes. This preliminary titration of the toxin 
will suffice for determining the dosage for horses, but in standardizing anti- 
toxin the technic must necessarily be more accurate. 
Immunizing the Animals.—The horses used should be young, vigorous, 
of fair size, and absolutely healthy. They should be severally injected 
with 3000 units of antitoxin and the following day with 5000 units of toxin, 
an amount sufficient to kill 5000 guinea-pigs, each weighing 250 grams, 
or 10 M. L. D. of toxin diluted to 50 c.c. with saline solution. If antitoxin 
is not given with the first doses of toxin, only one-tenth of the dose advised 
is to be given. After from two to four days, or as soon as the temperature 
reaction has subsided, a second subcutaneous injection of a larger dose is 
given, the amount of toxin increasing about 100 per cent, per dose for the 
first seven to eight injections. The rate of increase is 75 per cent, for the 
next seven to eight injections, and 50 per cent, for the next series. The 
rate of increase is then gradually lowered to 10 per cent., which is main- 
tained until the maximum for the horse is reached. At the end of this time 
a trial bleeding is made and the serum tested. 
There is absolutely no way of judging which horses will produce the 
highest grades of antitoxin. Roughly estimated, those horses that are 
extremely sensitive and those that react feebly are the poorest, but there 
are exceptions even in these cases. The only reliable method, therefore, 
is to bleed the horses at the end of six weeks or two months and test their 
serum. As shown by Park and Zingher, persons yielding negative Schick 
tests respond to toxin-antitoxin injections with the production of anti- 
toxin more readily than persons who react in a positive manner. Taking 
advantage of this data, Hitchens and Tingley1 have injected 0.2 c.c. diph- 
theria toxin equal to 3 M. L. D. for 250-gram pigs into the conjunctiva of 
one eye of horses under examination for the purposes of immunization, 
and found that many yielded negative tests read at the end of forty-eight 
hours, indicating the presence of natural diphtheria antitoxin in the blood 
of normal horses; these animals are probably to be preferred in the produc- 
tion of antitoxin, as they are likely to yield highly potent sera. If only 
high-grade serum is wanted, all horses that give less than 150 units per cubic 
centimeter should be discarded. The remaining horses should receive 
steadily increasing doses, the rapidity of the increase and the interval of 
time between the doses (three days to one week) depending somewhat on 
the reaction following the injection, an elevation of temperature of more 
than 3° F. being undesirable. 
For example, according to Park, a horse that yielded an unusually high 
grade of serum was started on 12 c.c. of toxin c.c. fatal dose), together 
with 10,000 units of antitoxin. Sixty days later a dose of 675 c.c. was given, 
and the serum contained 1000 units of antitoxin per cubic centimeter. 
Regular bleedings were made weekly for the next four months, at the end 
of which time the serum had fallen to 500 units in spite of weekly gradually 
increasing doses of toxin. At the end of three months the antitoxic serum 
of all the horses should contain over 300 units, and in about 10 per cent. 
1 Jour. Amer. Med. Assoc., 1917, 68, 1660. 214 
ANTITOXINS 
as much as 800 units in each cubic centimeter. Not more than 1 per cent, 
give above 1000 units and, according to Park, so far none has given him 
as much as 2000 units per cubic centimeter. The very best horses, if pushed 
to their limit, continue to furnish blood containing the maximum amount 
of antitoxin for several months and then, in spite of increasing injections 
of toxin, begin to furnish blood of gradually decreasing strength. If an 
interval of three months’ freedom from inoculation is allowed once every 
nine months the best horses will furnish high-grade serum for from two 
to four years. 
Kraus and Sordelli1 have recently described a new method for the prepara- 
tion of diphtheria and tetanus antitoxins and antivenins, consisting in the 
use of old horses, and making the injections of toxin neutralized with anti- 
toxin tw'ce a week in progressive doses. These investigators claim to have 
prepared potent sera in twenty weeks by this method. 
Collecting the Serum.—In order to obtain the serum the neck of the 
horse should be cleansed thoroughly as for an aseptic operation, and a special 
tourniquet applied to distend the jugular vein. A small slit is made through 
the skin over the vein and a special sharp-pointed cannula is passed up- 
ward under the skin for 2 inches or more, and then plunged into the vein. 
From 6 to 12 liters of blood are collected by a rubber tube into cylindric 
jars provided with special tops, facilitating filling with blood and subse- 
quent withdrawal of the serum. The cannula, tubing, jars, and everything 
used in collecting the blood and serum should be carefully sterilized, and 
the whole operation should be conducted with scrupulous aseptic care in 
order to avoid contamination (see Fig. 39). 
The jars are set aside (Fig. 90) for three or four days, and the serum is 
drawn off by means of sterile glass and rubber tubing and stored in large 
sterile bottles. When the globulins are to be separated the blood may be 
added directly to one-tenth of its volume of a 10 per cent, solution of sodium 
citrate, which prevents clotting of the blood. Penfold has recently des- 
cribed a method of bleeding directly into oxalate solution, and after sedi- 
mentation of the corpuscles has occurred the supernatant plasma is drawn 
off and the corpuscles reinjected intravenously into the horse. By this 
method Penfold states anemia and other bad effects from too frequent 
bleedings may be avoided, and that as much as 60 liters of high potency 
blood may be taken from a horse in ten days. 
The serum should be clear and free from blood and its sterility should 
be proved by culture tests. An antiseptic, such as 0.4 per cent, tricresol, 
0.5 per cent, phenol, or chloroform, may be added, but this is not neces- 
sary unless it is desired to keep the serum for some time. The serum is 
poured into small bottles fitted with rubber stoppers, or placed in special 
syringes labeled with the number of units contained. The whole process 
should be conducted with scrupulous aseptic technic. Diphtheria toxin 
varies too much to be used as a standard in determining the antitoxin con- 
tent of a serum; hence, a dried antitoxin is prepared by the Hygienic Labora- 
tory and is distributed for this purpose. The serum is evaporated and 
dried in vacuo by passing dry sterile air heated to 35° C. through it and, 
when perfectly dry, is preserved in special containers over anhydrous phos- 
phoric acid at a constant temperature of 5° C. Preserved in this manner, 
the antitoxin is quite stable. Just before use it is dissolved in the required 
amount of sterile normal salt solution. 
Method of Concentrating Serum by Isolating the Antitoxin Globulins.— 
The use of concentrated serum has lessened the incidence of serum sickness 
1 Prensa Medica, 1918, 4, 307. Tig. 90.—Preparation of Diphtheria Antitoxin. Separation of Blood-serum. 
The bottle on the left shows blood after standing about an hour; the bottle on the right shows the 
separation of serum about twelve hours after bleeding. THE PRODUCTION OF DIPHTHERIA ANTITOXIN 
and facilitates the administration of large doses. The first practical method 
for the concentration and refinement of diphtheria antitoxin was devised 
by Gibson.1 Banzhaf2 later developed a somewhat different method which 
was adopted by the American Public Health Association in 1915 with but 
one modification, consisting of holding the heated plasma-ammonium sul- 
phate mixture at 60° C. for fifteen minutes before filtering. Briefly, the 
method of Banzhaf is as follows: “The citrated plasma is diluted with half 
its volume of water and saturated ammonium sulphate solution is added 
up to 30 per cent, saturated solution. This mixture is heated up to 60° C. 
and held there for one hour. Then filtered while hot. The precipitate con- 
tains the native non-antitoxic proteins and a large amount of non-antitoxic 
proteins newly formed by the above method of heating. This precipitate 
is discarded. The filtrate is brought up to 50 per cent, saturated ammonium 
sulphate solution. The resulting precipitate contains only pseudoglobulin 
and antitoxin and is pressed to remove excess of fluid, followed by dialyza- 
tion until free from salts. After dialysis is completed 0.8 per cent, sodium 
chlorid is added for isotoxicity and 0.3 per cent, tricresol for preservation. 
It is then filtered through paper pulp and a Berkefeld clay filter, tested for 
sterility and potency, and filled into sterile syringes or bottles. This method 
gives a concentration of about six times the original potency” (Park). 
Heineman3 has recently described a process consisting of a combination 
of well-known methods and persistent repetition of certain details until the 
desired result is obtained. He states that more of the non-antitoxic proteins 
are eliminated with a product of higher concentration of antitoxic globulins, 
and that original plasma containing but 100 to 200 units of antitoxin can 
be used to advantage. 
As shown by Berg4 filtration through a Berkefeld type of filter does not 
remove appreciable amounts of antitoxin. 
Standardizing the Serum.—During the earlier investigations it was be- 
lieved that toxin was quite stable, and that it possessed a definite toxicity 
with a constant value in neutralizing antitoxin. Upon these suppositions 
the original Behring-Ehrlich antitoxin unit was based, consisting of 10 
times the amount of antitoxin that neutralized 10 fatal doses of toxin. 
For example, if the minimal lethal dose (M. L. D.) of toxin was 0.001 c.c., 
and 0.01 c.c. was neutralized by 0.01 c.c. of serum, then 0.1 c.c. of serum 
equaled one unit, or 10 units in a cubic centimeter. Later, stronger serums 
were found, and von Behring and Ehrlich modified the unit, which they 
now call the immunity unit, to be that quantity of antitoxin which will neu- 
tralize 100 times the minimal fatal dose for a 250-gram guinea-pig. 
It was soon discovered that toxins are unstable compounds and that 
almost immediately after their production they begin to change into toxoids, 
which are not acutely poisonous, but which retain their power to neutralize 
antitoxin. 
In order to standardize a serum it is necessary that the strength of the 
toxin be known and, since this is so variable, a standard antitoxin is supplied 
by the Hygienic Laboratory, by which the various antitoxin plants may 
measure the strength of their toxins. By mixing varying quantities of 
toxin with one unit of this standard antitoxin and injecting these into 250- 
gram guinea-pigs, the L+ (limes death) dose is obtained* which is the dose 
of toxin required to kill a pig in four days with the one unit of antitoxin. 
215 
1 Jour. Biol. Chem., 1906, 1, 161. 
2 Collected Studies from Research Laboratories, New York, 1912-1913, 7, 114. 
3 Jour. Infect. Dis., 1916, 19, 433. 
4 Jour. Infect. Dis., 1921, 29, 86. 216 
ANTITOXINS 
In order to accurately determine this dose many pigs may be required, 
but this method of titration is the key-note to successful standardiza- 
tion. 
Such a titration, for instance, has shown a toxin to react as follows: 
Table 1.—Method of Determining the L+ Dose of Diphtheria Toxin 
One antitoxin unit + 0.2 c.c. toxin = No visible symptoms. 
“ “ “ +0.22 “ “ = No symptoms. 
“ “ “ + 0.24 “ “ = Usually no symptoms or a very slight reaction. 
“ “ “ + 0.25 “ “ = Very slight congestion and edema. 
“ “ “ +0.26 “ “ = Slight edema at site. 
“ “ “ + 0.28 “ “ = Edema; sometimes late paralysis. 
“ “ “ + 0.3 “ “ = Acute edema and sometimes death. 
“ “ “ +0.32 “ “ = Always acute death about the fourth day. 
“ “ “ +0.34 “ “ = Death from second to third day. 
“ “ “ + 0.35 “ “ = Death about the second day. 
Here the L+ dose is 0.32 c.c. The dose of toxin that just neutralizes 
the antitoxin without causing symptoms has been called by Ehrlich the 
Lo (limes zero) dose, and in this instance it is about 0.24 c.c. This deter- 
mination, however, has not the same practical value as the L+ dose. 
Fig. 91.—A Hitchens Syringe. 
The needle is plugged by dipping the tip in carbolized vaselin. The side arm holds sterile salt 
solution; when the needle has been entered the injection is given by pressure on the bulb; the side 
arm is then turned upward, and the contents flow into the main barrel; when injected in this manner 
insures accuracy in dosage and uniform bulk of inoculum. 
Having determined the L+ dose of the toxin a series of six to eight 
guinea-pigs are injected with this constant dose of toxin and increasing 
amounts of the corresponding antitoxin serum; for instance, No. 1 would 
receive 0.001 c.c. of serum; No. 2, 0.002 c.c,; No. 3, 0.003 c.c.; No. 4, 0.004 
c.c.; No. 5, 0.005 c.c.; No. 6, 0.006 c.c., etc. If at the end of the fourth 
day Nos. 1, 2, 3, and 4 were dead, and Nos. 6 and 5 were alive, the serum 
would contain 200 units of antitoxin in a cubic centimeter. These injec- 
tions are best given with precision syringes, the one devised by Hitchens 
being particularly serviceable (Fig. 91). The syringes are sterilized and 
the needles are dipped in sterile vaselin to plug them. The mixtures are 
made in the barrel of the syringe and sufficient sterile salt solution is placed 
in the side arm to bring the total volume of the injection up to 4 c.c., and 
to wash in all traces of toxin and antitoxin. The mixtures are allowed to THE PRODUCTION OF DIPHTHERIA ANTITOXIN 
217 
stand for at least fifteen minutes (Park) before being injected (Fig. 92). 
The pigs must be of proper weight, i. e., about 250 to 300 grams; the ab- 
dominal wall is shaved and the injection given directly in the median 
abdominal line. The animals are placed two in a cage, and carefully ob- 
served for four or five days for symptoms of toxemia and edema about the 
site of injection. 
Romer and Sames’ Method1 of Determining Small Amounts of Diphtheria Antitoxin. 
—The principle of this method is based upon the observation that, when very small amounts 
of diphtheria toxin are injected intracutaneously into the abdominal skin of guinea-pigs, 
small areas of edema and necrosis result in about forty-eight hours. When such injections 
are made with mixtures of toxin and antitoxin the presence of free toxin is indicated by such 
tissue changes. It is chiefly used in determining the antitoxin content of human serums 
after active immunization with the toxin-antitoxin mixtures of von Behring. 
Technic.—I conduct this test in the following manner: The “limes-necrosis” (Ln) dose 
of a toxin is first determined, which is the amount of toxin which, together with 
unit of standard antitoxin, will still produce a minimal amount of necrosis in forty-eight 
hours after intracutaneous injection into guinea-pigs. A series of dilutions of the L+ dose of 
Fig. 92.—A Battery of Hitchens Syringes. 
a toxin is made, ranging from 1 : 5 to 1 : 100, and 0.2 c.c. of each mixed with 0.2 c.c. of anti- 
toxin so diluted that each 0.1 c.c. contains of a unit. These mixtures are made in small 
test-tubes, the cotton stoppers paraffined, and the tubes incubated for three hours and placed 
in the refrigerator for twenty-one hours, after which 0.2 c.c. of each is injected into guinea- 
pigs (prepared by pulling out the hairs); several injections may be made in each pig. 
When the Ln dose of the toxin has been determined this amount is mixed in a similar 
manner with varying amounts of the patient’s serum being tested. The amount of serum 
just neutralizing the toxin contains tuVu unit of antitoxin from which the amount of antitoxin 
per cubic centimeter of serum may be computed. For example, I have found that 0.00 3 c.c. 
of serum of a person reacting negatively to the Schick test neutralized this amount of toxin; 
therefore, each cubic centimeter of this person’s serum contained 0.33 unit of antitoxin. 
Kellogg’s Method for Determining the Presence of Natural Antitoxin in Human 
Serum as a Substitute for the Schick Test.—Kellogg,2 drawing attention to the practical 
difficulties of the Schick test for antitoxic immunity to diphtheria and especially the technical 
procedures surrounding the toxin, has advocated this guinea-pig intracutaneous tfest as a 
means for determining the presence or absence of antitoxin in human blood as a substitute 
for the Schick test. By means of his method it is claimed that specimens of blood may be 
sent to a laboratory and the work thereby centralized; Kellogg has given the following technic 
as employed in the California State Hygienic Laboratory; guinea-pigs are injected intra- 
cutaneously as in Romer’s method: 
1 Ztschr. f. Immunitatsf., 1909, 3, 344. 
2 Jour. Amer. Med. Assoc., 1922, 78, 1782. 218 
ANTITOXINS 
“One-three-hundredth minimal lethal dose of toxin will produce a definite sharp reaction, 
characterized by redness with some induration about 15 mm. in diameter, reaching its height 
in forty-eight hours and subsiding without necrosis. The reddened area fades quickly to a 
brownish color and there is some furfuraceous scaling of the skin. 
“The smallest amount of toxin that will produce a definite necrosis of the skin is about 
4*jy minimal lethal dose. In the test a predetermined strength of toxin is mixed with an 
equal volume of blood-serum from the person to be tested, the mixture allowed to stand 
for half an hour at room temperature, and then 0.2 c.c. of the mixture is injected intra- 
cutaneously into the shaved skin of a white guinea-pig. The toxin used is standardized so 
that the amount contained in 1 c.c. is one-thirtieth of the L+ dose. 
“Since the L+ dose of toxin is the amount that will neutralize one standard unit of anti- 
toxin, leaving 1 minimal lethal dose of toxin free, it follows that 1 c.c. of one-thirtieth the 
L+ dose mixed with 1 c.c. of serum containing fa unit of antitoxin will leave a balance of 
minimal lethal dose, and 0.1 c.c. of each will have a surplus of A '() „ minimal lethal dose of 
toxin. The volume of mixture injected, 0.2 c.c., therefore, represents one-three-hundredth 
of the L+ dose of toxin, plus whatever antitoxin may be in 0.1 c.c. of the serum. 
“The reading is made in from forty-eight to seventy-two hours. If the serum tested con- 
tains exactly %g unit of antitoxin per cubic centimeter, the amount stated by Schick as being 
the minimal protective amount, minimal lethal dose of toxin, will be free in the mixture 
and produce the reaction as noted above. 
“If the amount of antitoxin is greater, even by so small a margin as xJ-0 unit, no reaction 
will appear. If the amount of antitoxin is less than unit, even as much less as , gg, super- 
ficial necrosis will be plainly evident. 
“The reason for the sharpness of these differences with such small amounts of antitoxin 
can be appreciated when it is recalled that the L+ dose of toxin (which just kills a 250-gram 
pig when mixed with 1 unit of antitoxin) may contain mor« than 100 minimal lethal doses. 
“A variation of TV minimal lethal dose one way or the other from the :XJ 0 required to 
produce a definite reaction produces either necrosis or absolute lack of reaction, as the case 
may be.” 
PRODUCTION OF TETANUS ANTITOXIN 
The method used in the production of tetanus antitoxin is similar to 
that employed in producing diphtheria antitoxin, the horses being inocu- 
lated with increasing doses of a strong tetanus toxin. 
Tetanus Toxin.—The toxin is secured by inoculating large flasks or 
tubes of neutral veal broth containing 1 per cent, of sodium chlorid and 
peptone with abundant tetanus culture, and growing these anaerobically 
at 37° C. for two weeks. Anderson and Leake1 prepare 100 liters of broth 
with 50 kilograms of minced round steak and add 0.5 per cent, sodium 
chlorid and 1 per cent, peptone; after steaming for an hour the reaction 
is made neutral to phenolphthalein and the broth filtered through paper 
into liter Erlenmeyer flasks, followed by steaming without pressure for 
one and a half hours. The broth may be stored for a period of two weeks 
or less. Just before inoculating a 1 per cent, solution of C. P. glucose, 
(powdered) is added and the medium again heated for one and a half hours 
without pressure, cooled to 40° C., and immediately inoculated a few centi- 
meters below the surface with 1 c.c. of a twenty-four-hour broth culture 
of tejtanus bacilli which has been subcultured daily in 1 per cent, glucose 
broth for one to three weeks. No oil or other means is used to secure anaero- 
biasis; incubation is allowed to go on undisturbed for fifteen days at 37° C. 
The cultures are then filtered rapidly through Berkefeld filters, and the 
toxin preserved in fluid form with the addition of 0.5 per cent, phenol. As 
previously mentioned, the toxin rapidly deteriorates—especially tetano- 
spasmin—and for purposes of antitoxin standardization it is usually pre- 
served in a dry state after being precipitated with ammonium sulphate. 
The yellowish, crystalline masses are readily soluble in water or salt solu- 
tion, and should be used immediately after solution takes place. The strength 
of the toxin is determined by injecting increasing amounts into white mice 
or 350-gram guinea-pigs. 
1 Jour. Med. Research, 1915, 33, 239. PRODUCTION OF TETANUS ANTITOXIN 
219 
Immunizing the Animals.—According to Park, the “horses receive 5 c.c. 
as the initial dose of a toxin, of which 1 c.c. kills 250,000 grams of guinea- 
pig, and along with this twice the amount of antitoxin required to neutralize 
it. In five days this dose is doubled, and then every five to seven days 
larger amounts are given. After the third injection the antitoxin is omitted. 
The dose is increased at first slowly until appreciable amounts of antitoxin 
are found to be present, and then as rapidly as the horses can stand it, until 
they support 700 to 800 c.c. or more at a time. This amount should not be 
injected in a single place, or severe local and perhaps fatal tetanus may 
develop, or immunization may be conducted in exactly the same manner 
as described for diphtheria except that one only increases each dose by 
50 per cent. Horses withstand the effects of tetanus toxin better than 
diphtheria toxin. Good horses yield a serum containing 200 to 600 units 
per cubic centimeter” (Park). 
Collecting the Serum.—The horses are bled, and the serum is collected 
under strict aseptic precautions, in a manner similar to the collection of 
anti diphtheric serum. The serum should be clear and free from blood, 
and should be proved sterile by cultural tests. It may be preserved in the 
liquid state by adding 0.5 per cent, of phenol or 0.4 per cent, of tricresol. 
Standardizing the Serum.—The official immunity unit of tetanus anti- 
toxin of the United States Government is based largely upon the work of 
Rosenau and Anderson. These investigators, together with a Committee 
of the Society of American Bacteriologists, have defined the unit of tetanus 
antitoxin to be “ten times the least amount of serum necessary to save the life 
of a 350-gram guinea-pig for ninety-six hours against the official test dose of a 
standard toxin.” This test dose consists of 100 minimal lethal doses of a 
precipitated and dried toxin, tested out against 350-gram pigs, and pre- 
served in the Hygienic Laboratory from where it is sent to various anti- 
toxin plants for the purpose of securing a uniform method and unit of 
standardization. 
In standardizing tetanus antitoxin the L+ dose of toxin is employed. 
A standard toxin and an antitoxin, arbitrary in their first establishment, 
are preserved in the Hygienic Laboratory, and are kept constant by making 
frequent tests one against the other. In determining the L+ dose, increas- 
ing amounts of toxin are mixed with a constant amount of antitoxin equal 
to one-tenth of an immunity unit, and injected into 350-gram pigs. The 
L+ dose must contain just enough toxin to neutralize this amount of anti- 
toxin and kill a pig in four days. This L+ dose of toxin is sent out by the 
Hygienic Laboratory to those interested, commercially or otherwise, in the 
manufacture of antitoxin for purposes of standardization. 
For determining the strength of an unknown serum a large number of 
mixtures are made, each containing the L+ doses of the toxin and increas- 
ing quantities of antitoxin. The measurements are made with accurate 
volumetric pipets, and the total volume brought up to 4 c.c. with sterile 
salt solution in order to equalize concentration and pressure. The mix- 
tures are allowed to stand at room temperature for an hour, and are then 
injected subcutaneously into 350-gram pigs. This method of titrating the 
antitoxin is shown in the following example from Rosenau and Anderson: 220 
ANTITOXINS 
Table 2.—Method of Titrating Tetanus Antitoxin 
No. or 
Weight of 
Subcutaneous Injection of 
a Mixture of— 
Time of Death. 
Pig. 
Pig in Grams 
Toxin (Test 
Dose). 
Antitoxin. 
1 
360 
Gram. 
0.0006 
C.c. 
0.001 
Two days, four hours. 
Four days, one hour. 
Symptoms. 
Slight symptoms. 
No symptoms. 
2 
350 
0.0006 
0.0015 
3 
350 
0.0006 
0.002 
4 
360 
0.0006 
0.0025 
5 
350 
0.0006 
0.003 
In this series the animal receiving 0.0015 c.c. of antitoxin died in approxi- 
mately four days; this amount of serum, therefore, represents -j of one unit. 
The nature of the botulinus poison has previously been described. Was- 
sermann has recently immunized horses against this toxin, and the anti- 
toxin shows unmistakable value in animal experiments, although it has 
not been employed frequently enough in this form of poisoning in human 
beings to prove its value. 
BOTULINUS ANTITOXIN 
ANTIDYSENTERIC SERUM 
The Kruse-Shiga type of dysentery bacillus has been shown to produce 
varying amounts of a soluble toxin; and antiserums, which are partly anti- 
toxic and partly bactericidal in nature, have been prepared, and have ap- 
parently yielded good therapeutic results in the hands of several observers. 
Potent antiserums for the Flexner type of bacillus and for various strains 
isolated from the feces of cases of infantile ileocolitis have not been pro- 
duced. Even a virulent strain of the dysentery bacillus does not produce 
true soluble toxins in a manner comparable to those produced by tetanus 
and diphtheria. Potent toxins are seldom secured with less than two to 
three weeks’ incubation, and fresh cultures of whole or autolyzed bacilli 
are likewise quite too toxic, indicating that although a soluble toxin may 
be produced, considerable endotoxin is also present in the bacilli. 
Antidysenteric serum has very little prophylactic value, but in individ- 
ual cases it frequently exerts a curative action, and should be available 
for use in institutions and armies when dysenteric infection is prevalent. 
The older investigators, such as Kruse and Shiga, produced antiserums 
by immunization with whole bacilli. Later Kraus and Doerr prepared 
antitoxic serums with the toxin alone. At the present time the evidence 
would seem to indicate that the best serums are prepared by injecting both 
toxins and bacilli, producing a serum that is essentially antitoxic and bac- 
tericidal in action. 
Culture.—Young and healthy horses are best adapted for immuniza- 
tion. Two methods may be followed: (1) Immunization with toxin, or 
(2) with young cultures of whole bacilli. As previously mentioned, investi- 
gations have tended to show that the most potent serums are secured by 
using mixtures of both toxin and micro-organisms. 
Several strains of dysentery bacilli should be used in order that a poly- ANTIDYSENTERIC SERUM 
221 
valent serum may be prepared. Cultures should be grown for two weeks 
at 37° C., in alkaline broth similar to that used for preparing diphtheria 
toxin; this should be neutralized to phenolphthalein, and 7 c.c. normal 
soda solution to a liter added. The minimal lethal dose of the mixed un- 
altered cultures is determined by giving young rabbits increasing doses 
intravenously in order to obtain a guide as to the proper dose for immuni- 
zation. Fatal doses produce severe diarrhea and paralysis of the extremities, 
with rapid loss in weight. Rabbits and horses are quite susceptible to the 
toxin; guinea-pigs and mice are more resistant. 
Table 3.—Method oe Determining the Minimal Lethal Dose of Dysentery Culture 
No. 
Weight, Grams 
Dose in C.c. 
Result. 
1 
710 
0.025 
No symptoms. 
No symptoms. 
Diarrhea. Recovered. 
2 
690 
0.05 
3 
695 
0.1 
4 
690 
0.2 
Death third day. 
Death second to third day. 
5 
700 
0.3 
In this instance the minimal lethal dose was 0.2 c.c. and subsequent 
cultures of the same strains, grown under similar conditions, showed this 
dose to remain quite constant. 
It is good practice to keep the cultures growing during the entire time of 
immunization. Cultures may, howrever, be grown for three weeks, filtered 
through porcelain, and with the addition of 0.5 per cent, phenol, the toxin 
preserved for long periods of time. The minimal lethal dose of such a toxin 
is determined in the manner directed above. 
Immunizing the Animals.—Since horses are quite susceptible, the initial 
dose of unfiltered and unheated culture should not be larger than the mini- 
mal lethal dose for a young rabbit. The dosage is gradually increased, 
and the injections are given subcutaneously for from four to six months, 
after which several injections of from 300 to 350 c.c. may be given intra- 
venously at one time. If at any time diarrhea and other symptoms of 
toxemia are well marked, subsequent doses should be smaller and should 
be given at longer intervals until a higher immunity is produced. 
Instead of using bouillon cultures, young agar cultures may be used, 
the bacilli being grown for seventy-two hours, and one-tenth of an ordi- 
nary slant being given as the first dose. The early doses are heated to 
60° C. for an hour and injected subcutaneously; the later doses consist of 
cultures washed from 30 to 40 tubes, and are given intravenously. 
Flexner and Amoss’s Method for Rapid Production of Antidysenteric 
Serum.1—By this method potent antidysenteric sera can be safely pre- 
pared in the horse by the method of three successive intravenous injections 
of living cultures or toxin with intervening rest periods of seven days; and 
effective serum for therapeutic purposes may be prepared in about ten 
weeks. By inoculating alternately living bacilli belonging to the Shiga 
and Flexner groups a polyvalent serum of high titer may be secured which 
should be suitable for the serum treatment of acute bacillary dysentery, 
irrespective of the particular strain or strains of the dysentery bacillus 
causing the infection. 
Cultures are grown upon agar-agar slant surfaces in tubes 15 x 160 mm. 
in size for twenty-four hours, and the growth in each tube suspended in 
1 Jour. Exper. Med., 1915, 21, 515. 222 
ANTITOXINS 
2 c.c. of salt solution. Horses are injected intravenously; the first dose is 
1 c.c. of the suspension of Flexner bacilli after heating to 60° C. for thirty 
minutes; on each of the following two days 5 c.c. of heated suspension are 
usually given, followed by a rest of seven days, when living cultures are 
injected. The temperature is taken daily and used as an index of the reac- 
tion and subsequent doses. With the living bacilli the doses injected on 
each of three days in succession are 4, 10, and 30 loopfuls suspended 
in salt solution. Flexner and Shiga bacilli are inoculated alternately on 
three successive days, with intervening rest intervals of seven days, the 
doses being chosen so as to produce a sharp febrile reaction which sub- 
sides in twenty-four hours. At the end of eight to ten weeks’ immuniza- 
tion the serum contains immune agglutinins and a well-marked degree of 
antibacterial and antitoxic value. 
Collecting and Testing the Serum.—After three or four months a trial 
bleeding should be made and the serum tested as follows: The minimal 
lethal dose of a culture is determined and ten times this amount placed 
in a series of tubes or syringes with increasing doses of serum; the total 
quantity of injection is made up to 4 c.c. with sterile salt solution. The 
mixtures are set aside for one hour at 35° C. and injected intravenously in 
young rabbits. The animals are to be observed for at least five days for 
diarrhea, paralysis, and loss in weight. For determining the antitoxic 
value a toxin is prepared by cultivating the bacilli in sugar-free broth con- 
taining calcium carbonate "for three days; the bacilli are then killed with 
ether; the ether is removed and the culture filtered through hard paper or 
a Berkefeld filter and the toxin kept in the refrigerator. 
Table 4.—Method of Testing Antidysenteric Serum (Kruse-Shiga) 
No. 
Weight, 
Grams 
Culture, 
0.2 C.c. M. L. D. 
C.c. 
Serum C.c. 
Result. 
1 
600 
2 0 
0 00025 
Died second day. 
Died third day. 
Diarrhea, recovered. 
2 
610 
2 0 
0 0005 
3 
615 
2 0 
0 001 
4 
590 
2 0 
0 002 
Diarrhea, paralysis. 
No symptoms. 
No symptoms. 
5 
600 
2 0 
0 004 
6 
590 
2.0 
0.006 
In this instance 0.004 c.c. of serum was sufficient to protect young rab- 
bits against ten fatal doses of culture, and demonstrated that it is possible 
to secure a fairly potent serum against the toxins of the Kruse-Shiga micro- 
organism. 
According to Todd, if the antiserum is given at least one-half hour after 
administering the culture, it will protect the rabbit. If given twenty-four 
hours later, it affords no protection. Similarly, the mixtures of culture 
and serum must not be injected immediately after mixing, as the results 
are more irregular than if they are allowed to stand for one-half to one 
hour before injecting. 
If the trial bleeding shows a satisfactory serum the horse is bled asepti- 
cally, as was previously described, and the serum is separated and preserved 
with 0.5 per cent, phenol in quantities of 10 c.c. in sterile containers. As 
there is no official immunity unit, the serum is administered in doses of 
from 5 to 10 c.c. until a therapeutic effect is secured. ANTISTAPHYLOCOCCUS SERUM 
223 
ANTISTAPHYLOCOCCUS SERUM 
Both Staphylococcus pyogenes aureus and S. pyogenes albus have been 
shown to produce certain soluble toxins, such as a leukocidin and a hemo- 
lysin, wThich are partly responsible for the tissue destruction and symp- 
toms that accompany these infections. Severe staphylococcus infection is. 
probably due in part to the paralyzing effect and actual destructive action 
of the leukocidin upon the leukocytes, preventing for the time being the 
walling off of the lesion and effectual phagocytosis. Antistaphylococcus 
serums have been shown to counteract the action of the leukocidin and 
the hemolysin, and may be useful in the treatment of severe, spreading, or 
metastatic staphylococcus infections. 
According to Neisser and Wechsberg, during staphylococcus disease an 
antihemotoxin is produced against the hemotoxin of the cocci; later Bruck, 
Michaelis, and Schulze attempted to show that a demonstration of this 
antistaphylolysin in the serum may be regarded as evidence of a staphylo- 
coccus infection. 
Preparation of Antistaphylococcus Seium.—For immunization purposes 
several different cultures of the Staphylococcus aureus should be used in 
order that the antiserums may be polyvalent. Goats or horses may be em- 
ployed. Cultures may be grown on neutral agar for forty-eight hours, and 
an emulsion, equivalent to half an agar slant, heated to 60° C. for one hour 
and injected subcutaneously in an adult goat. If 10 different strains are 
used a 4-mm. loopful from each culture, emulsified in 5 c.c. of sterile salt 
solution, will be about the proper dose for the first injection. Subsequent 
doses are given at intervals of a week, and are rapidly increased in size 
until full, living, unheated cultures are injected intravenously without 
harm to the animal. The serum may be tested by determining its content 
of antilysin or of bacteriotropin. Complement-fixation tests are occasionally 
useful for obtaining an insight into the quantity of bacteriolysin present. 
Technic of the Antilysin Test.—The object of this test is to determine 
the amount of antihemolvsin present in a serum, which is dependent on 
the amount of serum necessary to protect the red blood-cells of rabbits 
against a solution of the staphylolysin. 
(a) Staphylolysin.—This is prepared by growing a known hemolysin- 
producing staphylococcus in slightly alkaline broth for three weeks, filtering 
through a Berkefeld filter, and preserving the filtrate with 0.5 per cent, 
phenol in the refrigerator. 
0b) Rabbit Blood.—Remove 2 or 3 c.c. of blood from the ear of a rabbit 
and place in 5 c.c. of a 1 per cent, sodium citrate in normal salt solution. 
Wash the corpuscles three times, and make up in a 1 per cent, suspension 
(dose 1 c.c.) or up to the original volume of blood (dose, 1 drop). 
(c) Patient’s Serum.—The serum is inactivated by heating to 56° C. for 
half an hour. 
(d) Control Serum.—As every normal serum contains a certain amount 
of antilysin,. it is necessary to use a normal control serum. Normal horse- 
serum, dried in vacuo to prevent deterioration, and freshly dissolved for 
each test in 10 volumes of sterile distilled water or salt solution, has been 
advocated by Bruck, Michaelis, and Schulze. 
(e) The Test.—It is first necessary to titrate the staphylococcus filtrate to 
ascertain the amount of lysin present. This is accomplished according to 
the following scheme: 224 
ANTITOXINS 
Amount of 
Staphylolysin 
Filtrate. 
Rabbit 
Blood 
1 Per Cent. 
Normal 
Salt Solution. 
Result of Hemolysis After Two 
Hours at 37° C. and Twenty- 
four Hours in Refrigerator. 
0.005 c.c 
1 c.c. 
q. s. 2 c.c. 
No hemolysis. 
No hemolysis. 
Slight hemolysis. 
Marked hemolysis. 
Complete hemolysis. 
Complete hemolysis. 
Complete hemolysis. 
Complete hemolysis. 
0.001 c.c 
1 c.c. 
q. s. 2 c.c. 
0.02 c.c 
1 c.c. 
q. s. 2 c.c. 
0.05 c.c 
1 c.c. 
q. s. 2 c.c. 
0.1 c.c 
1 C.C. 
q. s. 2 c.c. 
0.2 c.c 
1 c.c. 
q. s. 2 c.c. 
0.5 c.c 
1 c.c. 
q. s. 2 c.c. 
1 c.c. 
Table 5.—Method of Titrating Staphylolysin 
In this test 0.1 c.c. is the smallest amount of lysin that can completely 
hemolyze the given quantity of erythrocytes, and is taken as the unit for 
the second part of the test. 
The lytic dose of filtrate just determined is now placed in a series of 
small test-tubes, with increasing doses of serum to be tested and a constant 
dose of corpuscles. 
Table 6.—Method of Titrating Antistaphylolysin tn a Serum 
Amount of 
Filtrate. 
Inactivated 
Serum. 
1 Per Cent. 
Rabbit 
Corpuscles. 
Normal Salt 
Solution. 
Readings After Incubation at 37° C. 
for Two Hours and Twentv-four 
Hours in the Refrigerator. 
0.1 c.C 
0.001 c.c. 
1 c.c. 
q. s. 2 c.c. 
Complete hemolysis. 
Slight inhibition of hemolysis. 
Marked inhibition of hemolysis. 
Complete inhibition of hemolysis. 
Complete inhibition of hemolysis. 
Complete inhibition of hemolysis. 
0.1 c.c 
0.005 c.c. 
1 c.c. 
q. s. 2 c.c. 
0.1 c.c 
0.01 c.c. 
1 c.c. 
q. s. 2 c.c. 
0.1 c.c 
0.05 c.c. 
1 c.c. 
q. s. 2 c.c. 
0.1 c.c 
0.1 c.c. 
1 c.c. 
q. s. 2 c.c. 
0.1 c.c 
0.2 c.c. 
1 c.c. 
q. s. 2 c.c. 
In this instance 0.05 c.c. of the patient’s serum was sufficient completely 
to neutralize the lysin. 
A similar test is carried out with normal horse-serum. The antilytic 
dose of this serum is taken as 1, and the patient’s serum is compared with 
this unit. For example, if 0.1 c.c. of normal horse-serum was sufficient to 
neutralize the lysin in this experiment, then the antilysin value of the pa- 
tient’s serum is 2. 
According to Arndt and others, a high antilysin content of a serum is 
to be regarded as indicating a staphylococcic infection, even if it is impos- 
sible to establish fixed limits for the values. 
PRODUCTION OF ANTIVENIN 
Snake venom contains two toxins, one being largely neurotoxic and 
producing paralysis of the respiratory centers, and the other being hemo- 
toxic and irritant, and producing local necrosis of tissues, hemolysis, etc. 
In venom poisoning the neurotoxic effect is most dangerous. Largely as 
the result of the work of Calmette and Fraser an antivenin has been pre- 
pared that is capable of counteracting the neurotoxic action not only of 
cobra venom, but to a lesser extent of other venoms as ■well. These serums, 
however, appear to have no effect or but very little upon the irritant toxins. THE MEASURE OF ANTITOXINS 
225 
In the poisonous American snakes, such as the rattler, moccasin, and copper- 
head, the effects of the irritant toxins are largely in evidence, and satis- 
factory antiserums for these venoms have not been prepared (McFarland). 
In preparing antivenins the toxins, since they are thermolabilC, must 
be used unheated; subcutaneous injections are usually followed by exten- 
sive sloughing, and although a certain amount of immunity may be in- 
duced in the horse by intravenous injection, there is apparently no protec- 
tion against the local action of the toxins. 
Preparation of Antivenin.—According to Calmette, horses may be im- 
munized by giving them weekly subcutaneous injections of gradually in- 
creasing doses of cobra venom, heated to 70° C., for an hour which precipi- 
tates the irritant toxins without injuring the neurotoxin. The initial dose 
is usually 0.01 gram, gradually increased, until by the end of four months 
4 grams may be given at a single dose. The serum is then tested by mixing 
increasing doses with the minimal lethal dose for a young rabbit, and in- 
jecting the mixtures intravenously into a series of rabbits. 
Since the neurotoxin may prove dangerous in any case of snake bite, 
antivenin may be given to advantage, although the local pain and necrosis 
are not relieved by the serum. 
Flexner and Noguchi have successfully immunized rabbits and dogs 
with rattlesnake venom which had been treated with hydrochloric acid 
and iodin trichlorid, which deprived the venom of a large part of its toxicity 
while still preserving the power of causing the production of antivenin. 
Anticrotalus venin was found without appreciable antitoxic power for cobra 
and daboia venoms, and but feeble antitoxic activity for the water-moccasin 
venom. 
PRODUCTION OF POLLEN ANTITOXIN 
The pollen of certain plants is markedly toxic for susceptible individuals. 
In America the pollen of the golden-rod and of rag weed frequently produce 
a syndrome of distressing symptoms known as “autumnal catarrh.” The 
onset and character of the symptoms of pollen intoxication are strongly 
suggestive of an anaphylactic reaction. Dunbar has studied pollen toxins 
quite extensively, and considers them the etiologic factor in the production 
of hay-fever. 
Pollen antitoxin has been prepared by immunizing susceptible horses, 
the toxin being isolated by mixing the ground pollen with 5 per cent, sodium 
chlorid solution and 0.5 per cent, phenol at 37° C. for ten hours. In the 
form of a protein, it is then precipitated by adding eight to ten volumes 
of 96 per cent, alcohol, dissolving the resultant white precipitate in physio- 
logic salt solution (Citron). 
Antitoxin Unit.—A unit is the definite measure of antitoxin in any serum 
or solution that will neutralize a certain amount of toxin. Originally the unit 
of diphtheria antitoxin was determined according to the method of Behring 
and was defined as ten times the least quantity of serum that protected a 
300-gram guinea-pig against ten times the least certainly fatal dose of toxic 
bouillon. This method proved unsatisfactory on account of variations in 
the toxin as pointed out by Ehrlich, who showed that this toxin does not 
possess uniform combining affinity for the antitoxin. Ehrlich, therefore, 
devised a method by which a standard dried toxin was produced which is 
now widely used and described in the preceding pages. As previously 
stated the United States Government has established a definite unit for 
THE MEASURE OF ANTITOXINS 226 
ANTITOXINS 
the standardization of diphtheria and tetanus antitoxins, and frequently 
examines the serums made by various licensed manufacturers. Officers 
of the Public Health and Marine Hospital Service purchase from reliable 
pharmacists several grades of antitoxins made by each manufacturer, 
which are then sent to the Hygienic Laboratory at Washington, where they 
are tested for potency, freedom from contamination by bacteria, chemical 
poisons, especially tetanus toxin, and for excessive amounts of preserva- 
tive. Delinquencies are reported immediately, and steps are taken to with- 
draw that particular lot of serum from the market. 
A unit of diphtheria antitoxin may be defined as the “amount of antitoxin 
that will just neutralize 100 minimal fatal doses of toxin for a 250-gram guinea- 
pig” 
A unit of tetanus antitoxin may be defined as the “ amount of antitoxin 
which will just neutralize 1000 minimal fatal doses of toxin for a 350-gram 
guinea-pig.” 
The standardization of these serums is useful as a guide to their adminis- 
tration, especially when given for prophylactic purposes, where experience 
has taught that so many units usually confer protection; it also serves for 
purposes of record. In the treatment of diphtheria and tetanus, however, 
the serums are usually given until a therapeutic effect is noted, regardless 
of the number of units administered. If it were possible to determine 
quickly and accurately the amount of toxin in a given patient, then neu- 
tralization could be accomplished along the same lines that make this pos- 
sible in the test-tube. The indications are to administer at once sufficient 
antitoxin to neutralize all the toxin, giving subsequent doses large enough 
to overcome the toxin as it is produced until the focus of infection is removed. 
Antitoxin should be kept in a cold place and protected from air and 
light. When this is done, they usually do not deteriorate more than 30 
per cent, of their original strength, and often much less, within a year. All 
manufacturers place a large number of units in the container than the 
label calls for, in this way allowing for the gradual loss in strength up to 
the date specified on the label. According to Park the antitoxin in old 
serum is just as effective as that in fresh serum, except that there is less 
of it. 
PRACTICAL APPLICATION 
The employment of antitoxic serums, both in prophylaxis and in the 
treatment of infection, is considered in greater detail in the chapters on 
Passive Immunization and Serum Therapy. CHAPTER XIV 
FERMENTS AND ANTIFERMENTS 
Ferments.—The term “ferment” introduced in relation to infection 
and immunity has proved very confusing. Owing largely to the investiga- 
tions of Vaughan, Abderhalden, Jobling, and their associates this term 
has come into very general use, but in some instances appears to have been 
ill chosen. 
It is well known that some bacteria contain or elaborate true ferments 
or enzymes; also that an increase of true ferments like the proteases may 
be found in the plasma in pathologic conditions. Jobling has used the 
term in this strict meaning and in conformity with our general knowledge 
of true enzymes or ferments. 
Vaughan, Abderhalden, and others, however, have applied the term 
to other substances in serum which have not been clearly differentiated 
from the lytic antibodies or amboceptors and complement. Of course, 
complement may be a ferment, but this has not been definitely proved. 
And the use of the word “ferment” as practically synonymous with anti- 
body, has resulted in confusion with the true enzymes or ferments, as pro- 
teases, lipases, and others. 
Bacterial Ferments in Relation to Infection.—While many pathogenic 
bacteria are known to produce true ferments or enzymes, those of Bacillus 
pyocyaneus and the pneumococcus have received most attention in rela- 
tion to the production of disease. 
Bacillus pyocyaneus contains and produces a particularly active proteo- 
lytic ferment which has been studied by Emmerich and Low,1 and to which 
they have given the name pyocyanase. These investigators have claimed 
that this ferment is capable of agglutinating and digesting not only B. 
pyocyaneus but other bacteria as well, including anthrax, typhoid, diphtheria,, 
and cholera bacilli. On the basis of the positive results of in vitro aggluti- 
nation and bacteriolytic tests with pyocyanase, they advocated the use 
of this enzyme for the treatment of bacterial infections by local applica- 
tion and subcutaneous injection. According to these experiments proteo- 
lytic bacterial ferments may reduce infection by direct destruction of the 
infecting bacterium. 
Fermi,2 Petrie,3 and Dietrich,4 however, were unable to substantiate 
these claims for the bacteriolytic activity of pyocyanase. In their opinion 
bacterial enzymes are unable to digest bacteria from which they are de- 
rived or bacteria of other kinds, and that, generally speaking, proteolytic 
ferments exert no action on living plant or animal cells. These investiga- 
tors were inclined to believe that the morphologic and bactericidal changes 
produced by seeding bacteria in solutions of pyocyanase were brought 
about by osmotic changes or plasmoptysis. 
Proteolytic enzymes of pneumococci have been studied by Dick5 who 
found them in the blood during pneumonia about the time of crisis, and 
1 Ztschr. f. Hyg., 1899, 31, 1; ibid., 1901, 3691. Centralbl. f. Bakteriol., 1900, 27, 1. 
2 Ztschr. f. Hyg., 1894, 18, 83. 
3 Jour. Path, and Bacteriol., 1903, 8, 200. 
4 Centralbl. f. Bakteriol., 1901, 30, 574. 
5 Jour. Infect. Dis., 1912, 10, 383. 
227 228 
FERMENTS AND ANTIFERMENTS 
Rosenow,1 who demonstrated that extracts of virulent pneumococci and 
filtrates of broth cultures contain proteolytic enzyme capable of hydro- 
lyzing the proteins in broth and serum with a proportional increase of toxicity. 
Avery and Cullen2 found an intracellular enzyme or enzymes in pneumococci 
capable of hydrolyzing to some extent intact protein and especially pep- 
tones; also endolipolytic and various carbohydrate-splitting ferments. 
It is highly probable that the enzymes produced by pathogenic bacteria 
are for the primary purpose of nutrition and the production of favorable 
environmental conditions for multiplication. As shown by Diehl3 their pro- 
duction is greatly influenced by the conditions under which the bacteria 
are growing. In so far as their relation to infection is concerned bacterial 
ferments probably possess an important status for at least two reasons: 
1. Ferments elaborated by bacteria aid in their nutrition and enable 
them to survive in the tissues, thereby aiding in the production of infection. 
2. These ferments are probably able to bring about digestion of de- 
vitalized body and bacterial cells with the production of toxic substances 
capable of retarding phagocytosis, and when absorbed adding an element 
to toxemia. 
On the other hand, these ferments and especially the products of diges- 
tion, are inimical to bacteria of the same and different species, thereby 
tending to sterilize local collections of pus and offering an explanation for 
the instances in which pus from long-standing lesions are found to be free 
of living bacteria. 
The similarity of toxins to ferments has been previously discussed in 
the chapters on Infection. 
Immunity to Bacterial Ferments; Antiblastic Immunity.—Since the 
various ferments produced by some pathogenic bacteria may play an im- 
portant role in the production of infection and disease, the question natur- 
ally arises as to whether or not the body cells possess a means of defense. 
Dochez and Avery4 have found that antipneumococcus sera exert an 
inhibitory influence upon the growth of pneumococci which they believe 
is directed against the ferments produced by these bacteria. “Pneumococcus 
in order to grow must obtain a sufficient supply of protein and carbo- 
hydrate; these substances are furnished by the environmental medium, 
but probably require to render them suitable for absorption preliminary 
preparation in the nature of digestion. This change is effected at the sur- 
face of the bacterial cell and the integrity of this digestive zone is essential 
to the growth of the bacterium. Anti-enzymotic bodies such as have been 
demonstrated'in immune serum act at the point of contact of the cell with 
its environment, and influence in an unfavorable manner the nutritional 
processes there carried on, and the consequence of such action is retarda- 
tion or inhibition of growth.” To this possible type of immunity Dochez 
and Avery have applied the term “antiblastic immunity” indicating an- 
tagonism to the growth activities of a micro-organism. The term was 
coined by Ascoli5 several years ago to explain the action of antianthrax 
serum inhibiting the metabolic activities of anthrax bacilli, and particularly 
the formation of capsules. 
Similar investigations against the ferments of other bacteria do not 
appear to have been made except by Gheorghiewsky,6 who found immune 
1 Jour. Infect. Dis., 1912, 11, 286. 
2 Jour. Exper. Med., 1920, 32, 547, 571, 583. 
3 Jour. Infect. Dis., 1919, 24, 347. 
4 Jour. Exper. Med., 1916, 23, 61. 
5 Centralbl. f. Bakteriol., orig., 1908, xlvi, 178. 
6 Ann. de l’Inst. Pasteur, 1899, 13, 298. serum inhibited pigment production by Bacillus pyocyaneus, and by von 
Dungern,1 who found that immune serum may inhibit the liquefaction of 
gelatin by Staphylococcus aureus. Further researches may show that 
ferments greatly aid bacteria in infection and that forces may be mobilized 
by the animal body in opposition to them. 
Leukocytic and Serum Ferments in Relation to Infection.—In addition 
to the true ferments elaborated by bacteria either as exogenous ferments 
or as endoferments released upon disintegration of the bacterial cell, the 
serum during infection and disease is found to contain additional ferments 
probably derived from the body cells. 
Vaughan regards these ferments as called out by the bacterial protein 
and more or less specific for this protein; in his theory of infection and 
immunity the ferments are believed to split the bacterial protein with the 
formation of a toxic and a non-toxic portion. 
Jobling and Peterson2 have studied the leukocytic and serum ferments 
in relation to infection with particular care and in an exhaustive manner. 
They have never confused these ferments with antibodies and have ad- 
hered strictly to chemical methods in their studies. They regard digestion 
of foreign proteins as the basis of resistance to and recovery from bacterial 
invasion and infection; for overcoming intoxication not due to the soluble 
or true exotoxins, they look to the cells or fluids of the body for the elabo- 
ration of true ferments capable of digesting toxic protein fragments to their 
lowest degradation and non-toxic products. The importance of these 
serum ferments to infection is, therefore, in direct relation to the part played 
by the bacterial proteins and protein-split products in pathologic processes. 
Of these ferments the proteolytic ferments are of most importance in 
breaking down toxic complexes to non-toxic forms. Petersen3 has described 
several in serum as follows: (1) Leukoproteases embracing (a) an alkaline 
active ferment capable of splitting native protein largely to the proteose 
stage; (b) an acid-active ferment with a similar range of activity; (c) and 
ereptase, active in both acid and alkaline reaction and splitting partially 
hydrolyzed proteins to the amino-acid stage. These ferments are derived 
from disintegrating, but not living leukocytes. (2) Serum protease: a poly- 
valent, trypsin-like ferment, active in neutral or slightly acid or alkaline 
reactions. When antiferment is removed it is able to digest any native 
protein to the amino-acid stage. (3) Serum peptidase: a polyvalent fer- 
ment active under the same conditions as serum protease even in the presence 
of antiferment and capable of hydrolyzing proteins to the amino-acid stage. 
Of these three ferments peptidase is regarded as most important in- 
asmuch as it is not influenced by changes in antiferment, and is of the 
nature of a detoxicating agent hydrolyzing toxic albumoses and peptones 
to the non-toxic amino-acids. 
Immunity to Leukocytic and Serum Ferments; Nature of Antiferments.— 
According to some investigators, antiferments are to be found in large 
amounts in all normal serums, and are probably vitally concerned in the 
processes of life in preventing autodigestion. That they may be increased 
in number artificially by immunization up to a certain limit has been dis- 
puted; it is certain that they never attain the extreme amounts possible 
with the injection of toxins. This may be due to the formation of anti- 
antienzymes, produced by a regulating mechanism that prevents anti- 
enzymes from accumulating beyond a certain point and interfering with 
IMMUNITY TO LEUKOCYTIC AND SERUM FERMENTS 
229 
1 Centralbl. f. Bakteriol., orig., 1898, 24, 710. 
2 Jour. Exper. Med., 1912-16, 16-23. 
3 Archiv, Int, Med., 1917, 20, 515. FERMENTS AND ANTIFERMENTS 
230 
nutrition. It is possible that the body mechanism exerts a strict regulating 
effect between, the formation of enzymes and antienzymes. Furthermore, 
when free receptors, such as normal antienzymes, are present in the body 
fluids, the body cells are not stimulated to produce these antienzymes in 
excess, nor does the presence of the free receptors stimulate the cells to 
produce antibodies against their normal side chains. 
•Many investigators claim to have produced antiferments experimentally. 
Morgenroth1 believed that he obtained a specific antirennin by inoculating 
goats with rennin. Sachs2 and Achaline3 assert that they have produced 
specific antipepsin or antitrypsin by inoculating animals with these ferments. 
Antisteapsin and antilactase have been prepared by Schutze,4 antityrosinase 
by Gessard,5 and antiurease by Moll.6 
Jochmann and Muller have demonstrated the presence of an antiferment 
in the serum used against leukocytic ferments in diseases associated with 
great destruction of the leukocytes. Following these observers, Marcus 
Brieger and Trebing7 found that 90 per cent, of the patients suffering from 
carcinoma or sarcoma examined by them showed an increase of antitrypsin 
in the blood. Von Bergmann and Meyer8 confirmed this observation, 
although they found that a similar increase also occurred in 24 per cent, 
of non-cancerous patients. More recent work would indicate that the 
antitrypsin may be present in acute infections, such as pneumonia, typhoid 
fever, etc., in chronic infections, such as tuberculosis and syphilis, in exoph- 
thalmic goiter, and in severe anemias. As previously mentioned, Schwartz,9 
Suginoto,10 and Jobling and Peterson11 believe that the antitryptic influence 
of blood-serum is due to the lipoids, and especially to the compounds of the 
unsaturated fatty acids. 
The tryptic ferment liberated by disintegrating leukocytes and con- 
nective-tissue cells is largely responsible for the liquefaction of these cells 
and the formation of pus, as in abscess formation and autodigestion of 
infected surface wounds. On the other hand, an antitrypsin-like substance 
tends to limit the activities of the ferment and protect the surrounding 
tissues from progressive destruction. A deficiency of this substance may 
account for the rapid breaking down of infected glands and of a walled-off 
tuberculous lesion, the development of carbuncles, etc. A study of the 
antitryptic power of the blood may, therefore, prove of value in suppurative 
processes and in malignant disease, and considerably influence a prognosis. 
The nature of antiferment or antienzyme has been the subject of in- 
vestigation for many years, and with the result that many different theories 
have been advanced. For example, different protein fractions of the serum 
have been regarded as antitryptic by Landsteiner, Oppenheimer and Aaron, 
Cathcart, and more recently Fujimoto12; Baylis and Starling, Abderhalden 
and Gigon, and Rosenthal regarded the products of tryptic activity and 
especially the amino-acids as inhibiting ferment activity. Delezenne and 
Pozarski13 found that serum preserved with chloroform soon lost its anti- 
1 Centralbl. f. Bakteriol., 1899, xxvi, 349. 
2 Fortschr. d. Med., 1902, 20, 425. 
3 Ann. de l’Inst. Pasteur, 1901, xv, 737. 
4 Zeitschr. f. Hygiene, 1904, 48, 457; Deutsch. med. Wochen., 1904, 30, 308. 
6 Ann. de l’Inst. Pasteur, 1901, 15, 593. 
6 Hofmeister’s Beitr., 1902, 2. 
7 Berl. klin. Wchnschr., 1908, xlv, 1349. 
8 Berl. klin. Wchnschr., 1908, xlv, 1673. 
9 Wien. klin. Wchnschr., 1909, xxii, 1151. 
10 Arch. f. Exper. Path. u. Pharmakol., 1913, lxxii, 374. 
11 Jour, exper. Med., 1914, xix, 239, 459. 
12 Jour. Immunology, 1918, 3, 51. 13 Compt. rend. Soc. de biol., 1903, 55. tryptic "activity; Schwartz1 found that the antitryptic action of normal 
serum is due to the lipoids of the serum which are partially removed by 
extraction with ether. Similar results were reported by Sugimoto2 and 
this brings the subject down to the excellent work of Jobling and Petersen,3 
who have identified the antiferments of serum with the unsaturated fatty 
acids. 
According to these investigations an antiferment is not an antibody 
in the immunologic sense, but consists of the highly dispersed unsaturated 
lipoids of the serum and lymph. The amount of antiferment in a serum 
would depend, therefore, upon the amount of lipoids present, the degree 
of dispersion, and the degree of unsaturation. Lessening the dispersion by 
acidifying, salting, or heating inactivates the antiferment; physical absorp- 
tion of the lipoids removes antiferment as likewise solution in chloroform 
and other lipoid solvents. They believe that the antiferment lipoids are 
in more or less intimate physical combination with the serum albumin 
with which fraction they are thrown down by the usual methods of separa- 
tion of the serum proteins; in all probability this accounts for the earlier 
views that antiferment resided in certain proteins of serum. 
According to Jobling and Petersen antiferment is greatly increased 
during the acute infectious diseases, in pregnancy, in carcinoma, and cachexia; 
also in degenerative lesions of the central nervous system, after intravenous 
injection of proteins and in various other conditions. The antiferment 
of the lymph-stream is appreciably increased after feeding, while in starva- 
tion there is a progressive decrease. 
Removal of the antiferment (unsaturated fatty acids of the serum) by 
any of the physical or chemical means mentioned above results in increased 
activity of the proteolytic ferments which may attack and digest various 
proteins including those of the serum, with the production of toxic split 
proteins. 
Ferments vs. Antibodies in Disease; So-called “Protective Ferments” 
in Pregnancy and Disease.—It is largely to the researches of Abderhalden 
and his associates that we owe our knowledge of the fact that when food- 
stuffs are introduced into the body parenterally, i. e., by subcutaneous or 
intravenous injection, ferments are produced that, by process of cleavage 
and reduction, deprive them of their individuality. 
For example, as shown by Weinland,4 normal dog serum cannot reduce 
cane-sugar, whereas the serum of a dog immunized by several injections 
of this sugar is able to reduce it in vitro by means of a specific ferment of the 
nature of invertin. Similarly, normal serum is unable to cleave edestin 
(vegetable albumin), whereas the serum of an immunized dog will split 
this protein into simpler substances. 
After he had proved experimentally that the animal organism is able 
to mobilize ferments against foreign substances Abderhalden next took 
up the question whether ferments are produced when substances native 
to the body but foreign to the blood are introduced into the circulation. 
Having learned from the researches of Veit, Sc.hmorl, Weichard, and others 
that during pregnancy syncytial cells frequently enter the maternal circu- 
lation, Abderhalden used the serums of pregnant animals, and found that 
they contained a ferment-like substance capable of splitting placental pep- 
tone into amino-acids and coagulated placenta into peptones, polypeptids, 
and amino-acids. 
FERMENTS VS. ANTIBODIES IN DISEASE 
231 
1 Wien. klin. Wchnschr., 1909, 22. 
2 Arch. f. exper. Path. u. Pharmakol., 1913, Ixxiv, 14. 
* Jour. Exper. Med., 1914, 19, 459. 4 Ztschr. f. Biol., 1907, 279. 232 
FERMENTS AND ANTI FERMENTS 
It was apparently thus established that the body cells are harmonically 
attuned to one another, and if new or modified cells or their products are 
brought into relation with other cells, they are received as foreign invaders, 
and their entrance is followed by the production of wdiat Abderhalden has 
called “protective ferments” (“Abwehrfermente”) capable of bringing 
about their cleavage into simpler products. In this manner the presence 
in the circulation of some of the body cells may give rise to the production 
of these ferments if the cells in question are really foreign to the blood- 
plasma and other cells. 
Abderhalden has also stated that although he was led to make these 
investigations on the supposition that syncytial elements wrere present in 
the blood of pregnant women, it is not necessary that they be constantly 
in the blood, for every case of pregnancy has a complicated protein metabo- 
lism and there is a general exchange of substances between the placenta 
and the maternal blood that permits the entrance into the latter of protein 
products that have not been broken down completely into amino-acids, 
and that cause the organism to produce defensive proteolytic ferments. 
In cancer, where the production of new cells is so marked, some of these 
cells or their products may easily be swept into the general circulation, 
where they act as foreign invaders and cause the formation of protective 
proteolytic ferments. It is a noteworthy fact, moreover, that the serum 
of carcinoma cases reacts best with carcinoma cells and that of sarcoma 
with sarcoma cells. 
Similar ferments have been described in other conditions. Fauser has 
demonstrated that the blood-serum of dementia praecox patients contains 
ferments that act on the reproduction glands, so that the serum of males 
reacts with testicular extracts and that of females with ovarian extracts. 
These serums were, however, also found to react with thyroid tissue and 
brain cortex. In general paresis reactions were obtained with brain cortex 
and liver, also at times with thyroid gland, reproductive glands, and more 
rarely with kidney. 
Abderhalden has found ferments for the tubercle bacilli in the blood- 
serum of tuberculous persons, and they have also been found in the blood- 
serum of syphilitics for the Treponema pallidum, either in pure culture 
or in organs containing large numbers of the parasites; Smith1 has found a 
remarkable specificity of the ferments produced in rabbits following im- 
munization with typhoid and paratyphoid bacilli and cocci. 
The work of Abderhalden has been severely criticized and the existence 
of his “protective ferments” in pregnancy and disease denied. An enormous 
literature has accumulated on the application of his dialysis method in 
the diagnosis of pregnancy which has unfortunately detracted from interest 
and investigation in the fundamental principles. 
Some investigators have denied the existence and specificity of these 
“ferments.” However, that there may be an increase of true proteolytic 
ferments in pathologic conditions has been clearly established by Jobling 
and his associates as discussed above. The question is whether or not the 
“ferments” of Abderhalden are true ferments or a kind of antibody wdiose 
nature is unknown or which may be classed with the sensitizers acting with 
complement. 
The researches of Van Slyke and his associates2 employing Van Slyke’s 
method of amino-nitrogen determination, have shown that practically 
every serum, whether from a pregnant or non-pregnant female, or from a 
1 Jour. Infect. Dis., 1916, 18, 14. 
2 Jour. Biol. Chem., 1915, 23, 377. FERMENTS VS. ANTIBODIES IN DISEASE 
233 
male, gave protein digestion when incubated with placenta tissue prepared 
according to the method of Abderhalden; further evidence of non-specificity 
was seen in the fact that carcinoma tissue was digested apparently to about 
the same extent as was placenta. Taylor and Hulton,1 working with the 
protanin of salmon, found that normal serum and the serum of immunized 
animals gave about equal degrees of digestion, leading them to conclude 
that there is at present no reason for believing that the normal hydrolysis 
of the protein of the body occurs in the circulating blood, but that these 
metabolic changes presumably belong to the tissue cells. 
The investigations of Sloan,2 however, have shown that the serum in 
pregnancy may yield positive Abderhalden reactions without showing any 
evidences of digestion by Van Slyke’s method. In the opinion of Sloan, 
the latter method is not applicable for determinations of ferment activity 
in which a non-soluble, moist, complex protein substance is incubated with 
serum in a test-tube, but without dialysis. My own experiments with Asnis 
and Frees agree with these findings. Sloan also found that the serum of 
animals injected with a suspension of placental cells did not give rise to 
an increase of ferments, but that the injection of split products of placental 
cells was followed by an increase of proteolytic ferments. 
The work of Bronfenbrenner and his associates3 has also shown quite 
clearly that changes occur in the serum during pregnancy, with the produc- 
tion of a “ferment” or an antibody possessing a high degree of specificity. 
Too much excellent and careful work has been done to discard the fact 
that these substances may be found in the serum in pregnancy and patho- 
logic conditions, although it is now equally well established that the nature 
of these substances and the mechanism of their activity are probably en- 
tirely different from the conceptions of Abderhalden. 
Since these so-called “ferments” may not show their presence in chemical 
tests which are satisfactory for detecting the presence of such true ferments 
from leukocytes and serum as the proteases, the question of their nature, 
that is, whether they are true ferments or antibodies of the nature of ambo- 
ceptors or cytolysins requiring the presence and activity of complement, 
becomes one of much interest and importance. 
Unfortunately, this question cannot be definitely answered. Pearce 
and Williams4 were of the opinion that the “ferments” were not cytolysins, 
and Abderhalden has strenuously denied that his “ferments” and cytol- 
ysins were identical without, however, bringing forward conclusive proof. 
The experiments of Stephan,5 Hauptmann,6 Bettencourt and Menezes,7 
and Bronfenbrenner, however, have shown that complement bears an 
important relation to the activity of the so-called “protective ferments,” 
which indicates that the latter may be of the nature of amboceptors. It 
would appear that in the light of our present knowledge at least two con- 
clusions may be drawn on the nature of these so-called “protective fer- 
ments”: 
1. That there is no evidence of the production of true and specific fer- 
ments in pregnancy and disease in the meaning of Abderhalden. On the 
other hand, there may be an increase of such leukocytic and serum ferments 
as the proteases in at least some pathologic conditions. 
1 Jour. Biol. Chem., 1915, 22, 59; ibid., 1916, 25, 163. 
2 Amer. Jour. Physiol., 1915, 39, 1; ibid., 1916, 42, 558. 
3 Jour. Exper. Med., 1915, 21, 211. 
4 Jour. Infect. Dis., 1914, 14, 351. 
5 Munch, med. Wchn., 1914, 801. 
G Miinch. med. Wchn., 1914, 1167. 
7 Compt. rend. Soc. de biol., 1914, 162. 234 
FERMENTS AND ANTIFERMENTS 
2. That in pregnancy and certain diseases there may be produced more 
or less specific amboceptor-like antibodies demonstrable by Abderhalden’s 
methods, but not by ordinary chemical methods. 
Mechanism of the Abderhalden Reaction; Bronfenbrenner’s Theory.— 
Aside from the probable clinical value of the methods devised by Abder- 
halden in the serum diagnosis of pregnancy and various pathologic condi- 
tions, as malignancy, tuberculosis, lesions of the nervous system and duct- 
less glands, most interest concerns the question of the specificity of the 
“ferments” or antibodies concerned and the mechanism of their action. 
While Abderhalden and many of his pupils have claimed a high degree 
of specificity for the “protective ferments” and his pregnancy reaction, 
claiming from the beginning that errors of technic were largely respon- 
sible for the failure of others to obtain satisfactory results, the dialysis 
test as now conducted is not especially diflicult, and sufficient work has 
been done by other investigators who have followed Abderhalden’s technic 
with great care and exactness to give warrant to the claim that other factors 
aside from those purely technical may be responsible for the divergent 
and non-specific results obtained. 
As previously stated, Abderhalden bases his theory concerning the 
“protective ferments” upon the specific digestion of a substrat by specific 
ferments, claiming that these ferments are separate and distinct antibodies, 
and not to be classed with the cytolytic amboceptors or cytolysins of Ehrlich. 
That the substrat in the pregnancy test is a boiled tissue would seem 
to impair the specificity of the reaction and, indeed, certain physical factors, 
as the mechanical state of division of the substrat and its facility for acting 
as an absorbent in a purely mechanical capacity, likewise appear to be 
factors in the reaction on the basis of numerous investigations, showing 
that loose areolar placental tissue is frequently digested by normal sera 
and various pathologic sera irrespective of pregnancy, whereas digestion 
of a firm and compact tissue as that of malignant tumors is much less con- 
stant. In this connection the work of de Waele1 has a bearing, inasmuch 
as he found that any agent which would cause an alteration of the physical 
state of the serum globulins would cause an intense Abderhalden reaction, 
concluding that the reaction depended upon a globulinolysis having an 
origin in physical processes probably analogous to the precipitin reaction. 
While immunologic as well as chemical and physical reactions are more 
or less dependent upon quantitative factors, investigations by Flatow2 and 
Herzfeld,3 Plaut,4 and others show that while specific results may be obtained 
by proper manipulation of the material, non-specific results in either a nega- 
tive or positive reaction may be obtained with practically any serum, how- 
ever well controlled, with the same material. These investigations are sig- 
nificant not so much because of quantitative factors alone as they are by 
reason of indicating that pregnancy reaction is dependent upon the prin- 
ciples of mechanical absorption on the part of the substrata of something 
from the serum followed by a digestive process, rather than upon the simple 
digestion of a specific substrata by a specific ferment. 
For this conception of the mechanism of the Abderhalden reaction the 
investigations of Jobling and Peterson5 discussed above have been funda- 
mental and of great interest and importance. As previously stated, they 
1 Ztschr. f. Immunitatsf., orig., 1914, 12, 170. 
2 Munch, med. Wchnschr., 1914, lxi, 468; ibid., 608; ibid., 1168. 
3 Biochem. Ztschr., 1914, lxiv, 103. 
4 Miinch. med. Wchnschr., 1914, lxi, 238. 
6 Jour. Exper. Med., 1914, xix, 459 and 480. MECHANISM OF THE ABDERHALDEN REACTION 
235 
have shown that the digestive power of a serum is dependent upon non- 
specific proteolytic ferments or proteases normally present and held in 
check by an antiferment which, according to their work, is believed to 
reside in the unsaturated fatty acids of the serum. Upon removal of the 
antiferment by means of lipoidal solvents or saturation with various organic 
and inorganic substances, as boiled tissue, iodin, starch, kaolin, and the 
like, protease activity is released, followed by digestion, not of the so-called 
substrata but of the protein of the serum. Likewise, Plaut,1 Peiper,2 Fried- 
man and Schonfield,3 and Bronfrenbrenner4 have obtained positive Abder- 
halden reactions with guinea-pig and human sera, not only with placental 
tissue but also with such inert substances as kaolin, starch, barium sul- 
phate, chloroform, etc. These studies would, therefore, tend to show that 
the boiled placental tissue in Abderhalden’s reaction is not digested, but 
acts simply as an absorbent in a purely mechanical manner. 
Heilner and Petri5 and de Waele6 found the ferments in the blood-serum 
so quickly after the parenteral introduction of the protein, at intervals hardly 
sufficient for the elaboration of new and specific ferments, as to support the 
theory that the ferments are preformed and that the substrata serves to ac- 
tivate these rather than bring about the production of new ferments. 
It would appear, therefore, that the original theory of Abderhalden is 
untenable in that specific proteolytic ferments in the blood are not pro- 
duced during pregnancy, and that the Abderhalden reaction is not due to 
the digestion in vitro of specific antigen by specific ferments. 
As indicated by the researches mentioned above and especially the 
work of Bronfenbrenner, it is now generally accepted that the placental 
tissue employed in Abderhalden’s test simply removes the antiferment of 
the serum which is followed by the digestion of the serum proteins rather 
than of the placental tissue, by the non-specific proteases of the serum. 
During pregnancy, the infectious diseases, in cancer, cachexia, and other 
diseased states there may be an increase of proteolytic ferments of the 
serum, but these ferments are non-specific. 
The question then arises why the Abderhalden test properly conducted 
with placental tissue yields more ninhydrin reacting substances than when 
the test is conducted with kaolin, charcoal, or some other tissue? In so far 
as my own experiments with Williams7 are concerned, I have invariably 
found that the reactions in pregnancy were stronger when the tests were 
conducted with pregnant serum and placental tissue. A careful study 
of the Abderhalden dialysis reaction in Wells’ laboratory by Elsesser,8 
using Osborne’s purified vegetable proteins, showed that in spite of many 
atypical, irregular, and illogical results, “there is an obvious tendency for 
a substrata to react more often and yield stronger reactions when tested 
against its homologous immune serum, than when tested against a heterol- 
ogous immune serum.” 
The investigations of Bronfenbrenner9 apparently offer a satisfactory 
explanation. As previously stated his studies indicate, but do not definitely 
prove, that the so-called “defensive ferments” are antibodies of the ambo- 
1 Munch, med. Wchnschr., 1914, lxi, 238. 
2 Deut. med. Wchnschr., 1914, xl, 1467. 
3 Berl. klin. Wchnschr., 1914, li, 348. 
4 Proc. Soc. Exper. Biol, and Med., 1914, xi, 90. 
6 Munch, med. Wchnschr., 1911, lx, 1530. 
6 Ztschr. f. Immunitatsf., orig., 1914, xxii, 31. 
7 Amer. Jour. Obstet., 1915, lxxii, No. 1. 
8 Jour. Lab. and Clin. Med., 1915, 1, 79. 
9 Jour. Infect. Dis., 1916, 19, 655. 236 
FERMENTS AND ANTI PERM ENTS 
ceptor class. These antibodies contained in the serum are believed to sensi- 
tize the placental tissue antigen just as hemolytic amboceptors sensitize 
homologous corpuscles. The sensitized placental antigen then serves to 
remove the antiferment of the serum which releases the non-specific proteo- 
lytic ferments followed by autodigestion of the serum rather than of pla- 
cental tissue. 
According to this theory the process is both non-specific and specific 
—non-specific in that the plain placental tissue may absorb some anti- 
ferment and partially release proteolytic ferments of the serum, and specific 
in that the antibody of a pregnant serum sensitizes the placental antigen 
and this sensitized antigen removes antiferment by inducing a change of 
the colloids of the serum followed by autodigestion of the serum, and the 
production of ninhydrin reacting substances. According to Bronfen- 
brenner, the sensitized antigen will remove antiferment from any serum, 
pregnant or non-pregnant female, or, male serum, followed by autodigestion; 
on the bases of this theory he has devised a modification of the Abderhalden 
test to be described later. 
ANTITRYPSIN TEST 
The antitryptic activity of blood sera has been found to vary in patho- 
logic conditions and tests for measuring variation in antitryptic activity 
have been advocated as diagnostic procedures. 
Brieger1 originally asserted that 95 per cent, of cases of cancer evinced 
a marked increase in the antitryptic activity of the serum. He subse- 
quently found that in a large number of other conditions, including both 
acute and chronic wasting diseases accompanied by cachexia, similar changes 
may occur. Further investigations have confirmed these findings indicating 
that the change in the serum is not to be regarded as characteristic of new 
growths, as it occurs in too many other pathologic conditions and even in 
physiologic conditions, to have the value of a specific symptom. On the 
other hand, the absence of the antitryptic reaction in the blood may be 
taken generally as evidence against the existence of cancer. Weil2 has re- 
viewed the literature on this subject and given the results of his own experi- 
ences which are largely in accord with the above statements. 
Various methods for measuring antitryptic activity of sera have been 
described, those of Bergmann and Meyer,3 Muller and Jochmann,4 Fuld 
and Goss5 being best known. Of these tests that of Fuld and Goss is re- 
garded best, the technic being as follows: 
Solution of Trypsin.—This is made by dissolving 0.5 gm. of pure trypsin 
(Griibler) in 50 c.c. of NaCl solution and adding 0.5 c.c. of normal soda solu- 
tion; make up to 500 c.c. with physiologic salt solution. 
Casein Solution.—Dissolve 1 gm. of pure casein in 100 c.c. of decinormal 
sodium hydroxid solution with the aid of gentle heat. Neutralize to litmus 
with n/10 hydrochloric acid solution and dilute with physiologic salt solu- 
tion up to 500 c.c. Filter and sterilize in an Arnold sterilizer. Preserve 
in the refrigerator. 
Acetic Acid Solution.—To 5 c.c. of acetic acid (c. p.) add 45 c.c. of 
absolute alcohol and 50 c.c. of distilled water. 
The patient’s serum must be fresh, and should be diluted twenty times 
with salt solution. Dose, 0.2 c.c. 
Technic.—A titration of the trypsin solution must precede the test proper. 
1 Berl. klin. Wchn., 1908, 1349 and 2260. 
2 Amer. Jour. Med. Sci., 1910, 139, 714. 
3 Berl. klin. Wchn., 1908, No. 37. 
4 Miinch. med. Wchn., 1909, Nos. 29 and 31. 
5 Archiv. f. exp. Path., 1907, 137. ABDERHALDEN’S TESTS 
237 
Into each of several small test-tubes place increasing amounts of trypsin 
solution, as, for example, 0.1, 0.2, 0.4, 0.6, 0,8, and 1.0 c.c. Add 2 c.c. of 
the casein solution to each tube; shake carefully and place in an incubator 
or water-bath for half an hour at 50° C. Then add 3 or 4 drops of the acetic 
acid solution to each tube, and observe which tube first shows cloudiness 
after a few minutes. The tube containing the smallest amount of trypsin 
and which remains perfectly clear contains enough trypsin fully to digest 
the 2 c.c. of casein solution. 
Into each of six small test-tubes now place 0.2 c.c. of the 1 : 20 dilu- 
tion of the patient’s serum, and increasing amounts of the trypsin solu- 
tion, beginning with the completely digesting dose, as determined above, 
and increasing by 0.1 c.c. Add 2 c.c. of casein solution to each tube, and 
bring all tubes to a like volume by the addition of normal salt solution. 
Shake gently and incubate at 50° C. for half an hour. Add several drops 
of acetic acid solution to each tube, and again observe the tube containing 
the smallest amount of trypsin in which cloudiness can be seen. Thus the 
amount of trypsin neutralized by the antitrypsin of the serum is determined. 
For example, in an experiment the preliminary titration showed that 
0.5 c.c. of trypsin completely digested the casein. In the second part of 
the test the lower limit of trypsin was this 0.5 c.c. increased by 0.1 c.c. in 
successive tubes up to 1 c.c. It is now found that 1 c.c. of the trypsin solu- 
tion is required to bring about the complete digestion of the casein in the 
presence of the serum, or 1 c.c. — 0.5 c.c. = 0.5 c.c., which is the amount 
of trypsin neutralized by 0.01 c.c. of undiluted serum. 
A control experiment is conducted with the pooled serum of several 
normal persons, and a comparison of the value thus obtained shows whether 
the antitryptic power of the serum tested is altered. 
Robertson and Hanson1 have recently described a simple, accurate modi- 
fication of this method for measuring the antitryptic indices of blood-sera. 
The method of Marcus,2 which is a modification of the method of Muller 
and Jochmann, is described in the laboratory exercises on Experimental 
Infection and Immunity. This technic, however, is not as accurate as that 
of Fuld and Goss, described above. Both egg-albumens and the coagulated 
sera of different animals vary considerably in digestibility, so that there is not 
the required constant basis of comparison. Furthermore, the visual apprecia- 
tion of a minute depression on the surface of a serum plate is a very difficult 
and inexact procedure. 
ABDERHALDEN'S TESTS3 
Methods.—Two methods have been devised by Abderhalden for the 
demonstration of antibody or the “ferments” in the blood-serum of preg- 
nancy, cancer, and other conditions: 
1. The Dialyzation Method.—Specially prepared and coagulated placenta 
and fresh serum are placed in a dialyzing capsule so prepared that it will 
permit the passage of peptones and amino-acids only. The filled capsule 
is placed in sterile distilled water, and incubated for from sixteen to twenty- 
1 Jour. Immunology, 1918, 3, 131. 2 Berl. klin. Woch., 1908, No. 4; 1909. 
3 Abderhalden: Abwehrfermente des tierischen Organisims, Julius Springer, 3d ed., 1913. 
The author realizes that he is open to criticism for devoting space to a description of Abder- 
hafden’s methods, since the reaction has been proved to lack reliable diagnostic value in 
pregnancy at least. On the other hand, I doubt the wisdom of “scrapping” the immense 
literature (some of which is thoroughly reliable and acceptable) and dropping the subject; I 
still believe that immunologic changes may occur in the serum in pregnancy, and that, while 
Abderhalden’s theory and methods may be wrong, the subject is worthy of more investiga- 
tion. It is for this reason that one of his methods is described for the aid that may be given 
students of the subject. 238 
FERMENTS AND ANTIFERMENTS 
four hours, when the dialysate is tested by the biuret or ninhydrin test for 
peptones and amino-acids. Under proper conditions the presence of these 
substances indicates a positive reaction. 
2. The Optical Method.—This method is based upon the same principle 
as the dialyzation method. Into the tube of a polariscope place a solution 
of placental peptone and the serum to be tested. Warm the mixture to 
37° C., and after an hour note the degree of rotation and record it; repeat 
this at intervals during the following twenty-four to forty-eight hours. 
If the serum contains the antibody or “ferment,” the peptone is split into 
amino-acids and the degree of rotation 
increased from 0.05° to 0.5° C. and higher. 
This method requires an expensive polar- 
iscope, considerable practice in making 
the observations and readings, and ds 
only reliable in skilful hands. 
l The Dialyzation Method 
Testing the Dialyzing Shell. — The 
quality of the dialyzing shell largely de- 
termines the success of this method. It 
must fulfil two requirements: 
1. It must be absolutely non-perme- 
able for albumin. 
2. It must be evenly permeable for 
the protein cleavage products, such as 
peptones, polypeptids, and amino-acids. 
Special shells are made by Schleichter 
and Schull, No. 579a being recommended 
at the present time. The shells must be 
of correct size, and every one must be 
tested before being used. If a shell al- 
lows uncleaved protein to pass through, 
then all reactions would react positively 
regardless of the presence or the absence 
of the specific ferment. If the shell is too 
thick and too tight, and prevents the 
passage of peptones and amino-acids, 
then all reactions would be negative, 
even though the ferment were present in 
the serum and had digested the placental 
protein. Accordingly, each shell must be 
tested and standardized, and only those em- 
ployed that have proved satisfactory. 
Glassware.—It is highly important that all glassware should be free 
from clinging particles or traces of albumin, acids, and alkalies. Pipets 
and dialyzing cylinders should be washed in water, alcohol, ether, and 
finally in distilled water, and sterilized by dry heat. Boiling rods of solid 
glass (10 by 0.5 cm.) should be washed in alcohol, ether, and distilled 
water, wrapped in bundles of six in newspaper, and sterilized by dry heat. 
A very convenient dialyzing cylinder is showm in the accompanying 
illustration (Fig. 93). This cylinder measures 8 by 3 cm. It should be 
plugged with cotton and sterilized. When the shell is loaded wdth coagu- 
lated placenta and serum and covered with toluol, it will rest well beneath 
Fig. 93.—A Dialyzing Cylinder for the 
Abderhalden Ferment Test. 
The shell contains placental tissue and 
fresh serum; it is surrounded with 20 c.c. 
of distilled water and covered with toluol. 
The cotton plug prevents contamination. 
The cylinder is readily sterilized in a hot- 
air oven and affords a simple and efficient 
means for conducting the test by the 
dialyzation method. Fig. 94.—Ninhydrin Reaction (Abderhalden Ferment Test). 
The tube on the left shows a positive reaction with the serum of a pregnant woman; the tube on 
the right is the serum control and shows a faint violet color, due, presumably, to the passage of dia- 
lyzable substances in this serum. THE DIALYZA TION METHOD 
239 
the surface of the outside distilled water. The wide mouth of the cylinder 
facilitates all manipulations and the shell cannot upset. The apparatus 
is easily sterilized, and the cotton plug prevents bacterial contamination 
and undue evaporation of the contents. 
General Precautions.—According to Abderhalden, the work should be 
conducted in a special room, where there is no dust or fumes of acids, and 
where no bacteriologic work is in progress. This observer also recommends 
that a special incubator be used for this work. If, however, the working 
table is scrupulously clean and the glassware is clean and sterile, and if the 
shells are handled with sterile forceps and the dialyzing cylinder is stoppered 
with a plug of sterile cotton, all requirements are practically fulfilled. 
Reagents.—The presence of albumin or its split products may be de- 
tected by two color reactions: (1) the biuret reaction, and (2) the nin- 
hydrin reaction. The first is especially delicate for uncleaved albumin, 
and the latter for peptones and amino-acids. The technic of the biuret 
test is described with the technic of testing shells for permeability to albumins. 
Ninhydrin.—This is the trade name for triketohydrindenhydrate. It 
is a whitish yellow, readily soluble powder, dispensed in brown glass vials 
containing 0.1 gm. of the drug. A circular describing its method of use 
accompanies each package. As 0.2 c.c. of a 1 per cent, watery solution 
is the amount necessary for a test, the contents of the vial are dissolved 
in 10 c.c. of distilled water, and the vial rinsed with a portion of the solvent. 
This solution should be preserved in a brown bottle in a cold place, and 
precautions taken to prevent infection. Triketohydrindenhydrate has been 
described by Ruheman,1 who gives its formula as follows: 
C6H4<^^>C(OH)2 
Owing to the fact that it gives a blue color in the presence of any com- 
pound that possesses an amino-group in the alpha position of the carboxyl 
group, it is of great value as an aid in recognizing the products of protein 
digestion (Fig. 94). 
Testing the Shell for Non-permeability to Albumin.—1. New shells 
should be softened by soaking them for half an hour in sterile distilled 
water. A dozen or more may be tested at one time. 
2. The albumin solution is prepared by placing 5 c.c. of the albumin 
of fresh eggs in a mixing cylinder, and adding distilled water to make 100 
c.c. Mix well. There must be no flakes. Instead, a clear, hemoglobin- 
free serum which has been dialyzed against running water to remove dia- 
lyzable substances, may be used in doses of 2.5 c.c. for each shell. 
3. Carefully pipet 5 c.c. of the albumin solution into each shell. Great 
care should be exercised that none of the solution contaminates the outside 
of the shell. The preferable method is to hold the shell with a pair of broad- 
toothed sterilized forceps and carry the pipet to the bottom, in order that 
none of the albumin should contaminate the upper portion of the inside 
of the shell. The pipet may easily touch the edge of the shell and thus 
contaminate the dialysate. If in doubt, cover the upper end of the shell 
with the forceps and wash the outside with running water. 
4. The loaded shell is now placed in a sterile dialyzing cylinder con- 
taining 20 c.c. of sterile distilled water. Never load the shell in this cylinder, 
for some of the albumin may fall into the distilled water. 
5. Cover the contents of the shell and the surrounding distilled water 
1 Jour. Chem. Soc., London, 1910, xcvii, 2025. 240 
FERMENTS AND ANTIFERMENTS 
with a layer of toluol about \ inch in depth. Replace the cotton plug in 
the cylinder. 
6. Incubate at 37° C. for sixteen hours. 
7. Pass a sterile pipet quickly through the layer of toluol and remove 
10 c.c. of the dialysate to a clean sterile test-tube, and test for albumin by 
the biuret reaction. Add 2.5 c.c. of a 33 per cent, solution of sodium hy- 
droxid; shake gently, but remove the thumb from the top of the tube. 
The solution may become slightly cloudy. Carefully overlay with 1 c.c. 
of a 0.2 per cent, solution of copper sulphate in such manner that a sharp 
line of demarcation separates the alkaline dialysate from the copper sul- 
phate solution. A delicate violet tint at this line indicates that albumin 
is present, and that the shell is useless. If one cannot see this color or is 
in doubt, it is well to make the ninhydrin test. To do this dialysis should 
be continued for twenty-four hours; ninhydrin reacts with albumin in 
addition to peptones and amino-acids, but according to Abderhalden, this 
test is less sentitive than the biuret test. 
8. All shells should react negatively, i. e., they should not permit the 
passage of unchanged albumin. If the ninhydrin test is used, the tubes 
should be inspected one-half hour after boiling, and the contents should 
be as clear as water or show but the faintest blue tint. If this is not the 
case, shells should be discarded as being permeable to albumin. Those 
that are satisfactory in this respect should be tested further as follows: 
Testing the Shell for Permeability to Peptone.—1. The shells should 
now be thoroughly cleansed, but not with a stiff brush, washed in running 
water, and boiled for thirty seconds. 
2. Prepare a 1 per cent, solution of silk peptone (Hochst) in distilled 
water, and carefully pipet 2.5 c.c. into each shell, using every precaution 
against contaminating the upper portion on the inside, and especially of 
the outside, of the shell. 
3. Place the loaded shell in a sterile dialyzing cylinder containing 20 
c.c. of sterile distilled water, and cover the contents of the shell and water 
with toluol. Replace the cotton plug and incubate at 37° C. for twenty- 
four hours. 
4. Remove 10 c.c. of the dialysate (avoid removing toluol) to a clean, 
sterile, thin-walled test-tube, and add 0.2 c.c. of the 1 per cent, ninhydrin 
solution. Insert a sterile boiling rod and boil for exactly one minute. 
5. The boiling process is quite an important feature of this test. Al- 
ways boil in precisely the same manner. A high Bunsen flame should be 
used, and about one minute after air-bubbles first appear on the sides of 
the tube lively boiling commences. The flame should then be turned down 
and the boiling continued for exactly one minute. 
6. Place the tube in a rack. With a fresh sterile pipet remove 10 c.c. 
of dialysate from the next cylinder and test in the same manner, and repeat 
until the entire series have been finished. 
7. After half an hour inspect all the tubes; they should show a deep 
blue color; if they do not do so they are impermeable or partly permeable 
to peptone and should be discarded. There is usually a difference in the 
degree of color reaction among a number of shells, as their permeability 
varies. 
8. Those shells that have withstood both tests are now thoroughly 
washed in running water, boiled for thirty seconds, placed in a jar of sterile 
distilled water containing a few drops of chloroform, and covered with 
toluol. From this time on they should not be handled with the fingers, 
but only with forceps that have been sterilized by boiling. Of the entire THE DIALYZATION METHOD 
241 
number of shells, usually from 20 to 30 per cent, or more are found to be 
unsatisfactory. 
Preparation of the Placental Tissue.—This is the substratum, and 
should consist of coagulated placental protein free from dialyzable sub- 
stances that react with ninhydrin. 
1. A fresh normal placenta should be prepared soon after delivery. It 
is highly important to wash it free from all blood, Abderhalden having laid 
considerable stress upon this point. He explains that in the blood of all 
animals there is always a specific ferment for the red blood-corpuscles, 
as even the smallest hemorrhage into the tissue calls forth a protective 
ferment. For this reason all organs that contain blood may contain the 
substratum and ferment, and yield false positive reactions. 
2. The placenta should be placed in warm water and freed as far as 
possible of clots. The membranes and cord are removed, and the placental 
tissue cut into pieces about the size of a dime. These are placed in a sieve 
under running water, and each piece squeezed with the hand. From time 
to time the entire mass is thoroughly squeezed out in a towel. Tissues 
that cannot be freed from clots should be discarded. The tissues are now 
crushed in a mortar, connective-tissue strands removed, and the washing 
continued until the tissue is snow white. Decolorizing substances, such as 
H2O2, should not be used. If the tissue is not white and free from blood 
it should not be employed. Liver, spleen, and kidney tissue cannot be 
made perfectly white, although all traces of blood have been removed. 
3. Place 100 times as much distilled wrater as there is tissue in an en- 
ameled vessel; to each liter add 5 drops of glacial acetic acid and heat to 
boiling when the tissue is added and boiled for ten minutes. 
4. Wash the coagulated tissue wTith distilled water, and boil again with- 
out the addition of acid. This should be repeated six times in succession. 
If an interruption occurs, cover the tissue and water with a layer of toluol. 
5. After the sixth boiling add a small quantity of water to the tissues 
—just sufficient to enable it to boil for about five minutes without burning, 
for the water is now to be tested with the ninhydrin reaction and it is im- 
portant that this be as concentrated as possible. Filter the water, and to 
5 c.c. in a sterile test-tube add 1 c.c. of the ninhydrin solution. Boil vigor- 
ously for one minute. If there is the slightest discoloration within half an 
hour, the tissues must be boiled again, but with only five volumes of water 
and no longer than five minutes each time. These boilings should be re- 
peated as often as is necessary until the ninhydrin reaction remains water 
clear for at least one-half hour. 
6. The tissues are again gone over with a sterile forceps, and a search 
made for brown masses resembling blood-clots. These are to be discarded. 
7. The tissue is now preserved in a sterile jar containing sufficient sterile 
water and chloroform and covered with toluol. All tissue should be handled 
with sterile forceps, and when once removed from the jar, they should 
never be returned. The whole operation requires several hours and it 
should be conducted without interruption. If the process is interrupted, 
the tissue should be covered with a layer of toluol. 
8. It is well to try out the tissue with a known serum of pregnancy to 
make certain that it is a suitable substratum. 
9. Only normal placenta should be used, as in certain instances a normal 
organ may be satisfactory, whereas a diseased organ would be unsuitable. 
10. Animal placenta may be substituted for human placenta and vice 
versa, but Abderhalden cautions against this substitution until further 
work has been done. 242 
FERMENTS AND ANTIFERMENTS 
The Blood-serum.—The serum to be tested must fulfil three conditions: 
(1) It must contain the smallest amount of dialyzable substances that 
would react with ninhydrin. Blood is best drawn in the morning before 
breakfast. In all diseases accompanied by marked protein disintegration, 
such as cancer, the blood-serum may contain large amounts of dialyzable 
substances. 
(2) It must be absolutely free from hemoglobin and clear. 
(3) It must be free from cells. Even an apparently clear serum may 
contain millions of erythrocytes. 
1. From 10 to 20 c.c. of blood are withdrawn from a vein at the elbow 
with a dry sterile needle into a sterile centrifuge tube. This is placed aside 
at room temperature for several hours, when sufficient serum has usually 
separated out; if this has not occurred, centrifuge for several minutes. 
The serum is removed to a second sterile centrifuge tube, and centrifuged 
at high speed for several minutes until all corpuscles have been precipitated 
to the bottom of the tube. 
2. The serum should be used within twelve hours after the blood has 
been withdrawn. Abderhalden claims that heating a serum to 60° C. robs 
it of its digesting powers. Pearce and Williams have found that inactiva- 
tion considerably weakens the reaction, but does not abolish it altogether. 
3. Specimens of blood sent through the mails are really unsatisfactory 
for even if they are delivered within twelve hours after bleeding the amount 
of handling has usually resulted in the breaking up of a number of corpuscles, 
and the tingeing of the serum with hemoglobin. 
The Test.—1. Absolute cleanliness should be employed. The glass- 
ware should be sterile and dry, and everything should be in readiness. 
The technic should be aseptic and thoroughly understood. 
2. Remove a sufficient amount of the prepared placenta for the work at 
hand with sterile forceps and wash in a dish of sterile distilled water to 
remove toluol and chloroform. Boil with 4 or 5 volumes of sterile distilled 
water for two minutes and test the water with ninhydrin. If positive, 
the tissue must be boiled as described above until free of ninhydrin reacting 
substances. Place on sterile filter-paper and squeeze to remove any excess 
of water. Weigh and place 0.5 gm. in each of twro shells (one for a control). 
3. Holding each shell with a second pair of boiled forceps, pipet 1.5 c.c. 
of the patient’s serum into one shell containing placenta, and the same 
amount into a third shell which is to serve as a control on the serum. Place 
1.5 c.c. of sterile distilled water in the placental tissue control shell. 
4. Unless one is absolutely sure that neither the tissue nor the serum 
has touched the outside of the shells they should be held shut with sterile 
forceps and washed with sterile distilled water. 
5. Each of the three shells is now placed in cylinders containing 20 c.c. 
of sterile distilled water. Under no circumstances are the shells to be loaded 
while they are in the dialyzing cylinders. 
6. The contents of each shell and the water surrounding them are covered 
with a layer of toluol about £ inch in depth, and the cylinders plugged with 
cotton to prevent evaporation and contamination. The shell should be at 
least | to | inch above the level of the outside fluids, and due care must 
be exercised in carrying the cylinder back and forth from the incubator 
that the contents of the shell and the surrounding water do not become 
mixed. 
7. If it is at all possible, it is well to set up two more shells as controls, 
each containing placenta and normal serum and the serum of pregnancy 
respectively. THE DIALYZATION METHOD 
243 
8. All the cylinders are incubated at 37° C. for twenty-four hours. 
Ten c.c. of the dialysate are then removed from each tube with a separate 
sterile pipet and placed in sterile test-tubes of the same size and boiled with 
0.2 c.c. of the 1 per cent, ninhydrin solution for exactly one minute. After 
standing for half an hour the readings are made. 
Reading the Reaction.—The dialysate of the serum alone should be 
clear as water or show but the faintest blue tinge. The dialysate of the 
placenta alone should be clear; the dialysate of the patient’s serum plus 
that of the placenta may show a deep violet-blue color when the reaction 
is strongly positive, or a fainter blue when it is weakly positive. If this 
dialysate is water clear or has a faint blue color, comparable to the controls, 
the result is negative. If there is any doubt the test should be repeated. 
The negative control should be water clear or have a faint tinge comparable 
to its control. The positive control should show a deep violet-blue color. 
I generally control the result given by the shell containing tissue and 
patient’s serum by cleansing it thoroughly, boiling for a minute, and test- 
ing it with egg-albumen solution or a serum in case the reaction was posi- 
tive, to make sure that the shell has not allowed the passage of serum, 
or with peptone solution in case the reaction was negative, to make sure 
that it was not thick enough to block the passage of peptones and amino- 
acids. This procedure delays the report on a serum for another twenty- 
four hours, but the greater accuracy obtained warrants the delay. 
Readings should never be made by artificial light. Tubes should be 
held against a white background the better to appreciate the color changes. 
A pinkish- or brownish-yellow discoloration has nothing to do with the 
ninhydrin reaction. 
Sources of Error in the Dialyzation Method.—There are many sources 
of error, and until the technic has been improved sufficiently to eliminate 
these, Abderhalden’s directions should be followed minutely. 
1. The shells may become spoiled in time. They should not be cleansed 
with rough brushes or boiled too long. They should be cleansed at once 
after using, and tested every four weeks. If a wrong diagnosis results the 
shell should be retested at once. 
2. The placental tissue is an important source of error, due to the fact 
that it contains blood. 
3. The serum should be fresh and free from hemoglobin and corpuscles. 
4. The controls on placenta alone and each serum alone are absolutely 
necessary, as both may contain various substances capable of reacting 
with ninhydrin and thus yielding false positive reactions. 
5. The water used should be distilled and sterile. The glassware should 
be chemically clean and sterile, and the laboratory free from the fumes 
of acids and alkalies. It is very important that absolutely the same condi- 
tions should exist for the control tests as for the main test itself. 
Bronfenbrenner’s Modification.—Bronfenbrenner, Schlesinger, and 
Mitchell1 have sought to improve the test described above by some modifica- 
tion in technic embracing the principle that serum may contain an antibody- 
like substance capable of sensitizing the antigen, and that this sensitized 
antigen removes antiferment by producing a change in the colloids of the 
serum, followed by autodigestion (of the serum), and the appearance of 
ninhydrin reacting substances. The test may be conducted as follows: 
1. Serum is collected and placental tissue or other antigen are prepared 
as described above. 
2. Fresh serum (1.5 c.c.) and 0.5 gm. of antigen are placed in a sterile 
1 Jour. Amer. Med. Assoc., 1915, 65, 1268. FERMENTS AND ANTIFERMENTS 
244 
centrifuge tube, stoppered, and kept on ice for sixteen to eighteen hours. 
A control with normal serum should be set up in the same manner. During 
this time the tissue antigen is being sensitized by the antibody in the preg- 
nant serum and the sensitized antigen in turn removes the antiferment. 
At the same time the antigen in both pregnant and normal control serum 
is probably absorbing some antiferment in a non-specific manner. At 
this low temperature autodigestion of serum does not occur. 
3. The tubes are now thoroughly centrifuged and each serum trans- 
ferred to a thimble prepared as described. Dialyzation is now conducted 
in a thermostat for twenty-four hours as in the Abderhalden test, and the 
dialysate tested with ninhvdrin as described, which completes the test. 
The dialysate from the thimble carrying exhausted pregnancy serum yields 
a positive ninhydrin reaction due to autodigestion by reason of specific 
removal of antiferment by sensitized antigen in the preceding phase; the 
dialysate from the thimble carrying normal serum yields a negative nin- 
hydrin reaction, or a weak reaction due to the partial removal of antiferment 
by non-specific absorption in the preceding phase. 
In addition, the tissue antigen may be washed twice with saline solu- 
tion by centrifuging and, after the last washing, transferred to a thimble 
with 1.5 c.c. of a fresh normal serum (preferably the serum of a male guinea- 
pig unfed for at least eight hours previous to bleeding). Dialyzation is con- 
ducted as described and the dialysate tested with ninhydrin. Sensitized 
antigen (placental tissue sensitized with pregnant serum) produces nin- 
hydrin reacting substances in the dialysate by removal of antiferment, 
followed by autodigestion of the serum; plain antigen produces much less 
or no ninhydrin reacting substances. 
Practical Value of Abderhalden's Tests 
Pregnancy.—An enormous literature has accumulated on the applica- 
tion of Abderhalden’s methods in the diagnosis of pregnancy. Some of 
the reports by competent investigators are favorable to its practical value; 
others are unfavorable, while the majority report a high percentage of posi- 
tive reactions with the sera in pregnancy along with non-specific results 
with the sera of non-pregnant persons and lower animals. In general, it 
may be stated that the Abderhalden test has failed as a practical diagnostic 
procedure and has been largely discarded; the practical value of Bronfen- 
brenner’s modification awaits to be determined. 
Some authors have summarily dismissed the whole subject, but I am 
not able to do this and believe that it deserves the discussion devoted to 
it in the preceding pages. That Abderhalden is wrong in his conception 
of the nature of the so-called “protective ferments” and the mechanism 
of their action may be regarded as proved; on the other hand, I am con- 
vinced that in pregnancy and other conditions there may be found in the 
serum a specific antibody-like substance in addition to an increase of non- 
specific proteolytic ferments. 
The very technic developed by Abderhalden may not serve for the 
differentiation of sera containing the specific antibody-like substances 
from those that do not but, nevertheless, I believe that enough work has 
been done to show that a specific substance may be present worthy of con- 
tinued efforts in the way of improvement in technic for its detection. After 
the technic has been satisfactorily developed, and especially along the lines 
of Bronfenbrenner’s method, I believe that it is worth while to again study 
the sera in pregnancy from the standpoint of a practical test. PRACTICAL VALUE OF ABDERHALDEN’S TESTS 
245 
In addition to pregnancy the general results of a large number of in- 
vestigations indicate that similar antibody-like substances may be found 
in the serum in cancer, degenerative lesions of the central nervous system, 
and various bacterial infections. I have briefly mentioned in the following 
paragraphs some of these investigations, not because the present methods 
are generally acceptable for the detection of this antibody-like substance 
from the standpoint of practical tests, but to indicate that the antibody 
may be found in many different pathologic conditions. 
Cancer.—A large literature has also accumulated on the results of 
Abderhalden’s tests in cancer; Freund and Abderhalden1 claim to have 
found the “ferments” in the serum of cancer that will yield positive re- 
actions with cancer tissue. Frank and Heiman2 reported positive results 
in 53 of 54 cases of cancer; Markins and Munze,3 Epstein,4 Gambaroff,5 
Erpicum,6 Brockman,7 Lampe,8 Lowy,9 Ball,10 and others have reported 
highly favorable results. Frankie11 and Lindig12 have found the reactions 
generally non-specific in character. 
It is well to make the tests with a number of cancer tissues taken from 
various parts, also with sarcoma tissue and that of various benign tumors. 
Mental Diseases.—Fauser13 has studied the serums of 88 cases of de- 
mentia praecox and other mental diseases with various antigens composed 
of the ductless glands, testicles, ovaries, etc., and attained interesting re- 
sults, tending to show that in many of these brain affections there may be 
associated lesions in other organs, and that the symptoms may be due to 
perverted functions of certain ductless glands. Munzer,14 Bundschue and 
Roener,15 and Fisher16 have also found in the serums of mental and nervous 
diseases “ferments” reacting with the tissues of the ductless and genera- 
tive glands, tending to show that lesions of these organs may be operative 
in the symptomatology of these conditions. 
Syphilis.—Baeslack17 has reported having had exceptionally good re- 
sults with the serums of syphilitics and a substratum composed of coagu- 
lated syphilitic lesions of rabbit’s testicle. Using the dialyzation method 
he found the sero-enzyme test more constant and earlier than the Wasser- 
mann reaction. 
Tuberculosis and Acute Infections.—Abderhalden and Andryewsky18 
have suggested the use of the dialyzation or the optic method in the diag- 
nosis of acute infections. The peptone may either be prepared of the bacilli, 
or the boiled organisms used in the dialyzing shell. In preparing a bacterial 
substratum the material must be carefully centrifuged in order to facilitate 
washing. The tubercle bacilli are degreased by extraction in fat solvents. 
1 Munch, med. Wchnschr., 1913, 14, 763. 
2 Berl. klin. Wchnschr., 1913, 1, No. 14. 
3 Berl. klin. Wchnschr., 1913, 1, No. 17. 
4 Wien. klin. Wchnschr., 1913, xxvi, No. 17. 
5 Berl. klin. Wchnschr., 1913, No. 17. 
6 Bull, de l’Acad. Roy. de Belg., 1913, xxvii, 624. 
7 Lancet, London, November 15, 1913. 
8 Munch, med. Wchnschr., 1914, lxi, No. 9. 
9 Jour. Amer. Med. Assoc., 1914, lxii, 437. 
10 Jour. Amer. Med. Assoc., 1914, lxii, 599. 
11 Deutsch. med. Wchnschr., xl, No. 12. 
12 Miinch. med. Wchnschr., 1913, 60, 288. 
13 Deutsch. med. Wchnschr., 1913, xxxix, No. 7. 
14 Berl. klin. Wchnschr., 1913, 1, No. 5. 
16 Deutsch. med. Wchnschr., 1913, No. 42, 2069. 
16 Deutsch. med. Wchnschr., 1913, No. 44, 2138. 
17 Jour. Amer. Med. Assoc., 1914, lxii, 1002; ibid., Ixiii, 559. 
18 Miinch. med. Wchnschr., 1913, lxi, 1641. 246 
FERMENTS AND ANTIFERMENTS 
Abderhalden and Andryewsky found “ferments” present in the serum of 
cattle receiving injections of suspensions of dead tubercle bacilli and in 
experimental infections, and suggest that the test may prove efficacious 
in testing cattle. This work should receive further study in human infec- 
tions. Smith,1 employing Bronfenbrenner’s modification of the Abder- 
halden test, has reported highly specific results in the differentiation of 
various bacteria with rabbit immune sera. 
THE TWORT-D’HERELLE PHENOMENON OF BACTERIOLYSIS. 
While various enzymes of different sources may digest dead micro- 
organisms and play a part in resistance to infection and recovery from dis- 
ease, the bacteriolytic agent discovered by Twort and d’Herelle is capable 
of dissolving living bacteria of various kinds, and its discovery constitutes 
not only an important advance in our knowledge of bacteriology, but like- 
wise possesses great interest in relation to the processes of infection and 
immunity. I have already referred to the subject in previous chapters in 
relation to infection and immunity, and wish at this place to discuss more 
completely the source, nature, properties, and significance of this newly dis- 
covered bacteriolytic agent, which some investigators regard as a bacterial 
enzyme. 
In 1915 Twort2 observed in cultures of micrococci prepared by plating 
glycerinated cowpox vaccine certain transparent or degenerated areas in 
which the cocci did not grow. When material from these areas was streaked 
through a culture of the micrococci it was observed that the colonies became 
clear and transparent within a few hours. When material from a trans- 
parent area was passed through a Berkfeld filter and the filtrate added to 
cultures of the micrococcus it was found that the latter were killed and dis- 
solved. In other words, a filter-passing agent was discovered capable of kill- 
ing and dissolving living staphylococci; the source of this agent was prob- 
ably the micrococci or cowpox vaccine, but the bacteriolytic agent in the 
filtrate could be passed on to numerous generations by transferring it to 
successive fresh cultures of staphylococci and refiltering. 
Twort obtained similar results with a bacillus of the colon group, isolated 
from the intestinal mucosa of a dog with distemper; also with a large bacillus 
of the colon group from the intestinal discharges of a child suffering from 
diarrhea. The bacteriolytic agent maintained its activity for six months 
and was destroyed by heating at 60° C. 
This discovery did not attract the attention it deserved due in large part 
to the absorption of interest by the events and demands of the World War. 
About the same time d’Herelle3 may be said to have discovered the phe- 
nomenon anew and independently by finding in the intestine of locusts a 
principle antagonistic to the action of a certain pathogenic cocco-bacillus. 
With this suggestive observation as a basis, d’Herelle in 1915 systematically 
sought for a similar principle in the intestinal contents of patients with 
intestinal infections, and especially dysentery, which prevailed in a squadron 
of cavalry stationed in the neighborhood of Paris. It was observed that 
by inoculating 2 or 3 drops of a stool from a dysentery patient in the Pasteur 
Hospital (at the outset of clinical improvement) in 20 c.c. of bouillon and 
1 Jour. Infect. Dis., 1916, 18, 14. 2 Lancet, London, 1915, 2, 1241. 
3 Compt. rend. Acad. d. sc., Paris, 1917, 165, 373; ibid., 1918, 167, 970; ibid., 1919,168, 
631. Also a series of papers in Compt. rend. Soc. de biol., 1918,1919, 1920, and 1921, volumes 
81, 82, 83, and 84. For the latest summary of investigations and discussions consult Smith’s 
translation of d’Herelle’s monograph: The Bacteriophage—Its Role in Immunity, published 
by Williams & Wilkins Company, Baltimore, 1922. THE TWORT-D’HERELLE PHENOMENON OF BACTERIOLYSIS 
247 
filtering the culture a few hours later through a Chamberland No. 12 filter, 
that a trace of this filtrate added to a culture of the Shiga dysentery bacillus 
was followed by the death and dissolution of the organisms. When this 
material was filtered and a drop or two of the filtrate added to a fresh cul- 
ture the phenomenon was repeated. 
Since then d’Herelle has studied the subject very extensively, succeeding 
in discovering a bacteriolytic agent for various bacteria in the feces of human 
beings and some of the lower animals with various diseases, as well as in the 
feces of a few normal healthy individuals. He states that the phenomenon 
was probably first observed by Hankin,1 who found that the water of the 
Jumna River below the town of Agra was bactericidal for various micro- 
organisms and particularly the cholera vibrio; d’Herelle believes that these 
effects were due to the presence of this filterable bacteriolytic agent. 
d’Herelle’s Method of Securing the Bacteriolytic Agent.—For isolation 
from feces a portion of the size of a pea is inoculated into 50 c.c. of ordinary 
peptone bouillon about —0.6 in reaction to phenolphthalein (lysis will not 
occur in an acid medium). This culture is incubated at 37° C. for from 
twelve to eighteen hours, followed by filtration through sterile paper and 
finally through a sterile earthen filter, as the Chamberland L2 and L3 bougies. 
The day before the test is to be made an agar slant is inoculated with a 
strain of dysentery bacillus, and on the day of the test four tubes of broth 
medium are inoculated from this tube. To the first of these tubes is added 
1 drop of the filtrate, to the second, 10 drops, and to the third, 2 c.c. The 
fourth serves as a culture control. 
The tubes are incubated for eighteen to twenty-four hours. If all tubes 
carrying filtrate show a growth, about 0.02 c.c. from each is spread over the 
surfaces of three tubes of agar. If, after incubation, these tubes present a 
normal growth of the dysentery bacillus, the feces did not contain the bac- 
teriolytic agent. If, however, the cultures appear broken up, “moth- 
eaten,” and with areas of no growth, the results are positive. 
If, however, one or all three of the broth cultures remain clear while the 
control shows turbidity, the bacteriolytic agent is present, the amount in 
the filtrate bearing a relation to the results observed with the different 
amounts of filtrate. d’Herelle states that the bacteriophage may be present, 
therefore, without the slightest macroscopic evidence of lysis in the broth 
culture and that this is usually the case in the process of isolation. In his 
opinion the bacteriophage multiplies under certain conditions, and he has 
described methods for enhancing their virulence and for estimating their 
numerical strength. 
Sources and Specificity of the Bacteriolytic Agent.—As stated above, 
Twort originally discovered the agent in cultures of staphylococci from vac- 
cine lymph; Gratia2 has recently confirmed these findings and Bail3 has 
obtained the substance from Shiga dysentery bacilli. Twort, in addition 
to his work with staphylococcus cultures, obtained similar results with 
organisms of the colon-typhoid group and with cultures from cases of canine 
distemper and of infantile diarrhea. 
d’Herelle made daily filtrates of cultures of the stools of 34 cases of 
dysentery and obtained a bacteriolytic agent as soon as clinical improve- 
ment commenced. The filtrates of the stools of patients during the active 
stage of the disease, of patients that died, or of stools that did not contain 
Shiga bacilli were not bacteriolytic. He applied the same methods to the 
1 Ann. l’Inst. Pasteur, 1896, 10, 175, 511. 
2 Proc. Soc. Exper. Biol, and Med., 1921, 18, 192. 
3 Wien. klin. Wchn., 1921, 34, 555. 248 
FERMENTS AND ANTI FERMENTS 
stools of typhoid and paratyphoid patients and obtained bacteriophages 
active against Bacillus typhosus and B. paratyphosus. d’Herelle later 
made cultures from the stools of a healthy man every two weeks for ten 
months. Contrary to his previous belief that normal stool filtrates had not 
bacteriolytic power, he obtained from these 23 tests, 2 that were active against 
B. shigae, 2 active against Flexner bacilli, 1 against paratyphoid A, 3 against 
paratyphoid B, 3 against B. coli, and 1 against hog cholera. Eleven of the 
23 stool filtrates were inactive. d’Herelle also obtained a filtrate active 
against B. sanguinarium and another active against B. gallinarum from the 
stools of chickens suffering from avian typhoid. No active filtrates were 
found in the stools of normal fowls. In cases of human pyelonephritis, of 
flacherie in silkworms, of hemorrhagic septicemia in cattle, and of plague in 
rats, d’Herelle isolated bacteriolytic agents. 
Bordet and Ciuca1 introduced a new method for obtaining the bacterio- 
lytic agents. They injected a culture of Bacillus coli intraperitoneally into 
a guinea-pig. The day after the third injection they found that a small 
quantity of the resulting peritoneal exudate, as well as the culture of B. coli 
isolated from this exudate, would dissolve an eighteen-hour culture of the 
strain of B. coli that had been used for these inoculations. This lysis was 
not complete and a few colonies could be cultivated. If these surviving 
organisms were inoculated into another eighteen-hour colon culture, lysis 
would again result. This bacteriolytic action could in this way be repeated 
indefinitely. The bacilli exposed to the lytic substance acquired the ability 
to transfer the lytic property to subsequent generations. Wollstein2 repeated 
this work using a strain of the Shiga bacillus instead of B. coli, and obtained 
an agent bacteriolytic for dysentery and other bacilli. 
Dumas3 found filtrates active against Bacillus shigae and B. coli in the 
stools of 5 out of 8 normal individuals who never had had intestinal dis- 
ease, in the feces of guinea-pigs, in earth, in Paris city water, and in water 
from the Seine. 
Debre and Haguenau,4 using d’Herelle’s method, found an active bacterio- 
phage in 3 of 6 cases of acute dysentery, in 3 of 16 cases of enteric fever, in 
1 case of diarrhea, in 1 case of cancer of the stomach, in 1 case of rheumatic 
fever, in 1 case of phthisis, and in 2 cases of peritonitis. They were unable 
to obtain evidence of a bacteriolytic agent in filtrates from the stools of 
healthy or ill, breast, or bottle-fed infants. 
Kuttner5 from the stools of a typhoid convalescent obtained a filtrate 
active against typhoid, Shiga, and Flexner bacilli. She also found that a 
filtered glycerin extract of the small intestine of a guinea-pig and a saline 
extract of a guinea-pig’s liver were bacteriolytic for typhoid bacilli. This 
lytic principle could be transmitted by passage through successive typhoid 
cultures. Glycerin extracts of the large intestine and of muscle tissue were 
not bacteriolytic. 
Wollstein, using d’Herelle’s method, made filtrates of the stools of 23 in- 
fants including 6 with gastro-intestinal disturbances and 1 convalescent 
from Shiga dysentery. The stool filtrate of only 1 baby was bacteriolytic. 
This infant had a fatal B. coli peritonitis. The filtrate was active against 
colon and Shiga bacilli. 
1 Compt. rend. Soc. de biol., 1920, 83, 1293, 1296; ibid., 1921, 84, 276, 278, 280, 745, 
747, 748. 
2 Jour. Exper. Med., 1921, 34, 467. 
3 Compt. rend. Soc. de biol., 1920, 83, 1314. 
4 Compt. rend. Soc. de biol., 1920, 83, 1348. 
5 Proc. Soc. Exper. Biol, and Med., 1921, 18, 158, 222. THE TWORT-D'HERELLE PHENOMENON OF BACTERIOLYSIS 
249 
Davison1 has been able to demonstrate the bacteriolytic agent active 
against Shiga and Flexner bacilli in filtrates of the stools of a normal infant 
as well as from those suffering from bacillary dysentery (Flexner). If one- 
to sixty-day-old broth or peptone water cultures of recently isolated or old 
laboratory strains of B. dysenteriae (Shiga) or (Flexner) were filtered, the 
filtrate in many instances was slightly bacteriolytic for Shiga and Flexner 
bacilli. The bacteriolytic agents from the stools of these children and the 
filtrates of dysentery cultures retained their lytic power after passage through 
several successive cultures. 
From this review it is evident that the filter-passing bacteriolytic agent 
of Twort and d’Herelle may be secured from various and diverse sources; 
also that its origin in many instances bears no relation to the type of organ- 
ism attacked. 
While originally believed by both Twort and d’Herelle to be specific, 
it is now known that the lytic agent is non-specific and may attack several 
species of bacteria, although some digest only the bacteria causing the dis- 
ease and especially when isolated from severe infections. 
Properties of the Bacteriolytic Agent.—Considerable emphasis has been 
placed upon the influence of heat and other physical agents upon the bac- 
teriolytic agent as indicating its nature; that is, whether it is a living organ- 
ism of ultramicroscopic size or an enzyme. 
Twort found that the lytic action of a filtrate was destroyed when heated 
to 60° C., but not when heated to 52° C. d’Herelle found that it was still 
active after one hour at 64° to 65° C. Kabeshima2 found that a filtrate may 
be kept at 37° C. for four years, or at less than 0° C. for one hour, or be 
heated to 70° C. without losing its bacteriolytic power. He found that it 
was inactivated at 75° C. Davison found that filtrates were unaffected by 
being heated to 62° to 67° C. for one hour. Kuttner found that her filtrates 
would withstand being heated to 70° C. for thirty minutes, but that they were 
inactivated when exposed to 75° C. for thirty minutes. The temperature 
at which a filtrate and a culture are incubated affects the rate of lysis. 
d’Herelle noted that it proceeded very slowly at 15° to 16° C. Kuttner 
found that lysis occurred in about half the time when the filtrate and cul- 
ture were incubated at 41° to 42° C. as at 37° C. Lysis did not occur when 
the incubation temperature was 45° to 50° C. 
Gratia showed that the inhibition by the lytic agent on the growth of 
Bacillus coli was greatly influenced by the reaction of the medium, being 
faint at PH 6.8, 7.0, and 7.4, but much more pronounced at PH 8.0 and 8.5. 
Eliava and Pozerski3 stated that filtrates were active in media with reaction 
from PH 2.5 to 8.4, but lost their activity when the hydrogen ion concentra- 
tion was above PH 2.5 and below 8.4. Davison found that lysis was more 
complete when the reaction of the medium was at PH 8.0 and 8.2 than at 
Ph 6.0 to 7.7. Wollstein reported that the lytic action proceeded as rapidly 
and as completely in the absence of oxygen as in its presence. 
Kabeshima succeeded in chemically precipitating a substance having 
bacteriolytic power from a stool filtrate. His procedure was to add 3 volumes 
of acetone to a filtrate, shake, and allow to stand for forty-eight hours at 37° 
C., and then to decant to remove the acetone. The precipitate was a yellow- 
ish powder and had a more powerful bacteriolytic action than the original 
filtrate. It could be preserved unchanged for six weeks in pure acetone, 
but its strength diminished after ten weeks. It could be restored, how- 
1 Jour. Bacteriology, 1922, 7, 475, 491. 
2 Compt. rend. Soc. de biol., 1920, 83, 219, 471. 
3 Compt. rend. Soc. de biol., 1921, 84, 708. 250 
FERMENTS AND ANTIFERMENTS 
ever, by passage through broth cultures of dysentery bacilli. Alcohol 
also precipitated this bacteriolytic substance. The lytic agent was soluble 
in ether. By adding an equal volume of anhydrous ether to a filtrate, shak- 
ing, allowing to stand forty-eight hours, and then evaporating the ether, a 
waxy deposit having bacteriolytic activity was obtained. He also found 
that the addition of 1.0 per cent, sodium fluorid would not destroy the lytic 
activity of a filtrate, although it has been commonly believed that this 
amount of fluorid destroyed life and the fermentation associated with life. 
Bablet,1 however, found that sodium fluorid inhibited lytic activity and 
that chloroform and glycerin prevented lytic activity, although they would 
not prevent the growth of Shiga bacilli. Eliava and Pozerski2 stated that 
twenty-four hours’ contact with 2.5 per cent, phenol or 2.5 per cent, fluorid 
did not affect the bacteriolytic agent, but that 0.75 per cent, quinin chlo- 
hydrate reduced the lytic activity of a filtrate and 1 per cent, quinin chlor- 
hydrate destroyed it. They also claim that quinin salts do not affect soluble 
ferments. 
Maisin3 stated that the lytic substance was completely precipitated by 
saturating a bacteriolytic filtrate with ammonium sulphate and was almost 
completely precipitated by half saturation. The fact that the lytic sub- 
stance would not pass through a collodium membrane led him to assume 
that it was a colloid. Wollman4 has, however, shown that if the mem- 
branes are made of dilute collodion (less than 4 per cent.) they are per- 
meable to the bacteriolytic agent. Davison found that lytic activity is 
destroyed by the addition of 1 c.c. of N sodium hydrate to 4 c.c. of bac- 
teriolytic filtrate. 
These properties indicate that the bacteriolytic agent may be an enzyme, 
but if so, the enzyme requires activation by living bacteria because several 
investigations have shown that filtrates added to dead bacteria did not 
produce lysis, although living organisms were quickly dissolved. Apparently 
only living organisms are attacked and especially cultures less than twenty-four 
hours old in a suitable broth medium rather than suspensions in saline 
solution. If the bacteriolytic agent is a living organism of ultramicroscopic 
size it has not yet been successfully cultivated, although it may be active in 
cultures of dysentery bacilli or other bacteria for many months. 
d’Herelle, Davison, and others have found that filtrates containing the 
bacteriolytic agent are non-pathogenic for rabbits. Antibodies may be 
produced, however, against the bacterium employed in the preparation of 
the filtrate due to immunization by the dissolved proteins of the bacteria 
in the filtrate; antibody may also be produced against the bacteriolytic agent 
inasmuch as the sera are sometimes antilytic. These results are to be ex- 
pected, however, and do not yield any particular information regarding the 
nature of the lytic agent since an antilysin may be produced whether the 
lysin is a living microbe or a simple enzyme. 
Nature of the Bacteriolytic Agent (Bacteriophage; Bacteriolysant).— 
The exact nature of the bacteriolytic agent is unknown. Twort originally 
favored the view that it was an autolytic enzyme produced by bacteria and 
thought that it may arise spontaneously in cultures; more recently he has 
stated5 that d’Herelle’s explanation was the more probable. 
d’Herelle regards the lytic agent as a diastatic ferment produced in living 
1 Compt. rend. Soc. de biol., 1920, 83, 1322. 
2 Compt. rend. Soc. de biol., 1921, 85, 139. 
3 Compt. rend. Soc. de biol., 1921, 84, 467. 
4 Compt. rend. Soc. de biol., 1920, 83, 1478; ibid., 1921, 84, 3. 
6 Brit. Jour. Exper. Path., 1920, 1, 237. THE TWORT-D’HERELLE PHENOMENON OF BACTERIOLYSIS 
251 
bacteria by the entrance of an organism of ultramicroscopic size. He has 
called this organism bacteriophagum intestinale or bacteriophage, the name 
simply denoting its characteristic property and the place where he first 
found it. He has summarized his views as follows: “The bacteriophage is 
an ultramicroscopic organism, which is very widely disseminated in nature. 
It only grows in contact with living bacteria. It penetrates into the interior 
of an organism and forms a colony of 15 to 25 elements in the space of one 
and a half hours. The organism then bursts, liberating the young ultrami- 
crobes. These utilize for their development the bacteria which they dissolve 
with the aid of the lytic diastase, which they secrete. There is only one 
species of bacteriophage and this can acquire activity against any organism.” 
According to d’Herelle the bacteriophage is of an extremely small size 
capable of passing dense filters, although it may be partially concentrated 
by prolonged centrifuging at high speed. He has found that it possesses 
very great vitality, persisting in feces and filtrates properly sealed for sev- 
eral years and in a dried state for as long as six months. It is destroyed 
by heating at 75° C.; is sensitive to some antiseptics and resistant to others; 
for example, it has survived in 1 : 200 mercuric chlorid for four days and in 
phenol 1 : 100 for seven days, but is killed in 90 per cent, alcohol in two 
days and is highly susceptible to the products of bacteriolysis. 
When freshly isolated the bacteriophage usually is capable of attacking 
several bacterial species, but displays a variable virulence for these. Bac- 
teriophages have been described for the different strains of dysentery bacilli; 
also for typhoid, paratyphoid, colon, cholera, diphtheria, bubonic plague, 
and proteus bacilli, and various other bacilli of infectious diseases of the 
lower animals, as well as B. subtilis and staphylococci.. 
In mixtures of bacterial cultures and bacteriophage some organisms 
escape lysis and may acquire a resistance to bacteriophage accompanied by 
morphologic changes. d’Herelle believes that this occurs in the develop- 
ment of “carriers” of dysentery and typhoid bacilli. Some bacteria possess 
high natural resistance to bacteriophage and especially those micro-organisms 
possessing high virulence for man or the lower animals and resistant to phago- 
cytosis. According to d’Herelle: “Infection and death or immunity and 
recovery depend upon whether the bacteria or the bacteriophage triumphs 
in this battle. The products of the bacteria which have been dissolved by 
the bacteriophage also play an active role in stimulating the formation of 
antibodies. The outcome of an epidemic also depends upon these two 
forces for the active bacteriophage, the agent of immunity, as well as the 
bacteria causing the epidemic can spread from an individual to another.” 
The lysin or diastatic ferment which d’Herelle claims is produced when 
the bacteriophage penetrates a living bacterium is precipitable by alcohol, 
resists heating to 58° C., and is regarded as possessing a powerful opsonic 
action on the bacteria for which the bacteriophage possesses virulence. 
d’Herelle claims to have produced antisera for the bacteriophage by injecting 
filtrates into the lower animals; these antibacteriophagous sera were found 
to possess complement-fixing antibodies for bacteriophagous antigen and 
capable of neutralizing but not killing the bacteriophage. 
Kabeshima1 has suggested that the bacteriolysis is due to the interaction 
of a catalyst present in the feces of the patient and a proferment produced 
by the bacterium which is attacked and digested. He states that the 
catalyst is produced by some intestinal gland or by leukocytes as a protective 
measure against infection and found that the lytic substance could with- 
stand being heated to 70° C. and the effects of precipitating and antiseptic 
1 Compt. rend. Soc. de biol., 1920, 83, 219, 471. 252 
FERMENTS AND ANTIFERMENTS 
substances, all of which suggests that it is a ferment. d’Herelle has refused 
to accept this hypothesis not only because in his opinion a living microbe 
may possess these physical properties, but because it fails to explain serial 
action by the fact that the same lytic agent can act on diverse bacterial 
species, etc. 
Bordet and Ciuca have suggested that the ability to produce the lytic 
substance was acquired by bacteria as a result of contact with some external 
stimulus such as the leukocytic exudate of the peritoneum of a guinea-pig. 
This external influence possibly represented a defense mechanism on the 
part of the animal. The bacteria were then able to transmit to their de- 
scendants the aptitude to form this lytic substance or this aptitude for autol- 
ysis. Inasmuch as this substance was diffusible in a culture-medium, the 
mere contact with a medium in which such a microbe had grown would 
confer the ability to produce this substance upon allied bacteria placed on 
the medium, and these, in turn, to others. 
d’Herelle also refuses to accept this hypothesis for various reasons, one 
of which is that in his opinion it does not conform to the experimental facts. 
Bail1 accepts d’Herelle’s claim that the bacteriophage exists in the 
form of autonomous masses and conducts itself as an ultramicrobe, but states 
that these particles could only be constituted by the “splitter,” that is to 
say, by particles derived from the digested bacteria themselves. These 
organized particles, capable of reproduction under a filtrable form at the 
expense of the same bacteria, secrete a dissolving diastase. d’Herelle states 
that acceptance of this theory requires strict specificity on the part of the 
bacteriolytic agent and numerous investigators have proved its non-speci- 
ficity. 
It is apparent, therefore, that the various theories are divided into those 
regarding the bacteriolytic agent as an enzyme produced by bacteria and 
those agreeing in principle with d’Herelle, that lysis is indeed caused by a 
diastatic ferment, but that this ferment is produced by a newly discovered 
parasite of ultramicroscopic size characterized by its ability to penetrate 
only living bacteria. 
Davison2 leans toward the former hypothesis and summarizes the litera- 
ture and the results of his own experiments as follows: “According to the 
data available at present d’Herelle’s phenomenon probably depends upon 
a bacteriolytic enzyme produced by bacteria. The amount of this enzyme 
produced by a culture can be increased by external influences, such as in- 
testinal secretions, tissue extracts, leukocytes, etc. The action of these 
external influences is probably to favor the development of lysogenic organ- 
isms at the expense of the non-lysogenic. This enzyme not only dissolves 
organisms but also favors the multiplication of bacteria which produce this 
enzyme. In this way the bacteriolytic principle is carried from generation 
to generation. It is highly improbable that this phenomenon represents a 
defense mechanism on the part of an animal against bacterial invasion.” 
Relation of the Twort-d’Herelle Phenomenon to Immunity.—d’Herelle 
has placed great importance upon the bacteriolytic agent in relation to 
infection and immunity. To him the question of whether or not a bacterial 
disease develops resolves itself into the simple terms of a struggle between 
infecting microparasite and resisting bacteriophage. The “vicissitudes in 
the struggle between these two factors are reflected in the condition of the 
infected individual. Convalescence begins at the moment when the viru- 
lence of the bacteriophage is sufficient to give it, definitely, the upper hand. 
1 Wien. klin. Wchn., 1921, 34, 237. 
2 Abst. of Bacteriology, 1922, 6, 159. THE TWORT-D’HERELLE PHENOMENON OF BACTERIOLYSIS 
253 
The disease has a fatal outcome if the bacteriophage is inactive as a result 
of unfavorable conditions, or if the bacterium is able to acquire a refractory 
state.” Since the bacteriolytic agent is regarded as transmissible from one 
individual to another, d’Herelle believes that it is possible to confer re- 
sistance, a heterologous antimicrobial immunity, by administering filtrates 
containing the agent. He and others have found that these filtrates are well 
borne by animals in large doses by oral or subcutaneous injection, and from 
experiments with avian typhoid, barbone, and dysentery d’Herelle states 
that the administration of the bacteriolytic agent ought to prove curative 
when administered early in the treatment of disease. As previously stated, 
Davison, however, was unable to detect any curative activity in dysentery 
of children; d’Herelle’s claims are as yet unproved and the true relation of 
the bacteriolytic agent to infection, immunity, and the treatment of disease 
are still as obscure as the nature of the agent itself. CHAPTER XV 
BACTERIAL AGGLUTININS 
As previously stated, given any infection, several antibodies of different 
properties may be produced. If the infecting micro-organism produces 
characteristically an exogenous toxin, as, for example, that produced by 
the diphtheria bacillus, an antitoxin is produced as the most prominent 
of several antibodies. With other pathogenic bacteria that produce mainly 
an endogenous toxin various antibodies are formed, and one or more may 
play a prominent role in protecting the host, such as opsonins, agglutinins, 
precipitins, bacteriolysins, etc. 
If typhoid immune serum from an immunized animal or a patient suffer- 
ing from typhoid fever is added to an emulsion of typhoid bacilli in a test- 
tube and the mixture placed in an incubator, the following phenomenon 
will be observed: the bacteria, which previously formed a uniform emul- 
sion, clump together into little masses, settle at the sides of the test-tube, 
and gradually fall to the bottom, the fluid becoming almost clear. In a 
control test to which no active serum is added, the fluid remains uniformly 
cloudy. If the reaction is observed microscopically in a hanging drop, 
it is noted that with the addition of the serum the bacilli move nearer and 
nearer one another, this process being followed by a gradual loss in motility 
and the formation of clumps. The substance in the serum causing this 
phenomenon is called agglutinin, and the reaction is known as agglutination. 
Definition.—Agglutinins are antibodies that possess the power of causing 
bacteria, red blood-corpuscles, and some protozoa (trypanosomes) suspended in 
a fluid to adhere and form clumps. 
Historic.—Although Metchnikoff, Charrin and Roger1 had noticed 
peculiarities in the growth of Bacillus pyocyaneus when cultivated in im- 
mune serum which we now believe were due to agglutinins, Gruber and 
Durham,2 and Bordet3 were the first to recognize that the agglutination 
reaction was a separate function of immune serum. While investigating the 
Pfeiffer phenomenon of bacteriolysis with B. coli and the cholera vibrio, 
these investigators found that if the respective immune serums were added 
to bouillon cultures of these two species the cultures would lose their tur- 
bidity, flake-like clumps would form and sink to the bottom of the tube, 
and the supernatant fluid would become clear. Gruber at the same time 
called attention to the fact that agglutinins were not absolutely specific 
for their own antigen, but would agglutinate to a lesser extent closely 
allied species of bacteria. 
In 1896 Pfaundler4 drew attention to a peculiar phenomenon observed 
when bacteria were grown in an immune serum. Long and more or less 
interlaced threads of bacteria developed, which were regarded as due to 
agglutinins. At that time considerable emphasis was laid upon the im- 
portance of Pfaundler’s reaction, but at present the ordinary agglutination 
tests have superseded this reaction as a practical diagnostic procedure. 
1 Compt. rend. Soc. de Biol., 1889, 667; Ann. d. PInst. Pasteur, 1891, 5, 473. 
2 Munch, med. Wchn., 1896, 285; Jour. Path, and Bacteriol., 1897, 4, 13; ibid., 1901, 
7, 240; Brit. Med. Jour., 1898, 2, 588. 
3 Ann. de PInst. Pasteur, 1895, 9, 462. 
4 Centralbl. f. Bakteriol., Abt., 1898, 23, 71 and 131. 
254 NORMAL AND IMMUNE AGGLUTININS 
255 
In 1896 Widal1 and Griinbaum first turned these facts to practical use 
in the diagnosis of typhoid fever. These investigators found that the serum 
of patients suffering from typhoid fever acquires a high agglutinating power 
for Bacillus typhosus, and since this phenomenon generally manifests itself 
comparatively early in the disease, its recognition has considerable diag- 
nostic importance. It is purely accidental that we speak of the “Widal 
reaction” in typhoid fever, rather than of the “Griinbaum reaction,” for 
the latter observer conducted similar studies independent of Widal, but, 
owing to a lack of patients, Widal preceded him in the publication of a 
more extensive wrork. 
At the present time this diagnostic reaction is known as the Gruber- 
Widal reaction. It has proved of great value to a large number of different 
investigators, not only in making the serum diagnosis of typhoid fever, but 
in other infections as well. 
Kinds of Agglutinins.—Following the discovery of bacterial agglutinins 
by Charrin and Roger, Bordet2 discovered agglutinins in sera for red blood- 
corpuscles called hemagglutinins. Subsequent investigations showed that 
various substances as phytotoxins (abrin, ricin) and zootoxins (venoms) 
possessed the property of agglutinating red corpuscles and particularly 
those of certain animals called phytagglutins and zooagglutinins. 
In addition to these agglutinins for bacteria and red blood-corpuscles, 
myco-agglutinins may be found in the blood for the higher plants or fungi 
during mycotic infections, as for Sporotrichum beurmanni during sporo- 
trichosis; also protozoa agglutinins for such protozoa as trypanosomes and 
spirochetes. 
Of interest in this connection is the possibility of bacterial agglutinins 
for some motile bacilli being of two varieties, one for the bodies of bacteria, 
and the other for flagella. Smith and Reagh,3 working with the hog-cholera 
bacillus, found body and flagella agglutinins, the latter being much more 
easily demonstrated in immune sera and manifest in dilutions over twenty 
times greater than in those in which body agglutinins became visible. These 
observations were confirmed by Beyer and Reagh,4 who also claimed that 
the flagellar and body (somatic) agglutinins and agglutinable substances 
of the hog-cholera bacillus may be differentiated by heat. 
Hemagglutinins may be present in the blood-serum for the red blood- 
corpuscles of a different species (heterologous hemagglutinins), or for cor- 
puscles of animals of the same species (homologous hemagglutinins). The 
latter are also known as isohemagglutinins, as the agglutinin in the serum 
of one person for the corpuscles of another person. Autohemagglutinins 
occur in the serum for the corpuscles of the same person or lower animal 
and are rarely found. 
Normal and Immune Agglutinins.—Normal serums are frequently 
capable of agglutinating bacteria, such as the typhoid, colon, pyocyaneus, 
and dysentery bacilli. In some cases the typhoid bacillus may be agglu- 
tinated in dilutions as high as 1 : 30, a point of practical importance in the 
clinical use of the test. When a normal serum is found to have a high agglu- 
tinating power, it is probable that a previous infection by the micro-organism 
has oc'curred. Since the serum of a newborn child is largely devoid of 
agglutinins that are found in later life as shown by Savage5 and others, 
the so-called normal or natural agglutinins may, after all, be acquired 
properties. 
1 Semaine med., 1896, 259. 
2 Ann. d. PInst. Pasteur, 1898, 12, 688. 
3 Jour. Med. Research, 1903, 1904, 10, 89. 
4 Jour. Med. Research, 1904, 12, 313. 
6 Jour. Hyg., 1918, 17, 34. 256 
BACTERIAL AGGLUTININS 
The term immune agglutinin is applied to the agglutinating substance 
in a serum developed as the result of infection or of systematic immuniza- 
tion with the micro-organism. 
Studies in the bacterial agglutinins have been almost entirely confined 
to the sera of warm-blpoded animals; recently, Takenouchi1 has shown that 
the sera of several varieties of turtles may contain agglutinins for various 
bacteria of the typhoid-colon group. These observations are also of interest 
from the standpoint of origin of normal agglutinins, as their formation for 
these bacilli in cold-blooded animals by unrecognized infection appears 
improbable. 
As shown by Noguchi2 the sera of many cold-blooded animals may contain 
normal agglutinins and hemolysins for the corpuscles of cold-blooded ani- 
mals; since then Friedberger and Seeling,3 Landsteiner and Rock,4 Frankel,5 
Mazzetti,6 Takenouchi,7 and others have demonstrated hemagglutinins and 
Fig. 95.—Theoretic Formation of Agglutinins. 
hemolysins in the sera of cold-blooded animals for the corpuscles of various 
cold- and warm-blooded animals. 
Agglutinins and Agglutinoids.—According to Ehrlich’s side-chain theory, 
agglutinins are antibodies of the second order (Fig. 95). They resemble 
antitoxins or receptors of the first order in possessing an affinity-bearing or 
haptophore group that unites with the antigen, but they differ from them 
in having also a functional or agglutinophore group that agglutinates the 
antigen when this union has occurred (Fig. 96). 
Agglutinins that have lost their zymophore or agglutinophore group 
through the action of heat, age, acids, etc., but that still possess their hapto- 
phore group, are called agglutinoids, just as toxins that have lost their toxo- 
1 Jour. Infect. Dis., 1918, 23, 393. 
2 Bull. Univ. of Penna., 1902, 14, 438. 
3 Centralbl. f. Bakteriol., orig., 1908, 46, 421. 
4 Ztschr. f. Immunitatsf., 1912, 14, 14. 
5Ztschr. f. Immunitatsf., 1911, 10, 415. 
6 Ztschr. f. Immunitatsf., 1913, 18, 132. 
7 Jour. Infect. Dis., 1918, 23, 415. ORIGIN AND DISTRIBUTION OF AGGLUTININS 
257 
phore group are called toxoids. Such agglutinoids, then, may still combine 
with the bacteria or blood-cells without being able, however, to produce 
agglutination (Fig. 97). Heating for thirty minutes'at 65° to 80° C. usually 
changes agglutinins to agglutinoids; agglutinoids are destroyed at tempera- 
tures above 80° C. 
It is found, at times, that even a fresh serum when concentrated will 
cause less agglutination than when it is diluted. This is ascribed to the 
presence of agglutinoids, which have a stronger affinity for agglutinogen 
than has the agglutinin. When producing a reaction of this character 
they are called proagglutinoids. When the serum is diluted, the proagglu- 
tinoids become less concentrated and- finally, when they are diluted as 
to have no influence on the reaction, the agglutinins are still present in 
sufficient quantity to bring about 
agglutination. As a practical fact, 
in agglutination reactions the action 
of proagglutinoids is of much im- 
portance, for the inexperienced may 
be misled by the absence of, or by 
poor, agglutination in lower dilutions 
to neglect the use of higher dilutions. 
Agglutinogen.—The' substance in 
bacteria or other cells that produces 
agglutinin is called agglutinogen. It 
appears to be formed in the cell, and 
in some cases may be excreted into 
the surrounding medium. Certainly 
Fig. 96.—Theoretic Structure of Agglutinin 
AND AGGLUTINOID. 
1, Agglutinin: H, Haptophore group for 
union with antigen; A, the agglutinophore or 
zymophore group. 
2, Agglutinoid. Same structure as agglu- 
tinin, except that the agglutinophore or zymo- 
phore group is lost. 
Fig. 97.—-A Diagrammatic Illustration of the 
Action of Agglutinins and Agglutinoids. 
In the first tube (left) most of the bacilli (B) 
have been agglutinated and massed in the bottom 
of the tube by the agglutinins (a). 
In the second tube (right) the bacilli (B) are 
in combination with the agglutinoids (A), but 
agglutination does not occur because the agglu- 
tinophore groups are lost. A few bacilli have been 
agglutinated by the agglutinins (a). 
when bacteria die and become disintegrated, agglutinogen is liberated and 
the filtrates (entirely free from bacterial cells), when injected into animals, 
will cause the formation of agglutinins. The term agglutinogen is also used 
for designating the antigen or suspensions of cells used in conducting agglu- 
tination tests. 
Agglutinogen must be considered as having a simple haptophorous 
group, through which it may unite with the receptors of the tissue cells. 
This haptophore comes into play again in the union between agglutinogen 
and agglutinin, which precedes agglutination. It is a passive body, similar 
to the haptophore of antitoxin, and has no other function than that of unit- 
ing either with cell or wdth agglutinin. 
Origin and Distribution of Agglutinins.—The investigations that have 
been carried out for the purpose of determining the site of formation of 258 
BACTERIAL AGGLUTININS 
agglutinins have not thus far yielded conclusive results. The lymphoid 
tissues appear especially concerned, agglutinins being found early in the 
bone-marrow and the spleen by Pfeiffer and Marx1 and van Emden.2 Metch- 
nikoff believes that agglutinins may be derived from leukocytes and endo- 
thelial cells. It is more probable, however, that the formation is general, 
and is the result of wide-spread cellular activity. 
In accordance with the side-chain theory the ability of an animal to 
form agglutinins for a certain micro-organism would depend on its pos- 
session of receptors of the second order, which are able to unite with the 
agglutinogenic receptors of the micro-organism. It has been well estab- 
lished that the number of such suitable receptors vary in animals, and that 
different animals may not produce serums with equal agglutinating powers. 
Agglutinins do not appear in the serum immediately after inoculation, 
but require an incubation period of from two to four days for their produc- 
tion. 
Agglutinins are to be found in highest concentration in the blood. Dreyer 
and Walker3 found that the plasma contained slightly more normal or 
natural agglutinins than the corresponding serum; during immunization, 
however, more agglutinins were found in the serum. Cerebrospinal fluid 
is free of normal or natural agglutinins; during typhoid fever, and other 
infections accompanied by great production of immune agglutinins, small 
amounts of the antibody may be found in this fluid. Blister fluid, exudates, 
transudates, milk, and tears may contain agglutinins if these are present 
in the blood, as shown by Widal4 in typhoid fever. As recently shown by 
Little and Orcutt,5 the agglutinins toward Bacillus abortus found in the blood- 
serum of newborn calves are obtained from the mother through the colostrum. 
Properties and Nature of Agglutinins.—1. Agglutinins are fairly re- 
sistant substances that withstand heating to 60° C. for thirty minutes and 
lose their power only when heated to higher temperatures. It is possible, 
therefore, to make a serum bacteriolytically inactive by destroying comple- 
ment at 55° C., and still retain its agglutinating power. 
2. They resist drying, and their activity is best preserved in this state. 
They do not dialyze through animal membranes. 
3. As shown by Bechhold,6 Field and Teague,7 and others agglutinins 
are electro-positive, that is, bacteria move toward the anode under the 
influence of an electric current. Field and Teague have also shown that 
a combination of bacteria and agglutinin may be partly disassociated by 
means of the electric current. 
4. They are precipitated from a serum by magnesium or ammonium 
sulphates, when these salts are used in proper concentration, and are thus 
closely associated with the globulin fraction of serum. 
5. They are separate and distinct antibodies, and are not associated 
with bacteriolysins. Thus, the agglutinins of an immune serum may be 
lost, destroyed, or absorbed and the bacteriolysins retained. As previously 
mentioned the bacteriolytic power of a serum may be inhibited by heating 
it to 55° C. for a half hour without influencing the agglutinin content, 
and during disease processes the formation of agglutinins and that of bac- 
teriolysins are apparently not parallel processes. 
1 Ztschr. f. Hyg., 1898, 27, 272. 
2 Ztschr. f. Hyg., 1899, 30, 19. 
3 Jour. Path, and Bacteriol., 1910, 14, 39. 
4 Semaine Med., 1896, 259. 
5 Jour. Exper. Med., 1922, 35, 161. 
6 Ztschr. f. physik. Chem., 1904, xlviii, 385. 
7 Jour. Exper. Med., 1907, 9, 222. MECHANISM OF AGGLUTINATION 
259 
Stuber1 has claimed that agglutinins are lipoidal and may be extracted 
from sera with petroleum ether; Krumwiede and Noble,2 however, were 
not able to confirm these observations and found no evidence supporting 
the claim that agglutinins are lipoidal in character. In this connection 
it may be stated that Graham3 has shown that ether anesthesia has no 
influence upon agglutinins. 
Acid Agglutination.—Bacteria may be agglutinated by acids, and the 
method of acid agglutination was introduced by Michaelis4 for the differ- 
entiation of bacterial species on the basis that the hydrogen-ion concentra- 
tion at which agglutination is maximal is characteristic for various species 
of closely allied types. The results of considerable investigation on the 
acid agglutination of the typhoid-colon group of bacilli has shown that 
Bacillus typhosus and B. paratyphosus are readily distinguished by means 
of the reaction. Michaelis5 believes that his acetic-acid method has proved 
superior to serum-agglutination reactions for the differentiation of the 
typhoid-paratyphoid-dysentery coli group of bacilli. Definite differentia- 
tion between the paratyphoid bacilli A, B, and C have not been seen by 
Beniasch,6 Jaffe,7 Heinmann,8 and Grote9; Beniasch has also reported the 
resistance of B. coli to acid agglutination at any hydrogen-ion concentra- 
tion and, indeed, he has found certain strains of nearly all species of bacteria 
to be non-agglutinable within the tested reaction limits. Gillespie10 found 
that pneumococci belonging to the serologic types I and II have, as a rule, 
narrow zones of agglutination. The optimum hydrogen-ion concentration 
was found different in the two cases, while other pneumococci had broad 
zones or, in a few cases, narrow zones not coincident with those occupied 
by the typical organisms. For a technic of acid agglutination see page 289. 
Mechanism of Agglutination.—The true nature of the phenomenon of 
agglutination is unknown, as is shown by the number of theories advanced. 
Thus: 
1. Gruber’s11 idea of the mechanism of this phenomenon was that the 
agglutinin so changed the bacterial membrane as to render it more viscous, 
and that this increased viscosity caused the bacteria to adhere and form 
clumps. No visible changes in the organisms or red corpuscles can, however, 
be seen. 
According to the investigations of Malvoz,12 Dineur,13 Nicolle and Treuel,14 
the flagella of bacteria are intimately concerned in the phenomenon of 
agglutination and that the bacilli become attached to one another by means 
of these cilia. Ernst and Roby,15 however, have not been able to confirm 
these findings and do not regard the flagella of bacteria as being vitally 
concerned in the phenomenon. Furthermore, it is well known that non- 
motile and flagella-free bacteria may be agglutinated, but the subject is 
1 Munch, med. Wchn., 1915, 62, 1173; Biochem. Ztschr., 1916, 77, 273. 
2 Jour. Immunology, 1921, 6, 201. 
3 Jour. Infect. Dis., 1911, 8, 147. 
4 Deutsch. med. Wchn., 1911, 37, 969. 
5 Deutsch. med. Wchn., 1914, 41, 241. 
6 Ztschr. f. Immunitatsf., orig., 1912, 12, 268. 
7 Arch. f. Hyg., 1912, 76, 1. 
8 Ztschr. f. Immunitatsf., orig., 1913, 16, 127. 
9 Centralbl. f. Bakteriol., etc., orig., 1913, 69, 98. 
10 Jour. Exper. Med., 1914, 19, 28. 
11 Wien. klin. Wchn., 1896, 183, 204. 
12 Ann. de l’Inst. Pasteur, 1897, No. 6. 
13 Bull. d. l’Acad. Roy. d. Med. d. Belgique, 1898, 12, 705. 
14 Ann. d. l’lnst. Pasteur, 1902, 16, 562. 
15 Trans. Cong. Amer. Phys. and Surg., 1900, 26. 260 
BACTERIAL AGGLUTININS 
of interest in view of the investigations of Smith and Reagh,1 who found 
that agglutinin acting upon the bodies of hog-cholera bacilli may be different 
from those acting upon the flagella. 
2. Paltauf’s theory is somewhat similar, he believing that the agglutin- 
ogen is precipitated on the surface of the bacteria by union with the agglu- 
tinin, with the formation of a sticky substance. He cites evidence that 
tends to show that such substances are actually thrown out from the bacteria 
during agglutination, as may be seen in a properly stained preparation in 
the form of a precipitate surrounding the bacterial cells. 
3. The presence of some salt is necessary for the occurrence of agglu- 
tination. Bordet2 found that if the salts were removed from the serum 
and from the suspension of bacteria by dialysis and that the two were 
then mixed, agglutination did not occur, but that if a small amount of 
sodium chlorid was added, agglutination promptly took place. According 
to this view, therefore, agglutination is a phenomenon of molecular physics 
—the agglutinin acts on the bacteria or other cells and prepares them for 
agglutination by altering the relations of molecular attraction between 
them and the surrounding fluid, the second phase, the loss of motility, 
clumping, etc., being brought about by the presence of salt. This second 
phase, therefore, would be a purely physical phenomenon, the salts altering 
the electric conditions of the colloidal-like agglutinin-bacterium combina- 
tion, so that their surface tension is increased. To overcome this the par- 
ticles adhere together, presenting in a clump less surface tension than if 
they remained as individual particles. Bordet cites the precipitation of clay 
as an analogous case: if a little salt is added to a fine emulsion of potters’ 
clay in distilled water the clay immediately clumps and falls to the bottom, 
the resemblance between these flakes and the clump of agglutinated bacteria 
being very striking. Support for these findings has been furnished by the 
studies of Crendiropouls and Amos,3 of Landsteiner,4 and Lange.5 The 
physicochemical nature of agglutination is also indicated by the interesting 
observation of Bond,6 who found that the agglutinin content of sera for 
erythrocytes and bacteria may be altered (generally increased) by mechanical 
processes embracing friction and pressure of dried sera. 
Specificity of Agglutinins.—For a time after their discovery the agglu- 
tinins were regarded as strictly specific, i. e., a typhoid-immune serum would 
agglutinate only typhoid bacilli and no others. Gruber early pointed out 
that an immune serum will frequently agglutinate other closely related 
organisms, although not usually to so high a degree. 
Group or partial agglutinins, therefore, refer to the presence in a serum 
of certain agglutinins that agglutinate certain other micro-organisms that 
are morphologically, biologically, and often pathogenetically closely related 
to the homologous micro-organism (the bacterium causing the infection 
or used in artificial immunization). For example, a typhoid-immune serum 
possesses, besides its greatly increased agglutinating power for Bacillus 
typhosus, some agglutinin for B. paratyphosus, notably above that of 
normal serum. This is explained by the very close biologic relationship 
of these bacteria, together with the fact that the agglutinin-producing sub- 
stance (agglutinogen) is a complex and not a single chemical substance. 
This has been explained by Durham in the following example: If the typhoid 
1 Jour. Med. Research, 1903, 1904, 10, 89. 
2 Ann. d. l’Inst. Pasteur, 1899, 13, 225. 
3 Jour. Path, and Bacteriol., 1904, 9, 260. 
4 Ztschr. f. Immunitatsf., orig., 1910, 8, 397. 
5 Ztschr. f. Immunitatsf., orig., 1915, 24, 587. 
6 Brit. Med. Jour., June 14, 1919. NON-AGGLUTINABLE SPECIES OF BACTERIA 
261 
agglutinogen is composed of various elements, A, B, C, D, it is conceivable 
that the closely related paratyphoids might contain one or more of these 
four agglutinogens and, therefore, the agglutinating power of the typhoid 
serum for a paratyphoid bacillus, though not so great as on the typhoid 
bacillus, is still considerable. Accordingly, in an infection with one micro- 
organism a specific agglutinin will be formed for that micro-organism, and 
group agglutinins for other more or less allied micro-organisms, and conse- 
quently the specificity of the agglutinating reaction depends upon the 
principle of dilution, the specific agglutinin being present in largest amount 
and operative in dilutions above the range of the group agglutinins. 
Absorption Methods for Differentiating Between a Mixed and a Single 
Infection.—In 1902 Castellani1 discovered that if the serum of an animal 
immunized against a certain micro-organism contains that micro-organism 
in large numbers the serum will lose its agglutinating power, not only for 
that micro-organism, but also for all other varieties on which it formerly 
acted. If, however, the serum contains the organism corresponding to 
the group agglutinins, the agglutinating power of the serum for the homol- 
ogous organism is reduced but little or not at all. 
In a mixed infection, due to two or more varieties of bacteria, there will 
be specific agglutinins for each of the micro-organisms and group agglu- 
tinins for each of them as well. If the immune serum is saturated with 
one of these varieties its chief or major agglutinins and some or all of the 
group agglutinins will be removed, but the major agglutinin of the second 
species will remain. On the addition of the second bacterium to the immune 
serum agglutination occurs and its agglutinin is absorbed. Park, who has 
carefully investigated this subject, finds that the absorption method proves 
that when one variety of bacteria removes all agglutinins for a second, 
the agglutinins in question were not produced by the second variety. 
Non-agglutinable Species of Bacteria.—Certain species of bacteria, 
especially when freshly isolated from the animal body, may prove them- 
selves immune to the action of agglutinins; this is especially true of the 
bacillus of Friedlander. The isolation of inagglutinable or feebly agglutin- 
ating strains of typhoid bacilli has been recorded by Achard and Bensaude,2 
Kolle,3 Johnston and Taggart,4 Sacquepee,5 Klinger,6 Lesieur,7 Buxton and 
Vaughan,8 and others. It would appear that inagglutinability arises as the 
result of an increased resistance on the part of the bacilli to antibodies 
secreted against them; there is no evidence to show that there is any rela- 
tion between this state and chronicity of infection. As a rule, this resistance 
is lost when the micro-organism is grown for some time in artificial media. 
In some instances the typhoid bacillus, when freshly isolated from a patient, 
may resist agglutination until after it has passed a period of existence on 
artificial media. This variability is probably due to some change taking 
place in the agglutinable substance of the agglutinins during the sojourn 
of the bacilli in the animal body, and possess such an excess of agglutinogenic 
receptors as to require a much larger amount of agglutinin to cause agglutin- 
ation. 
McIntosh and McQueen9 have isolated a strain of inagglutinable typhoid 
1 Ztschr. f. Hyg., 1902, 11, 17. 
2 Compt. rend. Soc. de biol., 1896, xliv, 940. 
3 Deutsch. med. Wchn., 1897, 23, 132. 
4 Montreal Med. Jour., 1897, 25, 709. 
5 Ann. d. l’Inst. Pasteur, 1901, 15, 249. 
6 Centralbl. f. Bakteriol., orig., 1902, 32, 542. 
7 Jour. d. physiol, et de path, gen., 1903, 5, 539. 
8 Jour. Med. Res., 1904, 12, 115. 9 Jour, of Hyg., 1914, 13, 409. 262 
BACTERIAL AGGLUTININS 
bacilli from a case of typhoid fever and made experiments with it, the re- 
sults of which throw light on this question of inagglutinability. They 
found, and others have made similar observations, that the inagglutinable 
strain, when injected into animals, led to the production of typhoid agglu- 
tinins which had practically no action on itself but agglutinated heterologous 
typhoid strains freely; and, furthermore, that the strain in question would 
absorb typhoid agglutinins just like the agglutinable strains. Hence, the 
inagglutinability cannot be due to loss of affinity for the agglutinins, that 
is, loss of receptors in the terms of the side-chain hypothesis, but rather 
to some physical alteration by virtue of which the second step in aggluti- 
nation fails to take place, namely, aggregation or clumping. They appear 
to have no peculiarities with respect to complement fixation, and are clumped 
by chemical agglutinants, especially acids, an observation showing that 
acid and serum agglutination do not depend on the same factors. 
It should be remembered that agglutinins act on dead as well as on 
living bacteria, those killed by heat, formalin, phenol, etc., being similarly 
agglutinable. In making the microscopic test the use of dead bacteria is 
not so satisfactory as when the test is made with living motile bacteria, 
for the influence of the serum on motility alone is of value in interpreting 
a reaction. 
Variation in Agglutinating Strength of a Serum.—In a given infection, 
such as typhoid fever, there is usually a continued increase in the amount 
of agglutinin in the blood from the fourth day until convalescence is estab- 
lished, and then a decrease occurs. It is a fact of practical importance 
that the agglutinating power of a serum may vary from day to day, so that 
it is very strong one day, and may become weak or disappear entirely on 
the next day or two. Hence, the importance of making more than one test 
in a suspicious case when the first trial has been doubtful or negative. There 
is no satisfactory explanation for this variation, mixed infection, intestinal 
hemorrhage, etc., being regarded by some as responsible for it. 
Conglutination.—In 1906 Bordet and Streng,1 and Bordet and Gay2 des- 
cribed a colloidal substance in beef-serum heated to 56° C. (“bovine col- 
loid”) which has the property of causing a characteristic clumping and 
increased lysis of red blood-cells when treated with a heated specific hemo- 
lytic serum and fresh alexin (complement). Bordet and Streng, in later 
studies on this substance, gave to it the name “conglutinin.” Streng3 con- 
tinued these studies with bacteria and found that a typical clumping was 
produced by the mixture of bacteria, fresh complement, conglutinin, and 
a specific immune serum from which the agglutinins had been removed by 
absorption. By dialyzing the beef serum the conglutinin was shown to 
be present in the globulin fraction, and the reaction took place as well with 
bacteria killed by heat or 0.1 per cent, liquor formaldehydi as with live 
organisms. 
In a study of dysentery in infants Lucas, Fitzgerald, and Schorer4 first 
applied this reaction to clinical diagnosis. They found it more sensitive 
and specific than either the agglutination or fixation test. 
In their work cultures of the Flexner and Shiga dysentery bacilli treated 
with 0.1 per cent, liquor formaldehydi wTere used. They conclude that in 
the conglutination test we have a means of diagnosis far superior to any 
other form. 
1 Centralbl. f. Bakteriol., 1909, xlix, 260. 
2 Ann. d. l’Inst. Pasteur, 1906, xx, 467. 
3 Centralbl. f. Bakteriol., 1909, 1, 47; ibid., 19C9, 2, 415. 
4 Jour. Amer. Med. Assoc., 1910, lvi, 441. ROLE OF BACTERIAL AGGLUTININS IN IMMUNITY 
263 
Swift and Thro1 did not find the conglutination reaction of much greater 
value in the differentiation of various strains of streptococci than the agglu- 
tination reaction, 
Karvonen attempted to modify the Wassermann test for syphilis by 
adopting the principles of conglutination instead of hemolysis, but numerous 
investigations by Seibert and Mironescn,2 Hect,3 Leschly and Boas,4 and 
others have shown the method to be unsatisfactory. Maltauer and Johnston5 
have recently shown that the reaction is due to fibrinogen and a heat-sensi- 
tive serum constituent; further reference to conglutination will be made 
in the chapter on Cytolvsins. 
The technic of this reaction is given on page 285. Leschly6 describes 
technical details and gives a complete review of literature. 
Role of Bacterial Agglutinins in Immunity.—The agglutinins were 
formerly regarded as possessing a true protective and curative power by 
Max Gruber and others. However, Gengou7 was unable to establish 
any relation between the agglutinative power and the refractory state of 
animals to anthrax. In his experiments it was found that human serum 
may contain large amounts of agglutinin for attenuated anthrax bacilli 
and yet man is far from being immune to anthrax. Pigeon’s serum, on the 
other hand, is free of agglutinins for anthrax bacilli and yet this animal 
enjoys a high degree of natural immunity. 
It has previously been mentioned that bacteria may be grown in a 
specific agglutinating serum, and cultures made of agglutinated bacteria 
show them to be fully alive and as virulent as before agglutination took 
place. In certain eases agglutinins for a micro-organism may be entirely 
absent and yet the animal enjoy an immunity. Bacteria that have been 
acted upon by an agglutinin are apparently not altered in appearance, 
viability, or virulence, as shown by the experiments of Issaeff8 in Metchni- 
koff’s laboratory with the pneumococcus, and by Sanarelli,9 and Mesnil10 
with vibrios and the bacillus of swine erysipelas. 
Many observations tend to show that the agglutinating power of a serum 
gives no indication of the degree of immunity that exists. For instance, 
relapses may occur in typhoid fever at a time when the agglutinating power 
of the patient’s blood is at its highest, as first shown by Widal and Sicard.11 
At present agglutinins are regarded as playing a subsidiary role in im- 
munity, their presence being of diagnostic value, and an indication of the 
presence of more important factors, and as an aid to bacteriolysis and 
phagocytosis, as first shown by Besredka12 with guinea-pigs infected with 
typhoid bacilli. 
As shown by Bull13 experimentally agglutination may occur in vivo, and 
the power of the blood to cause agglutination determines, in large measure, 
whether, after their direct introduction in an experimental way, the bacteria 
are to be removed promptly from the circulation and bacteremia avoided, 
1 Archiv. Int. Med., 1911, 7, 24. 
2 Deutsch. med. Wchn., 1911, 37. 
3 Berl. klin. Wchn., 1912, 49, No. 2. 
4 Hospitalst., 1913, 57, 1201. 
5 Jour. Immunology, 1921, 6, 349. 
6 Ztschr. f. Immunitatsf., orig., 1916, 25, 219. 
7 Archiv. Intent, d. Pharmacodyn., 1899, 6, 299; Ann. d. l’Inst. Pasteur, 1899, 13, 642. 
8 Ann. d. I’Inst. Pasteur, 1893, 7, 260. 
9 Ann. d. I’Inst. Pasteur, 1893, 7, 225. 
10 Ann. d. l’Inst. Pasteur, 1898, 12, 481. 
11 Ann. d. l’Inst. Pasteur, 1897, 11, 411. 
12 Ann. d. l’Inst. Pasteur, 1901, 15, 209. 
13 Jour. Exper. Med., 1915, 22, 475, 484; ibid., 1916, 24, 7, 25, 35; ibid., 1916, 23, 419. 264 
BACTERIAL AGGLUTININS 
or whether they are to remain and produce bacteremia. Micro-organisms 
which are not agglutinated in the normal rabbit may be made to do so by 
the intravenous injection of homologous immune serum, and this probably 
explains the rapid disappearance of pneumococci from the blood following 
the intravenous injections of homologous antipneumococcus serum. The 
bacterial clumps accumulate in the organs in which they are phagocyted. 
In all of Bull’s investigations, including instances of both natural and pas- 
sive immunity, the agglutination of bacteria in the blood of the infected 
animal was followed by phagocytosis and destruction of the bacteria in 
the viscera of the body with a subsequent disappearance of the bacteria 
from the blood-stream. It wras observed, however, that when agglutination 
was incomplete, a few micro-organisms remained in the blood, later to multiply 
and cause a fatal bacteremia. In this way it appears that agglutinins, 
opsonins, and phagocytosis are closely related and probably exert an im- 
portant role in resistance to, and recovery from, infection. 
The agglutination reaction is used for the following purposes: 
1. For the diagnosis of disease, by identifying the bacterial infection 
from which the patient is suffering. To do this satisfactorily we must 
have on hand stock cultures of bacteria, and test the patient’s serum for 
agglutinins for these bacteria. For instance, if a patient presents symp- 
toms of typhoid fever, the serum is tested for typhoid agglutinins; ff the 
reaction is very weak or negative and continues so the serum is further 
tested for agglutinins for Bacillus paratyphosus A and B. 
2. Agglutination reactions are also of value as an aid to the identifi- 
cation of a micro-organism that has been cultivated from a patient. For 
this purpose we must have on hand various standard immune serums. 
For example, if a bacillus resembling the typhoid bacillus is isolated from 
the feces of a patient the diagnosis may be aided by a positive agglutina- 
tion reaction with typhoid immune serum. Reference has been made to 
the use of the test for the differentiation of pneumococci. The agglutina- 
tion test is also of great value in the identification of meningococci, and in 
the differentiation of the three groups of this micro-organism. Similar 
studies have been made in the serologic grouping of streptococci and gono- 
cocci to which further reference is made in the chapter on Serum Therapy. 
3. Agglutination tests are of value in determining whether in a case 
in which more than one micro-organism has been cultivated the condi- 
tion at hand is a single or a mixed infection. The absorption agglutinin 
test is made with the patient’s serum and the cultures are isolated from 
the patient. 
4. Agglutination tests are also of value for measuring the immunizing 
response that a patient is making to his infection or to artificial immuniz- 
ation. Thus, the test is of some value in determining the response to inocu- 
lation with typhoid vaccine, although it is probable that the agglutinin 
itself does not possess true antimicrobic properties. 
Agglutination in Typhoid and Paratyphoid Fevers.—In this group the 
agglutination test has proved of great value in diagnosis; probably the 
microscopic method employing serum or dried blood is mostly employed, 
but recent experiences with the macroscopic technic employing cultures 
prepared according to the Dreyer method have indicated that the latter 
is more delicate and more reliable and accurate. 
In typhoid fever the Gruber-Widal reaction may be positive as early as 
PRACTICAL APPLICATIONS PRACTICAL APPLICATIONS 
265 
the third day; usually, however, the positive reaction is obtained somewhat 
later—about the seventh or the eighth day. A day or so earlier the bacilli 
used in making the test may be seen to lose their motility and two or three 
may form a loose clump. This is the doubtful reaction, and it is well to 
test every day or every other day until a decisive reaction is obtained. 
According to Park, “about 20 per cent, of typhoid infections give posi- 
tive reactions in the first wTeek; about 60 per cent, in the second week; 
about 80 per cent, in the third week; about 90 per cent, in the fourth week, 
and about 75 per cent, in the second month of the disease.” In about 90 
to 95 per cent, of cases in which repeated examinations are made a positive 
reaction is to be found at some time during the patient’s illness. Moreschi1 
found that in 6.7 per cent, of cases agglutinins may not be found. 
Occasionally the reaction appears first during the stage of convalesc- 
ence, and at times it may even be absent, the diagnosis being confirmed 
by cultivating typhoid bacilli from the blood. The possibility of a given 
case reacting strongly one day and weakly or entirely negative a day or 
so later has been emphasized elsewhere. Dreyer and Walker2 found that 
1 the agglutinins reached the maximum in typhoid and paratyphoid fever 
between the sixteenth and twenty-fourth day of the disease. 
Usually the reaction is strongest during convalescence, remains posi- 
tive for several weeks, and then gradually returns to the normal. Occa- 
sionally the reaction remains positive for months or even years after the 
attack of typhoid fever; many such cases are “carriers” and harbor the 
bacilli in the gall-bladder, although the person appears to be quite well. 
Only very rarely does normal serum immediately agglutinate typhoid 
bacilli in a dilution higher than 1 : 10; where a time limit of one to two 
hours is given a few may show some agglutination in dilutions up to 1 : 30. 
If the typhoid bacillus is agglutinated by the patient’s serum in a dilu- 
tion of 1 : 100, or at least 1 : 30, the Widal reaction may be regarded as 
positive. It is not safe to use lower dilutions, as occasionally the serum of 
healthy persons may agglutinate Bacillus typhosus in dilutions up to 1 : 25. 
Ritchie3 has studied the normal agglutinins in human sera with much 
care and states: 
Complete agglutination in a dilution of 1 : 16 should be looked on with 
considerable suspicion; complete agglutination in dilution of 1 : 32 or above 
should be looked on as diagnostic. The same dilutions were given for the 
paratyphoid bacilli. 
Due care must be exercised not to mistake a pseudoreaction about 
detritus for true agglutination. 
Positive reactions are occasionally obtained in other diseases—acute 
miliary tuberculosis, malaria, malignant endocarditis, and pneumonia. It is 
also -well to bear in mind the possibility of a patient having been vaccinated 
against typhoid at some early date with resulting agglutinin formation. 
The agglutination test is of particular value as an aid in the diagnosis of 
the mild typhoid or paratyphoid infections so apt to be mistaken for some 
other disease and the detection of which is of great importance to the com- 
munity. Blood culture should be performed at once and agglutination tests at 
frequent intervals. 
Owing to the similarity of symptoms an infection with Bacillus para- 
typhosus A and B may be difficult to distinguish from typhoid fever. This 
difficulty is increased by the confusion of the Widal reaction owing to the 
1 Ztschr. f. Immunitatsf., orig., 1914, 21, 410. 
2 Lancet, London, 1916, September 2. 
3 Lancet, London, 1916, 1, 1245. 266 
BACTERIAL AGGLUTININS 
presence of group agglutinins in the serum if proper dilution is not practised. 
Bacillus A and Bacillus B are not identical in their agglutinable properties, 
the latter being more closely related to the typhoid bacillus than the former. 
In this country B. paratyphosus is usually held responsible for paratyphoid 
fever. Conclusions should not be drawn until tests have been made with 
both strains of the paratyphoid bacillus and with the typhoid bacillus. 
In conducting the tests for paratyphoid fever strains of both B. para- 
typhosus A and B should be employed; positive reactions With dilutions of 
1 : 30 or higher are significant and in most instances diagnostic, providing 
the patient has not previously received triple typhoid-paratyphoid vaccine. 
Dreyer1 believes that an agglutination of 1 in 10 is for all practical purposes 
diagnostic in cases of paratyphoid A. In case of mixed infection the ab- 
sorption method of Castellani will serve to clear up the diagnosis. 
Agglutination Reactions for the Detection of Typhoid Carriers.—Bigelow2 
and others have shown that a persistently high agglutinin titer of the serum 
after typhoid fever indicates a carrier condition; the agglutination reaction 
may prove of value in the detection of carriers in addition to bacteriologic 
examination of the bile, -feces, and urine. Sequelas of typhoid fever, as 
periostitis and cholecystitis, are usually accompanied by a persistently high 
titer of the serum in agglutinins. 
Agglutination Reaction After Typhoid-paratyphoid Immunization.—It 
is well established that the administration of typhoid or mixed typhoid- 
paratyphoid vaccine is followed by the production of agglutinins. Even 
a single dose of vaccine results in agglutinin production; after two or three 
doses the agglutinating titer of the serum is usually, but not always, further 
increased. 
The production of agglutinins by vaccines is of great importance in rela- 
tion to the use of the agglutination test for diagnosis purposes; the physician 
should always inquire into this phase of a patient’s history. If vaccines 
have been received the ordinary Widal test is apt to be worthless in diag- 
nosis; more accurate quantitative agglutination reactions are required as 
described below. 
Vaccinated individuals if quite recently inoculated will usually show a 
high titer of specific agglutination. A rapid rise in titer sets in within two to 
four days of inoculation. This is followed by a fall at first rapid, but subse- 
quently becoming very slow so that a relatively high titer is maintained 
for a long period (even for years). During this period examinations made 
at intervals of a few days give practically identical readings. 
Dreyer and Inman,3 Dakeyne,4 and Krumbhaar and Smith5 found that 
the agglutinins after vaccination persist for at least eight to twelve months 
and frequently for longer periods. However, the rate of disappearance 
of these agglutinins varies greatly in different persons. As stated by Meyer 
and Kilgore6 on the basis of their own work and a review of the literature 
up to 1917, it is not possible to make any statement in regard to the rate 
with which agglutinins disappear from the blood of vaccinated individuals. 
It is entirely likely that immunity to typhoid fever persists for some 
time after the agglutinins in the blood have decreased to normal propor- 
tions. It is customarily stated that vaccination protects for two or three 
years. Not infrequently individuals inquire at the expiration of this time 
1 Proc. Roy. Soc. Med., 1915, Medical Section, 10. 
2 Jour. Amer. Med. Assoc., 1911, 57, 1418; ibid., 1912, 58, 339. 
3 Lancet, London, 1915, 2, 225. 
4 Lancet, London, 1915, 2, 540. 
5 Jour. Infect. Dis., 1918, 23, 126. 
6 Archiv. Int. Med., 1917, 19, 293. PRACTICAL APPLICATIONS 
267 
if revaccination should be done; it is my practice to first titrate the agglu- 
tinin content of the blood and advise revaccination only in case the titer 
for typhoid bacilli is 1 : 30 or less, and for the paratyphoid bacilli if the 
titer is 1 : 20 or less. 
Agglutination Reactions in the Diagnosis of Typhoid and Paratyphoid 
Fevers in Vaccinated Individuals.—The wide-spread use of typhoid-para- 
typhoid immunization has introduced many new difficulties in the use of 
the agglutination test for the diagnosis of typhoid and paratyphoid fevers 
among immunized individuals. Blood, feces, and urine cultures should 
always be resorted to if possible for diagnosis, but these may show the 
presence of the bacilli in about slightly more than 50 per cent, of cases. 
The ordinary Widal test is almost worthless under these circumstances 
unless it is definitely known that the agglutinins had dropped to normal 
levels prior to the onset of enteric symptoms. Under these circumstances 
a more accurate quantitative method is required, employing a culture of 
uniform and unchanging properties as the formalized suspensions prepared 
by the Dreyer method, in order that the results of three or more successive 
observations shall be strictly comparative. 
Tidy1 has claimed that febrile conditions destroy the agglutinins pro- 
duced by vaccination, and that “ a positive agglutination reaction to Bacillus 
typhosus after the fifth day of pyrexia is a definite proof of typhoid fever 
in an inoculated man as in a non-inoculated one.” Dreyer and Walker,2 
Donaldson and Clark,3 Wilson,4 and others have challenged these state- 
ments and have apparently proved that Tidy’s comments are erroneous. 
Indeed, Conradi and Bieling5 have stated that just the reverse may occur, 
and that intercurrent infections, as tuberculosis, may actually increase the 
typhoid agglutinins to a slight degree and that agglutination tests may 
lead in this way to an erroneous diagnosis. The results reported by Perry6 
and other English investigators referred to above have indicated, however, 
that a definite increase of agglutinins detected by an accurate method is 
due to enteric infection. 
It follows that in the case of inoculated persons the diagnosis of active 
typhoid (or paratyphoid) infection will require two or more successive 
examinations of the serum. 
(a) If the individual is suffering from active typhoid infection his titer 
of typhoid agglutination will exhibit the usual rise and subsequent 
regular fall seen in non-inoculated subjects, but starting from and 
returning toward the higher base line of inoculated persons. 
(b) If the individual is suffering from active paratyphoid infection one 
of three things may occur as regards his typhoid agglutination titer, 
namely: 
1. No appreciable change may occur in the titer of typhoid agglutin- 
ation. 
2. A relatively slight rise may occur followed by a fall toward the 
former level. 
3. A marked rise may occur synchronous with the rise in paratyphoid 
agglutination titer, and subsequently followed by the usual fall 
toward the former level. 
lancet, London, 1916, 1, 241. 
2 Lancet, London, April 8 and September 2, 1916; March 10, 1917. 
3 Lancet, London, 1916, 2, 546. 
4 Lancet, London, 1917, 1, 263. 
5 Deutsch. med. Wchn., 1916, 42, 1280. 
6 Lancet, London, 1918, 1, 593. 268 
BACTERIAL AGGLUTININS 
Meanwhile the titer of paratyphoid agglutination runs the normal course 
of rapid rise to a maximum (usually exceeding the maximum typhoid titer), 
followed by a fall, at first rapid and then slower, as already described for 
typhoid subjects, and falling below the persistent base line of typhoid agglutin- 
ation of inoculated persons. 
In the case of mixed infections, whether in inoculated or non-inoculated 
persons, the agglutinin curves for the different infecting organisms are 
usually not synchronous, and they pursue their ordinary course independently 
of each other. 
Agglutination in Bacillary Dysentery.—In dysentery the agglutination re- 
action with the serum of patients shows great variability. In spite of the 
presence of bacilli in the feces the reaction is sometimes absent, often dis- 
appears rapidly during convalescence, and rarely is as high as in typhoid 
fever. The tests should always be performed with both the “Flexner” and 
“Shiga” types of bacilli, as the two do not possess identical agglutinable 
properties, and either may be the cause of infection in a given case. The 
absence of the reaction does not exclude a dysenteric infection. 
Martin, Hartley, and Williams1 tested the serum from 151 cases of 
dysentery from whose stools dysentery bacilli had been isolated, and found 
an increase of agglutinin in slightly less than 40 per cent, of those infected 
with the Flexner-Y group of bacilli. In all cases infected with the Shiga 
strain, however, marked agglutination was observed, indicating that the 
agglutination test possesses most value in these infections. Ritchie2 states 
that complete agglutination of Bacillus dysenteriae (Shiga) in dilution of 
1 : 64 or higher is diagnostic, but with the Flexner strains agglutination 
should be complete in dilutions as high as 1 : 128 to be of diagnostic import. 
Friedman and Steinbock,3 Soldin,4 Dunner,5 Ledingham and Penfold6 have 
found agglutination tests of diagnostic value in infections among soldiers 
in the recent war. 
Agglutination in Glanders.—In veterinary practice agglutination reac- 
tions are of value in the diagnosis of glanders, infected horses reacting in 
some instances to dilutions as high as 1 : 2000. For diagnostic purposes 
the agglutination test in glanders must be in dilutions higher than 1 : 800. 
A positive reaction in dilutions of 1 : 1000 is regarded as suggestive, and 
is controlled by a complement-fixation test; agglutination in dilutions of 
1 : 1500 practically always indicates an infection. The complement-fixa- 
tion test, however, is a better diagnostic reaction. 
Povitsky7 has recently described an improved method for the prepara- 
tion of the bacterial suspension and conduct of the test. 
In the diagnosis of glanders among human beings a positive reaction 
in dilution of 1 : 100 or higher is considered positive, normal serum not 
reacting above 1 : 50. 
Agglutination in Typhus Fever.—Weil-Felix reaction: In 1915 Weil and 
Felix8 isolated from the urine of typhus fever patients two strains of Bacillus 
proteus vulgaris which they found were agglutinated by the sera of individuals 
suffering with this disease. Later they succeeded in isolating a third strain 
(X 19) which proved more susceptible and has been extensively employed 
in agglutination tests. Friedberger9 believed that this bacillus may be 
the etiologic agent of typhus fever, but subsequent investigations by Land- 
1 Brit. Med. Jour., 1918, 1, 642. 
2 Lancet, London, 1916, 1, 245. 
3 Deutsch. med. Wchn., 1915, 42, 213. 
4 Deutsch. med. Wchn., 1915, 41, 845. 
5 Berl. klin. Wchn., 1915, 52, 1177. 
6 Brit. Med. Jour., 1915, 1, 37. 
7 Jour. Immunology, 1918, 3, 463. 
8 Wien. klin. Wchn., 1916, 29, 33, 927. 
9 Deutsch. med. Wchn., 1917, xliii, Nos. 42-44. AGGLUTINATION IN OTHER DISEASES 
269 
steiner and Hausmann,1 Doerr and Pick,2 Mollers and Wolff3 have not 
supported this claim. However, the bacillus is agglutinated by the sera of 
the majority of individuals with typhus fever according to the reports of 
Ribeyo,4 Cancik,5 Dietrich,6 Dienes,7 Sacquepee,8 Braun,9 Jacobitz,10 Oet- 
tinger,11 Schultz,12 and others. 
In applying the agglutination test for diagnosis the dilutions should 
range from 0 : 20 upward; agglutinating at 1 : 20 is regarded as suspicious. 
In typical cases agglutination may occur in dilutions as high as 1 : 250 
to 1 : 1000. Agglutinins are first detected about the sixth day of the disease 
and rapidly disappear after the crisis. 
Not all strains of Bacillus proteus are suitable for this reaction; indeed, 
the majority are unsuitable and laboratories conducting the test should 
secure strain X 19 for the work. Furthermore, this bacillus has been found 
in only a small percentage of cases of typhus fever. It is highly probable 
that the virus of the disease lowers resistance to Bacillus proteus and other 
bacilli of the colon group normally inhabiting the gastro-intestinal tract, 
thereby favoring antibody production. Analogous conditions are seen in 
scarlet fever favoring the streptococcus and poliomyelitis favoring various 
micrococci. As early as 1909 Wilson13 showred that heterologous agglutinins 
for colon and other intestinal bacteria were produced during typhus fever. 
As the subject stands today it would appear that during typhus fever 
there may be the production of agglutinins for Bacillus proteus and other 
bacteria, but especially for strain X 19 of B. proteus, and that the agglutina- 
tion test with this organism yields reactions of diagnostic value. Wolff14 has 
recently subscribed to Epstein’s theory that B. proteus vulgaris originating 
in the intestinal tract undergoes, through symbiosis with the typhus virus, 
a change in its internal structure and becomes agglutinable by the serum in 
typhus fever. The reaction is not specific and requires more investigation 
for establishing its exact status. 
AGGLUTINATION IN OTHER DISEASES 
In cholera the agglutination test has so far proved of doubtful aid in 
establishing a diagnosis of the disease. However, for the purpose of recog- 
nizing bacilli isolated from the feces of suspicious cases the reaction with 
known immune serum is of great value. 
In cerebrospinal meningitis the agglutination occurs within an hour in 
dilutions of 1 : 10. It is seldom that the patient’s serum agglutinates in a 
dilution higher than 1 ; 50. 
In tuberculosis the agglutination reaction is regarded as having little or 
no value as a diagnostic procedure. Koch recommended the agglutination 
test for the estimation of the degree of immunity conferred by tuberculin 
treatment. As pointed out elsewhere, agglutinins have apparently no anti- 
1 Med. Klin., 1918, 14, 515. 
2 Wien. klin. Wkhn., 1918, 31, 820. 
3 Deutsch. med. Wchn., 1919, No. 13, xlv. 
4 Cronica Med., 1919, 36, 75. 
5 Wien. klin. Wchn., 1916, 29, 1552. 
6 Deutsch. med. Wchn., 1916, 42, 1570. 
7 Deutsch. med. Wchn., 1919, 45, 14; Ztschr. f. Immunitatsf., orig., 1919, 25, 447. 
8 Bull. d. 1. Soc. Med. d. Hop., 1919, 43, 151. 
9 Centralbl. f. Bakteriol., orig., 1918, 8, 20, 475. 
10 Centralbl. f. Bakteriol., orig., 1918, 81, 251. 
11 Centralbl. f. Bakteriol., orig., 1918, 80, 304. 
12 Amer. Jour. Med. Sci., 1921, clxi, 78. 
13 Jour. Hyg., 1910, ix, 316; ibid., 1910, x, 155; ibid., 1920, 19, 115. 
14 Berl. klin. Wchn., 1920, 57, 834. 270 
BACTERIAL AGGLUTININS 
microbic influence, but, as with typhoid vaccination, may indicate the de- 
gree of reaction and the presence of other antibodies. 
Many strains of tubercle bacilli are almost non-agglutinable. The 
preparation of a homogeneous emulsion is not easily made, and the results 
are likely to be confusing and contradictory. Recently, however, Larson, 
Nelson, and Chang1 have described a method for preparing antigen by 
subjecting tubercle bacilli to the influence of carbon dioxid under high 
pressure, and believe that the test possesses diagnostic value. 
In plague the agglutination reaction becomes quite marked about the 
ninth day of the disease—too late, however, to be of much practical value 
in diagnosis. It is occasionally useful, however, for deciding whether a 
patient in the convalescent stage has really suffered from the disease. 
In Malta fever the agglutination reaction is of considerable value in 
making the diagnosis. Salvatore2 states that agglutination at 1 : 40 may 
be regarded as specific for Micrococcus melitensis if the typhoid reaction is 
negative. Agglutination should be positive in dilutions of 1 : 100 to be of 
significance in diagnosis among goats and other cattle. 
In pneumonia the reaction is of value in rapidly differentiating pneu- 
mococci and as an aid in specific serum therapy (see page 286). 
In syphilis agglutinins for Treponema pallidum by the sera of rabbits in- 
jected with a living and heat-killed culture furnished by Noguchi were first 
described by Kolmer3; Nakano,4 Kissmeyer,5 and Zinsser and Hopkins6 
have also described the agglutination of culture pallidum by immune serum. 
Kolmer, Broadwell, and Matsunami7 found that equal parts of normal human 
serum and pallidum culture furnished by Zinsser may result in partial 
agglutination, whereas with sera of syphilitics in the later stages agglutina- 
tion in dilutions of 1 : 5 and higher occurred with about 84 per cent, of sera, 
but that further studies are necessary to establish the practical value of 
agglutination in the diagnosis of syphilis. 
In pertussis agglutination tests have been found of some value in clinical 
diagnosis by Wollstein,8 Frankel,9 Seiffert,10 and Arnheim; Bordet11 finds it 
of value only when the micro-organism is freshly isolated from human 
sputum and grown on rich blood medium. According to Povitsky12 and 
Worth13 a dilution of serum not less than 1 : 200 is necessary for a practical 
positive diagnosis of pertussis, as normal human serum may agglutinate in 
dilutions up to 1 : 100. 
In sporotrichosis agglutination reactions have been described by Widal,14 
Davis,15 and others and are believed to possess diagnostic value, agglutina- 
tion occurring in dilutions of from 1 : 300 to 1 : 800 in an hour; Moore 
and Davis16 have recently reported favorably upon the specificity and diag- 
nostic value of agglutination and complement-fixation reactions in sporo- 
1 Proc. Soc. Exper. Biol, and Med., 1922, 19, 359. 
2 Policlinico, 1914, 20. 
3 Jour. Exper. Med., 1913, xviii, 18. 
4 Archiv. Dermat. u. Syph., 1913, cxvi, 265. 
5 Deutsch. med. Wchn., 1915, xli, 306. 
6 Jour. Exper. Med., 1915, xxi, 576. 
7 Jour. Exper. Med., 1916, xxiv, 333. 
8 Jour. Exper. Med., 1909, xi, 41. 
9 Munch, med. Wchn., 1908, lv, 1683. 
10 Munch, med. Wchn., 1909, 1561. 
11 Berl. klin. Wchn., 1908, xlv, 1453. 
12 Centralbl. f. Bakteriol., orig., 1912, lxvi, 276. 
13 Archiv. Int. Med., 1916, xvii, 279. 
14 Ann. d. l’Inst. Pasteur, 1911, 24. 
15 Jour. Infect. Dis., 1913, 12, 140. 
16 Jour. Infect. Dis., 1918, 23, 252. AGGLUTINATION IN OTHER DISEASES 
trichosis and point out that these tests are of much less value in the diag- 
nosis of blastomycosis, which is a closely related disease. 
Agglutination reactions have also been reported in leprosy by Harris 
and Lanford,1 Nakajo,2 and others. In dourine and other trypanosome in- 
fections by Mattes3 and Wehrbein,4 and in malaria by Biglieri.5 
The “group agglutinins” constitute a source of difficulty in making the 
differentiation among the numerous members of a group of micro-organisms, 
but if a highly potent agglutinating serum is used and the test is carried 
to the point of determining the highest dilution that will agglutinate the 
bacteria, it will in most cases be possible to differentiate the variously allied 
micro-organisms by this test. 
In conducting these reactions it is best to use a macroscopic method, 
and the agglutinating serum used must have been previously titrated against 
an easily agglutinable and known strain of the micro-organism in question. 
In this connection it is well to remember that freshly isolated cultures of a 
micro-organism may be not at all or but very slightly agglutinable. Thus, 
colonies of typhoid bacilli found in feces or in an abscess may, if picked from a 
plate, resist agglutination until subcultured several times in artificial media. 
The agglutination test has great value as a mode of differentiating 
among the members of the typhoid-colon group of bacilli. In the diagnosis 
of cholera, suspicious bacilli isolated from the feces may be tested with a 
known cholera immune serum, and the bacteriologic diagnosis thus be 
greatly facilitated. The test also has some value in making a biologic 
differentiation between meningococci and gonococci, and also between 
other groups of bacteria. 
Production of Immune Agglutinins.—In the preparation of immune sera 
for diagnostic agglutination reactions, rabbits are generally employed. The 
micro-organisms may be injected intravenously, intraperitoneally, or sub- 
cutaneously; as shown by McFarland,6 working with Bacillus coli, the route 
of injection has little or no influence upon agglutinin production. Intra- 
venous injections usually produce agglutinins more quickly and in larger 
amounts than other routes. 
Due care must be exercised against the injection of toxic amounts of 
culture. As shown by Perry and Kolmer,7 the administration of living 
cultures of typhoid bacilli produce most agglutinin. It is a good plan to 
start immunization with heat-killed suspensions, and as antibodies are 
produced living cultures may be used. The injections should be at intervals 
of five to seven days and the blood tested at intervals; final bleeding may 
be done about seven to nine days after the last injection. 
1. Use forty-eight-hour agar cultures of the organism, such as Bacillus 
typhosus, Spirillum choleras, etc. Bouillon cultures may be used, but are 
not recommended on account of the various other constituents present in 
the medium. 
2. With a sterilized 4-mm. platinum loop remove 1 loopful of culture, 
and rub up in 2 c.c. of sterile salt solution in a small test-tube until a homo- 
geneous emulsion is secured. 
3. Heat the emulsion for thirty minutes at 60° C. in a water-bath. 
4. Inject intravenously. 
271 
1 Jour. Med. Research, 1916, 34, 157. 
2 Jour. Infect. Dis., 1915, 17, 388. 
3 Centralbl. f. Bakteriol., 1912, 65, 538. 
4 Jour. Infect. Dis., 1915, 16, 461. 
5 Wien. klin. Wchn., 1915, 28, 1049. 
6 Ztschr. f. Immunitatsf., orig., 1911, 9, 451. 
7 Jour. Immunology, 1918, 3, 247. 272 
BACTERIAL AGGLUTININS 
5. Give four more injections at intervals of five days as follows: 
Second dose: 2 loopfuls in 2 c.c. NaCl solution, heated. 
Third dose: 4 loopfuls in 2 c.c. NaCl solution, heated. 
Fourth dose: 6 loopfuls in 2 c.c. NaCl solution, heated. 
Fifth dose: 1 agar slant in 4 c.c. NaCl solution, heated. 
Sixth dose: 1 agar slant in 4 c.c. NaCl solution, unheated. 
6. One week after the last injection has been made the blood is tested, 
and if found of satisfactory titer, the animal is killed and the serum se- 
cured. If the titer is found too low, one or more additional injections are 
given. 
Intraperitoneal Method (Rabbit).—1. Same as the preceding, except- 
ing that larger doses are given. 
First dose: 2 loopfuls in 4 c.c. NaCl solution, heated. 
Second dose: 4 loopfuls in 4 c.c. NaCl solution, heated. 
Third dose: 6 loopfuls in 4 c.c. NaCl solution, heated. 
Fourth dose: 1 agar slant in 5 c.c. NaCl solution, heated. 
Fifth dose: 1 agar slant in 5 c.c. NaCl solution, unheated. 
Sixth dose: 1 agar slant in 5 c.c. NaCl solution, unheated. 
2. The blood is tested one week after the last injection has been made. 
TECHNIC OF BACTERIAL AGGLUTINATION REACTIONS 
Two methods may be employed: 
1. The microscopic method which is generally employed where the Gruber- 
Widal reaction for typhoid fever has been employed as the reaction is quickly 
done and requires but a small amount of blood. 
This test is usually performed with serum separated from the clot and 
in various dilutions (wet method). The test may also be performed with 
dried blood (dry method), the agglutinins being preserved and redissolved 
with a diluent. The technic of the latter method is very simple. The blood 
is easily collected and may be sent for long distances, and for these reasons 
the method has been adopted by many boards of health. 
2. The macroscopic method is that generally preferred if sufficient blood 
is on hand, and is the method of choice in scientific research. Absorption 
tests must be performed with the macroscopic technic. 
Macroscopic methods were first introduced by Wright1 in 1897 as a 
substitute for Widal’s microscopic methods. Madsen and Jorgensen,2 how- 
ever, were first to devise an accurate quantitative method, although their 
technic has been long since superseded by simpler and equally accurate 
methods. Possibly the best and most generally useful method for accuracy, 
precision, simplicity, and safety is that devised by Dreyer,3 which has proved 
so valuable for diagnostic purposes during the recent war. 
Advantages and Disadvantages of Microscopic Methods.—The advan- 
tages are: (1) The reactions may be read within an hour after the tests are 
set up, thereby yielding quick results; (2) only small accounts of serum are 
required. 
The disadvantages are: (1) The living bacterial emulsions or broth cul- 
tures employed are likely to vary greatly in density from day to day and 
thereby reduce the accuracy of the tests. (2) Cultures are likely to vary in 
age and sensitiveness to agglutination which greatly influence accuracy. 
1 Brit. Med. Jour., 1897, 1, 139. 
2 Festkrift ved Indv. a. Stat. Seruminst., Copenhagen, 1902. 
3 Hospitalst., Copenhagen, 1906. Brit. Med. Jour., 1904, 2, 564. Jour. Path, and 
Bacteriol., 1909, 13, 332; ibid., 1906, 1. TECHNIC OF MICROSCOPIC AGGLUTINATION TEST 
273 
(3) Slight differences in the constitution of the culture medium employed 
may influence the accuracy of agglutination tests with living cultures. (4) 
It is not usually possible to conduct the tests with as great quantitative 
accuracy as the macroscopic test. (5) Weakly positive indefinite reactions 
are more difficult to read, and more subject to error in interpretation than 
macroscopic reactions. 
Advantages and Disadvantages of Macroscopic Methods.—The advan- 
tages are: (1) The technic permits the use of formalized cultures prepared 
after the method of Dreyer, which are highly sensitive to agglutination 
and of uniform density. (2) The bacterial emulsion prepared by this method 
is sterile and thereby safe. (3) The emulsion may be kept under suitable 
conditions for a year or longer and is always ready for use. (4) The method 
is more accurate and very simple. (5) The readings are quite sharp and 
definite and readily made. (6) Most important of all, repeated agglutination 
tests with the serum of the same individual may he rendered more uniform than 
is usual with microscopic tests; this is of particular value in the diagnosis of 
typhoid and paratyphoid fevers among individuals who have received vaccines. 
The only disadvantage is: The longer time required before readings 
may be made (two hours for typhoid reactions; four hours for dysentery; 
but sharper readings require twenty-four hours). An additional disadvan- 
tage for Board of Health laboratories is the difficulty in securing a sufficient 
supply of blood for serum. 
I am convinced that the macroscopic technic is the method of choice 
and especially in the diagnosis of intestinal infections among vaccinated 
individuals. The work of Dreyer and his associates has proved the super- 
lative merits of the method not only for research work but for routine ex- 
aminations as well. 
Technic of the Microscopic Agglutination Test (Wet Method).—The Widal 
Reaction in Typhoid Fever 
1. The bacterial emulsion should be prepared of young cultures, should 
be homogeneous and free from clumps, and of such density as to furnish 
a sufficient number of micro-organisms to give the reaction (Fig. 98). 
For the ordinary microscopic Widal test, eighteen to twenty-four-hour 
bouillon cultures of Bacillus typhosus, B. coli, and B. paratyphosus may 
be employed. An old laboratory culture—one that is known to be agglu- 
tinable—should be used. Broth cultures should be cultivated at a tempera- 
ture lower than body heat in order that long motile forms may be secured. 
During the summer and early autumn months the culture can be grown 
at room temperature; during the winter, on top of the incubator. 
Thick cultures are unsatisfactory for making the microscopic test, as 
there is always some false clumping and motility is not well marked (Fig. 
99). 
When these tests are done routinely, it is good practice to subculture 
in broth every day in order that a satisfactory culture may always be on 
hand. When performed at irregular intervals, a broth culture can be pre- 
pared from a stock agar culture and the test performed twenty-four hours 
later. 
Emulsions may be prepared of young cultures on solid media by re- 
moving portions of the growth with a platinum loop and emulsifying in a 
diluent, such as normal salt solution or broth. This may be performed by 
placing the diluent in a test-tube and rubbing the loop over the glass just 
at the margin of the fluid, the bacteria being gradually emulsified and 
floated into the diluent. The emulsion is gently shaken and removed to a 274 
BACTERIAL AGGLUTININS 
second tube, when unresolved bacterial clumps will sink to the bottom. 
In other cases the emulsion may be centrifuged for a short time or filtered 
through sterile filter-paper. Sufficient salt solution is added to give the 
emulsion a density equal to that of a rich twenty-four-hour bouillon culture. 
2. Sufficient blood may be obtained by pricking a finger and filling a 
Wright capsule; or 0.5 to 1 c.c. may be collected in a small test-tube. 
After standing a few hours the serum may be pipeted off the clot; or 
immediately after coagulation the clot may be broken up and the serum 
secured by centrifuging. 
The serum should be fresh, clear, and free from corpuscles. 
Liebermann and Acel1 have described a method of collecting 2 drops 
of blood in 1 c.c. of distilled water which hemolyses the red blood-corpuscles 
and gives an approximate dilution in 1 : 20. 
3. Dilute the patient’s serum by placing 1 drop from a capillary pipet 
in a small watch-glass and adding 19 drops of normal salt solution. This 
gives a dilution of 1 : 20. Mix thoroughly. 
Fig. 98.—A Satisfactory Culture for the 
Microscopic Agglutination Reaction. 
X 430. 
This shows a satisfactory culture of the 
proper density and free of clumps of bacilli. 
(Twenty-four-hour culture of Bacillus typhosus 
grown at room temperature.) 
Fig. 99.—An Unsatisfactory Culture for 
the Microscopic Agglutination Reaction. 
X 430. 
The culture is rather too dense and shows 
considerable spontaneous or false agglutination 
of the bacilli. (Twenty-four-hour culture of 
Bacillus typhosus grown at 37° C.) 
Because normal agglutinins may be active in dilutions as high as 1 : 20, 
for diagnostic tests in typhoid fever the serum should not be diluted lower 
than 1 : 10 (final dilution 1 : 20). For routine work dilutions of 1 : 50 
and 1 : 100 are well adapted for the microscopic test. Dilutions of 1 : 40 
and 1 : 80 are readily made and are equally useful. 
4. With a 3 or 4 mm. platinum loop place a drop on a clean cover-glass 
that is sufficiently thin to permit the use of an oil-immersion lens. The 
loop is better than a capillary pipet because the drop it gives is smaller,- 
and when it is later diluted with an equal quantity of bacterial emulsion 
it is not too large and is easily manipulated. 
5. With the same sterilized platinum loop add 1 loopful of a twenty- 
four-hour broth culture of Bacillus typhosus to the drop of diluted serum 
on the cover-glass. Mix gently and without spreading the drop. This 
gives a final dilution of 1 : 40. 
1 Deutsch. med. Wchn., 1914, 40, 2057. TECHNIC OF MICROSCOPIC AGGLUTINATION TEST 
275 
6. Edge a hanging-drop slide with vaselin, and invert the cover-glass 
slide over the hollow portion in such a manner that the drop will be sus- 
pended in its center. Care must be exercised not to spread the drop, for 
if this occurs and the fluid flows around the margins of the chamber a new 
preparation must be made. Inspect the slide, and add vaselin, if neces- 
sary, until it is sealed tightly. By means of a grease pencil label the slide 
with the name of the patient, the dilution, and the time when the prepara- 
tion was made. 
7. Place 5 drops of serum dilution 1 : 20 in a second watch-glass, and 
add an equal quantity of normal salt solution. Mix well. This gives a 
dilution of 1 : 40. 
8. Prepare a second slide by mixing a loopful of this dilution with an 
equal sized loopful of culture. Mix gently. This gives a final dilution of 
1 : 80. Mark the slide with the name, dilution, and the time. 
9. Prepare a third slide by placing a loopful of culture on a cover-glass 
and invert over a concave slide to which vaselin has been applied in the 
usual manner. This is the culture control. Label the slide. 
10. Place the slides in a dark place 
at room temperature and examine at 
the end of an hour with the | or oil- 
immersion lens. 
(a) First inspect the control. The 
bacilli should not be clumped, but 
should be motile, and preferably in 
the form of long slender rods (see Fig. 
98). 
(b) Examine the 1 : 40 and 1 : 80 
dilution preparations: a positive re- 
action is indicated by loss of motility 
and definite clumping (Fig. 100). A 
few free motile bacilli may be seen, 
or a clump may be seen to move, 
owing to the efforts of the bacilli to 
break away. A doubtful reaction is in- 
dicated by a partial loss of motility 
and a few indefinite clumps. A nega- 
tive reaction is indicated when there is 
no loss in motility or no clumping, or 
when the reactions resemble the control to which no serum has been added. 
In reporting upon agglutination tests always state the time at which the 
test was made and the dilution used. 
A 1 : 20 and a 1 : 40 dilution may be prepared and examined at the end 
of half an hour. Prompt agglutination is found practically only in typhoid 
fever. 
Dilutions may be conveniently prepared by drawing the serum up to 
the mark 0.5 in the white corpuscle pipet, and the distilled water up to the 
mark 11. Mix well. This gives a dilution of 1 : 20. One loopful of this 
diluted serum and 1 loopful of bouillon culture of the micro-organism to 
be tested give a dilution of 1 : 40. One loopful of the 1 : 20 diluted serum 
and 3 loopfuls of the culture give a dilution of 1 : 80. Having mixed the 
diluted serum and the bacterial suspension on a cover-glass, prepare the 
cultures on the vaselined concave slides in the usual manner. 
Precautions.—In bacteriologic technic due care should be observed to 
avoid contamination and possible infection when working with living cultures. 
Fig. 100.—A Positive Agglutination (Wi- 
dal) Reaction in Typhoid Fever. X 430. 
Serum from a patient ill about twenty-two days; 
a 1 : 100 dilution at the end of one hour. 276 
BACTERIAL AGGLUTININS 
(a) Agglutinated bacteria are not necessarily dead, and hanging-drop 
preparations, test-tubes, etc., should be immersed in 1 per cent, formalin 
before cleansing. 
(b) The working table or desk and the hands should be washed with 
a solution of lysol or 1 per cent, formalin after the reactions have been made 
and the work completed. 
(e) Early in typhoid fever the bacillus may be present in the blood, 
and consequently due care should be exercised in handling it, in diluting 
the serum, and in the disposal of the clot. 
(d) Great care must also be exercised against the error of falsely positive 
reactions due to the use of spontaneously agglutinating organisms. For this 
reason a control employing normal serum or simple saline solution should 
always be employed. Arkwright1 has recently shown that suspensions of 
these organisms in 0.42 to 0.1 per cent, saline solutions may prevent 
spontaneous agglutination. 
The Microscopic Agglutination Test (Dry Method) 
1. The culture is prepared as described above. 
2. Blood is secured by pricking the finger or lobe of the ear and collect- 
ing a few drops of blood upon aluminum foil, on a clean glass slide, or on 
partially glazed paper. The blood must not be heated to hasten drying, or 
agglutinins may be destroyed. Smears on aluminum foil and on glass 
slides are to be preferred to those on paper, as the blood can be moistened 
and portions removed without the likelihood of transferring extraneous 
material, such as paper fiber. While there are certain objections to this 
method to be pointed out later, yet practical experience has demonstrated 
its Value, as the serum does not readily deteriorate or become contaminated 
with bacteria, and the ease with which blood may be collected and mailed 
recommends the process for board of health laboratories. 
3. Place a loopful of a twenty-four-hour bouillon culture of Bacillus 
typhosus in the center of a clean cover-glass. 
4. Moisten the dried blood which has been collected on aluminum foil, 
glass slide, or paper with a loopful of normal salt solution. (A second and 
smaller loop may be used for this purpose.) Gently rub up the dried blood 
and transfer a sufficient amount to the drop of culture on the cover-glass 
until, when thoroughly mixed, it presents a delicate orange tint (Fig. 101). 
Avoid transferring too much debris with the solution of blood, especially 
if the blood has been collected on paper. It is good practice to mix the 
culture and solution of blood with the cover-glass held over a white surface, 
in order that the color may readily be observed. 
5. Having made the mixture on the cover-glass, invert it over a vaselined 
concave slide, label, and stand aside for an hour. 
6. Prepare the culture control in the usual manner and label. 
7. Examine at the end of an hour with the £ or oil-immersion lens. If 
minute fragments of fiber, etc., have been transferred, due allowance for 
false agglutination for these should be made. Otherwise the readings are 
made in exactly the same manner as in the “wet” method. 
8. Accurate dilutions are not possible with this technic. Satisfactory 
results are dependent largely upon the color; a faint orange tint of the 
suspension is desirable, and probably represents a dilution of about 1 : 40. 
This method, however, is very simple, and when carefully performed 
yields results in the practical serum diagnosis of typhoid fever almost as 
satisfactory as the serum-dilution method. 
1 Jour. Path, and Bacteriol., 1921, 24, 36. Fig. 101.—Microscopic Agglutination Test with Dried Blood. 
Shows the proper color of the suspended drop of typhoid culture when the solution of dried blood 
has been added. The tinge should be light orange or yellow, and a shade lighter than ordinary vaselin 
used in sealing the preparation. MACROSCOPIC AGGLUTINATION TEST 
277 
It is possible, however, to work with known approximate dilutions by 
the dried blood method if a good chemical balance is available. Blood 
must be collected on aluminum foil or glass, and is then scraped off and 
weighed. To each 5 mg. of dried blood 0.5 c.c. of salt solution is added 
which equals a dilution of 1 : 25 of whole blood or 1 : 100 of dried blood 
(Wesbrook). After permitting the mixture to stand for half an hour it is 
centrifuged for a short time. To 1 drop of the dilution thus obtained 
1 drop of culture is added, which gives a final dilution of about 1 : 50. 
At the end of an hour it is examined. Higher dilutions can be prepared 
from this stock dilution at the will of the operator. 
The Bacterial Suspension.—Living cultures may be employed by culti- 
vating the bacterium in broth or removing and emulsifying cultures from 
solid medium in physiologic saline solution as described under the micro- 
scopic method. 
The suspension should be free of macroscopic clumps and of a density 
equal to tubes 5 or 6 of McFarland’s nephelometer (see page 196). Further- 
more, the suspension must not show spontaneous agglutination. 
To emulsify a culture of the plague bacillus or any other micro-organism 
that displays a strong tendency to undergo “spontaneous” agglutination, 
distilled water or 1 : 1000 salt solution should be used. 
In the case of a culture of tubercle bacillus, the growth can be resolved 
into its elements by prolonged trituration in normal salt solution, and any 
residue or unresolved clumps removed by centrifugalization. A less laborious 
and dangerous method is to use the tubercle powder of Koch, which is 
obtained by reducing dried tubercle cultures to a fine powder by machinery. 
The powder may be made up into a suitable suspension by rubbing it in a 
mortar with normal salt solution. 
When it is necessary to work with highly dangerous micro-organisms, 
or to operate from day to day with the same bacterial suspension, one may 
employ suspensions that have been heated for one hour to 60° C., or sus- 
pensions in salt solution to which 1 per cent, formalin has been added. 
These will keep well in the refrigerator, but the sediment of dead bacteria 
must be well shaken before it is used. 
Nobel1 has described the following method for preparing a suspension 
of anthrax bacilli which are so prone to produce spontaneous agglutination: 
“The cultures are transplanted daily for ten days on plain agar and 
incubated at 42.5° C., until a sporeless and very vigorous growth is obtained. 
Each strain is then planted on plain agar in quart whisky flasks and incu- 
bated for twelve hours at 42.5° C. The growths are washed off in physio- 
logic salt solution containing 0.5 per cent, formalin (about 100 c.c. to a 
flask). The suspensions are shaken in a mechanical shaker for forty-eight 
hours. After standing for several days and being tested for sterility, equal 
parts of each suspension are mixed in a cylinder; shaken for twenty-four 
hours and allowed to stand over night. The larger clumps settle out leaving 
a homogeneous suspension above. This upper portion is poured off and 
filtered several times through four thicknesses of sterile cheese-cloth. The 
suspension is then diluted with physiologic salt solution plus 0.5 per cent, 
formalin to a density corresponding to a suspension of Bacillus typhosus 
containing 2,000,000,000 bacteria per cubic centimeter. A suspension of 
B. anthracis so prepared is perfectly homogeneous, stands up for at least 
forty-eight hours at 37° C., and shows no spontaneous agglutination.” 
Macroscopic Agglutination Test 
1 Jour. Immunology, 1919, 4, 105. 278 
BACTERIAL AGGLUTININS 
Dreyer Method.—As previously stated, Dreyer has found suspensions 
of typhoid, paratyphoid, dysentery, and other intestinal bacilli best pre- 
pared by using broth cultures sterilized and preserved with 0.1 per cent, 
formalin. This method has been warmly endorsed by numerous English 
workers; Sands,1 working in my laboratory, has also found formalized emul- 
sions best for the typhoid agglutinin reaction. 
Dreyer has described the preparation of formalized suspensions as 
follows2: 
(a) The bacillus (B. typhosus, B. paratyphosus, etc.) is grown for 
twenty-four hours at 37° C. in ordinary veal peptone bouillon in large Erlen- 
meyer flasks partly filled (1 liter of bouillon in a If-liter flask). • 
(b) Before use the flasks of bouillon are sterilized in the autoclave at 
115° C. for not more than fifteen minutes, and are then tested for sterility 
by incubation at 37° C. for forty-eight hours. 
(c) They are inoculated with a few drops each from a twenty- to twenty- 
four-hour old bouillon culture of the bacillus (B. typhosus or B. paratyphosus, 
etc.). 
(d) The culture used should be one which has been subcultivated daily 
in bouillon for one or two weeks (or longer). This continued subcultiva- 
tion has the effect of increasing its agglutinability and diminishing any 
tendency to spontaneous agglutination. 
(e) At the end of twenty-two to twenty-four hours’ growth at 37° C. the 
flasks are well shaken, and to each is added 0.1 per cent. (1 c.c. per liter) 
of commercial (40 per cent.) formalin. They are again shaken and placed 
in a cold chamber in the dark at about 2° C. 
(/) At intervals on the same day and on subsequent days for four or 
five days the flasks are again thoroughly shaken and replaced at once in the 
cold chamber. 
(g) After three or four days they will be found to be absolutely sterilized. 
Should it happen that the bacterial suspension is not entirely homogeneous 
it may be shaken for some hours in a mechanical shaker, or may finally be 
filtered through sterile cotton-wool. Cultures so prepared are put into 
sterile bottles with rubber stoppers and kept in a cold and dark place. 
The advantages of this method may be summarized as follows: 
(a) The bacteria are dead and fixed and their use devoid of danger. 
(b) The suspensions contain no excess of antiseptics which may be 
detrimental to agglutination. 
(c) There is either no loss of agglutinability at all or but a transitory 
loss of slight degree (B. dysenterise). 
• (d) The suspensions can be kept for six months or longer without change. 
By means of this method Dreyer has been able to standardize the agglu- 
tinin test for typhoid, paratyphoid, and other intestinal infections; he has 
shown quite conclusively that consistent results are only possible by con- 
ducting tests with suspensions of bacteria of uniform agglutinability and 
density. 
The suspension should always be practically transparent. It is stated 
by the Oxford Standard Laboratory (where Dreyer and his associates pre- 
pared standard emulsions for use in English laboratories and especially for 
army laboratories during the war) that “the growth of a mold in a bottle 
does not affect the agglutinability of the culture. If the mold be fished 
out and a drop or two of chloroform be added to the fluid to prevent further 
growth the culture is as good as ever. Bacterial growths in the cultures 
1 Jour. Immunology, 1920, 5, 97. 
2 Jour. Path, and Bacteriol., 1909, 13, 331. MACROSCOPIC AGGLUTINATION TEST 
279 
also occur, but are rare and almost invariably the result of careless handling.” 
Krumbhaar and Smith1 found that even with careful handling contamina- 
tion was apt to occur and they have advised removing from the stock bottle 
an amount sufficient for the work at hand, discarding any that is left over. 
I believe that more formalin may be added to lessen the risks of contami- 
nation without injury to the suspension. In my experiments 1 per cent. 
neutral formalin proved very satisfactory (10 c.c. neutral formalin per liter 
of culture). 
Serum.—Sufficient blood for the macroscopic test may be obtained by 
pricking the finger and filling a Wright capsule or collecting about 1 c.c. 
in a small test-tube. The serum may be allowed to separate or may be 
obtained at once by breaking up the clot and centrifuging. It should be 
free of corpuscles. 
The Test.—All dilutions and measurements are to be made with accu- 
rately graduated 1 c.c. pipets (dry and preferably sterile). 
Water instead of physiologic saline solution is employed for making the 
dilutions. Dreyer has recommended the use of distilled water and Krum- 
bhaar and Smith have found the tests superior to those conducted with saline 
solution. 
1. Place a row of seven small test-tubes (10 x 1 cm.) in a test-tube rack 
and add 1 c.c. of sterile distilled water to each. 
2. Dilute the serum 1 : 5 in the first tube as follows: 0.2 c.c. serum plus 
0.8 c.c. water. This now gives in this tube 2 c.c. of a dilution of 1 : 10. 
Mix well with the pipet. 
3. Place 1 c.c. of the serum from tube 1 into tube 2. Mix well, and 
place 1 c.c. of the mixture from tube 2 into tube 3, and so on. When the 
sixth tube has been reached discard 1 c.c., as no serum is to be added to the 
seventh tube which is the culture control; i. e., it will contain water plus 
bacterial emulsion. 
4. Add 1 c.c. of bacterial emulsion to each tube which doubles the 
serum dilution in each. Tube 1 now contains a serum in a dilution of 1 : 20, 
acting on the bacteria; tube 2, one of 1 : 40; tube 3, one of 1 : 80; tube 4, 
one of 1 : 160: tube 5, one of 1 : 320; tube 6, one of 1 : 640. Tube 7, as 
just stated, contains the bacterial emulsion in water and is the culture 
control. 
In determining the agglutination titer of a highly immune serum these 
dilutions may be continued to any degree. 
5. On each tube the final dilution is marked with a wax pencil. The 
tubes are then shaken gently, stoppered with cotton plugs, and placed in 
the incubator at 37° C. or in a water-bath at 55° C. for two hours. The 
tubes are then allowed to remain at room temperature for a few hours, 
or in the refrigerator for twenty-four hours, after which readings are made. 
When living cultures are employed the method of Kolle and Pfeiffer is 
very convenient, and may be safer than that of adding live cultures with a 
pipet. It is conducted as follows: 
1. Make dilutions of serum as described. 
2. Emulsify thoroughly a loopful (2 mg.) of culture from an eighteen- 
to twenty-four-hour-old agar culture in the first test-tube, repeating the 
process in the second tube, and so on through the series. In this method 
the serum dilutions are not doubled; thus in the foregoing series the dilu- 
tions would be 1 : 10, 1 : 20, 1 : 40, 1 : 80, 1 : 160, 1 : 320. 
3. The tubes are gently shaken, labeled, plugged, and incubated as 
directed in the preceding method. 
1 Jour. Infect. Dis., 1918, 23, 126. 280 
BACTERIAL AGGLUTININS 
The Readings.—The culture control should show a uniform cloudiness, 
with no sediment or flakes, or at most a very slight precipitate that is readily 
Fig. 102.—Macroscopic Agglutination Reaction. 
Serum of an individual who had received three injections of typhoid vaccine. This drawing was 
made twenty-four hours after the test was setup. The dilutions were: 1:20, 1:40,1:80, 1:160 and 
1:320; the sixth tube is the control. 
broken up by gentle agitation. A positive reaction shows masses and 
clumps of bacteria adhering to the sides and bottom of the tube, which 
Fig. 103.—Macroscopic Agglutination Reaction. Shows Action of Agglutinoids (Pro- 
agglutination). 
Note absence of agglutination in dilution 1 : 50; agglutination beginning in 1 : 100, and fairly 
well marked to 1 : 4000 inclusive. Note uniform cloudiness of control. This reaction was set up with 
a typhoid immune serum over six months of age; the drawing was made after the tubes had been in- 
cubated two hours and placed in a refrigerator overnight. 
are broken up with some difficulty (Fig. 102). The supernatant fluid should 
be clear. As dilutions become higher and the amount of contained agglu- MACROSCOPIC AGGLUTINATION TEST 
281 
tinin correspondingly less, agglutination becomes less and less complete. 
There is less sediment, and the turbidity of the supernatant fluid is greater, 
until the negative tube closely resembles the culture control. A micro- 
scopic examination of a deposit will show that the bacilli point in all direc- 
tions, whereas in a deposit of unagglutinated bacilli they lie horizontally 
side by side. 
When agglutinoids are present, agglutination is absent or incomplete 
in the lower dilutions of serum, and complete in the tubes containing the 
higher dilutions. This is called proagglutination (Fig. 103). 
The test-tubes are arranged in the rack and viewed from below in the mirror. In this manner the 
smallest deposits are easily seen and compared with the control. 
Fig. 104.—Agglutinoscope. (Altman.) 
Readings are facilitated by the use of a special instrument known as 
the agglutinoscope (Fig. 104). The tubes are placed in a rack having num- 
bered holes, and are viewed from beneath with the aid of a mirror. In this 
way one looks upward through the column of fluid, and secures a combined 
view of sediment and turbidity, and when examined with the culture control, 
fine and accurate readings may be made. 
If the readings are made within a few hours after incubation it is well 
to use a hand glass of about 2 to 4 diameters of magnification. Krumbhaar 
and Smith found that this facilitated the readings of reactions conducted 282 
BACTERIAL AGGLUTININS 
by the Dreyer method. If the tubes are allowed to stand over night before 
the readings are made the naked eye is sufficiently accurate. 
Hadley1 has recently advocated the adoption of a standard method of 
reading and recording reactions; the general adoption of his method would 
render results reported from different laboratories more uniform. 
Dreyer Standardized Agglutination Test.—1. The bacterial suspension 
is prepared as previously described consisting of broth cultures of strains 
selected for their high specificity, killed and preserved with 0.1 per cent, 
formalin. In England successive batches of standard agglutinable culture 
the relative sensitiveness to agglutination of the bacilli is indicated by a 
figure—the so-called Reduction Factor. 
2. The test is conducted by the drop method, the use of Dreyer’s pipets 
being preferable. These are made on the same lines as an ordinary dropping 
pipet (18.8 cm. in length; the end is drawn out to a length of 2.8 cm., with 
an external diameter of 0.235 cm. and a bore of 0.05 cm.). 
3. The test-tubes are conically pointed at the lower end: They are 
6 cm. long, 0.66 cm. in diameter; the lip is widened out to 1 cm. 
4. In conducting a test for typhoid and paratyphoid agglutinins proceed 
as follows: If a test for typhoid agglutinins only is to be done but one row 
of tubes is required: 
(a) Take a stand containing 15 agglutination tubes in 3 rows of 5 each, 
and a dilution tube. 
(b) With the proper dropping pipet measure out into the dilution tube 
54 drops of distilled water or normal saline solution (0.85 per cent, sodium 
chlorid in distilled water). 
(c) Wash the pipet with distilled water. 
(.d) Dry out the pipet with successive quantities of absolute alcohol, 
followed by successive quantities of ether, and get rid of the ether. 
(e) Take up the serum to be tested into the dried pipet. Measure out 6 
drops of the serum into the dilution tube already containing the 54 drops 
of saline solution, thus obtaining a dilution of 1 in 10. Mix thoroughly. 
Carefully wash out the pipet. 
With the pipet measure out into each row of tubes as follows: 
Number of 
Tube. 
Drops of Normal 
Saline8 Solution. 
Drops of Serum 
Dilution 1 'n 10 
1 
0 
10 
2 
5 
5 
3 
8 
2 
4 
9 
1 
5 
10 
0 
To each tube in row 1 add 15 drops of 
Bacillus typhosus culture. 
To each tube in row 2 add 15 drops of 
B. paratyphosus A culture. 
To each tube in row 3 add 15 drops of 
B. paratyphosus B culture. 
At each stage of the procedure the pipet is carefully washed and dried 
out with successive quantities of absolute alcohol followed by successive 
quantities of ether. 
(/) Shake each tube thoroughly in order from right to left, i. e., beginning 
each row with the highest dilution. 
(g) Place the stand for two hours in a water-bath at 50° to 55° C. (not 
in dry air). 
1 Jour. Immunology, 1917, 2, 463. 
2 Or, preferably, distilled water. MACROSCOPIC AGGLUTINATION TEST 
283 
In Tube 1 of each row the serum acts in a dilution of 1 in 25. 
In Tube 2 of each row the serum acts in a dilution of 1 in 50. 
In Tube 3 of each row the serum acts in a dilution of 1 in 125. 
In Tube 4 of each row the serum acts in a dilution of 1 in 250. 
Tube 5, containing no serum, is control against spontaneous agglutination. 
If the limit of agglutination is not reached within this series, higher 
dilutions are followed out in a similar manner. 
Thus, for example, 57 drops of normal saline solution plus 3 drops of 
a 1 in 10 serum dilution will give a serum dilution of 1 in 200, and, using 
the same quantities as before, one has the serum acting in dilutions of 1 in 
500, 1 in 1000, 1 in 2500, and 1 in 5000. And similarly for higher dilution. 
{h) The tubes are examined after two hours at 50° to 55° C., followed 
by fifteen minutes’ standing at room temperature. The reading is taken 
by comparing each tube in succession with the control tube, and is pref- 
erably made by means of artificial light against a black background. If 
daylight is used, the tubes inspected should be partly shadowed by passing 
a finger up and down behind them. 
A reading-glass aids in making finer and more delicate readings. 
If time permits it is preferable to place the tubes in a refrigerator over 
night making the readings next morning. 
Dreyer and Inman1 have described standard agglutination, standard 
agglutinin unit (the unit of agglutinating power), the reduction factor and 
method of readings as follows: 
Standard agglutination is the degree of agglutination present in the 
highest serum dilution in which marked agglutination without sedimenta- 
tion can be seen by the naked eye. 
The standard agglutinin unit is that amount of agglutinating serum 
which when made up to 1 c.c. volume with normal saline solution causes 
standard agglutination on being mixed with 1.5 c.c. of the original standard 
agglutinable culture and maintained at 55° C. for two hours (in the case of 
dysentery agglutination four and a half hours) in a water-bath, followed by 
fifteen to twenty minutes at the room temperature. 
The Reduction Factor.—The total volume in which the reaction occurs 
being 2.5 c.c. (1 c.c. of serum added to 1.5 c.c. of standard culture) the 
original standard agglutinable culture was given the reduction factor of 
2.5 to express the sensitiveness to agglutination of that particular culture. 
All subsequent batches of culture have been given reduction factors cal- 
culated on this basis, thus securing constancy in the agglutinin unit. For 
example, if a batch of standard culture proves to be twice as sensitive to 
agglutination as the original standard, so that half the amount of serum 
produces standard agglutination under test conditions, the new standard 
culture is given a reduction factor of double the size of the original factor, 
i. e., 5. Thus, whatever be the particular standard culture used to test 
any given serum the number of agglutinin units found per cubic centimeter 
of the serum remains always the same, although the dilutions in which 
standard agglutination occurs will be different. Since when standard agglu- 
tination occurs in a serum dilution of 1 in x, then x divided by the reduc- 
tion factor for the particular standard agglutinable culture used gives the 
number of standard agglutinin units contained in 1 c.c. of the serum con- 
cerned. 
Readings.—Owing to the rate at which the dilution increases in the 
series of tubes employed it will commonly happen that no tube in the series 
exhibits standard agglutination. If this be so, it will be found in looking 
1 The Lancet, London, March 10, 1917. 284 
BACTERIAL AGGLUTININS 
along the series that while one tube shows strong agglutination with sedi- 
mentation the next succeeding tube shows no agglutination or only a trace. 
In such cases standard agglutination lies approximately midway between 
the two dilutions. Though this method of making readings is amply ade- 
quate for diagnostic purposes it will be found that should a more precise 
determination of the limits of agglutination be required it can be obtained 
by using a stand of 12 tubes with the series of quantities given in the table 
contained in the directions for preparation and standardization of agglu- 
tinable cultures, where the successive dilutions of the serum only differ 
by about 20 per cent. Almost as accurate a reading can, however, be ob- 
tained with experience from the short series by taking note of the degree of 
agglutination present in each tube and by using a suitable interpolation 
table. 
The principal terms employed in describing the different degrees of 
agglutination met with are “total” (t), “standard” (S), “trace” (tr), and 
“mil.” (0). Total agglutination indicates the condition in which the whole, 
or practically the whole, of the agglutinated bacteria have settled down 
at the foot of the tube. Standard agglutination has already been described; 
the term Trace is applied to a very fine granulation recognizable by the 
naked eye. Around these main terms subsidiary differences congregate 
themselves as follows: Total minus (t —) marked deposit, but a number 
of floating flocculi remaining in the fluid. On each side of “standard” we 
find Standard plus (S +), and Standard minus (S —) respectively. In the 
former no deposit, but much larger flocculi than are seen in standard agglu- 
tination. In the latter finer agglutination than standard, with more the 
appearance of granulation in the fluid. Similarly, we recognize a Trace 
plus (tr +), and a Trace minus (tr —), the former representing something 
more than trace, but less than standard minus, the latter being on the limit 
of naked-eye visibility. Finally, on occasion it can be difficult to decide 
with certainty whether a given tube is absolutely nil or not, the term query 
trace (tr ?) is then applied. 
Modified Dreyer Test.—A drawback to the accuracy of Dreyer’s technic 
consists in the measurement of fluids by drops; Donald1 has emphasized 
the importance of this source of error, although Walker2 has replied to the 
criticisms and maintains that the drop method is accurate. 
I have found the use of accurately graduated 1 c.c. pipets preferable 
for measuring all fluids; likewise the use of larger amounts of serum dilu- 
tions and bacterial emulsion, the tests being set up as follows: 
Serum is diluted 1 in 10 by mixing 0.2 c.c. in a test-tube with 1.8 c.c. 
distilled water or physiologic saline solution. 
Tube 1 — 1.0 c.c. of serum 1 : 10 + 1.5 c.c. culture = 1 : 25. 
Tube 2 — 0.5 c.c. of serum 1 : 10 + 1.5 c.c. culture = 1 : 50. 
Tube 3 — 0.2 c.c. of serum 1 : 10 + 1.5 c.c. culture = 1 : 125. 
Tube 4 — 0.1 c.c. of serum 1 : 10 + 1.5 c.c. culture = 1 : 250. 
Tube 5 — 1.0 c.c. of water +1-5 c.c. culture = control. 
These give the same final dilutions as employed by Dreyer. The balance 
of the test is exactly as described above. 
Macroscopic Slide Methods.—For the rapid identification of bacteria 
in bacteriologic studies Coca3 has described an efficient method for con- 
ducting agglutination tests on slides; Krumwiede4 has found the method 
1 The Lancet, London, September 2, 1916, 423. 
2 The Lancet, London, September 23, 1916. 
3 Bull. Manila Med. Soc., 1910, 2, No. 1. 
4 Jour. Infect. Dis., 1918, 23, 275. MACROSCOPIC AGGLUTINATION TEST 
285 
satisfactory for the identification of typhoid, paratyphoid, and dysentery 
bacilli isolated from feces and meningococci from the nasopharynx. He 
has also described a dropping bottle for use in these tests.1 The technic 
is as follows: 
1. The tests are conducted with highly potent immune serum. The 
dilutions may be as low as 1 : 50 or 1 : 100. With a known and tested 
serum used in proper dilution the results are very reliable; with an unknown 
serum serious error may result because of group reaction (Krumwiede). 
Occasionally a slight delay in clumping due to the. proagglutination phe- 
nomenon will be noticed, but there is usually some evidences of agglutination. 
2. A drop of diluted immune serum is placed on a slide; a drop of saline 
solution is placed on the same or a second slide, due care being taken to 
avoid admixture with serum. 
3. A portion of the suspicious colony is picked off with a small loop of 
fine wire, and a sufficient amount rubbed off in the salt solution to give a 
slight clouding (this is the control). Some of the growth remaining on the 
loop is then rubbed off similarly in the drop of diluted serum. 
4. In a positive reaction clumping occurs almost immediately; spon- 
taneous agglutination is detected in the control. 
Bass and Watkins2 have described a slide method for conducting the 
Widal test for typhoid fever -which is very simple and may be conducted 
by the physician at the bedside or in his office. The technic is as follows: 
1. The culture is prepared by suspending twenty-four-hour growths of 
typhoid bacilli in distilled water, 10,000,000,000 per cubic centimeter, killed 
and preserved with 1 per cent, commercial formalin. This suspension is 
said to keep six to twelve months or longer. It must be well shaken before 
using. 
2. A blood film is made on a slide in the usual manner, using approxi- 
mately \ drop of blood. 
3. Place on the blood- 1 drop of water and dissolve the blood with 
the aid of a tooth-pick or other suitable instrument. 
4. Add 1 drop of the suspension of typhoid bacilli, and mix by tilting 
the slide from side to side and from end to end, causing the mixture to flow 
back and forth. 
5. A positive reaction occurs within two minutes, with the formation 
of small grayish clumps and fine granular sediment. When the test is nega- 
tive no such granular sediment forms. Dust particles must not be mistaken 
for agglutinated bacilli. 
Technic of Conglutination Test with Bacteria.—Fresh bovine serum is 
heated to 56° C. for one-half hour and tested for agglutinin for the bacteria 
under study. If agglutinin is present it is removed by adding to each 5 c.c. 
of serum about 10 loopfuls of the corresponding bacteria, mixing well, and 
removing the clumps by thorough centrifuging and filtration through paper 
after standing at room temperature for several hours. It may be neces- 
sary to repeat this step once or twice more. 
Dilutions of the patient’s serum in amounts of 1 c.c. are made in small 
clean test-tubes, as in the macroscopic agglutination test. To each tube 
add 0.1 c.c. of bovine serum (“conglutinin”); 0.1 c.c. of fresh guinea-pig 
serum (complement), and 0.1 c.c. of the emulsion of the bacteria of sufficient 
density to give well-marked emulsions in the test-tubes. Controls of bovine 
serum, complement serum, and patient’s serum should be included; also a 
culture control prepared with normal salt solution. After gentle mixing 
and standing at room temperature for twenty-four hours the results are 
1 Jour. Immunology, 1920, 5, 155. 
2 Archiv. Int. Med., 1911, 8. 286 
BACTERIAL AGGLUTININS 
read; positive results are indicated by complete clearing of the tube with 
the micro-organisms in flakes or granules, either clinging to the sides of the 
tube as granular material or at the bottom as flaky granular sediment. 
The Agglutination Test in the Differentiation of Pneumococci.—The 
investigations of Neufeld and Handel, and particularly of Cole, Dochez, 
and their associates in the Rockefeller Institute, have resulted in the group- 
ing of pneumococci into four groups on the basis of agglutination and pro- 
tective tests. Group III is composed of pneumococci belonging to the type 
of Pneumococcus mucosus, and an efficient antiserum has not yet been 
produced; Group IV is composed of all pneumococci not falling into the 
first three groups. 
The agglutination test is conducted for the purpose of differentiating 
these types; if it is decided to administer an immune serum, the serum corre- 
sponding to the type of infection must be given. The following method 
has been described by Avery, Chickering, Cole, and Dochez1: 
Collection of Sputum.—Care should be exercised in the collection of 
sputum to obtain a specimen from the deeper air-passages as free as possible 
from saliva. This can be done in practically all cases, even the most difficult, 
with a little persistence. The sputum is collected in a sterile Esmarch dish 
or other suitable container, and should be sent at once to the laboratory 
for examination. When delay is unavoidable the specimen should be kept 
on ice during the interval. 
Microscopic Examination of Sputum.—Direct films of sputum are stained 
by Gram, Ziehl-Neelsen, and by Hiss capsule stains. This serves to give 
an idea of the nature of the organisms present and an indication of the 
source of the sputum. Suitable lung specimens of sputum are relatively 
free in most instances from contaminating mouth organisms. It is fre- 
quently possible to identify Type III (Pneumococcus mucosus) organisms 
when they are present-, as they possess very large distinct capsules staining 
by both Gram’s and Hiss’ methods. 
Mouse Inoculation.—A small portion of the sputum about the size of a 
bean is selected and washed through three or four changes of sterile salt 
solution in sterile Esmarch or Petri dishes to remove surface contamina- 
tions. When the sputum is too friable or -when the specimen is relatively 
free from secondary organisms, this washing process may be omitted. In 
either event the kernel of sputum selected is transferred to a sterile mortar, 
ground up, and emulsified with about 1 c.c. of sterile bouillon or salt solu- 
tion, added drop by drop, until a homogeneous emulsion is obtained that 
will readily pass through the needle of a small syringe. With a sterile syringe 
0.5 to 1 c.c. of this emulsion is inoculated intraperitoneally into a wThite 
mouse. The pneumococcus grows rapidly in the mouse peritoneum, while 
the majority of other organisms rapidly die off with the exception of Fried- 
lander’s bacillus, Bacillus influenzse, and occasionally Micrococcus catar- 
rhalis, staphylococcus, and streptococcus. Pneumococcal invasion of the 
blood-stream also occurs early. Bacillus influenzae, if present, likewise 
invades the blood-stream; other organisms, as a rule, do not. The time 
elapsing before there is sufficient growth of pneumococcus in the mouse 
peritoneum for the satisfactory determination of type varies with the in- 
dividual case, depending upon the abundance of pneumococci in the speci- 
men of sputum and the virulence and invasiveness of the strain present. 
It may be from five to twenty-four hours, averaging six to eight hours with 
the parasitic fixed Types Nos. I, II, and III. As soon as the injected mouse 
appears sick, a drop of peritoneal exudate is removed by means of peritoneal 
1 Monograph No. 7, Rockefeller Institute, 1917, 23. MACROSCOPIC AGGLUTINATION TEST 
287 
puncture with a sterile capillary pipet, spread on a slide, stained by Gram’s 
method, and examined microscopically to determine whether there is an 
abundant growth of pneumococcus present. If there is an abundant growth 
of pneumococcus alone the mouse is killed and the determination of type 
proceeded with. If the growth is only moderate, or if other organisms are 
present in any quantity, further time must be allowed until subsequent 
examination of the peritoneal exudate shows an abundant growth of pneu- 
mococcus. It should be emphasized that undue haste in killing the mouse 
is time lost in the end. 
Mouse Autopsy.—As soon as the mouse is killed or dies, the peritoneal 
cavity is opened with sterile precautions and cultures are made from the 
exudate in plain broth and on one-half of a blood-agar plate. Films are 
made and stained for microscopic examination by Gram’s stain and Hiss’ 
capsule stain. The peritoneal exudate is then washed out by means of a 
sterile glass pipet with 4 to 5 c.c. of sterile salt solution, the washings 
being placed in a centrifuge tube. Cultures are then made from the heart’s 
blood in plain broth and on the other half of the blood-agar plate. 
Agglutination Test.—When the pneumococcus is present in pure culture 
in the peritoneal exudate, the determination of type may satisfactorily be 
made by macroscopic agglutination tests as follows: (1) The peritoneal 
washings are centrifuged at low speed for a few minutes until the cells and 
fibrin contained in the exudate are thrown down. The supernatant bacterial 
suspension is transferred into a second centrifuge tube and centrifuged at 
high speed until the organisms are thrown out. The supernatant fluid is 
saved for precipitin tests (see chapter on Precipitins), and the bacterial sedi- 
ment taken up in sufficient salt solution to make a moderately heavy suspen- 
sion. The concentration of bacteria should be similar to that of a good 
eighteen-hour broth culture of pneumococcus. 
(2) Highly potent antipneumococcus sera for Types I, II, and III must 
be available, at least for Types I and II; if these sera are cloudy they should 
be centrifuged before use. 
(3) The test is set up as follows in 6 small test-tubes: 
Tube 1: 0.5 c.c. Serum I (1 : 10) + 0.5 c.c. bacterial suspension = 1 : 20. 
Tube 2: 0.5 c.c. Serum II (undiluted) + 0.5 c.c. bacterial suspension = 
1:2. 
Tube 3: 0.5 c.c. Serum II (1 : 10) + 0.5 c.c. bacterial suspension = 1 : 20. 
Tube 4: 0.5 c.c. Serum III (1:5) + 0.5 c.c. bacterial suspension = 1 : 10. 
Tube 5: 0.1 c.c. Sterile ox bile + 0.5 c.c. bacterial suspension. 
Tube 6: 0.5 c.c. saline solution + 0.5 c.c. bacterial solution (control) 
suspension. 
Tube 5, containing bile plus bacterial suspension, is to determine the 
bile solubility of the strain and for differentiation of pneumococcus from 
streptococcus; pneumococcus is bile soluble and the contents become clear. 
Tube 6 is the control for spontaneous agglutination. 
(4) The tubes are gently shaken and incubated in a water-bath for one 
hour at 37° C., when the readings are made. 
(5) Positive reactions are usually readily detected and may be brought 
out clearly by gentle agitation of the tubes. If no agglutination has occurred 
and if the bacteria are bile soluble and otherwise resemble the pneumococcus, 
the type is IV. If agglutination occurs in tubes 2 and 3 there is present 
typical Type II pneumococci; if agglutination is partial in tube 2 and 
absent in tube 3 atypical Type II pneumococci are present. If agglutina- 
tion occurs only in tube 1 the pneumococcus is Type I; if only in tube 4 it 
belongs to Type III. Sometimes two types occur together and Type IV 288 
BACTERIAL AGGLUTININS 
may occur with any of the other types without being detected by this test. 
Sometimes the readings are facilitated by standing the tubes aside for sev- 
eral hours before making the readings. 
The test may be conducted with young white rats inoculated with two 
to four times the amount of sputum recommended for mice; the latter, 
however, are more satisfactory. 
Avery1 has recommended a method for conducting the test when animals 
are not available, consisting of inoculating the prepared sputum in special 
dextrose broth medium for securing the initial growth of pneumococcus for 
the tests; it is not as efficient as the mouse test described above. 
Microscopic tests are conducted by mixing on cover-slides platinum 
loopfuls of bacterial emulsion and undiluted and diluted immune sera. 
These slides are then suspended as hanging-drop preparations and the 
results read after one-half to an hour. A control should always be included. 
Results are not quite as accurate as those secured by the macroscopic test. 
The Agglutination Test in the Differentiation of Meningococci.—The 
specific agglutinins of the meningococcus have been the subject of much 
investigation. As a result it has been shown that differences exist, and 
from a study of these variations and the application of “absorption” methods, 
it has been possible to differentiate the majority of strains into groups or 
types. Gordon recognizes four types; according to Flexner, Gordon’s Type 
I appears to correspond with the “parameningococcus” of Dopter, and 
Type II with the normal or regular meningococcus. Types III and IV 
appear to conform to the more common intermediates. 
Type Sera.—These may be prepared by the immunization of horses, 
but for small amounts young rabbits are generally employed. Gordon 
and Murray inject 0.5 c.c. of a suspension (1,000,000,000 cocci) intravenously, 
and forty-eight hours later give three additional doses of 0.5 c.c. at hourly 
intervals. By this method within ten days a serum may be obtained with 
a titer as high as 1 : 800. Hine secured better results by injecting 1,000,- 
000,000 cocci as an initial dose followed by 500,000,000 one hour later, and 
six days later by 3,000,000,000. The serum is tested on the eighth day. 
Suspension.—A suspension of the coccus is made in sterile saline and 
heated for half an hour at 65° C. to kill the cocci and inactivate the auto- 
lysins. The suspension is standardized so that each cubic centimeter con- 
tains approximately 2,000,000,000 cocci, and 0.5 per cent, phenol added. 
This phenolated suspension may be used for agglutination tests and for the 
immunization of rabbits. Agglutination tests may be conducted with the 
cloudy spinal fluids of persons ill with meningococcus meningitis in which 
smears show large numbers of extracellular cocci after the fluid has been 
briefly centrifuged to remove pus-cells. 
Test.—Four dilutions of each serum are prepared with saline solution 
in small tubes in amounts of 1 c.c., 1 : 25, 1 : 50, 1 : 100, 1 : 200. To each 
tube and a control carrying 1 c.c. of saline solution is added 1 c.c. of bac- 
terial suspension. Results are read after sixteen hours’ incubation at 55° C. 
For the “absorption” of group agglutinins the sera are saturated with a 
suspension of the coccus to be tested for twenty-four hours, after which 
they are centrifugalized. The supernatant clear sera are then put up against 
their homologous cocci. 
Gates2 has recently described a centrifuge method for the serologic 
grouping of meningococci. Ellis3 and Gates4 have also described simple 
macroscopic methods in capillary tubes. 
1 Jour. Amer. Med. Assoc., 1918, 70, 17. 
3 Jour. Exper. Med., 1922, 35, 63. 
3 Brit. Med. Jour., 1915, 2, 881. 
4 Jour. Amer. Med. Assoc., 1921, 77, 2054. ABSORPTION AGGLUTINATION TEST IN MIXED INFECTION 
289 
Technic of the Absorption Agglutination Test in Mixed Infection (the 
Saturation Test of Castellani) 
The practical importance of partial agglutinins is recognized in the 
diagnosis of mixed infections. Thus the serum of a patient may agglutinate 
typhoid as well as paratyphoid bacilli in dilutions up to 1 : 100. This may 
indicate one of three possibilities: 
1. The patient may be infected with typhoid, but has formed an ex- 
ceptionally large quantity of group agglutinins for paratyphoid bacilli. 
Saturation of this serum with typhoid bacilli will remove all the typhoid 
and a portion, if not all, of the group agglutinins. Saturation with para- 
typhoid bacilli will remove the group agglutinins, but not the main or 
typhoid agglutinin. 
2. The patient may be infected with paratyphoid bacilli, but has formed, 
at the same time, many partial agglutinins for typhoid bacilli. Saturation 
of the serum with paratyphoid bacilli will remove all the paratyphoid and 
a large portion of the typhoid agglutinin. 
3. The patient may have a mixed infection of typhoid and paratyphoid, 
and therefore agglutinin for both may be present. Saturation of the serum 
with typhoid bacilli will remove the typhoid and probably a small portion 
of the paratyphoid agglutinin. After this reaction the serum will still show 
the presence of a decided quantity of paratyphoid agglutinin.- 
In selecting the most likely one of these hypotheses a decision may be 
reached by adopting the method of Castellani (Citron), which is as follows: 
1. Four rows of test-tubes are arranged, each row being made up of 
four small tubes each containing 1 c.c. of serum dilutions 1 : 20, 1 : 40, 
1 : 80, and 1 : 160, respectively. 
2. In each of the tubes of the first and second rows five loopfuls of 
typhoid bacilli are emulsified. An extra tube containing 1 c.c. of normal 
salt solution receives a similar amount of bacteria, and serves as the typhoid 
control. 
3. In each tube of the third and fourth rows five loopfuls of paratyphoid 
bacilli are emulsified. Arrange the paratyphoid culture control. 
4. Mix gently and incubate for four hours. Carefully record the presence 
or absence of agglutination in each test-tube. Centrifuge all the tubes 
excepting the two controls, and transfer the supernatant fluid of each to 
other test-tubes arranged in the same order. 
5. To each tube of the first and third rows add five loopfuls of typhoid 
bacilli; to each of the second and fourth rows, five loopfuls of paratyphoid 
bacilli. Mix well and incubate for four hours. 
(а) If typhoid is present, the agglutination titer in the first part of the 
test will be strong in the tubes of the first and second rows, and weak in 
those of the third and fourth rows. In the second part of the test the titer 
for typhoid will be weak or nil in the first, second, and fourth rows, whereas 
in the third row it will remain practically the same. 
(б) If paratyphoid exists, the agglutination titer in the first part of the 
test will be strong in the tubes of the third and fourth rows, and weak in 
those of the first and second rows. In the second part of the test the titer 
for paratyphoid will be less or negative in the fourth row, and strong or 
unchanged in the second row. 
(c) If a mixed infection exists, the agglutination titer in the first part 
of the test will be strong in the tubes of all four rows. In the second part 
of the test the titer in the first and fourth rows is much weaker or negative, 
and in the second and third rows it will remain the same. 
Technic of Acid Agglutination.—Mixtures of lactic acid and sodium 290 
BACTERIAL AGGLUTININS 
lactate have been generally employed. Gillespie has prepared these from 
stock solutions of one-third normal sodium lactate, with a small crystal 
of thymol, and normal lactic acid, according to the following scheme: 
y- sodium lactate in c.c. 
1.5 
1.5 
1.5 
1.5 
1.5 
1.5 
1.5 
1.5 
1.5 
1.5 
1.5 
0.12 
0.25 
0.5 
1.0 
2.0 
Normal lactic acid in c.c. 
0.5 
1.0 
2.0 
4.0 
8.0 
16.0 
8 
8 
8 
8 
8 
Distilled water in c.c.. . 
18.4 
18.2 
18.0 
17.5 
16.5 
18.0 
17.5 
16.5 
14.5 
10.5 
2.5 
Ratio of acid to salt 
1 
1 
1 
i 
4 
1 
2 
1 
2 
. 4 
8 
16 
32 
H-ion concentration in 
grams per multiplied 
by 104 
0.04 
0.1 
0.2 
0.4 
0.7 
1.4 
2.8 
5.5 
11 
22 
44 
A series of greater dilutions of lactates may be made by diluting each 
of the above with one or more volumes of distilled water. 
Young cultures of bacteria should be employed; if grown in broth, they 
may be prepared by rapid centrifuging and resuspension in sufficient distilled 
water to secure an even emulsion of such density as suitable for the macro- 
scopic agglutination test—0.3 c.c. of the bacterial emulsion is placed in 
each of a series of small and perfectly clean tubes in a rack, and 1 c.c. of 
the proper reaction mixture added to each tube. The tubes are then gently 
and briefly shaken, and placed in a water-bath at about 37.5° C. for an 
hour or two. Michalis2 uses acetic acid and reports sharp differentiation 
of Bacillus typhosus by the method of acid agglutination. 
1 Read 0.12 c.c. of normal acid freshly diluted 1 : 8. 
2 Deutsch. med. Wchn., 1915, xli, 241. CHAPTER XVI 
HEMAGGLUTININS 
As previously stated, agglutination like other immunity reactions is a 
manifestation of broad biologic laws and is not limited to bacteria. Agglu- 
tinins for other cells including erythrocytes, leukocytes, epithelium, and 
spermatozoa may be found in the sera of some animals either normally or 
naturally or as a result of immunization. From the standpoint of practical 
importance the hemagglutinins in human serum for human corpuscles 
(iso-agglutinins) are of most interest and especially in relation to the trans- 
fusion of blood. 
Substances Causing Hemagglutination.—The agglutination of erythro- 
cytes may be caused by many different non-specific and specific agents 
which may be classified as follows: 
(а) Various inorganic colloids may cause the agglutination of thin sus- 
pensions of blood corpuscles as shown by Landsteiner and Jagic,1 Hirsch- 
feld,2 and others with solutions of silicic acid and other substances. 
(б) Various plant substances or phyto-agglutinins, as abrin, ricin, and crotin. 
Abrin and ricin agglutinate the corpuscles of practically all warm- and cold- 
blooded animals; crotin agglutinates the corpuscles of the sheep, swine, 
horse, and to some extent the rabbit, but not of the dog. The seeds of 
many non-poisonous leguminous plants, and also of Solanacece, yield ex- 
tracts that are strongly agglutinative for red corpuscles, the active sub- 
stances being usually found in the proteose fraction. 
(c) Bacterial substances may cause hemagglutination as first shown by 
Kraus and Ludwig3 with bouillon cultures of staphylococci and various 
other bacteria. These have been studied extensively by Pearce and Winnie,4 
and the phenomenon is sometimes seen in tests for bacterial hemolysins. 
(d) Animal secretions, as snake venoms, contain agglutinins for the cor- 
puscles of the rabbit, guinea-pig, dog, sheep, and other animals as discovered 
by Mitchell and Stewart,5 and Flexner and Noguchi.6 These agglutinins 
are usually destroyed by heating to 75° C., and their agglutinating activity 
is usually in inverse ratio to their hemolytic activity. 
Various tissue extracts may contain hemagglutinins as shown by Sick7 
with extracts of organs of normal cats and dogs. Apparently these sub- 
stances are derived from the cells, as the blood-serum is free of agglutinins. 
Romer8 has observed agglutinins for rabbit corpuscles in extracts of the 
lens of the rabbit, and Landsteiner9 observed agglutinins in saline extracts 
of malignant tumors. 
Hemagglutinins have also been found in milk and colostrum by Land- 
steiner,10 Langer,11 Kraus,12 and others; in urine by Pfeiffer,13 Friedberger,14 
1 Wien. klin. Wchn., 1904, No. 3; Munch, med. Wchn., 1904, No. 27. 
2 Arch. f. Hyg., 1907, 63, 237. 
3 Wien. klin. Wchn., 1902, 59. 
4 Amer. Jour. Med. Sci., 1904, 128, 669. 
5 Trans. College of Phys. of Phila., 1897, 19, 105. 
6 Jour. Exper. Med., 1902, 6, 277. 
7 Deutsch. Arch. f. klin. Med., 1904, 138, 389. 
8 Arch, of Ophthal., 1905, 60, 239. 
9 Wien. klin. Wchn., 1908, No. 45. 12 Wien. klin. Wchn., 1901, 737. 
10 Munch, med. Wchn., 1903, 1812. 13 Ztschr. f. Hyg., 1907, 56, 488. 
11 Ztschr. f. Heilk., 1903, 24, 111. 14 Berl. klin. Wchn., 1900, 1236. 
291 292 
HEM A GGLU TIN I NS 
and others, and in cyst fluids by Brinkerhoff and Southard.1 They are only 
occasionally found in spinal fluid. 
Tallqvist2 has found hemagglutinins in extracts of Bothriocephalatus. 
(e) Serum hemagglutinins are of most importance, and have received 
most attention, and especially in relation to the subject of blood transfusion. 
Kinds of Serum Hemagglutinins.—1. Normal or natural hemagglutinins 
or those found in normal sera. These are of three kinds: 
(a) Autohemagglutinins, those that agglutinate the corpuscles of the 
same animal. 
(,b) Isohemagglutinins, those that agglutinate the corpuscles of another 
animal of the same species. 
(c) Heterologous hemagglutinins, those that agglutinate the corpuscles 
of animals of different species. 
2. Immune hemagglutinins, those produced by immunizing an animal 
with injections of blood from an animal of different species. A familiar 
example of these are the agglutinins for human erythrocytes to be found in 
the sera of rabbits immunized with human blood-corpuscles in the produc- 
tion of antihuman hemolysin. 
AUTOHEMAGGLUTININS 
Auto-agglutination, that is, agglutination of red blood-corpuscles by 
the serum of the same individual, is an extremely rare phenomenon. In 
1902 Klein3 observed auto-agglutination in the blood of a horse and pos- 
sibly also in the blood of a human being suffering with cirrhosis of the liver. 
Landsteiner4 noted auto-agglutinins in the sera of horses and other animals. 
Hektoen5 has reported the occurrence of these agglutinins in the blood of 
two individuals, and Ottenberg and Thalhimer6 in the blood of cats. 
Rous and Robertson7 have made the interesting observation that re- 
peated blood transfusions of rabbits with rabbit blood leads, to the produc- 
tion of auto-agglutinins; also that repeated bleedings lead to the same 
results.8 Both observations are of considerable interest in connection with 
blood transfusion. 
Recently Clough and Richter9 have reported and studied auto-aggluti- 
nation with the serum of a man suffering with bronchopneumonia. They 
found that this auto-agglutinin differed from ordinary agglutinins in various 
ways as follows: (1) It was active only at low temperatures, the agglutina- 
tion breaking up on wrarming confirming Landsteiner’s observations in this 
regard. (2) It was absorbed from the serum only at low temperature, and 
was liberated from the cells on warming. (3) It wras active on red blood- 
corpuscles of other animals. Absorption tests showed that this auto-agglu- 
tinin was distinct from other hetero-agglutinins and iso-agglutinins for 
corpuscles of Groups 1 and 2 contained in the serum of this individual. 
Auto-agglutinin wras also found in the serum of a daughter of the patient, 
indicating a hereditary peculiarity. Kligler10 has also described autohem- 
agglutinins in the serum of a pregnant woman suffering from a chronic heart 
1 Jour. Med. Research, 1903, 9, 28. 
2 Ztschr. f. klin. Med., 1907, 61, 427. 
3 Wien. klin. Wchn., 1902, 15, 413. 
4 Munch, med. Wchn., 1903, 40, 1812; ibid., 1902, 39, 1905. 
8 Jour. Infect. Dis., 1907, 4, 297. 
6 Jour. Med. Res., 1915, 33, 213. 
7 Jour. Exper. Med., 1918, 27, 509. 
8 Jour. Exper. Med., 1918, 27, 563. 
9 Johns Hopkins Hosp. Bull., 1918, 29, 86. 
10 Jour. Amer. Med. Assoc., 1922, 78, 1195. ISOHEM A GGL U TIN I NS 
293 
lesion, and in a condition of severe anemia due to repeated hemorrhages 
from hemorrhoids. Auto-agglutination occurred only at room temperature 
and not at 37° C.; furthermore, agglutinated masses of corpuscles at room 
temperature were dispersed when warmed. 
ISOHEMAGGLUTININS 
Agglutinins in human sera for the corpuscles of other human beings 
have assumed considerable practical importance in relation to blood trans- 
fusion and particularly of adults. 
Transmission and Influence of Age.—Halban1 was first to note that 
agglutinins may be absent from the blood of an infant while present in 
that of the mother. Unger2 found agglutinins in only 13 per cent, of infants 
under one month of age, although present in 97 per cent, of adults. The 
percentage of children whose sera contained agglutinins increased with 
age, but reached the adult average only between the second and fourth 
years of life. Likewise, the corpuscles of infants cannot be agglutinated 
by any serum in the majority of instances. According to Happ3 and Hess4 
the cells of newborn infants are rarely agglutinated, and the grouping present 
in adults rarely present in blood from the umbilical vein. The acquisition of 
susceptibility to agglutination occurs at about six months and far earlier than 
the appearance of agglutinins in the serum. At about two years of age all 
children have probably established their adult iso-agglutinin group (Happ). 
For these reasons Unger believes it is safe to transfuse an infant of six months 
or less with the blood of any group, but Happ states that the agglutinins 
in the blood of mother and child may be different and that it is not safe 
to transfuse an infant from its mother without making preliminary tests. 
Furthermore, Jones (A.) has recently reported that 78.7 per cent, of 197 
specimens of infant blood examined could definitely be placed in one of the 
four recognized groups. These results seemed to be dependent on a technic 
which permitted the recognition of weak agglutinins. Isohemolysins were 
also found in 13.7 per cent, of sera, and because of these iso-agglutinins 
and isohemolysins Jones also advises that compatibility blood tests be 
made in selecting a donor for transfusion of infants. 
Of interest in this connection it may be stated that Happ found agglu- 
tinins in the milk of 14 nursing women identical in grouping to those found 
in their sera. Of the 14 infants only 5 showed the presence of agglutinins, 
and Happ concludes that the infant does not acquire agglutinins through 
the mother’s milk. 
According to H. and L. Hirschfeld,5 Verzar and Weszecsky,6 and Lewis 
and Henderson7 the distribution of four groups of hemagglutinins varies 
extremely, but certain races could be grouped corresponding somewhat to 
the grouping made on an anthropologic basis and still more completely 
according to geographic distribution of the races. 
Classification and Grouping.—Landsteiner8 is credited with having dis- 
covered that human sera contain more than one iso-agglutinin; this dis- 
covery has been amply confirmed and has led to various classifications of 
the iso-agglutinins by different investigators and much confusion. 
1 Wien. klin. Wchn., 1900, 13, 545; Munch, med. Wchn., 1902, 49, 473. 
2 Jour. Amer. Med. Assoc., 1921, 76, 9. 
3 Jour. Exper. Med., 1920, 31, 313. 
4 Deut. med. Wchn., 1921, 47, 241. 
“Lancet, 1919, 2, 675. 
6 Biochem. Ztschr., 1921, 126, 33. 
7 Jour. Amer. Med. Assoc., 1922, 79, 1422. 
8 Centralb. f. Bakteriol., Abt., 1900, 27, 357. 294 
HEMAGGLUTININS 
Landsteiner expressed the belief that human sera contain two iso-agglu- 
tinins and that the corpuscles may be divided into two kinds according to 
their susceptibility to agglutinins and two agglutinogens. This work was 
quickly confirmed by the investigations of Shattock,1 Donath,2 Grunbaum,3 
von Descatello and Sturli,4 Eisenberg,5 and others who studied principally 
the bloods of sick persons, with the result that the presence of these iso- 
agglutinins were considered pathologic until Landsteiner6 showed that they 
occurred in normal blood and could be divided into three main groups, 
von Descatello and Sturli added a fourth group, and Hektoen7 also found 
three iso-agglutinins in human sera instead of the two described by Land- 
steiner. 
The next important contributions were made by Jansky8 and Moss,9 
who divided the blood of all human beings into four groups. At the present 
time the classification of Landsteiner plus the fourth group described by von 
Descatello and Sturli, the classification of Jansky and that of Moss, are usually 
described in text-books, and inasmuch as they do not correspond has led to 
a great deal of confusion, and the possibility of error and danger in blood trans- 
fusion. 
Jansky’s classification is largely employed in Europe and that of Moss 
in America; in order to standardize the practice of grouping blood and 
remove the danger of using both systems, a committee composed of mem- 
bers of the American Association of Immunologists, the Society of American 
Bacteriologists, and the Association of Pathologists and Bacteriologists 
have recently reviewed the subject, pointed out the dangers, and recom- 
mended the adoption of one classification in order to avoid confusion and 
the possibility of accident; on the basis of priority the classification of 
Jansky was recommended.10 
The following table shows how the groups of Jansky correspond to 
those of Landsteiner and Moss: 
Jansky’s. 
Corresponds to: 
Landsteiner’s. 
Moss’. 
Group I 
Group C 
Group IV 
Group II 
Group A 
Group II 
Group III 
Group B 
Group III 
Group IV 
Descatello-Sturli Group 
Group I 
Relation of Different Classifications of Human Iso-agglutinins 
From this table it will be noted that the classifications of Jansky and Moss 
are identical excepting that Groups I and IV are interchanged. 
Karsner11 has recently summarized the average incidence of these groups 
1 Jour. Path, and Bacteriol., 1900, 6, 303. 
2 Wien. klin. Wchn., 1900, 13, 497. 
3 Brit. Med. Jour., 1900, 1, 1089. 
4 Munch, med. Wchn., 1902, xlix, 1090. 
6 Wien. klin. Wchn., 1901, 14, 1020. 
6 Wien. klin. Wchn., 1901, 14, 1132. 
7 Jour. Infect. Dis., 1907, 4, 297. 
8 Shorn. Klin., 1907, 8, 85. 
9 Bull. Johns Hopkins Hosp., 1910, 21, 63. 
10 Jour. Amer. Med. Assoc., 1921, 76, 130. 
11 jour. Amer. Med. Assoc., 1921, 76, 88. ISOHEMAGGLUTININS 
295 
based upon more than 5000 tests made by five different investigators on 
the basis of Jansky’s classification; I have modified his table to show its 
relation to the Moss classification: 
Incidence of the Four Groups of Iso-hemagglutinins in Man 
Jansky. 
Corresponds to 
Moss. 
Incidence. 
Group I 
Group IV 
42.84 per cent. 
Group II 
Group II 
41.38 per cent. 
Group III 
Group III 
10.36 per cent. 
Group IV 
Group I 
5.42 per cent. 
It will be noted that according to the Jansky classification Groups I 
and II (corresponding to Moss’ Groups II and IV) preponderate, and col- 
lectively constitute about four-fifths of the race; the percentages observed 
by Culpepper and Ableson1 on the basis of 5000 tests were similar. 
The technic of grouping is described later in this chapter; the important 
relation of grouping to blood transfusion and the reactions following this 
operation are discussed under Blood Transfusion. 
Subgroups.—The possible existence of specific subgroups or minor agglu- 
tinins is still an open question. Langer2 claimed that one serum he examined 
contained six iso-agglutinins demonstrable by successive absorptions with dif- 
ferent agglutinable corpuscles. Culpepper and Ableson3 and Unger4 believe 
that the groups may “overlap” and thus explain reactions between members 
of the same group. For this reason Unger advises that recipient and donor 
be tested directly against each other, and that it should not be assumed 
that bloods are compatible merely because typing with Groups II and III 
sera show that they belong in the same group. 
Isohemagglutinins in the Lower Animals.—Hektoen was not able to 
find iso-agglutinins in the serum of rabbits, guinea-pigs, dogs, horses, and 
cattle; as far as I am aware these observations have not been disputed. 
The subject is worthy of further investigation by reason of the unique 
situation as compared with human sera, and especially so since heterol- 
ogous hemagglutinins occur in the sera of the lower animals. 
Heterologous Hemagglutinins in Human Blood.—As previously stated, 
these are agglutinins in human blood for the corpuscles of the lower animals. 
They have been studied largely in relation to the Wassermann test as in- 
fluencing the choice of corpuscles for the indicator antigen. In studies 
of this kind Kolmer, Matsunami, and Trist5 found agglutinins in the un- 
heated sera of human beings for the corpuscles of a number of the lower 
animals in the following percentages: 
For sheep corpuscles 16 per cent. 
For chicken corpuscles 10 “ 
For guinea-pig corpuscles 28 “ 
For ox corpuscles occasionally 
1 Jour. Lab. and Clin. Med., 1921, 6, 276. 
2 Ztschr. f. Heilk., Abt. Int. Med., 1903, 24, 111. 
3 Loc. cit. 
4 Jour. Amer. Med. Assoc., 1921, 76, 9. 
5 Amer. Jour. Syph., 1919, 3, 407. 296 
HEMAGGLUTININS 
These agglutinins are prepared by injecting the corpuscles of one animal 
into a second animal of a different species. Curiously the injection of rabbits 
with human corpuscles is followed by the production of agglutinins to a 
somewhat greater degree than occurs when rabbits are injected with sheep 
corpuscles or those of the ox. They are best produced by injecting washed 
corpuscles, but are also produced by injecting whole blood and even serum. 
Their presence in immune serum is disturbing in complement-fixation 
tests, and particularly antihuman agglutinin in rabbit antihuman hemo- 
lytic serum. So far no adequate method has been described for the com- 
plete removal of these agglutinins without removal of the hemolysins at 
the same time; desiccation, however, tends to destroy the agglutinins some- 
what more than the hemolysins as shown by the studies of Sands and West1 
in my laboratory. 
The immune hemagglutinins were apparently first observed by Belfanti 
and Carbone in the serum of a horse injected with rabbit blood. As shown 
later by Landsteiner2 and von Dungern3 heating immune serum for thirty 
minutes to 55° to 60° C. may remove hemolysins, but not hemagglutinins. 
Curiously the injection of rabbits with goose blood may produce agglu- 
tinins not only for goose corpuscles, but raise considerably the amount of 
normal or natural agglutinins in rabbit serum for the corpuscles of the 
guinea-pig, sheep, and other animals. A similar result is observed when 
rabbits are injected with guinea-pig corpuscles. 
Immune Isohemagglutinins.—Human sera must be used for the typing 
of human corpuscles. Since these do not keep well Kolmer and Trist4 
attempted to prepare immune hemagglutinins for the four groups of human 
corpuscles by injecting rabbits with cells of the four different groups. Our 
antisera often showed somewhat higher titers for cells of the group used in 
immunization, but absorption to remove group antibodies resulted in com- 
plete exhaustion of agglutinins and lysins. Coca, likewise, failed to pro- 
duce specific antihuman heterohemagglutinins and attributes these results 
to the presence of an antigenic complex common to all groups. Hooker 
and Anderson,5 however, have recently described the production of specific 
group hemagglutinins by some rabbits injected intravenously with small 
amounts of human corpuscles. These were demonstrated by absorption 
tests. No reasons were given for individual variation among rabbits in 
their response to injections with human cells. These investigators found 
that normal rabbit sera possess weak agglutinins for the four groups of 
human corpuscles and particularly for Group II and IV cells. Human 
sera likewise agglutinate rabbit corpuscles, but without group specificity. 
Group specific hemolysins for human corpuscles were found. 
The Influence of Temperature Upon Hemagglutination.—Agglutination 
of corpuscles is usually a rapid phenomenon. With microscopic tests con- 
ducted at ordinary room temperatures agglutination is well marked within 
fifteen minutes and usually within a few minutes when large amounts of 
agglutinins are present. Macroscopic tests are usually conducted by in- 
cubation for an hour at 37° C., but agglutination occurs at lower tempera- 
tures equally well and very rapidly. Jervell6 has recently shown that the 
most pronounced agglutination occurs in the refrigerator and that absorp- 
tion of agglutinin occurs more quickly and completely at low than at high 
temperature. 
IMMUNE HEMAGGLUTININS 
1 Jour. Immunology, 1919, 4, 275. 
2 Centralbl. f. Bakteriol., 1899, 25, 546. 
3 Munch, med. Wchn., 1899, 405. 
4 Jour. Immunology, 1920, 5, 89. 
5 Jour. Immunology, 1921, 6, 419. 
6 Jour. Immunology, 1921, 6, 445. MEDICOLEGAL APPLICATION OF HUMAN BLOOD GROUPING 
297 
MEDICOLEGAL APPLICATION OF HUMAN BLOOD GROUPING 
In 1908 Epstein and Ottenberg1 noticed that the groupings of human 
erythrocytes were hereditary and followed Mendel’s law. In 1910 Von 
Dungern and Hirschfeld2 examined 348 individuals belonging to 72 different 
families and proved the correctness of Ottenberg’s observations. Otten- 
berg3 has recently reviewed the subject very carefully and concludes that 
blood grouping may prove of aid in determining parenthood in medico- 
legal cases involving the charge of illegitimacy. 
Von Dungern and Hirschfeld have shown that the two agglutinogens 
A and B found in human corpuscles never occur in a child if not present 
in one of the parents; that when one of these is present in both parents it 
occurs in most of the children; when one (A or B) is present in only one 
parent, the child may inherit it; when neither parent has one of these sub- 
stances (A or B) the child never shows it. Medicolegally, if A or B is 
present in a child’s blood one of the alleged parents must possess it. 
In determining the legitimacy of a child its blood and that of the mother 
and father are tested and grouped according to the Jansky classification. 
In cases of illegitimacy a fourth person is to be included, the alleged father 
or mother, as the case may be. The commonest instance, of course, is that 
of disputed paternity. According to Ottenberg’s analysis of the work of 
Von Dungern and Hirschfeld, the results may be interpreted as follows: 
1. If the woman and the man belong to Group I the child must be Group 
I to be theirs; if the child belongs to II, III, or IV it is not the child of the 
supposed father or mother, as the case may be. 
2. If the woman is I and the man II or the woman is II and the man I 
the child must be I or II; if the child is III or IV it is not the child of the 
supposed father or mother, as the case may be. 
3. If the woman is I and the man is III, or the woman is III and the 
man is I, the child must be I or III; if the child is II or IV it is not the child 
of the supposed father or mother, as the case may be. 
4. If the woman is II and the man is II the child may be I or II; if the 
child is III or IV it is not the child of the supposed father or mother, as the 
case may be. 
5. If the woman is III and the man is III the child may be I or III; 
if the child is II or IV it is not the child of the supposed father or mother, 
as the case may be. 
The above unions comprise over 80 per cent, of all unions and are those 
on which the kind of possible offspring is definitely limited; they are the 
instances which may be of medicolegal value. The test, therefore, may not 
furnish evidence under the following two conditions: 
6. All unions containing a member of Group IV and unions of II and III, 
may produce offspring of any of the four groups. 
7. A child of Group I may result from any combination of parents. 
If the child’s blood is the correct group for the alleged parents, then 
the child could be their offspring, but not that it must be necessarily so. 
But, on the other hand, if the child’s group is wrong for the alleged parents, 
then according to Ottenberg’s analysis, one can say that the child must 
have a parent other than one of those asserted. 
In infants and very young children the test can only be relied upon if 
it shows definite group characteristics. As stated above, the corpuscles 
of children acquire susceptibility to agglutination at about six months of 
1 Trans. New York Pathol. Soc., 1908, 8, 117. 
2 Ztschr. f. Immunitatsf., 1910, 6, 284. 
3 Jour. Immunology, 1921, 6, 363; Jour. Amer. Med. Assoc., 1922, 78, 873. 298 
HEMAGGLUTININS 
age, which is considerably earlier than the appearance of agglutinins in 
their serum. Therefore this medicolegal test may not prove of value 
unless the child is at least six months of age; the test is apt to be most accu- 
rate when the child is two or more years of age. 
Buchanan1 has questioned this medicolegal application of blood grouping 
and states that on the basis of Mendelian laws, the only instances in which 
it appears that the blood group could be held as direct evidence would be 
in a family of four or more children of whom one was of a different group 
than the evident group represented in both parents, and all four grand- 
parents. Buchanan states that a grandparent might be a heterozygote, in 
virtue of which she might transmit a character to a son or daughter, who 
in turn might be a heterozygote, and finally, in a family at issue the long- 
concealed character or group might appear. Gichner,2 Learmouth,3 Keynes,4 
Weszecsky,5 Tebbutt, and McConnel,6 however, have confirmed the hered- 
itary nature of the blood group, and that when the group of the child and 
one parent is known, one may sometimes state to what group the other 
parent must belong. It is never possible, however, to say from the children 
alone to what group or groups the parents must belong. 
For the Medicolegal Identification of Blood-stains.—There are no 
available means for differentiating human blood-stains, although by means 
of precipitin and complement-fixation tests a stain of human blood may be 
identified and differentiated from the bloods of the lower animals. However, 
if fresh blood is obtained before drying agglutination tests are of aid to this 
extent: if a blood-stain is from one of two persons, and it is found that these 
two individuals belong to different groups, typing the corpuscles of the 
blood-stain would indicate from which of the two persons the blood was 
derived; if the corpuscles are agglutinated by the serum of the accused 
individual, the blood could not be his own except in rare instances of auto- 
hemagglutination previously discussed. Dried blood-stains cannot be used 
because of hemolysis of the corpuscles. 
TECHNIC OF TESTS FOR ISOHEMAGGLUTININS AND ISOHEMOLYSINS 
As discussed in the chapter on Blood Transfusion the bloods employed 
should be compatible, that is, the blood of the donor should not agglutinate 
or hemolyze the corpuscles of the patient, and vice versa. It is generally 
agreed that tests for agglutinins alone are sufficient preliminary to blood 
transfusion on the basis of the assumption that when agglutinins are absent 
hemolysins are likewise. 
However, in my experiments, I have found that hemolysins may be 
present in an occasional serum when the agglutinins are either entirely 
absent or present in such small amounts as to escape detection. For this 
reason I believe that tests for both agglutinins and hemolysins should 
be conducted, but of the two the former is much more important and the 
latter may be omitted. In addition to these tests the serum of the donor 
should be submitted to the Wassermann test if time and opportunity per- 
mit. The relation of syphilis to transfusion is discussed in the aforemen- 
tioned chapter. 
Macroscopic Test for Agglutinins and Hemolysins.—1. About 1 c.c. 
of blood is obtained from each donor from a prick of the finger in a small 
1 Jour. Amer. Med. Assoc., 1922, 78, 89; ibid., 1922, 79, 180. 
2 Jour. Amer. Med. Assoc., 1922, 79, 2143. 
3 Jour. Genetics, 1920, 10, 141. 
4 Blood Transfusion, Oxford Med. Publications, 1921, 90. 
5 Biochem. Ztschr., 1920, 107, 159. 
6 Med. Jour. Australia, 1922, 1, 201. TESTS FOR ISOHEMAGGLUTININS AND ISOHEMOLYSINS 
299 
tube containing 2 c.c. of a 1 per cent, sodium citrate in normal salt solu- 
tion. An equal amount is collected in a small, dry test-tube; when coagula- 
tion has occurred the serum is separated, or secured by breaking up the 
clot and centrifuging. 
2. From the recipient 1 c.c. of blood is placed in 2 c.c. of sodium citrate 
solution, and an equal amount is collected in a dry tube, allowed to coagu- 
late, and the serum collected. 
3. The sodium citrate tubes are centrifuged; the supernatant fluid is 
pipeted off, and the cells are washed again with normal salt solution. After 
the final washing enough normal salt solution is added to the sediment of 
cells to bring the total volume up to 1 c.c. 
4. The serum tubes are also centrifuged, so that clear serums are ob- 
tained. These should preferably be free from hemoglobin stain. 
5. The following mixtures should he set up within twenty-four hours of 
the time of collecting blood in order that native complements may not have 
undergone deterioration. Measurement may be made according to a drop 
from an ordinary 1 c.c. graduated pipet 
held vertically. Small sterile test-tubes 
(8 by 1 cm.) are to be used. 
Tube 1: 4 drops of donor’s serum 
4- 1 drop of recipient’s red-cell 
suspension. 
Tube 2: 4 drops of recipient’s serum 
+ 1 drop of donor’s red-cell sus- 
pension. 
Tube 3: Control: 4 drops of donor’s 
serum + 1 drop of donor’s red- 
cell suspension. Should show no 
agglutination and no hemolysis. 
Tube 4: Control: 4 drops of recipi- 
ent’s serum + 1 drop of recipi- 
ent’s red-cell suspension. Should 
show no agglutination or hemol- 
ysis. 
Tube 5: Control: 1 drop of donor’s 
red-cell suspension + 4 drops of 
normal salt solution. This serves 
as a control on the fragility of 
the corpuscles and isotonicity of 
the salt soluion. 
Tube 6: Control: 1 drop of recipi- 
ent’s red-cell suspension + 4 
drops of saline solution. 
One c.c. of salt solution is added to each tube and the tubes are 
gently shaken and placed in the incubator for two hours. They are to 
be inspected every half-hour. Agglutination is recognized macroscopic- 
ally by the clumping of the red blood-cells into small masses which cannot 
he easily broken up by agitation and that later sink to the bottom of the 
tube as a small clot (Fig. 105). Hemolysis is likewise easily detected, as 
corpuscles tend to become hemolyzed within two hours. If doubt exists, 
the finer grades of hemolysis may be detected after the tubes have been 
allowed to stand overnight in an ice-chest, or at once by thorough centrif- 
ugalization. This method requires considerable serum and weak agglutina- 
tion may be overlooked. 
Fig. 105.—Macroscopic Hemagglutina- 
tion. 
The tube on the left shows agglutinated 
masses of corpuscles; the middle tube shows a 
negative reaction, and the tube on the right is 
a corpuscle control. 300 
HEMAGGLUTININS 
Microscopic Test for Hemagglutinins; Direct Method.—In this test no 
attempt is made to group the bloods; it simply consists of a direct test of 
the serum of the patient (recipient) for agglutinins for the corpuscles of 
each donor and of the serum of each donor for agglutinins for the corpuscles 
of the patient. Therefore, it does not require the preservation of group 
sera, is simple and efficient. Unger1 has recently stated that even though 
the bloods of both recipient and donor are grouped, this direct test should 
be made in addition. I conduct the test as follows: 
1. About 5 to 10 drops of blood are secured from a finger of the patient 
in a small test-tube carrying about 1 c.c. of 2 per cent, sodium citrate in 
physiologic saline solution. At the same time about 0.5 to 1 c.c. of blood 
is secured in a small dry test-tube for serum. 
2. Blood is secured from each donor in the same manner. 
Shows a positive agglutination of human corpuscles by human serum. 
Fig. 106.—Microscopic Hemagglutination. 
3. The suspensions of corpuscles are centrifuged and the supernatant 
fluids discarded; to the corpuscles in each tube is added about two volumes 
of saline solution and gently shaken. This step is not absolutely necessary 
as the suspension in citrate may be used. 
4. The clots are gently broken up if necessary and centrifuged to secure 
a clear layer of serum in each tube. 
5. The tests are now set up with cover-glasses and vaselined hanging 
drop slides as follows: (a) Two loopfuls (ordinary 4-6 mm. loop of platinum 
wire) of patient’s serum is placed on a cover-glass; one loopful of donor’s 
corpuscles are added, mixed, and the slide suspended. Similar slides are 
prepared with the corpuscles of each donor. (b) Two loopfuls of the donor’s 
serum is placed on a cover-glass; one loopful of the patient’s corpuscle 
suspension is added, mixed, and the slide suspended. Similar slides are 
prepared with the serum of each donor, (c) Two loopfuls of saline solution 
1 Jour. Amer. Med. Assoc., 1921, 76, 9. TESTS FOR I SOU EM A GGLU TIN INS AND ISOHEMOLYSINS 
301 
are placed on a cover-glass; one loopful of patient’s corpuscles are added, 
mixed, and the slide suspended. A similar preparation is made with the 
corpuscles of each donor. These are the corpuscle controls. 
6. Each slide is examined microscopically (with -f objective) after 
standing fifteen minutes at room temperature. The controls are first in- 
spected and should show no agglutination. As a general rule there is no 
difficulty in deciding upon the results; Fig. 106 shows a negative reaction 
and Fig. 107 a positive reaction. As shown in Fig. 108 the corpuscles at the 
margin of the slide may show a tendency to clumping or rouleaux forma- 
tion from the effects of drying; these should not be mistaken for true agglu- 
tination. 
Rous and Turner1 have described a method similar to the above except 
that blood from patient and each donor is collected with a white corpuscle 
pipet and the tests set up in capillary tubes instead of test-tubes. 
Microscopic Test for Hemagglutins; Grouping of Blood.—For these 
tests it is necessary to have available known Group II and III sera; with 
these it is possible to determine to which of the four groups the blood of the 
Fig. 107.—Negative Hemagglutination Reaction. 
patient belongs, and a corresponding or compatible donor belonging to the 
same group is selected for transfusion. 
Preservation of the Agglutinating Sera.—A laboratory instituting these 
tests should secure small amounts of Group II and III sera from some other 
laboratory or from adults who have been reliably grouped. 
These sera are best preserved in small ampules with 0.2 per cent, tric- 
resol at a low temperature. Sanford2 has described a method of preserving 
the sera by placing a loopful on each of a large number of cover-glasses, 
drying, wrapping in paper, and keeping in a refrigerator. Karsner and 
Koeckert,3 however, found that the agglutinins deteriorated during drying 
and that group specificity was lost after several weeks. Kolmer4 likewise 
found that with some sera drying on cover-glasses destroyed the agglu- 
tinins; for this reason only sera containing relatively large amounts of 
agglutinin should be used for Sanford’s method, and it is well to test several 
glasses a few days after drying to make sure that agglutinins are present. 
With these precautions the method may be satisfactory; I have not clearly 
1 Jour. Amer. Med. Assoc., 1915, 64, 1980. 
2 Jour. Amer. Med. Assoc., 1918, 70, 122. 
3 Jour. Amer. Med. Assoc., 1919, 74, 1207. 
4 Jour. Amer. Med Assoc., 1919, 73, 1459. 302 
HEMAGGLUTININS 
observed the loss of group specificity described by Karsner and Koeckert, 
although it may be that I did not keep my preparations sufficiently long 
under observation. Holt and Reynolds1 have recently stated that the 
hemagglutinins are contained in the pseudoglobulin fraction of human serum, 
none being found in the euglobulin and albumin fractions. Desiccated 
pseudoglobulins were found to retain agglutinating properties for two and a 
half months. At any rate, I am quite sure that preservation of the sera in 
fluid form as described is just as simple and probably much better. 
After one secures a little Group II and III sera for starting, an ample 
supply of each is easily provided for future use. It is my practice to type the 
bloods sent to the laboratory for the Wassermann test, and secure sera in 
this manner. After drawing off the sera from a number of specimens, cor- 
puscles from each are easily secured, and the group to which each belongs 
determined by the method described below. For example, if the corpuscles 
Fig. 108.-Fai.se Clumping and Rouleaux Formation Due to Drying. 
of an individual belong to Group II the serum does likewise and is pre- 
served with tricresol in a refrigerator. In this manner fresh Group II and 
III sera are obtained at least once a month; even Group I and IV sera may 
be secured and kept in the same manner for controls, although these are 
not absolutely necessary. 
It is necessary to depend upon human sera; immune sera cannot be 
prepared. Kolmer and Trist2 have immunized rabbits with corpuscles 
belonging to the four groups; the immune sera agglutinated the corpuscles 
of the corresponding group slightly better than the corpuscles of other 
groups, but not sufficiently well to permit the use of such sera for grouping. 
In other words, rabbits injected with Group II corpuscles produced agglu- 
tinins for the corpuscles of all four groups, but slightly better for Group II; 
absorption of such sera with Group I, III, and IV corpuscles removed all 
agglutinin so that such sera could not be purified. 
Technic of Tests.—The following method is essentially similar to that 
1 Jour. Amer. Med. Assoc., 1922, 79, 1684. 
2 Jour. Immunology, 1920, 5, 89. TESTS FOR ISOHEMAGGLUTININS AND ISOHEMOLYSINS 
303 
described by Brem1 and Lee2 and found very useful during the war, especially 
by Karsner.3 
1. From 0.5 to 1 c.c. of 1 per cent, sodium citrate in physiologic saline 
solution is placed in a small test-tube; 2 or 3 drops of the blood to be tested 
are dropped in, secured by pricking the finger. 
2. Three cover-glasses are cleaned; upon No. 1 are placed 2 loopfuls 
of saline solution; upon No. 2, 2 loopfuls of Group II serum, and upon No. 3, 
2 loopfuls of Group III serum. One loopful of corpuscle suspension is 
added and mixed with the fluid on each slide. No. 1 is the corpuscle control. 
3. Each slide is inverted in position on a vaselined Widal hanging drop 
slide, and examined microscopically with a f objective after standing 
fifteen minutes. The control (No. 1) should show no agglutination; a nega- 
tive reaction is shown in Fig. 107, and due care must be exercised against 
regarding rouleaux formation as agglutination. A positive reaction is shown 
in Fig. 106, and is usually easily read. 
4. The grouping is made with slides No. 2 and 3 as follows: 
(a) No agglutination in either slide: The blood belongs to Group I 
of the Jansky classification or Group IV of the Moss classification. 
(b) Agglutination in both slides: The blood belongs to Group IV of the 
Jansky classification, or Group I of the Moss classification. 
(c) Agglutination with Group II serum; no agglutination with Group 
III serum: the blood belongs to Group III in both the Jansky and Moss 
classifications. 
(d) Agglutination with Group III serum; no agglutination with Group 
II serum: The blood belongs to Group II in both the Jansky and Moss 
classifications. 
Since both the Jansky and Moss classifications are in use, the worker must 
state in his report the classification employed; otherwise there is danger, if a 
donor has been grouped according to one method and the patient by the 
second, of confusing Group I and Group IV bloods. As previously stated, 
the Jansky classification has been recommended (purely on the basis of 
priority) for general adoption in order to avoid this confusion. 
Vincent’s Open Slide Method.—Vincent4 has described a simple slide 
method which Ottenberg5 has found very satisfactory: 
1. One drop of Group II serum is placed on the left-hand side of a slide 
and 1 drop of Group III serum on the right-hand side. 
2. The finger or ear of the patient is pricked and a small drop of blood 
transferred by the knife or lancet to the Group II serum and mixed; the in- 
strument is now cleansed and blood transferred to the Group III serum; 
or 2 to 3 drops of blood may be collected in 1 c.c. of 2 per cent, sodium 
citrate solution as described above, and a drop added to each of the two sera. 
3. The slide is tilted and rotated gently, so that the cells are uniformly 
distributed; this is repeated every few minutes. Agglutination is easily 
seen with the naked eye in from one to ten minutes at room temperature. 
Rouleaux formation usually appears more slowly than agglutination, and the 
clumps are broken up by rocking the slide, which inreases agglutination. In 
doubtful cases the reading should be confirmed by microscopic examination. 
As suggested by Culpepper and Ableson,6 the dried tests may be kept as 
permanent records. According to Ottenberg, false agglutination due to 
drying does not occur if the slides are not moved after the first ten or fifteen 
minutes, and chipping of the dried specimens may be prevented by painting 
them with a layer of collodion. 
1 Jour. Amer. Med. Assoc., 1916, 67, 191. 
2 Brit. Med. Jour., 1917, 2, 684. 
3 Jour. Amer. Med. Assoc., 1918, 70, 769. 
4 Jour. Amer. Med. Assoc., 1918, 70, 1219. 
5 Jour. Amer. Med. Assoc., 1922, 79, 2137. 
3 Jour. Lab. and Clin. Med., 1921, 6, 276. CHAPTER XVII 
PRECIPITINS 
Closely allied to the agglutinins are antibodies known as precipitins. 
They act on dissolved albuminous bodies in a manner quite similar to the 
action of agglutinins upon formed cellular elements. For example: (1) 
If typhoid immune serum is added to a bouillon culture of typhoid bacilli 
agglutination occurs; (2) if the culture is filtered and the immune serum 
is added to the clear sterile filtrate cloudiness appears and finally a precipitate 
forms. The first is an example of the action of agglutinins upon the formed 
bacilli, and the second illustrates the action of precipitins upon the albumins 
of dead and dissolved bacilli. 
Precipitins are formed not only for bacterial albumins, but for most 
any soluble animal {zooprecipitin) and vegetable protein (phytoprecipitin) 
as well. If the serum of a rabbit immunized with horse-serum is added to 
horse-serum a precipitate forms owing to the presence of a specific pre- 
cipitin in the immune serum. Normal rabbit-serum does not possess this 
power. 
Definition.—The precipitins are specific antibodies that develop in the 
serum of animals inoculated with bacteria or with solutions of animal or vege- 
table albumins which possess the power of producing a precipitate in a clear 
solution of the particular albumin or culture filtrate against which the animal 
has been immunized. 
Historic.—Kraus1 was the first to study and describe the bacterial 
precipitins. He observed that when the serums of animals that have been 
immunized against cholera, typhoid, or plague are added to a clear filtrate 
of the respective bouillon cultures of their bacteria, instead of to the bac- 
teria themselves, the clear solution becomes turbid and a precipitate forms. 
This reaction was found to be quite specific, i. e., it occurs best with the 
filtrate of the homologous bacteria, and to a much less extent with closely 
allied species. For example, the typhoid immune serum does not produce 
a precipitate with a filtrate of Spirillum cholerae, and similarly a cholera 
immune serum does not produce a precipitate with the filtrate of Bacillus 
typhosus. Kraus advocated the precipitin reaction as a means of identify- 
ing and differentiating certain species of bacteria, but the test possesses 
no advantage for these purposes over agglutination reactions and is not 
generally employed. 
Tchistovitch2 was the first to call attention to the non-bacterial pre- 
cipitins. This observer found that the serum of rabbits inoculated with 
eel-serum, when mixed with a small quantity of the eel-serum, caused a 
precipitate to form. 
About the same time (1899) Bordet found that the serum of rabbits 
inoculated with the serum of chickens, when mixed with the chicken-serum, 
gave a specific precipitate. A little later Bordet3 produced an antimilk 
immune serum (lactoserum) by inoculating rabbits intraperitoneally with 
milk partially sterilized by heating to 65° C. When this immune serum 
was mixed with the homologous milk, small particles appeared, which 
gradually formed larger flakes and sank to the bottom of the fluid. It was 
1 Wien. klin. Wchn., 1897, 10, 736. 
2 Ann. d. l’lnst. Pasteur, 1899, 13, 413. 
3 Ann. d. l’lnst. Pasteur, 1899, 13, 240. 
304 KINDS OF PRECIPITINS 
305 
found that the lactoserums were specific, i. e., cow lactoserum would pre- 
cipitate only cow casein, human serum only human casein, etc. 
Wladimiroff was the first; to use the bacterial precipitin reaction as a 
practical diagnostic test. He showed that the serum of a horse suffering 
from glanders would, when added to a clear filtrate of a culture of Bacillus 
mallei, produce a precipitate. The technic of these reactions is, however, 
more difficult than with the agglutination tests, and as the reactions are 
usually not more delicate or more advantageous than the latter, they are 
seldom employed. 
Wassermann1 and Schiitze2 made a very important practical demon- 
stration of the value of serum precipitins in differentiating the blood and 
secretions of man and animals. For example, the serum of rabbits im- 
munized with various bloods would react with solutions of old and dried 
specimens of their respective bloods, and although “group” precipitins 
were found present in the tests with the blood of closely allied species, yet 
the value of the reaction was not impaired to any extent when a p.roper 
technic, with correct dilutions, was employed. These discoveries were 
found to possess considerable value in forensic medicine, particularly in 
the recognition of the source of blood-stains. 
Nuttall,3 in a thorough and painstaking research with the blood from 
500 animals, was able to study the “blood relationship” of various animals 
as based upon group precipitins. For example, the serum of a rabbit im- 
munized with human blood will react best with human serum, then with 
the serums of the higher apes, and finally with the lower orders of monkeys. 
Similar reactions were found to occur among the lower animals. 
Nomenclature.—The antibody in an immune serum responsible for the 
phenomenon of precipitation is called precipitin; the substance or antigen 
responsible for the production of this antibody is known as the precipit- 
inogen; the precipitate is the end-product of the reaction between precipi- 
tinogen and precipitin. Just as toxoids and agglutinoids may be formed, 
so precipitin may be modified to precipitoid. 
Kinds of Precipitins.—Precipitins may be classed into two broad groups, 
namely, phy to precipitins, for the albumins of plants, and zoo precipitins, for 
the albumins of animals. 
In the group of plant precipitins are included those for bacteria, yeasts, 
fungi, and higher plants: 
Bacterioprecipitins include those found in the sera of animals normally 
or after immunization with bacteria, which act upon culture filtrates of the 
corresponding bacteria or upon a solution of the substance of these micro- 
organisms. They will be discussed in more detail later in this chapter. 
Mycoprecipitins include those for the albumins of yeasts and fungi. 
Schiitze4 claims to have produced them experimentally by injecting rabbits 
with the dissolved albumins of top and bottom beer yeasts, grain, and potato 
yeasts, prepared by rubbing up cultures in a mortar in sterile 25 per cent, 
soda solution with the addition of powdered glass and sand. A clear fluid 
was obtained by centrifuging, which was used for immunization and for the 
conduct of the tests; all of these yeasts reacted to the various antisera. 
Precipitins for the albumins of higher plants have also been produced 
experimentally probably first by Kowarski,5 who immunized rabbits with 
1 See article on Glanders in Kolle and Wassermann’s Handbuch, vol. 5, 2d ed. 
2 Ztschr. f. Hyg., 1901, 36, 5; Berl. klin. Wchn., 1901, 38, 187. 
3 Blood Immunity and Blood Relationship, Cambridge University Press, 1904. 
4 Ztschr. f. Hyg., 1901, 38, 493. 
5 Deutsch. med. Wchn., 1901, 27, 442. 306 
PRECIPITINS 
solutions of non-coagulable proteins from wheat flour and obtained a pre- 
cipitin for a saline extract of wheat flour. Weaker reactions were observed 
with extracts of the seeds of peas, rye, and bailey, but not with oats.1 
In the group of zooprecipitins are included those for milk, sera, bloods, 
inflammatory exudates, and transudates and meats. 
Lactoprecipitins or lactosera, first discovered by Bordet,2 occur in immune 
sera and precipitate milk casein. They will be discussed later. 
Hematoprecipitins include those for plasma, sera, and extracts of dried 
blood. Most investigations have been conducted with this large group 
and they have been found of great practical value and interest. 
In addition to these, precipitins have been produced for the albumins 
of meats, bones, egg-white, honey, cerebrospinal fluid, pleural exudates, 
urine, and other substances to which reference will be made later. 
The precipitins derive their names from their precipitinogens, as, for 
example, a precipitin produced by injecting rabbits with ox-serum is desig- 
nated anti-ox precipitin. 
Normal (Natural) Precipitins.—Although agglutinins may be found in 
normal serum, it is decidedly uncommon to find normal precipitins. By 
using low dilutions of normal sera Noguchi3 has demonstrated normal pre- 
cipitins in the blood of some of the cold-blooded animals. Ascoli4 has found 
ox-serum to contain precipitins for the sera of man, dog, pig, goat, rabbit, 
guinea-pig, and fowl; dog-serum to contain precipitins for the serum of the 
fowl and for egg-white; goat-serum for the sera of fowl and guinea-pig. 
These normal precipitins are very important when attempting to identify 
a blood, and may lead to error unless the specific precipitating antiserum 
is sufficiently powerful to work with the highest possible dilutions of the 
blood extract. 
I so precipitins have also been found by Ascoli, but are very rare; these 
are precipitins in the serum of an animal for the serum of another animal 
of the same species. Schiitze5 claims to have produced these precipitins 
in 2 rabbits by injecting rabbit-serum into 32 rabbits; he states that he 
found it more convenient to obtain isoprecipitins from goats treated with 
injections of goat milk over a period of a month or longer. 
Immune Precipitins.—These are ordinarily produced by immunization 
of animals with a foreign albumin; in man they may occur during the course 
of a bacterial infection or as the result of injections of bacterial vaccines, 
or the serum of the horse or other animal. 
Nature and Properties of Precipitins.—According to the side-chain 
theory, precipitins are antibodies or receptors of the second order, composed 
of a combining arm or haptophore group for the precipitinogen, and a zymo- 
phore or precipitinophore group that precipitates the antigen. Their struc- 
ture is, therefore, seen to be quite similar to that of agglutinin, the differ- 
ence being largely due to the different functions of the zymophore group. 
The properties of precipitins are quite similar to those of agglutinins. 
They are fairly resistant bodies, resist the effect of drying for prolonged 
periods, but are gradually destroyed by heating to 60° to 70° C. When 
inactivated by exposure or heat, they cannot be reactivated by the addi- 
tion of fresh normal serum, and, therefore, they bear no relation to the 
complements. 
1 Literature reviewed by Wells and Osborne, Jour. Infect. Dis., 1911, 8, 66. 
2 Ann. d. l’Inst. Pasteur, 1899, 13, 225. 
3 Univ. of Pennsylvania Med. Bull., 1902. 
* Munch, med. Wchn., 1903, No. 5. 
5Deutsch. med. Wchn., 1901, 28, 4; ibid., 804. ANTIPRECIPITINS 
307 
In their chemical nature precipitins are thrown down in the globulin 
fractions of the serum, although there is not a uniformity of opinion on the 
particular fraction carrying these antibodies. For example, Leblanc1 claims 
to have found them in the pseudoglobulins, while Eisenberg, Obermayer, 
and Pick2 identified them with the euglobulin fraction of immune serum, 
and this is the view most generally accepted. They are slowly destroyed 
by tryptic digestion, more slowly by pepsin. 
Precipitoids.—The haptophore or combining arm of precipitin is quite 
stable; the precipitophore group is more labile, and is affected by heat, 
and when this less resistant arm is lost, the receptor is called a precipitoid. 
Like agglutinoids, the precipitoids are of practical interest from the fact 
that their haptophore arm will not only combine with precipitinogen, but 
displays a greater activity in this direction than the whole receptor or pre- 
cipitin itself, and when union between precipitinogen and precipitoid has 
occurred, precipitation does not result. Hence in low dilutions of a precipitin 
serum the phenomenon of precipitation is slight or altogether absent, whereas 
in higher dilutions the reaction becomes evident. 
The precipitoids were first described by Muller,3 who heated a lacto- 
serum at 70° C. and found that it prevented precipitation when added to 
unheated and active lactoserum. They have been produced by Kraus and 
von Pirquet4 by heating typhoid and cholera bacterioprecipitins at 58° C.; 
these investigators also found them in old antisera. These sera cannot be 
activated by the addition of fresh serum, and the precipitoids are chiefly 
of interest in connection with the practical applications of the precipitin 
reaction, due care being necessary to avoid their neutralizing effect by using 
the precipitin sera in progressive dilutions. The mechanism of this phe- 
nomenon is very interesting and practically important. It finds a close 
analogy in the so-called protective colloids, the addition of precipitoid 
serum to a mixture of precipitin and antigen and the consequent preven- 
tion of precipitation being likened to the prevention of precipitation of 
mastic by normal serum by the addition of heated serum; however, this 
does not explain the specificity of the process. The prevention of precipita- 
tion of colloids by heated serum is non-specific but precipitoids are highly 
specific, that is, will only prevent the action of the corresponding precipitins. 
As shown by Welsh and Chapman,5 heating an immune serum from which 
the precipitin has been removed does not result in the formation of pre- 
cipitoids. According to these observers heated immune serum (precipitoid) 
prevents precipitation by active precipitin, and antigen by specifically dis- 
solving the precipitate which would otherwise make its appearance. 
Antiprecipitins.—These possess the power of neutralizing the precipitins, 
but occur in fresh unheated sera and should not be confused with the pre- 
cipitoids. Nuttall6 believes that they may occur in normal sera (natural 
antiprecipitins), although it is possible that the neutralizing effects of normal 
serum upon precipitation may be due to the solution of the precipitin in 
its homologous serum. 
Immune antiprecipitins have been produced by Kraus and Eisenberg7 
by treating goats with goat lactoserum obtained by injecting rabbits with 
goat’s milk; antilactoserum has also been prepared by Schiitze.8 
1 La cellule, 1901, 18, 337. 
2 Blood Immunity and Relationship, 96, 97. 
3 Miinch. med. Wchn., 1902, xlv, 1330. 
4 Centralbl. f. Bakteriol., 1902, 27, 60. 
5 Jour. Path, and Bacteriol., 1909, 13, 206. 
6 Blood Immunity and Relationship, p. 149. 
7 Wien. klin. Wchn., 1902, 27, 212. 3 Deutsch. med. Wchn., 1901, 28, 4. 308 
PRECIPITINS 
Group Precipitins.—These are not as prominent as group agglutinins, 
yet they are formed to a certain degree and are of much practical importance 
in attempting to differentiate bacteria and serums by the precipitation 
method. Although precipitins are highly specific, the principle of serum 
dilution, as emphasized under Agglutination, must be closely observed 
in order to dilute the group precipitins to such small amounts as to pre- 
vent them from interfering with the chief precipitin. This principle is of 
particular importance in differentiating the bloods of various animals, 
and especially in medicolegal cases, where the precipitin reactions are em- 
ployed for the diagnosis of blood-stains. From both a theoretic and prac- 
tical standpoint these group precipitins are of considerable importance, and 
I shall discuss them more fully later in a consideration of the specificity of 
precipitins. 
Origin of Precipitins.—Kraus and Levaditi1 have assigned the leuko- 
cytes as the chief cells concerned in the production of precipitins; Kraus 
and Scheffmann2 have expressed a similar claim for the vascular endothe- 
lium. Cantaguzene3 believes that these antibodies are formed chiefly in 
the lymphoid tissues and bone-marrow, and particularly by the mono- 
nuclear macrophages. The results of studies on the influence of x-rays 
upon precipitin production by Benjamin and Sluka,4 and upon antibody 
production in general by Hektoen,5 have shown that any agent that injures 
bone-marrow and lymphoid tissues reduces antibody production, indicat- 
ing the active participation of these tissues in antibody production. 
Production of Precipitins.—Immune serums for diagnostic purposes are 
produced by injecting the precipitinogenous fluid into the veins, peritoneal 
cavity, or subcutaneous tissues of animals, usually rabbits. 
As in the case of agglutinin formation, not all animals possess equally 
the power of forming a precipitin for a given albumin. Rabbits are generally 
employed; guinea-pigs produce unsatisfactory sera. While this point is 
of general interest with the bacterioprecipitins, it becomes of particular 
importance in relation to serum precipitins. For example, an animal will 
not form a precipitin active against its own serum. If formed, it would 
be an autoprecipitin, or iso precipitin, and, as a rule, animals do not form 
antibodies for their own tissue constituents. Furthermore, animals are 
unlikely to form precipitins for the proteins of other members of the same 
species, or if precipitins are produced, they are usually the result of pro- 
longed immunization of a number of animals. Precipitin formation is also 
slight for the proteins of other animals that are closely related either zoo- 
logically or biologically. For example, attempts at immunization of a 
guinea-pig with the serum of a rabbit, a pigeon with that of a hen, or a 
monkey with human serum, are procedures that do not usually yield good 
precipitating serums. The technic of preparing various precipitins is de- 
scribed later in this chapter under Practical Applications. 
Chemistry of Precipitinogens.—Only proteins are known to possess the 
property of inducing the formation of precipitins. Furthermore, they 
must be foreign proteins as previously stated. Almost any vegetable or 
animal protein may act as a precipitinogen under these conditions; of the 
serum proteins the globulins are apparently more active than the albumins. 
Obermayer and Pick6 and Jacoby7 claim to have produced non-protein 
1 G. r. Acad. d. Sci., 1904, 4, 5. 
2 Ann. d. l’lnst. Pasteur, 1906, 20, 225. 
3 Ann. d. l’lnst. Pasteur, 1908, 22, 54. 
4 Wien. klin. Wchn., 1908, 21, 311. 
5 Jour. Infect. Dis., 1915, 17, 415; ibid., 1918, 22, 561. 
6 Wien. klin. Wchn., 1904, 265. 7 Hofmeister’s Beitrage, 1901, vol. i. COCTOPRECIPITINS AND TIIERMOPRECIPITINOGENS 
309 
antigens by digesting egg-white and ricin respectively, to the stage when 
they no longer gave the usual protein reactions; Myers1 claims to have 
produced specific precipitins with Witte’s peptone, but Fink2 and others 
have failed to produce precipitins with these proteoses. Kurt Meyer3 
claims to have produced a precipitin for lecithin (lipoid precipitin), but 
these findings have not been confirmed. It is now the consensus of opinion 
that only the protein molecule can induce the production of precipitins, 
and that protein-free lipoids and proteins split to the peptones or further 
are not antigenic. r 
Purified proteins as caseinogen may act as precipitinogens as well as 
mixtures of proteins. Some of the fractions of protein cleavage are like- 
wise antigenic, and Corin4 considers the paraglobulins of serum to be the 
active seroprecipitinogens. 
As previously stated, precipitins are rarely produced by injecting an 
animal with its own serum or with the serum of another animal of the same 
species. As shown by Obermayer and Pick,5 however, rabbit serum treated 
with iodin or nitric acid are so altered that when these are injected into 
rabbits precipitins are produced; these precipitins reacted only with iodized 
protein and nitroprotein respectively, but not only with iodized and nitro- 
proteins of rabbit serum, but of beef serum as well. In other words, the 
precipitins were no longer specific for the species. 
Coctoprecipitins and Thermoprecipitinogens.—The precipitin produced 
by injecting an animal with a heated precipitinogen as heated or boiled 
serum, is called a coctoprecipitin. Precipitinogens are resistant to moderate 
heating and heated extracts of bacteria are commonly used for precipitin 
tests under the term thermo precipitins; this, however, is a misnomer and such 
heated extracts should he designated as thermo precipitinogens. In other words, 
thermoprecipitinogens produce coctoprecipitins, the former being the 
special name applied to the antigen and the latter to the antibody. 
This question of heated antigen possesses a great deal of academic in- 
terest as well as practical importance, the latter in reference to the detec- 
tion of adulteration of sausages with heated dog and cat flesh. 
Heating to coagulation does not destroy the precipitinogenic properties 
of a protein-like serum. Obermeyer and Pick6 found that beef-serum boiled 
for a short time served to induce the production of a precipitin (cocto- 
precipitin) which reacted with both unheated and boiled antigens; on the 
other hand, they found that precipitin induced by injecting rabbits with 
unheated serum reacted only with unheated serum and not with heated 
serum. 
Schmidt,7 who has studied this subject with particular care, found that 
the coctoprecipitin induced by injecting rabbits with serum heated at 70° C. 
for thirty minutes reacted well with a boiled antigen, and in practical tests 
for the detection of boiled meats advises this method for preparing the 
immune serum. He also found that antigens heated at 70° C. for thirty 
to sixty minutes are precipitable by precipitins, but to a lesser extent than 
unheated antigen. Fornet and Muller8 have found the coctoprecipitins 
non-specific, that is, the precipitins induced by injecting rabbits with heated 
1 Centralbl. f. Bakteriol., Abt., 1900, 28, 237, 244. 
2 Jour. Infect. Dis., 1919, 25, 97. 
3 Ztschr. f. Immunitatsf., orig., 1913, 19, 313. 
4 Deutsch. med. Wchn., 1902, 136. 
6 Wien. klin. Wchn., 1906, 12. 
6 Wien. klin. Wchn., 1906, 12. 
7 Ztschr. f. Immunitatsf., orig., 1912, 13, 166. 
8 Ztschr. f. Hygiene, 1910, 66. 310 
PRECIPITINS 
muscle extracts reacted not only with the homologous antigen, but with 
other foreign proteins as well; similar results have been reported by Zinsser 
and Ottenberg1 who injected rabbits with sera boiled for three to five minutes, 
and found that the precipitins acted upon boiled antigens, but were not 
specific. 
Since the precipitins (coctoprecipitins) induced by injecting animals 
with boiled antigens (thermoprecipitinogens) are non-specific it is necessary 
to prepare immune sera for the detection of boiled meats by injecting sera 
or meat extracts (preferably the former) heated at 70° C. for thirty minutes 
(Schmidt’s “70° C. precipitins”), or to rely upon the use of extra power- 
ful precipitins produced by injecting unheated serum (Fornet and Muller). 
Mechanism of Precipitation.—Of the various theories advanced to ex- 
plain the phenomenon of precipitation, none has received so much support 
experimentally as that regarding the reaction a colloidal phenomenon, first 
advanced by Landsteiner,2 and advocated by Bordet, Porges, Neisser, and 
Friedemann, Gengou, and others in explanation of agglutination. 
Both the precipitinogen and precipitin are colloids and closely follow 
the laws governing colloidal reactions. Electrolytes must be present in 
the form of salts and so alter the electric state of colloidal particles that 
their surface tension is decreased, and as a result of this change neighboring 
particles coalesce in such quantities as to produce a visible precipitate. 
Salts are likewise necessary for serum precipitation, and there is a close 
analogy between serum and colloidal precipitation. 
As stated by Krogh,3 however, we are not justified in assuming that 
specific serum precipitation is one solely of colloidal chemistry, but rather 
that colloidal phenomena are to be regarded as preliminary to a real chemical 
process that completes the reaction and gives it specific characters. 
The apparent coexistence of both antigen and antibody in antiserum is 
sometimes encountered, and especially when rabbits are rapidly immunized 
with large doses of serum and bled within a week or ten days after the last 
injection. This phenomenon is shown by the fact that the immune serum 
may be clear but show precipitation when mixed with dilutions of antigen 
(indicating the presence of precipitin) or when‘added to homologous pre- 
cipitating sera (indicating the presence of antigen). This paradoxic phe- 
nomenon of the presence side by side in the same serum of precipitinogen 
and precipitin, but incapable of reacting with each other, has been recorded 
by Linossier and Lemoine,4 Eisenberg,5 Ascoli,6 von Dungern,7 Gay and 
Rusk,8 Zinsser and Young,9 Weil,10 and others. 
Several explanations have been offered. In the first place von Dungern 
has doubted the actual coexistence of antigen and antibody in the same 
serum and believes that the apparent presence of the two side by side is 
due to the fact that complex antigens, as horse-serum, which have been 
used for immunizing rabbits in most experiments, contain several antigenic 
substances stimulating the production of several partial precipitins* In 
sera in which antigen and precipitin are found apparently side by side and 
1 Proc. New York Path. Soc., 1914. 
2 Centralbl. f. Bakteriol., orig., 1906, 41, 108; ibid., 1906, 42, 353. 
3 Jour. Infect. Dis., 1916, 19, 452. 
4 C. R. dela Soc. de Biol., 1902, 54, 85. 
5 Centralbl. f. Bakteriol., 1902, 31, 773. 
6 Munch, med. Wchn., 1902, 39, 1409. 
7 Centralbl. f. Bakteriol., 1903, 34, 355. 
8 Univ. of Calif. Publ. Path., 1912, vol. ii. 
9 Jour. Exper. Med., 1913, 17, 396. 
10 Jour. Immunology, 1916, 1, 19, 35, 47. SPECIFICITY OF PRECIPITINS 
311 
free, he believes that the antigen is of a nature that has no affinity for the 
particular partial precipitin present with it. Zinsser and Young have not 
accepted this view and believe that antigen and specific antibody may co- 
exist in the same serum, precipitation being prevented by protective col- 
loids. Weil, however, working with horse-serum as a complex antigen 
and crystalline egg-albumen as a purified antigen, has adopted the views 
of von Dungern, having found that with the latter antigen partial precipitins 
are not produced and the antisera do not present the paradoxic phenomenon 
of free precipitinogen and precipitin existing side by side. The question, 
however, is still an open one requiring further investigation. In the living 
animal antigen and antibody may be present in the serum, probably in 
loose combination and easily dissociated in the test-tube by the addition 
of fresh antigen whereby precipitin becomes available and precipitation 
results. 
The origin of the precipitate formed during the reaction is of interest. 
When a very potent immune serum is employed, the precipitinogen is so 
highly diluted that it no longer gives any of the chemical reactions for 
proteins, but when the precipitating serum is added it may yield, never- 
theless, a heavy precipitate. The precipitate can, therefore, hardly be 
regarded as due to the slight trace of albumin in the precipitinogen, and, 
furthermore, if the precipitating serum is diluted, the precipitate becomes 
smaller and smaller, and if the dilution is increased, it finally disappears 
altogether. For this reason the precipitate is generally considered as origi- 
nating mainly in the immune serum as indicated especially by the work of 
Welsh and Chapman,1 being the insoluble modification of the previously 
soluble proteins of the precipitin. 
However, when the reaction is maximum, a portion of the precipitate 
is apparently derived from the antigen. Wells2 states that the precipitate 
may contain more nitrogen than corresponds to the entire euglobulin frac- 
tion of the immune serum which carries the precipitin; however, it is al- 
ways less in amount than the total globulin of the immune serum. Further- 
more, as shown by Weil,3 the precipitate is able to sensitize guinea-pigs 
anaphylactically in both an active and passive manner indicating the co- 
existence of both antigen and antibody. The serum precipitinogen has been 
recovered by Weil by treating the precipitate with salt solution and solu- 
tions of sodium carbonate; these procedures, however, did not dissociate 
precipitin. Extraction with trypsin and leukocytes yielded both antigen 
and antibody. 
The precipitate is readily dissolved in weak acids and alkalies, has the 
power of binding complement as shown by Gay,4 and when digested or 
split, produces poisonous substances designated by Friedeberger as ana- 
phylatoxins. 
Specificity of Precipitins.—The high specificity of the precipitins is 
attested to by a very large number of investigations with different protein 
substances. The bacterioprecipitins may react with extracts of closely 
related bacterial species in exactly the same manner as the agglutinins, 
but in practical work the influence of these group precipitins may be avoided 
by using varying amounts of antiserum in a manner analogous to the agglu- 
tination reaction. Furthermore, the group precipitins are removable from 
immune serum by absorption methods as are the group agglutinins. 
1 Proc. Roy. Soc., 1908, 80, 161; Ztschr. f. Immunitatsf., 1911, 9, 517. 
2 Chemical Pathology, 4th ed., 186. 
3 Jour. Immunology, 1916, 1, 35. 
4 Univ. of Calif. Publ. Path., 1911, 2, 1. 312 
PRECIPITINS 
An antiserum obtained by immunizing a rabbit with human serum may 
react not only with human serum, but likewise with the sera or blood ex- 
tracts of the higher apes. This is shown by the extensive investigations of 
Nuttall1 upon this subject, the reactions becoming weaker as the species 
examined is farther removed from man zoologically, when the sera were 
tested with five different antihuman sera; 
34 Specimens of human blood (4 races) gave 100 per cent, precipitate. 
8 Simudae (3 species) gave 100 per cent, precipitate. 
36 Cercopithecidae (26 species) gave 92 per cent, precipitate. 
13 Cebidae (9 species) gave 73 per cent, precipitate. 
4 Hapalidae (3 species) gave 50 per cent, precipitate. 
2 Lemuroidea (2 species) gave 0 precipitate. 
These investigations by Nuttall, covering 16,000 tests, have proved 
very valuable for the study of blood relationships not only of the primates, 
but of relationships in the families of other carnivora, as the dog, cat, hyena, 
etc.; likewise among the insectivora, rodentia, ungulata, reptilia, aves, etc. 
Among the primates antihuman serum is most likely to react with the 
blood of the chimpanzee, the different members of the Simudas reacting 
as follows when their sera were tested with antihuman serum: 
Chimpanzee gave 100 per cent, precipitate. 
Gorilla gave 64 per cent, precipitate. 
Ourang-outang gave 42 per cent, precipitate. 
Antihuman sera may also react very slightly with 2 to 24 per cent, of 
the sera of other carnivora, rodents, and the lowTer animals in general when 
the tests are made with low dilutions of antigen, due to the presence of 
natural precipitins or group precipitins; as shown by Uhlenhuth and Wei- 
danz,2 however, these non-specific reactions may be avoided by using higher 
dilutions of antigen, experience having shown that with 1 : 1000 dilutions 
of antigen the reactions are absolutely specific. This is well shown in the 
following table by Hektoen3: 
Specificity of Precipitin in Serum of Rabbit Injected with Human Blood 
Blood. 
H’ghest dilution giving precipitate 
with antihuman serum after 
Fish 
twenty minutes at room 
temperature. 
1 :10 
Chicken 
1 : 10 
Rabbit 
0 
Guinea-pig 
1 : 10 
Rat 
1 : 10 
Cat 
1 : 10 
Gog 
1 : 10 
Swine 
1 : 10 
Sheep 
1 : 10 
Beef 
1 : 10 
Horse 
1 : 10 
Goat 
1 : 10 
Monkey (Macacus rhesus) 
1 : 100 
Human 
1 :5000 
Recently attention has been focused again on these cross reactions and 
particularly in connection with medicolegal work. Years ago Nuttall 
pointed out that precipitin reactions with the sera of animals which had 
1 Blood Immunity and Blood Relationship, Cambridge University Press, 1904, 165. 
2 Praktische Anleitung zur Ausfuhrung des biologischen Ehveissdifferenzierungsver- 
fahrens, Fischer, Jena, 1909. 
3 Jour. Amer. Med. Assoc., 1918, 70, 1275; Jour. Infect. Dis., 1922, 31, 72. SPECIFICITY OF PRECIPITINS 
313 
received many injections of a foreign protein were not strictly specific for 
that protein, as the antibodies produced reacted with the proteins of other 
animals of the same group or of the same class. Nuttall called this, for 
instance, the “mammalian reaction.” In spite of this, however, the titra- 
tion of the precipitin has shown sufficient quantitative differences to allow 
specific distinctions to appear. Lately, since Forssmann’s discovery of 
heterophile antigens (discussed on page 159), more doubt has been thrown 
upon the validity of the precipitin reactions. 
Thus Friedberger and Jarre1 found that comparative tests with the sera 
of rabbits immunized to the blood proteins of various animals showed group 
precipitin reactions which cannot be explained by a biologic relationship 
of these animals. A rabbit immunized to horse-serum developed pre- 
cipitins also for the serum proteins of ox, deer, goat, pig, sheep, and man. 
An antidog-serum from a rabbit precipitated horse and donkey proteins. 
These antisera were next absorbed with the red corpuscles and other 
organ cells of guinea-pig and sheep. The sera, first inactivated, were mixed 
with a suspension of cells which had been heated to 100° C. After fifteen 
minutes of incubation of the serum and heated cells at 37° C., all the heterol- 
ogous precipitins were absorbed, and were removed with the sediment of 
cells after centrifugalization. An antihorse-serum from a rabbit, which 
before treatment with sheep corpuscles precipitated ox-, sheep-, goat-, and pig- 
serum, after absorption with sheep cells contained only precipitin for horse- 
serum. Absorption with a heterologous antigen, therefore, rendered the 
serum strictly specific for its homologous antigen, as all the group antibodies 
were removed. 
This work, however, has not been confirmed by Manteufel and Beger.2 
They found that 63 per cent, of all precipitating sera, active against human, 
beef, and other proteins, were absolutely specific, that 24 per cent.. gave 
some precipitate with heterologous antigens diluted 1 : 100, and that 13 
per cent, precipitated heterologous antigens diluted 1 : 1000. All these 
sera, however, were practically useful, as their titers for their homologous 
antigens were high, 1 : 10,000 or 1 : 20,000. The longer the course of im- 
munization of an animal, the less specific the serum produced. 
They found also that the formation of heterologous precipitin is not the 
same process as the formation of the heterogenetic hemolytic amboceptor 
which occurs when certain animals are injected with extracts from the 
organs of other animals as described by Forssmann. 
Attempts were made to render sera strictly specific by absorbing from 
them the heterogenetic antibody with sheep corpuscles, as reported by 
Friedberger and his associates. These experiments, as well as other experi- 
ments on the absorption of precipitins with solutions of proteins, were 
unsuccessful. 
These experiments emphasize, therefore, the great importance of group 
precipitins and particularly in medicolegal work; also the necessity for using 
highly potent and properly titrated sera and an extremely careful technic in 
order to secure specific and reliable results. 
Precipitins do not permit of distinguishing different albumins of the 
same animal. For example, anti-ox precipitin prepared by injecting a rabbit 
with ox serum will react not only with ox-serum, but likewise with extracts 
of ox muscle, ox heart, ox kidney, ox pleural and pericardial fluids, etc. 
It would appear that every species of animal possesses throughout its tissues 
a common protein, but peculiar to its species. However, many organs un- 
1 Ztschr. f. Immunitatsf., 1920, 30, 351. 
2 Ztschr. f. Immunitatsf., 1921, 33, 348. 314 
PRECIPITINS 
doubtedly possess peculiar proteins not present in other organs because by 
carefully freeing the organs from all blood and using extracts for immuniza- 
tion, it is possible to secure antisera that will yield precipitates best with 
the particular extract employed and a slight or no precipitate with extracts 
of other organs. 
The most striking example of organ specificity is observed with the 
proteins of the lens of the eye. As shown by Uhlenhuth and since amply 
confirmed by Hektoen1 and others, immunization of rabbits with this sub- 
stance results in the production of a precipitin that does not react with the 
serum or any protein of the animal from which the lens was taken, but does 
react with the lens protein of the same animal and of all other animals in 
general. Hektoen2 has recently described specific precipitins for erythro- 
cytes—erythroprecipitins. Hektoen and Meune3 have also prepared specific 
leukoprecipitins for the leukocytes of the dog, guinea-pig, and human being. 
In other words, leukocytes appear to contain specific precipitinogenic sub- 
stances not found in serum, platelets, or erythrocytes. 
In medicolegal work, therefore, a diagnosis of “human blood stain” 
cannot be made without chemical evidence to prove that the stain actually 
consists of blood. For example, an extract of a stain of human albuminous 
urine may react with antihuman serum; or a stain of semen, blister fluid, 
pleural exudate, and according to Nuttall,4 even to a slight extent with nasal 
and lacrimal secretions (these may be avoided by using high dilutions of 
antigen or antiserum). 
Besides this animal specificity, precipitin reactions also demonstrate 
the “constitutional specificity” of proteins. If, instead of using a pure 
animal or plant albumin for immunization, variously altered albumins 
are used (heated albumins, acid albumin, formaldehyd albumin, and the 
like), the organism reacts by producing antibodies of a characteristic nature, 
differing from those developed after inoculation with pure albumin. For 
example, if a rabbit is immunized with normal horse-serum, the resulting 
immune serum will produce a precipitate when added to pure horse-serum, 
but not when added to horse-serum that has been heated. On the other 
hand, if a rabbit is inoculated with horse-serum that has been diluted and 
boiled for a short time, the resulting immune serum will react not only 
with normal horse-serum but also with heated serum and a group of its 
decomposition products with which the normal immune serum ordinarily 
never produces a precipitate. 
This observation is of practical importance in detecting meat substitu- 
tion by precipitin reactions. In order to render the detection difficult, the 
meat is commonly boiled; with the aid of precipitins produced by immuni- 
zation with heated proteins, this fraud is more easily detected than if a 
normal immune serum were used. 
Obermeyer and Pick have demonstrated that while animal specificity 
is not destroyed when the albumins are modified by heat, tryptic digestion, 
or oxidation, their specificity is lost when an iodin, nitro- or diazo-group 
is inserted into the protein molecule. Immunization with such transformed 
proteins, e. g., xanthoprotein, can produce a precipitating serum that will 
react with every xanthoprotein, even that of different animals. These in- 
vestigators conclude that species specificity is probably dependent upon a 
certain aromatic group of the protein molecule. 
1 Jour. Amer. Med. Assoc., 1921, 77, 32. 
2 Jour. Infect. Dis., 1922, 31, 32. 
3 Jour. Amer. Med. Assoc., 1922, 79, 1328. 
4 Loc. cit., 105. PRACTICAL APPLICATIONS 
315 
Role of Precipitins in Immunity.—Precipitins as they occur in the blood 
are probably not directly destructive for their antigens as are the anti- 
toxins and cytotoxins. Their role in resistance to infection and in the 
mechanism of recovery appear to be quite similar to the agglutinins. The 
latter, for example, do not kill or greatly injure the cells upon which they 
act, but yet agglutination greatly aids bacteriolysis or cytolysis in general, 
although these phenomena may occur without visible agglutination. The 
antigen may be recovered from a precipitate unaltered, as judged by its 
power of actively sensitizing guinea-pigs anaphylactically, just as living 
bacilli may be recovered from an agglutinated mass or found fully viable 
after being acted upon by specific opsonin. In other words, the precipitins 
appear to bear the same relation to dissolved albumins as agglutinins and 
opsonins do to formed elements, preparing or sensitizing the dissolved anti- 
gen for final destruction or albuminolysis. This subject is discussed more 
fully in the succeeding paragraph. 
Relation of Precipitins to Albuminolysins; Complement Fixation by 
Precipitates.—In 1902 Gengou1 showed that precipitates had the power of 
fixing or absorbing complement and concluded that immune sera prepared 
by injecting rabbits with milk, serum, egg-white, etc., contained not only 
precipitins but amboceptors or sensitizers as well, the latter being carried 
down in the precipitate and yielding the positive complement-fixation 
reactions. Gay,2 however, who has studied this problem with particular 
care, came to the conclusion that complement was fixed by a process of 
absorption by the precipitate alone independent of the possible coexistence 
of sensitizers or amboceptors. The work of Zinsser3 has yielded a great 
deal more information upon this interesting and important subject. His 
experiments indicate that in immunization with formed antigens as bacteria, 
the immune serum contains bacterioprecipitin, and bacterial sensitizer or 
amboceptor, but that the two antibodies are identical, the precipitin being 
a sensitizer for dissolved antigen (albumins), and the amboceptor a sensi- 
tizer for the formed antigen (cells). When animals are immunized with 
dissolved antigen (albumins) as serum or milk, precipitins alone are pro- 
duced, that is, sensitizers for the dissolved albumins. According to this 
view, therefore, the precipitins are to be regarded as protein sensitizers 
or amboceptors by which foreign proteins are rendered susceptible to the 
proteolytic action of alexin or complement. If this is correct, and I believe 
we may logically accept this view, complement fixation by precipitates is 
not one merely of mechanical absorption, but the fixation of complement 
by specific sensitizer which is the precipitin itself. Visible precipitation in 
such reactions is not necessary, and when it occurs is merely secondary 
because of the colloidal nature of the antigen and antibody and favorable 
quantitative and environmental conditions. 
PRACTICAL APPLICATIONS 
Phytoprecipitins 
Bacterial Precipitinogens in the Urine in Pneumonia.—Dochez and 
Avery4 have shown that patients suffering from lobar pneumonia may ex- 
crete in their urine at some stage of the disease a soluble substance of pneu- 
mococcus origin. This substance gives a specific precipitin reaction with 
antipneumococcus serum corresponding in type to the organism with which 
1 Ann. d. l’lnst. Pasteur, 1902, 16, 734. 
2Univ. Calif. Publ. Path., 1911, 2, 1; Jour. Immunology, 1916, 1, 83. 
3 Jour. Exper. Med., 1912, 15, 529; ibid., 1913, 18, 219. 
4 Proc. Soc. Exper. Biol, and Med., 1916—17, 14, 126. 316 
PRECIPITINS 
the individual is infected. A study of 111 cases of pneumonia has shown 
that in 65 per cent, of those due to pneumococcus Types I, II, and III, 
this precipitinogen is present in the urine, and can be detected by means 
of the appropriate antipneumococcic serum. The precipitinogen may 
appear in the urine as early as twelve hours after the initial chill, or it may 
appear for the first time at a later stage of the disease. It is a rule to find 
this substance when pneumococcus septicemia exists, being indicative of 
severe intoxication and an unfavorable prognosis. Quigley1 has found this 
precipitin reaction in 67 of a series of 82 cases of Types I, II, and III pneu- 
monia. The technic (described later) is simple, rapid, and accurate; while 
it should not supplant the mouse agglutination test for type diagnosis, 
I have found it of value as a confirmatory test. 
Bacterial Precipitinogens in Inflammatory Exudates.—Probably the 
earliest application of the precipitin test for diagnosis with inflammatory 
exudates was by Vincent and Bellot2 in the diagnosis of meningococcus 
meningitis. The spinal fluid is tested for meningococcus precipitinogen by 
overlaying with polyvalent antimeningococcus serum. Letulle and Legane,3 
Bruynoghe,4 and Robinson5 have found this test of some value. Worster- 
Drought and Kennedy6 observed 12 fluids from 28 cases to react positively. 
The test may prove of diagnostic value when confirmatory evidence is 
required. 
Schurman7 and Lacy and Hartman8 have found pneumococcus pre- 
cipitinogen in the cerebrospinal fluid in pneumococcus meningitis and have 
advocated the precipitin test as an aid in diagnosis. The technic is de- 
scribed later. 
Schurman, and later Floyd2 have also tested the pus from cases of pneu- 
monic pleuritis and found a large percentage to contain specific precipit- 
inogen corresponding to the type of pneumococcus and detectable by a 
precipitin test employing corresponding antipneumococcus sera. 
Blake10 has found a precipitinogen in the peritoneal exudates of mice 
inoculated intraperitoneally with pneumonic sputum for the purpose of 
differentiating types of pneumococci by the agglutination test. The clear 
peritoneal fluid secured by centrifuging the exudate is tested with anti- 
pneumococcus sera of Types I, II, and III and the test has proved a valu- 
able adjunct to the agglutination test. 
Bacterial Precipitins in Anthrax, Bubonic Plague, Glanders, Tubercu- 
losis, Gonorrhea, Echinococcus, Syphilis, and Other Diseases.—Wladimi- 
roff11 was apparently first to apply the precipitin test for the diagnosis of a 
bacterial infection, namely, glanders, finding a precipitin in the serum of a 
glandered horse for a filtrate of a forty-six-day-old culture of Bacillus mallei 
in glycerin-veal broth. Since then many investigators have described pre- 
cipitin reactions in other bacterial diseases, and in some instances have 
advocated the test for diagnostic purposes, notably Ascoli12 and Roncaglio13 
1 Jour. Infect. Dis., 1918, 23, 217. 
2 Bull, et mem. Soc. med. d. h6p., 1909, 27, 952. 
3 Compt. rend. Soc. de Biol., 1909, lxvi, 758. 
4 Centralbl. f. Bakteriol., 1911, 7, 38. 
5 New York Medical Journal, 1919, cix, 464. 
6 Cerebrospinal Fever, 1919, A. & C. Black, London, 272. 
7 Med. Klinik, 1915, 11, 741. 
8 Jour. Immunology, 1918, 3, 43. 
9 Jour. Immunology, 1920, 5, 321. 
10 Jour. Exper. Med., 1917, 26, 67. 
11 St. Petersburg Med. Wchn., 1900. 
12 Centralbl. f. Bakteriol., 1911, 58, 63. 
13 Ztschr. f. Immunitatsf., 1912, 12, 380. ZOOPRECIPITINS 
317 
for anthrax; Hecht1 for symptomatic anthrax; Isabolinsky and Patze- 
witsch2 for swine erysipelas; Muller3 for glanders; Berlin,4 Finzi,5 and Piras6 
for bubonic plague; Watabiki7 for gonorrhea; Baldwin,8 Porter,9 Fuchs,10 
Ferro,11 Smeeton,12 and others for tuberculosis; Welsh, Chapman, and Storey13 
have employed the test for the diagnosis of echinococcus disease, reporting the 
positive reaction as possessing diagnostic value; Coroma14 has employed it 
in leischmaniasis. But these tests for diagnostic purposes are not nearly 
as satisfactory as the simpler agglutination test, and quantitative reactions 
are not elicited as readily and decisively as with the agglutination and 
complement-fixation tes1,s, which have largely superseded the precipitin 
test. A possible exception is the precipitin reaction in the diagnosis of 
glanders by Ascoli’s method. Reinhardt15 has recently renewed interest in 
this subject and a large number of investigators have reported favorably 
upon the test as a diagnostic reaction. 
In syphilis Fornet and Schereschewsky16 have shown that in mixtures of 
syphilitic sera and sera from long-standing cases of syphilis, as paretics and 
tabetics, a precipitate may form. The serum to be tested is diluted five to 
ten times and placed in a small test-tube; undiluted paretic serum is over- 
laid and as it falls and mixes with the diluted serum sometimes produces 
a precipitate in two hours at room temperature. The reaction occurs in- 
frequently and has no diagnostic value. 
Bacterial Precipitins for the Identification and Differentiation of Bacteria. 
—Noble has employed the precipitin test with rabbit immune sera for a 
study of the biologic relationship of members of the typhoid-colon group of 
bacilli and concluded in a comparative study with the agglutination reac- 
tion that the former yielded more delicate results and greater differentia- 
tion. The concensus of opinion, however, does not support this conclusion; 
personally, I have found the agglutination and especially the complement- 
fixation reactions far more sensitive and decisive in their results. 
Yeast Precipitins.—Schiitze has immunized rabbits with different 
yeasts and obtained precipitins as previously stated, but no one to my 
knowledge has employed the precipitin test for the practical diagnosis of 
yeast and fungous infections. 
Precipitins in the Examination and Identification of Bloods, Blood-stains, 
and Sera (the Hematoprecipitins).—Since the simultaneous discovery of 
the medicolegal use of the precipitin test by Uhlenhuth17 and Wassermann 
and Schiitze18 a very large number of investigations have proved its prac- 
tical value for the identification of blood-stains. The voluminous literature 
has been summarized by Nuttall,19 Uhlenhuth and Weidanz20 together with 
reports of their own extensive investigations. In America Hektoen21 has 
ZOOPRECIPITINS 
1 Centralbl. f. Bakteriol., 1913, 67, 371. 
2 Centralbl. f. Bakteriol., 1913, 67, 284. 
3 Ztschr. f. Immunitatsf., 1910, 8, 626. 
4 Centralbl. f. Bakteriol., 1915, 75, 467. 
5 Centralbl. f. Bakteriol., 1913, 68, 556. 
6 Centralbl. f. Bakteriol., 1913, 69, 69. 
7 Jour. Infect. Dis., 1918, 22, 115. 
8 Jour. Med. Research, 1904, 12, 235, 243. 
9 Jour. Infect. Dis., 1910, 7, 87. 
10 Centralbl. f. Bakteriol., 1918, 81, 178. 
11 Riforma med., 1920, 36, 907. 
12 Amer. Rev. of Tuberculosis, 1922, 6, 588. 
13 Australian Med. Gaz., December, 1908. 
14 Ztschr. f. Immunitatsf., 1914, 20, 174. 
15 Monatsch. f. Prak. Tierhl., 1920, 31, 268. 
16 Berl. klin. Wchn., 1908, 18, 874. 
17Deutsch. med. Wchn., 1901, 27, 82. 
18 Berl. klin. Wchn., 1901, 38, 187. 
19 Blood Immunity and Blood Relationship, Cambridge University Press, 1904. 
20 Praktische Anleitung zur Ausfuhrung des biologischen Enveissdifferenzierungsver- 
fahrens mit besonderer Berucksichtigung der forensischen Blut und Fleischuntersuchung 
sowie der Gewinnung prazipitierender Sera, Gustav Fischer, Jena, 1909. 
21 Jour. Amer. Med. Assoc., 1918, 70, 1273; Jour. Infect. Dis., 1922, 31, 32. 318 
PRECIPITINS 
made numerous and valuable contributions to our knowledge of precipitins, 
especially methods for their production. He has also recently shown that 
precipitins prepared by immunization of rabbits with watery extracts of 
corpuscles are specific for the erythrocytic constituents, the hemoglobin 
acting as a species specific precipitinogen. 
In Germany the test has become popular for the medicolegal identifica- 
tion of blood-stains, being very successful in the hands of Uhlenhuth. Stains 
on clothing, wood, metal implements, on plaster walls, in earth, on straw, 
and hay, etc., are readily detected and identified. 
Contrary to popular opinion the test is not as simple as believed and 
in medicolegal work especially, requires potent antisera, experience, and a 
good working knowledge of the technic in order to secure reliable results. 
Group precipitin reactions must be scrupulously avoided and numerous 
controls included. The test, however, is more apt to err on the negative 
than on the positive side, a striking example being recorded by Hunt.1 
Personally I have found the complement-fixation test far more sensitive 
and satisfactory and have never consented to testify in medicolegal cases 
before conducting these tests in addition to the precipitin tests. 
I have already discussed the specificity of the reaction earlier in this 
chapter; with proper attention to technical details the reaction is specific. 
Only the question of monkey blood can enter under these conditions, but 
in the United States this possibility can usually be excluded. 
It is highly important to remember, however, that stains of other fluids 
containing human albumins cannot be differentiated from blood-stains by 
precipitin tests; microscopic or chemical tests must first establish that a 
stain is blood. With garments partly washed this is sometimes difficult. 
Furthermore, it is not possible to distinguish between different human races 
nor between individuals by the precipitin test. Therefore, in medicolegal 
cases involving the charge of murder these tests cannot determine whether 
a stain is due to the blood of the victim or the accused. 
Precipitins in the Examination and Identification of Seminal Stains 
(Spermatoprecipitins).—In medicolegal cases involving the charge of rape 
these tests are occasionally of value. Immune sera are best prepared by 
injecting rabbits with human semen secured by prostatic and vesicular 
massage for therapeutic purposes. 
Apparently the first specific antisemen precipitins were produced by 
Farnum2 at the suggestion of Hektoen. Strube3 also produced them, but 
found these precipitins acted upon serum proteins as well, and he was not 
able to remove these by adsorption. Pfeiffer4 succeeded in producing a 
specific antisemen precipitin serum for bull spermatozoa by removing all 
precipitins for serum and organs other than the testicle by adsorption. 
Hektoen5 has recently confirmed this work and prepared specific antisemen 
sera, and believes that this precipitin reaction may prove of value in de- 
termining the nature of spots suspected to be of seminal nature. The technic 
of the tests is described later in this chapter. 
Hektoen immunizes rabbits by giving four or five intramuscular injec- 
tions of human semen at intervals of three or four days, beginning with 2 c.c. 
and increasing the quantity by 2 c.c. each succeeding injection. The animals 
are bled from six to eight days after the last injection. 
1 Boston Med. and Surg. Jour., 1917, clxxvi, 48. 
2 Jour. Amer. Med. Assoc., 1901, 37, 1721. 
3 Deutsch. med. Wchn., 1902, 28, 425. 
4 Wien. med. Wchn., 1905, 18, 637. 
5 Jour. Amer. Med. Assoc., 1922, 78, 704. ZO0PRECIPITINS 
319 
Serum precipitins are removed from the rabbit immune serum by mix- 
ing equal parts of serum and a l : 200 dilution of human serum in saline. 
This mixture is left at room temperature for one hour in a refrigerator 
over night and thoroughly centrifuged. The supernatant fluid is employed 
for precipitin tests. 
Precipitins in the Examination and Identification of Meats (Musculo- 
precipitins).—Uhlenhuth1 has shown that the precipitin test may be utilized 
for the identification of different meats and especially for the detection of 
meat adulteration, as the presence of dog, cat, and horse flesh in sausages. 
Von Rigler,2 Schmidt,3 and numerous others have confirmed these results; 
Gay4 has been able to identify the heart of a deer by this method and differ- 
entiate it from the heart of a calf, being the first to apply the musculo- 
precipitin test in a medicolegal case with conviction for an infringement of 
the game laws of Massachusetts. 
Immune sera are prepared by injecting rabbits with extracts of flesh 
or with the corresponding sera. 
The precipitins are not specific for the flesh extracts alone, that is, they 
will react almost as well with extracts of the different organs of the same 
animal or with the serum. In other words, when the meat for examination 
is minced, the precipitin test cannot differentiate between the minced mus- 
cle of the heart—say of the dog—but can identify the presence of the dog 
muscle. 
Smoking of meat does not appear to interfere with the test; neither 
does salting, but cooking may. For the detection of cooked meat it is neces- 
sary to employ immune rabbit sera prepared by injecting sera diluted with 
saline and heated at 70° C. for thirty minutes or briefly boiled; very powerful 
antisera prepared by injecting unheated serum may suffice, this subject 
having been previously discussed under the coctoprecipitins. 
The musculoprecipitins are apt to show the same species of relationship 
as the blood precipitins; for example, it is not possible to differentiate among 
the meats of the horse, mule, and donkey. 
In practical work I have found the complement-fixation test much 
superior to the precipitin test for the identification and differentiation of 
meats; I have not been particularly successful with the musculoprecipitin 
reaction and prefer the former for the purpose of securing sharper and more 
conclusive results. 
Precipitins in the Examination and Identification of Milks (Lactoprecipi- 
tins; Lactosera).—The investigations of Bordet,5 Neoro,6 Wassermann and 
Schutze,7 Hamburger,8 Bauereisen,9 Valerio and Bornand,10 and numerous 
other investigators have shown that precipitins obtained by injecting cow’s 
milk into rabbits react with cow’s milk, but not with human milk, and 
vice versa. Amberg11 has shown that the precipitins secured by injecting 
whole milk are identical with those induced by the injection of casein. 
The lactoprecipitins show the same species specificity as the correspond- 
1 Deutsch. med. Wchn., 1901, 27, 780; Ztschr. f. Immunitatsf., Ref., 1909-10, 1, 525. 
2 Oest. Chem. Zt., 1902, 5, 97. 
3 Ztschr. f. Immunitatsf., 1912, 13, 166. 
4 Jour. Med. Res., 1908, 19, 219. 
6 Ann. d. l’lnst. Pasteur, 1899, 13, 225. 
6 Wien. klin. Wchn., 1901, 1073. 
7 Deutsch. med. Wchn., 1900, 178. 
8 Wien. klin. Wchn., 1901, 1202. 
9 Ztschr. f. Immunitatsf., 1911, 10, 306. 
10 Ztschr. f. Immunitatsf., 1912, 14, 32. 
11 Jour. Med. Research, 1904, 12, 341. 320 
PRECIPITINS 
ing seroprecipitins; for example, the precipitin for cow’s milk will not pre- 
cipitate human milk and vice versa; but cow precipitin reacts wdth the milk 
of the goat just as is true of the sera precipitins for these two animals. 
Human seroprecipitin is likely to precipitate human milk, but lacto- 
sera are better prepared by immunizing rabbits with milk. Reactions are 
commonly observed with boiled milk (thermoprecipitinogen). 
In my experience the complement-fixation test is very much superior 
to the precipitin test for the identification and differentiation of milks; 
the latter requires a large amount of immune serum for each test and the 
results have never been in my experience as clear cut and decisive as the 
complement-fixation reactions. 
Precipitins for the Examination and Identification of Bones (Osteo- 
precipitins), Eggs (Ovoprecipitins), and Other Substances.—Beumer1 and 
Schtitze2 have successfully identified bits of bone too small for recognition 
by inspection. Seroprecipitins were used with extracts of the bones, pre- 
pared as described in the section on Technic. Fine charred human bones 
have been identified, but I do not know the value of this method for the 
identification of decomposed bones; presumably the bone examined must 
contain some organic matter, as the reaction does not occur with the in- 
organic constituents. 
Uhlenhuth3 and others have prepared antisera for egg-albumen by in- 
jecting rabbits with the wrhite portions of hen eggs. These have been em- 
ployed for the detection of egg-white in prepared foods. 
Hen ovoprecipitin is likely to react with the albumins of the eggs of the 
pigeon and guinea-hen; in other words, their specificity is closely similar to 
the sero- or hematoprecipitins. 
By means of these precipitins Grulee and Bonar4 have detected egg- 
albumen in the urine of infants; negative results were observed in a study 
of feces.5 Hektoen, Fantus, and Portis6 in a study of the precipitin reac- 
tion with feces found positive reactions with antihuman sera and extracts 
of the feces of healthy persons, indicating the presence of human proteins 
probably derived from the blood- and epithelial cells. Tests with anti- 
beef, antichicken, antisheep, and other sera with extracts of feces of healthy 
men on unrestricted, full meat diet were generally negative, indicating that 
in health foreign proteins taken in food do not reach the feces as such. 
Berghausen,7 Hektoen, and Neymann8 have also employed the precipitin 
test for a study of the proteins in cerebrospinal fluid. The former reached 
the conclusion that the test was too delicate for clinical purposes; the latter 
believe that the test is adapted for routine use. Antisera wTere prepared by 
injecting animals with spinal fluid or the globulin and albumin fractions. 
With the fluids from paretics reactions were observed with dilutions 1 : 16 
to 1 : 512, and higher similar, results being observed wflth fluids from other 
cases of acute meningitis. Normal fluids did not react in dilutions higher 
than 1 : 3.7. Hektoen9 has also recently described a precipitin reaction for 
Bence-Jones albumin in urine. 
It may be stated that specific organic reactions have been secured by 
1 Quoted by Nuttall, Blood Immunity, etc., 406. 
2 Deutsch. med. Wchn., 1903, 29, 62. 
3 Deutsch! med. Wchn., 1900, 26, 734. 
4 Amer. Jour. Dis. Child., 1921, 21, 89. 
5 Amer. Jour. Dis. Child., 1920, 20, IS. 
6 Jour. Infect. Dis., 1919, 24, 482. 
7 Interstate Med. Jour., 1913, 20, 38. 
8 Jour. Amer. Med. Assoc., 1920, 75, 1332. 
9 Jour. Amer. Med. Assoc., 1921, 77, 929. Fig. 109.—Hemin Crystals. 
Prepared after the method described in the text. From a stain (over two months old) of sheep blood 
on a towel. TECHNIC OF PRECIPITIN TESTS 
321 
various investigators by prolonged immunization of rabbits with certain 
organ extracts. Thus it is possible to differentiate between the liver and 
kidney of the same animals; such tests have, however, but limited practical 
value. Maragliano attempted to apply this test of organic specificity to 
the serodiagnosis of malignant tumors by preparing immune serums by the 
injection of tumor juices, securing a serum that yielded a precipitate with 
the albumins of a cancerous tumor. The test is not absolutely specific, 
and its practical value in diagnosis requires confirmation. 
Freund and Kaminer have described a precipitin reaction in cancer 
with an extract of cancer tissue, but the practical value of the test has not 
been established. 
TECHNIC OF PRECIPITIN TESTS 
DIFFERENTIATION OF HUMAN AND ANIMAL BLOOD-FORENSIC BLOOD 
TEST 
Microscopic and Chemical Tests.—Unless a stain is definitely known 
to be a blood-stain it is first necessary to establish its identity by making micro- 
scopic and chemical tests before proceeding with the precipitin reactions. For 
example, old stains upon clothing may be due to substances other than blood, 
such as coffee and fruit juices; or, more importantly, they may be stains of 
some other human albuminous fluid which cannot be differentiated from 
old blood-stains. Blood-stains upon clothing, metal, wood, or glass may 
be used for making these reactions, and their source determined. 
To identify the stain as one of blood a portion may be taken into solu- 
tion in distilled -water, rendered slightly acid with dilute acetic acid, filtered 
until clear, and examined spectroscopically. Or the Teichmann hemin 
crystal test may be applied to the stain by transferring to a clean slide a 
small amount of material scraped from the stain; add a few small crystals 
of sodium chlorid, crush the crystals, and mix the powder with the dry 
material. Place a clean cover-glass over the stained material and run a 
small amount of glacial acetic acid under the cover-glass. Heat the prepara- 
tion to just about the boiling-point for a minute, replenishing the acid as 
may be necessary. The fluid turns brown. The specimen is allowed to 
cool a few minutes, and is then examined microscopically for the presence 
of brown rhombic crystals of hemin (Fig. 109). It may be necessary to 
reheat the specimen several times before the crystals are obtained. With 
stains in cloth and particularly those partially removed by washing, other 
chemical methods for detection, as the guaiac, benzidin, or Fiirth1 leuko- 
malachite-green tests must be employed. 
Kastle2 has found solutions of phenolphthalein in alkali containing the 
quantity of hydrogen peroxid required for the oxidation of the phenol- 
phthalein very satisfactory for the detection of blood. The reagent is pre- 
pared by dissolving 0.160 gm. of phenolphthalein in 105 c.c. of N/10 sodium 
hydroxid (prepared of pure product and accurately titrated), and diluting 
with redistilled water to nearly 500 c.c. To this solution is added 0.5 c.c. of 
M/1 hydrogen peroxid (0.5 c.c. of the 3 per cent, commercial product being 
sufficiently accurate), and the solution made up to exactly 500 c.c. with re- 
distilled water. This solution keeps in glass-stoppered bottles in a dark place 
and may show only a faint pink color, which does not interfere with its use. 
In conducting the tests 1 c.c. of the solution of substance containing 
blood is mixed with 2 c.c. of the reagent; a control is put up with 1 c.c. 
1 Fiirth-Smith, Physiological and Pathological Chemistry of Metabolism, 1916, Lippin- 
cott & Co., 546. 
2Hygienic Lab. Bulletin, 1909, No. 51. 322 
PRECIPITINS 
of distilled water and 2 c.c. of the reagent. After five to fifteen minutes 
at room temperature a deep pink color indicates a positive reaction. 
With this test Kastle was able to detect 1 part of blood to 8,000,000 
parts of water; of course, the test is not as sensitive as this in the presence 
of urine, feces, gastric contents, pus, etc., but still possesses a high degree of 
sensitiveness and is worthy of trial for the detection of occult blood. 
Having shown that a given stain is actually a blood-stain the source 
of the blood may be determined as the result of the precipitin reaction, 
which consists in extracting the stain in normal salt solution and mixing 
with antiserums prepared by immunizing rabbits with human and various 
animal serums. Since the antiserums are known, a precipitate with any 
one of the extracts indicates that the blood in the stain was derived from 
the same species of animal. The reaction is based upon the principle of 
the specificity of antigen and its antibody. Here the antibody is known, 
and is used in the test to detect the antigen. 
As mentioned elsewhere, because of the presence of group precipitins 
the reaction is fraught with certain technical difficulties of importance, 
especially in medicolegal cases. In most instances it may suffice to show 
that a stain is of human blood as will be indicated by a strong reaction 
with human blood, and negative reactions with the bloods of lower animals. 
If the reaction is negative with antihuman serum the antiserums of the 
domestic animals, such as that of the dog, cat, hog, ox, horse, etc., are tried. 
Although the blood of the higher apes, and even of the lower, orders of 
monkeys, may react slightly with human blood, this factor may be deter- 
mined by observing a proper technic of dilution, or the possibility of a given 
stain being one of monkey blood definitely ruled out. 
Preparation of the Extract of Stain (the Precipitinogen).—If the stain is 
on clothing or paper, a portion should be carefully removed and torn into 
shreds with forceps and scissors, and not with the fingers, and placed in 
normal salt solution. If the stain is upon wood, glass, or metal, the staining 
substance should be carefully scraped off with a clean, and preferably, 
sterile knife and placed in the salt solution. As a further control on the 
technic an unstained portion of the clothing should be extracted in the same 
manner in order to show that the latter alone does not give the reaction. 
The mixtures may be gently shaken, and should be stood aside for from 
two to twenty-four hours in a cold place to prevent bacterial multiplication, 
depending upon the rapidity of extraction. If the stain has been washed 
large portions of fabric must be extracted. Sometimes it is advisable to 
tease portions into single threads which may show incrustations, and espe- 
cially if the cloth is of a heavier sort. 
As a general rule, stains on fabrics and paper offer no difficulties, but 
small stains scraped from wood, metal, and leather surfaces may come 
away in small flinty flakes that dissolve very slowly in salt solution. With 
these it is sometimes advisable to extract with sterile distilled water diluting 
the final extract with an equal part of 1.8 per cent, saline to restore iso- 
tonicity. In the presence of insoluble bloods Uhlenhuth advises extracting 
with 0.1 to 1 per cent, solutions of soda. If these are employed it is neces- 
sary to carefully neutralize with decinormal hydrochloric acid using phenol- 
phthalein for an indicator to avoid false precipitation. Ziemke recom- 
mends extracting with 1 per cent, solution of potassium cyanid, correcting 
the alkalinity by adding crystals of tartaric acid until the solution becomes 
neutral to litmus. I have sometimes aided extraction with saline solution 
by heating on a water-bath at 40° C. (not higher) for thirty minutes. 
Small portions of blood-soaked plaster, earth, hay, straw, grass, etc., Fig. 110.—Precipitin Reaction. Preparation of Extract of Blood-stain. 
The tube on the left shows the color of an extract of a blood-stain; the middle tube shows this 
extract so diluted as to yield a faint albumin reaction with nitric acid; the tube on the right shows 
the foam test with the same diluted extract (about 1 : 1000). TECHNIC OF PRECIPITIN TESTS 
323 
are extracted in the same manner as described for fabrics, saline solution 
being preferred for the solvent. 
The reaction must be neutral to litmus. As a general rule, extracts are 
neutral to this indicator, but extracts of stains on leather, wood, bark, and 
earth may require neutralization with 0.1 per cent, solutions of sodium 
hydroxid or hydrochloric acid. As recently shown by Mason1 specific 
precipitation will take place if the pH of solutions is 4.5 to 9.5 inclusive. 
However, if the pH is higher or lower than this range the precipitates may 
not form or are dissolved. 
The extract should preferably not be stronger than 1 : 1000. The 
strength may be approximately estimated by the foam test by removing 1 c.c. 
of the extract into another tube and gently shaking or bubbling air through it. 
If a persistent froth appears upon the surface of the fluid sufficient extrac- 
tion has occurred. Place 2 c.c. in a test-tube, heat to boiling, and add a 
drop of a 25 per cent, solution of nitric acid. A faint opalescence indicates 
that the strength of the extract is about 1 : 1000 (Fig. 110). If a heavy 
precipitate forms, the amount of normal salt solution that must be added 
to a portion of the extract to reduce it to a dilution of 1 : 1000 should be 
determined. The extract should be almost colorless by transmitted light, 
and must be crystal clear; this may be accomplished by filtering it through 
paper or a clean sterile Berkefeld filter. The filter shown in Fig. 50 is simple 
and very efficient. 
When a blood solution is cloudy Nuttall drops some of it on a filter-paper, 
dries it for a half-hour or so in a thermostat, after which he places it in saline 
solution and thereby secures a clear solution without more ado. 
In preparing an extract of whole blood the corpuscles should be laked 
with distilled water, the normal salt content being restored by adding an 
equal part of double strength saline solution (1.8 per cent.). Further dilu- 
tions are made with 0.9 per cent, saline solution. 
When fowl antiserum is being employed, the antigen should be prepared 
by extracting with 1.8 per cent, saline solution, and the antiserum removed 
from a refrigerator should be allowed to stand at room temperature for an 
hour or two before use (Hektoen). 
The Immune Serum.—Production.—A highly potent, sterile, and abso- 
lutely crystal-clear serum immune against the protein to be recognized 
must be prepared. For the recognition of blood-stains it is not necessary 
that whole blood be injected, as the serum alone will suffice. 
Rabbits are generally employed; guinea-pigs produce precipitins very 
poorly. Sutherland2 has observed that the fowl is well adapted for the 
preparation of precipitin sera; 5 c.c. of serum is injected into a wing vein, 
followed four days later by the injection of 10 c.c. Four days later 10 c.c. 
are injected intraperitoneally, the animal being bled two weeks later. Hek- 
toen3 has found-this method satisfactory; also that a single intraperitoneal 
injection of 20 c.c. of defibrinated blood or serum in most cases yields a 
precipitating serum in ten to twelve days of sufficient strength and specificity 
for practical purposes. Hektoen states that these fowl antisera may yield 
non-specific reactions, especially on rapid transfer from low to higher tem- 
peratures, and advises that 1.8 per cent, salt solution be used in making all 
mixtures and dilutions. Roosters should be selected instead of hens in order 
better to avoid opalescent sera. 
Animals are injected with serum or blood (defibrinated or citrated); 
1 Johns Hopkins Hosp. Bull., 1922, 33, 116. 
2 Indian Jour. Med. Research, 1915, 3, 216. 
3 Jour. Infect. Dis., 1918, 22, 561. 324 
PRECIPITINS 
both are satisfactory, but serum is probably to be preferred, as the animals 
withstand better the effects of immunization. 
Antihuman precipitins are especially difficult to prepare in sufficient 
potency for medicolegal tests; rabbits frequently succumb before their 
sera are ready. Fresh and preferably sterile sera should be employed. 
Smith1 has described a useful method for preserving serum for purposes of 
immunization consisting of diluting 200 c.c. of defibrinated blood or serum 
with an equal volume of distilled water and adding 100 gm. of ammonium 
sulphate. The precipitate is collected next day by centrifuging, dried and 
ground into a fine powder which keeps indefinitely. When used, 0.5 gm. 
is suspended in 2 c.c. saline solution and injected intraperitoneally; this 
corresponds to approximately 10 c.c. of blood. 
It is advisable to immunize several rabbits at the same time in case one 
or more succumb during the process of immunization. This is particularly 
true when human serum or blood are being employed. 
NuttalVs Method.—Nuttall prefers the intravenous method, giving 3 to 
5 c.c. serum every four to five days for at least four doses. After this time the 
animals should be tested periodically, and bled fourteen days after the last 
injection when the serum is satisfactory. 
Uhlenhuth’s Method.—Inject intravenously 2 or 3 c.c. of serum every 
five days for three doses, testing the serum seven days after the last injection, 
and daily for two or three days, bleeding in quantity when the trials show a 
Valuable antiserum. 
Hekloen's Method.—Inject 1 to 2 c.c. of serum intravenously and repeat 
after six days. Six days later 4 or 5 c.c. intraperitoneally, or from 5 to 6 c.c. 
of blood or serum may be injected intraperitoneally four or five times at 
intervals of six days. 
The sera of the rabbits should be tested nine to twelve days after the 
last injection, and the animals bled while the precipitin content is high. 
Rapid Methods.—Fornet and Miller,2 Gay and Fitzgerald,3 and Hektoen4 
have reported good results following the intraperitoneal injection of serum 
or defibrinated blood in doses of 5, 10, and 15 c.c. one day apart, bleeding 
about the twelfth day after the last injection. 
Author's Methods.—The intravenous injection of 0.5 c.c. of defibrinated 
blood or serum every day for three weeks. A trial titration is made ten 
days after the last injection; if a serum is too feeble the animal receives 
six more daily injections. 
This method has yielded good results; the mortality has been lower 
than with other methods, and the yield of acceptable sera quite good. 
Or three intravenous injections may be given of 3 c.c. defibrinated blood 
or serum at intervals of three days followed by three intraperitoneal injec- 
tions of 10 c.c. each at intervals of five days. The animals are tested ten 
days after the last injection; if a serum is too feeble the series of intra- 
peritoneal injections is repeated. 
Collection and Storage.—Most investigators agree that precipitin pro- 
duction reaches its maximum about ten to twelve days after the last in- 
jection of serum; as a general rule, this is the proper time for bleeding. 
The sera of different animals should not be mixed, but kept separately in 
order to avoid clouding. 
Collection and Storage of Sera.—The serum must be absolutely clear. 
1 Jour. Med. Research, 1916, 34, 169. 
2 Ztschr. f. Biol. Technik u. Methodik, 1908, 1, 201. 
3 Univ. California Publ. in Path., 1912, 2, 77. 
4 Jour. Infect. Dis., 1914, 14, 403. TECHNIC OF PRECIPITIN TESTS 
325 
Animals should be bled after a period of fasting, as the opalescence of the 
serum following feeding cannot be removed by filtration and will interfere 
with the reaction. Precipitin immune serum should be collected with a 
scrupulous aseptic technic, and stored in ampules holding 1 c.c. Although 
it is best not to add a preservative the addition of 0.1 c.c. of a 1 per cent, 
solution of phenol to each cubic centimeter of serum does not render the 
fluid cloudy and aids greatly in its preservation. Nuttall advises against 
the use of preservatives and prefers to remove bacteria by filtration if the 
serum has become contaminated. 
If, after long standing, a precipitate has become deposited in an anti- 
serum, this should not be shaken up, but the ampule should be carefully 
opened and the clear supernatant serum drawn off with a capillary pipet. 
A serum that has become cloudy may be cleared partially or entirely by 
filtering it through a small candle filter (Fig. 50), although even an infected 
and offensive serum will give the reaction. 
Apparatus.—Long and narrow test-tubes, 10 by 0.5 cm., are used. These 
must be absolutely clean, and preferably sterile. 
Fig. 111.—Test-tube Rack for Precipitin and Agglutination Reactions. 
The strips of black material in the rear of the tubes facilitate reading the reactions 
The test-tube rack devised by Uhlenhuth in which the tubes hang sus- 
pended in beveled holes is quite satisfactory. Where a test is carried out 
with many controls, a rack similar to the one shown in the illustration 
(Fig. Ill) is quite serviceable. A strip of black material placed behind the 
tubes aids greatly in the detection of the finer degrees of opalescence or 
precipitation. 
Preliminary Titrations.—The precipitin content of an immune serum is 
titrated frequently during the process of immunization and after the animal 
has been bled. For medicolegal purposes Uhlenhuth advises the use of 
only highly valent serums. He considers an antiserum efficient if 0.1 c.c. 
of it, when added to its respective serum-antigen diluted 1 : 1000, pro- 
duces a distinct turbidity, either at once or in from one to five minutes at 
the latest. 
Into a series of six test-tubes place 1.0 c.c. of the following dilutions of 
serum-antigen, prepared with normal salt solution: 1 : 100, 1 : 500, 1 : 1000, 
1 : 2000, 1 : 4000, and 1 : 8000. To each tube add 0.1 c.c. of clear immune 
serum. The line of contact between serum and antigen should be sharp; the 
serum may be very carefully run down the side of each tube to collect at 
the bottom, but it is better to introduce the pipet to the bottom of each 
tube. The tubes must not be shaken. Within from one to five minutes 326 
PRECIPITIN S 
a faint, misty cloud appears at the bottom of the tubes reacting positively, 
and this becomes a distinct precipitate within one-half to one hour (Fig. 112). 
As previously stated, the immune serum should be of such strength 
as to produce a reaction within an hour with a 1 : 1000 dilution of antigen. 
Weaker sera react more slowly and may require as long as twenty-four 
Fig. 112.—Titration of a Precipitin (Serum). 
Not all tubes of the series are here shown. Note the well-marked precipitate in the bottom of 
the first two tubes (1 : 100 and 1 : 500); the third tube (1 : 1000) shows less precipitate; the fourth 
tube (1 : 2000) is negative (clear and no precipitate). The titer of this serum was recorded as 1 : 1000. 
hours; these are apt to be unsatisfactory. More powerful sera may react 
at once, and Uhlenhuth has cautioned against their use in medicolegal tests 
because of the increased danger of reacting with non-related bloods. 
If more prolonged incubation is required it is better to incubate at 55° C. 
than at 37° C. to prevent bacterial contamination. I have observed that 
Fig. 113.—Method of Placing Immune Serum in Bottom of Test-tube by Means of a Pipet 
to Secure a Sharp Line of Contact in Precipitin Tests. 
this intensifies a weak or indefinite reaction and does not materially reduce 
the activity of the precipitin or coagulate the precipitum. 
Before performing the actual test with the unknown blood-stain it is 
advisable to test the entire reaction with a similar known blood-stain in 
order to make sure that all ingredients are in good working order. In labora- 
tories equipped for medicolegal examinations stains upon filter-paper or TECHNIC OF PRECIPITIN TESTS 
327 
linen, of the blood of man, dog, cat, ox, horse, etc., and their respective 
antiserums are always kept in readiness for making the preliminary and 
actual tests. 
Technic of the Test.—The following mixtures are set up in a series of 
test-tubes. Fresh sterile pipets should be used in handling the various 
solutions. The immune serum should be added slowly, and in such a way 
that it will collect at the bottom with a sharp line of demarcation; for this 
purpose it is better to introduce the pipet to the bottom of each tube rather 
than flowing the serum down the sides (Fig. 113). 
Tube 1: 1 c.c. of extract of unknown substance in dilution of 1 : 1000 
+ 0.1 c.c. of immune serum. 
Tube 2: 1 c.c. of the unknown extract in dilution of 1 : 1000 + 0.1 c.c. 
of normal rabbit serum. 
Tube 3: 1 c.c. of an extract of bloodless part + 0.1 c.c. of immune serum. 
Fig. 114.—A Precipitin Reaction. Biologic Blood Reaction. 
Tube No. 1 contains an extract of blood-stain and antihuman serum (positive reaction). No. 2 
is same with normal serum (negative). No. 4 is 1 : 1000 human serum and antiserum (positive control). 
No. 5 is extract of stain and saline solution (negative). No. 6 is antiserum and saline solution (nega- 
tive). No. 7 is extract of sheep blood and antiserum (negative). 
Tube 4: 1 c.c. of a 1 : 1000 dilution of the serum of that species of animal 
whose blood is suspected to be present in the unknown solution + 
0.1 c.c. of immune serum (control). 
Tube 5: 1 c.c. of the extract of unknown substance + 0.1 c.c. salt solu- 
tion (control). 
Tube 6: 0.1 c.c. of the immune serum + 1 c.c. of normal salt solution 
(control). 
Tube 7: 1 c.c. of the extract of the blood of some other animal + 0.1 c.c. 
of immune serum (control). 
The tubes are not shaken, are kept at room temperature, and the re- 
sults are read after from ten to twenty minutes. Exposure to incubator 
temperature facilitates the reaction. With proper immune serums, and 
especially in medicolegal work, a positive reaction should appear within 
two to five minutes as a faint, misty cloud at the bottom of the test-tube. 328 
PRECIPITINS 
Within five minutes this becomes more definite, and in from ten to twenty- 
minutes the precipitate is seen (Fig. 114). Any cloudiness that develops 
later than twenty minutes after the beginning of the reaction has no signifi- 
cance. 
In tests other than those employed in medicolegal work, especially if 
the antiserums are weaker than desired, the reaction may be read after one 
to two hours. 
In the foregoing test, if positive results are obtained in tubes 1 and 4, 
and all the others react negatively the presence of the blood or protein of 
the species suspected in the unknown extract is established. If the entire 
test proves negative, the species to which the unknown specimen belongs 
must be determined with new antiserums prepared for each species, and 
the tests conducted in the manner described. 
Partial reactions between closely related species due to group precipitins 
seldom occur, and are easily detected when the technic described is em- 
ployed. The precipitin test, as determined by the extensive experience 
of Nuttall is highly specific, and it is only between very closely related 
animals, such as the hare and the rabbit, the horse and the mule, the sheep 
and the goat, etc., that any doubt can arise. 
Sources of Error.—(a) Opalescent Antisera.—Uhlenhuth has sounded a 
warning against their use in medicolegal tests. A slight opalescence is usually 
perceptible when any serum or antiserum is added to blood dilutions, the 
tube being viewed by strong transmitted light, but the clouding here re- 
ferred to is much more marked and takes place even in salt solution. 
(b) Weak antisera which may yield falsely negative reactions or require 
prolonged incubation with consequent bacterial contamination. I have 
secured satisfactory reactions, however, with weaker sera than advised 
above by incubating at 55° C. for twelve to eighteen hours instead of at 
38° C. This higher temperature prevents bacterial multiplication and does 
not interfere with the precipitin. 
(c) Too powerful antiserum which may react with antigens of non- 
related bloods. With 1 : 1000 dilutions of antigen, however, there is little 
danger and particularly with the average antihuman rabbit serum. 
(d) Too dilute extract of serum, blood, or other unknown substance 
being tested which leads to falsely negative reactions. 
(e) Too concentrated extracts of antigen which may dissolve the pre- 
cipitate or lead to non-specific group reactions. 
(/) Too much preservative in the antiserum which may result in slight 
clouding. 
(g) Too acid or too alkaline reaction of the antigen; particularly im- 
portant with extracts of stains on leather, plaster, and in earth. 
(h) Bacterial contamination which leads to clouding and collection of 
sediment. A source of danger only when the tests are incubated or left 
in a warm room for twelve to tw'enty-four hours. 
Capillary Tube Method.—When only very limited amounts of the un- 
known stain are available, the test, according to Hauser, can be carried 
out in slender, clean, and sterile capillary tubes. The piece of stained cloth- 
ing is torn into shreds, extracted, and filtered until clear. The tests are 
performed by drawdng a small amount of the unknown solution into a 
capillary tube, and underlying this with a small amount of immune serum. 
As many controls as possible are put up in the same manner. A distinct 
whitish ring will form in the positive tubes at the line of contact between 
the two fluids; this is best seen by holding the tube against a black back- 
ground. DETECTION OF MEAT ADULTERATION 
329 
The principle of this method is the same as that in the foregoing test. 
An extract of a meat will yield a precipitate when mixed with its antiserum, 
prepared by immunizing rabbits with an extract of the flesh or with the 
blood-serum of some other animal. 
The method is especially serviceable in food inspection for the detec- 
tion of horse, dog, or other foreign flesh in meat mixtures, such as sausage 
and the like. Even salted and cooked meats may be used in the test, al- 
though the latter may require the use of antiserums prepared by immunizing 
with heated or cooked antigen. 
Preparation of the Meat Extract.—To prepare this about 50 gm. of 
flesh are removed from the deeper parts of the specimen by means of a 
sterile knife, and through a fresh opening, as this portion has been least 
exposed to the methods of preservation, especially at the high tempera- 
tures to which the meat may have been exposed. It should contain as 
little fat as possible. It is then placed on a clean sterile tile and cut into 
smaller pieces, and finally minced by passing it through a perfectly clean 
meat-grinder or chopping it with a sterile chopping knife. After being 
finely minced the meat is placed in a sterile Erlenmeyer flask, and covered 
with 100 c.c. of sterile normal salt solution. The mixture of meat and salt 
solution is kept for about six hours at room temperature, or overnight 
in the refrigerator, the flask being gently shaken from time to time. 
Salted meat should be ground and freshened by placing it in a large 
sterile Erlenmeyer flask and covering it with sterile distilled water, re- 
newed several times in the course of fifteen minutes without shaking the 
flask. 
Graham-Smith and Sanger1 have found that with extracts containing 
more than 5 per cent, sodium chlorid the antiserum did not sink to the 
bottom, and that clouding occurred at the top of the tubes and took longer 
in forming. 
Since the presence of a great deal of fat interferes with the reaction, 
it is advisable to remove it beforehand by extracting it with ether and 
chloroform for from twelve to twenty-four hours (Miessner and Herbst). 
Pork sausages are usually quite fatty, and may require this preliminary 
treatment. To make the extraction, take about 75 to 100 gm. of the minced 
meat or sausage and place it in a large Erlenmeyer flask and cover with 
equal parts of ether and chloroform. After twenty hours the ether and 
chloroform are poured off, the meat is washed once or twice with sterile 
normal salt solution, and then extracted in 100 c.c. of salt solution as de- 
scribed above. 
To determine whether a sufficient quantity of protein substances has 
passed into solution place 2 c.c. in a test-tube and shake vigorously. If 
a foam develops and persists for some time the extraction may be said to 
be complete. It must then be filtered until it becomes perfectly clear. 
With extracts of fresh lean meat this is usually accomplished by filtering 
through a hard filter-paper moistened with salt solution. If it is not crystal 
clear, and especially if the meat to be examined is fat or salt, it may be 
necessary to filter through a sterile Berkefeld filter. 
To make the test the extract should contain about 1 part of protein 
in 500 parts of salt solution. To determine this, 2 c.c. of the clear filtrate 
are placed in a test-tube and heated, a drop of dilute nitric acid being added. 
If a marked cloudiness and a flocculent precipitate develops, the extract 
is too concentrated and must be diluted with clear normal salt solution 
Detection of Meat Adulteration 
1 Jour. Hyg., 1903, 3, 258. PRECIPITINS 
330 
until the heat and acid test causes only a diffuse, opalescent cloudiness 
that settles at the bottom of the tube after five minutes as a slight precipitate. 
Before proceeding with the experiment the reaction of the solution 
should be tested with litmus-paper, and if it is found to be acid, it should 
be neutralized very carefully with n/10 sodium hydroxid. 
Extracts of the meats that are known or likely to be present, such as 
extracts of pork and beef, should be prepared as controls. 
Preparation of Immune Serum.—An immune serum against that variety 
of flesh that is to be determined in the unknown specimen is prepared by 
injecting rabbits intravenously with the serum of an animal of that species. 
For example, if the object is to test for dog meat, an antidog serum is pre- 
pared by immunizing rabbits with sterile dog serum. 
As has repeatedly been mentioned it is advisable to immunize a number 
of rabbits at the same time, for only a small number will yield a satisfactory 
serum after the third injection. 
Immunization may be performed with extracts of flesh that have been 
filtered and heated at 56° C. for an hour to secure partial sterilization. 
Such injections when given subcutaneously are likely to produce extensive 
sloughing, and with any method of immunization the mortality is high. 
After the third inoculation it is well to remove a small amount of blood 
from the ear and make a preliminary titration. This is performed in exactly 
the same manner as in making the forensic blood test previously described. 
An antiserum is satisfactory if 0.2 c.c. produces a well-marked cloudiness, 
and a precipitate in ten minutes with 1 c.c. of a 1 : 500 or 1 : 1000 dilution 
of the serum or extract of flesh. 
In addition to being highly potent the immune serum must be crystal 
clear and sterile. To avoid opalescence the animal should be bled only 
after a period of fasting. 
When testing for cooked meats the antiserum should be prepared by 
immunizing rabbits with serum diluted with saline and heated at 70° C. 
for thirty minutes; antisera prepared by immunization with unheated sera 
may suffice, but better results are likely to occur with heated sera (see previous 
discussion on coctoprecipitins). 
Technic.—If, for example, the object is to determine whether a piece 
of meat is horse flesh, or if sausage contains the meat of this animal, the 
test is conducted as follows: . 
Tube 1: 1 c.c. of unknown extract, 1 : 500 + 0.2 c.c. of antihorse-serum. 
Tube 2: 1 c.c. of unknown extract, 1 : 1000 + 0.2 c.c. of antihorse- 
serum. 
Tube 3: 1 c.c. of unknown extract, 1 : 500 + 0.2 c.c. of normal rabbit- 
serum. 
Tube 4: 1 c.c. of horse flesh extract or horse-serum, 1 : 1000 + 0.2 c.c. 
of antihorse-serum (positive control). 
Tube 5: 1 c.c. of serum of some other animal, 1 : 500 + 0.2 c.c. of anti- 
horse-serum (control). 
Tube 6: 1 c.c. of unknown extract + 0.2 c.c. saline solution (control). 
Tube 7: 1 c.c. of sterile salt solution + 0.2 c.c. of antihorse-serum (control). 
The immune serum is added to each tube very carefully and preferably 
by placing the pipet in the bottom of each tube or by flowing the serum 
down the sides. The tubes should not be shaken. 
If the preliminary titration of the immune serum fulfils the ideal re- 
quirement of yielding a well-marked cloudiness within five or ten minutes 
with a 1 : 1000 extract, the foregoing test should be recorded at the end 
of half an hour at room temperature. If in tubes 1,2, and 4 a misty cloudi- BACTERIAL PRECIPITIN TESTS 
331 
ness should appear within five minutes, the extract is very probably one 
of horse flesh. If a definite precipitate forms within thirty minutes, the 
other tubes remaining clear, horse flesh or the flesh of some other single- 
toed animal is present. 
If the preliminary titration does not show a precipitate with the immune 
serum until at the end of one or two hours, this interval may be utilized 
for conducting the test. 
In a similar manner tests may be made for the meat of dogs, cats, or 
any other animals if the respective immune serums are used with the extract. 
Bacterial Precipitin Tests 
Bacterial precipitinogens are prepared by -filtering ten to twenty-one day 
bouillon cultures through Berkefeld filters. The filtrates must be abso- 
lutely clear and sterile for the reaction frequently requires a number of 
hours, and if bacteria are present they may grow quickly, produce turbidity, 
and mask a reaction. 
Ascoli prepares precipitinogens by washing off twenty-four-hour agar 
cultures with 5 to 10 c.c. saline solution, shaking for two hours and filtering 
through asbestos until clear. 
If broth cultures are employed it is important that the reaction is not 
too acid or too alkaline in order to avoid flocculation of serum proteins; 
Noble1 has found that broth + 1.5 to phenolphthalein produces slight 
flocculation. The reaction should not range more than — 0.5 to + 1.0. 
Thermoprecipitinogens (Methods of Ascoli, Krumwiede, and Noble).— 
Extracts of bacteria in tissues are prepared by Ascoli (for anthrax) by 
boiling 1 gm. of minced spleen in 5 c.c. saline solution and filtering. Krum- 
wiede and Noble2 have described the following method for preparing pre- 
cipitinogens of bacteria which, briefly, consists of growing the bacteria on 
large areas of agar; collecting and suspending them in distilled water; dis- 
solving them by the addition of alkaline hypochlorite solution (antiformin), 
and boiling to clearness; neutralization with hydrochloric acid; precipita- 
tion with alcohol and extraction of the sediment with 0.8 per cent, salt 
solution at 100° C. The final extract is clarified by centrifuging. 
Immune Serum.—'This is prepared according to the technic described 
under Active Immunization. Rabbits are given intravenous injections of 
increasing doses of cultures of the bacteria themselves or of filtrates, the 
inoculum being heated at 60° C. for an hour previous to making the injec- 
tion. After the third dose the serum is titrated and the injections continued 
unless the serum is satisfactory. 
Technic.—A known quantity of serum and varying amounts of pre- 
cipitinogen are employed. If too much precipitinogen is furnished, the 
precipitate will not form, and one that has already formed may dissolve 
on the addition of more precipitinogen. 
If, for example, one desires to determine if anthrax precipitin is present 
in a given serum, the test is conducted as follows: 
Tube 1: 0.5 c.c. antigen + 0.5 c.c. unknown serum very carefully and 
slowly run down the side of the tube in order to gather under the antigen 
with a sharp line of contact. 
Tube 2: 0.5 c.c. of 1 : 10 antigen + 0.5 c.c. unknown serum. 
Tube 3: 0.5 c.c. of 1 : 100 antigen + 0.5 c.c. unknown serum. 
Tube 4: 0.5 c.c. of antigen + 0.5 c.c. known positive serum (positive 
control). 
1 Jour. Infect. Dis., 1904, 1, 463, 
2 Jour. Immunology, 1918, 3, 1. PRECIPITINS 
332 
The tubes are not shaken, and are kept at room temperature. If the 
unknown serum contains considerable anthrax precipitins, a positive reac- 
tion will be noticed in the first three tubes in a short time—often within 
from ten to fifteen minutes. Tube 4 should show a strong reaction and the 
other tubes should remain clear. 
Quantitative Precipitin Test.—In studying the biologic relationship of 
an organism to others of the same group its immune serum may be used 
in amounts of 0.5 c.c. of varying dilutions with a constant dose of 0.5 c.c. 
of the bouillon filtrates of the various organisms studied. A comparison 
of the precipitates in the respective dilutions of the different filtrates indi- 
cates the relationship according to the amount of group precipitins present 
in the immune serum. The test may be conducted as follows: 
Tube 1: 0.5 c.c. undiluted serum + 0.5 c.c. antigen very carefully over- 
laid. 
Tube 2: 0.5 c.c. serum 1 : 2 + 0.5 c.c. antigen. 
Tube 3: 0.5 c.c. serum 1 : 4 + 0.5 c.c. antigen. 
Tube 4: 0.5 c.c. serum 1 : 8 + 0.5 c.c. antigen. 
Tube 5: 0.5 c.c. serum 1 : 16 + 0.5 c.c. antigen. 
Tube 6: 0.5 c.c. serum 1 : 32 + 0.5 c.c. antigen. 
Tube 7: 0.5 c.c. serum 1 : 64 + 0.5 c.c. antigen. 
Tube 8: 0.5 c.c. serum 1 : 128 + 0.5 c.c. antigen. 
Tube 9: 0.5 c.c. undiluted serum + 0.5 c.c. saline (control). 
Tube 10: 0.5 c.c. saline solution + 0.5 c.c. antigen (control). 
The tubes are not mixed but handled very gently and allowed to stand 
at room temperature for at least one hour, the first reading being made 
after fifteen minutes. 
Higher dilutions of immune serum may be employed. 
The immune sera are prepared by injecting rabbits with the respective 
micro-organisms. 
Absorption of Bacterioprecipitins.—Krumwiede and Cooper1 have de- 
scribed a method for removal of group precipitins from immune serum by 
absorption with precipitinogens or the bacteria themselves. They have 
found absorption of only limited application in the differentiation of closely 
related bacteria with a tendency toward non-specific results, that is, ab- 
sorption may remove not only the main precipitin, but the group precipitins 
as well. 
Dochez and Avery’s Precipitin Reaction with the Urine in Pneumonia. 
—The urine should be fresh and cleared by centrifuging. In each of a 
series of four small test-tubes place 0.5 c.c. urine with 0.5 c.c. clear anti- 
pneumococcus sera of Types I, II, and III; the fourth tube receives 0.5 c.c. 
saline solution and is a control. Mix the contents of each tube and place 
them in a water-bath at 37° C. for one hour. It is essential that the urine 
and sera be water clear. A positive reaction is indicated by a cloudy to 
heavy flocculent precipitate. 
In case the reaction is negative or indecisive the urine may be concen- 
trated by adding a few drops of acetic acid to 25 c.c. and boiling down to 
5 c.c. followed by paper filtration. To this are added 50 c.c. of 95 per cent, 
alcohol and the precipitate collected by centrifuging. The precipitate is 
rapidly dried in an incubator and extracted with 3 c.c. of salt solution. 
This extract is cleared by centrifuging and employed as above. 
Blake’s Precipitin Reaction with Mouse Peritoneal Exudate in Pneu- 
monia.—Mice are inoculated with sputum and the peritoneal exudate re- 
covered as described for the agglutination test. After thoroughly centrif- 
1 Jour. Immunology, 1920, 5, 547. BACTERIAL PRECIPITIN TESTS 
333 
uging, the clear supernatant fluid is pipeted off with care not to disturb 
the sediment, and is used as follows: 
I. (1 : 10) + 0.5 c.c. peritoneal fluid. 
II. (undiluted) + 0.5 c.c. peritoneal fluid. 
II. (1 : 10) + 0.5 c.c. peritoneal fluid. 
III. (1:5) + 0.5 c.c. peritoneal fluid. 
Incubation is usually not necessary. A positive reaction is shown by 
the development of a whitish ring at the line of contact or a diffuse cloudi- 
ness. A negative reaction in all tubes indicates pneumococcus belonging 
to Type IV. A positive reaction in Tube 2 and a negative reaction in Tube 
3 indicates atypical pneumococcus Type II; a positive reaction in both 
Tubes 2 and 3 indicates typical pneumococcus Type II. 
Oliver’s Precipitin Reaction with Sputum in Pneumonia.—Oliver1 has 
described the following method for the rapid and direct diagnosis of pneu- 
monia: “After a direct smear of the sputum has been made from 1 to 2 c.c. 
are placed in a clean centrifuge tube. From 3 to 5 drops of undiluted ox 
bile (or a 10 per cent, solution of sodium taurocholate) are added and a 
sufficient quantity of sterile physiologic sodium chlorid solution, if neces- 
sary, to insure a specimen of sufficient fluidity to allow of centrifugation. 
The mixture is then thoroughly stirred and broken up with a glass rod; 
grinding in a small mortar with a pestle may be necessary. The tube is 
then heated in a water-bath at 42° to 45° C. for twenty minutes which 
suffices for a solution of the pneumococci by the bile. The fluid is then 
centrifugalized. Of the centrifugate, from 0.3 to 0.5 c.c. quantities are 
carefully pipeted into each of three small, scrupulously clean tubes. To 
the first tube is added from 1 to 2 drops of undiluted Type I pneumococcus 
serum, and to the second and third tubes the same quantity of Type II and 
Type III sera, respectively. A positive reaction is shown by an almost 
immediate clouding and flocculation which is enhanced by heating at 42° C. 
in a water-bath for from ten to twenty minutes.” 
Precipitin Reactions with Spinal Fluid in Meningitis.—The spinal fluid 
is centrifuged and the clear superfluid employed. 
If pneumococcus meningitis is suspected the tests are conducted by 
placing 0.5 c.c. of fluid in each of three small test-tubes, adding 0.5 c.c. of 
Types I, II, and III antipneumococcus sera, respectively. These are mixed 
and incubated in a water-bath at 37° C. for one hour. Positive reactions 
are indicated by clouding and flocculation. 
If meningococcus meningitis is suspected the test is conducted in the 
same manner with monovalent sera for the identification of type of meningo- 
coccus, or with a polyvalent serum. 
Precipitin Reactions in Other Bacterial Infections.—Precipitin reactions 
have also been employed by Robinson and Meader2 for the diagnosis of 
gonococcus infections, but in the experience of Kelley3 and others the reaction 
has not proved of practical diagnostic value. 
Smith and Kaufman4 have recently described a precipitin reaction 
occurring with antigens prepared of swabs containing diphtheria bacilli 
from exudates prepared by a method described by the authors, and horse- 
or rabbit- immune sera. Of 74 cases controlled by cultural methods the 
precipitin reaction was found to yield satisfactory results. 
Other Uses of the Precipitin Test. The Identification of Bone.—Most 
1 Jour. Infect. Dis., 1920, 27, 310; ibid., 1921, 29, 518. 
2 Jour. Urology, 1920, 4, 551. 
3 Jour. Infect. Dis., 1922, 30, 623. 
4 Jour. Lab. and Clin. Med., 1922, 7, 619. PRECIPITINS 
334 
difficulty is encountered in the preparation of an extract. Soft tissues 
should be removed. The bone should be reduced to a powder; I have found 
sawing and collection of the dust most convenient unless small fragments 
may be pounded to dust with a clean hammer. Scrupulous cleanliness 
must be observed throughout to guard against accidental mixtures of pro- 
teins of other origin. It is better to handle the bone with forceps. The 
dust is extracted with salt solution as described for the preparation of ex- 
tracts of meats. The extract should give a positive foam reaction, and be 
clarified if necessary by filtration. 
Antiserum is prepared by injecting serum into rabbits. Usually the 
information sought is whether or not a fragment is human bone. The 
tests are conducted with an antihuman rabbit-serum exactly as described 
for the forensic blood test. 
For the Identification of Milk.—Antiserum is prepared by injecting milk 
intravenously into rabbits in the same amounts advised for serum. I gener- 
ally heat the milk at 60° C. for one hour to reduce the bacterial content 
before injection. The test may be conducted -with antisera prepared by 
injecting rabbits with serum instead of milk. 
In conducting the tests the milk antigen is prepared by diluting 1 : 25 
with saline solution and filtering. To 1 c.c. of this antigen is added 1 c.c. 
of immune serum, followed by gentle mixing; the antigen may be stratified 
over 0.5 c.c. of the serum, but usually the ring cannot be clearly seen. A 
positive reaction is indicated by flocculation and collection of a precipitate. 
Bordet conducted the test by adding 6 to 15 drops of milk to 3 c.c. of 
immune serum. 
For the Identification of Semen.—These are usually stains on clothing. 
The antigen is prepared by extracting with physiologic saline solution for 
several hours. The extract should be concentrated by using minimal amounts 
of saline and a portion centrifuged. The sediment should be examined for 
spermatozoa; the heads resist deterioration for much longer periods than 
the tails. The balance of the extract may then be diluted until it gives a 
satisfactory foam reaction and a slight cloud when boiled and a few drops 
of acetic acid added. The extracts usually require clearing by filtration. 
The antiserum may be prepared by injecting rabbits with human serum, 
semen, or testicular extracts. Immunization with semen yields the best 
serum. 
The tests are conducted in exactly the same manner as the blood tests. 
Control extracts of guinea-pig testicle or other animal should be included. 
As a general rule the immune sera also give precipitates with the serum 
of the animal corresponding to the species of spermatozoa employed for 
immunization. Pfeiffer,1 however, secured specific antisemen sera by ab- 
sorbing these group serum and organ precipitins. He injected rabbits with 
dried and powdered bull spermatozoa, suspended in salt solution; the result- 
ing antiserum acted strongly on semen solutions and testicular extracts, 
and only feebly or not at all on extracts of other beef organs, and by treat- 
ment of the antiserum with beef serum and certain organ extracts all pre- 
cipitins except those specific for semen could be removed. This treated anti- 
serum caused precipitates in dilutions of bull semen, and detected bull semen 
in mixtures with organ extracts. 
Hektoen2 has recently confirmed this work for human semen. He pre- 
pared immune sera by injecting rabbits with semen secured in the usual 
clinical manner by massage of the seminal vesicles, as follows: four or five 
1 Wien. med. Wchn., 1905, 18, 637. 
2 Jour. Amer. Med. Assoc., 1922, 78, 704. BACTERIAL PRECIPITIN TESTS 
335 
injections were made intramuscularly in rabbits at intervals of three or four 
days, beginning with 2 c.c. and increasing the quantity by 2 c.c. each suc- 
ceeding injection. As a rule the best time to bleed the rabbits for serum 
was found to be from six to eight days after the last injection. As was 
expected, the immune sera gave precipitates with human serum as well as 
with human semen. To remove the precipitin for human serum equal parts 
of antiserum and 1 : 200 dilution of human serum in normal saline solution 
were mixed, left at room temperature for about one hour and in the ice-box 
overnight, and then thoroughly centrifuged. As a rule this procedure 
removes all precipitin for human serum and leaves a serum specific for human 
semen, as shown in the following table taken from Hektoen’s paper: 
Specific Precipitins for Human Seminal Proteins in Serum of Rabbits Injected 
with Human Semen 
Titers of Antiserum in 
Serum of Rabbits 
Injected with 
Human 
Human 
Seminal 
Animal Seminal 
Fluids (Bull, Boar, 
Dog, Guinea-pig, 
Salt 
Human Semen. 
Serum. 
Fluid. 
Rabbit, Rat). 
Solution. 
1. Original 
6400 
800 
0 
0 
Treated 
0 
256 
0 
0 
2. Original 
3200 
256 
0 
0 
Treated 
0 
64 
0 
0 
3. Original 
6400 
640 
0 
0 
Treated 
0 
256 
0 
0 
4. Original 
6400 
640 
0 
0 
Treated 
0 
320 
0 
0 
Normal rabbit serum 
0 
0 
0 
0 
The figures give the highest dilution of serum and seminal fluid in which the antiserum 
caused distinct precipitates by the layer or contact method after one hour at room tem- 
perature. 
The clear fluid secured by centrifuging semen or salt solution extracts 
of stains is employed as antigen; the former should be diluted about 1 : 100. 
The tests are conducted by placing 1 c.c. of seminal solutions in test-tubes 
and introducing in the bottom of each 0.1 c.c. of immune serum by means 
of a pipet in order to secure sharp contact. The readings are made after 
one hour at room temperature. CHAPTER XVIII 
CYTOLYSINS 
General Considerations.—The cytolysins include a number of antibodies 
of considerable diagnostic and therapeutic importance, for example, the 
hemolysins and the bacteriolysins. It will be remembered that the various 
antibodies act differently upon their antigens, and that, according to the 
side-chain theory, as their antigens become more highly organized, their 
structure becomes more complicated. For example, the molecule of a 
soluble toxin may be considered as simple in structure, and accordingly 
its antibody has been conceived as being likewise simple, and composed 
of a plain cast-off receptor or side-arm that unites directly with the toxin 
and neutralizes it without further aid. Antitoxins and antiferments are 
antibodies of this nature. For more highly organized antigens, however, 
so simple an antibody will not suffice, and wre now find a more complicated 
antibody, composed of a portion that unites with the antigen and another 
portion, an integral part of the antibody, that exerts a special selective 
action upon the antigen, and either neutralizes its activity or prepares it 
for ultimate destruction. To this class of antibodies belong the agglutinins 
and precipitins, which agglutinate or precipitate their antigens preparatory, 
in a sense, to their final disintegration. For still more complex antigens 
nature has provided special ferment-like substances, always present in 
varying proportions in the blood, which, when united with the antigen, 
cause its disintegration and solution in a manner similar to the process of 
digestion as it takes place in the intestinal canal. These ferment substances 
are, however, powerless unless united with the antigens, and here we find 
that the antibody serves as the connecting link, binding antigen with fer- 
ment, which results in a form of digestion and final lysis or solution. The 
antibody is, therefore, simple in structure, and is composed of two binding 
or grasping arms—one for the antigen and one for a ferment. This inter- 
body, or amboceptor, is specific for the antigen, and will act only and specifically 
with this antigen. It is important to remember that the ferment or comple- 
ment is not an integral part of the antibody, but is free in the blood-stream; 
that the antibody is only a sensitizer or connecting link, but preserves its 
importance by being specific for its antigen; that the primary function of 
this antibody is to prepare (sensitize) the antigen, or unite antigen and 
complement, and that the latter then causes the lysis or solution of the 
antigen. The sensitizer or connecting link or antibody of this nature is 
known as an antibody or receptor of the third order. 
Different cells produce their own and specific interbodies or ambocep- 
tors. Thus bacteria or vegetable cells, blood-corpuscles, and various other 
cells, such as ciliated epithelium, spermatozoa, renal epithelium, etc., when 
present in the form of an infection, or when injected into an animal, generate 
different and specific sensitizers or amboceptors, which bring about their 
solution by binding them with the ferment or complement. One ferment 
or complement may not serve for all; there may be various ferments, which 
act with the different amboceptors, but all have properties so nearly alike 
that many believe with Bordet that but a single complement exists. 
Amboceptors (Sensitizers) and Complements (Alexins) 
336 AMBOCEPTORS AND COMPLEMENTS 
337 
Definition.—This special digestive and lytic process is known to occur 
with cells, and hence the antibodies capable of bringing about this action 
are called cytolysins, or substances that cause lysis or solution of the various 
cells that may be their antigens. 
According to Ehrlich the three orders of antibodies each have their 
counterpart, both in structure and in effect, in the receptors serving for 
the normal nutrition of cells. For the simplest molecule of food that is in 
solution the cell is provided with a simple receptor for union with the mole- 
cule which is then directly assimilated. This receptor is similar to an anti- 
toxin, or an antibody of the first order, which destroys its toxin directly 
and without further ado. More complex food material must first undergo 
Fig. 115.—Formation of Cytolysins (Hemolysins, Bacteriolysins, Cytotoxins). 
The central white area represents a molecule of a cell; the shaded portion represents the cell 
itself; the surrounding area represents the body fluids about the cell. 
R, Receptor of the molecule {third order)-, R-, overproduction of receptors, which are being cast 
off; A, a cast-off receptor which now constitutes the antibody or amboceptor; C, molecule of com- 
plement free in the body cells and body fluids; A-A*, amboceptors in combination with molecules of a 
cell (antigen) and a complement; A3, an amboceptor in combination with a molecule of a cell. The 
cell (antigen) is now said to be sensitized. Lysis does not occur because a complement is not united. 
some preparation by the cell before it can be assimilated, and accordingly 
we find receptors provided with a more complex structure which have their 
counterpart in the antibodies of the second order, or those possessing a 
special toxic portion that agglutinates or precipitates their antigen or pre- 
pares it for phagocytosis. It is possible thgt with physiologic substances 
this is all that the cell requires of its receptor, but so far as is known it would 
appear that for antibodies this action does not in itself injure the antigen, 
but is rather one step toward preparation for its further destruction. Or- 
ganized and complex food substances must be digested before assimilation 
can occur, and here we find that the receptor acts as a link in binding the 
food molecule to a ferment, with resulting dissolution and assimilation of 
the products of solution. These are called receptors of the third order, 338 
CYTOLYSINS 
and have their counterpart in similar antibodies—the cytolysins—which 
act as links or interbodies between antigen and a complement, the latter 
being entirely free and separate, and independent of the receptor or anti- 
body (interbody) itself (Fig. 115). 
Varieties of Cytolysins.—The cytolysins produced by bacteria are known 
as bacteriolysins, i. e., antibodies producing disintegration and lysis of 
bacteria. The cytolysins known as hemolysins cause lysis or hemolysis 
of the erythrocytes. Similar cytolysins may be formed for practically all 
cells, such as leukocytes, epithelium, liver, kidney, spleen, etc., and to these 
the general name cytoloxin has been given; thus we have leukotoxin, hepato- 
toxin, nephrotoxin, neurotoxin, etc., these terms being more nearly correct 
and expressive of the actual mechanism by which their action is produced. 
Nomenclature.—In no other field of immunity have so many different 
names been applied to the same substances as have been applied to this 
order of antibodies. This confusion of terms, added to the various inter- 
pretations placed upon their significance, has rendered the subject incom- 
prehensible to those not specially interested. 
The ferment-like and thermolabile substances present in all serums 
and actively concerned in lytic processes have been given the name of 
alexin by Bordet; Metchnikoff called it cytase, and Ehrlich designated it 
as complement or addiment because in the conception of the side-chain 
theory it completes the reaction after being linked with the antigen. The 
term “alexin” was first applied by Buchner to the germicidal substance 
found in normal serum. We now know that Buchner was working with 
both bacteriolysins and complement, although Bordet was the first to dis- 
cover the former, Buchner having been unconsciously most interested in 
the thermolabile complement. 
To the antibody itself the term substance sensibilisatrice has been applied 
by Bordet, for he believes that this antibody sensitizes or prepares the cell 
for the action of the alexin or complement. The following names have 
been applied to the antibody by various observers: fixative or fixateur, by 
Metchnikoff; preparator, by Muller; and amboceptor, interbody, and immune 
body by Ehrlich. Of these the terms “sensitizer” and “amboceptor” are in 
most general use, signifying a two-armed body that unites antigen on the 
one hand, with a complement on the other. 
When using the term “amboceptor,” care should be used to designate 
its specific character; thus, for example, a hemolytic amboceptor and a bac- 
teriolytic amboceptor mean respectively a hemolysin and a bacteriolysin. 
It is common practice to designate an amboceptor according to the cell 
for which it has a special affinity; thus antisheep amboceptor or hemolysin 
means an amboceptor for sheep cells, the prefix “anti” being affixed because 
it is specific for those cells. 
AMBOCEPTORS OR SENSITIZERS 
Although antitoxins have received considerable study from a thera- 
peutic standpoint, probably no order of antibodies has been given more 
attention than the cytolvsins have received, not only because of their vast 
therapeutic possibilities but also from their value as an aid to diagnosis. 
The hemolysins especially have been utilized in making the Wassermann 
test for syphilis and similar reactions, the very nature of the phenomenon 
offering a visible and fascinating method of study. 
Since the general structure, formation, and action of the various sensi- 
tizers or amboceptors, such as the bacteriolysins, hemolysins, and other AMBOCEPTORS OR SENSITIZERS 
339 
cytolysins, are essentially similar, the general character of sensitizers may 
be here considered, a study of the special characteristics of each being re- 
served for subsequent chapters on the more important members of the 
group. 
Historic.—The alexins or complements were first discovered through 
the researches of Nuttall and Buchner in 1889. The sensitizers were, of 
course, present in the various serums with which these observers were work- 
ing, but it was not until 1895 that Bordet showed quite clearly that two 
substances were concerned in the phenomena of bacteriolysis and hemolysis. 
At this time he demonstrated that the alexin or complement may be re- 
moved from a serum by heating it to 55° to 56° C., and that it may be re- 
activated by the addition of fresh serum from another animal; that an old 
bacteriolytic serum cannot produce bacteriolysis unless it is reactivated 
by a fresh normal serum or is placed in the peritoneal cavity of a living 
animal, from which it may derive the thermolabile alexin. In other words, 
the sensitizers in these serums withstood the effects of heating and age, 
but were unable to produce lysis without the aid of an alexin furnished by 
a fresh normal serum. 
Structure of Sensitizers or Amboceptors.—According to the theory of 
Ehrlich, an amboceptor is but a simple interbody furnished with two hapto- 
phore or grasping portions. One haptophore 
group attaches the antibody to its antigen what- 
ever that may happen to be—bacterium, ery- 
throcyte, epithelial cell, etc., while the other 
attaches a suitable complement (Fig. 115). The 
first is called the cytophil or antigentophil group, 
and the second, the complementophil group. 
The amboceptor is specific in the sense that it 
will unite only with its antigen or other very 
closely related body. For example, when a 
rabbit is injected with sheep corpuscles an 
amboceptor is formed that will unite only with 
sheep, and not with human, dog, ox, or other 
cells. 
As will be shown further on, Ehrlich believes that many different com- 
plements may be present in a serum, whereas Bordet believes that one 
complement exists that will act with the sensitizer or amboceptor, whether 
this is bacteriolysin or hemolysin. This view is based mainly upon the 
observation that the complement in a serum may be absorbed out by fur- 
nishing an excess of either bacteriolytic or hemolytic amboceptors, the 
one variety of amboceptor removing all the complement for the other. 
Although the results of experimental work would seem to indicate that 
Ehrlich’s belief in the plurality of complements is correct, and while this 
view is quite generally held, conclusive proof regarding this has not as 
yet been furnished. An amboceptor may have more than one comple- 
mentophil group, and may bind a number of different complements simul- 
taneously (polyceptor) (Fig. 116). Ehrlich and Morgenroth called atten- 
tion to this possibility when they stated: “Finally, it is possible that an 
immune body besides one particular cytophil group, contains two, three, 
or more complementophil groups.” Later Ehrlich and Marshall showed 
that in order to get a specific lytic effect it was not necessary for all com- 
plements to become active, but that only a few are necessary in any single 
instance to bring about effect. These complements are termed “dominant 
complements,” the remainder being known as “non-dominant complements.” 
Fig. 116.—Theoretic Struc- 
ture OF A POLYCEPTOR. 
A, Main portion of amboceptor 
in combination with a cell; C, 
dominant complement; c, lesser 
complements. 340 
CYTOLYSI NS 
Amboceptoids.—Whether amboceptors can undergo degenerative changes 
and lose their cytophil or complementophil groups and become amboceptoids, 
just as toxoids and agglutinoids are formed, is still doubtful. Reasoning 
from analogy to the toxins and agglutinins, it is probable that ambocep- 
toids may be produced by a loss of the complementophil group, the cyto- 
phil portion of all antibodies being more stable; such amboceptoids, by 
uniting with their antigens, may effectually block the action of an ambo- 
ceptor, just as agglutinoids prevent agglutination. 
General Properties of Sensitizers or Amboceptors.—Amboceptors are 
fairly resistant bodies, withstanding to a well-marked degree the effects of 
heat, acids, alkalies, exposure, and drying. A hemolytic serum, for instance, 
may be preserved in a sterile condition for many months and show but 
slight deterioration in its activity. Such a serum may be dried in vacuo 
or on suitable filter-paper, and preserve its activity for remarkable intervals 
of time with but slight and gradual deterioration. While a temperature 
of 55° C. will inactivate complement in from fifteen to thirty minutes, 
amboceptors can tolerate from 60° to 65° C. for an hour and show but 
slight depreciation in activity. 
Formation of Sensitizers or Amboceptors.—While experimental data 
are at hand to show that amboceptors may be produced by local tissues, 
it is entirely probable that in wide-spread infection or as the result of arti- 
ficial immunization there is general cellular activity with extensive anti- 
body formation. The spleen and hematopoietic tissues in general and the 
mononuclear leukocytes are regarded by many as being particularly active 
in the formation of hemolysins and bacteriolysins (Pfeiffer and Marx, 
Deutsch, Wassermann). 
As has been stated, Metchnikoff believes that antibodies of the class 
under consideration are the products of the leukocytes, thus tending to 
preserve the importance of the phagocytic theory. While there is little 
doubt that the various leukocytes, endothelial cells, and other phagocytic 
cells are sources of amboceptor production, there is no reason for accept- 
ing the belief that their formation is confined strictly to these cells. 
Specificity of Amboceptors (Sensitizers).—It has been stated elsewhere 
in this volume that amboceptors are highly specific bodies. This specificity 
is not, however, absolute, for just as group agglutinins are produced by 
one bacterium for closely allied species, so in like manner experimental 
investigation by Ehrlich and Morgenroth, von Dungern, and others has 
shown that immunization of an animal with the erythrocytes of another 
animal would produce one chief hemolysin for these cells and a secondary 
hemolysin for the cells of another animal. For example, on immunizing a 
rabbit with ox blood a hemolytic serum was obtained that was hemolytic 
not only for goat blood but also for ox blood. These secondary ambo- 
ceptors are known as group or partial immune bodies. Their production 
may readily be understood when it is remembered that the body cells are 
conceived as being provided with various side arms for many different 
blood-cells, bacteria, etc. Now, if the erythrocytes of the goat possess 
receptors not only for the particular goat-blood side arms of the body cells 
but also for the ox-blood side arms, both sets of side arms will be attacked 
and consequently two amboceptors are formed—one, the main one, for 
goat corpuscles, and a secondary one for ox corpuscles. Ehrlich and Morgen- 
roth, therefore, claim that the immune body of a hemolytic serum is com- 
posed of the sum of the partial immune bodies, which correspond to the 
individual receptors used to confer the immunity. Since the cells of various 
animals of the same and of different species vary in the number and variety AMBOCEPTORS OR SENSITIZERS 
341 
of side arms or receptors, which are not present in another, the different 
combining group possessed by a blood-cell or a bacterium will not, there- 
fore, find fitting receptors in every animal, and thus there may be a different 
variety of partial immune bodies in two animals. This would lead to the 
possibility of the occurrence of antibodies for the same blood-cell or bac- 
terium, differing from one another in the partial immune bodies of which 
they are composed, depending on the variety of the animals used in pre- 
paring the serum. 
This view is directly opposed to that of Metchnikoff and Besredka, 
who believe that a certain immune body is always the same no matter 
what species of animal was used in preparing the serum. As will be pointed 
out further on, in addition to theoretic interest, the subject possesses great 
practical importance, for as is well known, most curative serums are best 
prepared with many different strains of a particular micro-organism be- 
cause of certain differences in their antigenic properties, and if, in addition, 
the value of a bacteriolytic serum depends upon the sum total of the im- 
mune bodies it may be advisable to secure as many of these as possible 
by preparing the serum from various animals of the same and of different 
species. 
It will be understood, therefore, that the specific action of antibodies 
of this order is not limited to the cells used in the immunizing process, but 
extends to other cells that have receptors in common with these,- a condi- 
tion that is analogous to group agglutinins and precipitins for closely allied 
cells and bacteria or dissolved albumins. 
Natural or Native Amboceptors.—Just as small and varying amounts 
of native agglutinins and antitoxins may be found in normal serums so, in 
like manner, various native bacteriolytic and hemolytic amboceptors may 
be found. According to the side-chain theory, these various amboceptors 
are normally attached to body cells, hence it is probable that a few are 
being continually swept off into the blood-stream. In some instances the 
amount of a natural amboceptor may be quite high; thus, for example, 
many human serums contain relatively large amounts of antisheep hemo- 
lytic amboceptor. These natural amboceptors will be considered more 
fully in the chapters on Hemolysins and Bacteriolysins. 
The difference between a normal and an immune serum lies in the fact 
that the normal serum contains a number of amboceptors in small amounts, 
whereas the immune serum contains a greatly increased amount of at least 
one amboceptor for a particular cell. As has been shown by numerous 
investigators, this difference is not due to the complements, as these are 
not increased during the process of immunization. Since the presence of 
an amboceptor cannot be demonstrated unless complement is present, 
in testing a serum for an amboceptor we must furnish sufficient comple- 
ment to bring out the maximum activity of the amboceptor. If the serum 
of a rabbit before and after immunization is titrated with sheep erythrocytes, 
it may be found that the immune serum contains from a hundred to many 
thousand times the normal quantity of antisheep amboceptor. 
These facts bear a further practical relation to the treatment of in- 
fectious diseases with bacteriolytic serums. Ordinarily, when we inject 
an immune serum we furnish but one bactericidal substance, namely, the 
bacteriolytic amboceptor, and no complement at all. If the patient’s com- 
plement is decreased or at least insufficient to activate the amboceptor 
furnished, lysis will not occur, and accordingly an increased therapeutic 
effect may be secured by the injection simultaneously of an immune serum 
and a fresh normal serum. This procedure presents certain difficulties 342 
CYTOLYSINS 
and the subject is considered more fully in the chapter on Passive Immuni- 
zation. 
Antiamboceptors.—Just as antiagglutinins and antiprecipitins may be 
formed, so antiamboceptors may be produced experimentally by immuniz- 
ing an animal with an amboceptor-laden serum. An antiamboceptor is 
specific for the amboceptor that caused its production, and when these are 
mixed the activity of the amboceptor is impeded by the antiamboceptor, 
which unites with its cytophilic group. It is possible that old erythrocytes 
are destroyed by an autohemolysin present in the blood-stream under 
normal conditions, and that a physiologic equilibrium is maintained through 
the production of an antiamboceptor. 
COMPLEMENT OR ALEXIN 
Historic.—As early as 1876 Landois described the hemolytic action of 
fresh blood-serum upon the blood-corpuscles of animals of certain species. 
Traube and others observed that animals could withstand the injections 
of relatively large amounts of septic material, but it was not until 1886- 
90 that Fodor,1 Nuttall,2 Buchner,3 and others fully established the 
bactericidal properties of fresh blood-serum. Buchner demonstrated the 
fact that the active principle causng bacteriolysis or hemolysis is very 
labile, and can be inactivated by a temperature of 55° C., by dialysis, or 
by dilution with distilled water. He designated the active principle “alexin,” 
which means “protective substance.” 
Subsequently, in 1899, Bordet found that the alexin of Buchner was 
composed of two distinct substances—one a sensitizing substance, which 
is thermostabile, and a second, the thermolabile substance. Somewhat 
later (1899) Ehrlich and Morgenroth confirmed these observations, but 
applied the name “amboceptor” to the sensitizing substance and “com- 
plement” to the alexin. These terms are most widely employed at the 
present time. Bordet adheres to the term “alexin,” meaning thereby the 
thermolabile principle, and does not use it in the original sense of Buchner, 
which included both the sensitizing substance and the alexin. Metchni- 
koff’s cytases are practically the same as Ehrlich’s complement and Bordet’s 
alexin. 
Definition.—Complement or alexin [Lat., complementum, that which 
completes] is the substance, present alike in normal and in immune serum, 
which is destroyed by heating to 55° C., and which acts with an amboceptor 
or sensitizer to produce lysis. 
As mentioned in the discussion on amboceptors, the complement is the 
active lytic substance concerned in the phenomenon of cytolysis, but is 
powerless until united with the cell, corpuscle, or bacterium by means of 
the interbody or amboceptor, that is, with a sensitized antigen. 
Structure and General Properties of Complement.—Complement is 
ordinarily not attached to the body cells and is free in the blood-serum. 
According to Ehrlich, complement is simple in structure, and is composed 
of a haptophore portion for union with the complementophil haptophore of 
an amboceptor, and a second toxic or lytic portion, called the cytolytic 
group. In other words, the theoretic structure is similar to that of a toxin, 
although the function and action of the two are quite different. 
As will be discussed later, various investigators have claimed that com- 
plement may be split or fractionated into two portions, the midpiece being 
1 Deutsch. med. Wchn., 1886, 617. 
2 Ztschr. f. Hyg., 1888, 4, 353. 
3 Arch. f. Hyg., 1890, 10, 84. COMPLEMENT OR ALEXIN 
343 
identified with the haptophore portion or that which unites with a sensi- 
tized antigen, and the end piece with the active or cytolytic portion. 
Effect of Heat Upon Complement; Inactivation of Complement; Comple- 
mentoids.—The extreme sensitiveness of complement to heat is one of its 
prominent characteristics; as a general rule, heating a fresh serum in a water- 
bath at 55° C. for ten or fifteen minutes results in the complete inactiva- 
tion of hemolytic complement, although heating for thirty minutes is com- 
monly employed for this purpose. 
Complement deteriorates rapidly after bleeding unless the serum is 
frozen. At ordinary room temperature deterioration is rapid within twenty- 
four hours, and generally complete within a few days. 
When a serum contains complement it is said to be active and this must, 
under ordinary circumstances, be a fresh serum. On heating or exposure, 
the serum becomes inactivated; an inactivated serum may be reactivated 
by the addition of fresh serum. 
Inactivation of hemolytic complement is commonly believed to be due 
to inactivation of the cytolytic or active portion of the molecule which 
has been designated by Ehrlich as complementoid. 
Inactivation by heating, shaking, or standing is apparently not an 
irreversible process as commonly believed. Gramenitski1 has observed a 
gradual return to an active condition after moderate heating, the following 
protocol being one. published by him showing the effect of heating at 56° C. 
upon 1 : 10 complement and tested against sensitized beef corpuscles: 
Time After Heating at Which Test 
Was Made. 
Hemoglobin Gone Into Solution After 
Per cent., ten 
minutes. 
Per cent., 
twenty minutes. 
Per cent., 
thirty minutes. 
Per cent., 
forty minutes. 
Seven minutes 
20 
40 
70 
One and a half hours 
30 
60 
80 
Twenty-four hours 
20 
70 
80 
100 
Forty-eight hours 
10 
40 
70 
The largest amount of reactivated complement seemed to be present 
after twenty-four hours, followed by gradual deterioration. Gramenitski 
has explained this phenomenon on the basis of Traube’s2 work, showing 
that when a serum is heated at 56° C. there is a reduction of surface tension 
due to alteration of colloidal condition, that is, an aggregation of particles 
which, if not carried too far, may be reversible and followed by dispersion 
and reactivation as the serum is kept. I have been able to confirm the 
essential particulars of Gramenitski’s work on this interesting phenomenon. 
Since complementoids have their haptophore groups intact they will 
unite with amboceptors and to some extent prevent lysis by blocking the 
active complement just as toxoids unite with antitoxin and agglutinoids 
with their antigens. 
Effect of Acids, Salts, and Filtration Upon Complement.—Hemolytic 
complement is extremely susceptible to the effects of acids and alkalies; 
for this reason all glassware employed in tests employing complement 
must be scrupulously clean. Nolf,3 von Lingelsheim,4 Hektoen and Rue- 
1 Biochem. Ztschr., 1912, 38, 504. 
2 Ztschr. f. Immunitatsf., 1911, 9, 246. 
3 Ann. d. l’lnst. Pasteur, 1900, 14, 297. 
4 Ztschr. f. Hyg., 1901, 37. CY TOLY SI NS 
344 
diger,1 Manwaring,2 von Dungern and Herschfeld3 have shown that many 
inorganic salts in small amounts destroy or inactivate complement, and 
Cumming,4 Brown and Kolmer,5 Sherwood,6 and others, have shown the 
extreme destructiveness of acids and alkalies. 
The manner in which this inactivation or destruction is brought about 
is obscure. Strangely enough the inactivation by salt may be temporary, 
that is, full activity is restored when the solution is reduced to isotonicity, 
as shown by Muir and Browning.7 These investigators have also shown 
that complement inactivated by the addition of 5 per cent, sodium chlorid 
passes through Berkefeld filters, whereas plain complement serum is held 
back, probably by a process of absorption. Kyotoku, in my laboratory, 
has also shown that hemolytic complement of guinea-pig-serum is not filter- 
able if fresh, clean filters are employed, until a large amount of serum has 
been passed, confirming the work of Muir and Browning. Early investi- 
gations by Ehrlich and Morgenroth,8 and Vedder9 were to the effect that 
some complements were filterable and others not. In all probability a great 
deal depends upon the kind of filter employed and the influence of sodium 
chlorid upon inactivation and filterability of complement is probably one 
of colloidal dispersion. 
Effect of Shaking Upon Complement.—Jacoby and Schutze10 have shown 
that guinea-pig-serum complement is readily destroyed or inactivated by 
shaking and have pointed out the possible influence of this factor upon com- 
plement-fixation tests. Ritz,11 Kashisabara,12 Schmidt,13 and others, have 
confirmed these observations. Noguchi and Bronfenbrenner,14 however, 
found that shaking for one hour at 37° C. had almost no influence and 
conclude that the several shakings necessary for complement-fixation tests 
do not appreciably influence the reactions. 
Anticomplements.—Ehrlich and Morgenroth have claimed that these 
may be obtained by immunizing suitable animals with serums that con- 
tain complement, or complementoid. When an inactivated anticomplement 
serum is mixed with the homologous complement, the haptophores of the 
latter are bound by means of the haptophores of the anticomplements. 
A proof of this union lies in the fact that a complement serum that has 
been treated with its specific anticomplement is no longer able to activate 
an appropriate amboceptor. 
According to Gay, the production of anticomplements is only apparent; 
he explains the loss of complement activity wdien a fresh serum and its 
antiserum are mixed as due to the absorption of complement in the pre- 
cipitate which forms, although the latter may be invisible. 
Anticomplements may be of practical importance owing to the forma- 
tion of auto-anticomplements. The complements must exercise an important 
function, not only in the destruction of bacteria, but also in the digestion 
and solution of all kinds of foreign albuminous bodies that enter the organ- 
ism. As was shown by Wassermann, anticomplements may so bind up 
their complements as to render their host much less resistant to certain 
infectious diseases. The spontaneous development of auto-anticomplement 
in an animal has never been demonstrated, as there are no receptors in an 
organism of the complements of the same organism. The injection of the 
1 Jour. Infect. Dis., 1904, 1, 379. 
2 Jour. Infect. Dis., 1904, 1, 112. 
3 Ztschr. f. Immunitatsf., 1911, 10, 131. 
4 Jour. Infect. Dis., 1916, 18, 151. 
8 Amer. Jour. Syph., 1919, 3, 8. 
6 Jour. Infect. Dis., 1917, 20, 185. 
7 Jour. Path, and Bact., 1909, 13, 76. 
8 Berl. klin. Wchn., 1900, Nr. 31, 682. 
9 Jour. Med. Research, 1903, 9, 475. 
10 Ztschr. f. Immunitatsf., 1910, 4, 730. 
11 Ztschr. f. Immunitatsf., 1912, 15, 145. 
12 Ztschr. f. Immunitatsf., 1913, 17, 21. 
13 Ztschr. f. Immunitatsf., 1913, 19, 373. 
14 Jour. Exper. Med., 1911, 13, 229. COMPLEMENT OR ALEXIN 
345 
serum of another animal containing complements that are almost identical 
may, however, lead to the formation of an auto-anticomplement in the 
serum of the immunized animal. 
Source of Complement.—Despite a great amount of investigation upon 
this subject, owing to its interest and importance, the source or sources of 
complement production remains in doubt. 
Hankin,1 and shortly afterward Kanthack and Hardy,2 advanced the 
hypothesis that complement was a secretory product of the eosinophil 
leukocytes, but this theory could not be supported by solid arguments or 
experimental data. A similar theory was proposed by Buchner,3 who held 
that it is not the eosinophils only that secrete the alexins, but the leuko- 
cytes in general. Later Hahn,4 Schattenfroh,5 Laschtsckenko,6 and Tromms- 
dorff7 sought to confirm this theory by exact experiments and the sum total 
of their work justifies one in believing that the living leukocytes are at 
least one source of complement production. 
Metchnikoff,8 however, denied that living leukocytes secrete this sub- 
stance and maintained that alexin or complement was not present in the 
plasma and was produced by leukocytes and other body cells only upon 
leukocytic injury and disintegration or phagolysis. He wras led to adopt 
this view by reason of the experiments of Gengou,9 showing that alexin or 
complement was not present in plasma. The work of Hahn,10 and more re- 
cently and especially that of Hewlett,11 Lambotte,12 Addis,13 Dick,14 and Wata- 
nabe15 has shown, however, quite conclusively that complement is present 
in plasma where its presence is probably due to continual disintegration 
of leukocytes and liberation of complement during life. It is apparently 
increased to a slight extent, as serum is left in contact with the blood-clot 
as shown by Walker,16 Gurd,17 Kolmer,18 and others, indicating that disinte- 
gration of leukocytes may augment the complement supply. Gengou,19 how- 
ever, has recently shown that the endolysins or bacteriolytic substances ob- 
tained by the disintegration of leukocytes, do not have the properties of 
alexin or complement. It is highly probable that leukocytes contain or 
elaborate both as separate entities. 
The failure to obtain definite proof of the origin of complement in the 
leukocytes has led to search for the source in various organs and especially 
the liver. Morgenroth and Ehrlich20 observed a diminished amount of com- 
plement in dogs subjected to phosphorous poisoning and Nolf, by means 
of extirpation experiments with rabbits, found an extreme reduction. These 
results were confirmed by Muller,21 but Liefmann,22 who repeated Muller’s 
experiments, obtained negative results and especially with frogs. Dick,23 
however, as a result of experiments of this nature concluded that comple- 
1 Centralbl. f. Bakteriol., 1892, 12, 777, 809; ibid., 1893, 14, 852. 
2 Proc. Roy. Soc. Med. London, 1892, lii, 267. 
3 Munch, med. Wchn., 1894, 717 and 1897. 
4 Arch. f. Hyg., 1895, 25, 105; ibid., 1897, 28, 312. 
5 Arch. f. Hyg., 1897, 31, 1; ibid., 1899, 35, 135. 
6 Arch. f. Hyg., 1900, 37, 290. 
7 Arch. f. Hyg., 1901, 40, 382. 
8 Immunity in Infective Diseases, 1905, 190 et seq. 
9 Ann. d. l’Inst. Pasteur, 1901, 15, 232. 
10 Arch. f. Hyg., 1895, 25, 105. 
11 Arch. f. exper. Path. u. Pharmakol. 1903, 49, 307. 
12 Centralbl. f. Bakteriol., 1903, 34, 453. 18 Amer. Jour. Syph., 1919, 3, 407. 
13 Jour. Infect. Dis., 1912, 10, 200. 19 Ann. l’Inst. Pasteur, 1921, 35, 497. 
14 Jour. Infect. Dis., 1913, 12, 111. 20 Berl. klin. Wchn., 1900, 37, 683. 
15 Jour. Immunology, 1919, 4, 77. 21 Centralbl. f. Bakteriol., 1911, 57, 577. 
16 Jour. Hyg., 1903, 3, 52. 22 Weichhart’s Jahresbericht, 1912, 8, 155. 
17 Jour. Infect. Dis., 1912, 11, 225. 23 Jour. Infect. Dis., 1913, 12, 111. 346 
CYTOLYSINS 
ment is formed in the liver or is dependent upon this organ for its presence 
in the blood. Fassin1 found that extirpation of the thyroid gland was 
followed by a decrease of complement, while the subcutaneous injection of 
the extract to dogs and rabbits was followed by a rapid increase; Dick also 
observed a decrease after thyroidectomy. These results, however, are not 
conclusive, as it may well be that the thyroid is concerned in stimulating 
the production of complement by other tissues without itself being an 
important source of origin. 
Multiplicity of Complements.—Ordinarily a fresh serum, such as that 
of the guinea-pig, will furnish complement for either bacteriolytic or hemo- 
lytic amboceptors, and the question arises as to whether one complement 
unites equally well with all amboceptors, or whether several complements 
are present in one serum that act more or less specifically with different 
amboceptors. 
The question is whether one and the same serum may contain more 
than one alexin or complement and not whether the alexins or complements 
in the sera of different animals are functionally identical, as it is well known 
that the complements of different animals of the same and different species 
vary in their power to activate bactericidal and hemolytic systems. 
Bordet2 believes that only one complement is present, and bases this 
opinion mainly on the fact that a complement that can be shown to acti- 
vate either a hemolytic or a bacteriolytic amboceptor may be absorbed 
out of a serum by furnishing an excess of either amboceptor. 
Metchnikoff3 maintains that there are two cytases or complements, 
one being derived from macrophages and mainly hemolytic, and the second 
derived from microphages and chiefly bacteriolytic. 
Ehrlich and Morgenroth,4 Sachs,5 Wassermann,6 Wechsberg,7 and the 
German school in general believe that many different complements are 
present in amounts varying with the different serums. These observers 
have sought to prove this experimentally, and while the evidence is not 
absolutely convincing, because of the difficulty of working with substances 
that are so labile, yet the doctrine of the multiplicity of complements is 
quite generally accepted on the basis of such data, as follows: 
(o) By digesting 20 c.c. of fresh goat-serum that was found to activate 
different hemolytic amboceptors with 3 c.c. of a 10 per cent, solution of 
papain in the incubator for from thirty to forty-five minutes, it was found 
that the complement for one amboceptor was destroyed, whereas those 
remaining were left intact or but slightly impaired. 
(b) By treating 10 c.c. of this goat-serum with 1 c.c. of a 7 per cent, 
solution of soda for an hour it was found that some complements were 
destroyed and others were weakened. 
(c) By sensitizing different blood-cells with homologous amboceptors 
and adding these to a fresh serum for short and varying periods of time, 
some complements could be absorbed, whereas others would be left be- 
hind with undiminished or but slightly decreased activity. Prolonged 
exposure would remove all complements. 
(<d) As was previously stated, anticomplements may be produced by 
immunizing an animal with the complement of an animal of a different 
1 Compt. rend. Soc. de biol., 1907, 62. 
2 Ann. d. l’lnst. Pasteur, 1900, 14, 257. 
* Immunity in Infect. Dis., 1905, 185 et seq. 
4 Berl. klin. Wchn., 1900, 453, 677. 
6 Berl. klin. Wchn., 1902, No. 21. 
8 Ztschr. f. Hyg., 1901, 37 
7 Ztschr. f. Hyg., 1902, 39. COMPLEMENT OR ALEXIN 
347 
species. The anticomplements appear to be specific for the complements 
responsible for their production, and by means of these anticomplements 
different complements may be demonstrated in one serum. Since the 
formation of anticomplements would depend upon whether or not the 
body cells of the immunized animal possess suitable receptors for the various 
complements, in a series of animals it may be found that one does not pro- 
duce anticomplements for all the complements injected, a finding that 
would tend to support the theory of the multiplicity of complements. In 
addition, Marshall and Morgenroth actually found in ascitic fluid an anti- 
complement for at least one of two complements present in guinea-pig 
serum. 
(e) By passing normal goat serum through Pukall filters, a complement 
for sensitized guinea-pig cells passed, whereas a complement for sensitized 
rabbit corpuscles was withheld. 
By these and other experiments, including careful heating, Ehrlich and 
his school sought to prove that many different complements are to be found 
in normal serum. Neisser1 was among the first to adopt the view on the 
basis of his experiments that treatment of rabbit-serum with anthrax bacilli 
removed bacteriolytic complement, but not a complement acting upon 
sensitized goat and sheep corpuscles. Wilde,2 however, refuted these re- 
sults and showed that if a sufficient excess of bacilli were employed all of 
the complement could be removed; Bordet3 observed exactly similar re- 
sults when serum was treated with an excess of sensitized corpuscles, that 
is, all of the alexin or complement was absorbed. 
These experiments go to show that complements differ in this respect 
at least: that not all have identical haptophores. Whatever differences 
between complements exist must be slight; probably the cytophilic group 
of all are alike. At present the subject has more theoretic than practical 
importance. In the various diagnostic reactions guinea-pig-serum ordi- 
narily furnishes the complement for hemolysin, bacteriolysin, or other 
cytolysins, and in the therapeutic administration of bacteriolytic serums 
we are compelled in any case to depend for activation of the amboceptor 
upon the natural complement in the patient’s serum. 
Variation in Amount of Complement in Serum.—The complement con- 
tent of the sera of different animals varies within wide limits, and particu- 
larly that complement acting with sensitized blood-corpuscles, as shown 
by Noguchi and Bronfenbrenner,4 Kolmer, Matsunami, and Trist,6 and 
others. For example, the average human complement is five to seven 
times less hemolytic for sensitized sheep corpuscles than the average guinea- 
pig complement, and the sera of the white rat, rabbit, swine, sheep, and 
ox show similar wide variations. 
Even among animals of the same species slight variations in the com- 
plement content of the sera have been found, and, indeed, in the serum of 
the same person or lower animal at different times during the day. 
In addition to these quantitative variations in the complement content 
of normal sera, widely different qualitative variations are known to occur 
and especially in relation to the complement-fixation test. For these reasons 
it is customary to use a mixture of sera for complement in conducting these 
tests in order to reduce the chances of error. 
Apparently wide variations in the complement content of human sera 
occur in disease. These changes have been largely studied by Gunn6 and 
1 Deutsch. med. Wchn., 1900, 790. 
2 Berl. klin. Wchn., 1901, No. 34. 
* Ann. d. l’lnst. Pasteur, 1901, 15, 303. 
4 Jour. Exper. Med., 1911, 13, 78. 
5 Amer. Jour. Syph., 1919, 3, 407. 
6 Jour. Path, and Bacteriol., 1914-15, 19, 155. 348 
CYTOLYSINS 
others by titrating the serum with sensitized blood-corpuscles on the basis 
of Bordet’s work, that with an excess of sensitizer the total complements 
(bacteriolytic as well as hemolytic) may be measured. In typhoid fever 
Gunn observed that the amount of complement bore some relation to the 
severity of the disease, being least or weakest in patients who are very ill. 
Similar studies were made in erysipelas, diphtheria, and scarlet fever. 
The Nature of Complements.—The true nature of the complements is 
unknown. In many respects they bear a resemblance to ferments, and 
certainly the part they play in the processes of cytolysis suggests a ferment- 
like activity. Buchner,1 Bordet,2 Ehrlich and Morgenroth,3 and other of 
the pioneer investigators have subscribed to this view; Metchnikoff be- 
lieved that the evidence indicated that alexin is one of the numerous intra- 
leukocytic soluble ferments which passes into the fluids as the result of 
rupture or of damage to the phagocytes. Liefmann,4 who has studied the 
nature of complement very extensively, has concluded that most evidence 
supports the ferment hypothesis on the nature of complement. The chief 
objection to this view has been based upon the fact that complement, un- 
like the enzymes, is used up during cytolysis and that a definite quantita- 
tive relationship exists between the complement and sensitized cells upon 
which it acts. Recent experiments by Kiss,5 however, have shown that 
this quantitative relationship is not as strict as formerly supposed, and 
that the action of complement depends very largely upon its concentration, 
which strengthens the hypothesis of the ferment nature of complements. 
They differ from true proteolytic ferments such as trypsin in not digesting 
the stroma of corpuscles, although recent work by Dick6 would seem to 
indicate that proteolysis actually occurs, a process that increases the per- 
meability of the cell and permits the escape of hemoglobin. 
On the other hand, it is possible that the nature and action of -comple- 
ments may be placed upon a chemical basis. Following the discovery of 
the hemolytic power of cobra venom by Flexner and Noguchi,7 a power 
they ascribed to the presence of an amboceptor in the venom acting with 
serum complement, Kyes8 found that the amboceptor may be activated 
not only by a complement in the blood-serum but also by some constituent 
of the red blood-corpuscles themselves. This last observer speaks of the 
latter as endocom-plement, i. e., endocellular complement. 
In attempting to discover the nature of this endocomplement various 
substances existing normally in the erythrocytes, such as cholesterin and 
lecithin, were obtained in a pure state and their activating powers for cobra 
amboceptors tested. These investigations showed that lecithin has an 
activating power, whereas cholesterin is antihemolytic. Although all 
erythrocytes contain lecithin, yet all are not equally susceptible to the 
action of venom amboceptors, which is probably due to the fact that the 
lecithin in the cells of some animals is bound to other cell constituents in 
a loose way and is thus available as complement. In syphilitic infection 
the lecithin content of the erythrocytes is actually diminished or in some 
manner rendered less available, so that the inhibition or absence of venom 
hemolysis is characteristic of this infection. 
1 Munch, med. Wchn., 1900, 1193. 
2 Ann. d. l’lnst. Pasteur, 1898, 12, 688; ibid., 1899, 13, 273. 
3 Berl. klin. Wchn., 1899, 6, 481. 
4 Ztschr. f. Immunitatsf., 1913, 16, 503. 
5 Ztschr. f. Immunitatsf., 1909, 3, 558. 
6 Jour. Infect. Dis., 1913, 12, 111. 
7 Jour. Exper. Med., 1902, 6, 277. 
8 Berl. klin. Wchn., 1902, Nos. 38, 39; ibid., 1903, Nos. 2-4. COMPLEMENT OR ALEXIN 
349 
Kyes was able to obtain the union of cobra amboceptor and lecithin, 
forming what is known as cobra lecithid. Although lecithin is an unstable 
substance and is difficult to obtain free from fatty acids and soap, there 
is little doubt that Kyes’ lecithid is a phosphatid compound and is 
actively hemolytic after all traces of fatty acids have been removed. 
The next important observations were made by Noguchi,1 who found 
that soap isolated from blood and various tissues possessed active hemo- 
lytic properties. The salts of the fatty acids, and particularly of oleic acid, 
were found to possess similar hemolytic properties. Pure soluble oleates 
mixed with serum were found to produce compounds possessing many 
of the characteristics of true complements: (1) They are inactivated by 
heating to 56° C. for half an hour; (2) they are inactive at 0° C.; (3) the 
addition of acids, alkalies, and yeast renders them inactive. Von Lieber- 
mann,2 as the result of his own experiments, came to practically the same 
conclusions, and advanced the hypothesis that the complements of the 
blood are to be sought for in the soaps of the serum; that these soaps are 
united with serum albumin, and are inactive until liberated by the ambo- 
ceptors, when they become actively hemolytic. Objections were soon 
raised, however, by Hecker,3 Friedeman and Sachs,4 and Knaffi-Lenz,5 
whose experiments tended to show that the hemolytic action exerted by 
fatty acids or soap is quite incomparable to true complement action and 
are thermostable. Von Dungern and Coca6 have also shown that cobra 
venom contains a lipoid-splitting ferment which acts upon the lecithin, 
liberating hemolytic substances of a non-specific nature. Sachs and Alt- 
mann,7 Liefmann, Cohen, and Orloff,8 and others are likewise opposed to 
the lipoidal nature of the complements. 
Of further interest in this connection it was shown that oleic acid may 
act as an amboceptor, and when added to an inactive soap-albumin com- 
bination it would render this actively hemolytic. Von Liebermann and 
Fenyvessy9 have shown that a mixture of soap, serum, and oleic acid pos- 
sesses a striking resemblance to complements and amboceptors, and that 
the amboceptor-complement action is much more than a mere linkage of 
complement to antigen by means of an amboceptor. These observers sug- 
gest that an amboceptor may have an affinity for certain constituents of 
the cell or bacterial body, and, on the other hand, act upon the complement 
and separate one of its constituents, which then breaks up the cell. These 
artificial hemolysins, however, completely dissolve the stroma of the cor- 
puscles, whereas the immune hemolysins appear to dissolve out the hemo- 
globin, leaving the stroma undissolved. As has been mentioned elsewhere, 
recent work would tend to show that the stroma is also dissolved, at least 
in part, in specific hemolysis, so that the difference in action between the 
two is not quite so apparent. 
While the simplicity of the substances concerned in these observations 
does not harmonize with the great variety and complexity of the immune 
bodies, nevertheless, as Adami has pointed out, the points of resemblance 
between artificial and natural complements and amboceptor are so strik- 
1 Proc. Soc. Exper. Biol, and Med., 1907, 4, 107; Biochem. Zeitschr., 1907, 6. 
2 Biochem. Zeitschr., 1907, 4; Ztschr. f. Immunitatsf., 1912, 13, 695. 
3 Arb. a. d. k. Inst. f. Exper. Therap., 1907, 3. 
4 Biochem. Ztschr., 1908, 12. 
5 Biochem. Ztschr., 1909, 20. 
6 Berl. klin. Wchn., 1907, 46. 
7 Berl. klin. Wchn., 1908, 10. 
8 Ztschr. f. Immunitatsf., 1912, 13, 150. 
9 Biochem. Zeitschr., 1907, 5; Ztschr. f. Immunitatsf., 1911, 10, 479. 350 
CY TOLY SI NS 
ing that material advances in our knowledge of their nature and action 
may be gained by further researches into the chemistry of immunity. 
Complement Splitting.—The inhibiting effect of hypertonic (5 per cent.) 
solution of sodium chlorid upon the activity of complement and its filter- 
ability has been discussed in preceding sections; mention should also be 
made of the early observations of Buchner and Orthenberger1 to the effect 
that bacteriolysis is inhibited in the complete absence of salts. Some years 
later Ferrata2 showed that this inhibition of cytolysis was due to a failure 
of function on the part of complement and not to any failure of union of 
sensitizers or amboceptors with the antigens. This inactivation of com- 
plement in distilled water was ascribed to disturbances of colloidal equilib- 
rium or to the destruction of complement by a ferment which was believed 
by Sachs and Teruuchi3 to be active in a salt-free medium. 
Ferrata removed the salts by dialyzing for twenty-four hours against 
distilled water. This resulted in “splitting” the serum into a precipitate 
of globulins and the water-soluble albumins. By dissolving the former in 
isotonic saline solution and rendering the latter isotonic by the addition 
of salt, Ferrata found that neither portion was lytic, whereas a mixture 
of the two was actively lytic for sensitized corpuscles. 
These observations upon the splitting of complement were soon con- 
firmed and extended by Brand,4 who suggested the terms “midpiece” 
for the globulin fraction, because it unites with the sensitized antigen, and 
“end piece” for the albumin portion, because it cannot combine directly 
with the amboceptor and antigen, but is bound only after the globulin 
fraction has been previously attached. 
Brand found both portions thermolabile while free, but Michaelis and 
Skwirsky6 showed that the globulin or “midpiece” when united with sensi- 
tized antigen, was no longer affected by heating at 56° C. Hecker6 found 
that while the “midpiece” or solution of globulins in water became inactive 
after three or four hours, yet it was capable of uniting with sensitized antigen, 
and that when “end piece” or albumin fraction was subsequently added 
hemolysis occurred. This phenomenon has not been adequately explained. 
Hecker and, later, Sachs have suggested that “midpiece” or globulin frac- 
tion when inactivated by standing loses its avidity for the sensitized antigen, 
but acquires an increased combining power for the “end piece.” This 
theory, however, lacks proof, and in all probability the true mechanism is 
concerned with colloidal phenomena. 
Of great interest in connection with this subject of complement splitting 
is the question whether complement or alexin in normal serum exists as a 
combination of mid- and end pieces or as separate fractions. A definite 
answer cannot be given. Apparently the two portions may be absorbed 
separately from serum by sensitized blood-cells; Michaelis and Skwirsky, 
for example, have found that while acid reaction inhibits hemolysis, yet 
“midpiece” is bound, but the “end piece” or albumin fraction is unbound. 
They have recommended the use of strongly sensitized corpuscles on an 
acid medium as a method for securing the albumin fraction or “end piece” 
in pure form by removing the “midpiece” after union with sensitized cor- 
puscles by centrifuging. This would indicate the separate existence of these 
complement partitions in serum, but experiments are rendered difficult by 
1 Arch. f. Hyg., 1890, 10, 149. 
2 Berl. klin. Wchn., 1907, xliv, 366. 
3 Berl. klin. Wchn., 1907, xliv, 467, 520, 602. 
4 Berl. klin. Wchn., 1907, xliv, 1075. 
5 Ztschr. f. Immunitatsf., 1909-10, 4, 357, 629. 
6 Arb. a. d. k. Inst. f. Exper. Therap., 1907, No. 3, 39. COMPLEMENT OR ALEXIN 
reason of the fact that traces of “end piece” or the albumin fraction carried 
down with the “midpiece” or globular fraction will activate the latter, 
which would indicate that in the serum both portions are probably present 
in combination or as a complex. To secure the fractions in purer form other 
means of fractionating have been used in preference to the dialyzation 
method. Thus Sachs and Altman1 precipitate with n/300 to n/250 hydrochlo- 
ric acid. Frankel2 dilutes the serum ten times with distilled water and intro- 
duces carbon dioxid. Salting out methods are unsatisfactory because of 
the prolonged dialysis necessary for the removal of the salts. With these 
various methods it has been shown by Liefmann and Cohn,3 Marks,4 and 
others that quantitative factors are of great importance in a study of the 
action of these portions separately and together; all that can be regarded 
as demonstrated is that under certain circumstances complement may be 
separated into at least two parts; that one part largely identified with the 
globulin fraction, acts with the sensitized antigen, rendering it amenable 
to the lytic activity of the albumin fraction. 
This is not, however, accepted by all investigators. 
Bronfenbrenner and Noguchi5 have cast considerable doubt upon these 
views, and have shown that what is known as complement splitting is really 
nothing more than an inactivation of the active principle of complement, 
since both globulin and albumin fractions contain a part of the comple- 
ment, a fact that can be demonstrated by the removal of the inhibiting 
action of the acid or alkali used in the process. 
The Bordet-Gengou Phenomenon of Complement Fixation.—In the 
endeavor to demonstrate the unity of complement Bordet and Gengou6 
devised an experiment that has proved of great practical value in the serum 
diagnosis of syphilis and other infectious diseases. By mixing bacteria 
and their amboceptors with a little fresh serum containing complement 
and letting the mixture stand aside for an hour or so it was found that, 
upon the addition of corpuscles and their amboceptor, hemolysis did not 
occur, although the serum that had been used as complement was capable, 
in its original condition, of producing hemolysis of these corpuscles. Bordet 
advanced this experiment to show that the complement concerned in bac- 
teriolysis is the same as that at work in hemolysis, and consequently con- 
cluded that there is but one single complement. 
This experiment of Bordet is usually spoken of as the “Bordet-Gengou 
phenomenon,” and is now used extensively in determining whether or not 
a given serum contains certain amboceptors. The serum to be tested is 
first inactivated, treated with the antigen composed of an emulsion of the 
bacterium whose amboceptor it is desired to discover, and then mixed 
with a small quantity of a fresh normal complement serum. The mixture 
is placed in the incubator for an hour, during which time the bacterial 
antigen unites with its amboceptor, and the complement, i. e., fixes the 
complement, so that when red blood-cells previously sensitized with heated 
hemolytic serum are added, hemolysis does not occur because the comple- 
ment in the fresh serum, which was suitable for lysis of the sensitized cor- 
puscles, has been “fixed” by the bacteria by reason of the presence of specific 
amboceptors in the serum tested (Fig. 117). If these amboceptors were 
351 
1 Berl. klin. Wchn., 1908, xlv, 699. 
2Ztschr. f. Immunitatsf., 1911, 8, 781; ibid., 1911, 10, 388. 
3 Ztschr. f. Immunitataf., 1910, 6, 562; ibid., 1910, 7, 699. 
4Ztschr. f. Immunitatsf., 1911, 8, 508; ibid., 1911,11,18. Jour. Exper. Med., 1911, 13, 
590. 
5 Jour. Exper. Med., 1912, 15, 598, 625. 
6 Ann. d. l’lnst. Pasteur, 1901, 15, 289. 352 
CYTOLYSINS 
not present, then the complement would remain unfixed and be free to 
hemolyze the sensitized corpuscles, a negative reaction being indicated, 
therefore, by hemolysis, whereas the absence or inhibition of hemolysis 
indicates a positive reaction. A more complete description of this interest- 
ing phenomenon is given in Chapter XXIII. 
Wassermann, Neisser, and Bruch have applied this test to the serum 
diagnosis of syphilis, the technic of the test being considered in a subse- 
quent chapter. 
The Neisser-Wechsberg Phenomenon of Complement Deviation.— 
This is frequently confused with the phenomenon of complement fixation 
Fig. 117.—Mechanism of Complement Fixation. 
Tube 1 shows the hemolytic system: C, a red blood-corpuscle; A, a hemolytic amboceptor; Cp, 
complement; C.A.C., complement united to a corpuscle by means of the specific amboceptor. Hemol- 
ysis results. 
Tube 2 shows complement fixation by bacterial antigen and amboceptor: A, antigen; C, com- 
plement united to the antigen A, by the amboceptor A. When hemolytic amboceptors are added 
hemolysis does not occur because the complement has been previously fixed by the bacterial antigen 
and amboceptor. 
Tube 3 shows absence of complement fixation because the bacterial amboceptor A, is not specific 
for the bacterial antigen A2, and hence complement is not fixed; when hemolytic amboceptor and the 
corresponding corpuscles are added complement unites with these, C.A.C., and hemolysis occurs. 
just described. As previously mentioned in a discussion of the mechanism 
of cytolysis, Ehrlich and his associates believed that complement and ambo- 
ceptor may unite direct, which is in sharp contrast with the more generally 
accepted views of Bordet that complement or alexin is bound only by the 
antigen-sensitizer complex. Support for the theory of Ehrlich was supplied 
by Neisser and Wechsberg1 who demonstrated that in bactericidal tests an 
excess of amboceptors may reduce the degree of bacteriolysis. They ex- 
plained this phenomenon by stating that the amboceptors absorbed a por- 
tion of the free complement and thus prevented this portion from com- 
1 Miinch. med. Wchn., 1901, 697. COMPLEMENT OR ALEXIN 
bining with those amboceptors united with the bacteria, i. e., the comple- 
ment has been deviated or deflected from its natural course. In other 
words, that complement fixed by antigen and amboceptors may be said 
to be an example of complement fixation, whereas that fixed by ambo- 
ceptors alone is an example of complement deviation. 
There can be no doubt about the correctness of the fact that in bac- 
teriolytic tests an excess of antiserum (amboceptors) may yield less bac- 
teriolysis than smaller amounts of antiserum. This will be illustrated in 
the chapter on Bacteriolvsins. Many investigators have corroborated 
these results. Furthermpre, it is commonly encountered in quantitative 
complement-fixation tests, as the Wassermann test, employing varying 
amounts of serum. For example, the degree of complement fixation with 
0.1 c.c. serum of a syphilitic may be less than occurs with 0.02 c.c. 
But the explanation offered by Neisser and Wechsberg cannot be ac- 
cepted. There is no experimental data to support the statement upon 
which it is based, namely, that complement and amboceptors unite direct. 
Rather all data supports the contention of Bordet that complement unites 
with amboceptors only when the latter have previously united with the 
antigen. Analogous phenomena are commonly observed in the agglutina- 
tion and precipitin reactions when large amounts of immune sera produce 
less agglutination and precipitation than smaller amounts; these have been 
explained on the basis of the presence of agglutinoids and precipitoids. 
It is highly probable that these phenomena are colloidal reactions. Cer- 
tainly the different constituents are of the nature of colloids and an excess 
of one colloid is known to inhibit the precipitation of a second, a reaction 
taking place only when the reacting bodies are present in definite propor- 
tions. Since this can be shown for agglutination and precipitation it is 
reasonable to assume that similar conditions are operative in complement 
deviation and that the phenomenon is one of colloidal chemistry. 
Mechanism of Cytolysis or the Action of Complement (Alexin) and 
Amboceptor (Sensitizer).—With these general considerations regarding the 
amboceptors or sensitizers and the complements or alexins presented, we 
may now pass in review the opinions held regarding the mechanism of their 
action in the phenomenon of cytolysis. The more important amboceptors, 
as the bacteriolysins and hemolysins, will be taken up in more detail in 
subsequent chapters; likewise the properties of complement or alexin in 
special relation to complement fixation, but the mechanism of their action 
is regarded as identical and may be discussed at this time. 
Baumgarten’s Theory.—One of the earliest theories on the mechanism 
of bacteriolysis was that advanced by Baumgarten,1 who taught that the 
solution of bacterial cells in specific serum was brought about by a change 
in osmotic conditions, that is, the antibodies in the serum uniting with the 
cell membranes rendered the bacteria permeable for salts and other sub- 
stances which brought about swelling, increased intracellular pressure, and 
final destruction. This theory is now only of historical interest, and while 
salts play a role in the phenomenon of cytolysis, they do not occupy the 
important relation assigned them by Baumgarten as shown by the studies 
of von Lingelsheim2 and others. 
Buchner’s Theory.—Buchner looked upon the bactericidal activity of 
normal and immune serum as due to one constituent, namely, alexin (pro- 
tective substance), which he believed was of the nature of a proteolytic 
ferment because of its heat sensitiveness and instability on standing. Later 
353 
1 Lehrbuch der pathogenen Mikroorgan., 1911. 
2 Ztschr. f. Hyg., 1901, 37. CYTOLYSINS 
354 
researches by Bordet showed that this “alexin” was composed of two sub- 
stances, namely, antibody and a normal constituent of serum. Bordet 
adapted the term “alexin” for this latter substance and named the former, 
or antibody, “sensitizer.” 
Ehrlich-Morgenroth Theory.—Following the discoveries of Nuttall, 
Buchner, Bordet, Pfeiffer, and other pioneer investigators in cytolysis, 
Ehrlich and Morgenroth1 gradually evolved a theory of the mechanism 
of the phenomenon based largely upon the results of their own investiga- 
tions and those of their associates by enlarging upon Ehrlich’s side-chain 
theory to the third order of receptors. Briefly this is as follows: 
The antigen (blood-corpuscles, bacteria or other cells) undergoes lysis 
when acted upon by two substances; one of these is the specific antibody, 
heat resistant, and produced during immunization, while the other is a non- 
specific constituent of serum, heat sensitive, and not increased by immuni- 
zation as shown by Bordet2 and von Dungern.3 
To the antibody they gave the name “amboceptor,” believing that it 
had two combining affinities or arms, namely, one for union with the cell, 
and the second for union with the second substance, to which they gave 
the name of “complement.” 
The specific antibody unites with its antigen direct even at 0° C. and 
regardless of the presence or absence of the complement; the antigen, how- 
ever, shows no discernible changes, and in the case of living cells as bacteria, 
are not even killed. Complement must be present and united with the 
antigen-amboceptor complex before lysis can occur. Complement cannot 
unite with the antigen direct; it can do so only through the action of the 
interbody or amboceptor. The non-specific normal constituent of serum, 
namely, the ferment-like complement, is, therefore, the active lytic agent, 
but is powerless to dissolve the antigen until the latter is prepared by the 
specific amboceptors. 
According to this theory complement may unite direct with the ambo- 
ceptors without the antigen; this is how Neisser and Wechsberg explained 
their phenomenon of complement deviation, but, as previously stated, this 
direct union of complement and amboceptor lacks experimental proof. 
Bordet’s Theory.—Bordet, who analyzed Buchner’s “alexin” into the 
two components of thermostabile and specific antibody and thermolabile 
or normal serum constituent, designated the antibody as “sensitizer” and 
the latter as “alexin.” According to this theory4 the antibody unites with 
the antigen (blood-corpuscle, bacterial cell, etc.) direct, sensitizing or pre- 
paring it for the action of the alexin just as a mordant aids in the penetra- 
tion of a dye in the staining of cells. 
When antigen and antibody have united the antigen is regarded as 
“sensitized” or prepared for the lytic action of the alexin or complement. 
In the absence of the latter the sensitized antigen undergoes no appreciable 
changes. 
Alexin or complement cannot unite direct to the antigen; union can 
only occur with sensitized or “antibodyized” antigen. 
Alexin or complement cannot unite direct with the antibody or sensi- 
tizer; in this respect the theory of Bordet takes sharp issue with the theory 
of Ehrlich. This will be discussed a little later in connection with the sub- 
ject of conglutinins. 
1 Studies in Immunity, Ehrlich-Bolduan, 1910. 
2 Ann. d. l’Inst. Pasteur, 1898, 12, 688. 
3 Mttnch. med. Wchn., 1900, No. 20. 
4 Studies in Immunity, Bordet-Gay, 1909. COMPLEMENT OR ALEXIN 
355 
Metchnikoff s Theory.—Metchnikoff1 admits the production of the specific 
antibody (Bordet’s sensitizer or Ehrlich’s amboceptor), and has called it 
“fixator.” He asserted that the amount of antibody produced is propor- 
tional to the degree of phagocytosis and phagolysis that occur during the 
absorption of the antigen. In other words, he has sought to maintain the 
importance of his theory of phagocytosis by asserting that the antibody 
is a product of phagocytic cells, its production running parallel with the 
degree of leukocytosis, and injury during the process of immunization. 
Metchnikoff considers the “fixators” as analogous to enterokinase, and he 
believes that, like the latter, the fixation acts as an accessory digestive 
ferment, having for its object the linking of the more potent ferment, as a 
cytase (alexin or complement), to the cell (antigen) to be digested (lysis). 
Conglutinins in Relation to Theories of Cytolysis; Auxilysins.—As 
stated above, the views of Bordet and Ehrlich differ in regard to the power 
of alexin or complement for union direct to free sensitizer or amboceptor. 
As proof of their belief that this union may occur Ehrlich and Sachs2 have 
described their “Experimentum Crucis” as follows: 
(a) Fresh horse-serum + guinea-pig corpuscles gives slight hemolysis. 
(b) Fresh horse-serum + heated beef-serum + guinea-pig corpuscles 
gives marked hemolysis. Presumably this is due to the presence of anti- 
guinea-pig hemolytic amboceptor or sensitizer in the beef-serum; but if 
the corpuscles and heated beef-serum are first mixed and centrifuged the 
corpuscles do not hemolyze upon the addition of complement, that is, are 
not sensitized. Ehrlich and Sachs argued that this showed that the ambo- 
ceptor in the beef-serum unites first with the complement of the horse- 
serum independently and that this complex then unites with the corpuscles 
and results in hemolysis. If guinea-pig amboceptor is present in the beef- 
serum and fails to be removed when the serum is treated with these cells, 
it would also tend to prove that both Ehrlich’s and Bordet’s views on sensi- 
tization or direct union of antigen and antibody were incorrect, but further 
investigations by Bordet and Gay3 have shown that this is a peculiar phe- 
nomenon to be separated from the mechanism of hemolysis in general. 
They found that guinea-pig cells are hemolyzed very slowly by guinea- 
pig complement or alexin and antiguinea amboceptor or sensitizer; however, 
if heated beef-serum was added agglutination and hemolysis were very 
rapid. They, therefore, advanced the hypothesis that the complex of cells 
+ antibody + alexin (complement) attracts a colloidal substance from 
beef-serum which unites with them and produces two results: Powerful 
agglutination followed by hemolysis; plain cells or cells + antibody cannot 
attract it. 
Bordet and Gay named this substance in beef-serum “bovine colloid.” 
Bordet and Streng4 later suggested the name conglutinin. Streng5 later 
showed the presence of conglutinins in beef-serum for bacteria and also 
that conglutinins were present in the sera of goats, sheep, antelopes, and 
other herbivora, but not in the sera of cats, dogs, guinea-pigs, and birds. 
A further discussion of this substance and the technic of the conglutination 
test are given in the chapter on Agglutinins. 
Manwaring6 has found that normal goat-serum heated to 56° C. for 
three or four hours acquires the property of increasing the hemolytic and 
1 See Immunity and Infective Diseases, 1905. 
2 Berl. klin. Wchn., 1902, No. 21. 
3 Ann. d. l’Inst. Pasteur, 1906, 20, 467. 
4 Centralbl. f. Bakteriol., 1909, 49. 
5 Ztschr. f. Immunitatsf., 1909, 2, 415. 
6 Centralbl. f. Bakteriol., 1906, 42, 75. 356 
CYTOLYSINS 
hemagglutinative properties of goat antisheep hemolysin. He has called 
these substances in the heated goat-serum auxilysins (to increase), and 
apparently these are similar to the conglutinins described above. 
R6le of Cytolysins in Immunity 
The very important relation of certain of the cytolysins, notably those 
for pathogenic parasites of both vegetable and animal origin to resistance 
to disease and recovery from infection, is generally conceded. In general 
terms it may be said that all phases of immunity may be resolved into the 
activities of phagocytes and opsonins, antitoxins and cytolysins, the agglu- 
tinins and precipitins being accessory in their activities to the cytolysins. 
The bacteriolysins found normally in the blood are doubtless one im- 
portant means of natural immunity to certain micro-organisms, and in 
infections are greatly increased, particularly in diseases due to pathogenic 
bacilli. The complements apparently play an important role in physiologic 
processes of metabolism and repair, as also in the processes of natural im- 
munity and recovery from disease. While no means has been discovered 
for bringing about their increase by processes of immunization, their produc- 
tion is known to fluctuate during health and disease, and probably as de- 
mands are made for cytolytic processes. 
A complete discussion of these cytolysins in relation to immunity is 
given in Chapter XXVI; the role of some of the more important members 
of the group, as the bacteriolysins, hemolysins, and cytotoxins, is discussed 
in the separate chapters devoted to these antibodies. CHAPTER XIX 
BACTERIOLYSINS 
Having considered the general nature and properties of amboceptors 
and complements and the mechanism of their action in producing solu- 
tion or lysis of cells, we will now study more closely the bacteriolysins, which 
are antibodies belonging to this group and possessing diagnostic and con- 
siderable therapeutic importance. 
Historic; the Bactericidal Activity of Normal Blood and Serum.—The 
early history of the discovery of the bacteriolysins is closely associated with 
the history of immunity in general, for with the discoveries in bacteriology, 
and the establishment of the relation of bacteria to disease, it followed as 
a matter of course that investigations should be undertaken to ascertain 
the mechanism of resistance to and of recovery from an infection. 
In 1874 Traube and Gscheidlen1 showed that septic material may be 
destroyed in the blood of living animals, and in 1881 Lister demonstrated 
the same phenomenon in extravascular blood. These experiments were 
naturally somewhat crude, as they antedated the period in which the pyo- 
genic micro-organisms were isolated and studied in pure culture, but they 
served, nevertheless, to demonstrate the germicidal powers of the blood. 
In 1886 Fodor2 demonstrated the germicidal action of the defibrinated 
blood of the rabbit upon anthrax bacilli. This work was followed shortly 
after by that of Fliigge3 and Nuttall,4 who showed the germicidal powers 
of the body fluids in general independent of cells. Nuttall was able to 
follow step by step the destruction of anthrax bacilli on the warm stage 
of the microscope by defibrinated rabbit blood. The same phenomenon 
was shown by plating methods. Nuttall also discovered that when blood 
was heated to 55° C. the bactericidal properties were completely destroyed. 
These observations gave Fliigge and his assistant Bitter the opportunity 
to criticize vigorously the theory of phagocytosis, and the controversy be- 
tween the adherents of the cellular and humoral theories of immunity now 
began, as Metchnikoff was actively engaged in studying phagocytes and 
in formulating his phagocytic theory. 
Buchner5 and others took up the subject emphasizing the important 
germicidal powers of the body fluids and ascribing this function to the 
presence of “alexins” (substances that ward off disease). Buchner also 
found that if the serum was heated this germicidal power was lost; hence 
it followed that the active bacteriolytic agent was considered very unstable 
and was quickly destroyed outside of the body. He was able without diffi- 
culty to confirm Nuttall’s discoveries and added to them new facts of great 
value. 
In 1894 Pfeiffer6 demonstrated most clearly the phenomenon of bac- 
teriolysis which gave great encouragement and impetus to studies in im- 
munity, and incidentally strengthened the claims of the humoral theory. 
1 Jahresb. d. schles. Gesellsch. f. vaterl. Kult., Breslau, 1874. 
2 Deutsch. med. Wchn., 1886, 617; ibid., 1887, 745. 
3 Ztschr. f. Hyg., 1888, 4, 208, 223. 
4 Ztschr. f. Hyg., 1888, 4, 253. 
5 Arch. f. Hyg., 1890, 10, 84; Centralbl. f. Bakteriol., 1889, 5, 817; ibid., 6, 561; 1890, 
8, 65. 
6 Ztschr. f. Hyg., 1894, 16, 268; ibid., 1894, 18, 1. 
357 358 
BACTERIOLYSINS 
He showed that cholera vibrios introduced into the peritoneal cavity of a 
guinea-pig that had been immunized against cholera lost their motility 
and finally became disintegrated and passed into solution regardless of the 
presence of cells. 
As stated in the preceding chapter, it remained for Bordet, however, 
to show the mechanism of this interesting phenomenon. This observer 
demonstrated the fact that the thermolabile body was but one substance 
concerned in the reaction, and that the specific substance was thermostabile 
and the actual product of immunization, results that were later corroborated 
and elucidated by his researches upon hemolysis, and by those of Ehrlich 
and his pupils on cytolytic phenomena in general. As previously men- 
tioned, Bordet retained the name “alexin” for the thermolabile substance 
and applied the new term, “substance sensibilisatrice,” to the specific 
thermostabile antibody. Later both substances were renamed by Ehrlich, 
and called “complement” and “amboceptor” respectively. 
As will be pointed out further on, as the result of these observations 
Metchnikoff modified his phagocytic theory, and recognized the existence 
of both substances, which he named “cytases” and “fixateurs,” believing 
that both were derived from cells classed as phagocytes. 
All are agreed as to the presence of two different bodies in the body 
fluids concerned in bacteriolysis, although opinions vary as regards their 
origin and mechanism of action. The side-chain theory has been widely 
accepted in explanation of their action, and the terms applied by Ehrlich 
to the two substances concerned, namely, complements and amboceptors, 
are in general use. 
Definition.—Bacteriolysins are substances present in the serum and other 
body fluids that kill bacteria with or without lysis. 
The term itself would infer that solution or lysis of the bacterium is 
an essential property of an antibody of this order. Bactericidins are sub- 
stances that kill bacteria without lysis, and, strictly speaking, an effort 
should be made to differentiate between the terms, although from a practical 
standpoint this is not important. Certain micro-organisms may be killed 
and resist solution or digestion for a comparatively long time, whereas, on 
the other hand, the same bacteria, under different circumstances, may 
readily be lysed. 
Although the endotoxins liberated from the lysed bacteria may pro- 
duce symptoms of disease, followed by death, yet the bacterium itself 
is usually destroyed and unable to proliferate. A bacteriolysin is, there- 
fore, always bactericidal, although the converse is not necessarily true. 
Custom, however, has never strictly differentiated between the two 
terms, and bacteriolysis appears to be but a continuation of and a more 
nearly complete bactericidal process. Hence the definition just given covers 
both terms—bacteriolysins and bactericidins. 
Origin of Bacteriolysins.—Our knowledge regarding the origin of the 
bacteriolysins is quite fragmentary and has already been discussed in the 
preceding chapter. The investigations of Pfeiffer and Marx in cholera, 
and Wassermann in typhoid, have shown that the spleen and hematopoietic 
organs in general may be especially concerned. 
According to Metchnikoff the “bacterial fixateurs,” which are practi- 
cally the bacteriolysins, are secretory or excretory products of phagocytic 
cells, especially the polynuclear leukocytes or microphages. As early as 
1889 he1 expressed the opinion that a portion at least of the bactericidal 
power might come from substances that had escaped from the leukocytes 
1 Ann. de l’lnst. Pasteur, 1888, 3, 670. LEUKOCIDINS AND LEUKOCYTIC EXTRACTS 
359 
during the preparation of defibrinated blood and of serum. It is commonly- 
believed that during infection the bacteria cause phagolysis or disintegra- 
tion of these cells, with liberation of both complements (cytases) and ambo- 
ceptors (fixateurs), which produce extracellular lysis of the invading bac- 
terium (bacteriolysis). If, on the other hand, the phagocytes are fortified 
and phagolysis is prevented, the bacteria are phagocytized and undergo 
intracellular lysis, a condition that, according to Metchnikoff, may be in- 
duced experimentally by giving an animal an intraperitoneal injection of 
sterile bouillon twenty-four hours before bacteria or other cells are injected. 
Leukocidins and Leukocytic Extracts.—That leukocytes afford a bac- 
teriolytic substance is supported by observations showing that leukocytic 
exudates, secured by the injection of a sterile aleuronat suspension, possess 
a well-marked germicidal activity. Denys and Havet1 were the first to 
show that exudations rich in leukocytes exhibited a bactericidal power 
much higher than that of the corresponding sera. Shortly afterward Buchner2 
observed the same on comparing the bactericidal power of exudations rich 
in leukocytes with the blood-serum of the same animals. Issaeff3 found that 
the intraperitoneal injection of sterile bouillon and other mild irritants, 
by producing a leukocytic exudate that supplied certain bactericidal sub- 
stances and facilitated phagocytosis, increased the resistance of animals to 
bacterial infection. At one time surgeons made practical use of this ob- 
servation by injecting nucleinic acid and other substances into the peritoneal 
cavity before performing laparotomy in order to induce a local resistance 
to a possible infection. 
The bactericidal substance contained within leukocytes has been exten- 
sively studied by Schattenfroh,4 Petterson,5 Hiss,6 Kling,7 and others. It 
has been observed that when leukocytes are suspended in diluted blood- 
serum the bactericidal properties of the serum are increased without coin- 
cident destruction of the cells, showing that the leukocytes may secrete 
germicidal substances into the fluid. The same observation has been made 
with Bier’s congestive lymph, indicating that this activity takes place both 
in the test-tube and in living tissues. 
Watabiki8 has questioned the bactericidal activity of the endolysins 
or leukins of leukocytes. He could find no evidence of bactericidal activity 
of extracts of rabbit leukocytes for such bacteria as typhoid, cholera, colon, 
and dysentery bacilli, nor in extracts of spleen, liver, and bone-marrow of 
normal rabbits. Manwaring9 found that the active antibacterial substance 
of leukocytes resembled certain enzymes which can be isolated and purified 
bv alcoholic precipitation. This phase of the subject is considered more 
fully in Chapter XXXIX. 
Hiss10 and, later, Hiss and Zinsser11 found that autolyzed leukocytic 
exudates possess some bactericidal activity, and that they may profoundly 
modify experimentally induced infection of rabbits and guinea-pigs with 
the pneumococcus, staphylococcus, streptococcus, and other bacteria. 
In applying this method of treatment to man by means of subcutaneous 
1 La Cellule, 1894, 10, 7. 
2 Miinch. med. Wchn., 1894, 717. 
3 Ztschr. f. Hyg., 1894, 16, 287. 
4 Arch. f. Hyg., 1897, 28, 135. 
5 Ztschr. f. Immunitatsf., 1908, 1, 52; ibid., 1910, 7, 693. 
6 Jour. Med. Research, 1908, 19, 323. 
7 Ztschr. f. Immunitatsf., 1910, 7, 1. 
8 Jour. Infect. Dis., 1909, 6, 319. 
9 Jour. Exper. Med., 1912, 16, 249. 
10 Jour. Med. Res., 1909, 20, 245. 
11 Jour. Med. Res., 1910, 22, 397. 360 
BACTERIOLYSINS 
injections these investigators observed distinctly beneficial results in cases 
of epidemic cerebrospinal meningitis, lobar pneumonia, staphylococcus in- 
fections, and erysipelas (see Chapter XL). 
Gengou1 has recently reviewed this subject, and studied the relationship 
of these leukocytic bacteriolysins to alexin (complement). He found that 
a solution of washed polymorphonuclear leukocytes, made by dissolving the 
cells in centinormal HC1 and subsequently neutralizing the solution, has 
active bacteriolytic properties. Gengou thinks that this extract contains 
the substances which dissolve bacteria in the interior of the leukocytes after 
phagocytosis. The properties of these substances extracted from leukocytes 
were as follows: 
1. The extract is bacteriolytic for a number of different bacteria, show- 
ing no specificity. 
2. The range of temperature through which the substances act is very 
broad, from 20° to 60° C. These substances are quite resistant to heat. 
3. The bacteriolytic substances from leukocytes retain their activity in 
neutral and acid fluids for many days. 
4. The extracts, while energetically bacteriolytic, are not hemolytic, and 
they do not cause hemolysis of red cells sensitized with amboceptor. As 
the addition separately of the end-piece and mid-piece of complement does 
not cause the extract to take any part in the specific hemolytic reaction, it 
is concluded that the extract lacks both those parts of complement. 
5. These substances are not simply activated by the acid used to dis- 
solve the cells. They are themselves dissolved in an acid solution at first, 
but later cause bacteriolysis only in a neutral fluid. 
6. These extracts are proteolytic, in that they liquefy gelatin. The 
proteolytic properties, however, are less resistant to heat than the bacterio- 
lytic, and act best at different ranges of temperatures. Some actively bac- 
teriolytic extracts are devoid of demonstrable proteolytic action. 
7. The characteristic effect of these extracts upon bacteria is a solution 
or granular disintegration of the micro-organisms. 
Gengou concludes that the alexin or complement of serum is different 
from these bacteriolytic substances which he has extracted from leukocytes. 
While he has found that these endolysins were non-hemolytic, Carrel and 
Ebeling,2 on the other hand, found that cultures of leukocytes in vitro in 
hen serum sometimes elaborated a substance hemolytic for sheep and rabbit 
corpuscles, whether this substance was an alexin, a sensitizer, or a complex 
sensitizer alexin, as described by Hyde,3 was not determined. At any rate 
it is fairly well established that leukocytes are one source of hemolysins as 
well as of other cytolysins; this subject will be discussed in more detail in 
the following chapter. The experiments of Carrel and Ebeling also showed 
that leukocytes, cultivated in plasma, always secrete substances which in- 
crease the rate of growth of homologous cells. This indicates the impor- 
tance of these cells, not only in relation to processes of resistance and im- 
munity to infection, but to repair of injured tissues as well. Substances of 
a bacteriolytic nature are produced not only by leukocytes, but probably by 
other body cells as well, which are to be found not only in the blood but in 
the secretions and tissues, as indicated by the recent investigations of Flem- 
ing and Allison.4 
Preparation of Leukocytic Extracts.—Hiss and Zinsser have prepared 
1 Ann. d. l’lnst. Pasteur, 1921, 35, 497. 
2 Jour. Exper. Med., 1922, 36, 645. 
3 Amer. Jour. Hyg., 1921, 1, 358. 
4 Jour. Exper. Path., 1922, 3, 252. BACTERICIDAL ACTIVITY OF WHOLE BLOOD 
361 
leukocytic extract by giving rabbits intrapleural injections of aleuronat 
suspension. Manwaring has secured much larger quantities by making 
his injections into the horse. 
The aleuronat is prepared by making a 3 per cent, solution of starch 
in bouillon without heating, and adding 5 per cent, of powdered aleuronat 
to this emulsion. The starch helps to keep the aleuronat in suspension. 
The mixture is boiled for five minutes and placed in large sterile test-tubes, 
20 c.c. being placed in each tube. Final sterilization is done in an auto- 
clave. 
For making the injections large rabbits are selected. The hair over 
both sides of the thorax is removed, the skin is sterilized with tincture of 
iodin, and 10 c.c. of the aleuronat suspension are injected into each pleural 
cavity at a point in the anterior axillary line, at the level of the sternum, 
great care being taken to avoid puncturing the lungs. After twenty-four 
hours the animals are chloroformed and the pleural cavities carefully and 
aseptically opened. The cellular exudate is pipeted into centrifuge tubes 
containing at least 10 c.c. of sterile 1 per cent, sodium citrate in normal 
salt solution. One or more cubic centimeters of exudate may be obtained 
from each cavity. The exudate is usually tinged with blood. It is then 
centrifuged and the supernatant fluid removed. The sediment is broken 
up with a platinum spatula, and 20 volumes of sterile distilled water are 
added. The tubes are set aside in the incubator for twenty-four hours, 
after which cultures are made to insure sterility. A small amount of pre- 
servative may be added, and the extract placed in bottles or ampules ready 
for administration. It is given subcutaneously in doses of from 5 to 10 c.c. 
every four to six hours. 
Archibald and Moore1 have described a method of securing leukocytes 
from the circulating blood of normal animals. This consists of bleeding the 
horse, dog, or other domestic animal from a jugular vein into sterile flasks 
containing a sufficient quantity of 1 per cent, sodium citrate solution to 
prevent coagulation. To this suspension is added 0.5 per cent, acetic acid 
and the mixture centrifuged. The sediment is washed several times with 
saline solution, ground with sterile sand, and neutralized if necessary. To 
the resulting product is added four volumes of sterile water followed by 
heating at 58° C. for one hour, after which it is placed in the incubator 
until digestion is complete. The material is again centrifuged and the 
supernatant fluid employed preserved with tricresol. 
The endocellular bactericidal substances, or endolysins, mentioned in a 
previous chapter, which can be extracted from leukocytes, are not in the 
nature of complements, as they are not rendered inactive by temperatures 
below 80° C. They cannot apparently be increased by immunization, the 
quantity present in each leukocyte being probably at all times just sufficient 
for the digestion of a limited number of bacteria, which can be ingested at 
one time by the individual leukocyte. It is probable that the excess of 
bactericidal substance is thrown off into the blood-stream, representing the 
serum bacteriolysins, and at least indicating that the leukocytes are one 
source for the production of bacteriolytic amboceptors. 
Bactericidal Activity of Whole Blood Versus Plasma and Serum.—The 
maximum degree of bactericidal activity is observed with whole blood. 
Striking evidence of this has been recently supplied by the investigations 
of Heist, Cohen and Cohen,2 who showed that the natural immunity of the 
pigeon to virulent pneumococci was due in large part to the bactericidal 
activity of the whole blood. Rabbit and mouse bloods were found prac- 
1 Jour. Exper. Med., 1912, 16, 249. 
2 Jour. Immunology, 1918, 3, 261. 362 
BACTERIOLYSINS 
tically without bactericidal activity, and the high susceptibility of these 
animals to the pneumococcus is well known. Matsunami and myself1 
observed similar results in respect to natural immunity to virulent meningo- 
cocci. Matsunami2 has likewise observed that with the Heist-Lacey bac- 
tericidal test whole coagulable blood was more active than defibrinated 
and citrated blood and serum, probably because phagocytosis was opera- 
tive and aided in the former. Halm3 has reported a histon plasma just as 
bactericidal as serum, but Metchnikoff has maintained, largely on the basis 
of the experiments of Gengou,4 that plasma is without bactericidal activities, 
whereas serum is bactericidal due to the presence of substances derived 
from leukocytes during the process of coagulation. Watanabe,5 in my labora- 
tory, found plasma to contain about as much complement and antibody as 
serum. 
Mechanism of Bacteriolysis.—According to the side-chain theory of 
Ehrlich a bacteriolysin is an antibody of the third order, or an amboceptor 
furnished with two haptophore or grasping arms for uniting the bacterium 
on one side with a suitable complement on the other. The antibody, there- 
fore, acts simply as an interbody or connecting link; it is specific for the 
bacterium causing its production, but is unable itself directly to injure the 
bacterium, lysis being brought about by an 
attached complement. Bacteriolysis is, 
therefore, an interaction of amboceptor 
and complement upon the bacterial body 
(Fig. 118). 
While bacteriolytic amboceptors will 
unite with their antigens under practical- 
ly all conditions, nevertheless a suitable 
complement may not be present, and hence 
a bacteriolytic serum may not be active in 
all animals. 
The influence of bacteriolysins uponendotoxins is a question of con- 
siderable interest and importance. As the result of convincing experi- 
ments performed, especially by Pfeiffer, it is evident that a bacteriolytic 
serum does not neutralize the endotoxin at the time the bacterium under- 
goes disintegration. Highly immune serums appear to be unable to pro- 
tect an animal against the endotoxins, and, indeed, may even increase the 
intoxication, and, by liberating an excess of endotoxin, kill the animal. 
The bactericidal substances derived from leukocytes are, however, 
apparently capable of neutralizing endotoxins, to some extent at least, 
as Hiss and Zinsser were unable to ascribe the beneficial effects of leuko- 
cytic extracts to the bacteriolytic action alone. 
As mentioned elsewhere, both Metchnikoff and Bordet maintain that 
the bacteriolysin is in the nature of a “sensitizer,” preparing the bacterium 
for the action of the alexin or cytase, just as a mordant aids in the pene- 
tration of a dye-stuff. 
Bactericidal Activity Without Complement.—As stated above, comple- 
ment or alexin is ordinarily required for the phenomenon of bacteriolysis. 
Therefore, it is generally true that the antibacterial properties of both 
normal and immune sera are greater when perfectly fresh than after being 
heated or allowed to stand for weeks or months with the addition of a pre- 
servative, which permits the complete destruction of complement. 
Fig. 118.—Theoretic Structure of a 
Bacteriolytic Amboceptor. 
A, Amboceptor; C, complement; r, re- 
ceptor of bacillus; h, haptophore group of 
the amboceptor; h.a., complementophile 
group of the amboceptor. 
1 Jour. Immunology, 1918, 3, 201. 
2 Jour. Immunology, 1920, 5, 51. 
3 Berl. klin. Wchn., 1896, 864. 
4 Ann. d. l’lnst. Pasteur, 1901, 15, 232. 
5 Jour. Immunology, 1919, 4, 77. NATURAL BACTERIOLYSINS 
363 
Normal human serum, however, often shows some bactericidal activity 
even after heating to 55° C., as shown by Selter1 and others. The nature 
of this thermostabile bactericidal substance is unknown. Flexner,2 as like- 
wise Matsunami and myself,3 have observed the presence of these bac- 
tericidal substances in horse antimeningococcus sera kept many months 
and frequently heated and absolutely free of complement. Black, Fowler, 
and Pierce4 have also reported that heating typhoid and dysentery anti- 
sera did not materially reduce their bactericidal activity when tested by 
the Heist-Lacy method. I believe these bactericidal substances are de- 
rived from the leukocytes, being the bactericidal leukins or endolysins of 
these cells, which are known to be thermostabile and active in the absence 
of complement. 
General Properties of Bacteriolysins.—As with other cytolysins, the 
bacteriolysins are thermostabile and resist heating to 60° C., being gradu- 
ally destroyed at temperatures ranging from 70° to 80° C. They are like- 
wise highly resistant to acids and alkalies, and when preserved in a sterile 
condition with the addition of small quantities of a preservative bacterio- 
lytic serums for therapeutic and diagnostic purposes may remain active 
for long periods of time. Cholera immune serum as a diagnostic aid in mak- 
ing the agglutination and Pfeiffer bacteriolytic reactions is best preserved 
in dry form, the serum retaining its activity under these conditions for 
considerable periods of time. 
Natural Bacteriolysins.—As a result of the contention of Metchnikoff 
that bacteriolysins (fixateurs) are produced only upon the disintegration 
of leukocytes, and that the plasma is accordingly free from these anti- 
bodies, much experimental work has been done. The weight of evidence 
is against this view, as both amboceptors and complements have been 
demonstrated in plasma, although the quantity of bacterial amboceptors 
normally present in the body fluids is quite small. It is quite natural to 
expect that under normal conditions small amounts will be present, as 
receptors are being constantly thrown off into the blood, and leukocytes 
are, of course, being constantly formed and destroyed. 
The amounts of natural bacteriolysins in human blood and that of the 
lower animals vary greatly for different bacteria. For example, Nissen5 
showed many years ago that while normal rabbit blood was highly bac- 
tericidal for the anthrax bacillus, as demonstrated by Fodor and Nuttall, 
bactericidal activity varied greatly for other bacteria, being practically 
absent for staphylococci and streptococci and most active for the pneu- 
mococcus and bacilli of the typhoid-colon-cholera-dysentery group. 
Meltzer and Norris6 found that prolonged fasting had no influence upon 
the bactericidal activity of normal dog-serum upon typhoid bacilli, al- 
though Canalis and Morpurgo7 claim that starvation removes or diminishes 
the natural immunity of the pigeon to anthrax bacilli. 
As previously stated, the bactericidal power of defibrinated blood and 
fresh serum is due to the presence of complement (alexin) and natural 
antibody (bacteriolysin). After heating to 55° C. for ten to thirty minutes 
bactericidal activity is frequently (not always) lost, due to the destruction 
of complement. The presence of natural bacteriolysins may then be demon- 
strated by treating the serum with bacteria, removing them after allowing 
a suitable period for union with the bacteriolysin by centrifuging, and 
1 Ztschr. f. Hyg., 1918, 86, 313. 
2 Jour. Exper. Med., 1908, 10, 141. 
3 Jour. Immunology, 1918, 3, 157. 
4 Jour. Amer. Med. Assoc., 1920, 75, 915. 
5 Ztschr. f. Hyg., 1888, 4, 353. 
6 Jour. Exper. Med., 1899, 4, 131. 
7 Fortschr. d. Med., 1890, 8, 693, 729. 364 
BACTERIOLYSINS 
treating the sediment with guinea-pig serum as complement. Of course, 
the complement serum must be free of natural bacteriolysin for the bacteria 
under study in order to avoid erroneous results. 
The blood of man and many of the lower animals contains appreciable 
amounts of bacteriolysins for many different bacteria. As recently indi- 
cated by the investigations of Heist, Cohen and Cohen,1 Kolmer and Mat- 
sunami,2 Black, Fowler and Pierce,3 and others many examples of natural 
and acquired immunity are to be ascribed to these bactericidal agents. 
Specificity of Bacteriolysins.—The bacteriolysins are highly specific 
antibodies, and are useful in making the differentiation of bacterial species. 
Group bacteriolytic reactions are less common as compared to group agglu- 
tination, as was shown by Kolmer, Williams, and Raiziss4 wdth the typhoid 
colon group of bacilli. As a practical procedure, however, the agglutina- 
tion and complement-fixation reactions are so easily secured as to be the 
tests of choice in making a differentiation between closely allied organisms. 
As with the agglutinins, the influence of partial bacteriolysins may be re- 
moved by using highly immune serums in high dilutions. 
Immune Bacteriolysins.—These are produced during disease and by 
means of artificial immunization. Different bacteria vary greatly in the 
amount of bacteriolysins they are capable of inducing the body cells to 
produce. For example, the pathogenic cocci (staphylococci, streptococci, 
and pneumococci) induce the production of relatively small amounts of 
bacteriolysin and large amounts of opsonin, whereas the bacilli of typhoid- 
colon-dysentery-cholera group induce the production of large amounts of 
bacteriolysins as shown by the investigations of Stern and Korte,5 Lauben- 
heimer,6 Korte and Steinberg,7 Topfer and Jaffe,8 Neufeld and Hune,9 
Wright,10 Sluga,11 Richardson,12 and others. Torrey13 has shown that normal 
and immune rabbit sera may be highly bactericidal for gonococci. 
Role of Bacteriolysins in Immunity.—Doubtless the natural and im- 
mune bacteriolysins and bactericidans exert a very important and funda- 
mental influence upon natural and acquired resistance and recovery from 
bacterial infections. This is especially true of infections with pneumococci, 
meningococci, and bacilli of the typhoid-colon-cholera-dysenterv group and 
is discussed in more detail in subsequent chapters. 
All of the immune sera employed for the treatment of bacterial infec- 
tions contain some of the bacteriolysins which exert an important influence 
upon their curative properties; this is especially true of antipneumococcus, 
antimeningococcus, and antidysentery sera, and of lesser importance with 
diphtheria and tetanus antitoxic sera. This subject is discussed in more 
detail in those chapters dealing with the various immune sera employed 
in the treatment of disease. 
Practical Applications.—The bacteriolysins have considerable value in 
the following procedures: 
1. In making a differentiation of bacteria, especially when the presence 
of cholera vibrios is suspected. 
2. In the diagnosis of certain infections, such as cholera and typhoid 
fever. 
3. In the treatment of some infections with specific bacteriolytic serums. 
1 Jour. Immunology, 1918, 3, 261. 8 Ztschr. f. Hyg., 1906, lii, 393. 
2 Jour. Immunology, 1918, 3, 177. 9 Arb. a. d. Gsndhtsamte, 1907, 25, 16. 
3 Jour. Amer. Med. Assoc., 1920, 75, 915. 10 Lancet, 1901, 609. 
4 Jour. Infect. Dis., 1913, 13, 321. 11 Berl. klin. Wchn., 1904, xli, 79. 
5 Berl. klin. Wchn., 1904, xli, 213. 12 Jour. Med. Research, 1901, 6, 187. 
6 Ztschr. f. klin. Med., 1905, lvi, 170. 13 Jour. Med. Res., 1908, 19, 471. 
7 Deut. Arch. f. klin. Med., 1905, lxxxii, 321. TECHNIC OF BACTERIOLYTIC TESTS 
365 
TECHNIC OF BACTERIOLYTIC TESTS 
The essentials of this important test have been described at the begin- 
ning of this chapter. Briefly, it consists in making intraperitoneal injec- 
tions of a bacteriolytic serum mixed with living bacteria into a normal 
guinea-pig. The resulting bacteriolysis is studied microscopically by with- 
drawing small amounts of peritoneal exudate at varying intervals. By 
performing the experiment with varying dilutions of serum the bacterio- 
lytic titer may be determined by noting the dilution in which bacteriol- 
ysis just fails to occur in a specified time. 
Pfeiffer also showed that the phenomenon could be produced by inject- 
ing a mixture of serum from an immunized animal and the culture of cholera 
into the peritoneum of a normal guinea-pig. This phenomenon appeared 
when an old specimen of serum was used as well as when a fresh specimen 
that had been heated to 60° C. was employed. Later this observer found 
that if an old immune serum was injected into the peritoneal cavity and 
allowed to remain for a time, it regained its bactericidal powers. 
Pfeiffer believed that the bacteriolytic substance may exist in the serum 
of an immunized animal either in an active or in an inert state. In the 
blood-serum or peritoneal fluid of the living animal it occurs as an active 
substance, but when kept for a few days or when heated rapidly to 60° C. 
it becomes inert; it may be rendered active again by coming in contact 
with the lining endothelial cells of the peritoneum. 
The foregoing constitutes the classic Pfeiffer experiment. The bacterio- 
lytic amboceptor present in the immune serum is activated by the com- 
plement furnished by the guinea-pig. The same serum will not produce 
bacteriolysis in the test-tube in case it has been heated or the complement 
is lost through age unless fresh normal serum or peritoneal exudate is added. 
By immunizing guinea-pigs with gradually increasing doses of cholera 
serum and then introducing fatal doses of cholera culture intraperitoneally, 
the same phenomenon of bacteriolysis is observed. In this instance the 
guinea-pig furnishes, both amboceptor and complement. 
By these and similar experiments and observations Bordet was able to 
show the role played by the two bodies concerned in cytolysis in general, 
namely, the thermolabile alexin or complement and the specific sensitizer 
or bacteriolytic amboceptor. 
Bacteriolytic Test in Vivo for the Identification of Bacteria Recovered 
from Feces, Water-supplies, etc.—This method is employed chiefly in the 
identification of suspected cholera cultures. According to Citron, in Ger- 
many, the Pfeiffer test, made writh micro-organisms obtained in pure cul- 
ture from suspected patients is required for the official diagnosis of the 
first cases of cholera. 
As a rule, the agglutination test is first applied in making the diagnosis 
of a suspected micro-organism, as the technic of this test is more easily 
carried out. 
This bacteriolytic test may also be employed in the study of typhoid 
and paratyphoid bacilli, although bacteriolysis of these micro-organisms is 
less complete than that observed with cholera, and agglutination tests 
answer all practical requirements. 
The test consists in mixing varying dilutions of a known and highly 
immune serum with a constant dose of unknown micro-organisms, and 
injecting the mixtures intraperitoneally into guinea-pigs. After twenty 
minutes small amounts of exudate are withdrawn by means of fine capil- 
The Pfeiffer Experiments 366 
BACTERIOLYSINS 
lary pipets, and studied in hanging-drop preparations. In the presence of 
a positive reaction the bacilli are observed to lose motility, become swollen 
and coccoid in shape, and gradually form granules, ultimately disappearing 
in complete solution. 
Preparing the Immune Serum.—A highly immune serum is required. 
This may readily be prepared by giving rabbits a series of intravenous in- 
jections of a known culture, according to the technic described for the 
preparation of bacterial agglutinins. The official test in Germany demands 
that the serum be of such strength that 0.0002 gm. of dried serum will 
suffice to disintegrate completely within one hour 1 loopful (2 mg.) of an 
eighteen-hour-old culture of virulent cholera in 1 c.c. of nutrient bouillon 
when injected into the peritoneal cavity of a guinea-pig. The Hygienic 
Laboratory in Washington is prepared to furnish board of health labora- 
tories with a dried serum of high titer for diagnostic purposes. 
In testing an immune serum to determine its bacteriolytic titer the dose 
of micro-organisms should not be larger than 1 loopful, so that if any par- 
ticular strain of cholera or typhoid is not sufficiently virulent, necessitating 
the use of larger doses, the virulence should be increased by passing the 
organism through guinea-pigs. 
Method of Testing the Virulence of a Culture.—The unit of measure- 
ment is a 2 mm. platinum loop, which, when loaded, will usually hold about 
2 mg. of micro-organisms. All dilutions are made with sterile neutral broth, 
and not with salt solution. Mixtures are injected intraperitoneally into 
250-gm. guinea-pigs to determine the dose that will be fatal within twenty- 
four hours. 
Great care should be exercised in all manipulations to avoid accidental 
infection. The mouth-ends of pipets should be plugged with cotton. Suffi- 
cient assistants should be on hand to facilitate the making of injections 
and the examination of peritoneal exudates with ease, caution, and cer- 
tainty. All pipets, measuring glasses, test-tubes, and hanging-drop prepara- 
tions should be immersed after using in 1 per cent, formalin solution before 
cleaning. In other words, every precaution should be taken to carry out 
a thorough and conscientious bacteriologic technic. 
Guinea-pig No. 1: 4 loopfuls of agar culture emulsified in 4 c.c. of bouil- 
lon; inject 1 c.c. intraperitoneally (=1 loopful). 
Guinea-pig No. 2: 1 c.c. of foregoing emulsion + 1 c.c. of bouillon; 
inject 1 c.c. intraperitoneally ( = \ loopful). 
Guinea-pig No. 3: 1 c.c. of first emulsion + 4 c.c. of bouillon; inject 1 c.c. 
intraperitoneally ( = loopful). 
Guinea-pig No. 4: 1 c.c. of emulsion No. 3+1 c.c. of bouillon; inject 
1 c.c. intraperitoneally loopful. 
A satisfactory culture is one in which a dose of £ loopful will prove fatal 
within twenty-four hours. The immune serum is then titrated with five 
times this amount of culture, or 1 loopful. 
Method of Titrating a Bacteriolytic Serum.—The serum is inactivated 
by heating to 56° C. for half an hour, and dilutions are made with bouillon 
in sterile shallow glasses or test-tubes. One loopful of an eighteen-hour 
agar culture of the micro-organism is thoroughly emulsified in the diluted 
serum, and the mixtures are injected intraperitoneally in 250-gm. guinea- 
pigs. Higher dilutions than those given here may be employed until the 
limit of bacteriolytic activity is reached. 
1. Mix 0.5 c.c. of inactivated serum with 4.5 c.c. of bouillon (1 : 10). 
Place 2 c.c. of this mixture in a separate test-tube and add 2 loopfuls of 
culture. Inject 1 c.c. (=0.1 c.c. immune serum). TECHNIC OF BACTERIOLYTIC TESTS 
367 
2. Mix 2 c.c. of the first dilution (1 : 10) with 18 c.c. of bouillon 
( = 1 : 100). Place 2 c.c. in a separate tube and add 2 loopfuls of culture. 
Inject 1 c.c, (= 0.01 c.c. immune serum). 
3. Mix 1 c.c. of the second dilution (1 : 100) with 4 c.c. of bouillon 
( = 1 : 500). Place 2 c.c. in a separate tube and add 2 loopfuls of culture. 
Inject 1 c.c. (= 0.05 c.c. immune serum). 
4. Mix 2 c.c. of the third dilution (1 : 500) with 2 c.c. of bouillon 
( = 1 : 1000). Place 2 c.c. in a separate tube and add 2 loopfuls of culture. 
Inject 1 c.c. (= 0.001 c.c. immune serum). 
5. To 1 c.c. of the fourth dilution (1 : 1000) add 9 c.c. of bouillon 
( = 1 : 10,000). Place 2 c.c. in a separate tube, and add 2 loopfuls of 
culture. Inject 1 c.c. (= 0.0001 c.c. immune serum). 
Fig. 119.—Removing Exudate from the Peritoneal Cavity of a Guinea-pig (Pfeiffer Bac- 
teriolytic Test). 
A small incision is made through the skin, and the exudate is withdrawn by means of a pipet. 
6. Control: Emulsify 2 loopfuls of culture in 2 c.c. of bouillon. Inject 
1 c.c. This animal will probably succumb within twenty-four hours. 
7. Control: To 2 c.c. of a 1 : 10 dilution of inactivated normal rabbit- 
serum add 2 loopfuls of culture and inject 1 c.c. intraperitoneally. If goats 
or horses are used in preparing the immune serum, this control should be 
conducted with normal goat- or horse-serum. According to Kolle, 1 loop- 
ful of virulent cholera culture is destroyed in the peritoneal cavity of a 
guinea-pig by 0.1 to 0.3 c.c. of normal rabbit’s serum; 0.02 to 0.03 c.c. of 
normal goat’s serum; 0.005 to 0.01 c.c. of normal horse’s serum. 
8. Control: A pig may be injected with 1 c.c. of a 1 : 100 dilution of 
the immune serum and with a loopful of some other micro-organism, such 
as Bacillus coli. This control, however, is not absolutely necessary. 
An area of the abdominal wall of the guinea-pig about 1 inch in diameter 368 
BACTERIOLYSINS 
is shaved and cleansed with alcohol. After the injections have been made 
the bacteriolytic phenomena are observed. 
In making this test fine capillary pipets are prepared and used for with- 
drawing the peritoneal exudate. 
After permitting the animal to inhale a few drops of ether, to make sure 
that it will not suffer, a small incision is made through the skin of the abdo- 
men. The capillary pipet, the large end of which is kept closed with the 
index-finger, is then quickly passed into the abdominal cavity. The index- 
finger is released, and the tube is gently moved about and withdrawn. 
As the result of capillary attraction sufficient exudate usually passes into 
the tube without the aid of suction (Fig. 119). The tube may be fitted 
with a rubber teat in case suction should be necessary. Hanging-drop 
and smear preparations are made and studied microscopically (Fig. 120). 
Fig. 120.—Culture of Cholera Undergoing 
Bacteriolysis.’ A Positive Pfeiffer Re- 
action. 
A hanging drop of peritoneal exudate removed 
from a guinea-pig one-half hour after injection 
with 1 c.c. of the suspension shown in Fig. 121 
with 1 c.c. of 1 : 1000 dilution of cholera immune 
serum. Note that the bacilli are now quite short 
and coccoid in shape. At the end of seventy 
minutes the exudate was sterile. What appears 
to be two or three coccoid forms in apposition is 
really one bacillus undergoing lysis. 
Fig. 121.—Culture of Cholera Before 
Bacteriolysis, x 720. 
A hanging drop of cholera in normal salt solu- 
tion prepared from a twenty-four-hour culture 
of cholera on agar-agar. 
Stained smears are, however, less reliable and not so useful in determining 
the degree of bacteriolysis (Fig. 122). 
It is best to withdraw the exudate immediately after injection, and then 
at intervals of ten, twenty, thirty, forty, and sixty minutes. 
A hanging-drop preparation, made from the culture by emulsifying a 
minute quantity in bouillon, should be on hand as a control in studying 
bacteriolysis in the exudate (Fig. 121). A smear of the culture stained with 
dilute carbolfuchsin should also be on hand for making comparison with 
smears of the exudate (Fig. 123). 
Bacteriolysis is first manifested by loss of motility. As the process pro- 
gresses many of the bacilli become swollen and distorted, and later irregular 
and broken fragments or granules become apparent. Finally, at the end 
of an hour, the exudate is practically sterile. In those cases in which bac- 
teriolysis is complete in an hour the animal generally survives. In the 
higher dilutions of serum bacteriolysis is incomplete, and the animal be- Fig. 122—A Smear of Peritoneal Exudate Removed Twenty Minutes After Injection of 
Culture of Cholera in Figs. 121 and 123. 
Stained with 1 : 10 carbolfuchsin. 
Fig. 123.—Stained Preparation of Cholera Before Bacteriolysis. TECHNIC OF BACTERIOLYTIC TESTS 
369 
comes toxic and may succumb after twenty-four hours, double the virulent 
dose being used in all injections. 
In the foregoing titration a serum that is bacteriolytic in dilutions of 
1 : 1000 or over will be satisfactory for purposes of diagnosis. When pre- 
served in a sterile condition in separate ampules in a dark cold place the 
titer usually remains unaltered for several months. 
Dried Serum.—If the serum is dried in vacuo it will be necessary to 
titrate with dilutions of the dried products in bouillon, as during the process 
of drying the amboceptor content may be decreased. Dried serum is to 
be preferred, however, as it keeps much longer and there is no danger of 
rendering it worthless by contamination. After drying, the product is 
ground in a mortar and stored in amounts of 0.1 or 0.2 gm. in separate 
ampules. 
In determining the bacteriolytic titer of dried serum 0.1 gm. is care- 
fully weighed out and dissolved in 9.9 c.c. of sterile bouillon. From this 
stock dilution other dilutions may be prepared in the manner previously 
described, and injected with a loopful of the culture. 
After securing an immune serum of satisfactory potency the main test 
may be conducted. While the technic is more difficult than that of agglu- 
tination tests, the results, under proper conditions, are more conclusive, for 
there is less likelihood of group or partial amboceptor activity for closely 
allied bacteria taking place. 
Technic of the Pfeiffer Test.—The suspected culture to be tested should 
be grown on agar for from eighteen to twenty-four hours, and used in doses 
of 1 loopful (2 mg.). 
If cholera is suspected, cholera immune serum of known titer should be 
used. Dilutions of serum are made with sterile bouillon, and 2 c.c. of each 
dilution, representing two, five, and ten times the titer dose, are placed in 
separate glasses or small test-tubes. A fourth tube should contain 2 c.c. 
of a 1 : 100 dilution of normal serum, according to the animal used in pro- 
ducing the immune serum; a fifth tube should contain 2 c.c. of sterile bouillon 
(culture control). 
Two loopfuls of suspected culture are thoroughly emulsified in each of 
the 5 tfibes of the series, and 1 c.c. of each is injected intraperitoneally in 
5 guinea-pigs weighing about 250 gm. each. 
At intervals of ten, twenty, forty, and sixty minutes the peritoneal exu- 
date should be removed with capillary pipets and examined in hanging-drop 
preparations. Smears may be prepared and stained with dilute carbol- 
fuchsin, although they give less information than hanging-drop prepara- 
tions. 
If guinea-pigs Nos. 1, 2, and 3 showT granule formation at the latest 
after an hour, while in the fourth and fifth animals the bacteria remain 
whole, motile, and well preserved, the reaction may be regarded as positive 
and the diagnosis as established. 
Pfeiffer Bacteriolytic Test in Vivo in the Diagnosis of Disease.—Bacteri- 
olysins are usually produced somewhat later than agglutinins, and reach 
their highest point of production during convalescence. Bacteriolytic tests 
are used only for diagnostic purposes when agglutination reactions are nega- 
tive or doubtful. The most satisfactory results from these tests are obtained 
in cholera. Bacteriolysis with typhoid bacilli is less typical and more in- 
complete; wfith the paratyphoid and dysentery bacilli it is even more un- 
satisfactory, and in anthrax, pest, and the various diseases due to cocci 
it does not occur. 
The test is conducted in a manner similar to the preceding test, except 370 
BACTERIOLYSINS 
that instead of an immune serum the patient’s serum is used, with a known 
culture of typhoid, cholera, paratyphoid, etc., according to the infection 
suspected to be present. 
One c.c. of the patient’s serum is secured in a sterile manner, inactivated 
by heating to 55° C. for one-half hour, and dilutions of 1 : 20, 1 : 100, 
1 : 250, and 1 : 500 prepared with sterile bouillon. Two c.c. of these dilu- 
tions are placed in separate tubes, and 2 loopfuls of an eighteen-hour cul- 
ture of the test organism emulsified in each. A fifth tube contains 2 c.c. 
of bouillon with 2 loopfuls of culture, and serves as the culture control. 
One c.c. of each dilution and of the culture control is injected intra- 
peritoneally into 5 guinea-pigs, and the exudate examined after twenty, 
forty, and sixty minutes. 
If granule formation occurs in the first two or three animals, but is 
absent in the culture control, the reaction may be regarded as positive, 
and the diagnosis of typhoid, cholera, etc., considered as established. 
If granule formation occurs to any extent in the fifth animal (culture 
control), the culture is to be regarded as unsuitable and the test repeated. 
Measuring the Bactericidal Power of the Blood in Vitro by the Plate 
Culture Method (Stern and Korte).—Since the earliest days of bacteriology 
the aim and purpose have been to devise some means by which the bacteri- 
cidal power of the blood could be measured, just as is done in testing an 
ordinary germicidal substance, namely, by adding a bacterial suspension 
to the serum and observing the effect on the bacterial content. This is 
shown by counting the bacteria in a loopful immediately after adding the 
bacterial suspension, and repeating this at regular intervals, the counting 
being done by the method of plate culture. 
The use of the platinum loop in these tests is objectionable, since the 
dose is quite variable, depending upon whether the loopful is removed 
from the fluid edgewise, or with the loop held flat, like a spoon in use. If 
the serum contains agglutinins the results with any form of technic are 
likely to be irregular, and the number of colonies upon the plate stand in 
no relation to the actual number of micro-organisms present. The results 
may be masked by a rapid multiplication of the survivors, and accordingly 
several controls are necessary with any technic. 
Neisser and Wechsberg have recommended the so-called bactericidal plate 
culture method. By this method the patient’s serum is inactivated, and 
varying amounts mixed with definite and constant quantities of bacteria. 
To this a constant quantity of active normal serum is added as comple- 
ment, to reactivate the bacteriolytic amboceptors. These mixtures are 
then incubated for several hours. To determine whether and in what 
proportiop death of bacteria has resulted from the effect of the bacteriol- 
ysins, agar is added, and the mixtures are then plated and incubated for 
twenty-four hours or longer. The number of colonies are then counted 
and the results compared with control plates of the culture alone. 
Stern andi Korte have modified this technic slightly, and recommend the 
procedure as a substitute for the Pfeiffer test in the clinical diagnosis of 
typhoid fever. The method spares a certain number of animals, but it is 
somewhat more complicated than the Pfeiffer reaction, and its results are 
less trustworthy. As a clinical test, therefore, it is to be recommended 
only in cases in which! the agglutination reactions have yielded uncertain 
results, although with practice and care the test oftentimes yields uniform 
and satisfactory results, and may be employed in making special investiga- 
tions for determining the bactericidal power of the blood, as after typhoid 
immunization. TECHNIC OF BACTERIOLYTIC TESTS 
371 
Technic of the Test.—The requisites for success are that all vessels and 
diluting fluids as well as the serums employed should be absolutely sterile. 
To secure uniform and reliable results it is necessary to familiarize oneself 
with the technic by repeated practice. In order to carry out the steps in 
the technic according to strict aseptic bacteriologic principles the services 
of an assistant are required. 
Sterile bouillon is largely used throughout to maintain proper osmotic 
conditions. 
With so many tubes and controls it is important that all tubes and 
Petri dishes be properly labeled with a wax-pencil to avoid confusion. 
One c.c. of sterile patient’s serum and an equal amount from a normal 
and healthy person to serve as a control are inactivated by heating to 55° 
or 56° C. for half an hour. These serums are then diluted 1 : 50 by adding 
49 c.c. of sterile salt solution to each and mixing thoroughly. 
Complement is prepared by securing 4 to 5 c.c. of sterile rabbit’s blood 
and separating the serum. This serum is chosen because it should be from 
the same species of animal as that used in producing the immune serum 
to be tested. In determining the bactericidal titer of human serum either 
guinea-pig or rabbit complement serum may be used. Dilute 2 c.c. of this 
fresh serum (not over eighteen hours old) with 18 c.c. of sterile normal salt 
solution (1 : 10). Dose, 0.5 c.c. 
Secure a good twenty-four-hour bouillon culture of typhoid bacilli 
(grown in the incubator) and dilute 1 : 500 by thoroughly mixing 0.1 c.c. 
of the culture in a flask containing 50 c.c. of sterile normal salt solution. 
This dilution of culture, when used in constant doses of 0.5 c.c., generally 
yields satisfactory results, but it may be necessary to try it out before- 
hand, for the control plates must regularly show a uniformly good growth 
and contain many thousands of colonies before uniform results can be ex- 
pected. The bactericidal effect will then be distinctly shown by the reduc- 
tion, in the proper plates, of this large number of colonies to zero or near it. 
Place 1 c.c. of sterile neutral bouillon in each of 12 test-tubes which 
have been plugged with cotton, sterilized, and large enough to hold at least 
from 12 to 15 c.c. Place in the first of these tubes 1 c.c. of the diluted pa- 
tient’s serum and mix thoroughly by alternate sucking up and forcing out 
of the fluid; then, with the same pipet, draw up 1 c.c. and transfer it to the 
second tube of the series; mix as before, and transfer 1 c.c. to the third tube; 
continue in this manner to the last tube, from which, finally, 1 c.c. is dis- 
carded. 
Each tube now contains 1 c.c. of fluid, representing dilutions of 1 : 100, 
1 : 200, 1 : 400, 1 : 800, 1 : 1600, 1 : 3200, and so on up to 1 : 204,800 of 
the patient’s serum. Higher or lower dilutions than these may be em- 
ployed. It is very desirable that the dilutions be high enough to secure 
the limit of bactericidal activity, so that the last plates will show an in- 
crease in the number of colonies. 
Arrange four tubes with the normal control serum, each containing 
1 c.c. of fluid and representing the first four dilutions, viz., 1 : 100, 1 : 200, 
1 : 400, and 1 : 800. 
Label each tube carefully with the initials of the patient and the dilu- 
tion it contains. Add 0.5 c.c. of the diluted bacterial emulsion, and finally 
0.5 c.c. of the diluted complement serum to each tube. 
All manipulations should be made with strict bacteriologic care and 
with the aid of an assistant, as the introduction of contaminating micro- 
organisms that may cause spore formation will considerably vitiate the value 
of the test. 372 
BACTERIOLYSINS 
The following controls are necessary: 
1. 0.5 c.c. culture directly in a sterile Petri dish + 5 to 10 c.c. of neutral 
agar cooled to 42° C. and thoroughly mixed. This control will 
show the original number of bacteria contained in the emulsion. 
Mark as control No. 1. 
2. 0.5 c.c. culture + 1.5 c.c. bouillon. Mark as control No. 2. After 
three hours’ incubation this tube receives the usual amount of 
agar and is plated. It will show to what degree the culture has 
multiplied in the incubator. 
3. Since the complement serum frequently contains bacteriolysin, it is 
necessary to control this element carefully. To a series of four tubes 
add the following doses of the serum, diluted 1 : 10: 1 c.c., 0.5 c.c., 
0.2 c.c., 0.1 c.c. Add 0.5 c.c. culture to each tube, and sufficient 
bouillon to make the total volume in each tube 2 c.c. 
4. 0.5 c.c. complement + 1-5 c.c. bouillon. To control the sterility of 
the complement serum. 
5. 1 c.c. of patient’s serum (1 : 100) + 0 c.c. bouillon. To control the 
sterility of this serum. 
6. 1 c.c. of the control serum (1 : 100) + 1 c.c. of salt solution. To 
control the sterility of this serum. 
7. 1 c.c. of the patient’s serum in dilution of 1 : 100 + 0.5 c.c. culture 
+ 0.5 c.c. bouillon. A control on the possible presence of comple- 
ment in this serum. 
8. 1 c.c. of the control serum in dilution of 1 : 100 + 0.5 c.c. culture + 
0.5 c.c. bouillon. A control on the possible presence of complement. 
All tubes with the exception of the first control which has been plated 
are placed in an incubator at 37° C. for three hours. At the end of this 
time the contents of each tube are plated in neutral agar. Sterile Petti 
dishes should be properly and plainly labeled with a wax-pencil and ar- 
ranged in order. A flask of plain neutral agar is melted in boiling water 
and cooled to 42° C. The tubes are removed from the incubator and shaken 
gently, and with the aid of an assistant from 5 to 8 c.c. of agar are added 
carefully to each tube wdth a sterile 10 c.c. pipet. The contents are mixed 
by gentle rotation of the tube and then poured in the corresponding Petri 
dish, followed by an additional mixing according to the usual bacteriologic 
procedure. With ordinary speed the whole set of tubes may be poured in 
a satisfactory manner before the flask of agar has had time to harden. 
Another method of plating consists in pipeting the contents of each tube 
into its corresponding dish, and then washing the tube with an additional 
1 c.c. of sterile salt solution to remove all traces of serum and culture. Or 
the end of each tube may be flamed and the contents poured directly into 
a dish. If this method is adopted small test-tubes should be used. While 
the method is more convenient, it is usually n'ot so accurate as the first two 
methods. 
Neisser plates but 5 or 10 drops from each tube. The dose decided upon 
is the one employed throughout. For example, it would be erroneous to 
take 5 drops from one tube and 10 from another. Neisser, howrever, uses 
much smaller amounts of the serum, as 1, 0.3, 0.1, 0.03, and 0.01 c.c., in- 
stead of the much higher dilutions given in this technic. These differences 
must be remembered and the proper dilutions employed, and but 5 to 10 
drops should be plated in performing the bactericidal plate test according 
to Neisser’s technic. 
Topfer and Jaffe pour a thin layer of agar into a Petri dish and allow it 
to harden. Upon this they pour the culture-serum agar mixture, which TECHNIC OF BACTERIOLYTIC TESTS 
373 
after settling, is covered with another thin layer of agar. In this way a 
culture in the water of condensation is avoided. In the usual technic this 
may be avoided by turning the plates over soon after hardening occurs 
so that the water of condensation collects on the cover. 
All plates are incubated at 37° C. for from twenty-four to thirty-six 
hours and the colonies then counted. In some plates the number may be 
so large that counting will be inaccurate and unnecessary. Significance 
can be attached only to marked and easily recognizable differences. Accord- 
ing to Neisser, the growth is best and most rapidly described by means of 
approximate estimates, using a scheme somewhat like the following: 0 or 
almost none; about 100; several hundreds; thousands; very many thousands; 
infinite numbers. A distinct bactericidal action is then present only if the 
controls react normally, and if a reduction of colonies from an infinite num- 
ber or many thousands to 0 or very few has occurred. As previously stated, 
the test can then be regarded as satisfactory only if the lower limits of bac- 
tericidal activity have been reached and the last plates show an increase 
in the number of colonies. Examination of these plates is very much facili- 
tated by using a colony counter, that of Stewart being particularly service- 
able with agar plates. 
The controls are first examined and the results recorded, and finally 
those of the patient’s serum are set down. 
As a practical illustration, the results of a test with a rabbit typhoid 
immune serum are given, because this is in general an average example 
of those usually obtained: 
1 C.c. Dilutions 
of Immune 
Serum. 
Twenty-four- 
hour Culture 
of Bacillus 
Typhosus (Dilu- 
tion 1 : 500 C.c.). 
Rabbit Comple- 
ment Serum, 
1 : 10 C.c. 
Plates Poured After Three Hours 
at 37° C. Counted After 
Twenty-four Hours 
Immune Serum 
(Inactivated) 
Normal Rabbit 
Serum (In- 
activated). 
1 : 100 
0.5 
0.5 
About 500 colo- 
nies. 
Many thousand. 
1 :200 
0.5 
0.5 
About 125 colo- 
nies. 
Many thousand. 
1 :400 
0.5 
0.5 
Sterile. 
Many thousand. 
1 :800 
0.5 
0.5 
Sterile. 
Many thousand. 
1 : 1600 
0.5 
0.5 
Sterile. 
1 :3200 
0.5 
0.5 
Two colonies; 
practically 
sterile. 
1 :6400 
0.5 
0.5 
100 colonies. 
1 :12,800 
0.5 
0.5 
About 800 colo- 
nies. 
1 : 25,600 
0.5 
0.5 
About 2000 colo- 
nies. 
1 :51,200 
0.5 
0.5 
Many thousand. 
1 : 102,400 
0.5 
0.5 
Many thousand. 
1 -.204,800 
0.5 
0.5 
Many thousand. 
Controls 
Control 1: 0.5 c.c. culture plated immediately: many thousands of 
colonies. 
Control 2: 0.5 c.c. culture plated after being incubated three hours: 
plate very crowded; number of bacteria increased. BACTERIOLYSINS 
374 
Control 3: Varying amounts of normal rabbit serum used as complement. 
The first plate containing 0.1 c.c. serum showed a slight bactericidal 
action; the remaining plates showed many thousand of colonies and 
were comparable to control No. 1. 
Control 4: 0.5 c.c. complement serum: sterile. 
Control 5: 0.01 c.c. inactivated immune serum: sterile. 
Control 6: 0.01 c.c. inactivated normal control serum: plate shows 
several colonies of contaminating bacteria. 
Control 7: 0.01 c.c. inactivated immune serum plus culture: plate 
shows many thousands of colonies. 
Control 8: 0.01 c.c. of inactivated normal serum plus culture: many 
thousands of colonies. 
An examination of this experiment shows that the dose of culture was 
satisfactory, that the culture increased during the three hours of primary 
incubation, that the complement serum was very slightly bactericidal in 
a dose lower than that used in making the test, and that the technic was 
fairly satisfactory, slight contamination being present in but one plate— 
that of the inactivated normal control serum. 
The most striking result observed is the absence of complete bactericidal 
activity in the first two plates, which contained the largest amount of immune 
serum, and where one would naturally expect to find complete sterility. The 
titer of this serum was between 1 : 3200 and 1 : 6400. According to Neisser 
and Wechsberg these paradoxic results are caused by deviation or “deflec- 
tion of complement,” as was explained in a previous chapter. In bacteri- 
cidal experiments according to Neisser, the deflection is caused by an ex- 
cess of amboceptors in the immune serum. In a mixture of bacteria, com- 
plements, and large amounts of amboceptor the complement is bound not 
only by the amboceptors anchored to the bacteria but also in large measure 
by “free” amboceptors that are not anchored to bacteria. A portion of the 
anchored amboceptor, therefore, finds no complement at its disposal, and 
is, therefore, unable to exert any bactericidal action, which gives rise to a 
relative lack of complement. 
This phenomenon resembles the action of agglutinoids in the aggluti- 
nation reaction, where, in the lowest dilutions, agglutination is feeble or 
absent, but becomes manifest in the higher dilutions. 
In the foregoing experiment the immune serum used was several weeks 
old; perfectly fresh serums are not so likely to show this so-called “deflec- 
tion of complement.” In many instances, and especially if a fresh serum 
is used, one cannot help thinking that agglutinins may be responsible for 
the absence or diminution of bactericidal activity in the lower dilutions. 
It is certainly true that hemagglutinins considerably inhibit hemolysis, 
and this is especially the case with serums of low hemolytic activity. With 
more potent serums the agglutinins are diluted until activity ceases and 
hemolysis is ready and complete. Reasoning from analogy, therefore, the 
absence or diminution of bactericidal activity may be due to agglutinins, 
and the theory of “deflection of complements” may be emphasized a little 
too strongly. 
According to Halm, normal human serums show bactericidal activity in 
only about one-third of the cases, and the titer is only very exceptionally 
demonstrable in dilutions higher than 1 : 500. The serums of advanced 
cases of typhoid fever or of those but recently recovered are bactericidal 
in dilutions greater than 1 : 1000, and may reach 1 : 50,000 or higher. 
Similarly after typhoid immunization the patient’s serum may show a high TECHNIC OF BACTERIOLYTIC TESTS 
375 
bactericidal titer. Weston has found such serums active in dilutions of 
1 : 20,000, higher dilutions not being used. 
Besides being used in typhoid, the plate culture method has been em- 
ployed for experimental purposes in cholera, dysentery, paratyphoid, and 
other infections with bacilli of the typhoid-colon group. With these, how- 
ever. the test possesses but little diagnostic significance. 
The bactericidal titer does not run strictly parallel with the agglutinins 
or complement-fixing bodies. 
Measuring the Bactericidal Power of the Blood by Capillary Pipet 
Method (after Wright1).—By this technic it is sought to overcome the 
fallacies of the “loopful” method of measurement and those due to aggluti- 
nation of the test organism. 
The native complements of the patient’s own serum are used; hence 
the serums used in this technic must be fresh. Quantitative titration is 
accomplished by furnishing varying dilutions of culture, with a constant 
quantity of serum. A series of volumes of serum is taken, and to these 
are added equal quantities of progressively increasing dilutions of a counted 
bacterial culture. The mixtures are kept at 37° C. for twenty-four hours, 
after which each is introduced into nutrient broth and cultivated to see 
whether a complete bactericidal effect has been exerted. The largest number 
of bacteria that a constant quantity of serum has been able to kill furnishes a 
measure as to its bactericidal power. 
After a few technical details have been mastered this method is quickly 
performed and yields fairly constant and reliable results. It is not adapted 
for the titration of old immune serums, but is a ready clinical test, finding 
a special field of usefulness in determining the bactericidal powers of the 
blood after typhoid immunization. 
Requisites for Carrying Out the Test.—1. A specimen of the patient’s 
blood is collected aseptically, a process that may be accomplished by thor- 
oughly cleansing the finger with alcohol and collecting blood in a sterile 
Wright capsule or, better perhaps, by means of venipuncture, when from 
2 to 5 c.c. may be collected aseptically in a sterile centrifuge tube. The 
control blood from a healthy individual may be collected in a capsule. 
The serums are carefully separated and pipeted into small sterile test- 
tubes; 1 c.c. of each is ample for the test. 
2. A twenty-four-hour-old broth culture of the test organism (a young cul- 
ture is required, because such a culture contains a few dead micro-organisms 
and the absorption of bactericidal elements by dead organisms is thus avoided). 
3. About two dozen “looped pipets” made according to the directions 
given in Chapter I. 
4. Sterile neutral broth for making dilutions and cultivations. This 
is prepared and sterilized in the usual manner, from 5 to 10 c.c. being placed 
in test-tubes. When working with the typhoid bacillus a special broth 
containing 1 per cent, of mannite and sufficient litmus to color it a deep 
blue (Smallman) should be on hand. 
5. Two dozen small test-tubes, plugged and sterilized, for making dilu- 
tions of the culture. 
Preparation and Enumeration of the Bacterial Culture.—As the principle 
of the test depends upon measuring the bactericidal activity of the blood 
according to the number of organisms that are killed, it is necessary to pre- 
pare somewhat high dilutions and count the number of organisms in a unit 
volume in each dilution. 
1 Wright, A. E.: Technique of the Teat and Capillary Glass Tube, 1912, London, Con- 
stable & Co. BACTERIOL YS1NS 
376 
For this purpose place 10 sterile test-tubes in a rack, and arrange 10 
sterile Petri dishes on the table to correspond to these. To the first tube 
add 4.9 c.c. of plain sterile broth; to the second, fourth, sixth, eighth, and 
tenth tubes add 1 c.c., to the third, fifth, seventh, and ninth tubes add 
4 c.c. 
With a sterile graduated 1 c.c. pipet add 0.1 c.c. culture to the first 
tube and then discard the pipet by placing it in a cylinder of disinfecting 
solution or hot water. With a second sterile pipet mix the contents of tube 
No. 1 by carefully sucking it in and forcing it out of the pipet several times, 
and place 0.1 c.c. in Petri dish No. 1. Then with the same pipet transfer 
1 c.c. to the second tube, mix well, and place 0.1 c.c. in Petri dish No. 2; 
next transfer 1 c.c. to the third tube, mix, and place 0.1 c.c. in the third 
Petri dish. Continue in this way until the last tube is reached, when 1 c.c. 
is discarded into a germicidal solution and 0.1 c.c. placed in the tenth Petri 
dish. To each Petri dish now add from 8 to 10 c.c. of neutral agar at 41° C., 
and mix thoroughly. After the plates have hardened they are turned over 
in order that the water of condensation may collect on the cover. They 
are then incubated for twenty-four hours, and the colonies carefully 
counted. 
We now have the following dilutions of culture: 1 : 50, 1 : 100, 1 : 500, 
1 : 1000, 1 : 5000, 1 : 10,000, 1 : 50,000, 1 : 100,000, 1 : 500,000, and 1 : 
1,000,000. Since but 0.1 c.c. of these dilutions were plated, the total number 
of colonies in each plate must be multiplied by 10 in order to obtain the 
approximate number per cubic centimeter of the various dilutions. 
To secure fairly uniform results the various dilutions must be thor- 
oughly mixed and the pipeting be accurately performed. We have found 
that this method usually gives better results than are obtained by plating 
but one or two of the higher dilutions, which are used as a basis for calcu- 
lating the number of bacteria in the other dilutions. 
Filling the Pipets.—With a wax-pencil make a mark upon the capillary 
stem of a sterile looped pipet at a point 2 cm. from the lower end, and fit 
a rubber teat to the barrel. The point of the capillary stem is now broken 
off between the finger and thumb, the lower portion is sterilized in the 
flame, and the air is expelled from the teat. 
Mannite broth is then aspirated into the pipet until the bulb is about 
two-thirds full and the capillary portion contains an air column rising to 
at least 5 cm. from the end (Fig. 124). 
The end of the capillary stem is now inserted into the tube containing 
the patient’s serum, and the serum is allowed to flow in until it reaches 
the pencil mark. 
The orifice of the pipet is now raised above the surface of the serum, 
and a small bubble of air is admitted into the tube to serve as an index 
for the measurement. The end of the capillary stem is now carried into 
the tube containing the highest dilution of the organism, and the culture 
is allowed to flow in until the bubble of air has been carried just past the 
pencil mark. 
The serum and culture must now be mixed, being careful not to con- 
taminate the broth. This is effected by blowing these two volumes out 
into a sterile mixing tube and drawing up and blowing out the fluid several 
times in succession. By starting with the highest dilution the same mixing 
tube may be used throughout the series. With an air-bubble of at least 
5 cm. between serum and broth this can be quite easily achieved without 
driving the sterile broth down from the bulb of the pipet into the lower 
part of the capillary stem and contaminating it there. Fig. 124.—Bactericidal Test (Looped Pipet Method oe Wright). 
The pipet shown on the extreme left contains the mannite-litmus broth and an equal volume 
of patient’s serum and emulsion of typhoid bacilli; the second pipet shows the serum and bacillary 
emulsion mixed; the middle pipet shows the seru.m-bacillary mixture cultured in the broth; the fourth 
pipet shows acid production in the broth by living bacilli which escaped destruction after twenty- 
four hours’ incubation; the fifth pipet (extreme right) is the serum control. The broth, being clear and 
unchanged, shows that the serum was sterile. TECHNIC OF BACTERIOLYTIC TESTS 
377 
The column of mixed serum and culture is now drawn up into the middle 
region of the capillary stem and the lower end of the tube is sealed. The 
teat is then removed, and the dilution that has been used is written on the 
barrel of the pipet. 
The series of pipets are now filled wTith the nine measuring dilutions of 
the culture. 
Controls.—The serum from a healthy individual or, better still, the 
pooled serums of several persons may be used in precisely the same way, 
with at least the second, fourth, sixth, and eighth dilutions of culture. 
Several culture controls are advisable. The pipets are filled with man- 
nite broth in the usual manner, and then with one volume of at least four 
different dilutions, usually 1 : 50, 1 : 5000, 1 : 100,000, and 1 : 1,000,000. 
The tubes are sealed, labeled, and incubated together with those concerned 
in the test proper. 
One pipet is to contain the usual quantity of broth and one volume of 
the patient’s serum. This is a control on the sterility of the patient’s serum. 
A similar preparation is made with the control serum to serve as a control 
on its sterility. 
When the whole series of tubes have been filled, these are placed up- 
right in a large test-tube or cylinder labeled with the date and the source 
of the serum, and incubated at 37° C. for from eighteen to twrenty-four 
hours. 
Test for the Germicidal Activity of the Serum.—The serum and culture 
in the capillary portion of‘the tube must now be mixed with the broth in 
the barrel. This is accomplished by taking each pipet in hand singly, and 
heating the lower portion of the capillary stem in a “peep-flame” and draw- 
ing out into a small thread with a pair of forceps. 
A collapsed teat is now fitted over the barrel, and the negative press- 
ure carefully regulated by keeping the finger and thumb in position on the 
teat, and the finely drawn out end of the capillary stem gently snapped 
across. The column of serum and culture will then be carried up into the 
bulb of the pipet. The end of the tube is now sealed. 
When the whole series of pipets has been dealt with in similar fashion 
they are returned to the incubator for another twenty-four hours. 
Reading the Result.—The continued sterility of the broth may be de- 
termined from direct naked-eye inspection. 
When a complete bactericidal effect has been secured, the broth will 
have undergone no color change, but will still be clear. 
When a growth of the typhoid bacillus has occurred, a perceptible cloudi- 
ness will be apparent in the broth, and the color will have changed from 
blue to reddish blue, indicating that acid has been formed from the mannite. 
Turbidity of the broth without a change of color would indicate the 
admission of contaminating microbes that were unable to split mannite 
with the formation of acid. 
The controls are first examined and recorded. All the culture controls 
should show a growth and change of color. The serum controls should be 
sterile; the normal serum controls may or may not be sterile, depending 
upon the amount of natural bactericidal amboceptors which it contains 
for the test organism. 
The pipets containing the patient’s serum are now examined, and a 
simple numeric expression for the result is obtained by referring to the 
result of the enumeration of the various dilutions, as determined by the 
counting of the plates. In this manner the number of bacteria contained 
in 1 c.c. of the lowest dilution of the bacterial culture that has been com- 378 
BACTERIOLYSINS 
pletely sterilized is calculated. This will give the number of micro-organisms 
which 1 c.c. of serum would be capable of killing. 
Although this test affords a convenient basis for the comparison of 
serums, it must be understood that the expression is entirely arbitrary, 
and will vary according to the culture employed; this is true of any technic 
for determining the bactericidal titer of a serum. 
The Lacy-Heist Method of Determining the Bactericidal Activity of 
Coagulable Blood.—Heist, Cohen and Cohen1 have described this technic; 
it is simple and usually yields decisive results. These investigators have 
shown by means of this test that the pneumococcus is able to grow in the 
coagulable blood of susceptible animals, as the mouse, whereas, in the blood 
of insusceptible animals, as the pigeon, growth does not occur or is feeble. 
Kolmer and Matsunami2 have successfully employed the method in a study 
of natural immunity to the meningococcus. Increasing the virulence of 
pneumococci for rabbits was found to increase their ability to grow in rabbit 
blood in vitro according to the results observed with this method; the test 
possesses an approximate quantitative value, and is regarded by Heist and 
Cohen3 as a better measure of pneumococcus immunity than the aggluti- 
nation test. By means of this method it has been shown that whole blood 
before it coagulates is much more pneumococcidal than the serum, and 
immunization ;with pneumococcus bacterins has been found to increase 
this hemocidal activity. 
The technic is as follows: 
1. Heavy walled glass tubing is drawn out into capillary tubing having 
an inside diameter of 0.5 to 1 mm. The capillary tubing is cut in 15 cm. 
lengths and the ends trimmed square by nicking with a fine file or with the 
edge of a carborundum pocket hone, and breaking across. Five of these 
tubes are held side by side, palisade fashion, and a pellet of plasticine molded 
closely about them near one end. A short piece of glass tubing stuck into 
the plasticine so that it encloses the five protruding ends makes a convenient 
handle. Pressure on the plasticine spreads out the capillary tubes into a 
fan shape. With a wax-pencil a fiduciary mark is made upon each capillary 
tube about 5 cm. from its free end and also each one is given an identifica- 
tion mark. Tube 1 receives one dot with the wax-pencil, tube 2, two dots, 
and so on. This many stemmed pipet is a modification of that devised by 
Wright for estimating the coagulation time of blood (Fig. 125). 
2. Five small, sterile test-tubes are numbered from 1 to 5 and arranged 
in a row. They may be very conveniently held in a sloping position by 
thrusting their ends in a long strip of plasticine. In tube 1 is placed a quan- 
tity of a twenty-four-hour culture of pneumococci in blood broth. In tube 
2 one volume of the broth culture is diluted with four volumes of saline 
solution, making a 1 : 5 dilution. In a similar manner a 1 : 25 dilution is 
made in tube 3, and a 1 : 125 and a 1 : 625 in tubes 4 and 5. In other 
words, dilutions of culture are made in multiples of five. 
3. The many stemmed pipet is taken up and the tip of tube 1 touched 
to the surface of the undiluted culture in test-tube 1. The fluid runs up 
the slender tube by capillary attraction. When the upper level of the as- 
cending fluid reaches the fiduciary mark, the tip of the tube is withdrawn 
from the test-tube, the fluid which has run up the tube remaining in situ. 
The other capillary tubes are filled in the same manner from the test-tubes 
bearing corresponding numbers. If now the tip of one of the capillary 
tubes filled with diluted culture is touched to a piece of moist cheese-cloth 
the column of fluid will run out, the adhesion between the cloth and the 
1 Jour. Immunology, 1918, 3, 261. 
2 Ibid., 1918, 3, 201. 
3 Ibid., 1919, 4, 147. TECHNIC OF BACTERIOLYTIC TESTS 
379 
liquid used being greater than that between the same liquid and glass. 
If the tube should not completely empty itself, gentle blowing into the other 
end of the pipet will start the downward flow again. In this manner each 
capillary tube is thoroughly emptied in turn. 
4. The ear of a rabbit is rubbed briskly for a moment to increase the 
circulation and a vein is pricked. As the blood wells up the tip of capil- 
lary tube 1 is introduced into the drop and the blood flows up by capillary 
attraction. When the ascending column of blood has reached the fiduciary 
mark the tube is moved aside and tube 2 takes its place in the flowing blood. 
In less than a minute all five tubes are filled. 
5. When all five tubes have been filled the distal 
ends are sealed by dipping them in paraffin which 
has been melted and allowed to cool until it is on 
the point of solidifying. This method of closing the 
ends avoids bringing any appreciable amount of heat 
in contact with the freshly coagulated blood. The 
capillary tubes are then gently pulled out of the 
pellet of plasticine, dropped into a test-tube and in- 
cubated for twenty-four hours, at the end of which 
time they are taken out, the tips broken off, and a 
drop from each blown out upon a glass slide. All 
five drops can be placed in a row on the same slide. 
The dots which have been placed upon the capil- 
lary tubes before beginning operations enables one 
to identify them. The slide is fixed in formalin 
vapor, stained, and examined under the microscope. 
6. Under the microscope it is seen that some 
tubes contain no pneumococci. Numerous red and 
white blood-corpuscles are present and well stained, 
and strands of fibrin are interlaced among the cells. 
Other tubes contain great numbers of pneumococci 
mixed with debris, but very few cells; apparently 
the cells and fibrin have been digested by the pneu- 
mococci. These two pictures are practically the 
only ones encountered. Either the pneumococci 
have grown vigorously or they have not grown at all. 
Matsunami1 has applied this test to a study 
of the meningococcidal activity of normal and im- 
mune sera and found that it could not be accepted 
for the purpose of measuring the degree of meningococcidal activity of an 
immune serum, but that the method was useful for measuring the degree 
of natural immunity. Very probably the test has its limits for measuring 
the degree of bactericidal activity of immune serum or blood because it 
utilizes the native complement contained in the volume of blood being 
tested, which may be insufficient for complementing all of the bacteriolytic 
amboceptors present. 
More recently Borow and the writer2 have found the test of value in 
the course of chemotherapeutic studies in bacterial infections, medicaments 
being injected into rabbits by various routes and in doses per body weight, 
the tests being conducted before and at intervals after the administration 
for the purpose of determining if bactericidal activity of the blood was 
increased, and if so, how long this was maintained. 
Fig. 125.—Wright’s Many 
Stemmed Pipet. 
Used in estimating the 
bactericidal activity of whole 
coagulable blood according 
to the method of Heist and 
Lacy. 
1 Jour. Immunology, 1920, 5, 51. 
2 Jour. Infect. Dis., 1922, 31, 116. 380 
BACTERIOLYSINS 
BACTERIOLYTIC SERUMS IN THE TREATMENT OF DISEASE 
While bacteriolysis is readily demonstrated with some bacteria both 
in vivo and in vitro, nevertheless, when such serums are used therapeutic- 
ally, beneficial results are not dependent solely upon lysis of the infecting 
bacterium, but are usually the result of a combination of lysis with increased 
phagocytosis, due to the simultaneous presence of bacteriotropins (immune 
opsonins). When one recalls the close similarity in general properties that 
exists between bacteriolysins and bacteriotropins, the difference between 
extracellular and intracellular lysis is relatively slight, and tends to strengthen 
Metchnikoff’s views regarding the close relationship of the bacteriolysins 
to leukocytic products and phagocytosis. While extracellular lysis can 
occur in the absence of cells, especially in vitro, yet in the administration of 
antibacterial serums the activities of the leukocytes are much in evidence, 
extracellular lysis or bacteriolysis proper being rather subsidiary to intra- 
cellular lysis or phagocytosis. 
The preparation and methods of standardization of antibacterial serums 
are given in the chapter on Passive Immunization. Most effort in this 
direction has been expended on the preparation of antiserums for the treat- 
ment of meningococcic, streptococcic, pneumococcic, and gonococcic in- 
fections, and the greatest success has been attained with the various anti- 
meningococcic serums. CHAPTER XX 
HEMOLYSINS 
Hemolysis is the term applied to the solution or lysis of red blood-corpuscles. 
Strictly speaking, it would include the disintegration and solution of the 
stroma, although in practice the term is applied to any process in which 
the cells are so injured as to liberate hemoglobin into the surrounding fluids, 
with or without solution of the stroma. 
KINDS OF HEMOLYTIC AGENTS 
Hemolysis or erythrocytolysis may be caused by various non-specific 
physical and chemical substances which act alike upon the erythrocytes 
of a large group of animals, and specific agencies which act upon the cor- 
puscles of one kind of animal. For example, the bacterial hemolysins are 
non-specific and will hemolyze human corpuscles and those of the lower 
animals; serum hemolysins, howrever, are highly specific, and that for human 
corpuscles will act only upon these cells. These hemolytic substances may 
be grouped as follows: 
1. Mechanical agitation. 
2. Temperature changes (heat; alternate freezing and thaw- 
ing). 
3. Water; hypotonic and hypertonic saline solutions. 
4. Acids, salts, and other chemical substances. 
5. Tissue extracts (autolytic products). 
6. Inorganic colloids (silicic acid). 
7. Photodynamic substances (dyes, etc.). 
8. Vegetable poisons (ricin; abrin, crotin, curcin, and robin); 
“saponin substances.” 
9. Bacterial poisons (tetanolvsin; streptolysin; staphylolysin, 
etc.). 
10. Animal secretions (spider poison; cobra venom, etc.). 
Non-specific 
Specific 
1. Natural serum hemolysins. 
2. Immune serum hemolysins. 
Hemolysis by Mechanical Agitation; Temperature Changes.—Red blood- 
corpuscles collected by whipping blood to prevent coagulation are readily 
ruptured; these corpuscles also appear to become slightly more susceptible 
to the hemolytic activity of hypotonic salt solution, serum hemolysins, 
and other specific and non-specific hemolytic agents. For this reason cor- 
puscles are generally secured by collecting the blood in oxalate or citrate 
solutions when fragility tests are to be conducted. 
Alternate freezing and thawing results in hemolysis. Heating to 62° to 
64° C. causes hemolysis of the corpuscles of warm-blooded animals, the 
blood-cells of cold-blooded animals being slightly more susceptible. I 
have found that sheep corpuscles may be heated to 55° C. for fifteen minutes 
without undergoing hemolysis.1 
Hemolysis by Water, Hypotonic, and Hypertonic Saline Solutions.—As 
is well known, red blood-corpuscles undergo rapid hemolysis in water. The 
intravenous injection of large amounts of water may produce hemolysis 
1 Amer. Jour. Syph., 1921, 5, 628. 
381 382 
HEMOLYSINS 
in vivo with the production of a febrile reaction, hemoglobinuria, and other 
symptoms. 
Solutions for intravenous injection should be isotonic, and for ordinary 
purposes 0.7 to 1 per cent, solutions of sodium chlorid in water suffice. 
The average is 0.85 per cent, sodium chlorid in water, well known as physio- 
logic or normal saline solution, and employed for the preparation of red 
corpuscle suspensions in complement-fixation and other hemolytic tests. 
Hypotonic saline solutions contain less than 0.7 per cent, sodium chlorid 
and hypertonic solutions contain more than 1 per cent. The fresh red blood- 
corpuscles of healthy human beings do not begin to hemolyze in hypotonic 
saline solutions until the strength is reduced to approximately 0.45 per cent.; 
in hypertonic solutions hemolysis may be very slight at approximately 
7 per cent., but is not marked or complete even at 10 per cent. The re- 
sistance of corpuscles varies considerably among different persons and 
lower animals in health and disease as discussed in the succeeding para- 
graph; preserved red blood-corpuscles and blood prepared by defibrination 
are more fragile than fresh corpuscles collected in citrate solution. 
Hemolysis in water and hypotonic saline solutions appears to be essen- 
tially a physical process for establishing osmotic equilibrium. Water enters 
the corpuscles, injuring the stroma, and thereby permitting the escape of 
hemoglobin; before this occurs the suspension of cells is opaque because 
of the obstruction to light offered by the corpuscles, but on the completion 
of hemolysis (“laking”) the solution is perfectly transparent and the stroma 
settles to the bottom. 
Hemolysis by hypertonic saline solutions is apparently due to the action 
of sodium chlorid penetrating the stroma. 
Variation in the Resistance of Red Blood-corpuscles to Hypotonic Saline 
Solutions.—Many investigators have studied the tonicity of red blood- 
corpuscles by placing the cells in varying solutions of sodium chlorid and 
other salts in water to determine the concentration where hemolysis begins 
(point of minimal resistance) and where it is complete (point of maximal 
resistance). With solutions of sodium chlorid the following average values 
have been reported for the red blood-corpuscles of normal healthy human 
beings: 
Beginning hemolysis 
Maximal resistance (complete 
(minimal resistance) 
hemolysis). 
Ribierre1 
0.44 
0.34 
Chauffard and Rendu2 
0.46 
0.38 
Paoloni3 
0.44 
0.36 
Hill4 
0.457 
0.34 
Morris5 
0.47 
0.3 
Fontaine6 
0.45 
0.27 
Greenthal and O’Donnell7.... 
0.425 
0.325 
Smith and Brown8 found the red corpuscles of different healthy horses 
to vary in resistance to hypotonic saline solutions, and the same is true of 
the corpuscles of human beings. For this reason minor degrees of variation 
from the average are of little or no significance. 
Variation in results may be influenced by the temperature at which the 
1 Thesis de Paris, 1903, 106. 
2 Presse m6d., 1907, 15, 345. 
3 Policlinico, 1913, No. 6, 243. 
4 Archiv. Int. Med., 1915, 16, 809. 
5 Clinical Laboratory Methods, Appleton & Company, 1913, 264. 
6 Jour. Amer. Med. Assoc., 1921, 77, 1022. 
7 Amer. Jour. Dis. Child., 1921, 22, 212. 
8 Jour. Med. Research, 1906, 15, 425. KINDS OF HEMOLYTIC AGENTS 
383 
tests are made and the time elapsing before the reactions are read; also 
according to whether the corpuscles are used in citrated blood or after being 
washed. Serum appears to protect the corpuscles to a slight degree. In 
my own experience the corpuscles of healthy persons have yielded the 
following average results when tested by the method described in this chapter 
under Practical Applications: 
Washed corpuscles: 0.44 (initial); 0.32 (complete). 
Citrated blood: 0.45 (initial); 0.36 (complete). 
Whole blood: 0.45 (initial); 0.38 (complete). 
The point of initial or beginning hemolysis is more definite in my experi- 
ence than the point of complete hemolysis and the more valuable of the two. 
Many attempts have been made to place the tonicity test on a diag- 
nostic basis, but with little success. In 1907 Chauffard showed that there 
was an increased fragility of the erythrocytes in chronic family (hemolytic) 
jaundice which has been confirmed by numerous investigators. In obstruc- 
tive jaundice, however, resistance is generally increased, so that the fragility 
test possesses practical value in differentiating between hemolytic and 
obstructive jaundice. 
In secondary anemias there are apparently no constant values for the 
points of beginning and complete hemolysis, although in many cases there 
is a slight decrease of resistance. In pernicious anemia the values are like- 
wise inconstant, but generally the cells are more fragile than normal blood 
or in secondary anemias. Hill has found the following averages based upon 
a study of the blood of 19 normal persons, 24 with secondary anemia, and 
13 with pernicious anemia: 
Normal: 0.457 (initial); 0.340 (complete). 
Secondary anemia: 0.475 (initial); 0.322 (complete). 
Pernicious anemia: 0.477 (initial); 0.322 (complete). 
Butler1 has observed increased resistance in some cases of sepsis and 
acute infectious diseases; Greenthal and O’Donnell, however, were unable 
to observe constant changes in scarlet fever, diphtheria, and other infectious 
diseases, although in some cases of great severity a distinct increase in re- 
sistance was noted. Pel,2 Karsner and Pearce,3 and Hill4 have observed 
an increased resistance after splenectomy, and especially the point of max- 
imal resistance. The administration of arsenic (Fowler’s solution, arsphen- 
amin, etc.) has been observed to increase the resistance of erythrocytes 
to hypotonic saline solutions. 
A method for conducting the tonicity test is described in this chapter 
under Practical Applications. 
Hemolysis by Acids, Salts, and Other Chemical Substances.—Organic 
and inorganic acids and alkalies are extremely hemolytic for the corpuscles 
of all animals. The practical application is in relation to the preparation 
of glassware for complement-fixation and other hemolytic reactions. Brown 
and myself5 found that 1 c.c. of a 1 : 300 dilution of normal solution of 
hydrochloric acid, equivalent approximately to 0.000012 c.c. or about 0.000015 
c.c. of commercial acid (81 per cent.), proved hemolytic when added to 
1 c.c. of a per cent, suspension of washed sheep corpuscles. Approxi- 
mately 1 c.c. of a 1 : 100 dilution of normal sodium hydroxid (about 0.0004 
gm.) was likewise hemolytic. 
Chemical agents produce hemolysis by changes in the stroma probably 
by dissolving the lipoids or hydrolyzing the proteins and destroying the 
1 Quart. Jour. Med., 1912, 6, 145. 
* Deut. Arch. f. klin. Med., 1912, cvi, 592. 
3 Jour. Exper. Med., 1912, 16, 769. 
4 Arch. Int. Med., 1915, 16, 809. 
5 Amer. Jour. Syph., 1919, 3, 8. 384 
HEMOLYSINS 
power for retaining hemoglobin. Therefore, only such chemical agents as 
may penetrate corpuscles are actively hemolytic, as ammonium salts, alcohols, 
ketones, esters, antipyrin, amids, neutral amino-acids, urea, bile acids 
and their salts, etc.; likewise lipoid solvents, as chloroform, ether, alcohol, 
by dissolving lecithin and cholesterol of the stroma. Sodium bicarbonate 
solutions of 1 or 2 per cent, are hemolytic, but per cent, solutions are not 
hemolytic. Arseniuretted hydrogen, numerous anilin compounds, nitro- 
benzol, etc., are hemolytic when inhaled and bear an important relation 
to industrial medicine. Curiously, some of these substances, as AsH3, are 
hemolytic in the living body, but not in the test-tube, producing their effects 
through tissue changes, as shown by Friedberger and Brossa1 and others. 
Hemolysis by Tissue Extracts (Autolytic Products).—These are of 
practical importance in relation to immunologic reactions in connection with 
the preparation of tissue extracts employed as antigens for the Wassermann 
test. Saline extracts of tissues may contain specific hemolysins extracted 
from the blood contained in the organ or even from the cells themselves 
of such tissues as bone-marrow and spleen. Decomposition products are 
especially hemolytic, and particularly of the liver, due to the presence of 
fatty acids and bile-salts. Alcoholic extracts of fresh tissues, and particu- 
larly those of muscle, are but feebly hemolytic, whereas extracts of fatty 
tissues are highly hemolytic. 
Hemolysis by Inorganic Colloids.—Landsteiner and Jagic2 have ob- 
served the agglutinating and hemolyzing activities of silicic acid, and several 
investigators have studied its hemolytic activity and especially in combina- 
tion with lecithin and serum. So far all attempts toward utilizing mixtures 
of silicic acid and serum as hemolysin for complement-fixation tests have 
not met with success.3 
Hemolysis by Photodynamic Substances.—Sacharoff and Sachs,4 Pfeiffer,® 
Hausmann,6 and others, have observed the hemolytic effects of various 
dyes, chlorophyll, hematoporphyrin, urobilin, and other substances, but 
the subject does not appear to possess sufficient importance in relation to 
immunology to warrant further discussion. 
Hemolysis by Vegetable Poisons.—A number of plant poisons, the so- 
called “vegetable toxalbumins,” have been discussed in preceding chapters 
in connection with the toxins and antitoxins as owing a portion of their 
toxic effects to hemolytic activity. Of these, crotin and curcin are especially 
hemolytic, while ricin, abrin, and crotin are feebly hemolytic, but actively 
agglutinating. While their effects are non-specific, yet they vary greatly 
according to the species of animals whose blood is used. 
The “saponin substances,” classed as glucosids and found in at least 
forty-six different families of plants, are active hemolytic poisons. The 
mechanism of hemolysis is unknown, but it is probable that they injure 
the stroma by combining with or dissolving the lipoids. The addition of 
cholesterol and normal serum to saponin inhibits hemolysis, but compounds 
of saponin and lecithin are actively hemolytic. Their effects in vivo and 
in vitro have been studied extensively by Kobert.7 
1 Ztschr. f. Immunitatsf., 1912, 15, 506. 
2 Wien. klin. Wchn., 1904, No. 3; Munch, med. Wchn., 1904, No. 27. 
3 Literature reviewed by Landsteiner, Handbuch der Biochemie des Menschen und der 
Tiere, Fischer, Jena, 1909, 444. 
4 Munch, med. Wchn., 1905, 297. 
6 Wien. klin. Wchn., 1905, 221. 
6 Biochem. Ztschr., 1908, 12, 331; ibid., 14, 275. 
7 Archiv. f. exp. Path. u. Pharm., 1887, 23, 233. Also Die Saponinsubstanzen, Stutt- 
gart, 1904. KINDS OF HEMOLYTIC AGENTS 
385 
McNeil1 has applied the saponin hemolytic test to the study of red blood- 
corpuscles of human beings with various diseases. Resistance was greatly 
diminished in jaundice and the anemias, particularly in severe chronic 
jaundice and pernicious anemia. In diabetes mellitus and febrile condi- 
tions an increased resistance was frequently found. Neilson and Wheelon2 
have studied this reaction with particular care in relation to disease. With 
their method the average minimal degree of hemolysis of red blood-corpuscles 
in whole blood of healthy individuals was 1 : 13,937 sapotoxin solution, 
read after five minutes at 25° C. In cases of obstructive jaundice, preg- 
nancy, pulmonary tuberculosis, lead-poisoning, and cardiorenal diseases 
an increased resistance was commonly observed; in hemolytic jaundice 
resistance was decreased, as likewise in the anemia of syphilis. In secondary 
anemias of carcinoma and tuberculosis and in pernicious anemia increased 
resistance was sometimes found. Washed corpuscles were more susceptible 
than corpuscles suspended in serum, and the cholesterol content of the blood 
appeared to run parallel to the degree of resistance offered by the cells. 
In my experience saponin (Merck’s) produces initial hemolysis of the 
citrated blood of healthy persons in dilution of about 1 : 30,000 according 
to the method described in this chapter; with washed corpuscles hemolysis 
occurs in dilutions above 1 : 40,000. 
Hemolysis by Bacterial Poisons.—In the chapter on Toxins mention 
was made of the hemolytic poisons produced by many different bacteria, 
mostly pathogenic bacteria. Doubtless these hemotoxins are partly respon- 
sible for the production of the anemias developing during the course of 
many of the bacterial infections and notably those caused by streptococci 
and staphylococci. 
The bacterial hemotoxins were first described by Koch,3 followed by the 
extensive studies of Ehrlich4 and Madsen5 upon the hemolytic poison pro- 
duced by the tetanus bacillus (tetanolysin), and especially for the red blood- 
corpuscles of the horse and rabbit. Since then a very large literature has 
accumulated on this subject6 and the following bacteria are known to pro- 
duce hemolytic poisons under favorable conditions: 
Tetanus bacilli (tetanolysin). 
Staphylococci (staphylolysin). 
Streptococci (streptolysin). 
Pneumococci (endotoxin) (pneumotoxin). 
Many vibrios (vibriolysins). 
Proteus bacillus (proteuslysin). 
Bacillus megatherium (megatheriolysin). 
Bacillus pestis (pestolysin). 
Bacillus of malignant edema. 
Colon, typhoid, and dysentery bacilli (cololysin, typholysin, dysentero- 
lysin). 
Bacillus pyocyaneus (doubtful) (pyocyanolysin). 
Bacillus welchii. 
These bacterial hemolysins are non-specific in that they will act upon 
the corpuscles of different animals, although some are affected more readily 
than others. Some are heat sensitive, and the immune sera of many are 
capable of neutralizing the hemolysins. 
1 Jour. Path, and Bacteriol., 1911, 15, 56. 
2 Jour. Lab. and Clin. Med., 1921, 6, 454, 487. 
3 Berl. klin. Wchn., 1884, 477. 4 Berl. klin. Wchn., 1898, 273. 
5 Ztschr. f. Hyg., 1899, 32, 214; ibid., 32, 239. 
6 Summarized by Landsteiner, Handbuch der Brochemie des Menschen und der Tiere, 
Fischer, Jena, 1909, 451. 386 
HEMOLYSINS 
The properties of these lysins are more or less characteristic of the micro- 
organisms producing them. Staphylolysin, for example, is destroyed by 
heating to 56° C. for twenty minutes, while streptolysin requires 70° C. 
for two hours, and typholysin is said to resist boiling. Connell and Holly1 
believe that streptolysin and megatheriolysin are specific fat complexes of 
the bacteria in definite colloid states. 
Hemolysis by Animal Secretions.—Hemolytic poisons produced by 
spiders were observed in 1888 by Jousset and Sanarelli, and since then a 
large number of these poisons produced by spiders, various insects, fishes, 
and snakes have been studied. Especial attention has been given the hemo- 
lytic activity of cobra venom, first described by Stevens and Myers,2 and 
since then by Mvers,3 Stephens,4 Mitchell and Flexner,5 Flexner and No- 
guchi,6 McFarland and Weston,7 and many others. 
The venoms appear to resemble the hemolysins to be found in the blood 
and may be activated by serum (complement) in much the same manner 
as the serum hemolysins. Cobra venom is most active, followed in order 
by the poisons of the water moccasin, copperhead, and rattlesnake. 
The red blood-corpuscles of different animals vary in resistance to 
venom hemolysis, those of the dog being very susceptible. Among human 
beings the resistance varies in disease, being increased in the latter stages of 
syphilis, as shown by Weil,8 and after splenectomy, as observed by myself.9 
This subject is considered in more detail in the chapter of Venom Hemol- 
ysis. 
To this category of hemolytic substances belong the secretions of cer- 
tain parasites, as Bothriocephalus latus and Ankylostoma; also the bile 
and pancreatic juice of mammals. These hemolytic substances have been 
generally identified with lipoids, as likewise the hemolytic activity of watery 
and salt solution extracts of tissues. 
The hemotoxins in eel-serum are also to be classified in this category. 
This serum is extremely toxic and will kill the rabbit and guinea-pig when 
injected intravenously in amounts as small as 0.1 c.c. per kilo of body weight 
as shown by Mosso,10 Camus and Gley,11 Kossel,12 and others. Part of this 
toxicity is due to its remarkable hemolytic activity. The hemotoxin is 
thermolabile, being destroyed by heating to 55° C. for fifteen minutes; it 
resists the deteriorating influence of desiccation, but is easily destroyed 
by acids and alkalies. Animals may be immunized against this serum. 
Hemolysis by Serum.—Just as bacteria may be killed and possibly broken 
up by specific bacteriolytic amboceptors and complements, so, in like manner, 
hemolysis may he caused by specific hemolytic antibodies known as hemolysins. 
Working in unison with complements the mechanism of both bacteriolysis 
and serum hemolysis is probably identical. The simplicity of hemolytic 
experiments, the rapidity with which they may be performed and termi- 
nated, and the ease with which hemolysis may be observed by the naked 
eye have rendered the specific serum hemolysins particularly useful in the 
study of amboceptors and of complements, and of cytolytic phenomena 
in general. In fact, bacteriolysis was not thoroughly understood until 
Bordet discovered the hemolysins, and demonstrated the analogy that 
exists between bacteriolysis and hemolysis, a discovery that led to a vast 
1 Jour. Bacteriology, 1921, 6, 89. 
2 Brit. Med. Jour., 1898, 621. 
3 Jour. Path, and Bacteriol., 1899-1900, 415. 
4 Jour. Path, and Bacteriol., 1899-1900, 273. 
6 Nat. Acad. Sci., 1901. 
6 Jour. Exper. Med., 1902, 3, 277. 
7 Jour. Amer. Med. Assoc., 1909, 43, 845. 
8 Jour. Infect. Dis., 1909, 6, 688. 
9 Jour. Exper. Med., 1917, 25, 195. 
10 Arch. Exp. Path., 1889, 25, 101. 
11 Compt. rend. Soc. d. biol., 1898,126, 428 
12 Berl. klin. Wchn., 1898, 152. SERUM HEMOLYSINS 
387 
amount of research work and controversy, to many important discoveries, 
and to the final evolvement of diagnostic reactions of great value. 
SERUM HEMOLYSINS 
Historic.—For many years physiologists were aware that the bloods of 
various animals transfused into man or animals of a different species were 
more or less directly injurious, and incapable of replacing human blood. 
In 1875 Landois demonstrated experimentally that while transfusion of 
blood from one animal to another of a different species may prove injurious 
and even fatal, transfusion to an animal of the same or of very closely re- 
lated species produced no ill effects. The explanations offered were inade- 
quate, until later researches on the hemolysins showed that the normal 
blood-serum of one animal may contain hemolysins for the erythrocytes 
of other animals, and, consequently, upon transfusing this blood to another 
animal the hemolysin acting with the complement present produced hemol- 
ysis in vitro, thereby explaining in part the toxicity of the alien blood. 
In 1898 Belfanti and Carbone made the observation that the serum 
of a horse receiving several injections of rabbit blood was toxic for rabbits, 
whereas normal horse-serum was without injurious effects. 
At about the same time Bordet published his epoch-making discoveries. 
He observed that while normal guinea-pig-serum had little or no hemo- 
lytic action on rabbit erythrocytes, the serum of a pig that had received 
a few intraperitoneal injections of rabbit blood was able quickly and com- 
pletely to hemolyze rabbit blood, just as an animal may acquire, through 
immunization with cholera, the property of dissolving cholera vibrios. 
He demonstrated further that this acquired hemolytic activity was highly 
specific, for when animal A was immunized with the corpuscles of animal 
B the serum of A acquired the power of hemolyzing only the erythrocytes 
of B, and possibly of other animals closely related zoologically. It was 
found also that the hemolytic activity of an immune serum was lost by 
age or could be removed by heating; that in either case the serum could 
be reactivated by the addition of a little normal serum or peritoneal exudate 
—phenomena closely resembling that observed in bacteriolysis, and due to 
the action of a thermolabile body or alexin and a second and specific thermo- 
stabile antibody named by Brodet the “substance sensibilisatrice.” 
These observations were soon confirmed by Landsteiner and von Dungern, 
and were followed by very extensive studies by Ehrlich and Morgenroth, 
who likewise confirmed Bordet’s experiments, but offered a different ex- 
planation for the mechanism of the phenomenon, according to the side- 
chain theory, and renamed the alexin and sensitizing substance concerned 
in the process “complement” and “hemolytic amboceptor” or “immune 
body,” respectively. 
Definition.—Serum hemolysins (Gr., dlp.a = blood + XOscv = to dis- 
solve) are antibodies in a serum that, when acting with complement, have the 
power of “lysing” or breaking up red blood-corpuscles, or so altering their 
envelope as to allow the hemoglobin to escape. 
Nomenclature.— Normal or natural hemolysins are those found in normal 
serums; specific or immune hemolysins are those produced as the result of 
the injection of blood-corpuscles from an animal of a different species. 
Heterolysins are the hemolysins formed by immunization with corpuscles 
of a different species (the immune hemolysins). Isolysins are hemolysins 
for the corpuscles of animals of the same species. Autolysins are hemol- 
ysins that act upon the corpuscles of the same animal and are quite rare. 388 
HEMOLYSINS 
It should be remarked that isolysin and autolysin are not strictly synony- 
mous terms, as the former does not act upon the corpuscles of the animal 
producing the hemolysin, but may hemolyze the corpuscles of other animals 
of the same species. For example, Ehrlich has shown that the serum of a 
goat that had received several injections of blood from other goats, although 
actively hemolytic for the corpuscles of goats 1, 2, 4, 5, 6, 9, and less so 
for goats 3 and 8, was not able to hemolyze those of goat 7 or of itself at all. 
This immunity of the corpuscles of an animal to its own isolysin was subse- 
quently shown to be due to a complete absence of suitable receptors in its 
corpuscles. Therefore in cases in which a large internal hemorrhage occurs 
and the blood is absorbed an autohemolysin is not produced, or produced 
only in small amounts, and likely to be followed by the formation of an 
anti-autolysin which regulates the process of blood destruction within 
physiologic limits. Although little is known concerning the processes of 
normal blood destruction, as in the disposal of old erythrocytes, it is pos- 
sible that an autohemolysin is being produced, and that its activity is held 
within normal limits by an anti-autolysin. A disturbance of this physiologic 
equilibrium may then be the basis of certain types of primary anemia char- 
acterized by excessive blood destruction. 
The Nature of Hemolysinogens.—This refers to the nature or portion 
of erythrocytes concerned in the production of hemolysins when animals 
are injected with alien corpuscles. 
A large amount of investigation has been devoted to this subject.1 Bordet2 
has attributed the antigenic properties to the stroma of erythrocytes, while 
Nolf3 believes that this portion produces hemagglutinins and extracts of 
the cells, the hemolysin. Bradley and Sansum4 regard hemoglobin as anti- 
genic, but Ford and Halsey5 were unable to produce hemolysins with purified 
hemoglobin; Levene6 likewise found that injections of pure crystalline 
hemoglobin failed to produce hemolysins, while solutions of erythrocytes 
in | per cent, solutions of sodium carbonate were antigenic. It would 
appear, therefore, that some constituent of the erythrocyte other than 
hemoglobin is concerned in the production of hemolysins; the stroma are 
surely antigenic, and Bennett and Schmidt7 have recently found that a 
carbon dioxid globulin from washed erythrocytes or a substance intimately 
associated with it, is the antigen concerned in the production of hemolysins. 
The chemical nature of the hemolysinogens is of greater importance 
and interest. According to Guerrini,8 the nucleoproteins of dog’s erythro- 
cytes are antigenic; Landsteiner and Javic,9 Eisler,10 Bang and Forssmann, 
and others, however, believe that the lipoids are antigenic. The latter 
found that ethereal extracts of red blood-corpuscles gave rise to the produc- 
tion of hemolysins, but not of agglutinins on immunization of animals. 
Others have failed to confirm these observations. Jobling and Bull11 have 
found that the immunization of rabbits with hen corpuscles was followed 
by an increase of the serum lipases, and suggest that the lipoids may be 
1 Oppenheimer, Handbuch der Biochemie des Menschen und der Tiere, Fischer, Jena, 
1909, 521. 
2 Ann. de l’lnst. Pasteur, 1898, 12, 688; 1899, 13, 225, 273; 1900, 14, 257. 
3 Ann. de l’lnst. Pasteur, 1900, 14, 297, 656. 
4 Jour. Biol. Chem., 1914, 18, 497. 
6 Jour. Med. Research, 1904, 11, 403. 
6 Jour. Med. Research, 1904, 12, 191. 
7 Jour. Immunology, 1919, 4, 29. 
8 Munch, med. Wchn , 1904, No. 27. 
9 Wien. klin. Wchn., 1904 676; Centralbl. f. Bakteriol., 1905, 39, 309. 
10 Centralbl. f. Bakteriol., 1905, 40, 151. 
11 Jour. Exper. Med., 1912, 16, 483. SERUM HEMOLYSINS 
389 
antigenic. Vedder1 was unable to produce hemolysins with ether extracts 
of corpuscles, or with globulin from stroma, but the protein extract left 
after the removal of globulin as well as lipoid-free stroma, produced hemol- 
ysin; essentially similar results were obtained by Chodat.2 The consensus 
of opinion appears to be to the effect that protein-free lipoids are not anti- 
genic, that the pure lipoids are not hemolysinogenic, and that the active 
antigenic substance is contained in the stroma of corpuscles. 
Production of Hemolysins by Heterologous Antigens.—While antisheep 
hemolysin, for example, is best prepared by injecting the rabbit or some 
other animal with sheep corpuscles, Forssman3 has demonstrated that this 
hemolysin could be prepared by immunizing rabbits with such heterologous 
antigens as extracts of the organs of the guinea-pig, horse, cat, dog, mouse, 
hen, pigeon, turtle, and even of some bacteria, as Bacillus paratyphosus 
B. Orndschiew4 has demonstrated similar heterologous antigens in serum, 
and Doerr and Pick,5 in urine. Doerr and Pick and Morgenroth6 found that 
when the organs of an animal contained an antigen for antisheep hemolysin 
the corpuscles of this animal did not, and vice versa, but Kritchevsky7 found 
that nucleated corpuscles (hen) contained the hemolysin-producing antigen 
for sheep corpuscles just as the organs of the hen contain this substance. 
Kritchevsky also found that the connective tissue of guinea-pigs contain 
this antigen-producing hemolysin against sheep corpuscles. 
As shown by Forssman and Hintze,8 the immune sera may also contain 
hemagglutinins for sheep corpuscles, as the result of immunization with 
heterologous antigens. These have been discussed in the chapter on Agglu- 
tinins. 
These observations have not been adequately explained. The sera of 
the majority of rabbits show the presence of natural antisheep hemolysin, 
and in my opinion the production of this hemolysin may be increased by 
non-specific stimulation by immunization with a variety of heterologous 
antigens, including not only those mentioned above, but even by certain 
chemical compounds, as arsphenamin and mercuric chlorid, as shown by 
the experiments of Toyama and myself.9 I have not been able to produce 
hemolysin for human corpuscles by injecting rabbits with the heterologous 
antigens mentioned, and natural antihuman hemolysin is rarely found in 
these animals. 
The Nature of Serum Hemolysins.—The nature of these hemolysins 
is unknown. Evidently they are not ordinary chemical substances, since 
they do not act according to the laws of chemical equivalents and definite 
proportions. Many investigators have assumed that they were of the 
nature of lipolytic ferments; Bergel10 has stated that the sera of animals 
immunized to foreign erythrocytes are about twice as active in splitting 
foreign fats as the sera of untreated animals. Jobling and Bull11 likewise 
found that the sera of rabbits immunized with hen corpuscles acquired 
an increased lipolytic action, and that a portion of lipases were specific 
for the corpuscles. Later lipases were more closely identified with the com- 
plements than with the hemolysins. 
1 Jour. Immunology, 1919, 4, 141. 
2 Compt. rend. Soc. de biol., 1921, 85, 735. 
3 Biochem. Zeitschr., 1911, 37, 78. 
4 Ztschr. f Immunitatsf., 1912-13, 16, 268. 
6 Biochem. Zeitschr., 1913, 1, 129; Ztschr. f. Immunitatsf., 1913, 19, 251, 612. 
6 Berl. klin, Wchn., 1913, 1, 561. 
7 Jour. Exper. Med., 1916, 24, 233. 
8 Biochem. Zeitschr., 1912, xliv, 336. 10 Deut. Arch. f. klin. Med., .1912, cvi, 47. 
9 Jour. Immunology, 1918, 3, 326 11 Jour. Exper. Med., 1912, 16, 483. 390 
HEMOLYSINS 
Undoubtedly the serum lipases are concerned in the phenomenon of 
hemolysis, but appear to be more closely identified with the activity and 
nature of the complements than with the hemolysins. Investigations bear- 
ing upon the nature of the immunizing substances of erythrocytes discussed 
above indicate that the protein-free lipoids of these cells are not antigenic. 
The hemolysins produced by bacteria are regarded by Connell1 and Holby2 and 
others, however, as the bacterial fats in colloidal states. Hemolytic substances 
may be extracted from serum by ethyl alcohol and other fat solvents as shown 
by Woelfel5 and others, but these are not strictly specific. As discussed 
in the succeeding paragraphs, Metchnikoff has always regarded hemolysins 
as ferments derived from cells, but has failed to establish their identity and 
nature; they are commonly regarded as being identified with the globulin 
fraction of serum. At the present time, however, the hemolysins cannot 
be said to have been shown to be true ferments, proteolytic or lipolytic, 
although the process of serum hemolysis is apparently concerned with the 
action of serum lipases. 
The Mechanism of Serum Hemolysis.—The side-chain theory of Ehrlich 
affords a working hypothesis upon the interaction of corpuscles, hemolysin, 
and complement as described below, but does not throw any light upon 
the nature and mechanism of serum hemolysis. Bordet believes that the 
hemolysin injures the stroma and prepares it for the destructive action of 
complement resulting in changing the re- 
sistance of the cell to osmotic influences, 
but as far as I know he has not offered 
an explanation of the mechanism of these 
changes. 
The exact mechanism is not known and 
probably never will be until the nature 
of hemolysin has been determined. Un- 
doubtedly the lipases of the serum and 
particularly of that portion designated as 
alexin or complement are concerned. Land- 
steiner, Neuberg, Bergel, and others regard 
hemolysis as a phenomenon due to the splitting or solution of the lipoids 
out of erythrocytes. Bergel4 has advanced the following hypothesis: the 
lipoids of erythrocytes are antigenic, and the hemolysins are specific zymogens 
formed by the lipoids of lymphocytes. The zymogens (hemolysins) are 
specifically bound by the lipoids of the corresponding corpuscles, which are 
then activated by the non-specific complement followed by solution of the 
lipoids of the corpuscles and hemolysis. 
The Interaction of Hemolysin, Complement, and Corpuscles.—According 
to the side-chain theory, hemolysins are amboceptors or antibodies of the 
third order, requiring the action of a complement before hemolysis can be 
produced (Fig. 126). Bordet5 showed that two substances were concerned 
in the phenomenon of serum hemolysis, although his views on the mechanism 
of the process differ from those advanced by Ehrlich and Morgenroth.6 
Theory of Ehrlich and Morgenroth.—Ehrlich argued that the hemolytic 
amboceptor or hemolysin must be an antibody to the receptors of the red 
blood-corpuscles used in the process of immunization, and if this is true, 
it ought to unite with the corpuscles. Taking the serum of a goat that had 
Fig. 126.—Theoretic Structure of 
a Hemolytic Amboceptor (Hemol- 
ysin). 
A, Amboceptor; C, complement; r, 
receptor of the corpuscle; h, haptophore 
group of the amboceptor; h.a., comple- 
mentophile group of the amboceptor. 
1 Jour. Exper. Med., 1913, 17, 61. 3 Jour. Infect. Dis., 1905, 2, 97. 
2 Jour. Bacteriology, 1921, 6, 89. 4 Ztschr. f. Immunitatsf., 1918, 27, 441. 
8 Ann. d. l’lnst. Pasteur, 1899, 13, 273; 1901, 15, 312. 
6 Berl. klin. Wchn., 1899, 6, 481. SERUM HEMOLYSINS 
391 
been injected with and was hemolytic for the erythrocytes of a sheep, he 
destroyed the complement by heating the serum to 56° C. To this he added 
some sheep’s corpuscles, and allowed the mixture to stand for a short time 
at room temperature, after which it was centrifuged and the supernatant 
fluid pipeted off in another test-tube. No hemolysis had occurred, and the 
corpuscles were to all appearances unaltered, but it was now found that 
if a small amount of normal goat-serum, as complement, was added to 
the corpuscles and the mixture placed in the incubator, hemolysis occurred. 
By adding sheep’s corpuscles and normal goat-serum (complement) to the 
supernatant fluid that had been removed to a separate test-tube, hemol- 
ysis did not occur. This experiment indicated, therefore, that the red blood- 
had combined with all the antibody. That the action was specific 
was shown by the fact that the corpuscles of other animals, such as rabbits 
or goats, for example, exerted no combining power when used instead of the 
sheep’s cells. The union between cell and antibody was considered as 
being in the nature of a chemical combination and quite firm, as repeated 
washing of the cells with normal salt solution did not break it up. 
Having shown that the antibody had a combining affinity for the cells, 
it was important to solve the question of the relation of alexin or comple- 
ment to the process. In other words, does this substance unite directly 
with the cell or does it unite with the antibody and thus indirectly with the 
cell? 
Ehrlich and Morgenroth studied this by means of a similar experiment. 
Sheep’s corpuscles were mixed with normal goat-serum (complement), 
and after a time the mixture was centrifuged and the two portions tested 
separately. To the corpuscles heated immune serum was added, but hemol- 
ysis did not result. To the supernatant fluid corpuscles and heated im- 
muhe serum were added and hemolysis occurred. This indicated that the 
alexin or complement did not unite with the corpuscles, as did the anti- 
body in the first experiment, but remained free in the supernatant fluid. 
By mixing corpuscles, immune serum, and complement, and keeping 
the mixture at 0° to 3° C. for several hours, hemolysis did not take place. 
By centrifuging the mixture and separating the supernatant fluid from 
the corpuscles, a similar test showed that the red cells had combined with 
all the antibody, but had left the alexin practically undisturbed'. 
At higher temperatures these relations were more difficult to demon- 
strate, as hemolysis occurs rapidly, but by leaving the cells and serum in 
contact for short periods of time, centrifuging rapidly, and testing the 
corpuscles and supernatant fluid in the same manner, similar relations 
were found to exist. 
These experiments led Ehrlich to formulate the theory that complement 
will unite only with the antibody and not with the red corpuscles, but that 
it acts upon the corpuscles when united indirectly by means of the anti- 
body. As will be pointed out further on, this view is in direct opposition 
to that of Bordet, who does not accept this interpretation, but believes 
that the complement acts directly upon the corpuscles. 
Ehrlich, therefore, conceived the antibody as being in the nature of 
an amboceptor or of an interbody between an antigen and complement, 
with two combining arms—one the cytophil haptophore for union with 
the cell, and the second, the complementophil haptophore for union with 
complement. The amboceptor is unable in itself to injure the cell, but 
preserves its importance in being the only and specific means by which the 
ferment or complement can attack the cell and cause its destruction. 
The process of specific serum hemolysis is, therefore, supposed to be 392 
HEMOLYSINS 
as follows: In fresh immune serum containing both amboceptors and com- 
plement, or in a mixture of old or heated immune serum and fresh normal 
serum (i. e., of amboceptor and complement), the two substances occur 
independently of each other. When the corpuscles corresponding to the 
amboceptor are added, the amboceptor unites with these and the comple- 
ment unites with the amboceptor, the amboceptor, therefore, standing mid- 
way between corpuscles and complement. When these unions have taken 
place, hemolysis will result. The amboceptor has a greater affinity for the 
corpuscles than the complement has for the amboceptor, and wall unite 
with the cells at a low temperature, whereas the complement unites wdth 
the amboceptor only very slowly at low temperatures. Body temperature 
favors a quicker union of both, and especially that of complement with 
amboceptor. Hemolysis, therefore, may occur at low temperatures, but is 
hastened by higher temperatures, and occurs best at 37° C. 
As stated in a previous chapter, Ehrlich believes that a great many 
complements exist in normal serums, which view is in direct opposition 
to the “unitarian theory” of Bordet, which holds that the one alexin or 
complement will act with any sensitizer or amboceptor. Of the large num- 
ber of complements, each is especially adapted for the solution of one or 
more varieties of cells, which it can dissolve in conjunction writh a suitable 
amboceptor. This is known as the main or dominant complement; other 
complements that may aid in the process are termed “non-dominant.” In 
general, it is held that those complements that are especially active in 
hemolysis are but slightly active in bacteriolysis, and vice versa. Although 
the amboceptor is depicted as having but one haptophore arm for the domi- 
nant complement, it is really a polyceptor, and is so constituted that it 
combines with the cell to be dissolved, on the one hand, and with a number 
of complements, on the other. 
Theory of Bordet.—Bordet has never accepted these views. He holds 
that the antibody is not an amboceptor for uniting cell and complement, 
but that it sensitizes the cell and renders it susceptible to the direct lytic 
action of the alexin or complement. According to his viewrs, both antibody 
and complement may unite directly with the cell, and he has borne out this 
belief by making experiments almost exactly similar to those made by 
Ehrlich. 
In accepting Ehrlich’s view, it is a question of considerable practical 
importance whether the complement may unite directly with free ambo- 
ceptor. Ehrlich maintains that the two may enter into a loose and easily 
dissociated chemical combination, which is hastened by heat and retarded 
by cold. The union of hemolytic amboceptor in cobra venom with the 
lecithin of corpuscles (Kyes’ cobra lecithid), which acts as complement, 
while it tends to strengthen this view, can hardly be accepted as direct 
proof, as lecithin differs markedly from the ordinary complements found 
free in serum. Likewise, the theory of Neisser and Wechsberg regarding 
complement deviation, whereby it appears that an excess of amboceptor 
may combine directly with complement and in this manner rob those ambo- 
ceptors that are attached to cells of the complement necessary to produce 
lysis, is quite complicated, and is not universally accepted, the evidence of 
direct union of complement and amboceptor having not been proved be- 
yond the peradventure of a doubt. It follows, then, as Emery has stated, 
that we must either assume that the complementophil haptophore of an 
amboceptor united with its antigen has an increased affinity for complement 
over and above that of free amboceptors, or we must agree with Bordet 
that cell and complement unite directly after the former has been sensitized SERUM HEMOLYSINS 
393 
by the action of the antibody. This latter view of the separate union of 
cell with antibody and complement is supported by the observations of 
Muir, who found, upon saturating red blood-corpuscles with antibody and 
then with complement, that some of the former, but none of the latter may 
be dissociated from the combination and become free in the fluid. How- 
ever, the dissociated amboceptors found free of complement may be those 
that did not have time to unite with complement, and there does not appear 
to be any direct and positive experimental evidence to permit one to decide 
between Ehrlich’s and Bordet’s views. 
While Ehrlich believes that the union between cell and amboceptor 
is a chemical one and follows ordinary chemical laws, obeying the law of 
multiple proportions, Bordet holds that the antibody acts as a mordant 
and sensitizes the cell, comparing the process to the staining of filter-paper 
when immersed in a dye, or to the use of mordants preparatory to the 
staining of flagella of certain bacteria. For example, 0.4 c.c. of a hemo- 
lytic serum, if added at once, was found to dissolve 0.5 c.c. of corpuscles. 
If 0.2 c.c. of corpuscles were first added, and amounts of 0.1 c.c. were added 
subsequently, no lysis took place after that of the first portion added. Bordet 
cited this as an example of a physical process of the nature of absorption, 
just as filter-paper when added at once to a dye will be stained a uniform 
color, whereas if it be added a little at a time, the first pieces inserted will 
be stained deeply, the subsequent ones less and less so, until the dye is com- 
pletely absorbed. The process is probably not entirely a physical one as 
believed by Arrhenius,1 who states that “the immune bodies are probably 
not bound by the erythrocytes, but only absorbed by them.” The phenom- 
enon is greatly influenced by the concentration of the antibody in the serum, by 
the temperature at which corpuscles and hemolysins are brought together, by the 
length of exposure, reaction of the medium, and the amount of inorganic salts. 
Ordinarily the union of corpuscles and hemolysins is quite rapid, fifteen 
minutes at 20° to 37° C. being usually sufficient, and Cromwell2 states that 
absorption is more rapid with serum drawn early in the period of im- 
munization. Recent researches on the colloidal theory of antibodies 
would indicate that the hemolysins are governed by very complex chemico- 
physical laws, not as yet fully understood, which regulate the action of col- 
loids on one another and are probably intimately concerned in the processes 
under discussion. 
Theory of Metchnikoff.—As was stated in a previous chapter, Metchni- 
koff maintained that both substances concerned in hemolysis are ferments, 
and that both are adapted for intracellular digestion. He regards comple- 
ment or his cytase as a digestive ferment derived from leukocytes, and be- 
lieves that it is set free only when leukocytes are dissolved (phagolysis) 
either as the result of the injection of a foreign substance or during the 
process of coagulation. Amboceptor or his “fixateur” is likened to entero- 
kinase, and like it acts as an accessory ferment that unites the more potent 
ferment (cytase) to the particle to be digested. He also regards it as being 
derived from leukocytes, and considers that the amount formed depends 
upon the degree of phagocytosis that occurs during the absorption of the 
antigen. In the conception of immunity as being fundamentally a process 
of nutrition, and in the belief of the existence of more than one comple- 
ment, the similarity between the views of Ehrlich and those of Metchnikoff 
is indeed striking. 
Analogy Between Bacteriolysis and Hemolysis.—Studies in hemolysis 
aided greatly in a correct understanding of the mechanism of bacteriolysis. 
1 See Immunochemistry, MacMillan, 1907, 144, 257. 2 Jour. Immunology, 1922, 7,461. 394 
HEMOLYSINS 
It became apparent that two substances were concerned in bacteriolysis 
—one, the thermostabile amboceptor present in the immune serum, and 
the second, thermolabile alexin or complement, furnished by the peritoneal 
exudate in the Pfeiffer test, or by any fresh normal serum in the test-tube 
bacteriolytic reaction. The discovery and study of the specific serum hemol- 
ysins aided greatly in a better understanding of bacteriolysis and cytolysis 
in general. 
Specificity of Hemolysins.—The hemolysins are highly specific anti- 
bodies, and although partial or group hemolysins may be formed and act 
upon the corpuscles of closely related species, as antihuman hemolysin on 
the corpuscles of the higher apes, yet the main hemolysin may be so potent 
that high dilution of the immune serum practically rules out the activities 
of group hemolysins, the hemolytic amboceptor proving highly specific for 
the alien corpuscles responsible for its production. 
Distribution of Hemolysins.—Hemolysins are regularly found in the 
blood plasma as well as in the serum; the studies of Watanabe in my labora- 
tory have shown that this is true of both natural and immune hemolysins. 
When hemolysins are present in large amounts in the plasma they may 
occur in inflammatory exudates, but only exceptionally in transudates and 
such fluids as the normal cerebrospinal fluid and aqueous humors of the eye. 
In paresis, suppurative meningitis, and other conditions with increased 
permeability of the choroid plexus and meningeal hyperemia, hemolysins 
and, notably, natural antisheep hemolysin, may be found in the spinal fluid. 
Weil and Kafka have devised a test for this hemolysin in spinal fluid in 
connection with the diagnosis of these conditions. 
Secretions rarely contain hemolysins unless the organ is markedly hy- 
peremic, as colostrum and milk during the first few days before and after 
delivery. Urine is hemolytic due to acids and other substances, and only 
exceptionally contains specific hemolysins. 
NORMAL OR NATURAL HEMOLYSINS 
Just as small amounts of antitoxins, agglutinins, and opsonins may be 
found in normal serums, so also natural hemolysins may be present. Their 
most important practical significance is concerned with complement-fixation 
reactions, where a close and intimate quantitative relation exists between 
the complement and hemolytic amboceptor used. As an excess of hemo- 
lytic amboceptor may produce hemolysis with a decreased amount of com- 
plement in a given test for free complement, as in Wassermann’s syphilis 
reaction, the patient’s serum may contain so much natural antisheep ambo- 
ceptor as to make up for slight binding of complement and give undue 
hemolysis or even a false negative result. 
Hemolysis of alien corpuscles by a normal serum is found to depend 
upon the same mechanism of amboceptor and complement as in the artifi- 
cial immune serums. The amboceptors are easily removed by adding the 
corresponding corpuscles to the cold serum and centrifuging the mixture 
after allowing it to stand at 0° to 5° C. for an hour or two. The supernatant 
fluid will now be found to be free from amboceptors, whereas by adding a 
little normal complement serum to the corpuscles hemolysis results, indi- 
cating that the amboceptors had been bound to these. If a normal serum 
contains several different amboceptors for as many different bloods, all 
may be removed at the one time by adding the respective corpuscles to 
the serum and allowing sufficient time to elapse for the amboceptors to be- 
come linked to their corpuscles. NORMAL OR NATURAL HEMOLYSINS 
395 
Normal serum, therefore, probably contains numerous antibodies of the 
amboceptor type adapted to dissolve various foreign substances when they 
gain access to the blood. It is probable that, aside from hemolysins, various 
normal and immune cytolysins play an important part in the processes of 
immunity. 
While the normal hemolysins that may be found in the serums of various 
animals have not as yet been fully worked out, the following table, compiled 
by Sachs,1 is a resume of the work reported in the literature on the subject: 
Natural Hemolysins 
Hemolyze the 
The Serum of 
Erythrocytes 
OF THE— 
3 
rO 
J.£ 
fl 
d 
(U 
tub 
<L> 
0> 
s 
b4 
o 
o 
<u 
G 
o 
o 
§ 
o 
o 
C/3 
x 
X 
o 
Q 
ST 
X 
Rabbit 
0 
+ 
+ 
+ 
+ 
+ 
+ 
+ 
zb 
+ 
+ 
+ 
+ 
Guinea-pig 
+ 
0 
+ 
+ 
+ 
+ 
+ 
+ 
db 
+ 
+ 
db 
+ 
Dog 
(+) 
=b 
0 
Man 
=b 
+ 
+ 
0 
+ 
+ 
=±= 
db 
— 
+ 
+ 
=fc 
+ 
Goat 
+ 
+ 
+ 
0 
+ 
— 
+ 
Ox 
=±= 
=b 
=i= 
d= 
0 
— 
db 
— 
+ 
+ 
— 
— 
Sheep 
=fc 
+ 
+ 
+ 
=b 
— 
0 
=b 
— 
— 
— 
Hog 
d= 
d= 
— 
— 
0 
— 
+ 
+ 
— 
+ 
Horse 
d= 
+ 
+ 
+ 
+ 
db 
+ 
0 
— 
— 
Goose 
+ 
— 
— 
_ 
0 
_ 
_ 
_ 
Duck 
+ 
_ 
_ 
_ 
— 
— 
0 
— 
— 
Pigeon 
+ 
+ 
db 
— 
— 
— 
— 
— 
0 
— 
Hen 
+ 
+ 
— 
— 
— 
— 
— 
0 
+ 
= Well-marked hemolysis. 
(+) = Questionable or feeble hemolysis. 
=fc 
= Doubtful. 
- 
= No hemolysis. 
Kolmer and Casselman2 have titrated the natural thermostabile hemol- 
ysins for the corpuscles of various vertebrates in a large number of human 
serums. Most interest centers about the occurrence of natural antisheep 
hemolysin because of the wide-spread use of an antisheep hemolytic system 
in complement-fixation reactions, as, for example, the Wassermann syphilis 
reaction. In over 80 per cent, of human serums there is present sufficient 
natural amboceptor for sheep’s cells to give well-marked or complete hemol- 
ysis. Although this factor must be considered in using an antisheep hemo- 
lytic system in complement-fixation reactions, yet with a proper under- 
standing of principles and the employment of a satisfactory technic the 
danger of error is reduced to a minimum. In the following table the per- 
centages of serums showing 100, 75, 25, and 0 per cent, hemolysis, with the 
maximum dose of serum (0.2 c.c.) used in the method of titration, are shown: 
1 Sachs, H.: Handbuch der pathogenen Mikroorganismen, Kolle and Wassermann, 2. 
Auflage, 2, p. 799. 
2 Jour. Infect. Dis., 1915, 16, 441. 396 
HEMOLYSINS 
Blood. 
Number of 
Serums 
Tested. 
Per Cent. 
Showing 
100 Per Cent. 
Hemolysis. 
Per Cent. 
Showing 
75 Per Cent. 
Hemolysis. 
Per Cent. 
Showing 
25 Per Cent. 
Hemolysis. 
Per Cent. 
Showing No 
Hemolysis. 
Sheep 
125 
64 
20 
9 
7.5 
Dog 
25 
16 
36 
30 
18 
Ox 
85 
6 
20 
24 
50 
Goat 
25 
0 
8 
16 
76 
Hog 
40 
0 
1 
3 
96 
Rat 
25 
0 
0 
12 
88 
Chicken 
25 
0 
0 
8 
92 
Horse 
25 
0 
0 
4 
96 
Rabbit 
25 
0 
0 
4 
96 
Guinea-pig 
50 
0 
0 
2 
98 
Summary of Natural Thermostabile Hemolysins in Human Serum 
As discussed in section on the Properties of Hemolysins in this chapter, 
hemolysins are known to be thermolabile and thermostabile. Some of the 
natural hemolysins, and notably that for guinea-pig corpuscles in human 
sera, are almost entirely of the thermolabile variety, as shown in the follow- 
ing table by Kolmer, Trist, and Flick,1 based upon the results observed in 
titrations employing unheated and heated human sera in amounts of 0.1 c.c.: 
For erythrocytes of 
Percentage sera thermo- 
labile hemolysin. 
Percentage sera thermo- 
stabile hemolysin. 
Human 
4 to 44 
0 to 4 
Sheep 
80 to 95 
Ox 
72 to 88 
5 to 30 
Guinea-pig 
90 to 100 
20 to 25 
Rabbit 
94 to 100 
0 to 2 
Hog 
92 to 100 
0 to 15 
Rat 
4 to 32 
0 to 2 
The variable figures are due to the fact that the natural hemolysins in 
human sera for the corpuscles of the lower animals are present in groups anal- 
ogous to the groups of the isohemolysins and isohemagglutinins in human 
sera. For this reason percentages expressing the presence of natural hemol- 
ysins based upon studies made with the corpuscles of a single animal of 
each species are only approximately correct. 
Kolmer and Williams2 have likewise studied the thermostabile hemol- 
ysins found in normal rabbit-serum as preliminary to some work concerning 
the site of formation of immune hemolysins. The results obtained with the 
maximum dose of serum (0.2 c.c.) are given in the following table: 
Summary of Natural Thermostabile Hemolysins in Normal Rabbit-serum 
Blood. 
Number or 
Serums 
Tested. 
Per Cent. 
Showing 
100 Per Cent. 
Hemolysis. 
Per Cent. 
Showing 
75 Per Cent. 
Hemolysis. 
Per Cent. 
Showing 
25 Per Cent. 
Hemolysis. 
Per Cent. 
Showing No 
Hemolysis. 
Goat 
25 
4 
52 
88 
12 
Sheep 
50 
18 
56 
76 
24 
Dog 
25 
32 
40 
72 
28 
Human 
50 
0 
4 
20 
80 
Hog 
25 
0 
4 
20 
80 
Ox 
25 ' 
0 
4 
20 
80 
Chicken 
25 
0 
4 
8 
92* 
Guinea-pig 
25 
0 
0 
4 
96 
Rat, white 
25 
0 
0 
0 
100 
1 Amer. Jour. Syph., 1920, 4, 111. 
2 Jour. Infect. Dis., 1913, xiii, No. 1, 96. NORMAL OR NATURAL HEMOLYSINS 
397 
Of interest in this connection are the natural hemolysin and hemagglu- 
tinins in horse-sera for human corpuscles in relation to the therapeutic use 
of normal and immune horse-sera. Kolmer and Matsumoto1 found agglu- 
tinins in practically all horse-sera for human corpuscles, but hemolysins 
are practically absent, being found only occasionally in fresh unheated serum 
and exceptionally in older and preserved sera. 
Isohemolysins.—Hemolysins in sera for corpuscles of the same species 
are called isohemolysins; examples are hemolysins in human sera for human 
corpuscles and in guinea-pig-sera for guinea-pig corpuscles. These hemol- 
ysins, however, do not act upon the corpuscles of the individual from whom 
the serum is taken. 
The isohemolysins are mostly thermolabile, as stated above; they are 
likewise highly susceptible to the effects of drying and cannot be well pre- 
served in this state, as may the natural hemagglutinins. 
Isohemolysins apparently occur in groups similar to the isohemagglutinins. 
This has been found true not only for the isohemolysins in human sera, 
but likewise for the isohemolysins in horse-, guinea-pig-, and the sera of other 
of the lower animals. The groups of isohemolysins for human corpuscles 
have not been studied as thoroughly as the isohemagglutinins, and it is not 
possible to state at present whether or not subgroups occur. For blood 
transfusion it is usual to conduct only agglutination tests for determining 
blood compatability, and for this reason the isohemolysins have not been 
studied as thoroughly as the agglutinins. I shall discuss their relation to 
blood transfusion in more detail in the section devoted to Transfusion. 
As a general rule isohemolysins do not occur in human sera free of iso- 
hemagglutinins, so that bloods are usually tested for the latter only in prepara- 
tion for transfusion. However, I am quite sure that isohemolysins may 
occur occasionally in sera in which hemagglutinins are not demonstrable. 
Isohemolysins were apparently first discovered by Ehrlich and Morgen- 
roth2 by injecting a goat with an enormous amount of blood from other 
goats. The hemolysin was active for the corpuscles of other goats, but 
not for those of the immunized animal. Shortly afterward Ascoli3 dis- 
covered isohemolysins in human serum, and since then a large literature 
has accumulated, especially in relation to blood transfusion and the develop- 
ment of these substances in cancer and other diseases. 
In 1892 Maragliano4 directed attention to the fact that the blood-serum 
of patients afflicted with various diseases exerted a hemolytic influence on 
the blood-corpuscles of healthy persons. Later Ascoli5 found the serums 
of cancer, pneumonia, and Addison’s disease to be actively hemolytic for 
normal corpuscles. Kelling6 was first, however, to apply this test to the 
diagnosis of cancer. These reports stimulated many investigations, with 
the result that the presence of isohemolysins was regarded by many as a 
pathologic phenomenon, Weil7 early described the presence of hemolysins 
in the sera of dogs with a type of tumor known as infectious lymphosarcoma, 
active against the corpuscles of other dogs. Weil also observed, howrever, 
the presence of these hemolysins in normal dog-sera, but thought they under- 
went an increase in sarcomatous dogs, and particularly when the tumors 
showed necrosis. These hemolysins were not active against the dog’s own 
corpuscles; indeed, it was believed that these cells acquired increased re- 
1 Jour. Immunology, 1920, 5, 75. 2 Berl. klin. Wchn., 1900, 453. 
3 Munch, med. Wchn., 1901, 1239. 
4 Deutsch. med. Wchnschr., 1892, xviii, 411. 
5 Miinch. med. Wchn., 1901, xlviii, 1239. 
6 Arch. f. klin. Chir., 1906, lxxx, 77. 7 Jour. Med. Research, 1907, 287. 398 
HEMOLYSINS 
sistance as the result of some sort of immunizing process by the tumor 
products. 
Soon after this report Crile1 published reports indicating the develop- 
ment of isohemolysins in the sera of human beings during the early stages 
of cancer and regarded their presence as possessing diagnostic value. Ales- 
sandri,2 Janeway,3 Blumgarten,4 Scheiter,5 and others reported favorably 
upon the diagnostic value of the reaction, but Fischel,6 Whittemore,7 Smithies,8 
Agazzi,9 Ottenberg,10 Gorliani and Lisser,11 and others, were unable to con- 
firm the diagnostic value of this isohemolysin reaction in cancer. Krida12 
has collected from literature 472 cases of cancer; of these, 67 per cent, gave 
positive reactions, while 2.6 per cent, of 509 observations on normal in- 
dividuals yielded positive reactions. 
These observations were made at a time when the importance of group 
isohemolysins in normal sera was not generally appreciated; however, 
it would appear that in some cases of cancer and particularly those asso- 
ciated with necroses and the accompanying cachexia, isohemolysins may be 
produced, but these changes do not occur early enough or with sufficient 
regularity to place a test for isohemolysins upon a diagnostic basis. 
Autohemolysins; Paroxysmal Hemoglobinuria.—Hemolysins in the 
serum for the individual’s own corpuscles are called autohemolysins. They 
are rarely observed and always in disease. Human autohemolysin was first 
discovered by Donath and Lansteiner13 and Eason14 in a case of paroxysmal 
hemoglobinuria, and these observations have since been generally con- 
firmed. This hemotoxin or isohemolysin is capable of sensitizing the red 
blood-corpuscles of the patient or those of a normal person at a low tempera- 
ture, hemolysis occurring in the presence of fresh serum, presumably with 
complement, and best at body temperature. As previously stated in the 
chapter on Agglutinins, autohemagglutinins have been demonstrated in 
the sera of cases of paroxysmal hemoglobinuria by Clough and Richter.18 
According to Cook, about 90 per cent, of hemoglobinurics show a posi- 
tive Wassermann reaction. Landsteiner found that about 10 per cent, of 
paretics showed the presence of isohemolysins, although these were not al- 
ways autohemolysins. Malaria and trypanasomiasis have also been re- 
garded as causes of paroxysmal hemoglobinuria, the most evidence, however, 
according to Matsuo16 and others, indicating that the disease usually has a 
syphilitic origin. Berghausen17 has suggested that the cold, trauma, and 
passive congestion leading to an attack are associated with the production 
of an excessive acidity of the tissues and that the organic acids thus formed 
may play some part in the hemolytic processes. 
It is possible that the hemotoxin is similar to the hemolysin of cobra 
1 Jour. Amer. Med. Assoc., 1908, 40, 1883; ibid., 41, 158. 
2 Jahresbericht f. Immunitat., 1909, 5. 
3 Jour. Amer. Med. Assoc., 1909, 43, 408. 
4 Med. Record, 1909, lxxv, 61. 
5 Jour. Amer. Med. Assoc., 1909, 43, 1481. 
6 Berl. klin. Wchn., 1908, 882. 
7 Boston Med. and Surg. Jour., 1909, clx, 77. 
8 Med. Record, 1909, Lxxv, 90. 
8 Berl. klin. Wchn., 1910, No. 31. 
10 Archiv. Int. Med., 1909, 3, 467. 
11 Amer. Jour. Med. Sci., 1912, 144, 103. 
12 Albany Med. Ann., 1911, 31, 259. 
13 Munch, med. Wchn., 1904, 1590; Ztschr. f. klin. Med., 1906, 58, 173. 
14 Edinb. Med. Jour., 1904, 19, 43. 
16 Bull. Johns Hopkins Hosp., 1918, 29, 86. 
16 Arch. f. klin. Med., 1912, 107, 335. 
47 Arch. Int. Med., 1912, 9, 137. NORMAL OR NATURAL HEMOLYSINS 
399 
venom, being in the nature of an amboceptor complemented by the fatty 
acids or lecithin of red corpuscles (endocomplement) or by a serum com- 
plement. 
As previously stated, the hemolysin unites with the corpuscles at a low 
temperature, but hemolysis occurs only at higher temperatures. As shown by 
Meyer and Emmerich, and confirmed by Cook,1 the complement also unites 
with the corpuscle-hemolysin complex at low temperatures, but, as just 
stated, higher temperatures are required for hemolysis. Persons with par- 
oxysmal hemoglobinuria may have an attack merely by holding their hands 
in cold water. The autohemolysin will also sensitize the corpuscles of normal 
persons, as shown by Lorant2 and others. Immediately after a paroxysm 
the hemolysin may be absent from the serum, as reported by Roberts.3 
Moss4 has shown that the corpuscles of the patient may possess an in- 
creased resistance to hypotonic salt solutions, although just the reverse 
would be expected, and resistance to carbon dioxid is slightly increased,fas 
shown by Berghausen. 
Methods for conducting the autohemolysin test for paroxysmal hemo- 
globinuria are described in this chapter under Practical Applications. As 
shown by Meyer and Emmerick5 the test may fail owing to a lack of sufficient 
complement or owing to an anticomplementary action of the serum, as shown 
by Kumagai and Inoue.6 
The Removal of Hemolysins from Serum by Absorption.—Natural or 
immune hemolysins may be removed from a serum by absorption with the 
homologous corpuscles. If the serum is active, that is, contains complement, 
it is first necessary to chill the serum to about 0° C. and add chilled cor- 
puscles, the mixture being kept at 0° to 2° C. for several hours. At this 
temperature the hemolysin will unite with the corpuscles (sensitization 
occurs), but hemolysis does not occur or is very slight because complement 
is inactive at low temperatures. The mixture is now rapidly centrifuged; 
the supernatant fluid is found free of hemolysin under proper technical 
conditions when tested by adding complement and corpuscles, whereas the 
corpuscles are sensitized as found by adding complement and incubating 
at 38° C., when hemolysis occurs. 
By first heating the serum at 55° C. for fifteen minutes the native com- 
plement is inactivated. Old sera need not be heated, as complement is 
destroyed in a few days unless extra precautions are taken. With inacti- 
vated sera the hemolysin may be removed by adding the corpuscles and 
keeping the mixture at room temperature for one-half hour or in a water- 
bath at 38° C. for fifteen minutes. At these higher temperatures sensiti- 
zation is more rapid. The mixture is now centrifuged, when the superfluid 
will be found free of hemolysin under proper technical conditions. 
This process, however, frequently renders a serum anticomplementary, known 
as the “Sach’s-Friedberger phenomenon.”7 If absorbed sera are to be used 
in the Wassermann or other complement-fixation tests they should be 
reheated to 55° C. for fifteen minutes in order to remove the antilysins. 
Methods for the removal of hemolysins from active and inactivated sera 
are given under Practical Applications. 
Absorption of sera with corpuscles does not remove syphilis “reagin”- 
1 Ztschr. f. Immunitatsf., 1914, 21, 623. 
2 Deut. Archiv. klin. Med., 1918, 126, 148. 
3 Brit. Med. Jour., 1915, 2, 398. 
4 Bull. Johns Hopkins Hosp., 1911, 22, No. 245. 
6 Deut. Arch. f. klin. Med., 1909, 96, 287. 
6 Duetsch. med. Wchn., 1912, No. 8. 
7 See Bauer, Deutsch. med. Wchn., 1908, 34, 698. 400 
HEMOLYSINS 
or antibodies other than the hemolysin corresponding to the corpuscles 
employed. Absorption with barium sulphate as originally advocated by 
Wechselmann1 has been found by Noguchi and Bronfenbrenner2 to remove 
not only natural hemolysin, but syphilis antibody as well; these results 
were confirmed by Kytoku3 in my laboratory. 
Natural Hemolysins in Relation to Complement-fixation Reactions.— 
Many investigators have expressed the opinion that the presence of natural 
antisheep hemolysin in human sera reduces the sensitiveness of the Wasser- 
mann and other complement-fixation tests conducted with an antisheep 
hemolytic system. Theoretically at least the presence of large amounts of 
natural hemolysin in some sera may be expected to disturb the important 
quantitative relations among corpuscles, complement, and hemolysin and 
thereby reduce the delicacy of complement-fixation tests. As shown by Wil- 
liams,4 the same may occur with an antihuman hemolytic system due to iso- 
hemolysins, a fact commonly overlooked by those who employ this system 
for complement-fixation tests. 
Practically, however, the subject is not of great importance, as the in- 
fluence of natural hemolysins can be neutralized by simple technical pro- 
cedures which will be discussed in more detail in the chapters devoted to 
the Complement-fixation Test. In my experience5 antisheep hemolysin is 
apt to be disturbing and reduce the sensitiveness of the Wassermann reac- 
tion only when present in large amounts in those sera containing small 
amounts of syphilis antibody. 
Production of Hemolysins.—Because of their use in the serum diag- 
nosis of syphilis and other infections hemolysins possess great practical 
value. They are best produced by injecting rabbits with washed human 
or sheep erythrocytes, or with those of some animal of another species. 
Rabbits differ considerably in their power to form hemolysins, and for 
some unknown reason hemolysins are more readily produced with some 
erythrocytes than with others. It is not possible, however, to immunize 
every kind of animal against every type of erythrocyte. As a general rule, 
an animal produces a better hemolysin the more remote its relationship is 
to the animal from which the erythrocytes for making the injection are 
taken. 
Hemolysins may be prepared as the result either of intraperitoneal or 
of intravenous injections of erythrocytes washed at least three or four 
times to remove all traces of serum. Unless an antihuman hemolysin is 
required within a short space of time, better results are obtained, as a rule, 
by using the slower, intraperitoneal method for its production. 
In immunizing rabbits it must be remembered that the quantity of 
hemolysin produced bears no direct relation to the size of the doses of cor- 
puscles given. Thus, a highly potent hemolytic serum may be prepared 
by three intravenous injections of from 3 to 5 c.c. of a 10 per cent, suspen- 
sion of washed sheep cells. 
It must be emphasized that as far as possible an aseptic technic should 
be employed, and that corpuscles used for making intravenous injections 
should be filtered to remove small particles of fibrin, and preferably washed 
four times with sterile salt solution. The method of preparing corpuscles 
for injection has been given in Chapter II. After the third or fourth 
injection the animal should be bled from the ear and the serum tested, 
as animals not infrequently succumb after the fourth injection, and many 
1 Ztschr. f. Immunitatsf., 1909, 3, 528. 
2 Jour. Exper. Med., 1911, 13, 217. 
3 Jour. Immunology, 1919, 4, 239. 
4 Jour. Exper. Med., 1920, 32, 159. 
5 Amer. Jour. Syph., 1920, 4, 135. NORMAL OR NATURAL HEMOLYSINS 
401 
possess serums of high potency after receiving three injections. By this 
method success is better assured. 
Production of Antihuman Hemolysin.—Noguchi and Bronfenbrenner1 
have found the rabbit best adapted for the preparation of antihuman 
hemolysin; more recently Sanford2 has recommended the use of small dogs, 
claiming that these animals produce hemolysin equally well, but with the 
decided advantage of less coincident production of hemagglutinins. 
It is difficult to prepare powerful antihuman and antichicken hemolytic 
sera; many rabbits succumb during the process of immunization. A variety 
of methods have been advocated, as follows: 
Fatalities have been commonly ascribed to anaphylaxis, toxicity of 
human corpuscles for rabbits, agglutination, and hemolysis in vivo. The 
results of a study of these factors in my laboratory by Matsumoto3 indi- 
cated that primary and direct toxicity of human corpuscles for the rabbit 
was of most importance followed by agglutination and hemolysis in vivo 
and anaphylaxis. The toxic effects are most apparent when attempts are 
made to immunize rabbits with human, chicken, and guinea-pig corpuscles; 
sheep and beef corpuscles are much less toxic. 
1. Noguchi’s Intravenous Method.—Inject 4, 3, 4, and 3 c.c., and pos- 
sibly another 4 c.c., with four- or five-day intervals; bleed nine or ten days 
after last injection. 
2. Noguchi’s Intraperitoneal Method.—Inject at four- or five-day inter- 
vals, bleeding nine or ten days after last injection. Doses 5, 8, 12, 15, and 
20 c.c. of washed human corpuscles. 
3. Thompson’s Method.—Inject intravenously 0.1 c.c. washed cells 
every day for three or four weeks. This method was adopted after the 
work of Coca, showing that powerful hemolytic sera for sheep cells may be 
produced by this procedure. 
4. Craig’s Method.—Inject intravenously 1 c.c. of washed erythrocytes 
every other day until five or six injections have been given; two or three 
more injections may be required. 
5. Vedder’s Method.—Inject intravenously 0.5, 1, 2, and 3 c.c., washed 
and packed human cells at five to seven days, interval; the last injection of 
3 c.c. may have to be repeated several times. 
6. Bronfenbrenner’s and Schlesinger’s Method.—Rabbits are injected 
intravenously with washed red cells at four-day intervals; in order to avoid 
anaphylactic reactions each intravenous injection, beginning with the third 
one, is preceded by a desensitizing intraperitoneal injection of the same 
cells given one-half hour in advance. 
In a comparative study of the above methods Miss Rule and the author4 
found the intravenous methods superior to the intraperitoneal, inasmuch 
as hemolysin production occurred more quickly, required fewer injections, 
and smaller doses of cells. Best results were secured in the production of 
antihuman, dntichicken, antiguinea-pig, and antiox hemolysins by the daily 
intravenous injection of 0.1 c.c. washed packed cells diluted with 0.9 c.c. sterile 
saline solution, after the method of Thompson. 
Army Method.—I have also secured good antihuman hemolysins by the 
following method employed in the Army Medical School: Rabbits are given 
intravenous injections of 3 c.c. of packed, washed corpuscles on three suc- 
cessive days and are then allowed to rest for twenty-one days, when they 
are given daily intravenous injections of 0.5 c.c. of washed cells. If the titer 
is too low, a series of the 0.5 c.c. injections is repeated. A useful modification 
1 Jour. Exper. Med., 1911, 13, 78. 
2 Amer. Jour. Syph., 1920, 4, 697. 
3 Jour. Immunology, 1920, 5, 507. 
4 Amer. Jour. Syph., 1920, 4, 484. 402 
HEMOLYSINS 
of this method consists of three intravenous injections of 1 c.c. of a 50 per 
cent, suspension of washed cells at daily intervals, followed by three injec- 
tions of 2 c.c. each daily and three more injections of 3 c.c. each daily. The 
animal is now rested for three weeks, when three injections of 3 c.c. 
each are given at daily intervals and the serum tested about one week after 
the last injection. 
Sanford’s Method.—Small dogs of the fox terrier size are recommended 
and given a series of intraperitoneal injections at weekly intervals of 50 
per cent, suspensions of washed human corpuscles in doses of 30, 40, 50, 
60, and 60 c.c.; sometimes a good hemolysin is secured with three in- 
jections. 
Production of Antisheep Hemolysin.—The production of antisheep he- 
molysin in rabbits is so simple and easy as compared with the production of 
antihuman hemolysin that the subject has not attracted much attention; 
practically any method devised along generally accepted lines and with proper 
attention to the usual technical details will yield a satisfactory serum. Sheep 
cells are but slightly toxic for rabbits, and the antibody-producing tissues 
are promptly and powerfully influenced so that the majority of animals 
survive and produce highly potent immune sera. Agglutinin production 
for sheep cells is tardy and never reaches a high degree as compared with 
hemolysin production; no serologist of experience should consent to use any 
but highly potent sera in view of the ease of production, and no laboratory 
should be without the simple means of providing an adequate supply of 
this hemolysin. 
Of the large number of methods advocated for the production of anti- 
sheep hemolysin, mention may be made of the following: 
1. In the Hygienic Laboratory, Neill injects rabbits intravenously with 
1 c.c. of fresh sheep cells every three days for four injections and tests the 
serum five days after the last injection; in a second method he injects intra- 
venously 1,1, and 2 c.c. washed cells on three successive days, with a second 
series of injections after an interval of five days, after the method of Gay 
and Fitzgerald. 
2. Schweitzer and Stevens in the New York Department of Health 
Laboratories inject 0.25 c.c. of a 50 per cent, suspensions of washed cells 
every three days, increasing the dose by 0.25 c.c. with each injection until 
four to eight injections have been given. 
3. Coca has advocated the daily administration of small doses (0.1 c.c.) 
of cells over a period of several weeks. 
4. The author has used for several years an intravenous method con- 
sisting in the injection of 5 c.c. of a 10 per cent, suspension of washed cells 
every three days until four to six injections have been given. This method 
is very simple and has uniformly yielded powerful antisheep hemolysin. 
Intraperitoneal injections may also be used, consisting in the injection 
of 5 c.c. of 50 per cent, suspensions of washed corpuscles every five days 
until four to six injections have been given. 
The Collection of Serum.—Most authors advise bleeding ananimal four 
to nine days after the last injection of cells when preliminary titrations 
conducted with serum secured by collecting blood from an ear vein have 
shown that the titer is satisfactory. According to our experiments the best 
time for bleeding appears to be about seven days after the last injection by the 
intravenous route, inasmuch as hemolysin production is likely to occur for 
about this time after an injection of the antigen. 
A preliminary titration of serum secured from a small amount of blood 
from an ear vein may not give an exact idea of the content in hemolysin, NORMAL OR NATURAL HEMOLYSINS 
inasmuch as the serum of the whole blood secured by bleeding the animal 
to death may show less or slightly more hemolysin. 
Preservation of Serum.—Methods have been described in Chapter IV. 
I have found that the addition of an equal part of chemically pure glycerin 
to unheated serum constitutes an excellent method. Antihuman serum is 
also well preserved dried in filter-paper, as described by Noguchi. 
As previously stated, the portion of the erythrocyte that is responsible 
for the production of hemolysins is a moot question. Bordet and von Dun- 
gern maintain that the stroma is the exciting agent; Nolf and others believe 
that the stromata produce hemagglutinins, and that the hemoglobin is 
chiefly concerned in the production of the hemolysin. 
General Properties of Hemolysins; Thermostabile and Thermolabile 
Hemolysins.—Immune hemolysins are highly resistant antibodies, and are 
easily preserved. Sterile immune serum may be inactivated by heating 
in a water-bath for half an hour at 56° C., and may be preserved for many 
months if small amounts are placed in ampules and kept in a cold place. 
If an equal quantity of neutral glycerin is added to the clear inactivated 
serum it will aid greatly in its preservation. The immune hemolysins resist 
drying to a well-marked degree, and filter-paper saturated with the serum 
and dried, after the method of Noguchi, preserves the hemolytic activity 
in a remarkable degree. Heating an immune serum at 56° C., for half an 
hour, as in the process of inactivating complement, does not materially 
injure the hemolytic activity of a potent serum. A temperature of 70° C. 
or above may cause deterioration and finally destroy the hemolysins. 
A portion of immune hemolysin, however, is thermolabile and easily 
destroyed by heating serum to 55° C. for half an hour. Many of the natural 
hemolysins and especially those in human serum for human corpuscles and 
the corpuscles of the guinea-pig, rabbit, hog, and ox are of the thermolabile 
variety. 
Thiele and Embleton1 have sought to prove that these thermolabile 
hemolysins are only differentiated complements, that is, immune hemol- 
ysin as it is produced is first thermolabile and gradually acquires thermo- 
stability as production is continued. Sherman,2 on the other hand, regards 
all hemolysins as thermostabile and that the reduction in hemolytic activity 
of a serum as a result of heating is due to “masking” of the hemolysin rather 
than an actual destruction or inactivation, that is, the hemolysins are in- 
capable of sensitizing corpuscles in the presence of complement. In my 
own experiments, however, I found that the thermolabile hemolysins may 
be differentiated from the complements, being more resistant to heat; in 
this respect my results do not agree with those of Thiele and Embleton. 
On the other hand, these experiments have indicated that heat may actually 
destroy some of these hemolysins, that for sheep cells being more resistant 
than that for guinea-pig corpuscles.3 
While immune hemolysins are largely resistant to the deteriorating 
influence of desiccation, the natural hemolysins are more susceptible. I 
have found that the isohemolysins in human sera deteriorate rapidly upon 
drying, and under ordinary conditions are more susceptible than the iso- 
hemagglutinins.4 
As stated in the preceding chapter, the hemolysins have been generally 
found in the englobulin and pseudoglobulin fractions of the serum, and not 
in the albumin fraction, by Fuhrmann5 and others. 
1 Ztschr. f. Immunitatsf., 1914, 20, 1. 
2 Jour. Infect. Dis., 1918, 22, 534. 
3 Jour. Immunology, 1919, 4, 403. 
4 Jour. Immunology, 1919, 4, 393. 
5 Hofmeister’s Beitr., 1903, 3, 417. 404 
HEMOLYSINS 
As was previously mentioned, hemolytic amboceptors possess a great 
affinity for the receptors of their homologous corpuscles, and will readily 
unite with them at a low temperature. At incubator temperature the union 
is quite rapid, so that corpuscles may be “sensitized” within half an hour. 
Source of Hemolysins.—As has been stated elsewhere, Metchnikoff re- 
gards the leukocytes as the source of fixateur or amboceptor formation. 
Recently Carrel and Ebeling1 have furnished direct experimental evidence of 
leukocytes as a possible source of hemolysin production. These investigators 
have found that cultures of leukocytes in hen plasma elaborate a hemolytic 
substance for rabbit and sheep erythrocytes in about 50 per cent, of cultures. 
Bullock found that the amount of hemolytic amboceptor in a serum runs 
parallel with the number of mononuclear leukocytes, and he regards this 
as an indication of the activity of the lymphoid tissue in general, which 
he considers as the main source of amboceptor formation. While it is prob- 
able that endothelial cells and mononuclear leukocytes are especially con- 
cerned in the process, our own investigations in this field would indicate 
that the process is more general, being participated in by cells of other 
tissues which possess suitable combining affinities for the alien corpuscles. 
Antihemolysins.—A further step in the study of hemolysins, but one 
more of theoretic than of practical interest, was the discovery of antihemol- 
ysins. Besredka2 first discovered these antihemolysins in normal sera. 
By injecting guinea-pigs with normal rabbit-serum containing amboceptors 
for ox blood Bordet3 secured a serum that inhibited the action of anti-ox 
immune serum. Ehrlich and Sacks,4 by injecting a goat with normal rabbit- 
serum, likewise secured a serum that acted as an anti-amboceptor against 
immune hemolytic amboceptors for ox blood. Ehrlich argued that the 
anti-amboceptor acted against the complementophil group of the ambo- 
ceptor, which prevented union with a complement from taking place. This 
view was advanced in support of his theory concerning the two-armed 
character of the amboceptor and that an anti-amboceptor may be produced 
against either the cytophil or the complementophil group, or both. In these 
particular serums, however, the investigators may have been working with 
an anticomplement instead of an anti-amboceptor. 
A specific hemolysin—one, for example, specific for dog blood, derived 
from treating a rabbit with dog cells—is highly toxic for dogs, being capable 
of producing hemolysis in vitro and a clinical condition known as hemolytic 
jaundice. It is possible, however, gradually to immunize a dog against 
this amboceptor for his own cells by starting with very small doses and 
gradually increasing these until it is found that the animal tolerates amounts 
that would be fatal to non-immunized animals. If a portion of this serum 
is now added to the specific hemolytic serum, it will be found that the power 
of the latter is inhibited. Although this action may likewise be due to 
anticomplement, it is probable that an anti-amboceptor against the cytophil 
group of the amboceptor is also formed, which prevents the amboceptor 
from uniting with the red blood-cells, although conclusive experimental 
evidence of this has not been adduced. 
The Protective or Antihemolytic Activity of Sera; Relation to Anemia.— 
While sera may be hemolytic for the corpuscles of various animals, depend- 
ing upon the presence of natural hemolysins and complements, it is now 
well known that normal serum may be antihemolytic for various hemo- 
1 Jour. Exper. Med., 1922, 36. 645. 
2 Ann. d. l’lnst. Pasteur, 1901, 15, 785. 
3 Ann. d. l’lnst. Pasteur, 1899, 13, 273; 1900, 14, 257. 
4 Berl. klin. Wchn., 1900, 681; 1901, 569, NORMAL OR NATURAL HEMOLYSINS 
405 
toxins. Muller1 showed that heated serum may prevent hemolysis by 
serum hemolysins. Neisser and Friedemann,2 Bergmann and Keuthe,3 and 
others have noted this property in human sera, and Marshall and Morgen- 
roth4 have attributed the effects to a destructive effect upon complement. 
For this reason the process has been designated as the anticomplementary 
activity of sera and is of great practical importance in relation to the reac- 
tion of complement fixation; the phenomenon will be discussed in greater 
detail in the chapters devoted to that subject. 
In addition to this anticomplementary action, normal sera may protect 
corpuscles to a marked degree against the hemolytic activity of saponin, 
bacterial hemotoxins, sodium oleate, lactic acid, and other substances. 
Ransom5 has expressed the opinion that this neutralizing effect is due to 
the presence of cholesterol in serum and corpuscles in so far at least as 
saponin hemolysis is concerned. Bruere,6 however, was unable to confirm 
this opinion. Goldberger7 found that the sera of cancerous persons ac- 
quired an increased antihemolytic activity over oleic and lactic acids, but 
Sweek and Fleischer8 did not observe that these sera were any more anti- 
hemolytic than the sera of normal individuals. Clark and Evans9 found 
the sera of normal persons and of patients with diseases without anemia 
remarkably constant in antihemolytic activity for sodium oleate on guinea- 
pig corpuscles; in anemias, however, and especially in pernicious anemia 
and conditions associated with lesions of the spleen, a marked reduction 
in protective power was found. The authors have suggested that this 
reduction may constitute a factor in the mechanism of the so-called hemo- 
lytic anemias, Addisonian anemia, and hemolytic jaundice. Subsequent 
studies on the nature of the antihemolytic substances in serum for sodium 
oleate and saponin have shown that this property does not depend alone 
upon the serum cholesterol, lecithin, or a combination of these, and the dimin- 
ished protective power associated with the above-mentioned anemias can- 
not be explained simply on the basis of the diminution of the total cholesterol. 
Hemolysins in Relation to Complement Fixation.—There is probably no 
other group of antibodies that possesses greater diagnostic value than do 
the hemolysins, a fact that was demonstrated in the practical application 
of the Bordet-Gengou phenomenon of complement fixation in the diagnosis 
of syphilis and other infections. By adding sensitized corpuscles—i. e., 
corpuscles with their homologous amboceptors—to a fluid, the presence 
or absence of complement may be determined. If complement is present, 
hemolysis will occur and be complete or partial, depending upon the amount 
of complement available; if hemolysis does not occur, it may be concluded 
that free complement is absent. This is the basis of the complement-fixation 
diagnosis of syphilis, gonorrhea, glanders, differentiation of proteins, etc. 
When a proper amount of complement is added to a mixture of antigen 
and its immune serum (containing amboceptors), it is bound to these ambo- 
ceptors, so that when corpuscles and hemolytic amboceptor are subse- 
quently added, hemolysis does not occur, since there is no available com- 
plement, it having been “fixed” by the first amboceptors. If, however, 
1 Centralbl. f. Bakteriol., 1901, 29, 860. 
2 Berl. klin. Wchn., 1902, 677. 
3 Ztschr. f. Exp. Path., 1906, 3, 255. 
4 Deutsch. med. Wchn., 1901, March 28th. 
6 Ibid. 
6 Jour. Med. Research, 1902, 8, 362. 
7 Folia Serologica, 1911, 7, 941. 
8 Jour. Med. Research, 1913, 27, 383. 
9 Bull. Johns Hopkins Hosp., 1920, 31, 354; 1921, 32, 113, 328. 406 
HEMOLYSINS 
amboceptors for the antigen in the first instance are not present, as where 
a normal serum is used, the complement remains free and acts with the 
hemolytic amboceptor to produce hemolysis of the test corpuscles. In 
this manner the hemolysins and their corresponding corpuscles are em- 
ployed as indicators or tests for the presence of free complement, so that 
if an antigen is known, the antibody may be determined; or vice versa, 
by using a known antibody, the antigen may be determined, the criterion 
in each instance being whether complement is or is not bound or “fixed,” 
a fact that is determined by the subsequent addition of a hemolysin and 
its homologous corpuscles. 
The practical applications and technic of these reactions are given in 
subsequent chapters. 
Quantitative Relationship Between Hemolysin, Corpuscles, and Com- 
plement.—From a practical as well as a theoretic standpoint an important 
property of hemolysin and of complement is the quantitative relation- 
ship that each bears to the other. This is especially important in hemolytic 
reactions, where an excess of either may compensate for a decrease of the 
other and yield fallacious results. If a certain amount of guinea-pig’s com- 
plement is necessary to lyse 1 c.c. of a 2.5 per cent, suspension of sheep’s 
cells, along with hemolytic amboceptor, then double this amount of com- 
plement will be required to lyse 2 c.c. of the same blood, and so on. If a 
constant quantity of corpuscles and hemolysin are added to a series of 
test-tubes, and increasing amounts of complement, after incubating the 
mixtures for an hour the smallest amount of complement that produces 
complete hemolysis is called a unit, and in this manner the strength or 
activity of a serum complement is measured or titrated. In a similar manner 
the hemolytic activity of a serum or its measure of hemolysin may be de- 
termined by placing in a series of tubes, as previously directed, a definite 
and equal amount of corpuscle suspension, and to each tube is then added 
an amount, also definite and equal, of a normal serum as complement; 
there are next added decreasing and graduated amounts of the immune 
serum whose native complement has been destroyed by inactivation. After 
incubating the mixtures for an hour, the smallest amount of inactivated 
immune serum that will just produce complete hemolysis is known as the 
amboceptor or hemolysin unit of the serum. In other words, there are three 
substances concerned in serum hemolysis: the hemolysin, the corpuscles, and 
the complement. By taking two of these as constants, e. g., the corpuscles 
and the complement, the unit of hemolysin may be determined; or by taking 
the corpuscles and hemolysin as constants the unit of complement may be 
determined. Since the corpuscles and hemolysin are most stable, these may 
be used as constants, and the unit of complement determined under these 
conditions as preliminary to complement-fixation reactions. 
It is important to bear in mind, in this connection, that the titer of an 
immune hemolytic serum will vary with the complement used. For ex- 
ample, an antisheep hemolysin is much more active when guinea-pig 
serum is used as complement than it is when tested with the same quantity 
of rabbit-serum as complement. 
After determining the unit of hemolysin or complement—that is, after 
adjusting the hemolytic system to exact proportions—the results that 
follow the varying quantities of complement and hemolysin require the 
most careful consideration. Less than one unit of amboceptor with one 
unit of complement cannot yield complete hemolysis; likewise if with one 
unit of hemolysin less than a unit of complement is combined hemolysis 
is incomplete; with one unit of hemolysin and one unit of complement PRACTICAL APPLICATIONS 
407 
and a double dose of corpuscles hemolysis will also be incomplete. With 
less than a unit of complement and an excess of hemolysin, however, 
hemolysis may be complete. The complement may be reduced to so small 
an amount that hemolysis is incomplete no matter how much hemolysin 
is used, but the important fact to be borne in mind is that a slight decrease 
in complement may be compensated for by the presence of many units 
of hemolysin, so that complete hemolysis results and a false reaction is 
secured. The converse of this is true to a less marked extent—i. e., an 
excess of complement may compensate for a decrease in hemolysin, but 
is less capable of doing so. 
These facts are of the utmost importance in making hemolytic experi- 
ments, as in complement-fixation reactions, where the entire test depends 
upon demonstrating whether or not a portion or the whole of the comple- 
ment used has been fixed. Unless, in a series of hemolytic reactions, the 
amount of hemolysin employed is the same throughout, the amount of 
complement acting in these cannot be determined by comparing the de- 
gree of hemolysis. This is true especially in cases where a small amount 
of complement is fixed, as in the Wassermann reaction, with a serum con- 
taining a small amount of syphilitic antibody, when the presence of an 
excess of hemolytic amboceptor may give complete hemolysis and over- 
shadow the fact that a small amount of complement has been actually 
fixed by syphilitic antibody and antigen. 
PRACTICAL APPLICATIONS 
Method of Titration of Immune Hemolysin.—Various methods have 
been employed by different workers in this field, but all are based upon 
the same principles as have been here outlined. 
A small amount of immune serum is inactivated by heating in a water- 
bath at 56° C. for half an hour. In testing the serum of a rabbit during 
the process of immunization 2 or 3 c.c. of blood are easily secured from 
the ear, and the serum is separated. After it has been inactivated the 
serum is diluted to 1 : 100 (1 c.c. of serum of 99 c.c. of salt solution, or 0.1 
c.c. serum + 9.9 c.c of salt solution). 
Fresh guinea-pig serum is secured for complement by bleeding a healthy 
pig under ether anesthesia into a Petri dish or centrifuge tube. This serum 
is diluted 1 : 20, making a 5 per cent, solution, by adding 1 c.c. of serum 
to 19 c.c. of salt solution. Each cubic centimeter of this dilution contains 
0.05 c.c. of undiluted serum, which experience has shown is a satisfactory 
amount to use. 
The corpuscle suspension is then prepared. The blood used depends 
upon the kind of hemolysin that is to be titrated. With sheep and ox 
blood, a 2.5 per cent, suspension of washed corpuscles may be employed. 
With antihuman hemolysin, the corpuscles are usually used in 1 per cent, 
suspension. (See Noguchi Modification of Wassermann Reaction.) After 
the corpuscles have been washed three times, 1 c.c. is placed in 39 c.c. of 
salt solution, or sufficient salt solution is added to 2.5 c.c. of the corpuscles 
to make the total volume equal 100 c.c. 
To a series of six sterile test-tubes increasing doses of the diluted im- 
mune serum are now added, together with 1 c.c. of complement dilution, 
1 c.c. of corpuscles suspension, and sufficient normal salt solution to make 
the total volume in each tube about 3 or 4 c.c. The following table shows 
the method of preliminary titration of a hemolytic serum: 408 
HEMOLYSINS 
Tube. 
Amount or Inactivated 
Immune Serum in C.c. 
(1 : 100). 
Dose of 
Comple- 
ment, C.c. 
(1 : 20). 
Dose of 
Corpuscles, 
C.c. (2.5 Per 
Cent.). 
Normal 
Salt 
Solution. 
Result of Hemolysis 
After One Hour in 
the Water-bath 
(37° C.). 
1 
0.1 (0.001 c.c. undiluted) 
1 
1 
q. s. 3 c.c. 
No hemolysis. 
2 
0.2 (0.002 c.c. undiluted) 
1 
1 
q. s. 3 c.c. 
Partial hemolysis. 
3 
0.4 (0.004 c.c. undi uted) 
1 
1 
q. s. 3 c.c. 
Complete hemolysis. 
4 
0.6 (0.006 c.c. undiluted) 
1 
1 
q. s. 3 c.c. 
Complete hemolysis. 
5 
0.8 (0.008 c.c. undiluted) 
1 
1 
q. s. 3 c.c. 
Complete hemolysis. 
6 
1.0 (0.01 c.c. undiluted) 
1 
1 
q. s. 3 c.c. 
Complete hemolysis. 
Preliminary Titration of a Hemolysin 
In this instance the unit of amboceptor is 1 : 250, which is too low for 
a satisfactory antisheep serum. The rabbit should, therefore, receive an- 
other dose or two of corpuscles, and the serum be titrated again in from 
four to seven days after the last injection has been given. In this titration 
it will be well to use a higher dilution of the inactivated immune serum, 
as 1 : 1000. This may be prepared by adding 1 c.c. of a dilution of 1 : 100 
with 9 c.c. of normal salt solution and mixing well. The titration is then 
proceeded with as follows: 
Tube. 
Amount of Inactivated 
Immune Serum in C.c. 
(1 : 1000). 
Dose 
of 
Com- 
ple- 
ment 
in C.c. 
(1 : 20). 
Dose 
of 
Cor- 
pus- 
cles, 
Cc. 
(2.5 
Per 
Cent.). 
Normal 
Salt 
Solution. 
Result of Hemolysis 
After One Hour in the 
Water-bath at 37° C. 
1.... 
0.05 (0.00005 c.c. undiluted) 
1 
1 
q. s. 3 c.c. 
No hemolysis. 
2.... 
0.1 (0.0001 c.c. undiluted) 
1 
1 
q. s. 3 c.c. 
No hemolysis. 
3.... 
0.15 (0.00015 c.c. undiluted) 
1 
1 
q. s. 3 c.c. 
Beginning hemolysis. 
4.... 
0.2 (0.0002 c.c. undiluted) 
1 
1 
q. s. 3 c.c. 
Partial hemolysis. 
5.... 
0.25 (0.00025 c.c. undiluted) 
1 
1 
q. s. 3 c.c. 
Just complete hemolysis. 
6.... 
0.3 (0.0003 c.c. undiluted) 
1 
1 
q. s. 3 c.c. 
Complete hemolysis. 
7.... 
0.35 (0.00035 c.c. undiluted) 
1 
1 
q. s. 3 c.c. 
Complete hemolysis. 
8.... 
0.4 (0.0004 c.c. undiluted) 
1 
1 
q. s. 3 c.c. 
Complete hemolysis. 
9.... 
0.45 (0.00045 c.c. undiluted) 
1 
1 
q. s. 3 c.c. 
Complete hemolysis. 
10.... 
0.5 (0.0005 c.c. undiluted) 
1 
1 
q. s. 3 c.c. 
Complete hemolysis. 
Titration or Hemolysin 
The following controls should be set up at the same time: 
1. 1 c.c. of corpuscles in 1 c.c. of amboceptor dilution. This tube should 
show no hemolysis, as the serum has been inactivated and is too 
highly diluted for complement activity, even though native com- 
plement were present. 
2. 1 c.c. of corpuscles in 1 c.c. of complement dilution. This tube may 
show a trace of hemolysis, due to the presence of a small amount 
of natural hemolysin for the corpuscles used. As a general rule, 
guinea-pig serum is free from natural antisheep hemolysin, or the 
amount is so small under these conditions that it is not necessary 
to remove it. 
3. 1 c.c. of corpuscles in 3 c.c. of salt solution. This tube should show 
no hemolysis, and serves to show that the diluent was isotonic. Fig. 127.—Titration of Hemolytic Amboceptor. 
The tube containing 0.3 c.c. hemolytic amboceptor is the smallest amount showing complete hemol- 
ysis, and this amount is the unit. PRACTICAL APPLICA TIONS 
409 
In the foregoing titration it is found that 0.25 c.c. of 1 : 1000 dilution 
of amboceptor is the unit, or the titer is 1 : 4000. The method requires 
accurate pipets and careful work, but yields uniform and satisfactory re- 
sults (Fig. 127). 
This method is given only as an example of the technic and principles 
involved. The method of titration varies according to the hemolytic system 
being employed for conducting complement-fixation tests, and will be re- 
ferred to again in the chapter devoted to this technic. 
Instead of using varying amounts of a stock solution of hemolysin as 
described, a series of dilutions of hemolysin may be prepared, as 1 :1000, 
1 : 2000, 1 : 3000, etc., and employed in a constant dose of 0.5 or 1.0 c.c. 
If the serum is satisfactory the rabbit may now be bled under anes- 
thesia. The serum is separated and inactivated and again titrated, as the 
final titration, for some unknown reason, is likely to be a little lower than 
in the primary tests. 
When suitably preserved, a hemolytic serum will maintain its activity 
for long periods of time; it should always, however, be titrated before com- 
plement-fixation tests are undertaken. 
Methods for Removing Hemolysins from a Serum.—In general, these 
aim to remove the natural hemolysins, such as natural antisheep hemol- 
ysin, from human serums preliminary to making the Wassermann test, 
or from a guinea-pig-serum that is to be used as complement. 
The method of removal consists simply in adding corpuscles to the 
serum, and allowing sufficient time for the corresponding hemolytic ambo- 
ceptor to become attached, and then removing both by centrifuging the 
mixture. If the serum is fresh, it should be cooled to 0° to 3° C. in order 
to inhibit complement activity, which would hemolyze a portion of the 
corpuscles. 
Method for Removal of Natural Hemolysins from Active Serum.—(1) Chill 
the serum in a mixture of ice and salt; (2) for each cubic centimeter add 1 
drop of sediment of washed corpuscles thoroughly chilled; (3) mix and 
keep at or near 0° to 2° C. for at least one hour; (4) centrifuge rapidly and 
preferably with the tube surrounded with ice. Remove the supernatant 
serum. 
In removing a natural hemolytic amboceptor from a guinea-pig-serum 
that is to be used as complement a measured amount of serum is first re- 
moved to a separate tube and thoroughly chilled in a glass of cracked ice. 
If a large amount of serum is to be used, for example, 5 c.c., it is well to 
place about 0.1 to 0.2 c.c. of pure undiluted corpuscles, after their last 
washing, in the bottom of a centrifuge tube. This quantity of corpuscles 
does not materially affect the dilution of the serum. If a smaller amount 
of serum is used, such as 1 c.c., it is well to add 8 c.c. of a 2.5 per cent, sus- 
pension of chilled corpuscles, and after keeping the mixture at or near 0° 
to 2° C. for an hour, centrifuging the mixture; the final dilution of 1 : 20 is 
secured by adding 10 c.c. of salt solution to the supernatant diluted 
serum. 
Method of Removal of Natural Hemolysins from Inactivated Serum.—This 
method refers especially to the removal of natural antisheep hemolysin 
from human sera for the Wassermann and other complement-fixation tests. 
(1) Heat the serum in a water-bath to 55° C. for ten minutes.. This suffices 
for the inactivation of complement. (2) For approximately each cubic 
centimeter of serum add 1 drop of sediment of washed corpuscles. Mix 
and stand at room temperature for about half an hour or in a water-bath 
or thermostat at 38° C. for fifteen minutes. (3) Centrifuge and remove the 410 
HEMOLYSINS 
supernatant serum. (4) Heat the sera to 55° C. for fifteen minutes to re- 
move anticomplementary substances. 
The time allowed for absorption of hemolysin may be less than stated 
unless a serum contains relatively large amounts of hemolysin. Kahn1 
uses a period of ten minutes; Simon2 has advocated a method employing 
ten minutes in a water-bath at 37° to 40° C. 
Method of Determining Natural Hemolysins in Serum.—To ascertain 
whether or not a certain natural hemolysin is present in a serum it is 
merely necessary to inactivate the serum, and to a measured amount, for 
example, 0.2 c.c., add 1 c.c. of complement serum (1 : 20) that is known 
to be free of the particular hemolysin in question, and 1 c.c. of a 2.5 per 
cent, suspension of the corresponding corpuscles. Sufficient salt solution 
is added to bring the total volume to 3 or 4 c.c. The mixture is then in- 
cubated at 37° C. for one or two hours, when the occurrence of hemolysis 
indicates the presence of the hemolysin for the corpuscles employed. 
If the serum is fresh the native complement may be employed. In 
this case the test is conducted by placing 0.1 or 0.2 c.c. serum in a test- 
tube with 1 c.c. of suspension of washed corpuscles homologous with the 
hemolysin being sought, followed by incubation in a water-bath at 38° C. 
for one hour, when the presence or absence of hemolysis will be evident. 
To determine the amount of natural hemolysin by titration dilute 
the serum with 9 parts of normal salt solution, and to a series of test- 
tubes add increasing amounts of 0.05 c.c., 0.1 c.c., 0.2 c.c., 0.4 c.c., 0.8 c.c., 
1 c.c., and 2 c.c., corresponding respectively to 0.005, 0.01, 0.02, 0.04, 0.08, 
0.1, and 0.2 c.c. of the undiluted serum. Add 1 c.c. of a 5 per cent, dilution 
of fresh hemolysin-free guinea-pig serum as complement, and 1 c.c. of a 
2.5 per cent, suspension of the corpuscles; sufficient salt solution is added 
to make the total volume about 3 c.c. After shaking, the tubes are placed 
in the incubator at 37° C. for one hour, removed, and the results read, or 
the tubes may be placed in a refrigerator overnight and the results read 
in the morning. 
SERUM DIAGNOSIS OF PAROXYSMAL HEMOGLOBINURIA 
First Method.—According to Ehrlich, a small tourniquet should be 
applied about the base of one of the patient’s fingers, and this is then kept 
immersed in ice-cold water for half an hour. Blood from the finger thus 
constricted is then collected in a Wright capsule, and blood from a finger 
of the other hand is used as a control. Both are allowed to clot and are 
then centrifugalized. The serum from the finger held in iced water is tinged 
red from dissolved hemoglobin, whereas the control serum is not tinged 
or at least not tinged so deeply. 
Second Method.—Donath and Landsteiner have applied Ehrlich’s method 
in vitro. Their method consists of collecting blood in a small test-tube, 
cooling to 0° C. for half an hour, heating subsequently to 37° C. for three 
hours. The presence or absence of hemolysis is observed, and the results 
compared with those obtained from normal blood treated in the same manner 
and at the same time. 
Third Method.—This technic is carried out in vitro in the following 
manner: Pipet 2 c.c. of the patient’s blood in a small test-tube and separate 
the serum. At the same time place 1 c.c. of blood in a centrifuge tube con- 
taining 9 c.c. of a 1 per cent, solution of sodium citrate in normal salt solu- 
tion. Wash the corpuscles twice and suspend the sediment in 10 c.c. of 
1 Jour. Lab. and Clin. Med., 1921, 6, 218. 
2 Jour. Amer. Med. Assoc., 1917, lxxii, 1535. SERUM DIAGNOSIS OF PAROXYSMAL HEMOGLOBINURIA 
411 
normal salt solution. Then, secure a cubic centimeter of a fresh serum from 
a normal person. Proceed to make the following mixtures: 
Tube 1: 0.2 c.c. patient’s serum + 1 c.c. corpuscle suspension. 
Tube 2: 0.1 c.c. patient’s serum + 1 c.c. corpuscle suspension. 
Tube 3: 0.2 c.c. normal serum + 1 c.c. corpuscle suspension. 
Tube 4: 0.1 c.c. normal serum + 1 c.c. corpuscle suspension. 
Tube 5: 1.0 c.c. corpuscle suspension. 
Add sufficient norma1 salt solution to each tube to make the total volume 
measure 2 c.c. Shake gently, and place in the refrigerator at a low tem- 
perature (not higher than 4° C.) for an hour. Shake each tube gently and 
place them in the incubator at 37° C. for two hours. The tubes are then 
centrifuged and the presence or absence of hemolysis is noted. Usually 
the patient’s serum shows hemolysis of greater or less degree. 
Similar mixtures may be made with the patient’s serum and the cor- 
puscles of a normal person. The hemolytic substance is capable of lysing 
these to the same degree that it does the patient’s own cells. 
The Tonicity Test for Measuring the Resistance of Red Blood-corpuscles 
to Hypotonic Saline Solutions.—The corpuscles are exposed to a series of 
solutions of varying strengths of sodium chlorid in test-tubes. If the tests 
are made only occasonally it is better to employ freshly prepared solutions 
in order to avoid errors due to evaporation of stock solutions. 
1. Chemically pure sodium chlorid is dried in a hot-air oven for one 
hour at about 170° C. A 1 per cent, solution is then prepared by dissolving 
5 gm. in 500 c.c. distilled water. 
2. A convenient series is 0.7 to 0.2 per cent, solutions of salt with varia- 
tions of 0.02 per cent. These may be prepared in amounts of 10 c.c. in 
test-tubes by mixing varying amounts of the stock 1 per cent, solution of 
salt and distilled water, great care being necessary in pipeting to secure 
accurate mixtures and results: 
0.7 per cent. (7 c.c. of 1 per cent, salt solution + 3 c.c. of distilled water. 
0.68 per cent. (6.8 c.c. of 1 per cent, salt solution + 3.2 c.c. of distilled 
water. 
0.66 per cent. (6.6 c.c. of 1 per cent, salt solution + 3.4 c.c. of distilled 
water. 
0.64 per cent. (6.4 c.c. of 1 per cent, salt solution + 3.6 c.c. of distilled 
water. 
0.62 per cent. (6.2 c.c. of 1 per cent, salt solution + 3.8 c.c. of distilled 
water. 
0.6 per cent. (6.0 of 1 per cent, salt solution + 4.0 c.c. of distilled water. 
0.58 per cent. (5.8 c.c. of 1 per cent, salt solution + 4.2 c.c. of distilled 
water. 
3. Similar dilutions are made, until the final dilution is reached. In 
many instances it may not be necessary to use so large a number of dilu- 
tions, as from 0.5 to 0.2 per cent, may be sufficient range to indicate the 
tonicity. 
4. Five c.c. of blood are aspirated, under aseptic precautions, from 
an arm vein of the patient, and immediately placed in 25 c.c. of sterile 
1 per cent, sodium citrate in 0.85 per cent, sodium chlorid to prevent co- 
agulation. The flask or large centrifuge tube is gently shaken, and the 
mixture is centrifuged at sufficient speed to throw down the corpuscles. 
After this one washing the supernatant fluid is removed, leaving the ery- 
throcytes at the bottom of the tube. 
5. A series of small test-tubes (10 by 1 cm.) are marked appropriately 
and placed in a rack. To each tube 3 c.c. of the various hypotonic salt 412 
HEMOLYSINS 
solutions and 0.05 c.c. of the red blood-corpuscles (about 1 drop) are added. 
The salt solution and corpuscles in each tube are well mixed and the whole 
series is placed in a water-bath for one hour, followed by one or two hours 
at room temperature, after which the readings are made. 
As shown by Lewis1 the temperature exerts an important influence upon 
the results, hemolysis becoming progressively more marked as the tempera- 
ture is decreased from 37° to 5° C. Some observers read one hour after 
adding the blood and allowing the mixtures to stand at room temperature, 
the tubes being centrifuged in order to permit accurate readings. Formerly 
I used a period of eighteen hours in the refrigerator, but have found that 
the above method permits the settling of the corpuscles to a sufficient de- 
gree for accurate readings on the same day as the test is set up and without 
the extra labor of centrifuging. 
6. The tube of lowest dilution—even if it shows but a trace of hemolysis 
—represents the point of initial hemolysis or minimal resistance. The 
strength of salt solution in which all the corpuscles are hemolyzed repre- 
sents the point of complete hemolysis or maximal resistance. 
7. If the test is being frequently used it is good practice to prepare 
26 stock solutions of salt varying from 0.7 to 0.2 per cent, in gradations of 
0.02 per cent. Used bottles of good grade glass, scrupulously clean and well 
stoppered to reduce evaporation to a minimum, should be employed. 
Short Method.—A short method adapted for occasional use in the clinical 
aboratory is as follows: 
(1) Arrange 11 small test-tubes and carefully pipet the following amounts 
of a 1 per cent, solution of salt, accurately prepared as described above: 
1.4, 1.3, 1.2, 1.1, 1.0, 0.9, 0.8, 0.7, 0.6, 0.5, and 0.4 c.c. 
(2) Add to each tube in order the following amounts of distilled water: 
0.6, 0.7, 0.8, 0.9, 1.0, 1.1, 1.2, 1.3, 1.4, 1.5, and 1.6 c.c. Mix the contents 
of each tube. 
(3) This prepares a series of dilutions of salt as follows: 0.7, 0.65, 0.6, 
0.55, etc., to 0.2 per cent, in gradations of 0.05 per cent. 
(4) These tubes are carried to the patient. A finger is pricked to secure 
a free flow of blood and 1 drop is added to each tube. The contents are 
gently mixed. 
(5) The readings are made after incubation at 37° C. for one hour followed 
by two hours at room temperature. Or the tubes may be allowed to stand 
at room temperature for three hours, when the readings are made. In case 
of doubt regarding the points of initial and maximal resistance the tubes 
may be centrifuged. 
Method for Measuring the Resistance of Red Blood-corpuscles to 
Saponin.—(1) A stock 1 : 1000 solution of saponin (Merck’s) should be 
prepared in physiologic saline solution. It is well to prepare this as re- 
quired by dissolving 0.1 in 100 c.c. saline or 0.01 gm. in 10 c.c. 
• From this solution twelve dilutions are prepared with saline solution 
varying from 1 : 25,000 to 1 : 36,000, as follows: 
0.1 c.c. of 1 : 1000 + 2.4 c.c. saline = 1 : 25,000. 
0.1 c.c. of 1 : 1000 + 2.5 c.c. saline = 1 : 26,000. 
0.1 c.c. of 1 : 1000 + 2.6 c.c. saline = 1 : 27,000. 
0.1 c.c. of 1 : 1000 + 2.7 c.c. saline = 1 : 28,000, etc. 
(2) In a series of test-tubes place 1 c.c. of each solution. 
(3) Blood should be collected in citrate solution as described in the 
tonicity test with hypotonic saline solution. To each tube add 0.1 c.c. of 
citrated blood, mix, and place in a water-bath for one hour. The readings 
1 Jour. Exper. Med., 1909, 11, 593 SERUM DIAGNOSIS OF PAROXYSMAL HEMOGLOBINURIA 
413 
may be made at once with the aid of centrifuging or after the corpuscles 
have settled for a few hours in a refrigerator. 
(4) If washed corpuscles are employed the series of dilutions of saponin 
should be higher, as, for example, 1 : 35,000 to 1 : 46,000. One drop of 
washed corpuscle sediment may be employed. 
(5) Whole blood may be used, but in this case the dilutions of saponin 
should be lower. A drop of blood from the finger may be added to each 
tube. The technic described by Neilson and Wheelon1 is excellent. 
Method for Determining Bacterial Hemolysins.—Two methods may be 
employed. In the plate method 1 c.c. of sterile defibrinated blood is added 
to 6 to 10 c.c. of 2 per cent, agar, previously boiled and cooled to 40° to 
45° C.; the mixture is well shaken, inoculated with the bacteria, and poured 
into a sterile Petri dish, followed by incubation. Hemolysis is indicated by 
the presence of clear zones around the colonies. 
In the tube method the micro-organism is cultivated in an appropriate 
broth medium and preferably one that is not hemolytic due to protein 
extractives and salts. Young cultures are to be prefered (twenty-four to 
forty-eight hours) and may be used unfiltered or the organisms removed 
by centrifuging. Varying amounts are placed in sterile test-tubes with 1 
c.c. of a 1 per cent, suspension of washed rabbit, human, sheep, horse, 
guinea-pig, or other corpuscles. These mixtures should be incubated in 
a water-bath at 38° C. for about two hours, when the results may be read 
at once or after the tubes have rested in a refrigerator over night. 
It is well to include a set of controls set up in exactly the same manner, 
but using a portion of the same broth (sterile) instead of the culture. This 
will give a check on the hemolytic activity of the culture-medium. 
1 Jour. Lab. and Clin. Med., 1921, 6, 454. CHAPTER XXI 
VENOM HEMOLYSIS 
Nature of Venom Hemolysis.—In a previous chapter the statement was 
made that certain snake poisons, and especially cobra venom, are actively 
hemolytic. Flexner and Noguchi1 first demonstrated that the blood-cor- 
puscles of certain species of animals undergo hemolysis when a suitable 
serum is present, and believed that the venom contained an amboceptor 
that was active with serum complement. 
Shortly afterward Kyes2 discovered that venom may hemolyze the 
corpuscles of certain animals without the presence of serum, and believed 
that the complement-like activator was contained within the corpuscles, 
to which he accordingly applied the name endocomplement. 
Later Kyes2 confirmed Calmette’s observation that practically any 
serum, when heated to 65° C. and higher, showed an increased activity in 
the process of venom hemolysis. Kyes and Sachs3 then concluded that 
endocomplement was not of the nature of a thermolabile complement, 
but was, rather, a combination of lecithin and the stromata of erythro- 
cytes. 
Kyes later succeeded in combining cobra venom and lecithin by shaking 
a watery solution of venom with a solution of lecithin in ether, forming 
cobra-lecithid, which was found to be actively hemolytic. 
The erythrocytes of various animals differ in their susceptibility to 
venom hemolysis. For instance, those of the dog and guinea-pig are most 
susceptible to the process; those of the ox, goat, and sheep are entirely 
refractory, whereas those of the horse, rabbit, rat, pig, and man occupy 
an intermediate position. Sacks suggested that the variation in hemolytic 
resistance of red blood-cells from these species of animals was dependent 
on the amount of lecithin contained in the cells. Kyes, on the other hand, 
believes that since all erythrocytes contain sufficient lecithin to activate 
cobra venom, the varying susceptibility depends rather on the availability 
of the intracellular lecithin for the reaction, i. e., whether the lecithin in the 
cell is available in a free state. 
According to this theory, therefore, any factor that modifies the avail- 
ability of the cell lecithin may modify the susceptibility of the cells for 
hemolysis with cobra venom. 
Noguchi1 has questioned the correctness of this view. He holds that 
although lecithin exists in the stroma of all kinds of corpuscles, it is not 
present in a form available for venom activation, and that the degree of 
susceptibility to hemolysis depends chiefly upon the amount of ether- 
soluble activators present in the cells, as, for example, fatty acids, par- 
ticularly oleinic acid, and their soluble soaps. In his opinion heating an 
inactive serum to 65° C. and higher renders it active with venom, owing 
to the presence of a protein compound of lecithin. 
1 Jour. Exper. Med., 1902, vi, 277. 
2Berl. klin. Wchnschr., 1902, xxxix, 886; ibid., 1903, xlii, 21; ibid., 1903, xlii, 956; 
Biochem. Ztschr., 1907, iv, 109; Jour. Infect. Dis., 1910, vii, 181. 
3 Berl. klin. Wchnschr., 1903, xliii, 21, 57, 82. 
4 Jour. Exp. Med., 1907, ix, 436. 
414 VENOM HEMOLYSIS IN SYPHILIS 
415 
A normal serum may, therefore, contain two activators, one being 
thermolabile and resembling complement (inactivated by calcium chlorid), 
and the other being thermostabile and a protein lecithin. By adding oleinic 
acid or its soluble soap to a non-activating serum the latter is rendered 
highly active so far as venom hemolysis is concerned. Hence while Kyes 
regards lecithin as the chief component of endocellular complement, Noguchi 
regards the fatty acids, neutral fats, and soluble soaps as the active agents. 
Other observers consider the fatty acids and soap as indirect activating 
agents in venom hemolysis, in that they possess the power of modifying the 
cell and rendering the intracellular lecithin available for the formation of 
complete hemolysin. 
On the other hand, in susceptible cells the union of cobra venom and 
lecithin occurs directly with the formation of the complete hemolysin, Kyes’ 
cobra-lecithid, due to the splitting of the fatty acid radical from the lecithin. 
Coca,1 von Dungern and Coca,2 and Manwaring3 regard this product as a 
venom-free lecithin derivative, and not as a lecithin. They prefer to call 
the active principle “cobralecithinase,” and the complete hemolysin “mono- 
fatty-acid-lecithin. ’ ’ 
According to Kyes, the relative amounts of lecithin and venom ambo- 
ceptor show quantitative relationship comparable to serum amboceptors 
and complements, namely, that within certain limits, the larger the amount 
of venom, the smaller the amount of lecithin necessary to effect hemolysis; 
and, conversely, the larger the amount of lecithin, the smaller the amount 
of venom required. 
Further reference to the intimate relationship that exists between lipoids 
and complements and hemolysis is also made in the discussion on the nature 
of complements. 
VENOM HEMOLYSIS IN SYPHILIS 
The first application of venom hemolysis was made by Weil,4 who found, 
in testing the hemolytic powers of cobra venom with cells derived from 
persons suffering from different diseases, that the red cells of syphilitic in- 
dividuals offered a characteristic resistance. Various explanations have 
been offered for this phenomenon: 
1. It was argued that the quantity of red-cell lecithin is actually dimin- 
ished in syphilis after the primary stage because pallidum toxin attacks 
the same lipoidal substances of tissue cells as does cobra venom, in this 
way accounting for the diminished amount of lecithin that can be extracted 
in syphilis, as compared with that obtained from normal tissues. Accord- 
ingly, the increase in resistance is only apparent, and is due rather to the 
fact that there is insufficient endocomplement for the venom amboceptor. 
2. Another explanation offered was that the increased resistance of the 
red cells of syphilitic persons to venom hemolysis is due to the fact that 
pallidum toxin attacks endocomplement, and that the cells become specific- 
ally immunized to this deleterious influence in much the same way that 
repeated injections of such a hemolytic agent as saponin leads in rabbits 
to the production of red cells, which show a marked resistance to saponin 
hemolysis, but not to any other hemolytic agent. 
3. Pallidum toxin was believed to so affect the lecithin content of red 
1 Ztschr. f. Immunitatsf., 1912, 12, 134. 
2 Jour. Infect. Dis., 1912, 10, 57. 
3 Ztschr. f. Immunitatsf., 1910, 6, 513. 
4 Proc. Soc. Exper. Biol, and Med., 1909, vi, 49; ibid., 1909, vii, 2; Jour. Infect. Dis., 
1909, vi, 688. 416 
VENOM HEMOLYSIS 
cells as to render a smaller quantity of it available in a free state for union 
with the venom amboceptor to form the hemolysin. 
4. Another theory advanced was that pallidum toxin effects a dissocia- 
tion of red cells between lecithin and cholesterin, the latter substance causing 
inhibition of hemolysis. 
Whatever may be the true explanation, the fact has been quite well 
attested that the red cells of a large percentage of persons in the tertiary 
stage of syphilis exhibit a characteristic increased resistance to venom hemol- 
ysis, and while the cobra hemolysis test in this disease is of secondary im- 
portance to the Wassermann reaction as a diagnostic procedure, yet it 
represents one of the most interesting of biologic phenomena, and may 
possibly be employed in other clinical methods. 
Technic of the Cobra Venom Tests 
Preparation of Venom Solution.—A 1 : 1000 stock solution of dried 
cobra venom is prepared by accurately weighing out 0.01 gm. of dried 
pulverized venom and dissolving this in 10 c.c. of normal saline solution 
(1 : 1000). This stock dilution is best preserved in amounts of 1 c.c. in 
sealed ampules, kept in the frozen state in the ice chest in a wide-mouthed 
well-stoppered vacuum bottle containing salt and ice (Schwartz). 
Each cubic centimeter is sufficient for making three tests, so that the 
10 ampules will be enough for 30 tests, or 1 gm. of venom for 3000 reactions. 
Or the dried venom may be weighed out in amounts of 0.001 gm. in test- 
tubes, and diluted, just before being used, with 1 c.c. of normal salt solu- 
tion (1 : 1000). Immediately before the tests are conducted subdilutions 
are prepared of the stock dilution (1 : 1000), using separate pipets for each, 
as follows: 
Solution A: 1 : 10,000 = 1 c.c. stock solution + 9 c.c. normal saline 
solution. 
Solution B: 1 : 15,000 = 2 c.c. solution A + 1 c.c. normal saline solu- 
tion. 
Solution C: 1 : 20,000 = 1 c.c. solution A + 1 c.c. normal saline solu- 
tion. 
Solution D: 1 : 30,000 = 1 c.c. solution B + 1 c.c. normal saline solu- 
tion. 
Solution E: 1 : 40,000 = 1 c.c. solution C + 1 c.c. normal saline solu- 
tion. 
These amounts are sufficient for making three tests; if more tests are 
to be made, larger amounts of the various dilutions will keep fairly well 
in a good refrigerator for several days, but it is always well to plan the work 
so that the exact amount will be prepared and no waste occur. 
Each lot of stock solution should be tested occasionally with the cells 
of known normal and positive persons, to make certain that the venom 
is active in these dilutions. These titrations are conducted in the same 
manner as the test. 
Preparation of Blood-cells.—With a sterile syringe blood is drawn from 
a vein at the elbow and 2 c.c. placed in an accurately graduated centrifuge 
tube containing 5 c.c. of a 2 per cent, solution of sodium citrate in normal 
saline solution. The suspension should be shaken gently to insure mixing 
and the prevention of coagulation, but defibrination by means of whipping 
should never be practised. The cells may be prepared at once or placed in 
the ice-chest overnight. Sufficient normal saline solution is added to bring 
the total volume to 15 c.c. Mix gently and centrifuge at low speed until Fig. 128.—Venom Hemolysis. 
First row strongly positive; second row moderately positive; third row weakly positive; fourth and 
fifth rows negative. PRACTICAL VALUE OF THE VENOM TEST IN SYPHILIS 
417 
the supernatant fluid is clear. Draw off the fluid, add more normal saline 
solution, mix up the cells, and centrifuge again until clear. Repeat this 
process once more so that all traces of serum will be removed. After com- 
pleting the last centrifugalization, which should be thorough (ten minutes), 
the cells are diluted with 25 volumes of normal saline solution, which makes 
a 4 per cent, suspension—for instance, 0.8 c.c. of corpuscles would require 
20 c.c. of diluent. 
Just what influence sodium citrate has upon the process is not known, 
but it is certain that satisfactory results are seldom obtained with blood 
defibrinated by whipping with rods of glass beads. Similarly, if centrifuged 
too rapidly, cells are broken up or rendered more susceptible to the venom 
amboceptor. 
The Test.—Into a series of 5 test-tubes (4 by | inches) place 1 c.c. of 
each dilution of venom, and label each tube correctly; add 1 c.c. of the 4 
per cent, suspension of cells to each tube, shake gently, and incubate for 
one hour at 37° C. Except in cases where the venom dilutions are being 
frequently used and are known to be reliable, controls should be included. 
I usually add a normal control of cells from a healthy person with the 1 : 
30,000 or 1 : 40,000 dilution and expect complete hemolysis to occur. A 
preliminary reading is made at the end of an hour; the tubes are shaken 
gently and placed in the refrigerator overnight; the final reading is made 
next morning, and should tally quite closely with the preliminary reading. 
Reading the Results.—Unless the cells are derived from a very strongly 
reacting case of syphilis, the 1 : 10,000 dilution will be hemolyzed. The 
normal control tube should show complete hemolysis. If the 1 : 10,000 
tube is not hemolyzed and some of the higher dilutions show hemolysis, 
an error in technic has occurred, and the test should be repeated with fresh 
dilutions. 
The reactions are read and recorded as follows (Fig. 128): 
A 
. B 
C 
D 
E 
H. 
M. H. 
— 
— 
— 
= strongly positive. 
H. 
H. 
M. H. 
— 
— 
= moderately positive. 
H. 
H. 
H. 
S. H. 
— 
= weakly positive. 
H. 
H. 
H. 
H. 
M. H. 
= negative. 
H. 
H. 
H. 
H. 
H. 
= certainly negative. 
H. = complete hemolysis; M. H. = marked hemolysis; S. H. = slight hemolysis; the 
dash (—) = no hemolysis. 
If complete hemolysis has occurred in all tubes after an hour’s incuba- 
tion, the cells are regarded as being hypersensitive to cobra venom. This 
occurs, as a rule, in primary syphilis and in active tuberculosis. 
1. While the test is much simpler than the Wassermann reaction and 
there is less possibility for errors in technic to creep in, it possesses but 
two other advantages, namely: (1) It may react positively in latent or 
tertiary syphilis when the Wassermann reaction may be negative, and 
(2) it may react positively in treated syphilitic cases when the Wassermann 
reaction is negative, and thus point to a continuation of treatment. Corson- 
White and Ludlum1 found 94 per cent., Schwartz2 69.3 per cent., and Stone 
Practical Value of the Venom Test in Syphilis 
1 Jour. Nervous and Mental Diseases, 1910, xxxvii, 721. 
2 New York Medical Journal, 1912, xcv, 23. 418 
VENOM HEMOLYSIS 
and Schottstaedt1 90.9 per cent, of positive reactions in the active stages 
of syphilis. 
2. The test is positive in but about 20 per cent, of cases of tabes dorsalis 
and general paralysis (White and Ludlum), a finding obviously inferior to 
the Wassermann reaction. 
3. During primary syphilis the cells are hypersensitive and positive re- 
actions are but occasionally obtained. 
4. Positive reactions may occur in cancer, but otherwise the test is 
quite specific, and may, in selected cases, prove a valuable adjunct to the 
Wassermann reaction. However, with a more improved technic in per- 
forming the Wassermann reaction, and especially if antigens reinforced 
with cholesterin are used, the venom test is inferior to the Wassermann. 
In cases where syphilis or tuberculosis of the lungs is to be differentiated, 
a negative venom test would indicate tuberculosis, as in this disease the 
cells are hypersensitive. 
THE PSYCHOREACTION OF MUCH AND HOLZMANN 
Normal serum, when added to a lytic dose of cobra venom and human 
red blood-cells, will not interfere with hemolysis. According to Much and 
Holzman,2 however, if the serum obtained from a patient suffering from de- 
pressive mania or dementia prsecox is added to the mixture of venom and 
human red blood-cells, the expected hemolysis does not take place. 
Technic.—A 1 : 5000 dilution of cobra venom is prepared by diluting 
1 c.c. of the stock dilution (p. 416) with 4 c.c. of normal saline solution. 
Enough of the patient’s blood is collected from a vein at the elbow to yield 
at least 1.5 c.c. of serum; heat the serum to 55° C. for an hour. Prepare 
a 5 per cent, suspension of washed human blood-cells. An effort should be 
made, if possible, to use as a control the cells of a person which are known 
to be readily hemolyzed by 1 c.c. of the 1 : 5000 dilution of venom in half 
an hour. 
Into a series of three small test-tubes place 0.4 c.c. of the patient’s serum 
and decreasing amounts of venom—1, 0.8, and 0.5 c.c. respectively. Next 
add to each tube 1 c.c. of the blood-corpuscle suspension. The total volume 
in each tube is brought up to 2.5 c.c. by the addition of normal saline solu- 
tion. 
A similar series of tubes should be set up as controls, the patient’s serum 
being omitted. The following table shows the arrangement: 
Tube 1: 0.4 c.c. serum -j- 1 c.c. venom + 1 c.c. blood. 
Tube 2: 0.4 c.c. serum + 0.8 c.c. venom + 1 c.c. blood. 
Tube 3: 0.4 c.c. serum + 0.5 c.c. venom + 1 c.c. blood. 
Tube 4: 1 c.c. venom -f- 1 c.c. blood. 
Tube 5: 0.8 c.c. venom + 1 c.c. blood. 
Tube 6: 0.5 c.c. venom + 1 c.c. blood. 
Tube 7: 0 1 c.c. blood. 
The tubes are shaken gently and incubated for an hour at 37° C., when 
a preliminary reading is made. If the control tubes 4, 5, and 6 are hemo- 
lyzed, a positive reaction would be indicated by the inhibition of hemolysis 
in the first three tubes, 1, 2, and 3. Control tube 4 at least should 
pletely hemolyzed at the end of half an hour, and in a positive reaction 
tube 1, containing the same amount of venom with the patient’s serum, 
will show inhibition of hemolysis. 
1 Arch, of Int. Med., 1912, x, 8. 
2 Miinch. med. Wchn., 1909, 20. VENOM HEMOLYSIS IN CANCER 
419 
Practical Value.—Corson-White and Ludlum1 have found the reaction 
positive in 80 per cent, of cases of the catatonic form of dementia praecox 
and in 62 per cent, of the hebephrenic type. Only 3 out of 37 cases of manic- 
depressive insanity reacted positively. Among controls with serums of 
other diseases positive reactions were secured in one case each of cerebro- 
spinal syphilis, tertiary lues, exophthalmic goiter, and confusional insanity, 
and in 2 cases each of general paralysis and epilepsy. The aforementioned 
observers claim, however, that if the venom is carefully standardized with 
blood-cells that are completely hemolyzed in 1 : 5000 dilution of venom 
in thirty minutes, the reaction possesses some diagnostic value, having 
been found, under these conditions, to yield positive reactions in 87 per 
cent, of cases of dementia praecox and in 100 per cent, of catatonics. 
In a study of this reaction RosanofP found that practically all human 
sera possess the power, in a more or less pronounced degree, of inhibiting 
the hemolytic action of cobra venom on human corpuscles. A high degree 
of inhibition of hemolysis by serum was found among 8.1 per cent, of appar- 
ently healthy persons and was not specific for any psychosis, although 
the reaction was present in 57.9 per cent, of a series of cases of dementia 
praecox. No relationship was found between the Wassermann and Much- 
Holzmann reactions. 
Moss3 has likewise reported unfavorably upon the diagnostic value of 
this reaction. The action of different sera in favoring or retarding venom 
hemolysis was inconstant. 
According to Citron, the reaction is probably due to interference with 
hemolysis by an increase in the cholesterin of the serum—a possibility more 
apt to occur in diseases of the central nervous system than in any physio- 
logic or other pathologic condition. 
VENOM HEMOLYSIS IN TUBERCULOSIS 
Calmette found that the blood of tuberculous patients may activate 
cobra hemolysin in very small doses, and upon this observation he devised 
a test that yielded about 65 per cent, of positive reactions in tuberculosis. 
Positive reactions have, however, been found in other diseases, and the 
practical value of the test has not been established. 
VENOM HEMOLYSIS IN CANCER 
Although the red blood-corpuscles of the horse may be hemolyzed by 
venom without the aid of serum, Kraus, Graff, and Ranzi4,5 found that 
about 70 per cent, of cancer serums considerably hastened and aided the 
hemolytic process. 
A 1 : 5000 dilution of venom is used. In two series of 4 tubes each place 
respectively 0.1, 0.2, 0.3, and 0.5 c.c. of the patient’s serum (heated); to 
the first series add 0.3 c.c. of the venom solution, and to the second series 
0.15 c.c. of the same. To each of the tubes in the series add 5 drops of a 
10 per cent, suspension of washed horse corpuscles; shake thoroughly and 
incubate at 37° C. Inspect the tubes at the end of fifteen and thirty min- 
utes, and then after one, two, and three hours. 
Positive reactions have also been found in pregnancy after the fourth 
month, in icterus, advanced tuberculosis, and other diseases. 
1 Jour. Nerv. and Mental Diseases, 1910, xxxvii, 721. 
2 Arch. Int. Med., 1909, 4, 405. 
3 Bull. Johns Hopkins Hosp., 1911, 22, 278. 
4 Wien. klin. Wchn., 1911, No. 28. 
5 Miinch. med. Wchn., 1912, No. 59, 574. CHAPTER XXII 
PRINCIPLES OF THE BORDET-GENGOU PHENOMENON OF 
COMPLEMENT FIXATION 
Historic.—With Bordet’s discovery of the hemolysins in 1898, and his 
demonstration of the role of the antibody or sensitizer and alexin in the 
process, new light was thrown upon the bacteriolysins, and the close analogy 
between hemolysis and bacteriolysis soon became apparent. Bordet’s 
discoveries were quickly verified by Ehrlich and Morgenroth and the Ger- 
man school in general, although his views regarding the mechanism of the 
processes were questioned. The controversy soon centered upon the ques- 
tion of the unity or the multiplicity of alexins or complements. Bordet 
at this time advanced his belief in the existence of one alexin or comple- 
ment that would act with any sensitizer or amboceptor, and he still main- 
tains this view. One of the experiments conducted by him, and later made 
in conjunction with his pupil, Gengou, in support of his theory, is now 
known as the Bordet-Gengou phenomenon of complement fixation.1 This has 
become widely known as the precursor of all complement-fixation tests, 
and is the basis of the well-known and invaluable Wassermann reaction 
for the diagnosis of syphilis. 
In devising the technic of this important method Bordet’s main object 
was to show that the complement in a normal serum would unite with 
either a bacteriolytic or a hemolytic amboceptor, and that, by furnishing 
sufficient of either amboceptor, all the complement may be “fixed.” He 
argued that if two or more complements existed in the same serum, as was 
held by Ehrlich, they would demonstrate their presence by exhibiting 
different affinities for these widely varying amboceptors. 
Prior to this Bordet had shown that the addition of a small amount of 
normal serum to an immune hemolysin wTould result in lysis of the homol- 
ogous corpuscles, and that the process could not take place without the 
alexin. He then mixed an emulsion of pest bacilli with antipest serum and 
added a small amount of normal, unheated guinea-pig serum to supply 
the alexin or complement. After allowing the mixture to stand for four 
hours at room temperature, it was sought to determine whether the alexin 
had been fixed by pest antigen and pest amboceptor, or whether it was 
free in the fluid. Bordet knew that the normal alexin serum used in the 
experiment could produce hemolysis of corpuscles with a homologous ambo- 
ceptor, so he tested for free alexin by subsequently adding to the mixture 
antirabbit hemolysin and rabbit corpuscles. Hemolysis did not occur 
because the alexin or complement had been bound by the pest antigen 
and amboceptor. When a normal serum wTas substituted for the antipest 
serum, hemolysis occurred, because the normal serum contained no sensi- 
tizers or amboceptors that could unite with the pest bacilli and “fix” the 
alexin, which, therefore, remained free, and wdien the hemolysin and red 
corpuscles were subsequently added, united with them to lyse the red cells. 
In this way the corpuscles and hemolysin served as indicators for free or 
unfixed alexin or complement, just as litmus or phenolphthalein may be 
used as a test for the presence of an acid or an alkali. 
1 Ann. d. l’Inst. Pasteur, 1901, 15, 289. 
420 THE ORIGINAL COMPLEMENT-FIXATION REACTION 
421 
The Original Complement-fixation Reaction.—In this paper Bordet and 
Gengou1 gave the results attained with three different antigens and their 
respective antiserums—pest, typhoid, and Proteus vulgaris. The details 
of the technic employed with an antigen of pest bacilli and an antipest 
horse-serum are given: 
(a) The antigen consisted of twenty-four-hour cultures of pest bacilli in 
normal salt solution, making a somewhat concentrated emulsion. 
(b) The antipest horse-serum was heated for half an hour at 56° C., 
to remove the alexin or complement. Normal horse-serum (heated) was 
also employed as a negative control. 
(c) As alexin, the fresh serum of a normal guinea-pig wTas used. 
(d) The substance sensibilisatrice or hemolysin consisted of the inac- 
tivated serum of a guinea-pig injected with rabbit’s red blood-cells. 
(e) The corpuscles of the rabbit were washed, to free them of alexin, 
and were used as the indicator antigen. 
To 0.4 c.c. of the pest emulsion, 1.2 c.c. of inactivated antipest serum 
and 0.2 c.c. of guinea-pig alexin were added. This mixture was allowed to 
remain at the ordinary laboratory temperature (18°-20° C.) for several 
hours. In order to ascertain whether or not the alexin had been absorbed, 
hemolysin and erythrocytes were added to the mixture. This was accom- 
plished by sensitizing about 20 drops of washed rabbit’s cells with 2 c.c. 
of inactivated hemolysin for about fifteen minutes, and adding 2 drops to 
each of the test-tubes. Hemolysis did not occur because the alexin had 
been fixed by the pest antigen and antibody. A similar test, conducted 
with normal serum, hemolyzed in a few minutes because the complement 
or alexin remained free in the mixture. Even at this early stage Bordet 
included the important controls on his antigen and serums that are so 
necessary in all complement-fixation tests. 
The following table gives the original details and results of the first 
complement-fixation experiment with pest antigens and antipest serum: 
The Oeiginal Bordet-Gengou Complement-fixation Reaction 
Tube. 
Antigen, 
C.c. 
Serum. 
Alexin, 
C.c. 
Hemolysin and Erythro- 
cytes. 
Results. 
(a)... . 
0.4 
1.2 c.c. inactivated anti- 
pest serum. 
0.2 
2 drops sensitized rab- 
bit’s blood. 
No hemol- 
ysis. 
(&).... 
0.4 
1.2 c.c. inactivated normal 
serum. 
0.2 
2 drops sensitized rab- 
bit’s blood. 
Complete 
hemolysis. 
(«).... 
1.2 c.c. inactivated anti- 
pest serum. 
0.2 
2 drops sensitized rab- 
bit’s blood. 
Complete 
hemolysis. 
(*)-.. 
1.2 c.c. inactivated normal 
serum. 
0.2 
2 drops sensitized rab- 
bit’s blood. 
Complete 
hemolysis. 
(e).... 
0.4 
1.2 c.c. inactivated anti- 
pest serum. 
2 drops sensitized rab- 
bit’s blood. 
No hemol- 
ysis. 
(/)•••• 
0.4 
1.2 c.c. inactivated normal 
serum. 
2 drops sensitized rab- 
bit’s blood. 
No hemol- 
ysis. 
By showing in this manner that the complement of a serum could be 
fixed by either bacteriolytic or hemolytic amboceptors, Bordet endeavored 
.to support his views on the unity of complement. Ehrlich and Morgen- 
roth later verified his findings, and, in addition, by more delicate and com- 
plicated experiments, showed that many complements may be present 
1 Ann. d. l’lnst. Pasteur, 1901, xv, 289. 422 
PRINCIPLES OF THE BORDET-GENGOU PHENOMENON 
in a serum, a fact manifested by a different rate of absorption, by the action 
of specific anticomplements, etc., as mentioned in a previous chapter. 
While Ehrlich’s theory as to the multiplicity of complements has been 
widely accepted, the subject really possesses greater academic than practical 
interest, for experience has shown that the results are the same in comple- 
ment-fixation tests at least, regardless of whether we believe in the unity 
or in the multiplicity of complement, as under ordinary conditions the 
complement or complements in a given quantity of serum are capable of 
being absorbed by bacteriolytic, hemolytic, or other amboceptors. 
History of Applications of the Bordet-Gengou Phenomenon; Discovery 
of the Wassermann Reaction.—Gengou1 showed later that not only cellular 
antigens, such as bacteria and red blood-cells, are capable of stimulating 
the production of amboceptors, but that the proteins in solution, such as 
serum and milk, may produce complement-binding amboceptors in addition 
to precipitins. This subject was later studied more extensively by Moreschi,2 
whose interest became aroused as the result of his theoretic studies upon 
anticomplements. This investigator observed that upon mixing a soluble 
protein with its antiserum precipitation occurred and the existing comple- 
ment disappeared, a coincidence that led him to assert that the complement 
disappeared because it was carried down mechanically in the precipitate. 
This explanation naturally had the effect of leading many to assume that 
the Bordet-Gengou phenomenon of complement fixation may be the result 
of a similar precipitation process, and led to many interesting and valuable 
investigations, especially those made by Gay.3 It is now generally agreed, 
however, that protein amboceptors are formed, and that actual comple- 
ment fixation occurs independently of precipitation. Later Neisser and 
Sachs4 elaborated on Gengou’s studies, and perfected a complement-fixation 
technic for the differentiation of proteins that is much more delicate than 
the precipitin test, and serves to demonstrate and differentiate traces of 
protein, as in blood-stains, so minute in quantity as not to be appreciable 
by the precipitin test. 
Widal and Lesourd5 applied the Bordet-Gengou reaction to the diag- 
nosis of typhoid fever, using an emulsion of typhoid bacilli and the serum 
of a typhoid fever patient, and found that a positive reaction occurred 
more frequently and earlier than the agglutination test. These observa- 
tions were made soon after Bordet and Gengou’s discovery, and were prob- 
ably the first direct and practical application of a complement-fixation 
technic in diagnosis. It was not until several years later, however, that the 
possibilities of the method were seriously considered. 
Hitherto most experiments were conducted with known antigens and 
their antibodies. It was shown, especially in the work of Neisser and Sachs 
on protein differentiation, that when an antigen and its specific antibody 
are present complement is absorbed, and the specific relation existing be- 
tween these bodies was again emphasized. Hence in a complement-fixation 
test, if the antibody is known the antigen may be found, or if the antigen is 
known the antibody may be found, the detection in either instance depending 
upon whether or not complement is absorbed, this being decided by adding 
corpuscles and their amboceptors to the mixture, the absence or the occurrence 
of hemolysis determining this point. In this manner Neisser and Sachs were 
able to diagnose the nature of blood-stains by using solutions of the sus- 
pected stains as antigen, and adding, in different experiments, known antb 
1 Ann. d. l’lnst. Pasteur, 1902, 16, 734. 
2 Berl. klin. Wchn., 1905, No. 37, 1181. 
3 Centralbl. f. Bakteriol., 1905, 39, 603. 
4 Berl. klin. Wchn., 1905, xlii, 1388. 
5 Comp. rend. Soc. biol., 1901, 53, 841. EARLY WORK ON THE WASSERMANN REACTION 
423 
serums secured by injecting rabbits with various bloods. When a positive 
complement-fixation reaction occurred, they concluded that the antigen 
of the blood corresponded to the known antibody, and they were thus 
able to identify the species of animal from which the blood in the stain was 
derived. 
Wassermann and Sachs, encouraged by these results, endeavored by 
complement-fixation tests to show the existence of antigens in diseased 
organs, using tuberculous glands and lungs with an antituberculous serum 
and the serums of tuberculous persons. These investigations finally led to 
the serum diagnosis of syphilis, in which the antigen was supplied by tissues 
containing large numbers of the Spirocheta pallida. By furnishing the 
antigen it was hoped that the luetic antibody could be detected in the body 
fluids through the absorption of complement by the union of the antigen 
and its antibody in the complement-fixation test. 
At this time Schaudinn and Hofmann discovered the spirochete of 
syphilis, a finding that served to focus the attention of the medical world 
upon this disease. In co-operation with Neisser, who was conducting ex- 
tensive researches on experimental syphilis in monkeys, Wassermann and 
Bruck1 applied the complement-fixation method to the study of these experi- 
mental infections, and published a report of their work on May 10, 1906. 
At first monkeys were immunized with aqueous extracts of human 
chancre, condylomata, syphilitic placenta, etc., and their serums, mixed 
in vitro with these extracts, were found to give the complement-fixation 
reaction. Since these results may have been due to protein amboceptors 
or precipitins produced simultaneously by the injection of human serum 
contained in the extracts, the experiment was carried out with extracts 
of bone-marrow and other organs of syphilitic monkeys used to obviate 
this error. It was found, however, that the inactivated serums of syphilitic 
monkeys reacted positively with antigens of either human or monkey lesions, 
and regardless of whether the monkeys had been injected with human 
extracts, since, after ordinary cutaneous infection, their serum would show 
complement fixation. These early reports also showed a high specificity 
for complement fixation, as monkey immune serum did not react with ex- 
tracts of normal organs or normal monkey serum with extracts of syphilitic 
organs. 
Just fourteen days after Wassermann, Neisser, and Bruck published their 
report, a second paper on the same subject appeared, showing the work of Detre.2 
Using aqueous extracts of luetic papules, liver, pancreas, and tonsillar 
exudate as antigens Detre performed the complement-fixation method 
with the serums of 6 syphilitic and 4 normal persons, finding positive reac- 
tions with two of the six luetic serums. 
Early Work on the Wassermann Reaction; Discovery of Its Non-specific 
Character.—-In 1906 Wassermann and Plaut3 studied the cerebrospinal fluids 
of 41 cases of paresis, and found positive reactions in 32, 4 cases reacting 
doubtfully and 5 negatively. In the following year Levaditi and Marie4 
and Schutze5 observed positive reactions with the cerebrospinal fluid of 
tabetics, whereas Morgenroth and Stertz6 confirmed the previous finding 
in paresis. Since then numerous investigators have corroborated these 
observations, and while all the evidence tended to strengthen the belief 
1 Deutsch. med. Wchn., 1906, 32, 745. 
2 Wien. klin. Wchn., 1906, 19, 619. 
3 Deutsch. med. Wchn., 1906, 32, 1769. 
4 C. R. Soc. Biol., 1907; Sem. med., 1907, No. 21. 
5 Allg. Ztschr. f. Psych, u. Psychiat.-gerichtl. Med., 1908, lxv, 565. 
'Virchow’s Archiv., 1907, clxxvii, 166. 424 
PRINCIPLES OF THE BORDET-GENGOU PHENOMENON 
in the luetic origin of general paralysis and tabes, decisive confirmation 
was lacking until Noguchi and Moore,1 in 1913, demonstrated the presence 
of the Treponema pallidum in sections of the cerebral cortex. 
In 1906 Wassermann, Neisser, Bruck, and Schucht2 applied the com- 
plement-fixation test to a large number of cases of syphilis in Neisser’s 
clinic. Aqueous extracts of luetic liver, placenta, glands, chancres, and 
gummata were used as antigens. Of 257 cases in all stages of the disease, 
only 49 reacted positively. With but 19 per cent, positive reactions, the 
method did not appear to have a promising future, although at the present 
time, with a better understanding of the technic and of the importance 
of quantitative factors that greatly influence the results, the value of the 
test has been greatly enhanced. 
As it appeared that after all no method of diagnosis was to be secured 
as the result of the demonstration of the syphilitic antibody in the body 
fluids, Neisser and Bruck determined to return to earlier methods and 
attempt to discover if luetic antigen could be demonstrated in the serums 
of luetics through complement fixation. Antigens prepared of the red 
corpuscles of syphilitic persons gave positive reactions with the serums of 
highly immunized monkeys. Of 160 luetic patients, in 70 per cent, either 
antigen or antibody was found. Later, however, Citron showed that the 
extracts of corpuscles of normal persons yielded similar results which, in 
the light of subsequent discoveries, was due to their content in lipoidal 
substances. 
Up to this time the syphilitic reaction was considered as but a simple 
and direct application of Bordet’s phenomenon, requiring a specific syphilitic 
antigen before complement could be fixed with the syphilis antibody. In 
January, 1907, Weygandt3 reported that he had obtained a positive reac- 
tion in tabes with an extract of normal spleen. Marie and Levaditi, using 
an aqueous extract of normal fetal liver, secured positive reactions with 
the cerebrospinal fluid of paretics, but observed that it was necessary to 
use larger doses than when extracts of syphilitic organs were used. Subse- 
quently other investigators, as, for example, Fleischmann,4 Michaelis,5 
Landsteiner, and Plaut, found that watery extracts of normal organs served 
to fix the complement with luetic antibody. Finally, in December, 1907, 
a profound impression wras created by the discovery made by Landsteiner, 
Muller, and Potzl6 that an alcoholic extract of guinea-pig heart yielded 
results equal to those obtained with an aqueous extract of syphilitic liver. 
These results indicated that the antigenic principle was soluble in alcohol, 
and a prolonged series of investigations on the various lipoids and their 
relation to the reaction was begun. These included the employemnt of 
lecithin by Porges and Meier7; sodium taurocholate and glycocholate by 
Levaditi and Yamonouchi8; cholesterin and vaselin by Fleischmann; oleic 
acid by Sachs and Altman9; acetone-insoluble fractions of alcoholic extracts 
by Noguchi and Bronfenbrenner,10 and many other combinations of various 
lipoidal substances by different investigators. 
These dealt a blow to the theory of the Wassermann reaction, which, 
taken in conjunction with the wide-spread use of the test by inexperienced 
and unskilful persons and the many sources of error, tended to retard an 
earlier appreciation of the great value of the test, and served to swing the 
1 Jour. Exper. Med., 1913, 17, 232. 
2 Ztschr. f. Hyg., 1906, lv, 451. 
3 Munch, med. Wchn., 1907, liv, 1557. 
4 Dermat. Centralbl., 1908-09, 11, 226, 258. 
5 Berl. klin. Wchn., 1907, xliv, 1103. 
6 Wien. klin. Wchn., 1907, 20, 1421, 1565. 
7 Berl. klin. Wchn., 1908, xlv, 731. 
8 C. R. Soc. Biol., 1908, lxiv, 814. 
9 Berl. klin. Wchn., 1908, xlv, 494. 
10 Jour. Exper. Med., 1911, 13, 43. PRODUCTION OF COMPLEMENT-FIXING ANTIBODIES 
425 
pendulum of medical opinion so far in the wrong direction that it is only 
now, with a better understanding of its possibilities and limitations, that 
the method is being established in its proper sphere. Positive reactions 
were said to have occurred in frambesia, leprosy, malaria, pellagra, pneu- 
monia, scarlet fever, typhoid fever, malignant tumors, and practically 
every other disease liable to afflict humanity. The careful work of Citron1 
was largely instrumental in preserving the importance of the reaction until 
a better understanding of the technic resulted in improved and more care- 
ful work, with a greater respect for the real value of the Wassermann re- 
action. 
Although the true explanation of the mechanism of complement fixation 
in syphilis is still lacking, sufficient work has been done to show that the 
specific nature of the reaction is dependent upon a peculiar luetic anti- 
body, and that the older belief in the specificity of antigen, in so far as it 
insisted upon the presence of the Treponema pallidum in the tissues ex- 
tracted for “antigen,” is largely disproved. While the method cannot be 
said to be absolutely diagnostic of syphilis, since positive reactions were 
had in frambesia (yaws), yet it is practically so, especially in those countries 
where the latter disease is unknown or relatively infrequent. 
Kinds of Antigens Inducing the Production of Complement-fixing Anti- 
bodies.—From this brief historic survey it is apparent that many different 
substances are capable of acting as antigens in inducing the production 
of complement-fixing antibodies (for which I propose the name alexofixa- 
gens). Practically every foreign protein is capable of antigenic stimulation, 
although substances belonging to the same class may vary greatly in anti- 
genic activity; for example, among the bacteria, the bacilli of the typhoid- 
colon group are actively antigenic, while the pneumococcus is quite feeble. 
Treponema pallidum (syphilis) and T. pertenue (frambesia tropica) are 
the most active antigens. The substances known to be antigenic in induc- 
ing complement-fixing antibodies may be classified as follows: 
' T. pallidum 
T. pertenue 
B. mallei 
B. abortus (Bang) 
B.typhosus 
B. diphtheriae 
Gonococcus 
Sporotrichia 
Helminths 
Active 
1. Foreign cells < 
1. Bacteria, fungi, and pro- 
tozoa 
B. tuberculosis 
Pneumococcus 
Streptococcus 
Staphylococcus 
Spore-bearing bacilli 
Feeble 
2. Erythrocytes, leukocytes, spermatozoa, and other cells. 
2. Foreign soluble substances <! 
1. Sera. 
2. Albuminous exudates and transudates. 
3. Milk and other albuminous excretions. 
4. Plant proteins. 
In other words, all substances known to induce the production of cytolysins, 
albuminolysins, and precipitins during infection or by artificial immuniza- 
1 Deutsch. med. Wchn., 1907, 33, 1165; Berl. klin Wchn., 1907, xliv, 1370; 1908, xlv, 
518. 426 
PRINCIPLES OF THE BORDET-GENGOU PHENOMENON 
tion may act as alexofixagens, that is, induce the production of alexin or 
complement-fixing antibodies; some substances, however, may not induce 
the production of demonstrable amounts of precipitins and yet act as 
alexofixagens. 
Chemical Nature of Alexofixagins.—As discussed in previous chapters 
antigens are regarded as protein molecules, and those producing complement- 
fixing antibodies are apparently of the same nature. 
The physical and chemical state of the protein, however, bears an im- 
portant relation to its activity in inducing the production of these anti- 
bodies. Solubility in the tissues appears to be an essential character and 
proteins that have been boiled (coagulated) appear to lose their antigenic 
activity. 
Purified vegetable and animal proteins may act as antigens, as shown 
by White and Avery for edestin,1 Lake, Osborne, and Wells2 for wheat and 
rye gliadins and hordein of barley, and Kahn and McNeil3 for edestin, 
phaseolin, casein, and lactalbumin. Incomplete proteins appear to be 
devoid of antigenic activity. Gelatin which lacks cystine, tyrosin, and 
tryptophan was found by Starin,4 Kahn and McNeil without antigenic 
activity, and the same was found true of the protamins and histones by 
Wells,6 Schmidt,6 and others. 
The cleavage products of proteins have not been studied in relation 
to the production of complement-fixing antibodies as thoroughly as in 
relation to precipitin production and anaphylaxis. Schmidt7 has found 
deutero-albumose obtained from Witte’s peptone without antigenic activity. 
Gay and Robertson8 found the split products of casein inactive; Kahn and 
McNeil9 and Fink10 have reported similar results with proteoses. The animo- 
acids and simple polypeptids are not antigenic. These investigations have 
indicated, therefore, that the antigenic activity of proteins depends upon 
their chemical structure, and if split or modified by racemization they lose 
in£whole or part their antigenic activity. 
Some investigators have declared that the “nucleoproteins” are actively 
antigenic and highly specific. So-called “nucleoproteins” have been pre- 
pared from bacteria and tissue cells, but, as pointed out by Wells,11 a large 
part of the reported results of experiments with these is to be disregarded 
because of the indefinite chemical nature of the substances employed. 
According to Wells, the nucleoproteins concern three substances from the 
immunologic standpoint: (1) Nucleic acid, which is non-protein, probably 
a glucosid and non-antigenic; (2) the nucleins, which seem to consist of 
nucleic acid firmly bound to proteins and especially the histones and perhaps 
to protamins. The histons, protamins, and nucleic acid are not antigenic, 
and the nucleins composed of these radicals probably are not. (3) The 
nucleoproteins, which appear to be indefinite and loose compounds of any 
or all of the proteins of the cell with either nucleic acid or with the nucleins. 
These are antigenic to the degree depending upon the amount of protein 
present, but these are not an essential part of nucleoprotein and the amount 
present depends largely upon the method employed. In other words, pure 
nucleoproteins from the chemical standpoint do not appear to be antigenic; 
so-called “nucleoproteins” loosely prepared are antigenic to the degree 
depending upon the amount of protein present. 
1 Jour. Infect. Dis., 1913, 13, 103. 4 Jour. Infect. Dis., 1918, 23, 139. 
2 Jour. Infect. Dis., 1914, 14, 364. 8 Ztschr. f. Immunitatsf., 1913, 19, 599. 
3 Jour. Immunology, 1918, 3, 277. 6 Jour. Biol. Chem., 1920, 41. 
7 Univ. of Californ. Pub. in Path., 1916, 2, 157. 
8 Jour. Exper. Med., 1912, 16, 470. 10 Jour. Infect. Dis., 1919, 25, 97. 
9 Jour. Immunology, 1918, 3, 277. 11 Ztschr. f. Immunitatsf., 1913, 19, 599. NATURE OF COMPLEMENT-FIXING ANTIBODY 
427 
Great interest attaches to the possibility of lipoids acting as antigens 
in inducing the production of complement-fixing antibodies. Since various 
lipoidal substances serve as extracts for the Wassermann reaction (erro- 
neously designated as “antigens” in this connection) it is commonly be- 
lieved that they may be antigenic. Warden1 reports that the bacterial 
fats are antigenic and that they are even more antigenic and specific than 
the bactericidal proteins. Meyers2 has reported that acetone-insoluble 
extracts (presumably phosphatids) of tapeworms and tubercle bacilli pro- 
duce complement-fixing antibodies when injected into rabbits, and Bergel3 
has found the same after injecting rabbits with lecithin. Meinicke4 has 
likewise subscribed to the lipoidal nature of the complement-fixation reac- 
tion. Against these positive findings are numerous negative reports by 
Ritchie and Miller,5 Kleinschmidt,6 Thiele,7 and others who were unable 
to produce complement-fixing antibodies with lipoids. The mere fact that 
complement fixation in syphilis occurs with lipoidal substances is no criterion 
of the antigenic activity of lipoids; immunization of rabbits by Fitzgerald 
and Leathes8 with these substances obtained from liver, which were capable 
of acting as “antigens” in the Wassermann reaction, did not engender the 
production of complement-fixing antibodies. 
While it cannot be stated definitely that lipoids are without the power 
of stimulating the production of complement-fixing antibodies, the pre- 
ponderance of evidence is in favor of this view. The presence of even traces 
of protein in the lipoids employed for immunization may suffice for inducing 
the production of these antibodies, and a sufficient amount of work with 
protein free lipoids has not been done to warrant the statement that pure 
lipoids from bacteria, protozoa, or other sources are true antigens capable 
of engendering complement-fixing antibodies either during the course of 
disease or by artificial immunization. 
Very little work has been done with the glucosids. The work of Ford 
and others discussed in previous chapters, indicates that toxic glucosids, 
and particularly those which may be hydrolyzed by enzymes, may pro- 
duce antibodies of antitoxic nature, but purified glucosids are probably 
unable to induce the production of complement-fixing antibodies. 
The antigenic activity of glycoproteins has been studied by Elliott9 
with mucins of ox tendon, ox submaxillary gland, and swine stomach. 
These were found antigenic, but not to the same degree as the simple pro- 
teins. In all probability the antigenic portion of these resided in the pro- 
tein and not in the carbohydrate group. 
Nature of Complement-fixing Antibody (Alexofixagins); Relation to 
Precipitins, Cytolysins, and Albuminolysins.—In a strict sense hemolysis, 
bacteriolysis, and other cytolytic reactions are examples of complement 
fixation in that complement becomes fixed or absorbed by a sensitized 
antigen (antigen + amboceptor or sensitizer). With the discovery of the 
complement-fixation reaction by Bordet and Gengou it was concluded 
that the mechanism was the same, that is, complement was fixed by antigen 
and antibody and, therefore, was not available for hemolysis when cor- 
1 Jour. Infect. Dis., 1918, 22, 133; ibid., 23, 504; 1919, 24, 285. 
2 Ztschr. f. Immunitatsf., 1910, 7, 732; 1911, 9, 530; 1912, 14, 355. 
3 Deut. Arch. klin. Med., 1912, 106, 47. 
4 Ztschr. f. Immunitatsf., 1918, 27, 350; 1919, 28, 280. 
5 Jour. Path, and Bact., 1913, 17, 429. 
6 Berl. klin. Wchn., 1910, 47, 57. 
7 Ztschr. f. Immunitatsf., 1913, 16, 160. 
8 Univ. of Californ. Pub. in Path., 1912, 2, 39. 
9 Jour. Infect. Dis., 1914, 15, 501. 428 
PRINCIPLES OF THE BORDET-GENGOU PHENOMENON 
puscles and hemolysin were subsequently added. According to this view 
the complement-fixing antibody (which may be designated as alexofixagin) 
is of the nature of amboceptor (called the “Bordet amboceptor”), and the 
reaction has been advocated as a more sensitive test for bacteriolytic ambo- 
ceptors in serum than the bacteriolytic tests of Pfeiffer and others described 
in a preceding chapter. 
Neufeld and Haendel,1 however, have reported that at 0° C. sensitized 
cholera spirilla fixed hemolytic complement, while at 37° C. they fixed both 
hemolytic and bacteriolytic complements. They concluded from these 
experiments that the fixation of complement occurring at or near 0° C. 
was due to the complement-fixing antibody and that it is distinct from the 
bacteriolytic amboceptor operative at 37° C. 
Neufeld and Hune2 have, furthermore, found that bacteriolysins and 
complement-fixing antibodies in typhoid immune sera were not identical; 
Haendel3 reported similar findings with typhoid and cholera immune sera, 
and Torrey4 with antigonococcic serum. In other words, while complement- 
fixing antibodies may be of the nature of amboceptors or sensitizers, there 
is little or no evidence to support the vie\V that they are identical with the 
bacteriolysins and proteolysins. 
In 1905 Moreschi,5 in a study of the mechanism of the anticomplementary 
activity of serum to be discussed shortly, came to the conclusion that this 
phenomenon was not due to the development of anticomplements, but 
to a fixation of complement by precipitate formed by precipitin and homol- 
ogous antigen. At the same time and independently of Moreschi the 
investigations of Gay6 showed that precipitates may fix complement and 
that the degree of complement fixation bore a relation to the amount of 
precipitate formed, and this to the amount of antiserum present. 
This explanation, however, was soon challenged and has resulted in a 
large number of investigations bearing upon the relation of precipitins and 
precipitates to the phenomenon of complement fixation. Neisser and Sachs,7 
Muir and Martin,8 Friedberger,9 Sobernheim,10 Altmann,11 and others gener- 
ally observed fixation of complement when precipitation occurred, but 
likewise noted that complement fixation occurred upon mixture of homol- 
ogous antigens and antisera when visible precipitation was not apparent. 
Dean,12 however, in a thorough study of the subject came to the conclu- 
sion that precipitation and complement fixation occur together, although 
visible precipitation may not be apparent under ordinary conditions. He 
believes that the two reactions represent two phases of the same reaction 
and that “flocculent precipitation represents the final stage of a change 
which can be recognized in its earliest and incomplete stage by means of a 
complement-fixation reaction.” 
The question now arises: Is complement fixation a mere matter of phys- 
ical absorption of complement by precipitates or do precipitins and a separate 
complement-fixing antibody occur together or separately in immune sera? 
Gay13 has declared himself in favor of the latter view and that complement- 
fixing lysins and precipitins are separate and distinct, although likely to 
occur in protein antisera. 
1 Arb. a. d. kais. Gesund., 1908, 28. 5 Berl. klin. Wchn., 1905, No. 37. 
2 Arb. a. d. Gesundh., 1907, 25, 164. 6 Centralbl. f. Bakteriol., 1905, 93, 603. 
3 Deutsch. med. Wchn., 1907, 2030. 7 Berl. klin. Wchn., 1905, 1388. 
4 Jour. Med. Res., 1910, 22, 95. 8 Jour. Hyg., 1906, 6, 265. 
9 Deutsch. med. Wchn., 1906, 578. 
10 Centralbl. f. Bakteriol., Ref., 1906, 38, 114 (Beiheft). 
11 Centralbl. f. Bakteriol., 1910, 54, 174. 
12 Centralbl. f. Bakteriol., 1912, 13, 84. 13 Univ. of Californ. Publ., 1911, 2. SPIROCHETIC COMPLEMENT-FIXING ANTIBODY 
429 
In the chapter on Precipitins we discussed the nature of these anti- 
bodies in relation to amboceptors (proteolysins or albuminolysins) according 
to the results of the investigations of Zinsser1 who regards them as identical. 
In relation to the nature of complement-fixing antibodies Zinsser believes 
that the precipitins unite with the antigen forming a complement-fixing 
complex and that since both complement and sensitized antigen are col- 
loidal in nature, they follow the laws governing other mutually precipitating 
colloids and precipitation occurs only as a purely secondary phenomenon 
when concentrations lie within the proper zones. 
Undoubtedly precipitation bears an important relation to complement 
fixation not only in reactions by soluble proteins, as sera and milk and their 
antisera, but likewise in complement-fixation reactions in syphilis, which 
are the most pronounced reactions of this kind. Whether precipitation is 
always caused by precipitins and especially in relation to the Wassermann 
reaction is an open question. Precipitins bear a close relationship to proteol- 
ysins, but the nature of this relationship is not clear. Precipitation appears 
to be an integral part of the complement-fixation reaction with cells, soluble 
proteins and in syphilis and frambesia, but whether precipitation is brought 
about by precipitins or is the result of physicochemic changes in serum 
and antigen by a separate and peculiar complement-fixing antibody can- 
not be definitely stated. Complement fixation in syphilis and frambesia 
is particularly complex and is discussed separately in the following para- 
graphs. 
Specificity of Complement-fixing Antibodies (Alexofixagins).—These 
antibodies possess an extremely high degree of biologic specificity for the 
antigen engendering their production. First discovered in relation to bac- 
terial infections, the complement-fixation test has proved a reliable and 
sensitive means for the diagnosis of some infections and for differentiating 
among those substances capable of acting as alexofixagins, that is, capable 
of stimulating the tissues with the production of a complement-fixing anti- 
body or alexofixagin. Some micro-organisms, as the pneumococcus, engender 
very little complement-fixing antibody; others, as the Gram-negative bacilli 
and particularly those of the typhoid-colon-dysentery-cholera group, induce 
the production of these antibodies to a well-marked degree. Under proper 
technical conditions the complement-fixation reaction is a more accurate 
means for studying biologic relationships than the agglutination and pre- 
cipitin reactions. 
The same may be said of the alexofixagins and complement-fixation 
reaction with soluble proteins of animal and plant origin, as sera, milks, 
extracts of meats, etc. The Wassermann reaction in syphilis and frambesia 
tropica is the only exception to this general rule of biologic specificity on 
the part of alexofixagins and the fixation reaction. 
The Nature and Specificity of the Spirochetic Complement-fixing Anti- 
body (Syphilis “Reagin”).—The unique character of the complement- 
fixation reaction in syphilis and frambesia tropica requires that its nature, 
specificity, and mechanism be discussed separately from other complement- 
fixation reactions. When first discovered by Wassermann, Neisser and 
Bruck, and independently by Detre, it was considered a fixation of comple- 
ment by syphilis amboceptor and extracts of Treponema pallida contained 
in the syphilitic tissues employed for the preparation of antigens; in other 
words, a direct application of the Bordet-Gengou phenomenon. With the 
discovery that “antigens” for this complement-fixation reaction may be 
prepared of non-syphilitic tissues, the whole nature and mechanism of the 
1 Jour. Exper. Med., 1912, 15, 529; ibid., 1913, 18, 219. PRINCIPLES OF THE BORDET-GENGOU PHENOMENON 
430 
reaction in syphilis became sharply differentiated from other complement- 
fixation reactions because the reaction could no longer be regarded as bio- 
logically specific. 
As is now well known, the lipoids1 have always been chiefly concerned in 
the reaction, although Wassermann and Detre and their co-workers naturally 
ascribed the complement-fixing powers of their extracts to the presence 
of the Treponema pallidum. It is, indeed, fortunate that pure cultures 
of the treponema were not available at the time the original studies were 
made, for these would naturally have been employed as antigen, and as 
subsequent work with pallidum antigens has shown complement fixation 
to be quite irregular and less reliable than when lipoidal tissue extracts are 
used, this result, coupled with the imperfect understanding and faulty technic 
of the earlier investigations, would probably have yielded results so dis- 
couraging as to constitute weighty drawbacks to the full development of the 
reaction. 
Notwithstanding the large amount of work that has been done in an 
effort to ascertain the true nature of the syphilitic reaction, a correct ex- 
planation of its mechanism is still lacking, as the large number of theories 
advanced tend to show. 
While lipoidal extracts, as well as normal and luetic serums, may sepa- 
rately absorb or fix small amounts of complement, a mixture of a suitable ex- 
tract and syphilitic serum is capable of fixing large amounts of complement, 
and this constitutes the main principle and all that is definitely known of the 
syphilitic reaction. 
The serum of a syphilitic is characterized, therefore, by the presence of 
this lipodotropic, antibody-like substance, called “reagin” by Neisser, which 
has a great affinity for lipoids and in mixture with them will cause the 
absorption or fixation of complement to a well-marked degree. Instead 
of being an example of complement fixation in a mixture of specific antigen 
with specific antibody, as originally believed, it is technically a non-specific 
reaction, but practically it is highly specific, since this peculiar antibody is 
found in largest amount and most constantly in syphilis, and to a lesser 
extent in practically only one other disease, namely, frambesia. In countries 
and districts where this disease is infrequent or unknown the reaction for 
syphilis is highly specific. 
As will be shown further on, the presence of this lipodotropic substance 
is dependent upon the activities of the Treponema pallidum, and when 
repeated tests continue to show its presence, there is every reason to believe 
that a cure has not been effected, but that the patient still harbors the living 
parasite. 
Citron has advanced the hypothesis that the antibody-producing antigen 
is a toxolipoid, which would explain the fact that while pure lipoids, such 
as lecithin, cannot stimulate antibodies (Bruck), as the toxolipoid does, 
they can, nevertheless, react with the lipodotropic antibodies in vitro, with 
fixation or absorption of complement. As Sachs and Altman point out, 
an equally tenable theory would be that in syphilis the tissues undergo 
such alterations that they can produce antibodies to the lipoid substances 
as may be contained in the spirochetes themselves. 
While the production of this lipodotropic antibody is still unexplained, 
the fact remains, nevertheless, that it forms the basis of the biologic syphilitic 
reaction, and in a mixture with a suitable lipoid is capable of absorbing or 
inactivating complement to a marked degree. Whether or not it is a true 
1 The term “lipoid” (“fat-like”) is applied to compounds that are soluble in ether, al- 
cohol, chloroform, and benzol, but every lipoid is not soluble in all these reagents. SPIROCHETIC COMPLEMENT-FIXING ANTIBODY 
431 
antibody in the sense that it is inimical to the spirochete is doubtful; by 
many it is regarded as a secondary product of cellular activity, and, as 
stated above, has been called syphilis “reagin.” 
Investigators in this field anxiously awaited the isolation of the Tre- 
ponema pallidum in pure culture, believing that if this result were secured 
it would be possible to work with a specific antigen, determine the nature 
of the true syphilis antibody, and possibly establish a complement-fixation 
test specific for syphilis. 
In 1909 Schereschewsky,1 using an antigen of an impure and non-patho- 
genic culture of a spirochete regarded as the Treponema pallidum, reported 
positive complement-fixation reactions with the majority of serums tested. 
In 1912 Noguchi,2 having undoubtedly isolated the spirochete in pure 
culture, prepared antigens and found that, whereas certain long-standing 
or treated cases of syphilis yielded positive reactions with the pallidum 
antigens, the reactions were uniformly negative when the lipoidal extracts 
were used. In primary and secondary syphilis the reactions with pallidum 
antigens were uniformly negative, whereas with the lipoidal extracts they 
were uniformly positive. As a result of his experiments Noguchi concluded 
that in syphilis there is produced a true antibody that reacts specifically 
with pallidum antigen, in addition to the lipodotropic “reagin,” which 
reacts with lipoidal extracts, and whereas the latter indicates activity of the 
infecting agent, the former is a gage of the defensive activity of the infected 
host. 
Craig and Nichols,3 using alcoholic extracts of pure cultures in ascite- 
kidney agar of Treponema pallidum, Spirochete pertenuis, and S. micro- 
dentium, found similar positive reactions in all stages of syphilis with the 
three antigens, but the reactions were weaker and less constant as compared 
with those obtained with a stock of lipoidal extract. 
Similar studies conducted by Kolmer, Williams, and Laubaugh4 with 
aqueous and alcoholic extracts of pallidum cultures showed positive reac- 
tions in secondary, tertiary, and congenital syphilis. The aqueous extracts 
yielded better reactions than the alcoholic extracts; in practically all in- 
stances, however, the reactions were weaker than those obtained with the 
ordinary tissue extracts. Control antigens of typhoid and cholera bacilli 
and sterile culture-mediums demonstrated that all contained lipoidal sub- 
stances that may give weak reactions with the lipodophilic syphilis “reagin.” 
This may explain Schereschewsky’s positive reactions with an antigen of a 
spirochete that in all probability was Spirochete microdentium (Noguchi). 
The true nature of the Wassermann-Detre reaction, therefore, cannot 
be said to have been determined. It is probable that the ordinary syphilis 
reaction is in itself not dependent upon a true antibody, and that the reac- 
tion is not an immunity reaction, but due rather to the presence of peculiar 
tissue products (reagins) altered by the presence and activities of the spiro- 
chetes themselves, and that the Wassermann reaction is an expression of 
this active injury to tissue cells. In addition to this secondary product 
there is probably a true syphilis antibody that may yield specific complement 
fixation with pallidum antigens. 
In so far as the Wassermann reaction is concerned, the true antibody 
is entirely secondary in importance, and the whole question is intimately 
concerned with the chemistry of lipoids. While future researches in im- 
1 Deutsch. med. Wchn., 1909, xxv, 1652. 
2 Jour. Amer. Med. Assoc., 1912, Ivin, 1163. 
3 Jour. Exper. Med., 1912, xvi, 336. 
4 Jour. Med. Res., 1913, xxviii, 345. 432 
PRINCIPLES OF THE BORDET-GENGOU PHENOMENON 
munochemistry may reveal the mechanism of the reaction, the principles 
are at least well understood at present, so that the syphilis reaction is proving 
of great diagnostic and practical value. 
Mason1 has also recently discussed the mechanism of the Wassermann 
reaction in relation to colloidal reactions, believing that the reaction may 
in some way depend on certain physical differences in the sera as showm 
by the increased flocculability of syphilitic serum in the presence of certain 
colloidal solutions. 
Mechanism of Complement Fixation.—The divergent views of Bordet 
and Ehrlich on the mechanism of antigen-amboceptor action have been 
given elsewhere. Bordet believes that the antibody unites directly with 
the antigen, and serves to sensitize and prepare it for direct union with the 
alexin or complement in a manner similar to that of using a mordant in 
aiding the penetration of a dye-stuff. In the absence of the homologous 
and specific antibody (sensitizer) the antigen is incapable of absorbing 
more than very small amounts of complement or none at all. In the absence 
of antigen the sensitizer and complement do not unite, or unite to but a 
very slight degree. The important requirement for complement fixation 
is, therefore, an antigen that has been sensitized by the antibody, and in 
this manner has an increased combining affinity for complement. 
According to Ehrlich and Morgenroth, however, the complement does 
not unite directly with the antigen, but only indirectly through the anti- 
body, which acts as a connecting link or amboceptor between antigen and 
complement. Antigen alone, or even amboceptor alone, binds the comple- 
ment only very slightly or not at all. The important requirement for com- 
plement fixation is an amboceptor attached to its homologous or suitable 
antigen, which increases the affinity and fixing power of the amboceptor 
for the complement (see Fig. 85). 
Complement Fixation a Colloidal Reaction.—Many observations support 
the view that complement fixation by a specific antigen and its anti- 
body is really complement absorption by a precipitate that forms when 
antigen and antibody are mixed. As previously stated, all antigens are 
probably protein in character. While there is some evidence to show 
that lipoids, and even carbohydrates, may act as antigens, there is no 
doubt that the chief antigenic principles of any antigen of pro- 
tein structure; hence when mixed with an immune serum containing spe- 
cific antibodies, it is believed that an invisible precipitate is formed that 
absorbs the complement. With serum antigens the quantity of protein 
is so large that a precipitate can readily be seen (the precipitin test). A 
serum antigen and its antibody may, however, be so highly diluted that, 
when mixed, a precipitate is not visible, although complement may be fixed 
(complement-fixation test for the differentiation of proteins). Moreschi and 
Gay have contended for many years that complements may become entangled 
and absorbed in such precipitates. Reasoning on the basis of the colloidal 
theory, it is possible that transition compounds of very diverse nature are 
formed when antigen, antibody, and a complement are mixed. This view 
concerning the action of complements and anticomplements is supported 
by numerous investigators who have examined the question from the stand- 
point of colloidal reactions. Thus in a complement-fixation test a mixture 
of antigen, antibody, and a complement in definite proportions results in 
the formation of new compounds of opposite electric charge, which tend to 
aggregate in masses (although these may be so small as to be invisible) 
and reduce their surface tension in just the same manner as agglutination 
1 Bull. Johns Hopkins Hosp., 1920, 31, 234. ANTICOMP LEM ENT ARY ACTIVITY OF ANTIGENS AND SERA 
433 
and precipitation are brought about after the colloidal theories. When 
corpuscles and hemolytic antibody are subsequently added hemolysis does 
not occur because free complement is absent. 
A process similar to complement absorption by a specific antigen and 
its antibody is the Wassermann reaction. According to the colloidal theories, 
this reaction may be explained as due to the formation of an invisible precipi- 
tate by interaction between some substance in the serum of a luetic person 
(probably in the nature of an altered globulin), complement, and lipoidal 
substances contained in an alcoholic or watery extract of a normal or a dis- 
eased organ. This view is supported by the fact that euglobulin is known 
to be generally increased in the body fluids of syphilitics, and by analogy 
with the various flocculation tests that have been devised for the diagnosis 
of syphilis, as, for example, the reaction of Porges and Meier, Meinicke, 
Sachs and Georgi, Vernes, etc., which are dependent upon the appearance 
of a precipitate when luetic serum is mixed with an emulsion of lecithin, 
sodium glycocholate, alcoholic extracts of tissues, etc. The exact nature 
of the antibody in syphilitic serums that forms these new compounds with 
lipoids and complement, resulting probably in the absorption of complement, 
is unknown. It is most likely in the nature of a globulin, its main char- 
acteristic being the power it possesses of reacting with lipoids. Schmidt1 
ascribes the reaction to the physicochemic properties of the globulins of 
the syphilitic serum, wdiich he believes possess a greater affinity for the 
colloids of the antigen than do normal globulins. This view is supported 
by the common observation that the turbidity of the antigen emulsion 
is closely related to its efficiency, since clear solutions are less active. 
THE ANTICOMPLEMENTARY ACTIVITY OF ANTIGENS AND SERA 
Practically all substances employed as antigens for the immunization 
of animals and the complement-fixation test, including the living micro- 
parasites producing disease and antibody production, are capable of ab- 
sorbing, fixing, or inactivating hemolytic complement in the test-tube 
independent of the presence or absence of specific complement-fixing anti- 
bodies. The degree of this complement absorption or fixation varies greatly 
with the kind and amount of different antigens, but is so important in con- 
nection with the complement-fixation test that this property, of the sub- 
stance being employed as antigen must be carefully determined by a process 
of titration. 
Under certain conditions all sera and other body fluids employed in 
the complement-fixation test and irrespective of whether or not they are 
normal or immune, may possess or acquire a similar property for absorbing, 
fixing, or inactivating complement by themselves and in the absence of 
antigen. This property is likewise so important in relation to the practice 
of the complement-fixation test that controls on this property of each 
serum being examined are always included (the serum controls). 
The presence of this property of antigens, sera, and other substances 
is detected by the interference of hemolysis when they are placed in test- 
tubes along with complement followed after a suitable period by the addition 
of corpuscles and specific hemolysin. Hemolysis is absent or incomplete 
because of the absorption, fixation, or destruction of all or a portion of the 
complement. Hence the process has been designated as the anticomple- 
mentary activity of these substances. 
The phenomenon may be divided into two entirely different processes, 
1 Ztschr. f. Hygiene, 1911, 69, 513. PRINCIPLES OF THE BORDET-GENGOU PHENOMENON 
434 
namely, (a) destruction or interference of activity of complement, and (b) 
absorption or fixation of complement. In the test-tube, however, this dis- 
tinction may not be detectable because the end-result of absent or incomplete 
hemolysis is the same for both. 
Destruction of complement or interference with its activity is readily 
understood and may be caused by such factors as changes in the hydrogen 
ion concentration of sera (acids) brought about by autolytic changes and 
the activity of ferments in the sera or furnished by contaminating bacteria. 
In bacterial antigens prepared with nutrient broth, the latter may contain 
various acids, alkalies, or protein split products which are known to be 
very potent in the inactivation and destruction of complement. In the 
alcoholic or salt solution extracts of tissues employed as “antigens” in the 
Wassermann test the anticomplementary activity may be caused by various 
lipoidal constituents, protein substances, or even the alcohol itself. Very 
probably the anticomplementary activity oj lipoids, carbohydrates, and un- 
organized substances is due to actual destruction oj complement or interference 
with its activity. 
Aside from the destruction or inactivation of complement due to changes 
in its active portion, it would appear that sera, suspensions of bacteria, 
and other cells and extracts of these as well as solutions of proteins of higher 
plants, are able to absorb or fix complement in the absence of homologous 
antibodies. 
These effects are sometimes designated as non-specific complement- 
fixation reactions, but I believe it is well to regard them as examples of 
anticomplementary action. As will be discussed in the succeeding section 
normal human, rabbit-, dog-, mule-, and other sera of the lower animals may 
not be anticomplementary in ordinary amounts, but when set up in com- 
plement-fixation tests with antigens composed of non-anticomplementary 
amounts of bacterial extracts or alcoholic extracts of tissues, may yield 
strongly positive complement-fixation reactions. I believe these are properly 
designated as examples of non-specific complement fixation. 
Kinds of Anticomplementary Substances (Antilysins).—As previously 
stated, many substances employed in complement-fixation tests may prove 
to be anticomplementary (also called “antilytic”) if they are employed in 
sufficiently large amounts; these antilysins may be found in: 
1. Suspensions and extracts of bacteria as shown by Wilde.1 All bac- 
terial antigens are anticomplementary, although they vary greatly in the 
amounts required to show these effects. Bacteria growing in serum render 
the latter highly anticomplementary, as shown by Craig2 and others. 
2. Extracts of tissues, as shown by von Dungern3 and tissue constituents, 
as shown by Levene and Baldwin.4 These include the watery, saline, and 
alcoholic extracts of tissues commonly employed as “antigens” in the Was- 
sermann test. 
3. Sera of various animals. Fresh sera are generally free of anticomple- 
mentary substances. Sterile sera may become anticomplementary and 
especially when kept at 20° to 38° C., but these substances are generally 
removed by heating to 55° C. for thirty minutes. Sera deeply tinged with 
hemoglobin are apt to become anticomplementary. Sera contaminated 
with bacteria become highly anticomplementary. Heating the sera of some 
animals to 60° C. and higher renders them anticomplementary and I shall 
refer to this subject shortly in more detail. 
Certain pathologic conditions may render serum anticomplementary, 
1 Berl. klin. Wchn., 1901, 38, 878. 
2 Jour. Exper. Med., 1911, 13, 521. 
3 Munch, med. Wchn., 1900, xlvii, 677. 
4 Jour. Med. Research, 1904, 12, 205. ANTICOMPLEMENTARY ACTIVITY OF ANTIGENS AND SERA 
notably uremia, as shown by Neisser and Doring,1 Neisser and Friedemann,2 
Bergmann and Keuthe,3 Laquer,4 and others. 
Absorption of sera by corpuscles to remove hemolysins or absorption 
by bacteria to remove agglutinins, opsonins, and bacteriolysins may render 
them anticomplementary; these antilysins are generally removed by heating 
the sera to 55° C. for thirty minutes. 
4. Precipitates may absorb complement and simulate the effects of anti- 
complementary sera as shown by Moreschi5 in a study of the anticomple- 
ments of sera. I will later refer to these in a discussion of the mechanism 
of action of antilysins. 
The Influence of Heat Upon Anticomplementary Substances (Thermo- 
labile and Thermostabile Antilysins); General Properties.—Human sera may 
prove anticomplementary when unheated, but lose this property completely 
or partly after heating in a water-bath to 56° C. for fifteen to thirty minutes; 
these antilysins are designated as thermolabile and to remove them is one 
reason why sera are heated routinely for the Wassermann and other com- 
plement-fixation tests. The antilysins may not be influenced by heating to 
56° C., and these are known as thermo stabile. According to Manniger6 horse- 
sera must be heated to 56° C. for thirty minutes and the sera of mules and 
asses to 63° to 64° C. for thirty to forty minutes to remove the thermolabile 
antilysins. 
Heating to 56° C. usually has very slight influence upon the anticom- 
plementary effects of bacterial suspensions, and extracts and tissue extracts 
employed as antigens in complement-fixation tests; these antilysins are 
usually of the thermostabile variety. 
Curiously, heating the sera of some animals is said to develop in them 
anticomplementary effects. Camus and Gley7 found this true of eel-sera 
heated to 56° C.; Ehrlich and Sachs8 observed that heating dog- and ox- 
serum to 60° C. had the same effects, and Sachs9 has described similar changes 
in rabbit-serum. Noguchi10 has observed that heating normal sera of the 
dog, sheep, and ox to 56° C. results in the development of anticomplementary 
activity. 
Anticomplementary substances are non-specific, that is, when present 
in a serum they are likely to interfere with hemolysis in different hemolytic 
systems. 
Noguchi has observed that absorption of dog-sera with corpuscles may 
remove the thermostabile antilysins, but Kyotoku11 was unable to accom- 
plish this in my laboratory working with anticomplementary human sera. 
Absorption with barium sulphate, kaolin, bone, ash, and charcoal removed 
them to a slight degree. Filtration of diluted human sera through Kitasato 
filters was found to remove the antilysins; Kyotoku noted but slight in- 
fluence of filtration upon the antibodies, but I have since observed that a 
large portion of the complement-fixing antibodies are likewise removed by 
this process. As far as I know there is no method for removal of thermo- 
stabile antilysins from sera wdiich does not remove completely or partly 
the antibodies at the same time. 
The Chemical Nature of Anticomplementary Substances.—The blood 
lipoids have been observed by many investigators to possess antihemolytic 
435 
1 Berl. klin. Wchn., 1901, 38, 593. 2 Berl. klin. Wchn., 1902, 39, 677. 
3 Ztschr. f. exper. Path, und Therapie, 1906, 3, 255. 
4 Deutsch. med. Wchn., 1901, 27, 744. 8 Berl. klin. Wchn., 1902, 39, 492. 
5 Berl. klin. Wchn., 1905, xlii, 1181. 9 Deutsch. med. Wchn., 1905, 31, 705. 
6 Ztschr. f. Immunitatsf., 1921, 31, 222. 10 Jour. Immunology, 1919, 4, 239. 
7 Compt. rend. Soc. de biol., 1901, liii, 732. 11 Jour. Immunology, 1919, 4, 239. 436 
PRINCIPLES OF THE BORDET-GENGOU PHENOMENON 
properties. Landsteiner and v. Eisler1 and Bang and Forssmann2 claimed 
to have secured them by ether extraction of corpuscles and serum; Noguchi3 
has likewise reported that ether extracts of blood-serum and corpuscles, 
freed from lecithin and certain related bodies, contains the thermostabile 
antilysin in concentrated but not pure form, which can be dissolved in 
saline solution and to which solution he has applied the name “protectin.” 
Zinsser and Johnson4 were unable to extract the thermolabile antilysins of 
human sera with ether, but found that they were removed by precipitation 
of the globulins. With normal rabbit- and dog-sera I have observed that 
a portion of the antilysins were removed by extraction with ether and chloro- 
form, but that both the globulin and albumin fractions were likewise con- 
cerned and especially the globulins.5 Kyotoku in his study with human 
sera, was.likewise unable to remove the antilysins with ether and found them 
closely allied with the protein constituents and especially the globulin 
fraction. Manniger has6 likewise identified the antilysins of the sera of 
horses, mules, and asses with the globulins of serum. 
It is well established that serum cholesterol is antihemolytic not only 
by effect upon the complement in serum hemolysis, but likewise upon the 
hemotoxins of bacteria, venom, saponin, etc. It is highly probable that 
the antilysins of human and other sera are identified with the protein frac- 
tions under certain circumstances and with the lipoids in others. They 
appear to be a group of substances capable of affecting complement. Their 
exact chemical nature is as yet unknown and the subject urgently requires 
further investigation. 
The Mechanism of Action of Anticomplementary Substances.—As first 
shown by Gay7 and Moreschi8 the anticomplementary activity of serum 
may not be caused by the development of antilytic substances, but be the 
result of complement absorption or fixation by the development of precipi- 
tates. For example, Sachs has showed that when sheep corpuscles are 
added to inactive rabbit-antisheep serum for the absorption of hemolysin 
that the serum may become anticomplementary. According to Gay this 
is caused by traces of sheep-serum adhering to the corpuscles producing 
precipitates with antisheep precipitin likewise present in the immune serum. 
Not infrequently the addition of washed sheep corpuscles to human serum 
for the absorption of natural antisheep hemolysin renders the serum anti- 
complementary. The anticomplementary activity is removable, however, 
by heating the serum to 55° C. for fifteen to thirty minutes. Whether these 
thermolabile anticomplementary effects are due to precipitates I am unable 
to state, but doubt that such is the case. 
The anticomplementary effects of suspensions or extracts of bacteria 
and tissue cells may likewise be caused by the development of invisible 
amounts of precipitate capable of absorbing complement due to the presence 
of natural precipitins in the complement serum acting with the bacterial 
or other cellular protein; or it may be that natural amboceptors for these 
bacterial or cellular substances are present in the complement serum, re- 
sulting in the specific fixation of complement and simulating the effects of 
an anticomplementary action. 
1 Wien. klin. Wchn., 1904, 27, 676; Centralbl. f. Bakteriol., 1905, 39, 309. 
2 Centralbl. f. Bakteriol., 1905, xl, 150. 
3 Jour. Exper. Med., 1906, 8, 726. 
4 Jour. Exper. Med., 1911, 13, 31. 
5 Jour. Infect. Dis., 1916, 18, 46. 
6 Ztschr. f. Immunitatsf., 1921, 31, 222. 
7 Centralbl. f. Bakteriol., 1905, 39, 603. 
8Berl. klin. Wchn., 1905, xlii, 1181. NON-SPECIFIC COMPLEMENT FIXATION 
437 
The lipoid antilysins are probably in large part composed of cholesterol 
and may inhibit hemolysis by directly neutralizing or destroying complement 
or, according to Noguchi, by increasing the resistance of corpuscles to he- 
molysis. 
In some instances at least the anticomplementary action of sera, bacterial 
suspensions and extracts, and alcoholic extracts of tissues may be due to 
physical conditions of a colloidal nature. Manniger believes that the serum 
globulins are responsible for anticomplementary effects by absorbing com- 
plement, and that the removal of this property by heating serum is due to 
the formation of compound of globulin and water, resulting in the production 
of an emulsoid colloid without ability to absorb complement. It may be 
that the anticomplementary effects of bacterial antigens are due to the 
absorption of complement by fine particles comparable to the antilytic 
effects of such inorganic emulsions as quartz, sand, and kaolin. 
Importance of Anticomplementary Action of Sera and Antigens in Rela- 
tion to Complement-fixation Tests and Reactions.—As previously stated, 
antilysins in serum or antigen employed in a complement-fixation test may 
interfere with hemolysis and thereby resemble a specific complement- 
fixation reaction. For this reason every substance employed as antigen 
must be titrated for its anticomplementary effects; the smallest amount 
that just begins to interfere with hemolysis is known as the anticomple- 
mentary unit, and in the main tests the antigen is used in a small fraction 
of this amount in order to avoid non-specific reactions. 
The anticomplementary activity of an antigen is determined by placing 
increasing amounts in a series of test-tubes with a fixed amount of comple- 
ment, and incubating the mixtures for the same length of time and at the 
same temperature as employed in the main tests. Hemolysin and corpuscles 
are now added to each tube and the mixtures reincubated. 
The occurrence and degree of hemolysis is then read and indicate the 
presence or absence of anticomplementary effects in the range of doses of 
antigen employed in the titration. The details of technic will be described 
in the succeeding chapters. 
Serum controls are included in every complement-fixation test to deter- 
mine whether or not the serum in the amount employed is free of anticom- 
plementary effects. If hemolysis is incomplete with a satisfactory hemo- 
lytic system, the inference is that the serum contained antilysins and the 
results of the complement-fixation test should be interpreted with great 
care and, better, discarded, with a repetition of the tests with fresh sera. 
By referring to the original Bordet experiment it will be observed that 
this investigator controlled any non-specific absorption of complement by 
both the immune and the normal serum in tubes C and D of the series by 
using the full dose of these serums with a similar amount of complement, 
and noting that hemolysis was complete. His controls, E and F, were to 
determine if the process of inactivation or removal of native complement 
from the two serums was complete, and the total absence of hemolysis 
showed that it was. His control on the anticomplementary action of the 
antigen was also included in tube D, for if the emulsion alone had absorbed 
complement to any degree, hemolysis would have been incomplete. 
NON-SPECIFIC COMPLEMENT FIXATION 
Of considerable interest and practical importance is the possibility of 
normal human sera and those of some of the lower animals of yielding non- 
specific complement-fixation reactions with various bacterial and tissue 438 
PRINCIPLES OF THE BORDET-GENGOU PHENOMENON 
antigens. Both sera and antigens are free of anticomplementary effects 
in these reactions in the amounts employed, and for this reason the phenome- 
non is to be separated from the effects of the antilysins previously discussed. 
Non-specific Complement Fixation by Normal Human Sera (Proteo- 
tropic Reactions).—When unheated human sera are employed in the Wasser- 
mann test with alcoholic extracts of tissues for “antigens,” positive reactions 
may occur with sera from non-syphilitic individuals. Seligman and Pinkus,1 
Noguchi,2 Browning and McKenzie,3 and others have drawn attention to 
this phenomenon, and Noguchi believes that it is caused by the presence 
of protein in the alcoholic extract of tissue being employed for “antigen.” 
For this reason he has designated the reaction as proteotropic complement 
fixation. Heated human sera do not yield these reactions. When unheated 
sera are employed in a complement-fixation test for syphilis, the tissue 
extract antigen should be free of protein, and for this reason Noguchi devised 
a method for preparing an extract of alcohol-soluble but acetone-insoluble 
lipoids which are to be preferred to plain or crude alcoholic extracts for 
tests of this kind, inasmuch as the latter usually contain protein. In my 
experience the use of properly prepared acetone-insoluble lipoids reduces 
the percentage of pseudopositive or proteotropic reactions to less than 2 
per cent., and this kind of “antigen” is to be preferred in all complement- 
fixation tests employing unheated serum;4 heating to 55° C. for fifteen 
minutes has been found sufficient for removing this property from human 
sera.5 
Non-specific Complement Fixation by Normal Rabbit-, Dog-, and Mule- 
sera.— The fresh normal sera oj several species of animals may absorb or fix 
complement with various lipoidal and bacterial antigens in a non-specific 
manner. Schilling and Hoesslin6 were probably among the first investi- 
gators to note this phenomenon with normal rabbit-serum. Manteufel 
and Woithe7 and Browning and McKenzie8 have also noted the phenomenon 
during studies in experimental trypanosomiasis. Dohi9 examined the sera 
of 74 normal rabbits, using as antigen an alcoholic extract of syphilitic 
liver, and found that 39 reacted positively and 35 negatively. He also 
observed that heated serum was more likely to show this non-specific ab- 
sorption of complement than unheated serum. Browning and McKenzie, 
however, state that they have not observed this phenomenon when using 
serum in a fresh, or active, condition. Blumenthal10 has likewise observed 
positive reactions with normal rabbit-serum, using an alcoholic extract 
of syphilitic liver as antigen; Craig and Nichols,11 on the other hand, have 
reported uniformly negative results with an alcoholic extract of syphilitic 
liver. Similar observations on this power of normal rabbit-serum to ab- 
sorb complement in the presence of lipoidal antigens have been made by 
Emmanuel12 and by Epstein and Pribram,13 the former stating that the 
administration of salvarsan removes this property temporarily, and the 
1 ZtscMr. f. Immunitatsf., 1910, 5, 377. 
2 Serum Diagnosis of Syphilis, Lippincott Co. 
3 Diagnosis and Treatment of Syphilis, Lea & Febiger, 1913, 20. 
4 Jour. Immunology, 1916, 2, 23. 
5 Amer. Jour. Syph., 1920, 4, 641. 
6 Deutsch. med. Wchn., 1908, 34, 1422. 
7 Centralb. f. Bakteriol., R., 1909, 43, 359. 
8 Jour. Path, and Bacteriol., 1909-11, 15, 182. 
9 Beitr. z. path. u. Therap. der Syph., 1911, 514. 
10 Berl. klin. Wchn., 1911, 48, 1462. 
11 Jour. Exper. Med., 1911, 14, 206. 
12 Berl. klin. Wchn., 1911, 48, 2335. 
13 Ztschr. f. exper. Path. u. Therap., 1909, 7, 549. QUANTITATIVE FACTORS IN COMPLEMENT-FIXATION TESTS 
439 
latter making similar claims for the mercurials. Casselman and myself1 
observed a large percentage of positive reactions with the sera of 117 normal 
rabbits and various lipoidal extracts used as antigens in the Wassermann 
reaction. In a further study Miss Trist and myself2 found that while heated 
rabbit-serum yielded positive reactions with 38 to 49 per cent, of sera with 
lipoidal antigens, active or unheated sera yielded but 5 to 15 per cent, posi- 
tive reactions. Bacterial antigens yielded even higher percentages of posi- 
tive reactions. Similar results have been reported by Kritchewsky,3 Hellens,4 
and Huddleson.5 Pearce and myself6 have also found that the sera of animals 
reacting positively or negatively generally continue to react in the same 
manner over long periods of observation. Rossi,7 Miss Trist and myself8 
have also found that the sera of a large percentage of normal dogs tends 
to yield non-specific fixation with various lipoidal and bacterial antigens. 
The mechanism of this phenomenon is not understood; it would appear 
that the complement-absorbing body is in both the serum lipoids and pro- 
teins.9 In my experience heating these sera at 56° C. for half an hour greatly 
increases their property for non-specific fixation of complement; heating 
at 62° C. for the same period lessens the tendency.10 The subject is of great 
importance in view of the frequency with which complement-fixation studies 
are conducted with rabbit-, dog-, and mule-sera. 
QUANTITATIVE FACTORS IN COMPLEMENT-FIXATION TESTS 
From what has been said it will, therefore, readily be appreciated that 
complement-fixation tests are largely quantitative. Equally fallacious 
results may be obtained by using too large or too small amounts of the 
various ingredients. 
Necessity for Using Optimum Amount of Antigen.—While it is possible 
to use too large quantities of antigen, so that non-specific absorption of 
complement occurs, leading to false positive reactions, it is also possible 
to use an amount so small that any specific absorption of complement by 
antigen and antibody cannot readily be detected. The proper amount to 
use must be determined by titration. 
The same is true, but to a much less extent, of the serum being tested, 
for while too large amounts of serum may lead to non-specific fixation of 
complement, surprisingly small amounts may give well-marked specific 
fixation, this factor depending, of course, upon the quantity of antibodies 
contained in the serum. 
Necessity for Proper Adjustment of the Hemolytic System.—Of even 
greater importance are the quantity of complement employed and the 
proper adjustment of the hemolytic system, composed of complement, hemol- 
ysin, and corpuscles. 
Too large an amount of complement may furnish sufficient to satisfy the 
complement-fixing antibodies of an immune serum united with the antigen, 
with enough free complement left over to produce partial or complete hemol- 
ysis when corpuscles and hemolysin are subsequently added. In this man- 
ner specific complement fixation would be overlooked and a false negative 
reaction secured. 
It is also possible to use too small an amount of complement, with rela- 
tively large doses of serum and antigen, so that the complement becomes 
1 Jour. Med. Research, 1913, 28, 369. 
2 Jour. Infect. Dis., 1916, 18, 20. 
3 Ztschr. f. Immunitatsf., 1914, 20, 238. 
4 Ztschr. f. Immunitatsf., 1913, 17, 156. 
8 Jour. Immunology, 1916, 2, 147. 
6 Jour. Infect. Dis., 1916, 18, 32. 
7 Ztschr. f. Immunitatsf., R., 1909, 1, 429. 
8 Jour. Infect. Dis., 1916, 18, 27. 
9 Jour. Infect. Dis., 1916, 18, 46. 
10 Jour. Infect. Dis., 1916, 18, 64. 440 
PRINCIPLES OF THE BORDET-GENGOU PHENOMENON 
unduly susceptible to non-specific fixation, and consequently false positive 
reactions may be secured. 
It has previously been explained that an excess of hemolysin may offset 
any slight deficiency in the amount of complement. For instance, if a 
small amount of complement is specifically fixed by an antigen and its 
amboceptor, the addition of too large an amount of hemolysin may result 
in complete hemolysis of the corpuscles, and thus overshadow the slight 
but specific fixation of complement. 
On the other hand, hemolysis cannot be complete if the dose of hemol- 
ysin is too small. With a given dose of corpuscles and complement a cer- 
tain amount of hemolysin is necessary to produce hemolysis, this dose being 
determined by a process of titration, as described in a previous chapter. 
If less than this dose is used, but the amounts of corpuscles and comple- 
ment remain the same, hemolysis will be correspondingly incomplete and 
lead to false positive reactions. 
A very important feature of all complement-fixation tests will be seen 
to be a proper and accurate adjustment of the hemolytic system. Taking 
arbitrary amounts of corpuscles and hemolysin as constants, the quantity 
of complement necessary to produce hemolysis may be determined (titra- 
tion of complement); or, taking corpuscles and complement as constants, 
the amount of hemolysin necessary to effect complete hemolysis may be 
determined. One or the other or both titrations should be made before 
the main test is attempted, in order to avoid using an excess or too little of 
either ingredient. If the exact unit of complement and hemolysin are used, 
the results must be very carefully guarded, because in a general way all 
antigens and serums exert a slight anticomplementarv action that may 
yield results that will be interpreted as weak positive reactions. For this 
reason the original complement-fixation tests invariably called for a slight 
excess of complement or hemolysin, or both, to allow for possible non-specific 
complement fixation, and this is a good general rule that makes the reaction 
somewhat less delicate, but more reliable in the long run, especially for 
inexperienced workers. 
Complements of different species of animals act differently in activating 
a hemolytic amboceptor and toward fixation by antigen-antibody combi- 
nations. For instance, a complement from one animal may readily enough 
combine with a hemolytic amboceptor to produce hemolysis, but will not 
lend itself for fixation, and is, therefore, unfit for complement-fixation tests. 
Noguchi and Bronfenbrenner have found guinea-pig-serum most suitable 
from all standpoints, but it is important to remember that the comple- 
mentary activity of the serums from different guinea-pigs varies, and, 
therefore, it is necessary to titrate each complement serum or hemolysin, 
i. e., adjust the hemolytic system, before the main test is conducted. 
These quantitative factors are of great importance, and complicate 
any complement-fixation method, but efforts to circumvent or ignore them 
are likely to lead to errors in technic. A proper understanding and appre- 
ciation of these factors constitutes the basis for reliable work, whereas 
less essential details may be altered to conform to the ideas and convenience 
of the individual worker. 
PRACTICAL APPLICATIONS 
It will be understood, therefore, that complement-fixation reactions may 
serve two primary purposes: 
1. With a known antigen, the antibody may be found. This is the PRACTICAL APPLICATIONS 
441 
usual order in diagnostic tests. For example, in the reaction for syphilis 
the antigen is furnished and the antibody sought for in the body fluids. 
So specific has this test proved in the diagnosis of this disease that a posi- 
tive reaction secured with a proper technic is regarded as strong evidence 
of the existence of lues, even though the primary lesion had occurred years 
before and the person is at the time in apparent good health. In the gono- 
coccus fixation test and other tests of a similar nature the antigen is known 
and is furnished, and the antibody is tested for in the serum. 
2. With a known antibody the corresponding antigen may be found. 
This order of events has less practical application, and is used principally 
in the diagnosis of blood-stains and in the differentiation of proteins in 
general. It is also used in making special bacteriologic investigations, when 
an organism may be identified by specific complement fixation with its 
known antibody serum. In these instances the antibody serum is secured 
by immunizing rabbits with a known antigen, the immune serum then 
being used for selecting the antigen in unknown substances and mixtures. 
Complement-fixation methods have their greatest value, and are prob- 
ably best known, in the serum diagnosis of syphilis—the biologic syphilitic 
reaction of Wassermann, Neisser and Bruck, and Detre. Although originally 
believed to be a direct application of the specific Bordet-Gengou phenom- 
enon of complement fixation, subsequent investigations have shown that 
the antigen need not be specific, in the sense of containing the Spirocheta 
pallida, but that lipoidal substances in general may serve as “antigen,” 
the peculiar and specific character of the reaction depending upon the nature 
of the antibody, which has a strong affinity for lipoids, and in such a mixture 
is capable of absorbing or fixing a considerable amount of complement. 
In no other disease has the method been so widely employed as in syphilis, 
although it possesses value in the serum diagnosis of various bacterial in- 
fections, such as gonorrhea, glanders, typhoid fever, echinococcus disease, 
etc., ana in the diagnosis of blood-stains and in the differentiation of proteins 
in general. 
In the following chapter the Wassermann syphilitic reaction will be con- 
sidered in some detail, as a thorough working knowledge of this test is of 
great value, and serves as the foundation of complement-fixation technic in 
general. CHAPTER XXIII 
COMPLEMENT FIXATION IN SYPHILIS 
THE TECHNIC AND PRACTICAL VALUE OF THE WASSERMANN AND OTHER 
COMPLEMENT-FIXATION TESTS 
The Phrase “Wassermann Test.”—Since the application of the Bordet- 
Gengou phenomenon to the diagnosis of syphilis by Wassermann, Neisser 
and Bruck and Detre, many advances have been made in our knowledge 
of the properties of the several biologic reagents employed and numerous 
modifications of the original test have been proposed. The phrase “Wasser- 
mann test” is used by most persons as a short term for the phrase “comple- 
ment-fixation test for syphilis,” but this is an error and the term should be 
limited to the test conducted after the method of Wassermann. As far as 
I can ascertain, Wassermann has made no important changes in his technic 
as described in his early papers and still uses salt solution extracts of syph- 
ilitic liver for antigen, although he cannot have failed to note the success 
of alcoholic extracts of normal tissues and, indeed, the general superiority 
of these extracts; apparently he continues to use the same hemolytic system 
and the same general technic described in 1906 and 1907. The term “Wasser- 
mann reaction,” therefore, should, strictly speaking, be confined to a test 
conducted with Wassermann’s original technic, but practically the latitude 
may be broader and embrace those modifications emploveing alcoholic 
extracts of luetic and non-luetic tissues as antigens, and such changes as 
using each constituent of the test in one-half, one-quarter, or one-tenth the 
amounts of the original Wassermann test, inasmuch as these changes do 
not alter the principles and are used mainly for economy. 
However, all tests employing a different method for adjusting the hemo- 
lytic system, a method for titrating complement, a different hemolytic 
system, the use of active instead of heated serum, the utilization of natural 
complements and amboceptors in patients’ sera, etc., cannot claim the 
designation “Wassermann test,” being based upon principles and technic 
too remote from Wassermann’s technic; modifications of this nature are 
now quite numerous and are usually designated as modifications of the 
Wassermann test. 
The Demand for Standardization of Technic.—Probably no laboratory 
test has been subject to as much favorable and unfavorable criticism and 
to as many modifications in technic as Wassermann’s serum test for syphilis; 
owing to the fact that the reaction is not strictly or biologically specific 
for syphilis and requires several biologic reagents of varying properties, 
it is readily subject to error in both a positive and negative way unless 
carefully and intelligently understood and conducted. 
Many of the . proposed modifications of the original technic have been 
demanded by an increasing experience and knowledge of the reaction, and 
several have undoubtedly served to improve its delicacy and value, but the 
employment of different methods by various persons has led to wide varia- 
tion in the results, confusion in regard to the actual value of the test, and 
insistent demand for one standard method. 
Within recent years several publications have drawn particular atten- 
tion to discrepancies in the results of Wassermann reactions with the same 
442 WASSERMANN AND OTHER COMPLEMENT-FIXATION TESTS 
443 
serum in different laboratories and, indeed, to varying results reported by 
the same serologist on specimens of blood from the same individual with- 
drawn at the same time and under identical conditions. For example, 
Uhle and MacKinney,1 who submitted specimens of blood from each of 
292 individuals to from four to ten laboratories, found that “there is one 
chance in five that the tests will agree”; Phelps2 in a similar analysis of the 
results of Wassermann tests with 358 specimens of blood divided and sent 
to from two to four laboratories, found that the results agreed in but from 
52 to 65 per cent, and Wolbarst,3 in a study of 37 cases by two or more serol- 
ogists, found that the results agreed in 70 per cent, and disagreed either 
slightly or absolutely in 30 per cent. One of the most thorough studies of 
this kind has been more recently made by Palmer4 who submitted 75 serums 
to eight different laboratories with considerable variation in the reports. 
Palmer is of the opinion that a safe and efficient standardized technic should 
be adopted and required of all laboratories and that a higher degree of 
efficiency should be demanded of technicians. 
I believe all serologists of experience agree that a certain percentage of 
discrepancies are to be expected with the use of different hemolytic systems 
and reagents and particularly different antigens, even granting that all 
tests were conducted with the requisite technical care and attention to 
details. All of the above-mentioned authors in addition to Pusey,5 Shrop- 
shire,6 and others have emphasized the need of a standardized technic; 
Brem,7 Faller,8 Schlesinger,9 and numerous others have directed attention 
to sources of error in the technic. 
The Author’s Investigations on Standardization of Technic.—Possibly the 
easiest way of standardizing the Wassermann test would be to adopt the 
original technic with a change in the antigen. But such a proposal would 
receive little recognition, not only because the original technic is defective 
in sensitiveness but also because the antigen is the greatest single source 
of contention and disagreement. 
In my opinion “standardization” means a real, earnest, and unbiased 
study of different methods and of each and every phase of the complement- 
fixation test, for the purpose of determining by actual trial what is best 
and incorporating the facts into a test which will have for its purposes the 
establishment of a technic (1) of superlative sensitiveness; (2) of practical 
specificity; (3) of technical accuracy and uniformity of results; (4) yielding 
a true quantitative reaction; (5) as simple, and (6) as economical as possible. 
Realizing my complex nature of the Wassermann reaction and the 
variable properties of its several biologic reagents, our almost complete 
ignorance of its mechanism and the absence of a specific and wholly satis- 
factory antigen, the task of even attempting standardization was considered 
a serious and laborious problem; knowing that the majority of serologists 
had an individual way of conducting certain steps in the technic and par- 
ticularly that many had learned from experience to rely so firmly upon 
their own method as to be very loath to accept any other, the hope of build- 
ing up a widely acceptable technic would appear almost hopeless unless an 
1 Jour. Amer. Med. Assoc., 1915, lxv, 863. 
2 Boston Med. and Surg. Jour., 1915, clxxiii, 391. 
3 New York Med. Jour., 1913, xcv, 378. 
4 Jour. Amer. Med. Assoc., 1922, 79, 724. 
5 Jour. Amer. Med. Assoc., 1913, lxi, 1920. 
6 Southern Med. Jour., 1916, ix, 205. 
7 Californ. State Jour. Med., 1912, x, 362. 
8 Lancet-Clinic, 1914, cxii, 536. 
9 Med. Rec., 1915, lxxxvii, 61. 444 
COMPLEMENT FIXATION IN SYPHILIS 
unexpected discovery bestowed upon the standard technic an indisputable 
quality of excellence. The method pursued in this investigation, which 
has covered a period of five years,1 was to become acquainted with all exist- 
ing methods by thoroughly reviewing the available literature, and by means 
of personal communications and interviews with a large group of serol- 
ogists and submitting the whole to careful unbiased experiment and choosing 
that proving best on the basis of actual trial and receiving endorsement 
on the basis of experience. 
This work has resulted in the building up of a new test2 embracing a 
new antigen3 and many technical improvements. The antigen and technic 
of the test are given in this chapter, but owing to lack of space detailed 
accounts of the work conducted cannot be given. These results as well 
as an account of investigations in complement fixation in bacterial infec- 
tions and for the differentiation of proteins, will be presented by the writer 
in a separate monograph. 
General Technic 
In the preparation of glassware and biologic reagents employed in the 
Wassermann and other complement-fixation tests for syphilis, the general 
technic is the same and may be described collectively; the details covering 
amounts to employ, technic of titration, etc., are given in the descriptions 
of the various tests. 
The Glassware.—Test-tubes should be of convenient sizes, perfectly clean, 
free from acids and alkalies, and preferably sterile. They need not be plugged 
with cotton, as it suffices to sterilize them in a wire basket with their mouth- 
ends downward. A method for cleaning test-tubes and other glassware 
is described on page 7. 
Pipets should be perfectly clean and preferably sterile. Three kinds 
are required: The ordinary 1 c.c. pipet, graduated to 0.01 c.c. and cali- 
brated to the tip; 5 and 10 c.c. pipets divided into 0.1 c.c. Care should be 
exercised in handling pipets to avoid breaking the tips. After use they 
should be washed free from blood, serum, etc. The handling and care of 
pipets are described on page 4. It is well to have a separate 1 c.c. pipet 
for each serum to be tested. If one pipet is employed for all sera, due care 
must be exercised in washing it out very thoroughly. 
The technic for preparing saline solution is described on page 9. 
The Serum and Cerebrospinal Fluid.—(a) Serum.—As a general rule 
all specimens of blood submitted for complement-fixation tests should be 
collected aseptically in sterile containers. This is especially necessary when 
there has been delay in transmitting the fluid to the laboratory, is when 
sent through the mails from distant points. When the reactions are to be 
conducted on the same or on the following day, the specimen of blood may 
be collected in chemically clean but not necessarily sterile containers. Bac- 
terial contamination renders a fluid anticomplementary and unfit for com- 
plement-fixation tests. Specimens should be kept in a cold place until 
used. 
Collecting Blood for Complement-fixation Tests.—In collecting blood 
for the Wassermann reaction the following points should be remembered: 
1. That during active antisyphilitic treatment the blood may react 
negatively, wdiereas at a later period a true positive reaction is observed. 
1 See series of thirty-two papers published in the Amer. Jour, of Syph. commencing 1919, 
iii, 1 and ending 1922, vi, 680. 
2 Amer. Jour. Syph., 1922, vi, 82. 
3 Amer. Jour. Syph., 1922, vi, 74. GENERAL TECHNIC 
445 
It is well, therefore, not to collect blood until all specific treatment has been 
suspended for at least two weeks. 
2. That blood collected during or immediately after an alcoholic de- 
bauch may yield a false negative reaction (Craig and Nichols). 
3. That blood should not be collected just after anesthesia or while 
the patient has a high temperature. 
As a general rule, at least 1 c.c. of serum and 2 c.c. of cerebrospinal fluid 
are required for making the syphilitic reaction. From 2 to 3 c.c. of blood 
are needed, these amounts being easily collected from adult persons by 
pricking the finger deeply and filling a small test-tube or vial, as shown on 
p. 15. This method is very convenient, especially for physicians, hospitals, 
and dispensaries where direct access to a laboratory can be had. When 
the treatment is to be guided by the Wassermann reaction, a number of 
tests are required, and patients may object to repeated venipuncture, 
whereas no objections will be raised to simple puncture of the finger. 
Larger amounts of blood are collected from 
a vein at the elbow under aseptic precautions, 
as described on p. 18. As a rule, it is well to 
collect at least 5 c.c. of blood, especially if the 
specimen is shipped from a distant point (Fig. 
129). An excess of serum permits the technician 
to repeat a test when necessary, or to apply 
more than one method, and thus at times both 
the physician and the patient are saved the time 
and annoyance incident to collecting another 
specimen. 
The Keidel tube, wdiich is sterilized and 
ready for use, is quite a convenience (p. 20). 
However, a’ test-tube or a centrifuge tube may 
be used, or, when a specimen is to be mailed, a 
5 or 10 c.c. vial of thick glass, stoppered with a 
cork or a rubber stopper, is quite satisfactory 
(Fig. 108). Vial, stopper, and needle are readily 
sterilized in boiling water, drained, and cooled, 
the specimen collected, the vial tightly stoppered, 
and the whole sent at once to the laboratory. 
Cotton stoppers are unsatisfactory, as unless 
the tube or vial is maintained in an upright 
position, the fluid may be absorbed and ren- 
dered anticomplementary. When specimens of 
blood are to be mailed, it is better to fill a small 
vial than to place the same amount in a large 
container, for in the latter case agitation through 
handling may result in so much mechanical hemolysis taking place as to 
render the serum unsatisfactory for use. Specimens so collected may be 
sent for long distances, even in warm weather, and undergo no change. 
In collecting blood from children, or where the veins are small, a pro- 
portionately smaller needle may be used. In infants the cupping apparatus 
of Blackfan is quite satisfactory (p. 22); frequently sufficient blood may be 
obtained from a large toe. 
Preparing the Serum.—The specimen of blood should be kept in a cold 
place, and the serum removed at the end of twenty-four hours. Serum 
that is allowed to remain with the clot for longer periods is more likely to 
become anticomplementary, especially if it becomes deeply tinged with 
Fig. 129.—A Vial to Contain 
Blood for the Wassermann 
Reaction. 
This is an ordinary glass vial 
fitted with a rubber stopper. It 
holds 5 c.c. to the mark, and is 
readily packed for mailing. Never 
stopper with cotton. A good cork 
stopper may be used. 446 
COMPLEMENT FIXATION IN SYPHILIS 
hemoglobin. In cases where the serum does not separate, the clot may be 
broken up gently with a sterile glass rod and centrifugalized. The serum 
should be clear and free from corpuscles. Opalescent and milky serums, 
obtained during the period of digestion and from nursing women, usually 
do not interfere with the reaction; bile-stained serum may at times give 
marked non-specific fixation of complement. 
Heating the Serum.—In the original Wassermann test the serum was 
heated to 56° C. for thirty minutes and in many modifications of the test 
this practice has been adopted. Some complement-fixation tests, however, 
employ unheated serum in order to increase the sensitiveness of the reac- 
tions by conserving all the syphilis “reagin.” 
Serum is heated for these purposes: (a) To remove thermolabile anti- 
complementary substances; (b) to remove the possibility of proteotropic 
or non-specific reactions, and (c) to inactivate the native or natural comple- 
ments. Heating serum to 56° C. for thirty minutes results in the destruction 
of a portion of the syphilis “reagin”; unless a serum is strongly anticomple- 
mentary, I believe that heating to 55° C. for fifteen minutes is sufficient and 
results in much less destruction of “reagin.”1 
In order to conduct a reliable test it is usually necessary to secure fresh 
serum. This anticomplementary action of serums is so important that in every 
complement-fixation test there is a serum control tube containing all the in- 
gredients except antigen, the object being to discover any inhibitory action of 
the serum itself upon the complement. 
Wechselmann’s method of converting syphilitic serums showing a nega- 
tive or weakly positive reaction to those exhibiting a marked positive reac- 
tion depends upon removing the excess of indifferent and inhibiting serum 
components and upon the destruction or diminution of the natural anti- 
sheep amboceptor present in so large a percentage of human serums. As 
Noguchi and Bronfenbrenner have pointed out, this method may likewise 
remove the antibodies concerned in the reaction. To 1 c.c. of heated serum 
add 3.5 c.c. of saline solution and 0.5 c.c. of a 7 per cent, suspension of 
freshly precipitated barium sulphate; shake, and let it stand for one hour 
at 37° C.; centrifugalize, and pipet off the diluted serum, which is now ready 
to be tested (1 c.c. = 0 2 c.c. of undiluted serum). 
Cadaver serums are likely to be highly discolored with hemoglobin and 
quite anticomplementary. Such serums may be tested in half the usual 
dose, and while the results are quite specific, they are not so reliable or 
constant as those obtained from the living. 
The doses of serum used in testing for the syphilis reaction are given 
with each method. In the original Wassermann test 0.2 c.c. was used. As 
a rule, from 0.05 to 0.2 c.c. of serum are satisfactory; higher doses may 
occasionally show a stronger positive reaction, but the serum must be 
perfectly fresh to avoid non-specific complement fixation. 
(.b) Cerebrospinal Fluid.—In certain nervous diseases the cerebrospinal 
fluid is examined for the syphilis reaction. Fluid is secured by lumbar 
puncture, according to the method described on p. 25. If the specimen 
contains blood, it should be centrifuged until it is clear. It should not be 
heated before use, as it does not contain hemolytic complement, and fresh 
fluids from cases other than syphilitics do not react positively. Cerebro- 
spinal fluids, as a rule, possess weaker fixing powers than the corresponding 
blood-serum, and hence it is necessary to use larger doses—at least 0.5 
to 1 c.c., instead of 0.05 to 0.2 c.c.—as in the case of blood-serum. 
(c) Other Fluids.—Positive syphilitic reactions have been described as 
1 Amer. Jour. Syph., 1920, 4, 641. COMPLEMENT 
447 
occurring with milk, pleural and peritoneal exudates, and albuminous urine 
(Bauer and Hirsh) from luetic cases. The material should be perfectly 
fresh, as anticomplementary action is likely to occur and all require titration 
for this property in order to avoid non-specific reactions. 
Complement 
While complement is to be found in the fresh normal serum of practically 
all warm-blooded animals, not all are suitable for complement-fixation tests. 
A suitable complement must possess two important properties: (1) Hemo- 
lytic activity, or the power of activating a hemolytic amboceptor, and (2) 
fixability, or the power of being “fixed” by antigen and antibody. Noguchi 
and Bronfenbrenner1 have studied the complements of the dog, sheep, hog, 
ox, rabbit, and other animals, and found that the complement of the guinea- 
pig was best adapted, from all standpoints, for the complement-fixation 
test. The complements of pigs and sheep are quite fixable, but their weak 
hemolytic action and rapid deterioration render them unsuitable for fixa- 
tion purposes. Rabbit complement is quite active, but is not easily fixable. 
Similar results have been observed by Matsunami, Trist, and myself.2 
Kolmer, Yui, and Tyau3 found rat complement fairly well suited for making 
the syphilitic reaction with an antihuman hemolytic system. 
The hemolytic power of guinea-pig complement is not constant. In 
unhealthy animals it is likely to be low, and even among normal animals 
it may show some variation. For this reason the hemolytic power of each 
serum is determined by a method of titration before complement-fixation 
reactions are conducted. Fixed doses of hemolysin and corpuscles may 
be used, and the amount of complement necessary for effecting complete 
hemolysis may be determined, or a fixed dose of complement and corpuscles 
may be used with different amounts of hemolysin, the chief object being 
to adjust all three factors of the hemolytic system, namely, complement, 
corpuscles, and hemolysin, to exact and known proportions. 
The complement in the serums of different guinea-pigs may show con- 
siderable variation in fixability. The amount of complement inhibited 
by serum alone and organic extract alone, or by given constant quantities 
of serum and extract, may vary more markedly than their complementary 
activity. To reduce this error to a minimum it is advisable, whenever 
possible, to use the pooled serums of two or more pigs for making comple- 
ment-fixation tests. 
Preparation of Guinea-pig Complement Serum.—The animal is bled under 
ether anesthesia into a Petri dish or centrifuge tube. The large vessels on 
both sides of the neck are quickly severed with a pair of sharp-pointed 
scissors or a scalpel, care being exercised not to sever the trachea and esoph- 
agus. A funnel is used for collecting blood in centrifuge tubes. It is well 
finally to sever the spinal cord in order that the animal may not recover 
from the anesthetic and thus insure a painless operation throughout (see 
p. 37). 
The following technic is employed by the author4: 
1. Select large well-nourished guinea-pigs which have not been fed 
within twelve to twenty-four hours of the time for bleeding; avoid pregnant 
animals. 
2. Collect blood in a chemically clean and preferably sterile centrifuge 
tube by means of a clean funnel about 4 inches in diameter, or in a Petri 
1 Jour. Exper. Med., 1911, 13, 78. 
2 Amer. Jour. Syph., 1919, 3, 407. 
3 Jour. Med. Research, 1913, xxviii, 483. 
4Amer. Jour. Syph., 1919, 3, 407. 448 
COMPLEMENT FIXATION IN SYPHILIS 
dish. We prefer the former because if separation of serum is unsatisfactory 
the clot may be gently broken up with a glass rod and the serum secured 
by centrifuging the material. Chemical cleanliness is essential because 
traces of acids or alkalies are destructive of complement. 
3. If guinea-pigs are to be bled to death, stunning the animals with a 
blow at the base of the skull is preferable to ether and chloroform anesthesia 
because cardiac activity is not interfered with and the maximum amount 
of blood is obtained. Guinea-pigs may also be bled from the ears without 
anesthesia, or from the heart (p. 37) or an external jugular vein under 
light ether anesthesia. 
4. In bleeding a guinea-pig to death it is not necessary to shave the 
neck; a few hairs in the blood do not alter the properties of the serum. 
The great vessels on one or both sides should be quickly severed and pref- 
erably with a pair of stout sharp scissors with one pointed blade to facilitate 
piercing of the skin and underlying tissues. It is easy to avoid cutting the 
trachea and esophagus, but it does not appear to make any difference whether 
the trachea is cut; the esophagus should not be cut, because in squeezing 
the abdomen to pump out blood in the great vessels stomach contents 
may escape, although even this accident is rare in our experience. 
5. In bleeding from the heart a chemically clean and sterilized 5 to 10 
c.c. all glass or glass-metal syringe fitted with a short needle of about gage 
20 may be used; in bleeding from an external jugular vein the hairs are 
plucked just above the region of the clavicle and a small incision made 
through the skin and vein, with collection of blood by means of a funnel 
into centrifuge tubes. As soon as the animal is released bleeding ceases and 
the wound heals promptly and without infection. 
6. Do not use the blood immediately after bleeding; if the animals are 
bled on the day the serum is to be used, place the clots in the incubator at 
37° C. for an hour, followed by breaking up each with a glass rod and cen- 
trifuging, or let the clots stand at room temperature for two or three hours 
for separation of serum which if not complete may be finished by centri- 
fuging. Animals may be bled late in the afternoon of the day before and the 
clots placed in the incubator for one hour or left at room temperature for 
two hours and then placed in a refrigerator until the following day. If the 
sera have separated poorly, gently break up the clots and centrifuge. It 
is preferable to leave the serum on the clot in the refrigerator until ready for use; 
unused serum .should be returned to the refrigerator until required. Traces of 
hemoglobin in the serum do not interfere, but the serum should be free of 
corpuscles. 
Moore1 and, more recently, Ecker2 have described guinea-pigs naturally 
deficient in complement. According to Moore, this property is inheritable 
and apparently is not due to anything interfering with the action of hemol- 
ysin. 
As a general ride, therefore, it is good practice to bleed the animal late in 
the afternoon preceding the day on which the experiment is to be made, or at 
least some hours before the regular work of the day begins. The serum should 
be clear and contain no corpuscles. 
Preservation of Complement.—Various methods have been advocated 
from time to time for the preservation of complement serum. These have 
been subjected to a comparative study by Kolmer, Matsunami, and Trist,3 
and the following are to be recommended: 
1. In a Frozen State.—Serum is placed in ampules in amounts of 1 c.c., 
1 Jour. Immunology, 1914, 4, 425. 
2 Jour. Infect. Dis., 1921, 29, 611. 
3 Amer. Jour. Syph., 1919, 3,513. COMPLEMENT 
449 
sealed, and kept in a refrigerator at or near 0° C. or in mixtures of ice and 
salt. Vacuum bottles have been advocated. A home-made “Frigo” is 
easily constructed as described on page 56. 
2. With Sodium Chlorid (Author’s Method).—To each cubic centimeter 
of serum add 0.2 gm. chemically pure sodium chlorid, dissolve, and keep 
at a low temperature. When to be used dilute each cubic centimeter with 
19 c.c. of distilled water which gives a 1 : 20 dilution in 1 per cent, saline. 
This method is particularly useful when complement is employed in a 1 : 20 
dilution. 
3. With Sodium Chlorid After the Method of Thompson}—An 8.1 per 
cent, solution of pure sodium chlorid in distilled water is prepared and auto- 
claved; equal parts of complement serum and this salt solution are pre- 
pared, placed in ampules, and kept at a low temperature. When required, 
1 c.c. of the mixture is diluted with 4 c.c. of distilled water which gives a 
1 : 10 dilution of complement in isotonic solution. Dilutions of 1 : 20 or 
1 : 30 may be made from this 1 : 10 solution. 
4. With Sodium Chlorid After the Method of Austin?—A 25 per cent, 
solution of pure sodium chlorid in distilled water is prepared and sterilized; 
to 1 part of complement serum is added 1| parts of this solution and the 
mixture stored in ampules at a low temperature. When required 1 c.c. 
is diluted with 3 c.c. of 0.5 per cent, salt solution which gives a 1 : 10 dilu- 
tion of complement in an isotonic solution. 
5. With Sodium Chlorid After the Method of Neill?—A saturated solution 
(approximately 36 per cent.) of pure sodium chlorid is prepared in hot 
distilled water, cooled, filtered, and sterilized; to each cubic centimeter of 
complement serum is added 0.1 c.c. of this salt solution and the mixture 
kept at a low temperature in ampules. When required 1 c.c. is diluted 
with 3 c.c. of distilled water and further diluted with 6 c.c. of physiologic 
saline solution which gives a 1 : 10 dilution. 
6. With Sodium Acetate After the Method of Rhamy?—A 10 per cent, 
solution of pure sodium acetate in 0.9 per cent, solution of sodium chlorid 
is prepared and sterilized; to 1 part of complement serum is added 1| parts 
of acetate solution, and the mixture kept at a low temperature in ampules. 
When required 1 c.c. is diluted with 3 c.c. of saline solution which yields 
a 1 : 10 dilution. Higher dilutions may be prepared as desired. 
Complement serum preserved by these methods will usually maintain 
its hemolytic and fixing properties for several weeks; loss in fixability or 
delicacy in the complement-fixation test is usually apparent before an appreci- 
able deterioration in hemolytic activity. 
The Amount of Complement to Employ.—And now we come to a very 
important question, namely, the amount of complement that is to be used 
in conducting complement-fixation tests. The hemolytic system should be 
so adjusted that the amount of complement employed is neither unneces- 
sarily large, which tends to falsely negative reactions in the presence of 
small amounts of antibody, nor too small, which does not allow for the anti- 
complementary activities of serum alone and antigen alone and for comple- 
ment destruction during the primary incubation. For practical purposes 
in routine complement-fixation tests when the complement is titrated plain 
(that is, in the absence of antigen and normal serum) it is necessary to use 
at least 2 units of complement and 2 units of hemolysin to allow for anti- 
1 Jour. Amer. Med. Assoc., 1916, lxvi, 652. 
2 Jour. Amer. Med. Assoc., 1913, lxii, 868. 
3 Public Health Reports, August 23, 1918, 1387. 
4 Amer. Med. Assoc., 1917, lxix, 973. 450 
COMPLEMENT FIXATION IN SYPHILIS 
complementary action of serum and antigen and destruction of complement, 
as shown in the comparative studies of Kolmer, Matsunami, and Rule.1 
Some present-day observers use exactly one unit of complement and 
one unit of hemolysin. This is permissible, providing the complement 
is titrated in the presence of a constant dose of antigen and a constant 
dose of serum, in order that due allowance for the anticomplementary 
action of these may be made in the titration. Under these circumstances, 
however, it is necessary to titrate each patient’s serum with the comple- 
ment, because one serum or even the pooled serums of different persons 
should not be taken as a standard in the titration, for two important reasons: 
(1) The patient’s serum which we are about to test may be more anti- 
complementary than the serum used in the complement titration, and 
hence when used in the main test, with exactly one unit of complement, 
mild degrees of inhibition of hemolysis will be secured that may be inter- 
preted as slightly positive reactions; or (2) the serum used in the comple- 
ment titration may contain more or less natural hemolytic amboceptor 
than the patient’s serum, and this factor exerts an influence on the titration, 
so that the unit varies with different serums. For these reasons I have 
included here a fourth method for using a single unit of complement under 
conditions where the anticomplementary action of each serum and the antigen 
are determined, in preference to titrating the complement with a serum 
and using this unit for a number of other serums that are sure to differ 
from each other. 
In complement titration, therefore, we determine the amount or unit of 
complement necessary to produce hemolysis with fixed amounts of hemol- 
ysin and corpuscles. It will be observed that I use two units of hemol- 
ysin, as determined by a previous titration. These two units are equivalent 
to one dose, and the same would be true whether three, four, five, or more 
units were used, because in this titration the corpuscles and hemolysin 
are arbitrary and fixed constants, and are used for determining the amount 
of complement necessary to bring about complete hemolysis. 
In conducting the main tests the dose of corpuscles and that of hemol- 
ysin are the same as those used in the complement titration, but instead 
of using exactly one unit of complement it is necessary to use two units, 
as stated above, to allow for the anticomplementary action of antigen and 
patient’s serum. 
It is important to remember that, in conducting this titration, hemol- 
ysin, complement, and corpuscles are to be mixed at once; if hemolysin 
and corpuscles are mixed and allowed to stand for ten minutes or more 
before receiving the complement the corpuscles become “sensitized,” and 
the amount of complement required for effecting hemolysis will be less' than 
if all three are mixed one after another. If this rule is not adhered to an 
error in technic may result. 
Since the original work of Wassermann appeared the antisheep hemo- 
lytic system has been most widely used in complement-fixation tests. 
Antisheep hemolysin is readily prepared by immunizing rabbits with 
washed sheep’s corpuscles. A simple and efficient method is to give intra- 
venous injections of five doses of 5 c.c. each of a 10 per cent, suspension 
every three or four days. Other methods and the details of the preparation 
and preservation of hemolysin are given in previous chapters. 
Objections have been made to the use of the antisheep hemolytic system 
Hemolysin 
1 Amer. Jour. Syph., 1920, 4, 518. RED BLOOD-CORPUSCLES 
451 
because of the presence, in a large proportion of human serums, of variable 
amounts of natural hemolysin for sheep’s cells. In about 90 per cent, of 
fresh inactivated human serums sufficient hemolysin is present partially 
or completely to hemolyze the usual dose of sheep-cell emulsion with the 
customary amount of guinea-pig complement. In fact, the Bauer and 
Hecht modifications of the Wassermann reaction utilize this natural hemol- 
ysin, but, as will be pointed out further on, this factor is too variable 
to be employed in conducting the reaction, as non-specific or false positive 
results are quite likely to occur. 
As has been stated in the preceding chapter, the delicacy and accuracy 
of any complement-fixation test depend to a large extent upon proper ad- 
justment of the hemolytic system. It will readily be understood that the 
presence of an unknown quantity of natural hemolysin in a serum is a 
drawback to accurate quantitative estimations. The importance of this 
lies in the fact that for some unknown reason an excess of hemolysin may 
completely hemolyze the corpuscles, even though a small amount of the 
necessary complement has been specifically fixed by antigen and syphilis 
antibody. In this manner negative reactions may result with serums that 
would otherwise show a slight positive reaction. To remove this source 
of error Noguchi has advocated the use of an antihuman hemolytic system, 
which renders the reaction more delicate. However, comparative studies 
between antisheep and other hemolytic systems by Kolmer, Matsunami, 
and Rule1 demonstrate that, with proper technic, the influence of natural 
hemolysins may be reduced to a minimum and rendered almost negligible.2 
At any rate it is very simple if desired, and but little trouble to remove the 
antisheep hemolysin routinely from human serums previous to making 
the tests (for technic, see p. 399). Furthermore, the influence of natural 
hemolysin may be prevented by heating the tubes in a water-bath to 55° C. 
for ten minutes after the secondary incubation, as I have recently described.3 
This inactivates complement and thereby breaks up the hemolytic system. 
Methods for titrating immune hemolysin will be dealt with in giving a 
detailed description of the various methods that follow. 
Red Blood-corpuscles 
Defibrinated sheep blood is washed three times with an excess of sterile 
normal salt solution to remove all traces of serum. Methods for collecting, 
washing, and preserving blood-cells are described on pages 11-13. 
In the original Wassermann reaction a 5 per cent, suspension in salt 
solution is used in doses of 1 c.c. This emulsion is quite heavy, and sharper 
and clear results are secured by using just half this amount and at the same 
time sufficient cells are used to make the readings easy and distinct. Either 
0.5 c.c. of a 5 per cent, or 1 c.c. of a 2.5 per cent, suspension may be used. 
I have used the latter with entire satisfaction for several years. 
The suspension of cells is prepared with sterile 0.85 per cent, sodium 
chlorid solution. To 2.5 c.c. of corpuscles sufficient salt solution is added 
to bring the total volume of the emulsion up to 100 c.c., or smaller amounts 
may be prepared by suspending 1 c.c. of corpuscles in 39 c.c. of salt solution. 
Before each day’s work the amount of corpuscle suspension needed 
should be calculated, and sufficient for the day prepared at one time, for 
if a fresh suspension is prepared later, titration with the complement and 
hemolysin would be required. Attempts to count the corpuscles in sus- 
pension can only be regarded as approximate and are unreliable. By titrat- 
i Amer. Jour. Syph., 1920, 4, 278. 
2 Ibid., 1920, 4, 135. 
* Ibid., 1921, 5, 628. 452 
COMPLEMENT FIXATION IN SYPHILIS 
ing each suspension with the complement and hemolysin to be used, that par- 
ticular emulsion is thereby adjusted, so that it is immaterial whether a few more 
or a few less corpuscles are present. 
Sheep’s blood is obtained either from an abattoir or by bleeding an 
animal from the external jugular vein (p. 38). The latter method is prefer- 
able, but due care must be exercised not to bleed too frequently or in ex- 
cessive amounts, as if anemia occurs the corpuscles become unduly fragile. 
Sheep’s cells are easily preserved in a satisfactory condition for forty- 
eight hours by first washing them and then storing the sedimented cor- 
puscles in a good ice-chest. Suspensions are less easily preserved. It is 
best to use fresh corpuscles, and those that are several days old and un- 
duly fragile should never be used. 
. The following technic for preparing suspensions may be employed as 
recommended by Kolmer and Brown1: 
1. A satisfactory electric or water centrifuge should be available for 
washing the blood; the former is to be preferred. The instrument should 
be started and stopped slowly. 
2. Washing the blood is best accomplished in accurately graduated 
centrifuge tubes. Those tubes may be sterilized prior to use, but this is 
not necessary; they should be chemically clean and rinsed with physiologic 
saline solution before use. 
3. For washing blood physiologic saline is employed (0.85 per cent, 
sodium chlorid in distilled water). 
4. In washing defibrinated blood one volume of blood is placed in a 
centrifuge tube, two volumes of saline solution added and gently mixed; 
citrated blood collected in the proportion of 1 part blood to 4 parts 
citrate solution is used without further dilution. Each tube must be accu- 
rately counterbalanced in the centrifuge and whirled until all corpuscles 
have been thrown down, the time required depending upon the speed of 
the centrifuge (first washing). 
5. The supernatant fluid is carefully removed down to the corpuscle 
mass with a capillary pipet (attached to suction pump or rubber teat), 
and at least three to five volumes of saline solution added; by capping 
the tube and inverting all the corpuscles are thoroughly but gently stirred 
and mixed in the saline solution, and the tubes again centrifuged (second 
washing). 
6. The supernatant fluid is again removed, replaced with saline solution, 
the corpuscles thoroughly mixed and again centrifuged for twice as long 
as found necessary to throw down the corpuscles in the second washing. 
This longer period of centrifuging is advisable to firmly and evenly pack 
the washed cells and the speed and duration of centrifuging should be uni- 
form in each laboratory as based upon experience with the particular centrifuge 
in use. 
7. If the supernatant fluid is discolored with hemoglobin the cells should 
be washed again until the supernatant fluid is practically colorless. 
8. With the last washing the centrifuge is stopped slowly so as not to 
disturb the corpuscle mass, and the volume of corpuscles read in the centrifuge 
tube before removing the supernatant fluid; the latter is now removed and a 
suspension of proper strength prepared. A 5 per cent, suspension, for 
example, is prepared by washing the cells from the centrifuge tube into a 
proper flask with nineteen times as much saline solution as corpuscle mass. 
9. As the corpuscles in suspension tend to settle rapidly, the suspension 
should be gently and thoroughly shaken by hand before being used and dur- 
1Amer. Jour. Syph., 1919, 3, 169. ANTIGENS 
453 
ing the time of use, if several minutes are required for pipetting the suspen- 
sion; otherwise the doses will be uneven and inaccurate. 
10. So far as practicable the suspension should be kept in a refrigerator 
when not being used. 
Antigens 
As was previously stated, the ordinary alcoholic extracts of syphilitic 
livers used as “antigens” in conducting the Wassermann reaction are not 
biologically specific. It is generally accepted that even in watery extracts 
of syphilitic livers the main antigenic principles are lipoidal substances, 
independent of the Treponema pallidum itself. Next to pure cultures of 
pallida, these aqueous extracts of luetic livers come closest to being a specific 
biologic antigen. Although alcoholic extracts of luetic liver may contain 
special lipoidal substances that enhance their efficacy as antigens, yet, as 
shown by Noguchi, and as confirmed later by us, alcohol does not serve 
well to extract pure cultures of pallida, and therefore these extracts can 
hardly be regarded as specific, in the sense that they contain antigenic 
principles of the spirochetes themselves. The only specific biologic antigen 
is an aqueous extract of a pure culture of pallida; this antigen is, however, 
much less serviceable than an ordinary organic extract because the Wasser- 
mann reaction depends upon the peculiar lipodophilic “reagin,” which 
absorbs complement with lipoids in a characteristic but biologically non- 
specific manner. 
The term “antigen,” as ordinarily used in the Wassermann reaction, 
must therefore be regarded as a misnomer. It is, however, so generally 
used that it may be retained with a distinct understanding as to its actual 
meaning. 
With the discovery that alcoholic extracts of normal organs may serve 
as antigen and that the chief antigenic principles reside in the lipoids, it 
followed that extensive researches were undertaken in the hope of dis- 
covering a lipoid, or a combination of lipoids, that would prove sufficiently 
delicate to act specifically in the serum diagnosis of syphilis, and not with 
normal serums or those of persons suffering from other diseases. As a 
result a large number of different extracts are in use. Each has its own 
advocates, so that the general subject of antigens is the most complicated 
one with which we have to deal in performing the Wassermann reaction. 
While various organic extracts may be used, practical experience has 
shown that some are better than others. It is important to remember: 
1. That all antigens are capable in themselves of absorbing a certain amount 
of complement. This is due to the presence of undesirable extractives, some 
preparations containing more than others. In certain doses, however, 
all antigens are capable of exerting this anticomplementary action, and, 
other things being equal, that antigen is best which shows this non-specifi- 
cation in the smallest degree. Before an antigen may be used in conducting 
any complement-fixation test it is necessary to ascertain its anticomple- 
mentary dose, for if this dose were used, a portion of or all the complement 
would be fixed in a non-specific manner, so that hemolysis, being partial 
or absent, yields false positive reactions. 
2. Most antigens, when in sufficiently large amounts, are hemolytic, i. e., 
they may hemolyze corpuscles in a non-specific manner. This hemolytic 
action is usually due to the presence of undesirable extractives, and ex- 
tracts of organs that have undergone advanced autolysis or fatty degenera- 
tion are known to contain more of these hemolytic substances than do 
extracts of normal organs. As a general rule, a highly anticomplementary 454 
COMPLEMENT FIXATION IN SYPHILIS 
antigen is likely to be correspondingly highly hemolytic. The hemolysis 
may be due to the presence of lipoidal substances or to the alcohol used 
in preparing the extract. If an antigen were used in an amount equal to 
its hemolytic dose, partial or complete hemolysis would occur in all tubes, 
so that a false negative result would be secured. As a rule, the hemolytic 
dose of an antigen as determined by titration in the presence of serum is 
larger than the anticomplementary dose, so that if the latter is known, 
it is not always necessary to determine the former. 
3. Practically every alcoholic organic extract will serve, in certain amounts, 
to absorb complement in the presence of the serum of a syphilitic person. Some 
extracts, however, will do this better than others. The Wassermann re- 
action depends upon the fact that a larger amount of complement is fixed 
by the syphilis antibody and extract than is fixed by normal serum or the 
serum of a person with some disease other than syphilis and this same ex- 
tract. The only notable exception to this general rule is to be found with 
the serum of frambesia. The amount of antigen that is found, by a process 
of titration, to fix a large amount of complement with a constant dose of 
syphilitic serum is known as its antigenic unit. Not every lipoid serves 
equally well as antigen, and therefore considerable research work has been 
done in the hope of discovering an extract or a combination of lipoidal 
substances that would show a constant reaction and would react only with 
the syphilis antibody. Thus far this has not been accomplished; un- 
fortunately, pure pallida antigens are not entirely specific or serviceable, 
and if the specific and ideal antigen is discovered in the future it will 
probably be of the nature of a lipoidal substance, altered or produced in a 
specific manner by the Treponema pallidum itself. In the meantime we 
have antigens sufficiently delicate and specific, when properly used, to 
render the Wassermann reaction of great value in the diagnosis of syphilis 
and to serve as a guide to its treatment. 
From a practical standpoint, therefore, to be suitable for use as antigen 
in the syphilitic reaction any extract or preparation must fulfil the follow- 
ing requirements: 
1. It should be largely free from anticomplementary action. 
2. It should likewise be free from hemolytic action, in small doses at least. 
3. It should possess a high degree of sensitiveness for the syphilitic 
antibody, i. e., be capable of absorbing relatively large amounts of com- 
plement in the presence of syphilitic serum. A good antigen is one that, 
in small amounts, is perfectly antigenic, and that does not become anti- 
complementary or hemolytic until from four to ten times this amount is used. 
4. It should be quite stable and not difficult to prepare, and different 
preparations should bear a certain relationship to one another in their 
properties—that is, they should keep well, and different extracts prepared 
in the same manner should show fairly constant antigenic, anticomple- 
mentary, and hemolytic doses. 
Preparation of Antigens.—The following antigens have been most widely 
used and recommended: 
1. Aqueous extracts of syphilitic livers. 
2. Alcoholic extracts of syphilitic livers. 
3. Alcoholic extracts of normal organs. 
4. Alcoholic extracts of normal organs reinforced with cholesterin. 
5. Acetone-insoluble lipoids. 
6. Alcoholic extracts of heart reinforced with lecithin and cholesterin 
(author’s new antigen). 
- 7. Aqueous extracts of pallidum culture. ANTIGENS 
455 
1. Aqueous Extracts of Syphilitic Livers.—This is the original antigen, 
as employed by Wassermann, Neisser, and Bruck; Wassermann still uses 
these extracts in preference to others. They may contain spirochetes or 
their direct derivatives, and, as shown originally by Wassermann and 
Neisser, may be true biologic antigens, for when injected into monkeys 
antibodies are formed. 
No satisfactory analyses of these extracts have been made. Chemically 
they differ in no essential respect from the liver of acute yellow atrophy 
(Ehrmann and Stern; Seligman and Pinkus). As antigen they are more 
efficient than similar extracts of normal liver. The nature of the specific 
factor has not yet been demonstrated with certainty. They react with 
the serums of yaws, and, as in the case of other antigens, their main anti- 
genic principle is apparently due to the presence of lipoids. 
They are less stable than alcoholic extracts, and are likely to become 
highly anticomplementary and lose their power of reacting with syphilitic 
serums. Citron is convinced that these changes are brought about by 
careless handling of the extract or its exposure to the light. He recom- 
mends that the extract be kept constantly in the ice-chest, and that it be 
kept out only long enough to remove sufficient for the day’s work. 
Preparation.—The fresh liver taken from a syphilitic fetus, and showing the presence 
of spirochetes by dark-ground illumination, is weighed and cut into fine pieces. Four times 
its weight of 0.5 per cent, phenol in physiologic salt solution is added. The mixture is placed 
in a brown bottle and shaken mechanically at room temperature for twenty-four hours. It 
is then filtered through gauze, to remove the larger particles, and stored in a brown bottle 
in an ice-chest. After several days of sedimentation the fluid assumes a yellowish-brown 
opalescence and is ready for the preliminary titration to determine its anticomplementary 
and hemolytic doses. The sediment should not be disturbed, but the supernatant fluid should 
be carefully removed by means of a pipet. According to Citron, extracts that must be 
used in quantities of less than 0.1 c.c. are, as a general rule, unsatisfactory. Only such 
extracts should be used as in doses of 0.4 c.c. will not interfere with hemolysis. The method 
of making these titrations is given later. 
2. Alcoholic Extracts of Syphilitic Livers.—These antigens are exten- 
sively used. They are not true biologic antigens, for they do not give rise 
to antibodies (Schatilof and Isabolinsky; Seligman and Pinkus); they are, 
however, usually better antigens than similar extracts of normal liver, a 
fact that may be explained, in part at least, by chemical changes, namely, 
fatty changes, autolysis, soaps (Beueker), excess of cholesterin (Piglini), 
etc., which, while not specifically syphilitic in nature, are often produced 
to a striking degree in congenital syphilis. 
Preparation.—Fetal liver known to contain numerous spirochetes is used in preparing 
this extract. Fresh organs may be examined at once by dark-field illumination, or if this is 
impossible and the fetus shows signs of syphilis, the liver may be cut into large pieces and 
preserved in 70 per cent, alcohol. After a few days a section is removed and stained by the 
Levaditi method for spirochetes. If these micro-organisms are numerous, the liver is suit- 
able for preparing the antigen; otherwise it should be discarded. Very fatty livers are to be 
avoided, and those of stillborn fetuses are to be preferred. 
Ten grams of liver are minced, ground with quartz sand, and treated with 100 c.c. of 
absolute ethyl alcohol. The mixture is shaken mechanically with glass beads for twenty- 
four hours, and extracted in the incubator for ten days. The containing flask or bottle 
should be well stoppered to prevent undue evaporation, and should be shaken up at least 
once a day. The extract is then filtered through fat-free paper or paper washed with ether 
and alcohol to remove the hemolytic substances that may be present. The filtrate is measured, 
and the loss by evaporation is made up by the addition of more alcohol. If a shaking ap- 
paratus is not at hand, extractions may be left in the incubator a few days longer. After 
standing a few days a sediment forms, which should not be removed or disturbed. 
3. Alcoholic Extracts of Normal Organs.—These are used extensively 
at present, and apparently yield results equal to those obtained with ex- 456 
COMPLEMENT FIXATION IN SYPHILIS 
tracts of luetic liver. It is certainly true that a good extract of a normal 
organ is superior to a poor one prepared from luetic liver. Many, with 
the idea of specificity uppermost in their mind, adhere to the use of the 
latter, whereas the results of research and of practical work shows that 
lipoids from normal organs serve equally well as antigen in making the 
syphilitic reaction, and, indeed, may prove superior if luetic liver is used 
that has undergone advanced fatty changes or autolysis, when undesirable 
hemolytic and anticomplementary derivatives are extracted in excess. 
Human, guinea-pig, and beef-heart muscle are usually employed. The 
first is especially efficient and is to be preferred. 
Preparation.—The organ is obtained fresh from the autopsy room. It is freed from 
fat, and to each 10 grams of minced muscle 100 c.c. of absolute ethyl alcohol are added. 
Extraction is conducted in exactly the same manner as was described in the preparation of 
alcoholic extract of luetic liver. 
If guinea-pig heart is employed, as much of the fat as possible should be removed, other- 
wise the extract may be quite anticomplementary. 
Boas prepared an extract of human heart by treating the ground muscle with nine parts 
of absolute alcohol, shaking for an hour at room temperature, filtering, and storing in 
a stoppered bottle. He found that different extracts so prepared are remarkably constant 
in their properties, although they deteriorate rapidly and should be prepared freshly every 
few weeks. 
4. Alcoholic Extracts of Normal Organs Reinforced with Cholesterin. 
—Sachs1 advocated the addition of pure cholesterin to alcoholic extracts 
of normal heart as a means of rendering these antigens more delicate with- 
out materially increasing their anticomplementary and hemolytic properties. 
He found that such preparations possess properties equal to the best syph- 
ilitic extracts. This work has been confirmed by McIntosh and Fieldes,2 
Walker, and Swift,3 Laubaugh, Williams, Casselman, and myself4 and others. 
We have studied the subject with much interest, comparing the results 
with those secured from other antigens, as alcoholic extracts of syphilitic 
liver and acetone-insoluble lipoids. It is true that these preparations are 
highly sensitive—so much so that I never employ them alone in making 
diagnostic reactions, but always in conjunction with other extracts as 
controls, in order to detect and avoid non-specific reactions with non-luetic 
serums. We have found that they occasionally give faint positive reactions 
with normal serums; on the other hand, not infrequently, they react strongly 
positive in cases where, with other extracts, the reactions are negative; 
in the majority of such cases the serum is from a long-standing or a treated 
case of lues that needs further treatment until the reaction with a cholesterin 
extract becomes negative. These alcoholic extracts of normal organs have 
their greatest value, therefore, with known syphilitic serums when the 
reaction is conducted as a guide to treatment. In diagnostic reactions, 
however, it is my opinion that they should not be used alone, but together 
with less sensitive antigens. In other words, one should use every precaution 
and exercise great care before making a diagnosis; when lues is known to be 
present, however, the treatment should be thorough, and there would seem to 
be no better criterion for judging the state of the infection than repeated negative 
reactions with cholesterin extracts. 
Preparation.—These extracts are prepared of human, ox, and guinea-pig heart. Hu- 
man heart usually yields the best extract. Care should be taken to use only muscle and 
to avoid fat. To 10 grams of minced muscle add 100 c.c. of absolute ethyl alcohol. Shake 
1 Berl. klin. Wchn., 1911, 48, 2066. 
2 Ztschr. f. Chemotherapie, 1912, 1, 76. 
3 Jour. Exper. Med., 1913, 18, 75. 
4 Archiv. Int. Med., 1914, 12, 660. ANTIGENS 
457 
in a mechanical shaker for twenty-four hours, and continue the extraction in the incubator 
for ten days or two weeks. Then filter through fat-free filter-paper, and add absolute alcohol 
to make up for the loss through evaporation. Add 0.2 gram of Kahlbaum’s cholesterin 
(0.2 per cent.); shake well and stand aside in the refrigerator for a few days. The cholesterin 
goes into solution slowly in cold alcohol. After a week the extract may be again filtered 
and stored in a tightly stoppered bottle. The slight sediment that may form should not be 
disturbed. 
These extracts keep fairly well. Different preparations are quite similar 
in their properties; they are usually found to be highly antigenic and no 
more anticomplementary than crude alcoholic extracts. They constitute, 
therefore, inexpensive and very sensitive antigens. 
5. Acetone-insoluble Lipoids.—As previously stated, crude alcoholic 
extracts may contain an excess of undesirable constituents, such as neutral 
fats, fatty acids, soaps, and certain protein materials, which are responsible 
for the untoward anticomplementary and hemolytic effects. To eliminate 
these Noguchi and Bronfenbrenner1 advised the exclusive use of the acetone- 
insoluble fraction instead of the entire unfractionated alcoholic extract, 
especially if unheated human serums are used in conducting the syphilis 
reaction. These extracts are composed essentially of lecithins, which, 
when prepared from any one source, consist of a mixture of analogous 
bodies; lecithins from different sources vary in their composition. In speak- 
ing of lecithins, one is prone to regard them as chemicals, and to overlook 
their biologic properties. Noguchi no longer employs the term “lecithin” 
to designate the acetone-insoluble fraction of tissue lipoids. 
These antigens are readily prepared of ox heart or of human liver, the 
former being preferable for use. Their main disadvantage is the expense 
of preparation, for it may be necessary to prepare several extracts before 
one that is satisfactory is secured. A good extract will, however, keep 
well, and is a reliable and valuable antigen for the testing for the syphilitic 
reaction. 
Preparation.—A mashed paste of the muscle of ox heart is extracted with 10 parts of 
absolute alcohol at 37° C. for four days. It is then filtered through filter-paper and the 
filtrate collected and brought to a state of dryness by evaporation. The use of the electric 
fan is not necessary, for if poured into large flat dishes, the filtrate will evaporate in from 
twelve to twenty-four hours. The residue is then taken up with a sufficient quantity of ether, 
and the turbid ethereal solution is allowed to stand for a few hours in a cool place until cleared. 
The clear ethereal portion is then carefully decanted off into another clean evaporating dish, 
and then allowed to become concentrated by evaporating the ether off. The concentrated 
ethereal solution is now mixed with about 10 volumes of pure acetone. A light yellow pre- 
cipitate forms, which is allowed to settle, and the supernatant fluid is decanted off. Dis- 
solve each 0.3 gm. of this substance in 1 c.c. of ether and add 9 c.c. of pure methyl alcohol. 
As a rule, the greater part of the substance goes into solution. This alcoholic solution re- 
mains unaltered for a long time, and is kept as a stock solution from which the emulsion for 
immediate use may be prepared at any time by mixing 1 c.c. with 19 c.c. of saline solution. 
This solution is then titrated for antigenic, anticomplementary, and hemolytic action. 
According to Noguchi, if the extract is anticomplementary or hemo- 
lytic in doses of 0.4 c.c. of a 1 : 10 dilution, it is unsuitable. If it produces 
complete inhibition of hemolysis with 0.1 c.c. of syphilitic serum in doses 
of 0.02 c.c. or less of the same dilution (= 0.2 c.c. of a 1 : 100 dilution), it 
is suitable. In making the fixation test, 0.1 c.c. of a 1 : 10 emulsion is to 
be used, thus employing five times the minimal antigenic dose which does 
not cause non-specific fixation and is not unduly sensitive. 
6. A Cholesterolized and Lecithinized Alcoholic Extract of Heart Muscle 
(Author’s New Antigen).—As a result of investigations upon the subject 
1 Jour. Exper. Med., 1911, 13, 43. 458 
COMPLEMENT FIXATION IN SYPHILIS 
of antigens for the Wassermann test1 a new method has been found very 
satisfactory. Due to the investigations of Noguchi,2 Browning, Cruick- 
shank and McKenzie,3 Erlandsen,4 Neymann and Gager,6 and others we 
now know that the lecithins (diaminomonophosphatids) are highly anti- 
genic and very slightly hemolytic and anticomplementary, and Noguchi’s 
method for extracting these from tissues (acetone-insoluble lipoids) is highly 
efficient and somewhat preferable to the more complicated procedure of 
Browning, Cruickshank and McKenzie. As shown by Erlandsen and 
Neymann and Gager, ether removes from tissues a large portion of the 
hemolytic substances, principally soaps, neutral fats, and fatty acids; also 
bile salts in extracts of liver tissue. Alcohol also removes some of these 
in addition to substances possessing marked anticomplementary activities, 
as proteins and protein cleavage products, cholesterol and other lipoidal 
bodies, and various salts. No doubt some of the latter are also antigenic, 
but at least the best of them, namely, cholesterol, can be prepared in a high 
state of purity and added to any extract for the advantage of its influence 
upon antigenic activity. 
The following method utilizes these observations and the results of 
previous studies on antigens,6 and yields extracts of remarkable antigenic 
sensitiveness and slight anticomplementary and hemolytic activity; the 
principles of the method are as follows: 
1. Dried human or beef-heart muscle is employed; there is no choice 
between the two tissues if both are perfectly fresh. Human heart muscle, 
however, is frequently fatty or may have undergone more or less postmortem 
change; if so, fresh lean beef-heart muscle is to be preferred. Dried tissue 
is employed for the following reasons: (a) It permits of the use of very 
finely divided tissue; (b) tissues are easily kept for long periods of time; 
(c) three or more heart powders may be kept, permitting of the use of mix- 
tures of tissues and polytropic extracts which are probably better than 
preparing an extract of a single heart muscle. 
2. The dried tissue is first extracted with ether, which removes a large 
portion of the hemolytic substances; the phosphatids (lecithins), which are 
also partly removed by the ether, are recovered by precipitation with acetone 
and returned to the extract. 
3. The tissue is now dried and extracted with alcohol, which removes 
some antigenic and hemolytic substances and considerable anticomple- 
mentary substances. This alcohol is evaporated, extracted with ether, and 
precipitated with acetone, which removes the highly antigenic fraction of 
lecithins (phosphatids), which are later returned to the extract. 
4. The tissue is now extracted for a second time with absolute ethyl 
alcohol which removes some antigenic principles and slight or almost negli- 
gible amounts of anticomplementary and hemolytic substances. The 
acetone-insoluble lipoids (lecithins, phosphatids) recovered from the pre- 
liminary ether and alcohol extractions are dissolved in ether and returned 
to this alcoholic extract with 0.2 per cent, cholesterol; the resulting ex- 
tract possesses high antigenic activity, slight anticomplementary, and no 
hemolytic activity. The details of preparation of this extract are as follows: 
Preparation.—(a) Fresh beef or human hearts are washed free of blood, dissected free 
of fat and large blood-vessels, and the muscle passed through a meat grinder three or four 
times. The minced tissue is then rapidly dried in a vacuum apparatus or equally well by 
spreading in thin layers on clean glass plates and drying by rapid fanning, preferably in a dust- 
1 Amer. Jour. Syph., 1922, 6, 74, 
2 Jour. Exper. Med., 1909, 11, 84. 
3 Jour. Path, and Bacteriol., 1910, 14, 484. 
4Ztschr. f. physiol. Chem., 1907, 51, 71. 
6 Jour. Immunology, 1917, 2, 573. 
6 Amer. Jour. Syph., 1922, 6, 289. ANTIGENS 
459 
proof box, for eighteen to twenty-four hours, turning the layers after the first ten to twelve 
hours. The cakes of dried material are now broken up, placed in an incubator over night 
and ground into a fine powder which is kept in tightly stoppered bottles of colored glass 
at room temperature. Three or more hearts may he prepared at one time and a mixture of the 
powder used in preparing extracts. Rapid drying is essential to prevent decomposition which 
results in greatly increasing the hemolytic activity of the extracts. 
(b) Twenty-five grams of powdered muscle are extracted with about 100 c.c. of ether 
in a Sohxlet for eighteen hours; if this apparatus is not available the powder is extracted 
with 200 c.c. of ether in a tightly stoppered bottle at room temperature for five days, being 
shaken occasionally each day. The ether is carefully removed and saved for the time being 
in a tightly stoppered bottle. 
(c) The powder is now dried by fanning for a few minutes or by spreading on a glass 
plate for several hours, placed in a bottle, and extracted with 200 c.c. of 95 per cent, alcohol 
in an incubator for four days. 
(id) The alcohol is carefully decanted, poured into a flat shallow dish, and fanned dry. 
The residue is extracted with 30 to 50 c.c. of ether for the ether-soluble portion; this ethereal 
extract is covered and allowed to stand for an hour or two for the heavy insoluble particles to 
settle out. 
The ethereal extract is now mixed with the ether of primary extraction and the mixture 
concentrated by fanning until reduced to about one-quarter volume or about 25 to 30 c.c. 
Six volumes or about 150 c.c. of pure acetone are now added to the concentrated ether, which 
throws down a whitish precipitate; the mixture is covered and placed aside for several hours 
or over night for the complete separation of the acetone-insoluble portions. On the following 
day the supernatant acetone is decanted and the sticky residue of acetone-insoluble lipoids 
removed and kept in acetone in a tightly stoppered wide-mouthed bottle for future use. 
(<?) The muscle powder is now extracted for a second time with 100 c.c. of absolute 
acetone-free ethyl alcohol in an incubator for six days, guarding against evaporation and 
shaking the mixture once or twice a day, and if possible for at least one day in a mechanical 
shaker; the extract is now filtered through fat-free paper. 
(/) Pure cholesterol, 0.2 gram (Kahlbaum’s preferred), and all of the acetone insoluble 
lipoids previously prepared are dissolved in 10 c.c. of pure ether and the cloudy brownish 
mixture slowly added to the filtered alcoholic extract and well shaken. The extract is now 
placed in the incubator for a few hours or over night, being shaken occasionally, and then 
placed in a refrigerator for a day or two; the light brownish precipitate is now removed by 
filtration through fat-free paper or by decanting the extract. The finished antigen is kept 
in a tightly stoppered brown glass bottle in a refrigerator; any precipitate forming after this 
t ime is left undisturbed. 
7. Aqueous Extracts of Pallidum Culture.—This antigen was prepared by 
Noguchi, who used pure cultures of Treponema pallidum in ascites kidney 
agar. Preferably several strains should be used in the preparation, which 
corresponds quite closely to luetin. Cultures grown seven, fourteen, twenty- 
one, twenty-eight, thirty-five, and forty-two days are chosen and examined, 
and those that show the best growths in the agar columns are selected. 
The oil is poured off, the tubes cut just above the kidney, and the column 
of ascites agar between the piece of kidney and the oil removed with par- 
ticular care, so as not to include the kidney or the oil. This substance is 
ground in a mill until the spirochetes show disintegration. The thick emul- 
sion is then diluted with normal salt solution and heated to 60° C. for one- 
half hour; 0.4 per cent, phenol is added as a preservative, and the emul- 
sion titrated for its anticomplementary dose. In conducting complement- 
fixation reactions with pallidum antigen one-half of the anticomplementary 
dose is used, and the serum must be inactivated. 
Comparative Antigenic Values of Various Extracts.—For several years 
past I have been particularly interested in studying, from a practical stand- 
point, antigenic values of the extracts most commonly employed in the 
serum diagnosis of syphilis, and comparing them with suitable alcoholic 
extracts of syphilitic liver as a standard antigen. 
For this purpose antigens were carefully chosen after titration, and 
only those were employed that were safely free from anticomplementary 
action and whose antigenic dose was known. A large number of serums 460 
COMPLEMENT FIXATION IN SYPHILIS 
and cerebrospinal fluids from syphilitic and non-syphilitic persons were 
tested with numerous different extracts at the same time and under similar 
conditions. A suitable alcoholic extract of syphilitic liver was always in- 
cluded among the antigens in testing each serum or fluid, and the other 
extracts compared with it in determining their antigenic value. 
In the following table the results of such studies, covering a period of 
two years, are given: 
Comparative Antigenic Values or Various Tissue Extracts 
Antigenic Properties as Compared to Alcoholic Ex- 
tracts of Syphilitic Liver. 
Extracts. 
Equal. 
Stronger. 
Weaker. 
Negative 
(positive 
with alco- 
holic extract 
of syphilitic 
liver). 
Positive 
(negative 
with alco- 
holic extract 
of syphilitic 
liver). 
Cholesterinized alcoholic ex- 
tracts of human, pig, and beef 
heart 
50.0. 
30.0 
20.0 
Acetone-insoluble lipoids 
73.0 
10.8 
9.7 
3.2 
1.8 
Alcoholic extract of pig and beef 
heart 
71.1 
1.9 
13.4 
5.7 
7.6 
Acetone extract of syphilitic 
liver 
68.7 
4.4 
18.9 
6.6 
1.3 
Alcoholic extract of normal liver 
73.2 
23.5 
3.5 
The results of these studies have shown: 
1. That cholesterinized alcoholic extracts of human, beef, and guinea- 
pig heart are far more sensitive than simple alcoholic extracts of syphilitic 
liver. Another peculiar feature of these antigens is the fact that in syphilis, 
if they react at all, they usually do so quite strongly. 
2. That the addition of cholesterin to crude alcoholic extracts of syph- 
ilitic liver and of normal liver doubles their antigenic sensitiveness with- 
out materially increasing their anticomplementary and hemolytic action. 
3. We have practically never found a serum that reacted negatively 
with a cholesterin extract and positively with an alcoholic extract of syph- 
ilitic liver. On the other hand, in about 20 per cent, of cases the choles- 
terinized antigens will react positively, whereas with the plain antigen 
of syphilitic liver the reactions are negative. In the majority of such in- 
stances the person was known to be luetic, but had received treatment 
and was regarded clinically as cured, or the serum wras that of a long-stand- 
ing and unrecognized case of lues. 
4. It is highly important that cholesterolized extracts be carefully 
standardized, and that any serum, even if but slightly anticomplementary, 
be discarded. 
5. In our experience repeated negative reactions with satisfactory 
cholesterinized antigens constitute the best evidences of the absence of 
lues or testify to the recovery from a luetic infection. The treatment of 
syphilis should be continued until the patient’s serum reacts negatively 
with alcoholic extract of syphilitic liver, and finally with cholesterinized 
extracts. The disease cannot be regarded as cured until the reaction has 
remained negative for a year or two at least, and treatment must not be 
discontinued until this result is secured, or it is shown that the serum is 
“Wassermann fast” and that it is impossible to secure a negative reaction. ANTIGENS 
461 
6. For the less experienced worker, or when but one antigen is being 
used in conducting the reaction, a properly prepared alcoholic extract of 
syphilitic liver is to be recommended. One drawback to the use of this 
extract is the difficulty of obtaining suitable tissues for the preparation 
of the antigen. It has been my practice for many years to preserve the 
livers of as many stillborn fetuses as I could obtain in 70 per cent, alcohol, 
and to discard them later unless on section they showed the presence of 
numerous spirochetes. These antigens are usually more sensitive than 
similar extracts of normal liver, but it is important to remember that not 
every extract is satisfactory simply because it is prepared of syphilitic tissues. 
7. A suitable preparation of acetone-insoluble lipoids, prepared after 
the method of Noguchi, constitutes a sensitive, reliable, and satisfactory 
antigen. When properly titrated and standardized and used with inac- 
tivated serums this antigen may prove quite sensitive and safe. Noguchi, 
in his efforts to simplify the technic of the syphilis reaction, impregnated 
filter-paper with this antigen and allowed it to dry. This preparation is 
unstable and generally unsatisfactory. The antigen is best preserved in 
a stock bottle or in ampules, and is diluted with salt solution just before 
being used in the test. Under these conditions we have found these extracts 
to be quite stable. 
8. Plain alcoholic extract of human, guinea-pig, and beef heart are 
easily prepared, are quite inexpensive, and when properly titrated serve 
as satisfactory antigens. Boas uses alcoholic extracts of human heart 
(Michaelis) exclusively, and has found that they yield better results than 
alcoholic extracts of syphilitic liver. Garbat and others use and recom- 
mend similar extracts of guinea-pig heart. 
9. Aqueous extracts of pure cultures of pallida have not thus far yielded 
results equal or superior to ordinary non-specific antigens. As compared 
with lipoidal extracts, they have generally yielded reactions that are much 
weaker, and in primary and secondary syphilis may react entirely negatively. 
Much is yet to be learned, however, of bacterial antigens in general, and the 
subject must be regarded as still in the experimental stage. 
It may be said to be well proved that extract antigens in the syphilis reac- 
tion are not biologically specific, and need not be extracts of syphilitic tissues. 
An antigen cannot give reliable or satisfactory results unless it is carefully 
titrated and its properties determined. Antigens may serve as a frequent source 
of error when the complement-fixation reactions are conducted by those possessing 
insufficient knowledge of their good and bad properties. The test for the syph- 
ilitic reaction should not be undertaken by any one not competent to titrate 
and judge of the qualities of the antigen to be employed. 
Method of Diluting Antigens.—As a general rule, all organic extracts 
must be diluted with normal salt solution before being used. 
If extracts and diluent are mixed quickly, the emulsion is clear of slightly 
opalescent. If the diluent is added slowly to the organic extract, the re- 
sulting mixture becomes quite turbid and milky. As shown by Sachs and 
Roudoni,1 the antigenic power of the extract is more marked with the turbid 
than when the clear or opalescent emulsion is used. For this reason, in 
testing for the syphilis reaction the emulsion of organic extract should be made 
so as to secure the maximum amount of turbidity. The required amount of 
antigen is placed in a test-tube, and the salt solution is added slowly with a 
pipet; or the salt solution may be placed in a tube and the extract added drop 
by drop and gently shaking with each addition. This is the preferred method. 
Although with each new extract it may be necessary to titrate with 
1 Berl. klin. Wchn., 1908, 45, 1968. 462 
COMPLEMENT FIXATION IN SYPHILIS 
various dilutions before one that is satisfactory is reached, experience has 
shown in the majority of instances that the following dilutions are usually 
correct: 
1. Alcoholic extract of syphilitic liver: 1 part with 9 parts of salt solu- 
tion. Extracts of German manufacture are usually diluted six or seven 
times. 
2. Cholesterinized extracts and acetone-insoluble lipoids in methyl 
alcohol require higher dilution, as 1 part of extract with 19 parts of salt 
solution. 
3. Plain alcoholic extracts of human, beef, and guinea-pig heart: 1 part 
and salt solution, 19 parts. 
4. Alcoholic extract of human heart, prepared after the method of Boas 
(quick method), is used in lower dilution, as 1 part of extract with 9 parts 
of salt solution. 
5. The alcoholic extract of heart reinforced with lecithin and choles- 
terin used by the author is diluted with varying amounts of salt solution 
as described in the text. 
6. Aqueous extracts of syphilitic liver and extracts of pallida culture are 
used undiluted, or may require dilution with 4 parts of salt solution. 
Although these emulsions will keep for a few days if placed in the refrigerator, 
it is advisable to make up only the amount required for immediate use, as freshly 
prepared emulsions are better than older ones. 
Principles for the Titration of Antigens.—Three values are to be deter- 
mined: 
1. The anticomplementary unit, or that amount of antigen that in itself 
is capable of fixing or inactivating the complement. 
2. The hemolytic unit, or that amount of antigen that in itself is capable 
of lysing red blood-cells. This action is probably due to the presence of 
certain lipoids and alcohol. 
3. The antigenic unit, or that amount of antigen that serves to absorb 
or fix a certain and constant dose of complement with a definite amount 
of syphilitic serum. 
1. Anticomplementary Titration.—(a) Emulsions should be prepared by 
adding the extract slowly drop by drop from a 1 c.c. pipet to the measured 
amount of saline and shaking gently. (b) Fresh non-syphilitic serum should 
be used, but is not absolutely necessary; a mixture of sera is preferable in 
order to equalize the content of natural hemolysins. The serum should be 
heated in exactly the same manner as in the main tests and used in an aver- 
age amount, (c) The mixtures of antigen, serum, and complement should 
be given exactly the same primary incubation as used in the main comple- 
ment-fixation tests; also receive the same amounts of hemolysin and cor- 
puscles and secondary incubation. (d) The titration should be conducted 
with increasing amounts of a given emulsion with constant amounts of 
serum and complement to ascertain the smallest amount of extract just produc- 
ing beginning inhibition of hemolysis which is designated as the anticomple- 
mentary unit. 
2. Hemolytic Titration.—(a) An average amount of heated human- 
or guinea-pig-serum should be used, (b) Increasing amounts of antigen 
emulsion with a constant amount of serum and corpuscles should be in- 
cubated in exactly the same manner as the secondary incubation of the 
main complement-fixation tests to ascertain the smallest amount of extract 
just producing beginning hemolysis which is designated as the unitr 
3. Antigenic Titration.—(a) Emulsions should be prepared by adding 
the extract slowly drop by drop from a 1 c.c. pipet to the measured amount VARIOUS METHODS FOR CONDUCTING THE SYPHILIS REACTION 
463 
of saline and shaking gently. (b) A mixture of strongly positive syphilitic 
sera should be used; this serum should be heated in exactly the same manner 
as in the main tests and used in an average amount, (c) The mixtures of 
antigen, serum, and complement should be given exactly the same primary 
incubation as used in the main complement-fixation tests; also receive the 
same amounts of hemolysin and corpuscles and secondary incubation. 
(<d) The titration should be conducted with increasing amounts of antigen 
with constant amounts of serum and complement to ascertain the smallest 
amount of extract giving complete inhibition of hemolysis which is designated 
as the unit. 
In these titrations the hemolytic system should be exactly as employed 
in the main complement-fixation tests. The complement should be pre- 
pared in the same manner and used in the same number of units; likewise 
the hemolysin and corpuscles and the total volume should be the same. 
VARIOUS METHODS FOR CONDUCTING THE SYPHILIS REACTION 
First Method; the Original Wassermann Reaction.—The simplest 
technic, and the one best adapted for inexperienced workers, is the original 
Wassermann reaction, performed with alcoholic instead of aqueous extracts 
of syphilitic liver as antigen. It is true that this method is not an exact 
quantitative reaction, and that it is probably less delicate than some of the 
modified methods, but its advantages are that it is easy of manipulation, 
is readily learned, and is especially recommended for persons who perform 
these tests at irregular intervals, as false positive reactions are less likely 
to occur than when the more delicate methods are used. 
Second Method; the Author’s Modification of the Wassermann Reac- 
tion with Multiple Antigens.—In this method the technic is essentially the 
same as in the first method, except that three different antigens are used 
instead of one, namely, cholesterinized extract of normal heart, plain alcoholic 
extract of syphilitic liver, human or beef heart, and acetone-insoluble 
lipoids. 
With strongly reacting serums all antigens may yield positive reactions. 
With the serums of long-standing or treated cases of syphilis the choles- 
terinized extracts may react strongly positive, whereas with the plain 
alcoholic extracts the reactions are weakly positive, or negative with one 
and positive with the other. In cases of syphilis that have received con- 
siderable treatment the reaction may be negative at first with the plain 
alcoholic extracts, and as treatment is continued it may finally be negative 
with the cholesterinized extracts. In a certain percentage of cases, 
especially those of old infections of the central nervous system, the reaction 
is positive with the cholesterinized extract and negative with the other 
extracts; strong reactions of this character usually indicate syphilitic 
infection. 
Third Method; the Author’s Modification of the Wassermann Test 
with Varying Amounts of Serum and Based Upon Studies in the Standard- 
ization of Technic.—This complement-fixation technic is based upon the 
results of nearly six years of continuous investigation upon the subject of 
standardization of technic. 
The method of study pursued was to become acquainted with all exist- 
ing methods by thoroughly reviewing the available literature and by means 
of personal communications and interviews with a large group of serol- 
ogists and submitting the whole to careful unbiased experiment, and choosing 
that proving best on the basis of actual trial. 464 
COMPLEMENT FIXATION IN SYPHILIS 
Requirements of a Standard Wassermann Reaction and How These Have 
Been Fulfilled by the New Technic.—In my opinion the essential require- 
ments of a standard technic may be summarized1 as follows: (1) As high 
degree of sensitiveness as is permissible with specificity; (2) practical spec- 
ificity; (3) technical accuracy and uniformity in results; (4) yield a quan- 
titative reaction; (5) simplicity in technic; (6) economy. 
An attempt has been made to fulfil all of these requirements in the 
new technic by the following procedures: 
Meeting the Requirement of Sensitiveness.—1. By the use of a highly 
sensitive antigen. This is of paramount importance and the new extract 
previously described2 has been found almost free of hemolytic activity, 
very low in anticomplementary activity and very highly antigenic, per- 
mitting: 
2. By the use of large amounts of antigen. At least ten antigenic units of 
the new antigen may be used with entire safety, this amount being six 
times or more less than the anticomplementary and hemolytic units and 
yielding very sensitive but true reactions. A large amount of time has been 
devoted to a study of different kinds of antigen, their manufacture,3 and 
preservation4; also to the proper amount to use from the standpoint of 
securing the greatest degree of sensitiveness consistent with true reactions.5,6 
With the new extract 5 to 10 antigenic units have proved to be the optimum 
amounts to use. 
3. By using relatively large amounts of serum and spinal fluid in the test. 
This phase has required a great deal of investigation7 because there are 
limits to be placed upon the amounts to employ because: (1) The introduc- 
tion of serum constituents other than antibody may interfere with com- 
plement fixation; (2) excessive amounts may result in non-specific reactions 
due to the anticomplementary activities of serum and spinal fluid, and (3) 
reduce sensitiveness by the introduction of natural hemolysin and hemag- 
glutinins. To these may be added: (4) The question of economy. 
Most serologists use 0.1 to 0.2 c.c. of serum, and 0.4 to 1.0 c.c. spinal 
fluid; the original Wassermann test requires 0.2 c.c. serum, and 0.4 c.c. 
spinal fluid. Calculated in relation to the amount of complement employed, 
the largest amount of spinal fluid used in the new test, namely, 0.5 c.c., 
corresponds to 1.5 to 2.0 c.c. in the original Wassermann test. Extensive 
trials have shown that for routine work the use of these amounts results in 
the greatest possible degree of sensitiveness to be expected on the basis 
alone of amounts of serum and spinal fluid employed. 
4. By heating sera for only fifteen minutes at 55° C. instead of for thirty 
minutes. As stated in a previous paper8 we found it necessary in routine 
work to heat sera for the purpose of (1) removing anticomplementary sub- 
stances; (2) removing the possibility of non-specific proteotropic reactions, 
and (3) removing native complement and depending upon a mixture of 
guinea-pig complements. These objects are secured by heating for fifteen 
minutes at 55° C. unless a serum is older than four days or improperly pre- 
served, when thirty minutes may be required for the removal of antilysins. 
Heating for fifteen minutes results in much less destruction of syphilis anti- 
body, and has been found uniformly satisfactory when tests are conducted 
every three or four days, the specimen of blood kept in a refrigerator, and the 
serum left on the clot until used. Spinal fluids are used unheated. 
1 Amer. Jour. Syph., 1919, 3, 1. 
2 Ibid., 1922, 6, 74. 
3 Ibid., 1922, 6, 289. 
4 Ibid., 1922, 6, 319. 
6 Ibid., 1922, 6, 481. 
6 Ibid., 1922, 6, 651. 
7 Ibid., 1922, 5, 439, 
8 Ibid., 1920, 4, 641. VARIOUS METHODS FOR CONDUCTING THE SYPHILIS REACTION 
465 
5. By using a mixture of guinea-pig complements prepared in a manner 
tending to increase sensitiveness to fixation. As is well known, guinea-pig 
complement varies in fixability by syphilis antibody and tissue extracts1; 
we have not found it necessary to titrate individual sera for fixability, a 
mixture meeting the requirements. Human complement also varies in 
fixability and explains why a syphilitic serum may give a falsely negative 
reaction in tests employing the patient’s own complement; I am convinced 
that this may occur2 and it constitutes an important reason for conducting 
the tests with a mixture of the sera of healthy guinea-pigs. 
6. By mixing serum and antigen for a brief period before the addition of 
complement in setting up the test; this appears to slightly increase the sensi- 
tiveness of reactions? 
7. By using a primary incubation of fifteen to eighteen hours in a refrigerator 
at 6° to 8° C. This is of great importance from the standpoint of increasing 
the delicacy of reactions.4,5l6> 7 It must be emphasized, however, that refrigerator 
incubation increases non-specific complement fixation and that the hemolytic 
system must be adjusted accordingly; a hemolytic system adjusted for conduct- 
ing the primary incubation in a water-bath or thermostat for one hour is not 
likely to prove satisfactory for the refrigerator method. 
Refrigerator incubation of two hours or less with one hour water-bath 
is better than one hour in water-bath alone, but inferior to fifteen to eighteen 
hours in a refrigerator. The adoption of the latter necessarily requires two 
days for the conduct of tests; this is to be regretted, but the fact remains 
that it results in better work and is, therefore, recommended. 
8. By close adjustment of the hemolytic system in order to avoid ex- 
cessive amounts of complement and hemolysin.8,9 This is accomplished 
by titrating both hemolysin and complement before the main tests. Guinea- 
pig complement occasionally contains natural hemolysin10 and this is ad- 
justed for in the new technic by daily titration of hemolysin, the extra work 
and time required being negligible factors. Complement is titrated in 
the presence of the antigen, permitting a close adjustment of the dose to 
employ. 
Extensive experiments have shown that under these conditions and with 
a primary incubation of fifteen to eighteen hours at 6° to 8° C. it is necessary 
to use 2 f ull units of complement and 2 units of hemolysin in order to obtain 
sharp, clear, and decisive reactions without danger of non-specific results?1 
9. By using an antisheep or antiox hemolytic system. I have reached 
this decision reluctantly and only after a very large number of comparative 
tests. There is so much theoretic evidence in favor of the antihuman hemo- 
lytic system that only extensive comparative tests have shown most conclusively 
that with the new technic an antisheep system yields the best and most sensitive 
reactions. An antiox system ranks second. 
In the new test the question of the influence of natural antisheep hemol- 
ysin in the sera and complement is rendered practically negligible. Further- 
more, the antisheep system permits the use of such powerful hemolysins 
easily prepared by the immunization of rabbits, that the amount of com- 
plement and hemolytic sera required are greatly reduced. This is an im- 
portant factor, for it appears that the use of relatively large amounts of 
guinea-pig complement and rabbit hemolytic serum demanded by an anti- 
1 Amer. Jour. Syph., 1919, 3, 407. 
2 Ibid., 1919, 3, 541. 
3 Ibid., 1921, 5, 290. 
4 Ibid., 1920, 4, 675. 
5 Ibid., 1921, 5, 30. 
8 Ibid., 1921, 5, 44. 
7 Ibid., 1921, 5, 63. 
8 Ibid., 1920, 4, 518. 
9 Ibid., 1920, 4, 616. 
10 Ibid., 1920, 4, 484. 
11 Ibid., 1920, 4, 518. 466 
COMPLEMENT FIXATION IN SYPHILIS 
human system as compared with an antisheep, introduce enough other 
serum constituents to reduce the degree of complement fixation by syphilis 
antibody and antigen. Whether or not this is the true explanation, the 
fact remains that actual comparative tests under rigid conditions have 
shown the superiority of the antisheep and the antiox hemolytic systems. 
10. By reading the reactions within three hours after the conclusion of the 
secondary incubation. This permits the partial and sufficient settling of non- 
hemolyzed corpuscles for the purpose of accurate readings without allow- 
ing an excessive amount of hemolysin sometimes represented by natural 
antisheep hemolysin in a human serum1,2 to continue as is apt to occur 
when the tests are placed in a refrigerator over night before the readings 
are made.3 
Meeting the Requirements of Practical Specificity.—The phrase “practical 
specificity” is used purposely because the Wassermann reaction cannot be 
rendered biologically or absolutely specific for syphilis alone; positive re- 
actions undoubtedly occur in frambesia or yaws. 
However, the new test must avoid non-specific reactions due to avoid- 
able errors in technic: 
1. By close adjustment of the hemolytic system to a primary incuba- 
tion of fifteen to eighteen hours at 6° to 8° C. in order to supply sufficient 
complement and hemolysin for non-specific fixation and yet to detect the 
slightest degrees of specific fixation by syphilis and antigen. 
2. By careful titration of antigen under conditions rendering the dose 
employed suitable for a primary incubation of fifteen to eighteen hours at 
6° to 8° C. 
3. By incubating controls in every test and especially serum, antigen, 
and hemolytic controls to detect anticomplementary activities of serum 
and antigen or defects in the hemolytic system; also corpuscle controls to 
check the tonicity of the saline solution and fragility of cells. Test with 
sera from healthy normal and from syphilitic individuals should be in- 
cluded as positive and negative controls. 
Meeting the Requirements of Technical Accuracy and Uniformity in Re- 
sults.—This has been fulfilled in the new technic by the following pro- 
cedures : 
1. By adopting the principle that pipeting relatively large amounts of 
fluid (0.2 to 1.0 c.c.) tends to greater accuracy than measuring smaller 
amounts (less than 0.2 c.c.). This appears justified in view of the known 
inaccuracy of ordinary pipets and other measures, as well as a wide differ- 
ence in the skill and care of individual workers. In the new technic the 
smallest amount of patient’s serum or spinal fluid to be measured is never 
less than 0.2 c.c.; this permits of the use of 1 c.c. pipets divided into 0.1 c.c. 
2. By using a total volume of 3 c.c. with sufficient corpuscles and test- 
tubes of suitable size to yield clear, sharp, and easily read reactions. 
3. By using a reading scale4 furnishing hemoglobin in solution and non- 
hemolyzed corpuscles for reading the finer differences in the degree of hemol- 
ysis or no hemolysis. 
In regard to uniformity in results it must be emphasized that the anti- 
complementary activity of serum or spinal fluid is very important in rela- 
tion to reactions. For this reason tests conducted with portions of the 
same specimen of blood in different cities cannot be expected to yield abso- 
lutely similar results, nor even in the same city, if serologists vary in their 
method of preserving blood until the tests are conducted. 
1 Amer. Jour. Syph., 1920, 4, 111. 
2 Ibid., 1920, 4, 135. 
3 Ibid., 1920, 4, 135. 
4 Ibid., 1922, 6, 64. VARIOUS METHODS FOR CONDUCTING THE SYPHILIS REACTION 
467 
Two or more serologists working in the same or different laboratories 
testing portions of a sample of blood or spinal fluid from one person should 
agree at least upon the question of positive or negative reactions; in my 
experience most variation occurs with serums yielding weakly positive 
reactions. Slight discrepancies in the reports on the degree of complement 
fixation must be expected, inasmuch as the personal equation plays an 
important part in reading the degree of hemolysis, as it does in matching 
colors in other lines of work, for example, in hemoglobin estimations and 
color reactions in general. Slight discrepancies, however, do no harm as 
long as the primary question of whether a serum does or does not yield a 
positive or negative result is untouched and particularly with serums yield- 
ing the borderline weakly positive or doubtfully negative reactions. 
The new test has been found to fulfil this primary requisite, and further- 
more has yielded remarkably similar reactions in the hands cf different 
serologists working with portions of the same serum or spinal fluids in two 
different laboratories in Philadelphia. 
Meeting the Requirements of a Quantitative Reaction.—This is an im- 
portant requirement in relation to the Wassermann test as a serologic 
guide on the treatment of syphilis. The ordinary test employing a single 
dose of serum or spinal fluid is only roughly quantitative and is better 
designated as qualitative as previously described.1 
A complement-fixation test may be made quantitative by any of three 
procedures as follows: 
(a) Using varying amounts of serum or spinal fluid with constant amounts 
of complement and antigen. 
(b) Using varying amounts of complement with constant amounts of 
serum or spinal fluid and complement. 
The first two are much more satisfactory than the third; the first has 
been adopted because most economic and equally satisfactory as the others. 
In the new technic patient’s serum is used in the following amounts: 
0.1, 0.05, 0.025, 0.005, and 0.0025 with 0.1 c.c. in the serum control. Spinal 
fluid is used in amounts of 0.5, 0.25, 0.125, 0.0625, and 0.03125 c.c. with 
0.5 c.c. in the control. Extensive trials with varying amounts of serum in 
the different stages of syphilis and with spinal fluids from cases of neuro- 
syphilis, have shown that these amounts are satisfactory. One object was 
to adopt as the largest doses of serum or spinal fluid amounts yielding the 
most sensitive reactions and for the smallest doses, amounts yielding less 
than total inhibition of hemolysis in the great majority of cases of syphilis. 
Between these extremes are three graded doses making five in all, the 
control being the sixth tube of the series. Of course, a finer quantitative 
test can be secured by introducing eight tubes carrying 0.1, 0.5, 0.25, 0.125, 
0.0625, 0,03125, and 0.015 c.c. serum with 0.1 c.c. in the control, but six tubes 
has proved satisfactory and materially reduces both time and materials 
required. The reading scale permits reading the results with each of the 
five different amounts of serum or spinal fluid according to the + + + + > 
+ + + , + + , + and — scale. This technic is quantitative, therefore, in 
two directions, namely > by using five graded amounts of fluid to be tested 
with five possible readings on each. 
Meeting the Requirements of Economy.—This refers to both time and 
materials. From the standpoint of time required the new test cannot 
qualify as being economical; from the standpoint of materials it easily 
qualifies. 
The new technic is not a short-cut method; I am convinced that the 
1 Amer. Jour. Syph., 1922, 6, 64. 468 
COMPLEMENT FIXATION IN SYPHILIS 
principles involved in complement fixation are too intricate, the reagents 
too subject to variation, and our knowledge of the mechanism of the re- 
action too meager to permit the evolvement of a short cut and simple test 
fulfilling the requirements of a standard test. Doubtless the time and labor 
involved for conducting the new test will prevent its adoption by many 
serologists, but I have endeavored to adhere to the principle that accuracy 
should never be sacrificed for speed and labor saving. The new technic 
provides for a quantitative and a qualitative test; I use the former routinely 
because it requires but little more time. 
In so far as materials are concerned the quantitative test requires but 
0.3 c.c. serum and 1.5 c.c. of spinal fluid; the qualitative test requires but 
0.2 c.c. serum and 1.0 c.c. of spinal fluid. 
In the new quantitative test 1 c.c. of guinea-pig serum is usually sufficient 
for examining 6 to 7 sera or spinal fluids; in the qualitative test this amount 
suffices for at least 15 sera or spinal fluids including all controls. In the 
original Wassermann test, which is a qualitative test only, 1 c.c. of com- 
plement is sufficient for testing 8 sera or spinal fluids including the usual 
controls. Complement is the most expensive reagent and the new test 
easily meets the requirements of economy in this and all other materials. 
The amounts of blood corpuscles and hemolysin required are so small as to 
not be worthy of discussion. 
Meeting the Requirements of Simplicity.—As previously stated, simplicity 
is but a relative term in as much as the simplest technic is a complicated 
problem for the inexperienced and insufficiently trained worker, whereas, 
a more complicated technic is perfectly simple to the experienced serologist. 
A new technic introduces only well-known principles and I hope will be 
accepted as relatively simple; certainly the test can be carried out by a 
careful and conscientious worker in any laboratory supplied with accurate 
glassware, a water-bath, and a refrigerator. 
The simple examination of urine for albumin is a procedure capable of 
yielding different results in the hands of different workers; the Wassermann 
test requires a worker who has a working understanding of the principles, 
who refuses to compromise with something almost as good or almost satis- 
factory, and who conducts his or her work with a reasonable degree of accu- 
racy and skill, refusing to sacrifice these for mere speed. 
Fourth Method; the Author’s Modification of the Wassermann Reaction 
Employing Varying Amounts of Complement.—It has previously been pointed 
out that the syphilis reaction is dependent upon the fact that while hemolytic 
complement may be rendered inactive or fixed by serum alone and organic 
extract alone, it is characteristic of syphilis that a mixture of serum and 
extract will absorb or fix more complement than the sum of the amounts 
absorbed by these two substances alone. In the foregoing methods no 
attempt has been made to measure the amount of complement absorbed 
by serum and antigen alone, but sufficient complement has been furnished 
to allow for this non-specific fixation, and we are content to show that the 
serum and antigen alone do not absorb enough complement to interfere 
with hemolysis, so that any inhibition of hemolysis may be interpreted as 
specific complement fixation. 
Browning and Mackenzie and Thomsen have devised a technic wffiere- 
by it is possible to estimate the actual amounts of complement absorbed, 
first, by the serum and antigen alone, and second by these two substances 
combined. The complement absorbed is measured in terms of hemolytic 
doses. This method consumes a little more time and more of the various 
reagents is required. It is, nevertheless, a good quantitative method, shows Fig. 130.—Titration of Hemolytic Complement. 
The tube containing 0.4 c.c. of complement is the smallest amount producing complete hemolysis, 
and this amount is the unit. TECHNIC OF THE FIRST METHOD 
469 
exactly the degree of complement fixation in each case, and is especially 
valuable for research work rather than for routine purposes. 
In conducting any complement-fixation test the following are essential 
factors if success is to be achieved: (1) Reliable reagents, particularly a good 
antigen must be had, for no matter how much care is exercised, good results 
cannot be secured with indifferent reagents; (2) the observer must possess 
a thorough working understanding of the underlying principles and par- 
ticularly of the quantitative relations of the various reagents; (3) there 
must be an accurate adjustment of the hemolytic system; (4) he must have 
a careful, painstaking and accurate habit of pipeting small amounts. Accu- 
racy should never be sacrificed for speed, as the latter is properly acquired 
only with experience. 
Technic of the First Method 
THE ORIGINAL WASSERMANN REACTION 
This is the original Wassermann reaction, except that an alcoholic, in- 
stead of an aqueous extract of syphilitic liver, is used as antigen. This 
is the simplest of all technics, and, when properly performed, constitutes, 
in the final analysis, a reliable test and one especially adapted for those not 
constantly engaged in this work. 
1. Complement.—Fresh clear serum (not over twenty-four hours old) 
of a healthy guinea-pig. Dilute 1 : 10 by adding 9 c.c. of sterile normal 
saline solution to each 1 c.c. of serum. Dose, 1 c.c. (=0.1 c.c. of undiluted 
serum). 
2. Corpuscles.—Sheep’s blood washed three times and diluted to make 
a 5 per cent, suspension. For example, 1 c.c. of corpuscles in 19 c.c. of salt 
solution makes up sufficient for a number of tests. 
3. Hemolytic Amboceptor.—Serum of a rabbit immunized with washed 
sheep’s corpuscles. As stated elsewhere, this serum is heated to 55° C. for 
half an hour, and an equal part of chemically pure glycerin is added. Mix 
wTell and preserve in sterile 1 c.c. ampules. Each ampule will, therefore, 
contain 0.5 c.c. of serum. One stock dilution is prepared in such manner 
that about 0.2 c.c. represents one hemolytic unit. One may otherwise 
prepare a whole series of flasks with various dilutions, and in making a 
titration to use 1 c.c. of each dilution; I have found it much more accurate, 
simple, and economical, however, to prepare one stock dilution, which is 
titrated with each complement and corpuscle suspension before each day’s work. 
For example, if a serum is known to have a titer of 1 : 2000, an ampule 
(0.5 c.c. of serum) is diluted with 200 c.c. of salt solution; this gives a dilu- 
tion of serum approximately 1 : 400, of which 0.2 represents one hemolytic 
unit. The titration must be repeated each time to make sure of this, be- 
cause the complement of different pigs may vary in activity, and the chief 
object is to adjust the hemolysin and complement to each other. 
Titration of Hemolysin.—Into a series of six test-tubes place increasing 
amounts of the amboceptor dilution: 0.05, 0.1, 0.15, 0.2, 0.25, and 0.3 c.c. 
Add 1 c.c. of complement (1 : 10) and 1 c.c. of corpuscle suspension to each 
tube, and sufficient salt solution to make the total volume in each tube 
about 4 c.c. Shake gently and incubate for one hour at 37° C. At the end 
of this time the tube showing just complete hemolysis contains one hemo- 
lytic dose, or unit of amboceptor. In the tests double this amount, or two 
units, is used. 
The amboceptor titration is very important. Under no circumstances 
should the same dose be used day after day without titration, because the 
complement of different guinea-pigs may vary in its activity, and these 470 
COMPLEMENT FIXATION IN SYPHILIS 
variations would be detected and would be adjusted in this titration. For 
example, with a weaker complement the dose of amboceptor required to 
effect complete hemolysis becomes higher; each new corpuscle suspension 
may also vary slightly in the actual number of cells contained in 1 c.c., 
but this makes no difference when each suspension is titrated with the 
complement and amboceptor to be used in the day’s work. This titration 
is set up first, and while it is in the incubator, the main tests are arranged. 
4. Antigen.—Alcoholic extract of syphilitic liver or acetone-insoluble 
lipoids of proved value may be used. It is well to estimate just how much 
antigen will be required for the tests on hand, so that no waste will occur, 
as fresh emulsions are better than old ones carried over from day to day. 
The dose should be at least double the titrated antigenic unit, or one-fourth 
of the anticomplementary dose. For instance, if an alcoholic extract of 
syphilitic liver diluted 1 : 10 is found on titration to be perfectly antigenic 
in doses of 0.2 c.c., and not anticomplementary in amounts under 2 c.c., 
then 0.4 c.c. may be used in making the tests, as this amount is still about 
five times less than the anticomplementary dose, and well within the range 
of safety against non-specific complement fixation. If 10 tests are to be 
made, then at least 4.4 c.c. of diluted antigen are required, including sufficient 
for the antigen control, or in round numbers, 0.5 c.c. of antigen plus 4.5 
c.c. of salt solution slowly added in order to secure the maximum turbidity. 
5. Serum.—This should be fresh and clear and heated in a water- 
bath to 55° C. for half an hour before using. The temperature should not 
go above 56° C. nor below 55° C. Dose, 0.2 c.c. 
6. Cerebrospinal Fluid.—This should be fresh and free from blood. 
It is used unheated, as spinal fluid contains little or no hemolytic comple- 
ment. The dose should be at least four times that of the serum, or 0.8 c.c. 
The Test.—A front and a rear tube for each serum are placed in a rack. 
Each tube is marked plainly with the patient’s name or initials, and in 
addition the front tube is marked with the number of the antigen or with 
the letter “A,” or the word “antigen” is written on it, the rear tube being 
marked “control” (serum control). The necessity for carefully marking 
each tube is nowhere more important than in conducting Wassermann 
reactions with a number of serums, as the slightest error or lapse of memory 
may result in confusion and prove to be quite a serious matter. 
In each series of reactions the serum from a known case of syphilis that 
has given a positive reaction and the serum of a known non-syphilitic per- 
son are included as positive and negative controls respectively. 
Into each front tube the proper dose of antigen is placed; to the front 
and rear tubes 0.2 c.c. of the patient’s serum is added. To all tubes 1 c.c. 
of the complement (1 : 10) and sufficient normal salt solution are then 
added to bring the total volume in each to about 3 c.c. 
The rear tube of each set is the serum control; the positive and negative 
serums are treated in just the same manner as the patient’s serum. In 
addition to these there are three other important controls that should not 
be omitted: 
1. The antigen control: Dose of antigen plus 1 c.c. of complement (1 : 
10) and a sufficient quantity of salt solution. 
2. Hemolytic system control: 1 c.c. of complement (1 : 10) and 2 c.c. of 
salt solution. 
3. Corpuscle control: 1 c.c. of corpuscle suspension plus 3 c.c. of salt 
solution. 
Each tube is gently shaken and incubated at 37° C. for an hour, when 
two units of amboceptor and 1 c.c. of corpuscle suspension (5 per cent.) Fig. 131.—Wassermann Reaction (First Method). 
Fig. 132.—Reading the Wassermann Reaction. TECHNIC OF THE FIRST METHOD 
471 
are added to each tube except the corpuscle control. Tubes are shaken 
and reincubated for an hour or an hour and a half, depending upon the 
hemolysis of the serum controls, after which a preliminary reading is made 
and recorded. With partially positive reactions the tubes may be centri- 
fuged in order to read the relative amounts of hemolysis, and the final 
reading made at once, or the tubes may be placed in the refrigerator (just 
above freezing-point) and the final readings made the next morning. 
Unknown Serum, 
Mr. B. 
Unknown Cere- 
brospinal Fluid, 
Mr. C. 
Known Positive 
Syphilitic 
Serum. 
Known Negative 
Normal Serum. 
Controls. 
2. 
Serum, 0.2 c.c. 
+ • 
Complement (1 
c.c. of 1 :10) +. 
Salt solution (q. s. 
3 c.c.). 
'4. _ 
Cerebrospinal 
fluid, 0.8 c.c. +• 
Complement (1 
c.c. of 1 :10) +• 
Salt solution (q. 
s. 3 c.c.). 
6. 
Serum, 0.2 c.c. 
+• 
Complement (1 
c.c. of 1 :10) +. 
Salt solution (q. 
s. 3 c.c.). 
8. 
Serum, 0.2 c.c. 
+• 
Complement (1 
c.c. of 1 : 20) +. 
Salt solution (q. 
s. 3 c.c.). 
10. 
Antigen control: 
Antigen, 0.4 c.c. 
+• 
Complement (1 
c.c. of 1 :10) +. 
Salt solution (q. 
s. 3 c.c.). ’ 
1. 
Antigen, 0.4 c.c. 
+• 
Serum, 0.2 c.c. 
+• 
Complement (1 
c.c. of 1 :10) +• 
Salt solution (q. 
s. 3 c.c.). 
3. 
Antigen, 0.4 c.c. 
+• 
Cerebrospinal 
fluid, 0.8 c.c. 
+• 
Complement (1 
c.c. of 1 :10) +. 
Salt solution (q. 
s. 3 c.c.). 
5. 
Antigen, 0.4 c.c. 
+• 
Serum, 0.2 c.c. 
+• 
Complement (1 
c.c. of 1 :10) +. 
Salt solution (q. 
s. 3 c.c.). 
7. 
Antigen, 0.4 c.c. 
+ • 
Serum, 0.2 c.c. 
Complement (1 
c.c. of 1 : 20) +. 
Salt solution (q. 
s. 3 c.c.). 
9. 
Hemolytic con- 
trol. 
Complement (1 
c.c. of 1 ; 10) +. 
Salt solution (q. 
s. 3 c.c.). 
Scheme for Conducting aWassermann Reaction (First Method). (See Fig. 131.) 
Tubes are shaken gently and incubated at 37° C. for an hour, after which two hemolytic 
doses of amboceptor and 1 c.c. of corpuscle suspension are added to each. They are then 
gently shaken and reincubated for an hour or an hour and a half, after which a preliminary 
reading is made. 
All the tubes in the rear row (upper row in table) (serum controls), the antigen and 
hemolytic system controls, and the front tube with the negative normal serum, are com- 
pletely hemolyzed. The front tube with the unknown serum and cerebrospinal fluid and the 
positive serum control show inhibition of hemolysis or positive reactions. 
This scheme illustrates the technic employed with an unknown serum 
and cerebrospinal fluid. The proper dose of diluted antigen is taken as 
0.4 c.c., and two doses of hemolytic amboceptor determined by titration 
as equivalent to 0.4 c.c. of the stock dilution. 
Reading and Recording the Wassermann Reaction.—1. The hemo- 
lytic system control is inspected first. It should show complete hemolysis, 
indicating that the complement and amboceptor were active and have been 
used in sufficient amounts. If a few corpuscles are found in the bottom 
of the tube, some error in pipeting has probably occurred, too many cor- 
puscles or too little complement or amboceptor having been introduced. 
2. The corpuscle control should show no hemolysis, indicating that 
the solution is isotonic and that the corpuscles are not unduly fragile. 
3. The antigen control should show complete hemolysis, indicating 
that the dose used was not anticomplementary. If this tube shows incom- 
plete hemolysis, due to the anticomplementary action of the antigen, all 
the front two tubes will also show some inhibition of hemolysis, due to 
this non-specific complement fixation, and it is necessary to repeat the 
tests with another extract. 472 
COMPLEMENT FIXATION IN SYPHILIS 
4. The rear tubes of all serums should be completely hemolyzed, indi- 
cating that the serums were practically free from anticomplementary action 
as previously stated, most antigens and serums are usually very slightly 
anticomplementary if small amounts of complement are used with a close 
single unit of amboceptor, but in this technic the complement and 2 units 
of amboceptor are sufficient, under ordinary circumstances, to offset this 
influence. If, however, a serum is more than normally anticomplementary, 
the rear tube will show some inhibition of hemolysis, and, of course, in the 
front tube a similar inhibition, and probably to a greater degree, will be 
seen. If the serum is very slightly anticomplementary and the front tube 
shows complete inhibition of hemolysis, the reaction is in all probability 
positive. If the rear tube, however, shows marked inhibition of hemol- 
ysis, indicating that it is highly anticomplementary, the result cannot be 
determined, but a retest with fresh serum must be made. This indicates 
the great importance of the “serum control,” and it may be stated that a test 
should never be made without it. 
5. The front tube containing the known syphilitic serum should show 
inhibition of hemolysis, indicating that the extract possesses antigenic 
properties. 
As the complement is “fixed” by the syphilis antibody and extract, 
hemolysis could not occur when the corpuscles and amboceptor were added. 
If a portion of the complement is fixed by antibody and extract, then the 
unfixed portion will hemolyze some of the corpuscles, the reaction being 
moderately positive, slightly positive, etc., depending upon the degree of 
hemolysis that takes place. This illustrates the importance of observing 
exactness in pipeting, and the great influence of quantitative factors in 
testing for the Wassermann reaction, for if an excess of complement is used, 
there may be sufficient for all the syphilis antibody, and enough unbound 
complement to hemolyze all the corpuscles. In this manner a false negative 
reaction will result. Corpuscles and sufficient hemolytic amboceptor are 
added merely in order to test for any free complement. Under proper 
conditions a total lack of hemolysis indicates that there is no free comple- 
ment, but that it has been fixed by syphilis antibody and extract, constitut- 
ing a positive reaction (+ + + +). Complete hemolysis indicates that 
complement was not bound and that syphilis antibody was, therefore, 
absent from the fluid tested—a negative reaction (—). Partial hemolysis 
indicates that a portion of the complement has been fixed by smaller amounts 
of syphilis antibody and of the extract, yielding partially positive reactions 
(H—I—b; H—b; +; =*=)• 
6. The front tube containing the known normal serum should show 
complete hemolysis because, in the absence of syphilis antibody, the com- 
plement remains free to hemolyze the corpuscles with the hemolytic ambo- 
ceptor. 
7. Various methods have been proposed for recording the results of 
hemolytic tests. The following scheme, after Citron, is widely used (Fig. 
132). 
+ + + + = complete inhibition of hemolysis = strongly positive. 
+ + + = 75 per cent, inhibition of hemolysis = moderately positive. 
+ + = 50 per cent, inhibition of hemolysis = weakly positive. 
+ = 25 per cent, inhibition of hemolysis = very weakly positive. 
=»= = less than 25 per cent, inhibition of hemolysis = delayed hemol- 
ysis or doubtful reaction. 
— = complete hemolysis = negative reaction. TECHNIC OF THE SECOND METHOD 
473 
Under the third method a scale is given that is easily prepared for mak- 
ing these readings. However, after some experience they are readily made, 
and at first should be attempted only after the non-hemolyzed corpuscles 
have been centrifuged or allowed to settle to the bottom of the tube. As 
stated elsewhere, this method is not an accurate measure of the amount of 
syphilis antibody, but constitutes a relative and convenient gage of value 
within certain limits. In reporting reactions to the clinician, the plus signs 
should not be used, or if used, should be interpreted by the terms “strongly 
positive,” “weakly positive,” etc. 
Technic of the Second Method 
AUTHOR’S MODIFICATION OF THE WASSERMANN REACTION WITH MULTIPLE ANTIGENS 
Practically the same technic is used in this as in the first method, except 
that three different antigens, instead of one, are used with each serum, for 
the reasons previously stated; for economy the amounts of each reagent 
are just one-half those originally employed. 
This method can be strongly recommended, as it is simple, accurate, 
and reliable. Although a little more work is demanded and a larger quantity 
of the various reagents is required, the results warrant the expenditure 
of a little more labor, and the second objection is readily overcome by using 
half the quantities prescribed in the original Wassermann technic, as given 
in the first method. 
1. Complement.—Fresh clear sera of two or more guinea-pigs collected 
as previously described. If only a small number of tests are to be done suffi- 
cient blood may be obtained by bleeding two or more pigs from the heart 
as described on p. 37. Dilute the serum 1 : 20 (1 c.c. serum + 19 c.c. 
saline solution). 
2. Corpuscles.—Sheep’s corpuscles washed three times and made up 
into a 2.5 per cent, suspension with saline solution as previously described. 
Hemolysin.—Antisheep hemolysin titrated each time the tests are con- 
ducted. 
The serum is so diluted that the unit will be approximately 0.1 to 0.2 
c.c. In a series of 6 test-tubes place increasing amounts of this stock solu- 
tion as follows: 0.05, 0.1, 0.15, 0.2, 0.25, and 0.3 c.c. Add 1 c.c. comple- 
ment 1 : 20 and 1 c.c. of 2.5 per cent, corpuscle suspension. Mix gently 
and place in a water-bath at 38° C. for one hour. The unit is the smallest 
amount showing complete hemolysis and 2 units are employed for the 
antigen titrations and main tests. 
In conducting routine tests by this method the hemolysin titration may 
be put up last and incubated along with the primary incubation of the 
main tests; at the end of the hour the unit is read and 2 units added to the 
tubes of the main test. 
3. Antigens.—I generally use the following three antigens: (1) A choles- 
terinized alcoholic extract of human heart; (2) alcoholic extract of syphilitic 
liver or alcoholic extract of heart; (3) acetone-insoluble lipoids. 
As previously stated, these extracts are used in amounts equal to from 
two to four times their titrated antigenic unit, providing these doses are at 
least four times smaller than the anticomplementary units. The amount 
of each antigen required for the wrork at hand is calculated, placed in test- 
tubes, and slowly diluted with the requisite amount of salt solution to 
secure maximum turbidity of the emulsions. 
1. Anticomplementary Titration.— The antigen extract is diluted 1 : 10 by 
placing 1 c.c. in a test-tube and slowly adding 9 c.c. of normal salt solution 
or 1 : 20 (1 c.c. antigen + 19 c.c. saline). 474 
COMPLEMENT FIXATION IN SYPHILIS 
Increasing amounts of this emulsion are placed in a series of test-tubes: 
0.2, 0.4, 0.6, 0.8, 1, 1.2, 1.5, and 2 c.c. To each tube are now added 1 c.c. 
of the diluted complement serum ( = 0.05 c.c. undiluted serum), and sufficient 
normal salt solution to bring the total volume up to 3 c.c. Shake each tube 
gently and incubate for one hour at 37° C. Then add to each tube 1 c.c. 
of the corpuscle suspension and a dose of amboceptor equal to 2 units, 
as just determined by previous titration. Shake gently and reincubate for 
another hour and a half, when a preliminary reading of the results may be 
made. That amount oj antigen that shows beginning inhibition of hemolysis 
is regarded as the anticomplementary unit. The final readings are made 
after the tubes have stood over-night in a refrigerator at low temperature 
(Fig. 133). 
This titration may also be made in the presence of normal serum, al- 
though this is not absolutely necessary. The serum must be perfectly fresh, 
and must be that from a person known to be free from lues. It is inactivated 
by heating to 55° C. for half an hour, and 0.2 c.c. is added to each tube. 
Complement and salt solution are now added, and the titration conducted 
in the manner just described. Normal serum may absorb a small amount 
of complement in itself, and hence a titration conducted with serum may 
show a slightly lower anticomplementary dose. 
The following controls are included: 
1. A hemolytic system control, containing the complement, corpuscles, 
and amboceptor in the same amounts as were used in conducting the titra- 
tion. This control should show complete hemolysis. 
2. A serum control, which is the same as the hemolytic system control 
plus 0.2 c.c. of the serum. This should show complete hemolysis, and indi- 
cates that the serum was not anticomplementary. This control test should 
never be omitted. 
3. A corpuscle control, including 1 c.c. of the corpuscles in salt solution. 
This tube should show no hemolysis. 
The following table gives the results of a titration with an alcoholic ex- 
tract of syphilitic liver diluted 1 : 10 (see Fig. 133). 
Anticomplementary Titration of a Tissue Extract 
Tube. 
Antigen 
(1 : 10), 
C.c. 
Comple- 
ment 
0 : 20), 
C.c. 
Normal 
Serum, 
C.c. 
43 
S 
"n ° 
.S3 § 
2xi 
u 2 
Anti- 
sheep 
Hemol- 
ysin 
Units. 
Sheep’s 
Corpus- 
cles 
2.5 Per 
Cent., 
C.c. 
Results. 
1.... 
0.2 
1 
0.2 
a 
jj 
2 
1 
Hemolysis. 
2.... 
0.4 
1 
0.2 
2 
1 
Hemolysis. 
3.... 
0.6 
1 
0.2 
2 
1 
Hemolysis. 
4.... 
0.8 
1 
0.2 
o S 
2 
1 
Hemolysis. 
5.... 
1.0 
1 
0.2 
2 
1 
Hemolysis. 
6.... 
1.2 
1 
0.2 
n 
2 
1 
Slight inhibition of hemol- 
7.... 
1.5 
1 
0.2 
S3 
'o 02 
2 
1 
ysis. 
Marked inhibition of hemol- 
8.... 
2.0 
1 
0.2 
C/3 C/3 
43 
73 =! 
2 
1 
ysis. 
Complete inhibition of hemol- 
9.... 
Control. 
1 
C/3 **-* 
2 
1 
ysis. 
Hemolytic control: complete 
to.... 
Control. 
1 
0.2 
43 o 
So 
2 
1 
hemolysis. 
Serum control: complete he- 
in 
molysis. Fig. 133.—Titration of Antigen for Anticomplementary Unit. 
In Fig. 133 the tube containing 1.5 c.c. of antigen shows slight inhibition of hemolysis, and this 
amount is the anticomplementary unit. 
In Fig. 134 the tube containing 0.15 c.c. of antigen is the smallest amount producing complete 
inhibition of hemolysis, and is the antigenic unit. 
Fig. 134.—Titration of Antigen for Antigenic Unit. TECHNIC OF THE SECOND METHOD 
In this titration tube No. 6, containing 1.2 c.c. of the antigen emulsion 
showed beginning inhibition of hemolysis and was recorded as the anti- 
complementary dose. 
2. Hemolytic Titration.—As previously mentioned, organic extracts are 
capable in themselves of hemolyzing red cells; this is due to the hemotoxic 
action of lipoids and alcohol. Extracts of organs that have undergone 
advanced autolysis and decomposition are very likely to be hemolytic. 
Serum exerts an inhibiting influence on the lytic action of an organic 
extract. Hence the hemolytic dose of an extract depends largely on whether 
or not complement serum is used in the titration. 
When an organic extract is titrated in the presence of complement, the 
hemolytic dose is higher than the anticomplementary dose. In the fore- 
going titration 3 c.c. of the extract emulsion showed beginning hemolysis, 
and when 4 c.c. was used, hemolysis was complete. These large amounts 
of emulsion give the tube contents quite a milky appearance, but close 
inspection shows that all the cells are broken up. 
As a general rule, the hemolytic titration is not absolutely necessary. 
It may be conducted with the anticomplementary titration by adding 
another tube or two to the foregoing series, with higher doses of extract; 
or this titration may be conducted separately, and without complement 
hemolysin, by using the same doses of antigen with 1 c.c. of corpuscle sus- 
pension and sufficient salt solution to bring the total volume in each tube 
up to 3 or 4 c.c. 
3. Antigenic Titration.—As previously stated, this titration is not abso- 
lutely necessary,, as one-fourth the anticomplementary dose of an extract 
may be used in the main test. For instance, in the foregoing titration 0.3 
or 0.4 c.c. may safely be used in making the test for the syphilitic reaction. 
Different extracts vary, however, in their antigenic value. Some may be 
highly anticomplementary and have a comparatively low antigenic value; 
purer extracts, such as acetone-insoluble lipoids or cholesterinized alcoholic 
extracts of heart, are largely free from anticomplementary action, and at 
the same time possess a high antigenic value. It is advisable, therefore, to 
use an antigen whose full antigenic as well as anticomplementary doses are 
known, for, while it is necessary to use sufficient antigen, it is not advisable 
to use a larger amount than is necessary. 
For this titration all antigens except alcoholic extracts of syphilitic 
liver, should be diluted 1 : 20 with normal salt solution. Usually the anti- 
genic unit is so much lower than the anticomplementary unit that it is best 
determined with a more dilute antigen. 
The titration is conducted in a manner similar to the anticomplementary 
titration except that 0.2 c.c. of fresh and inactivated serum from a known 
and untreated syphilitic person is added to each tube. Increasing doses of 
antigen, patient’s serum, and complement are mixed, shaken, and incubated 
for one hour. Two units of hemolysin and corpuscles are then added, the 
tubes shaken and incubated for another hour, after which the preliminary 
reading is made. The final reading is taken after the tubes have been placed 
over night in a refrigerator at low temperature. That amount of antigen 
that shows just complete inhibition of hemolysis is taken as the antigenic unit 
(Fig. 134). In conducting the syphilis reaction two to four times this unit 
is used, providing that these amounts are at least four or five times less than 
the anticomplementary dose. This larger antigenic dose is advisable, because 
the exact unit may not be sufficient with serums containing but small 
amounts of syphilis antibody such as those of treated or long-standing 
cases of lues. 
475 476 
COMPLEMENT FIXATION IN SYPHILIS 
The following table illustrates this titration with the same alcoholic 
extract of syphilitic liver (see Fig. 134): 
Antigenic Titration of a Tissue Extract 
Tube. 
Anti- 
gen 
1 : 10, 
C.c. 
Syphi- 
litic 
Serum 
(Inactive) 
C.c. 
Com- 
ple- 
ment 
1 : 20, 
C.c. 
a s 
o 
> g 
a O 
3 ”8 
<U rt 
■5 
-- 
t>C cj 
Anti- 
sheep 
Hem- 
olysin, 
Doses. 
Sheep 
Cor- 
puscles 
2.5 
Per 
Cent., 
C.c. 
i 
Results. 
d 
o 
a> 
d 
1 
0.05 
0.2 
1 
2 
1 
° sb 
”8.5 Slight inhibition of he- 
2 
0.01 
0.2 
1 
C 
o £5 
2 
1 
nS molysis. 
'a £ Marked inhibition of 
3 
0.15 
0.2 
1 
a 
P 
2 
1 
a >> hemolysis. 
5 Complete inhibition of 
4 
0.2 
0.2 
1 
o 
2 
1 
e ‘g hemolysis; unit. 
~ a No hemolysis. 
5 
0.25 
0.2 
1 
-4_> <D 
2 
1 
§ No hemolysis. 
6 
0.3 
0.2 
1 
m £ 
2 
1 
H No hemolysis. 
7 
Control. 
0.2 
1 
d T 
2 
1 
Serum control: hemol- 
8 
Control. 
0 
1 
#<L> CJ 
*2 
bs co 
2 
1 
s ysis. 
Hemolytic control: he- 
cnS 
H molysis. 
In this instance 0.15 c.c. of the emulsion represents the antigenic unit. 
In performing the Wassermann reaction 0.3 or 0.4 c.c. was used, and these 
amounts were about one-fourth the anticomplementary dose. 
It is not unusual to find cholesterinized alcoholic extract and acetone- 
insoluble lipoids perfectly antigenic in 0.05 c.c. of a 1 : 20 dilution, and 
not anticomplementary under 1 or 2 c.c. of a 1 : 10 dilution. In these in- 
stances four times the antigenic dose, or 0.2 c.c. can be used, and yet this 
amount is at least ten times smaller than the anticomplementary dose— 
a condition of affairs that constitutes a safe and desirable antigen. 
Each new antigen should be tested with a number of serums and con- 
trolled by an older antigen of known value before being finally accepted 
as satisfactory. 
Antigen containers should be well stoppered and kept in the refrigerator. 
Deterioration may set in suddenly, and they should, therefore, be retitrated 
every few weeks. 
4. Serum.—Heated in a water-bath to 55° C. for thirty minutes; dose 
0.2 c.c. This amount is placed in each of the three tubes carrying the anti- 
gens and in the serum control. 
5. Cerebrospinal Fluid.—Used unheated. Dose, 0.8 c.c. with each 
antigen; 1.0 c.c. may be used in the control. 
The Test.—For each serum and spinal fluid four test-tubes are arranged 
in a row and marked with the patient’s name or initials. The first tube is 
marked “C. H.,” and receives the cholesterinized heart extract; the second 
tube is marked “S” for the alcoholic extract of syphilitic liver or plain 
alcoholic extract of heart; the third is marked “A” for acetone-insoluble 
lipoids, and the fourth is not marked at all or simply marked with the 
letters “S. C.” (serum control). 
To each of the four tubes 0.2 c.c. of the patient’s serum is added, or 
0.8 c.c. of cerebrospinal fluid. 
To each tube 1 c.c. of the diluted complement (1 : 20) and sufficient 
salt solution to bring the total volume in each up to 3 c.c. are now added. Fig. 135.—Wassermann Reaction (Second Method). 
Shows a + + + + reaction with the cholesterinized extract (C. H.); a + reaction with the alcoholic ex- 
tract of syphilitic liver (S). and a + + reaction with the extract of acetone-insoluble lipoids (A). TECHNIC OF THE SECOND METHOD 
477 
Controls.—A known positive and negative serum should be included, un- 
less one is performing a large number of tests with reliable antigens every 
week, in which case, among many serums, a few at least are likely to be 
positive. Under these circumstances these controls may be omitted; as 
a general rule, however, they should be included. 
To the hemolytic system control tube 1 c.c. of complement dilution, and 
2 c.c. of salt solution are now added. Three antigen control tubes are set up 
for each antigen with the dose employed, plus 1 c.c. of complement dilution 
and sufficient salt solution to make the total volume about 3 c.c. The 
corpuscle control receives 1 c.c. of the suspension plus 3 c.c. of salt solution. 
All the tubes are shaken gently and placed in a water-bath for an hour 
at 37° C. Instead of the water-bath an incubator at 38° C. may be em- 
ployed for one hour (not less) for the primary and secondary periods of 
incubation. At the end of primary incubation 2 units of the amboceptor 
and 1 c.c. of the corpuscles are added to each tube except that containing 
the corpuscle control. Each tube is shaken gently and reincubated for an 
hour or longer, depending upon the hemolysis of the controls, when the 
readings are made. By making the readings at this time the influence of 
an excess of hemolysin due to the presence of natural antisheep hemolysin 
in the human sera is avoided and lesser degrees of complement fixation 
detected, which may become completely hemolyzed if the tubes are set 
aside over night. 
Reading the Results.—The readings are made in the same manner as 
described in the first method, the controls always being inspected first. 
The hemolytic, antigen, and serum controls and known negative serum 
tubes should all be hemolyzed. The antigen tubes containing the positive 
syphilitic serum should not be hemolyzed. Results with the unknown 
serums are dependent upon whether or not the serums are luetic, and if 
they are, upon the quantity of syphilis antibody present. 
With strongly positive serums there is complete inhibition of hemolysis 
with all three antigens. With serums of long-standing or treated cases of 
syphilis containing smaller amounts of antibody the reaction with the 
cholesterinized extract is usually strongly positive, whereas with the other 
two antigens the degree of inhibition of hemolysis is less marked and vari- 
able (see Fig. 135). In from 15 to 20 per cent, of cases the cholesterinized 
extract shows a 50 per cent, or more inhibition of hemolysis, whereas with 
the other two antigens the reactions are negative. In our experience the 
majority of such serums were taken from patients giving a frank history 
of syphilis of many years’ standing and from known cases undergoing treat- 
ment, further therapy being indicated until the reaction finally becomes 
negative when cholesterinized extracts are used. In a small proportion 
of cases a feebly positive reaction of 25 per cent, or less inhibition of hemol- 
ysis may be found with the cholesterinized extract alone. Many of these 
reactions occur with serums of treated cases of syphilis; on the other hand, 
a similar reaction may occur with about 5 per cent, of normal serums, so 
that if the history and clinical conditions are clearly negative, a slight de- 
gree of inhibition of hemolysis (5 to 10 per cent.) with the cholesterinized 
extract and marked hemolysis with the other two antigens may be inter- 
preted as a negative reaction. 
After a new antigen has been prepared and titrated, it should be tested 
out in this manner by placing it in the series along with at least two other 
older antigens of proved value, and used in the examination of a large 
number of serums before it is finally accepted as reliable. 478 
COMPLEMENT FIXATION IN SYPHILIS 
Technic of the Third Method 
AUTHOR’S MODIFICATION OF WASSERMANN REACTION BASED UPON STUDIES IN THE 
STANDARDIZATION OF TECHNIC 
Glassware and Apparatus.—Good pipets are essential; certified pipets are 
best, of course, but it is advisable to have a separate 1 c.c. pipet for each 
serum and the item of expense is apt to be prohibitive. As previously 
stated, the new technic calls for pipeting amounts from 0.2 to 1.0 c.c. in 
order to reduce error due to inaccurate pipets. 
Fig. 136.—A Wire Rack Used by the Author for Complement-fixation Tests; Carries 72 
Tubes. 
One c.c. pipets graduated to the tips are preferred as long as the tips 
are not chipped. 
Five and 10 c.c. pipets divided into 0.5 cm. are also required (Fig. 4). 
The test-tubes should have rounded bottoms, no lips, and measure 85 
mm. in length with an internal diameter of 11 to 13 mm. It is important 
that the diameter be within these limits on account of the color scale. Their 
size is shown in Fig. 140. 
Fig. 137.—A Larger Wire Rack Used by the Author for Complement-fixation Tests; 
Carries 144 Tubes. 
For mixing large volumes, as in the preparation of corpuscle suspen- 
sion or dilutions of complement serum, volumetric flasks rather than the 
ordinary graduated cylinders should be used because of greater accuracy. 
The glass-stoppered, graduated cylinders (50 to 100 c.c. capacity) are more 
convenient, however, for measuring intermediate amounts, and may be 
used if carefully selected on the basis of accuracy in graduations. For TECHNIC OF THE THIRD METHOD 
479 
measuring any amount of fluid under 50 c.c. it is better to use an accurate 
10 c.c. pipet and reserve the graduated cylinders or flasks for measuring 
larger volumes. 
It is imperative that all glassware including new glassware should be 
chemically clean, that is, free of all traces of acids or alkalies and preferably 
sterile; test-tubes do not require cotton plugs, but may be sterilized in baskets 
open ends down. A full description of technic for cleaning test-tubes and 
pipets is given on p. 7. 
Test-tube Racks.—Each test requires six test-tubes. Galvanized wire 
racks (Fig. 136) carrying 12 rows of six tubes each have been found very 
serviceable; also racks carrying 24 rows of six tubes each (Fig. 137). 
Water-bath.—A simple and inexpensive water-bath for heating sera and 
conducting the secondary incubation is shown in Fig. 138. A much simpler 
pan made by any tinsmith is shown in Fig. 139. When carrying water to 
the depth of 8 cm. the temperature of either pan may be maintained at 
55° or 37° C. with very little care and attention. 
Fig. 138.—A Very Simple and Efficient Water-bath Used by the Auihor for the 
Inactivation of Sera and for Secondary Incubation. Size is indicated. (Amer. Jour. Syphilis 
1922, 6, 92.) 
Refrigerator.—Any refrigerator maintaining a temperature of between 
6° and 8° C. suffices for the primary incubation. 
Saline Solution.—Sodium chlorid (0.85 per cent.) in water prepared as 
described on p. 9. 
Corpuscles (Indicator Antigen).—Two per cent, suspension of freshly 
collected and washed sheep corpuscles; dose 0.5 c.c. 
Abattoir blood may be used, but occasionally a worker will encounter 
corpuscles more resistant to hemolysis than average cells; for this reason it is 
better to keep a sheep for furnishing blood, but this is by no means neces- 
sary. Blood preserved with formalin (see p. 13) may be employed under 
certain conditions,1 but freshly collected blood is better. 
Corpuscle suspensions should be fairly uniform and the method de- 
scribed on p. 452 has been found satisfactory; it is not necessary to count 
the erythrocytes or estimate the hemoglobin as some workers advise, in- 
asmuch as the color scale is prepared of each suspension. 
Hemolysin.—Antisheep hemolysin diluted with saline and titrated daily 
1 Amer. Jour. Syph., 1919, 3, 169. 480 
COMPLEMENT FIXATION IN SYPHILIS 
with 0.3 c.c. of 1.30 guinea-pig complement, and 0.5 c.c. of 2 per cent, sheep 
corpuscles. 
Antisheep hemolysin is easily prepared by the immunization of rabbits, 
five intravenous injections of 5 c.c. of 10 per cent, suspensions of washed 
sheep blood every five days being a satisfactory method.1 The animal should 
be bled seven to nine days after the last injection and the serum preserved 
by adding an equal part of best grade neutral glycerin. 
Most serologists employing an antisheep hemolytic system have re- 
ported that the hemolysin requires only an occasional titration; in this 
technic I have not found this to be the case, owing to variation in the hemo- 
lytic activity of complement, variation in the resistance of sheep corpuscles 
to hemolysis and variation in guinea-pig sera. For these reasons I have 
found it necessary to titrate the hemolysin each time the complement-fixation 
tests are conducted using the same complement and corpuscles to he employed 
in the main tests.2 
Fig. 139.—A Large Water-bath Used by the Author for the Secondary Incubation in 
Complement-fixation Tests. Size is indicated. (Amer. Jour. Syphilis, 1922, 6, 92.) 
Complement.—Dilution (/ : 30) of the mixed sera of several healthy guinea- 
pigs. 
The complement serum for the fixation test must be: (1) Highly sensitive 
to fixation by antibody and antigen; (2) possess a high degree of hemolytic 
activity for the erythrocytes of the indicator antigen, and (3) be free, or 
largely so, of agglutinins and hemolysins for the cells of the indicator antigen, 
Studies of the complements of different animals have shown that guinea- 
pig serum is best, and a mixture of the sera of three or more pigs should he 
used as described on p. 447. 
Dilute 0.2 c.c. of serum with 5.8 c.c. of saline solution (1 : 30); this is 
sufficient for the hemolysin and complement titrations (these require a 
total of 5.7 c.c.). The balance of serum should be placed in the refrigerator 
and diluted later (described under complement titration) for the main 
tests. 
1 Amer. Jour. Syph., 1920, 4, 484. 
2 Ibid., 1920, 4, 616. TECHNIC OF THE THIRD METHOD 
481 
Titration of Hemolysin.—1. Arrange a series of ten test-tubes and place 
0.5 c.c. of varying dilutions of hemolysin in each tube respectively. 
2. Ordinarily a range of dilutions from 1 : 1000 to 1 : 16,000 is sufficient, 
but depending upon the hemolytic activity of the complement and resistance 
of the corpuscles higher or lower dilutions may be required. A 1 : 100 
dilution preserved with phenol against bacterial contamination may be 
prepared as follows and kept in a refrigerator for several weeks, from which 
the higher dilutions are prepared as needed: 
Glycerolized serum, 2.0 c.c. 
Saline solution, 94.0 c.c. 
Five per cent, phenol solution, 4.0 c.c. 
3. The dilutions are prepared as follows in a separate set of large test- 
tubes: 
0.2 c.c. of 1 : 100 + 1.8 c.c. saline = 1 : 1,000. 
0.2 c.c. of 1 : 100 + 3.8 c.c. saline = 1 : 2,000. 
0.2 c.c. of 1 : 100 + 5.8 c.c. saline = 1 : 3,000. 
0.2 c.c. of 1 : 100 + 7.8 c.c. saline = 1 : 4,000. 
0.2 c.c. of 1 : 100 + 9.8 c.c. saline = 1 : 5,000. 
0.5 c.c. of 1 : 3000 + 0.5 c.c. saline = 1 : 6,000. 
0.5 c.c. of 1 : 4000 + 0.5 c.c. saline = 1 : 8,000. 
0.5 c.c. of 1 : 5000 + 0.5 c.c. saline = 1 : 10,000. 
0.5 c.c. of 1 : 6000 + 0.5 c.c. saline = 1 : 12,000. 
0.5 c.c. of 1 : 8000 + 0.5 c.c. saline = 1 : 16,000. 
Mix contents of each tube very thoroughly. 
4. To each tube, carrying 0.5 c.c. of these various dilutions of hemolysin, 
add 0.3 c.c. of 1 : 30 dilution of the same complement and 0.5 c.c. of a 2 
per cent, suspension of the same corpuscles as used in the complement- 
fixation tests; add 1.7 c.c. saline to each tube to make the total volume in 
each 3 c.c. 
5. Mix the contents of each tube and place in the water-bath at 38° C. 
for one hour; the unit is the highest dilution of hemolysin showing just com- 
plete hemolysis. Two units are employed in the titration of complement 
and antigen and in the complement-fixation tests. 
The following table shows the ensemble and results of a titration: 
Tube. 
Hemolysin. 
Comple- 
ment, 
1: 30. 
Corpus- 
cles, 2 
Per Cent. 
Saline. 
After Water-bath Incubation for 
One Hour. 
1.... 
0.5 c.c. 1 : 1,000 
0.3 c.c. 
0.5 c.c. 
1.7 c.c. 
Complete hemolysis. 
2.... 
0.5 c.c. 1 : 2,000 
CC 
CC 
CC 
Complete hemolysis. 
3.... 
0.5 c.c. 1: 3,000 
CC 
cc 
cc 
Complete hemolysis. 
4.... 
0.5 c.c. 1 : 4,000 
cc 
cc 
cc 
Complete hemolysis. 
5.... 
0.5 c.c. 1 : 5,000 
cc 
cc 
cc 
Complete hemolysis. 
6.... 
0.5 c.c. 1 : 6,000 
cc 
cc 
cc 
Complete hemolysis; unit. 
1.... 
0.5 c.c. 1 : 8,000 
cc 
cc 
“ 
Marked hemolysis. 
8.... 
0.5 c.c. 1 : 10,000 
CC 
cc 
cc 
Marked hemolysis. 
9. ... 
0.5 c.c. 1 :12,000 
cc 
“ 
cc 
Slight hemolysis. 
10.... 
0.5 c.c. 1 : 16,000 
cc 
cc 
cc 
No hemolysis. 
Titration of Hemolysin 
In the above titration the unit was 0.5 c.c. of 1 : 6000; 2 units were 
contained in 0.5 c.c. of 1 : 3000, etc. 482 
COMPLEMENT FIXATION IN SYPHILIS 
6. Sufficient hemolysin is now prepared so that each 2 units are contained 
in 0.5 c.c. 
Titration of Complement.—1. Arrange ten test-tubes and place 0.1, 
0.15, 0.2, 0.25, 0.3, 0.35, 0.4, 0.45, and 0.5 c.c. of 1 : 30 complement in 
each respectively; the tenth tube serves as a corpuscle control for this and 
the hemolysin titration. 
2. Into each of the first nine tubes place 10 units of antigen so diluted 
that this amount is contained in 0.5 c.c. 
3. Place sufficient saline solution in each tube to make the total volume 
about 2 c.c. 
4. Mix contents of each tube and place in a water-bath at 38° C. for 
one hour. 
5. Add 0.5 c.c. hemolysin (2 units) and 0.5 c.c. corpuscle suspension 
(2 per cent.) to each tube; mix and re-incubate one hour. 
6. Ordinarily the smallest amount of 1 : 30 complement giving com- 
plete hemolysis is taken as the unit, but experience has shown that with the 
method of primary incubation employed in the complement-fixation tests, 
namely, fifteen to eighteen hours at 6° to 8° C., this is insufficient; in this test 
the unit is taken as the amount of complement in the next higher tube. For 
example, if hemolysis is just complete with 0.25 c.c. the unit is taken as 
0.3 c.c. and double this amount for the antigen titrations and complement- 
fixation tests. For convenience I have designated this amount as two full 
units of complement} 
The following table shows the ensemble, the results of a titration, and 
the method of reading: 
Titration or Complement 
Tube. 
Comple- 
ment. 
Antigen 
(10 Units). 
Saline. 
Hemolysin 
(2 Units). 
Corpus- 
cles (2 
Per Cent.). 
After Water-bath Incu- 
bation for One Hour. 
1.... 
0.1 c.c. 
0.5 C.C. 
1.4 c.c. 
0.5 c.c. 
0.5 c.c. 
No hemolysis. 
2.... 
0.15 c.c. 
<< 
1.4 c.c. 
U 
<< 
Slight hemolysis. 
3.... 
0.2 c.c. 
U 
1.3 c.c. 
tt 
U 
Marked hemolysis. 
4.... 
0.25 c.c. 
U 
1.3 c.c. 
u 
it 
Complete hemolysis; the 
exact unit. 
5.... 
0.3 c.c. 
n 
1.2 c.c. 
u 
u 
Complete hemolysis the 
full unit. 
6.... 
0.35 c.c. 
u 
1.2 c.c. 
u 
u 
Complete hemolysis. 
7.... 
0.4 c.c. 
1.1 c.c. 
Complete hemolysis. 
8.... 
0.45 c.c. 
u 
1.1 c.c. 
(C 
it 
Complete hemolysis. 
9.... 
0.5 c.c. 
a 
1.0 c.c. 
ci 
u 
Complete hemolysis. 
10.... 
2.0 c.c. 
No hemolysis. 
In practice, the hemolysin titration may he placed in the water-bath at the 
same time as the complement titration; at the end of the first incubation of the 
complement titration the unit of hemolysin is available and 2 units added to 
all tubes of the complement titration, etc. 
7. Each 2 full units of complement are diluted with sufficient solution to 
make 1 c.c., called the dose of complement: for example, if 0.3 c.c. of 1 : 30 
complement is the full unit, the dose is 0.6 c.c. 
1 Experience has shown that in this test the reactions are unsatisfactory if less than 
0.4 c.c. of 1 :30 complement is employed; occasionally hyperactive sera yield a unit 
with 0.1 c.c. of 1 :30, but when this occurs it is necessary to use 0.4 c.c. for the dose of 
complement. TECHNIC OF THE THIRD METHOD 
483 
A convenient scheme for diluting the dose of complement to 1 c.c. is 
as follows: Divide 30 by the dose = the dilution to employ in amount of 
1 c.c. For example: 
Exact unit = 0.25 c.c. of I : 30 dilution. 
Full unit = 0.3 c.c. of 1 : 30 dilution. 
Dose = 0.6 c.c. of 1 : 30 dilution. 
— = 1 • 50 
0.6 
If 75 doses of complement were to be provided (sufficient for testing 
12 sera or spinal fluids) this would require 75 c.c. of 1 : 50 prepared by 
diluting 1.5 c.c. guinea-pig serum with 73.5 c.c. of saline solution. 
Titration of Antigen.—The antigen employed in the complement-fixation 
test for syphilis introduces the most important single factor of variation; 
the adoption of a certain kind of antigen fulfilling certain requirements 
is the most important factor in relation to standardization of technic. 
In my opinion an antigen should be an alcoholic extract of a fresh tissue 
and preferably heart-muscle reinforced with 0.2 per cent, cholesterin1; 
this amount of cholesterin “stabilizes” the extract and greatly increases 
antigenic activity with very slight or no increase of anticomplementary 
activity and without increasing the chances for non-specific positive reac- 
tions with heated normal human sera. A superior antigen in my experi- 
ence is that previously described on p. 458, in which a mixture of dried 
powder of several heart-muscles are used. This powder is first extracted 
with ether, then with alcohol and again with alcohol; the second alcohol 
constitutes the base of the antigen which is reinforced with 0.2 per cent, 
cholesterin and all the acetone-insoluble lipoids recovered from the ether 
and primary alcoholic extracts. 
Whatever antigen is employed it should be used in a dose of 10 antigenic 
units, and this amount should be at least six times less than the anticomple- 
mentary and hemolytic units. A plan for establishing a uniform unit of anti- 
gen for use in a standardized test has been developed.2 
Antigen should be carefully preserved3 and titrated at least once a 
month unless it shows evidences of losing in antigenic activity or acquiring 
increased anticomplementary activity. 
Antigen should be diluted by placing the required amount of physiologic 
saline in a test-tube or Erlenmeyer flask and adding the required amount of 
extract drop by drop or in amounts of 0.1 c.c. and shaking by rotating after 
each addition.4 
In a series of ten test-tubes prepare the following dilutions of antigen: 
1.0 c.c. antigen to 3.0 c.c. saline =1:4. 
1.0 c.c. antigen to 4.0 c.c. saline =1:5. 
0.5 c.c. antigen to 2.5 c.c. saline =1:6. 
1.5 c.c. antigen 1 : 4 to 1.5 c.c. saline =1:8. 
2.0 c.c. antigen 1 : 5 to 2.0 c.c. saline = 1 : 10. 
1.5 c.c. antigen 1 : 6 to 1.5 c.c. saline = 1 : 12. 
1.5 c.c. antigen 1 : 8 to 1.5 c.c. saline = 1 : 16. 
1.0 c.c. antigen 1 : 10 to 1.0 c.c. saline = 1 : 20. 
1.0 c.c. antigen 1 : 12 to 1.0 c.c. saline = 1 : 24. 
1.0 c.c. antigen 1 : 16 to 1.0 c.c. saline = 1 : 32. 
1 Amer. Jour. Syph., 1922, 6, 74, 289. 
2 Ibid., 1922, 6, 651. 
3 Ibid., 1922, 6, 319. 
4 Ibid., 1922, 6, 461. 484 
COMPLEMENT FIXATION IN SYPHILIS 
Hemolytic Titration.—1. In a series of ten regulation test-tubes place 0.5 
c.c. of the above dilutions of antigen respectively. 
2. To each tube add 0.5 c.c. of a 1 : 10 dilution of normal human serum 
previously heated for fifteen minutes at 55° C. and 1.5 c.c. of saline solution. 
3. Mix the contents of each tube and place in a refrigerator at 6° to 8° C. 
for fifteen to eighteen hours. 
4. Add 0.5 c.c. of 2 per cent, corpuscle suspension to each tube; mix 
and place in a water-bath at 38° C. for one hour. 
5. Allow tubes to stand several hours in a refrigerator and read the 
results, the smallest amount of antigen just beginning to produce hemolysis 
is the hemolytic unit. 
The following table shows the ensemble, the results of a titration, and 
the method of reading. 
Hemolytic Titration of Antigen 
Tube. 
Antigen, 
0.5 C.c. 
Heated 
Human 
Serum,1 
1 : 10. 
Saline. 
o 
+-> 
P 
<v 
Corpus- 
cles, 2 
Per Cent. 
Water-bath One Hour. 
1. . . . 
1 : 4 
0.5 c.c. 
1.5 c.c. 
O 
*4H P 
0.5 c.c. 
Marked hemolysis. 
2.... 
1 : 5 
U 
“ 
UrS 
U 
Slight hemolysis; unit. 
3.... 
1 : 6 
U 
U 
° c 
u 
No hemolysis. 
4.... 
1 : 8 
U 
u 
OC O 
L c-> 
u 
No hemolysis. 
5.... 
1 :10 
U 
u 
f3 Jc 
u 
No hemolysis. 
6.... 
1 : 12 
U 
u 
o.S£ 
(C 
No hemolysis. 
7.... 
1 : 16 
“ 
“ 
c3 
4- 
u 
No hemolysis. 
8.... 
1 : 20 
u 
u 
(( 
No hemolysis. 
9. . . . 
1 : 24 
3s 
(( 
No hemolysis. 
10.... 
1 : 32 
u 
X 
No hemolysis. 
Anticomplementary Titration.—1. In the first ten tubes of a second 
series of twelve test-tubes place 0.5 c.c. of the above dilutions of antigen 
respectively. 
2. To each of the first eleven tubes add 0.5 c.c. of a 1 : 10 dilution of 
normal human serum previously heated at 55° C. for fifteen minutes. 
3. Add 1 c.c. of diluted complement (carrying 2 full units) to each of 
the twelve tubes. 
4. To the eleventh tube add 0.5 c.c., and to the twelfth tube, 1 c.c. 
saline and place in a refrigerator at 6° to 8° C. for fifteen to eighteen hours. 
5. Mix all tubes. 
6. Add 0.5 c.c. hemolysin (2 units), and 0.5 c.c. of the 2 per cent, sus- 
pension of corpuscles to each tube; mix, and place in a water-bath at 38° C. 
for one hour. Place the tubes in a refrigerator for a few hours and read the 
results. 
7. The anticomplementary unit is the smallest amount of antigen pro- 
ducing some inhibition of hemolysis. The eleventh tube is the serum control; 
the twelfth tube is the hemolytic system control and both should show 
complete hemolysis. 
The following table shows the ensemble, the results of a titration, and 
method of reading: 
1 May be omitted, in which case 2 c.c. saline are added to each tube instead of 1.5 c.c. TECHNIC OF THE THIRD METHOD 
485 
Anticomplementary Titration of Antigen 
Tube. 
Antigen. 
Heated 
Human 
Serum 1 
1 : 10. 
Comple- 
ment. (2 
Full 
Units) 
C/2 
U-i 
3 
O 
A 
o 
Hemol- 
ysin (2 
Units). 
Corpus- 
cles. 
Water-bath One Hour. 
1.... 
1 : 
4 
0.5 c.c. 
1.0 c.c. 
£ 
to 
0.5 c.c. 
0.5 c.c. 
Slight hemolysis.2 
2.... 
1 : 
5 
"3 
o 
tt 
it 
Complete inhibition of hemol- 
ysis. 
3.... 
1 : 
6 
CJ 
tt 
It 
Marked inhibition of hemolysis. 
4.... 
1 : 8 
tt 
tt 
<5 
tt 
ft 
Slight inhibition of hemolysis; 
unit. 
5 
1 : 10 
tt 
tt 
o 
it 
tt 
Complete hemolysis. 
6.... 
1 : 12 
d 
tt 
“ 
Complete hemolysis. 
7.... 
1 : 16 
tt 
Complete hemolysis. 
8.... 
1 : 20 
00 
1 
tt 
tt 
Complete hemolysis. 
9.... 
1 : 24 
SO 
tt 
tt 
Complete hemolysis. 
10.... 
1 :32 
O 
4-> 
tt 
tt 
Complete hemolysis. 
11.... 
0.5 
saline. 
tt 
»- 
to 
C( 
it 
Complete hemolysis. 
12.... 
1.0 
saline. 
it 
tt 
L 
*3 
tt 
tt 
Complete hemolysis. 
Antigenic Titrations.—1. In a series of ten test-tubes prepare the follow- 
ing dilutions of antigen starting with the remainder of the 1 : 10 dilution 
prepared above for the hemolytic and anticomplementary titrations: 
0.1 c.c. antigen 1 : 10 to 2.9 c.c. saline = 1 : 300. 
0.1 c.c. antigen 1 : 10 to 3.9 c.c. saline = 1 : 400. 
0.1 c.c. antigen 1 : 10 to 4.9 c.c. saline = 1 : 500. 
1.0 c.c. antigen 1 : 300 to 1.0 c.c. saline = 1 : 600. 
1.0 c.c. antigen 1 : 400 to 1.0 c.c. saline = 1 : 800. 
1.0 c.c. antigen 1 : 500 to 1.0 c.c. saline = 1 : 1000. 
1.0 c.c. antigen 1 : 600 to 1.0 c.c. saline = 1 : 1200. 
1.0 c.c. antigen 1 : 800 to 1.0 c.c. saline = 1 : 1600. 
1.0 c.c. antigen 1 : 1000 to 1.0 c.c. saline = 1 : 2000. 
1.0 c.c. antigen 1 : 1200 to 1.0 c.c. saline = 1 : 2400. 
2. Arrange a series of twelve regulation test-tubes and place 0.5 c.c. 
of the above dilutions of antigen into the first ten tubes respectively. 
3. In each of the first eleven tubes place 0.5 c.c. of a 1 : 10 dilution of a 
mixture of equal parts of four or more freshly collected syphilitic and Wasser- 
mann positive sera previously heated at 55° C. for at least fifteen minutes. 
4. In each tube place 1 c.c. of diluted complement (carrying 2 full units). 
5. To the eleventh tube add 0.5 c.c. and to the twelfth tube 1 c.c. of 
saline solution. 
6. Mix contents of all tubes and place in a refrigerator at 6° to 8° C. 
for fifteen to eighteen hours. 
7. Add 0.5 c.c. hemolysin (2 units) and 0.5 c.c. of the 2 per cent, corpuscle 
suspension to all tubes; mix, and place in a water-bath at 38° C. for one 
hour. Place tubes in a refrigerator for a few hours and read the results. 
8. The antigenic unit is the highest dilution of antigen giving complete 
inhibition of hemolysis. The eleventh tube is the serum control, and the 
twelfth tube the hemolytic system control; both should show complete 
hemolysis. 
1 May be omitted and 0.5 c.c. saline added instead. 
2 Due to the hemolytic activity of the antigen. 486 
COMPLEMENT FIXATION IN SYPHILIS 
The following table shows the ensemble, the results of a titration, and 
method of reading: 
Antigenic Titration of Antigen 
Tube. 
Antigen, 
0:5 C.c. 
Heated 
Syphi- 
litic 
Sera, 
1 : 10. 
Comple- 
ment (2 
Full 
Units). 
C/3 
H 
G 
O 
rG 
Hemol- 
ysin (2 
.Units). 
Corpus- 
cles, 2 
Per Cent. 
Water-bath One Hour 
1.... 
1 : 300 
0.5 c.c. 
1.0 c.c. 
<L> 
CJ 
+-> 
rG 
bp 
0.5 c.c. 
0.5 c.c. 
Complete inhibition of hemol- 
ysis. 
2.... 
1 : 400 
cc 
CC 
<D 
O 
CC 
cc 
Complete inhibition of hemol- 
ysis. 
3.... 
1 : 500 
cc 
CC 
G 
S 
03 
it 
cc 
Complete inhibition of hemol- 
ysis. 
4.... 
1 : 600 
CC 
cc 
<3 
o 
cc 
cc 
Complete inhibition of hemol- 
ysis. 
5.... 
1 : 800 
cc 
(C 
d 
cc 
cc 
Complete inhibition of hemoly- 
ysis. 
6.... 
1 : 1000 
cc 
cc 
o 
2 
cc 
cc 
Complete inhibition of hemol- 
ysis. 
7.... 
1 : 1200 
Cl 
cc 
4-> 
cc 
cc 
Complete inhibition of hemolysis. 
8.... 
1 :1600 
cc 
cc 
u 
O 
cc 
cc 
Complete inhibition of hemolysis 
{unit). 
9.... 
1 : 2000 
a 
cc 
s 
cc 
cc 
Marked inhibition of hemolysis. 
10.... 
1 : 2400 
cc 
cc 
#bf 
a 
cc 
Slight inhibition of hemolysis. 
11.... 
0.5 
saline. 
cc 
cc 
*3 
cc 
cc 
Complete hemolysis. 
12.... 
l.o 
saline. 
cc 
cc 
cc 
cc 
Complete hemolysis. 
9. Ten antigenic units are used in conducting the complement-fixation 
tests. For example, if the unit is 0.5 c.c. of a 1 ; 1600 dilution, as shown in 
the above table, the dose of 10 units would be contained in 0.5 c.c. of 1 : 160 
dilution. 
The Quantitative Complement-fixation Test.—1. Sera should be properly 
prepared and heated in a water-bath at 55° C. for fifteen minutes; spinal 
fluids are used unheated. The tests should be set up in the following order: 
Serum first, followed by antigen, and lastly by complement. 
2. For each serum arrange six regulation test-tubes and place saline 
solution in the first five in the following amounts respectively: 1.2, 0.5, 
0.5, 2.0, and 0.5 c.c. 
Into the first tube place 0.3 c.c. serum and mix; transfer 0.5 c.c. to tubes 
Nos. 2 and 6. 
Mix No. 2 and transfer 0.5 c.c. to tube No. 3. 
Mix No. 3 and transfer 0.5 c.c. to tube No. 4. 
Mix No. 4 and transfer 0.5 c.c. to tube No. 5; discard 1.5 c.c. 
Mix No. 5 and discard 0.5 c.c. 
Each tube now contains 0.5 c.c. carrying 0.1, 0.05, 0.025, 0.005, 0.0025, 
and 0.1 c.c. (serum control), as employed by Detweiler1 in his quantitative 
test. 
3. For each spinal fluid arrange six regulation tubes and place 0.5 c.c. 
saline in Nos. 2, 3, 4, 5, and 6. 
Into the first, second, and sixth tubes place 0.5 c.c. spinal fluid; mix No. 2 
and transfer 0.5 c.c. to No. 3; mix No. 3 and transfer 0.5 c.c. to No. 4; mix 
No. 4 and transfer 0.5 c.c. to No. 5; mix No. 5 and discard 0.5 c.c. 
1 Canad. Med. Assn. Jour., January, 1916. Fig. 140.—A Positive Reaction with the New Complement-fixation Technic. 
Shows test-tubes slightly reduced in size, the volume in each and depth of color. The reaction is 4421 (very strongly positive). 
(Amer. Jour. Syphilis, 1922, 6, 92.) TECHNIC OF THE THIRD METHOD 
487 
Each tube now contains 0.5 c.c. carrying 0.5, 0.25, 0.125, 0.0625, 0.03125, 
and 0.5 c.c. (control). 
4. Into the first five tubes of each set add 0.5 c.c. antigen dilution (carry- 
ing ten antigenic units). 
5. After an interval of five to thirty minutes add 1.0 c.c. of complement 
to each tube (carrying two full units). 
6. Include the following controls: (a) Antigen control tube carrying 0.5 
c.c. of the diluted antigen, 1 c.c. of the diluted complement, and 0.5 c.c. of 
saline solution; (b) hemolytic system control carrying 1 c.c. of the diluted 
complement and 1 c.c. of saline solution; (c) corpuscle control carrying 2.5 
c.c. of saline solution; (d) positive and negative serum controls should be 
included using syphilitic and normal sera respectively, set up in the various 
amounts as described above. 
7. Prepare a reading scale as follows: 
(a) Heat 6 c.c. of the diluted complement in a water-bath at 55° C. for 
fifteen minutes. 
(b) Prepare a solution of hemoglobin by dissolving 2 c.c. of the 2 per 
cent, corpuscle suspension in 4 c.c. of plain water. 
(c) Arrange a series of five regulation test-tubes numbered 1 to 5 and 
place in each: 0.5 c.c. of the diluted antigen and 1 c.c. of the heated diluted 
complement. 
(d) Add hemoglobin solution to the first four tubes as follows: 1.5, 1.13, 
0.75, and 0.38 c.c. 
(e) Add 0.24 c.c. physiologic saline solution to No. 2; 0.5 c.c. to No. 3; 
0.74 c.c. to No. 4, and 1 c.c. to No. 5. 
Corpuscles are added the following day. 
8. Mix the contents of all tubes gently, but thoroughly, and place in a 
refrigerator at 6° to 8° C. for fifteen to eighteen hours. 
9. Warm the tubes in a water-bath at 30° C. for five to fifteen minutes, 
but not longer1 and add 0.5 c.c. hemolysin (carrying two units) to all tubes 
except the corpuscle control; thoroughly mix the 2 per cent, corpuscle 
suspension (which has been carried over in a refrigerator from the previous 
day) and add 0.5 c.c. to all tubes. 
Tube. 
Patient’s 
Serum in 
0.5 C.c.* 
Antigen, 
10 Units. 
c/i 
O> 
'Complement 
(2 Full 
Units). 
U 
O 
«4H 
6 . 
Hemolysin 
(2 Units). 
Corpuscles, 
2 Per Cent. 
4-) C/1 
cj *3 
gj 
1 
2 
0.1 c.c. 
0.05 c.c. 
0.5 c.c. 
it 
"3 
P 
‘a 
1.0 c.c. 
it 
o *-< 
00 P 
0.5 c.c. 
tt 
0.5 c.c. 
tt 
<u <u 
c a 
3 
0.025 c.c. 
tt 
>* 
it 
4-> 
d £ 
tt 
tt 
.2 3 
4 
0.005 c.c. 
it 
tt 
tt 
tt 
5 
0.0025 c.c. 
a 
Ip 
4-> 
it 
.2 m 
tt 
tt 
§ 
6 
0.1 c.c. 
o 
tc 
a « 
tt 
tt 
u° • 
7 
control. 
Control 
0.5 c.c. 
0) 
> 
tt 
as 
tt 
it 
Secondary ii 
58° C. Read 
.ater with seal 
8 
antigen. 
Control 
c3 
£ 
OJ 
tt 
tt 
9 
hemolytic. 
Control. 
saline. 
a 
PL, 
it 
corpuscle. 
The Quantitative Complement-fixation Test for Syphilis (Fig. 140) 
1 This preliminary warming may be omitted; if more than fifteen minutes are used some 
non-specific reactions may occur. 
2 Spinal fluid doses: 0.5, 0.25, 0.125, 0.0625, 0.03125 and 0.5 c.c. (control). 488 
COMPLEMENT FIXATION IN SYPHILIS 
10. Add corpuscle suspension to the tubes of the reading scale as fol- 
lows: 0.13 c.c. to No. 2, 0.25 c.c. to No. 3, 0.38 c.c. to No. 4, and 0.5 c.c. to 
No. 5. 
11. Mix the contents of all tubes gently but thoroughly, and place in 
a water-bath at 38° C. for one hour (the water must reach above the level of 
the contents in the tubes). 
12. Place all tubes in a refrigerator for one to three hours to permit 
the partial settling of non-hemolyzed corpuscles; the degree of inhibition 
hemolysis is then read off and recorded for each tube with the aid of the 
reading scale as —, + (1), ++ (2), + + + (3), or + + + + (4). All serum 
controls, antigen, and hemolytic controls should show complete hemolysis. 
13. Tube No. 1 of the color scale shows a — reaction; tube No. 2 shows 
a +; tube No. 3 shows a + + , tube No. 4 shows a + + + , and tube No. 5 
shows a -f + + + reaction. 
Experience has shown that the reactions may be interpreted as follows: 
Very strongly positive if there is partial or complete fixation of comple- 
ment in the first four or five tubes. Occasionally a serum will show 
less fixation of complement in the first tube carrying 0.1 c.c. serum than the 
second tube carrying 0.05 c.c. as a + + , + + + + , + + + , + and — (con- 
trol) reaction (recorded as 2, 4, 3, 1, —). It may be assumed that this is 
due to the presence of natural antisheep hemolysin, but I have seen the 
phenomenon occur with hemolysin-free sera and believe that it is due to the 
presence of other serum constituents in this relatively large amount of serum 
interfering with the fixation of complement by antigen and syphilis anti- 
body. So far I have not seen this occur with spinal fluids. 
Strongly positive if there is a partial or complete fixation of complement 
in the first three tubes. 
Moderately positive if there is partial or complete fixation of complement 
in the first two tubes. 
Weakly positive if there is partial or complete fixation of complement in 
the first tube. 
Negative when all tubes show complete hemolysis. 
The following results of tests with a syphilitic serum and spinal fluid 
shows the method of recording and reporting: 
Quantitative Reaction.—Moderately positive (42 — ). 
Serum 0.1 c.c. = + + + + • 
Serum 0.05 c.c. = + + . 
Serum 0.025 c.c. = —. 
Serum 0.005 c.c. = —. 
Serum 0.0025 c.c. = —. 
Serum 0.1 (control) c.c. = —. 
Quantitative Reaction.—Very strongly positive (44442). 
Spinal fluid 0.5 c.c. = + + + + . 
Spinal fluid 0.25 c.c. = + + + + . 
Spinal fluid 0.125 c.c. = + + + + • 
Spinal fluid 0.0625 c.c. = + + + + • 
Spinal fluid 0.03125 c.c. = + + • 
Spinal fluid 0.5 (control) c.c. = —. 
The Qualitative Complement-fixation Test.—The qualitative test is con- 
ducted in the same manner as the quantitative test except that two tubes, 
instead of six, are employed; otherwise the technic is the same and may be 
described more briefly: 
1. For each serum arrange two regulation test-tubes and place 0.8 c.c. TECHNIC OF THE THIRD METHOD 
489 
saline and 0.2 c.c. serum in the first; mix the contents and transfer 0.5 c.c. 
to the second tube (control). 
2. For each spinal fluid arrange two regulation test-tubes and place 
0.5 c.c. spinal fluid in each. 
3. Add 10 units of antigen (0.5 c.c. dilution) to the first tube and 0.5 
c.c. of saline solution to the second tube or control tube, of each series. 
4. After waiting five to thirty minutes add 1 c.c. of diluted complement 
(carrying two full units) to both tubes of each set. 
5. Include corpuscles, hemolytic system, and antigen controls as pre- 
viously described; also a syphilitic serum and normal serum for a positive 
and negative reaction, as described above. 
6. Set up first part of the reading scale as described with the quanti- 
tative test. 
7. Mix the contents of all tubes gently but thoroughly, and place all 
tubes in a refrigerator at 6° to 8° C. for fifteen to eighteen hours. 
8. Warm the tubes in a water-bath at 38° C. ior five to fifteen minutes 
{not longer), add 0.5 c.c. hemolysin (2 units) to all tubes except the cor- 
puscle control. Gently but thoroughly mix the 2 per cent, corpuscle suspen- 
sion carried over in the refrigerator from the previous day and add 0.5 c.c. 
to each tube. 
9. Finish the reading scale by adding corpuscles as previously described. 
10. Mix the contents of all tubes gently but thoroughly, and incubate 
in a water-bath at 38° C. for one hour. 
11. Place the tubes in a refrigerator for one to three hours to permit 
the partial settling of non-hemolyzed corpuscles and read the results with 
the aid of the scale. All serum, the antigen, and hemolytic controls should 
show complete hemolysis; the corpuscle control should show no hemolysis. 
The Qualitative Complement-fixation Test for Syphilis 
Tube. 
Patient’s 
Serum in 
0.5 C.c. 
Antigen 
(10 Units). 
tn 
CL) 
Complement 
(2 Full 
Units). 
o 
+-> 
CD 
Hemolysin 
(2 Units). 
Corpuscles, 
2 Per Cent. 
0) 
3 <L> 
jz; X a) 
3 
r, O 
<3 
1 
0.1 c.c. 
0.5 c.c. 
S 
1.0 c.c. 
0.5 c.c. 
0.5 c.c. 
.2 33; 
2 
0.1 c.c. 
(control). 
C 
<D 
C( 
O 
C/5 
3 s 
U 
(( 
ts oa 
J-uj 
3 b! J 
3 
Antigen 
0.5 c.c 
-2 
u 
3 
u 
(i 
<J <D 
s 
(control). 
0) 
> 
.S c 
4 
Hemolytic 
(control). 
<43 
a 
<u 
CC 
h • D 
5 
Corpuscle 
(control). 
> 
2.5 c.c. 
u 
8=2 
*c ° 
Ph 
33 
C/3 4-J O 
C3 
12. The results are read with the aid of the reading scale as previously 
described and recorded according to the + + + + , + + + , + + , +, and 
— method of recording complement fixation in the first tube of each set: 
+ + + += strongly positive (100 per cent, inhibition of hemolysis), + + + 
= moderately positive (about 75 per cent, inhibition of hemolysis), ++ = 
weakly positive (about 50 per cent, inhibition of hemolysis), and + = very 
weakly positive (about 25 per cent, inhibition of hemolysis), — = negative 
(complete hemolysis). 
The Relative Value of the Quantitative and Qualitative Tests.—Experi- 
ence has shown that the quantitative test is more satisfactory than the 
qualitative test. It is true that more test-tubes and materials are required, COMPLEMENT FIXATION IN SYPHILIS 
490 
but these and the slightly greater time required for conducting the tests 
are more than compensated for by the results. With about \ per cent, of 
sera the first dose of 0.1 c.c. may yield a weaker reaction than the smaller 
amounts, as 0.05 and 0.025 c.c.; indeed, with weakly positive sera the 0.1 
c.c. amount may yield a falsely negative reaction, whereas true positive reac- 
tions occur with the 0.05 c.c. amount. As previously stated, Detweiler has 
noticed this phenomenon in his quantitative test and believes that it is due 
to the influence of natural antisheep hemoglobin, but I have observed the 
reaction with hemolysin-free sera and have never seen it with spinal fluids 
containing some hemolysin. In my opinion it is due to interference of 
complement fixation by some other serum constituent. Whatever may be 
the true explanation, the fact remains that it occurs and constitutes the main 
reason for using varying amounts of patient’s serum in conducting the 
syphilis complement-fixation test; furthermore, the quantitative method 
gives a better serologic guide to treatment. 
For example, a qualitative test yielding a + + + + reaction may not 
show any reduction in the strength of the reaction despite considerable 
treatment because the reagin content of the serum has not been reduced 
below the + + + + level; under these circumstances the physician may 
surmise that there has been no serologic improvement even though clinical 
improvement has taken place. On the other hand, the quantitative test 
employing smaller amounts of serum will show in a clear manner the degree 
of serologic improvement, as demonstrated in the following record of a case 
of syphilis in the secondary stage treated with neo-arsphenamin: 
Serum. 
Before 
Treatment. 
After 6 
Injections. 
After 12 
Injections. 
After 15 
Injections. 
0.1 
+ + + + 
+ + + + 
+ 
0.05 
+ + + + 
+ + 
— 
— 
0.025 
+ + + + 
— 
— 
— 
0.005 
+ + 
— 
— 
— 
0.0025 
— ' 
— 
— 
— 
By means of this quantitative reaction it is possible to plot a curve for 
each case under treatment. Such curves show fluctuations similar to those 
observed by Vernes with his precipitation reaction, with a gradual tendency 
for reduction to a negative reaction. 
The Quantitative and Qualitative Tests with an Antiox Hemolytic 
System.—In those laboratories where a sheep cannot be maintained or 
sheep blood regularly obtained from an abattoir, beef blood may be used 
with entire success and satisfaction. The technic is exactly the same as 
described except that beef corpuscles are used instead of sheep. 
Not infrequently it is more difficult to prepare highly potent antiox 
hemolysin by immunization of rabbits than antisheep; under these cir- 
cumstances it may be necessary to dilute the complement 1 : 20 for the 
hemolysin and complement titrations instead of 1 : 30 as described for the 
antisheep system. 
When the hemolysin is of such strength as permits the use of 1 : 30 
complement, the results are almost identical to those observed with the 
antisheep system1; hemolysins of this strength can and should be used rather 
than a weak hemolytic serum requiring the use of 1 : 20 complement. 
1 Amer. Jour. Syph., 1922, 6, 667. TECHNIC OF THE FOURTH METHOD 
491 
The Quantitative and Qualitative Tests with Antihuman and Anti- 
chicken Hemolytic Systems.—Experience has shown that tests conducted 
with antisheep and antiox hemolytic systems are slightly more sensitive than 
tests conducted with the antihuman and antichicken systems.1 I believe that 
these differences are due to the dilution and amount of complement and not to 
natural hemolysins in human sera. It appears to be a fundamental prin- 
ciple that the higher the dilution of complement, that is, the smaller the amount 
of complement serum we may use, the more sensitive are the complement-fixation 
reactions. The difficulties encountered in the preparation of antihuman and 
antichicken hemolysins are such that it is usually necessary to use the com- 
plement diluted 1 : 5 or 1 : 10, and these larger amounts of guinea-pig serum 
complement reduce the sensitiveness of reactions with the new technic probably 
by the introduction of serum proteins interfering with specific complement 
fixation. The presence or absence of natural hemolysins in the complement 
serum appear to be of much less importance because the results are the 
same regardless of the presence or absence of these substances in the com- 
plement or patients’ sera. 
Techinc of the Fourth Method 
AUTHOR'S MODIFICATION OF THE WASSERMANN REACTION WITH VARYING AMOUNTS 
OF COMPLEMENT 
In this method the amount of syphilis antibody in a serum is measured 
according to the number of hemolytic doses of complement absorbed or 
fixed with a constant amount of antigen. As previously stated, any organic 
extract used as antigen may of itself fix a certain amount of complement; 
a non-syphilitic serum may do the same, and a mixture of the two may fix 
still more, though the amounts may be relatively small. A peculiarity 
possessed by a syphilitic serum is that it fixes a large amount of complement 
when mixed with antigen; as a result the test becomes a quantitative and 
not a qualitative reaction. The only practical means we possess of esti- 
mating the amount of complement in a fresh serum is to ascertain the he- 
molytic dose, i. e., to find the smallest quantity of serum that is just suffi- 
cient completely to lyse the test amount of corpuscles with the hemolysin. 
When this has been done, the quantity of complement used in the reaction 
may be expressed in terms of hemolytic doses fixed, and not in terms of the 
amount of fresh serum. 
When properly performed according to this method, which has been 
modified after the technic of Browning and Mackenzie2 and Thomsen,3 the 
syphilis reaction becomes quite delicate and accurate, but is more com- 
plicated than the other methods, and should not be attempted until one is 
accustomed to the simpler test and thoroughly understands the underlying 
principles of the syphilis reaction and knows the many sources of fallacy. 
The greater amount of work that it entails and the larger quantities of com- 
plement-serum and amboceptor that are required may serve as factors against 
its adoption as a routine method. On the other hand, the sources of error 
are well under control, and the test has yielded remarkably uniform results 
in the hands of my colleagues in their respective laboratories, working with 
the serums of the same persons. 
1. Corpuscles.—Sheep corpuscles are washed three times with salt 
solution and made up in a 2| per cent, suspension; dose, 0.5 c.c. 
1 Amer. Jour. Syph., 1922, 6, 667. 
2 Ztschr. f. Immunitatsf., orig., 1909, 2, 459. 
3 Ztschr. f. Immunitatsf., orig., 1910, 7, 389. 492 
COMPLEMENT FIXATION IN SYPHILIS 
2. Hemolysin.—Antisheep hemolysin is titrated by placing increasing 
amounts of diluted serum, as 0.1 c.c., 0.15 c.c., 0.25 c.c., 0.30 c.c., and 0.35 
c.c. of a 1 : 300 dilution in a series of test-tubes and adding to each tube 
0.025 complement serum (= 0.5 c.c. of a 1 : 20 dilution), 0.5 c.c. of 2b per 
cent, corpuscle suspension, and sufficient salt solution to make the total 
volume in each tube about 3 c.c. Each tube is gently shaken and incubated 
in the water-bath at 38° C. for one hour, when the reading is made. The 
unit of hemolysin is the smallest amount giving complete hemolysis. In- 
stead of using increasing amounts of one dilution of hemolysin as above, a 
series of dilutions may be made in flasks, as 1 : 2000; 1 : 2500; 1 : 3000; 
1 : 3500; 1 : 4000, etc., and 1 c.c. of each used in the titration. 
The unit need not be determined more than once in two weeks, pro- 
viding the hemolysin is kept at a low temperature. 
3. Complement.—It is advisable to use the mixed sera of at least two 
or more healthy guinea-pigs. When only a small amount of complement 
serum is required, sufficient blood may be obtained by aspirating 2 c.c. of 
blood from the hearts of several large animals (see page 37). The com- 
plement serum should be clear, collected a few hours before use, and pref- 
erably from fasting animals. The mixed serum is diluted with 9 parts of 
normal salt solution (1 : 10) and titrated. This titration must be performed 
with every complement serum before the main tests are conducted. In 
order to allow for the anticomplementary action of the antigen the com- 
plement is titrated as suggested by Thomsen, in the presence of the dose of 
antigen, as determined by titration, to be used in the main tests, as follows: 
In each of a series of six test-tubes place the dose of antigen (as, for example, 
0.1 c.c. of 1 : 20 dilution of acetone-insoluble lipoids); add the following 
amounts of complement 1 : 10 with an accurate pipet: 0.1 c.c., 0.2 c.c., 
0.3 c.c., 0.4 c.c., 0.5 c.c., and 0.6 c.c. Add sufficient salt solution to make the 
total volume in each tube about 2 c.c., mix gently, and incubate in a water- 
bath at 38° C. for one hour, then one unit of hemolysin and 0.5 c.c. of 2\ 
per cent, suspension of cells are added, the tubes shaken and re-incubated in 
the water-bath for an hour, when the reading is made. The unit of comple- 
ment is the smallest amount producing complete hemolysis. The results of a 
titration are shown in the following table and Fig. 141. 
Results of Complement Titration in the Presence of Antigen 
Tube 
No. 
Antigen 
1 : 20, 
C.c. 
Comple- 
ment 
1 : 10, 
C.c. 
Add salt solution to make 
2 c.c.; shake gently and incu- 
bate in water-bath at 38° C. 
for one hour. 
i _ 
Hemol- 
ysin. 
Cells 2 yi 
Per Cent.. 
C.c 
Shake gently and incubate 
in water-bath at 38° C. for 
one hour. 
Results. 
1... 
2. .. 
3.. 
4.. . 
5.. . 
6.. 
0.2 
0.2 
0.2 
0.2 
0.2 
0.2 
0.1 
0.2 
0.3 
0.4 
0.5 
0.6 
1 unit. 
1 unit. 
1 unit. 
1 unit. 
1 unit. 
1 unit. 
0.5 
0.5 
0.5 
0.5 
.5 
0.5 
Very weak hemolysis. 
Marked hemolysis. 
Almost complete he- 
molysis. 
Complete hemolysis 
the unit. 
Complete hemolysis. 
Complete hemolysis. 
4. Antigen.—A large number of comparative tests with the same sera 
and a variety of antigens have shown that the author’s new cholesterolized 
and lecithinized alcoholic extract of beef heart is best adapted for this 
reaction. As a general rule these extracts possess a marked degree of anti- 
genic sensitiveness and are usually free of anticomplementary action except Fig 141.—The Wassermann Reaction after the Quantitative or Fourth Method. 
Shows the fixation of four units of complement. There is partial fixation with five units, but the serum alone fixed 
a unit (tube 7). TECHNIC OF THE FOURTH METHOD 
493 
in very large doses. Each extract must be titrated to determine the proper 
dose to employ; as a general rule it is sufficient to titrate an extract once 
every two months providing it is carefully refrigerated and is yielding 
satisfactory results in complement-fixation tests. 
In conducting these titrations a mixed complement serum diluted 1 : 10 
is titrated in the following doses with one unit of hemolysin and 0.5 c.c. of 
a 2\ per cent, suspension of cells: 0.1 c.c., 0.15 c.c., 0.2 c.c., 0.25 c.c., 0.3 
c.c., and 0.4 c.c. In the antigenic titration the following amounts of antigen 
diluted 1 : 20 are placed in a series of ten test-tubes: 0.001 c.c., 0.005 c.c., 
0.008 c.c., 0.01 c.c., 0.05 c.c., 0.08 c.c., 0.1 c.c., 0.2 c.c., 0.3 c.c., and 0.4 
c.c.; 0.1 c.c. of fresh inactivated syphilitic serum, preferably from several 
persons mixed, plus 2 units of complement and sufficient salt solution to 
make 3 c.c., are added to each tube and incubated in a water-bath at 38° 
C. for one hour, when 1 unit of hemolysin and 0.5 c.c. of corpuscles are 
added, tubes shaken, and re-incubated for an hour. The reading may be 
made at once or after the corpuscles have settled; the unit is the smallest 
amount of antigen yielding complete inhibition of hemolysis. 
In the anticomplementary titration the extract is diluted 1 : 20 and the 
following amounts placed in a series of ten test-tubes with 0.1 c.c. of fresh 
inactivated normal serum and 2 units of complement: 0.1, 0.2, 0.3, 0.5, 
0.8, 1.0, 1.2, 1.5, and 2.0 c.c. The titration is conducted in the same man- 
ner, and the anticomplementary unit is the smallest amount of extract producing 
inhibition of hemolysis. 
Hemolytic and serum controls on both the syphilitic and normal serum 
should be included and show complete hemolysis. 
A satisfactory extract is one in which the antigenic unit is at least ten 
times less than the anticomplementary unit, and in conducting the Wasser- 
mann reaction 2 units of antigen are employed as the dose. 
5. Fluid to be Tested.—Serum is heated at 56° C. for half an hour and 
used in dose of 0.1 c.c. Cerebrospinal fluid should be fresh and is used un- 
heated in dose of 1 c.c. 
6. The Test.—For each serum eight test-tubes are arranged in a row. 
One is labeled with the patient’s name and all are numbered. Into each 
is placed 0.1 c.c. of the patient’s serum. Into each of the first six tubes are 
placed the dose of antigen and 2, 3, 4, 5, 6, and 8 units of complement re- 
spectively. The last two tubes are the serum controls, to determine the 
amount of complement fixed by serum alone, and receive 1 and 2 units of 
complement respectively. Sufficient salt solution is added to each tube to 
bring the total volume up to 3 c.c., and all are incubated in the water-bath 
at 38° C. for an hour. One unit of hemolysin and 0.5 c.c. of corpuscle sus- 
pension are now added to each tube and re-incubated in the water-bath for 
one hour, when the readings are made. Sharper readings may be made after 
the corpuscles have settled either by centrifuging or standing the tubes in 
the refrigerator overnight. 
An antiox or antihuman hemolytic system may be employed if 
sufficiently potent hemolysins are at hand. 
Controls.—1. The anticomplementary action of each serum is controlled 
in the last two tubes of each series; unless a serum is markedly anticom- 
plementary, the use of 1 and 2 units of complement is sufficient. 
2. A known positive and negative serum may be included. 
3. A hemolytic control is set up with 1 unit of complement and antigen; 
after the primary incubation 1 unit of hemolysin and the cells are added. 
This tube is a control on the unit of complement and should show complete 
hemolysis. 494 
COMPLEMENT FIXATION IN SYPHILIS 
4. A corpuscle control may be included, containing 0.5 c.c. of corpuscles 
and 3 c.c. of salt solution. It controls the tonicity of the salt solution and 
should show no hemolysis. 
Reading the Results.—The controls are first inspected. The corpuscle 
control should show no hemolysis and the hemolytic control be just he- 
molyzed. The last two tubes of each series are the serum control tubes, 
and the first tube containing 1 unit of complement may show incomplete 
hemolysis, while the second tube containing 2 units of complement shows 
complete hemolysis unless the serum is quite anticomplementary. 
Results and Method of Reading the Wassermann Reaction (Quantitative Method) 
No. 
Diag- 
Results or Complement-fixation Tests. 
Serum 
Controls. 
Amount of Comple- 
ment Absorbed; 
Readings. 
NOSIS. 
2 Units. 
3 Un ts. 
4 Units. 
5 Units. 
6 Units. 
8 Units. 
l«Unit. 
2 
Units. 
1 
Normal. 
_ 
_ 
_ 
_ 
Negative. 
2 
Normal. 
+ 
— 
— 
— 
— 
— 
=fc 
— 
Negative. 
3 
Syphilis. 
++++ 
+ + + + 
+++ 
4- 
— 
— 
=4= 
— 
Positive; 4 units. 
4 
Syphilis. 
+++ 
+ + 
=fc 
— 
• — 
— 
=1= 
— 
Positive; 3 units. 
5 
Syphilis. 
+ + 
— 
— 
— 
— 
— 
— 
— 
Positive; 2 units. 
6 
Syphilis. 
++++ 
+ + + + 
+ + + + 
4-4-4-4- 
4-4-4-4- 
4-4-4-4- 
4- 
Positive; at least 8 
units. 
7 
Syphilis. 
++++ 
+ + + + 
4—1—f 4- 
4—1—1—1- 
4-4-4-4- 
4-4-4-4- 
4- 
db 
Positive; at least 8 
units. 
8 
Syphilis. 
+++ + 
+ + 
4- 
— 
— 
4- 
db 
Pos tive; 2 to 3 units. 
The first six tubes of each series show whether or not the reaction is 
positive or negative, and if positive, the amount of complement absorbed. 
If tube No. 7 of the serum controls shows some inhibition of hemolysis, 1 
unit is subtracted from the number of units of complement absorbed in the 
first six tubes, and the difference represents the amount of complement 
absorbed by antigen and syphilitic antibody. I regard the reaction as 
positive when lysis is incomplete with 2 units of complement, in addition to 
the amount absorbed by the serum alone. More strongly reacting serums will 
absorb from 3 to 8 units of complement, and not infrequently more; these 
sera may be retested with 8, 10, 12, etc., units of complement, but the 
employment of these large doses routinely is not justified because of the 
large amount of complement used, and the complement-fixing power of the 
majority of sera are readily measured with 2 to 8 units. The method of 
reading is best shown in the preceding table and Fig. 141. 
Cases of syphilis progressing favorably with the administration of 
specific remedies show less and less complement fixation until a negative 
reaction is secured. A large number of comparative tests employing the 
original Wassermann and this reaction with the same antigen have shown 
that a serum yielding a + + + + result in the original Wassermann reaction 
may show anywhere from 2 to 8 units of fixation with this technic. 
The following method is an abbreviated description of the method 
which has been described by Neill1 as employed in the Hygienic Laboratory 
of the United States Public Health Service: 
Saline Solution.—0.9 per cent, solution of pure salt in distilled water; 
sterilized. 
Corpuscles.—Washed sheep corpuscles suspended in sufficient salt solu- 
tion to restore the volume of defibrinated blood employed. For use 5 c.c. 
Hygienic Laboratory Method 
1 Reprint No. 483 from the Public Health Reports, August 23, 1918, 1387. HYGIENIC LABORATORY METHOD 
495 
of this suspension is placed in 95 c.c. of salt solution, giving a 5 per cent, 
suspension of whole blood (corresponds to about a 2 per cent, suspension of 
washed packed cells). 
Hemolysin.—Antisheep hemolysin prepared by the immunization of 
rabbits. Dilute 0.1 c.c. of serum with 19.9 c.c. salt solution (= 1 : 200). 
In a series of six test-tubes place increasing amounts as follows: 0.1, 0.2, 
0.3, 0.4, 0.5, and 0.6 c.c. Add 1 c.c. of 1 : 20 dilution of the pooled sera of 
5 guinea-pigs for complement and 1 c.c. of 5 per cent, suspension of sheep 
corpuscles. Incubate in a water-bath at 37° C. for one hour and stand in 
a refrigerator overnight. The smallest amount of hemolysin giving com- 
plete hemolysis is the unit; if the unit is more than 0.4 c.c. of 1 : 200 the 
hemolysin should be rejected. The titration should be done at least once 
in six weeks. 
Complement.—The pooled sera of several guinea-pigs; titrated daily just 
before the main tests are set up. 
“Estimate, in round numbers, the number of cubic centimeters of red- 
cell suspension needed for the day’s work: for example, 100 c.c. Multiply 
the unit of amboceptor by 200 and place that amount of amboceptor serum 
in a 100 c.c. glass-stoppered graduated cylinder. Add about 50 c.c. of 0.9 
per cent, sodium chlorid solution, taking care to wash down the serum 
adhering to the sides of the cylinder; next add 5 c.c. of the undiluted sheep 
corpuscles which have been made up to the volume of the defibrinated 
blood. Then make up to 100 c.c. with 0.9 per cent, sodium chlorid solution. 
Invert fifty times to mix thoroughly. Set aside for fifteen minutes.” 
Dilute complement 1 : 10 and place 0.6, 0.5, 0.4, 0.3, 0.2, and 0.1 c.c. 
in a series of test-tubes; add 1 c.c. of the corpuscle-hemolysin suspension 
and sufficient salt solution to make 4 c.c. Water-bath incubation one 
hour. The unit is the smallest amount of complement giving complete 
hemolysis. 
Antigen.—Acetone-insoluble lipoids titrated in two series of test-tubes 
for anticomplementary and antigenic activities in the following amounts 
of 1 : 10 dilution: 2.0, 1.6, 1.4, 1.0, 0.8, 0.6, 0.4, 0.2, 0.1, 0.06, and 0.04 c.c. 
In the anticomplementary titration 0.2 c.c. of heated known negative serum 
is added to each tube; in the antigenic titration 0.2 c.c. of heated syphilitic 
serum. Add two units of complement to each tube and sufficient salt solu- 
tion to make the total volumes about 4 c.c. Mix the contents of the tubes 
and place in a water-bath at 37° C. for one hour. Add 1 c.c. of the ambo- 
ceptor-corpuscle suspension to each tube, mix, and re-incubate for a half 
hour; set aside in a refrigerator overnight. In the main tests the dose of 
antigen employed is several times the smallest amount giving complete 
inhibition of hemolysis in the titration with syphilitic serum providing this 
dose is not more than one-half the anticomplementary unit. 
Serum.—Heated in a water-bath to 54° to 56° C. for one-half hour; dose, 
0.2 c.c. 
Tests.—For each serum arrange two tubes; of each place 0.2 c.c. in the 
front tube and 0.4 c.c. in the rear (control). Add one dose of antigen to the 
front tubes and two doses of complement to both tubes. Add sufficient 
salt solution to make the total volume 3 c.c. in each tube. 
Include controls with known positive and negative sera; also an antigen 
control carrying two doses, a hemolytic system and corpuscle controls. 
Mix well by individually agitating each tube; incubate in a wrater-bath 
at 37° C. for one hour. Add to each tube 1 c.c. of the amboceptor-sheep 
corpuscle suspension. Mix well and re-incubate for one-half hour. Place 
in a refrigerator overnight. 496 
COMPLEMENT FIXATION IN SYPHILIS 
Readings are made as follows: 
70 to 100 per cent, fixation = “strongly positive.” 
40 to 70 per cent, fixation = “positive.” 
20 to 40 per cent, fixation = “weakly positive.” 
0 to 20 per cent, fixation = “negative.” 
Noguchi Method 
Among the large number of modifications of the original syphilis reac- 
tion that have been devised, that of Noguchi has proved of distinct value. 
In this method an antihuman hemolytic system is employed that eliminates 
one possible source of error due to the natural antisheep amboceptor in 
human serum, although the importance of this has been overemphasized 
and the possible influence of natural antihuman hemolysins in some serums 
overlooked. 
Noguchi advocated the use of active serum for this test, with an antigen 
of acetone-insoluble lipoids. Active serum yields a more delicate reaction, 
but may give false or proteotropic complement fixation, especially when 
crude alcoholic extracts are used as antigens. I may state, however, that 
when a good antigen of acetone-insoluble lipoids is used, the percentage of 
false reactions is relatively small, being less than 2 per cent. The Noguchi 
test, on the other hand, may be conducted with inactivated serums when 
the danger of false reactions is removed, but the delicacy of the test is 
likewise diminished, so that it more closely approaches the Wassermann 
reaction. 
1. Complement is furnished by the fresh, clear serum of the guinea-pig, 
put up in 40 per cent, solution, prepared by diluting 1 part of serum wfith 
1| parts of normal salt solution. Dose, 0.1 c.c. (5 drops from a capillary 
pipet). Whenever possible it is always best to use a mixture of the serums 
from two or more guinea-pigs. 
2. Human Corpuscles.—These are washed three times with normal 
salt solution, and used in dose of 1 c.c. of a 1 per cent, suspension. To a 
graduated centrifuge tube containing 9 c.c. of sterile 2 per cent, sodium 
citrate in normal salt solution add 1 c.c. of blood and shake gently. This 
amount of blood is easily secured by pricking the finger with a lancet and 
collecting the blood in the centrifuge tube up to the mark 10. This is 
centrifuged thoroughly, and the supernatant fluid drawn off. More saline 
solution is then added to the corpuscles, stirred up, and the mixture centri- 
fuged. The washing is repeated once more, and the supernatant fluid dis- 
carded. The corpuscles are then suspended in sufficient salt solution to 
make a total volume of 100 c.c. 
3. Hemolytic Amboceptor.—Antihuman hemolysin is prepared by im- 
munizing rabbits with human cells, as described on p. 401. It is a difficult 
matter to secure a potent amboceptor; human erythrocytes are more toxic 
than sheep’s cells for rabbits, and most animals produce but small amounts 
of the amboceptor. Hemagglutinins are produced in large amounts, and 
when using a low titer hemolytic serum, the test corpuscles are quickly 
agglutinated in small dense masses that are broken up with difficulty and 
that interfere greatly with hemolysis. With serums having a titer of 1 : 100 
or over the agglutinins are not so much in evidence; a satisfactory reaction 
is best observed, therefore, with a potent amboceptor (1 : 100) serum. 
The hemolytic serum may be preserved in 1 c.c. ampules after adding 
an equal part of glycerin, and a stock dilution prepared and titrated in the 
usual manner. The serum is also well preserved dried on filter-paper, as NOGUCHI METHOD 
497 
described on p. 57. A trial titration should always be made to determine 
the potency of the serum before the paper slips are prepared. 
If paper amboceptor is used, the uniform rule of titrating it with the com- 
plement and corpuscles on hand should be observed before the actual tests are 
made. This is done chiefly because, where one guinea-pig serum is used 
for complement, it may occasionally happen that the serum is less active 
than usual, so that if fixed doses of complement and amboceptor are used, 
the reactions may at times prove to be incomplete and inaccurate. The 
process of titration is so simple that any one may readily conduct it, and 
thus fulfil the most important requirement of any complement-fixation test, 
namely, adjustment of the complement, amboceptor, and corpuscles to one 
another. 
Titration of Serum Hemolysin.—Prepare a 1 : 100 dilution by mixing 
0.1 c.c. of immune serum (inactivated) with 9.9 c.c. of saline solution. To 
a series of six small test-tubes add increasing amounts of this diluted serum 
as follows: 
Tube 1: 0.1 c.c. amboceptor serum (1 : 100) + 0.1 c.c. complement 
(40 per cent.) + 1 c.c. corpuscle suspension (1 per cent.). 
Tube 2: 0.2 c.c. amboceptor serum (1 : 100) -f- 0.1 c.c. complement 
(40 per cent.) + 1 c.c. corpuscle suspension (1 per cent.). 
Tube 3: 0.4 c.c. amboceptor serum (1 : 100) + 0.1 c.c. complement 
(40 per cent.) + 1 c.c. corpuscle suspension (1 per cent.). 
Tube 4: 0.5 c.c. amboceptor serum (1 : 100) + 1 c.c. complement 
(40 per cent.) + 1 c.c. corpuscle suspension (1 per cent.). 
Tube 5: 0.8 c.c. amboceptor serum (1 : 100) + 0.1 c.c. complement 
(40 per cent.) + 1 c.c. corpuscle suspension (1 per cent.). 
Tube 6: 1 c.c. amboceptor serum (1 : 100) + 0.1 c.c. complement 
(40 per cent.) + 1 c.c. corpuscle suspension (1 per cent.). 
Sufficient saline solution is added to the first tubes of the series to bring 
the total volume up to 2 c.c. The tubes are then shaken gently and placed 
in the incubator at 37° C. for two hours (or one hour in water-bath at the 
same temperature), during which time they should be inspected and shaken 
gently several times. At the end of the period of incubation that tube which 
shows just complete hemolysis contains one amboceptor unit, and double 
this amount is used in making the main tests. If the serum has a titer of 
less than 1 : 500, it should not be used either in preparing the amboceptor 
slips or in conducting the reaction. 
Titration of Dried Amboceptor Paper— After the paper (S. & S. No. 
597) has been evenly saturated with immune serum and dried, the sheets are 
cut into 5 mm. strips and standardized by placing increasing lengths of 
paper into a series of tubes as follows: 
Tube 1: 1 mm. paper + 0.1 c.c. complement (40 per cent.) (5 drops) 
+ 1 c.c. corpuscle suspension (1 per cent.). 
Tube 2: 2 mm. paper + 0.1 c.c. complement (40 per cent.) (5 drops) 
+ 1 c.c. corpuscle suspension (1 per cent.). 
Tube 3: 3 mm. paper + 0.1 c.c. complement (40 per cent.) (5 drops) 
+ 1 c.c. corpuscle suspension (1 per cent.). 
Tube 4: 4 mm. paper + 0.1 c.c. complement (40 per cent.) (5 drops) 
+ 1 c.c. corpuscle suspension (1 per cent.). 
Tube 5: 5 mm. paper + 0.1 c.c. complement (40 per cent.) (5 drops) 
+ 1 c.c. corpuscle suspension (1 per cent.). 
Tube 6: 6 mm. paper + 0.1 c.c. complement (40 per cent.) (5 drops) 
+ 1 c.c. corpuscle suspension (1 per cent.). 
One cubic centimeter of saline solution is added to each tube, and the 498 
COMPLEMENT FIXATION IN SYPHILIS 
mixture shaken gently and incubated at 37° C. for two hours or one hour 
in a water-bath. At the end of this time the tube just completely hemolyzed 
contains one amboceptor unit, and in performing the test double this amount is 
used. (See Fig. 142.) 
This titration should always be conducted before the actual tests are 
set up, as is the rule in conducting the Wassermann reaction. When the 
titer of the paper is known, it may not be necessary to set up all the tubes 
of the foregoing series, as a few only are necessary to determine if the same 
amount of paper as was used in the previous tests will suffice with the new 
complement and corpuscle suspension at hand. 
All titrations and the main tests may be conducted in a water-bath 
(37° C.). With the aid of a good thermometer a satisfactory bath is easily 
improvised. In fact, I believe that better results are secured in the water-bath 
than in the incubator. It is possible, therefore, to conduct these reactions 
in a perfectly satisfactory manner without the aid of an expensive incubator. 
4. Antigen.—Acetone-insoluble lipoids (Noguchi) are to be used ex- 
clusively if the tests are conducted with active serums. When heated serums 
are used, any extract may be employed, as in making the Wassermann reac- 
tion, but the same antigen gives excellent results, and I use it exclusively 
in conducting the Noguchi reaction with both active and inactivated serums. 
The antigen must be titrated as usual, and its anticomplementary, he- 
molytic and antigenic properties determined. According to Noguchi, an 
extract is satisfactory if it is antigenic in 0.02 c.c. of a 1 : 10 dilution, and 
not anticomplementary or hemolytic in amounts under 0.4 c.c. (1 : 10). 
In conducting the tests five times the antigenic unit, or 0.1 c.c., is employed; 
this dose is at least four times smaller than the anticomplementary unit, 
and is therefore safe and satisfactory. 
The antigen is best preserved in methyl alcohol, as described on p. 457. 
Dried on paper and properly preserved in sealed tubes in a cold place it will 
retain its activity for several months, but as a general rule it is best to use 
fresh emulsions of the alcoholic solution. 
Titration of Antigen.—The anticomplementary, hemolytic, and antigenic 
doses of an extract are determined in the same general manner as was 
described under the Wassermann reaction. 
1. Anticomplementary Titration.—A portion of the stock alcoholic solu- 
tion of acetone-insoluble lipoids is diluted with 9 parts of saline solution. 
This is the emulsion that is employed in conducting the Noguchi reaction, 
and contains 0.3 per cent, of the original lipoidal substances. 
Sufficient emulsion for these titrations is prepared by diluting 0.4 c.c. of 
the alcoholic solution with 3.6 c.c. of saline solution. 
Into a series of seven small test-tubes place increasing amounts of this 
emulsion as follows: 0.1, 0.2, 0.3, 0.4, 0.6, 0.8, 1.0 c.c., add 0.1 c.c. (5 capil- 
lary drops) complement (40 per cent.) to each; also 1 c.c. of a 1 per cent, 
suspension of corpuscles and sufficient saline solution to make the total 
volume in each tube about 2 c.c. Incubate at 37° C. for one hour (one- 
half hour in water-bath), and add two units of amboceptor. Shake the 
tubes gently and re-incubate for two hours (one hour in water-bath). That 
tube showing beginning inhibition of hemolysis contains the anticom- 
plementary dose, which should not be under 0.4 c.c. In the tubes containing 
the larger doses slight hemolysis may be noticed, which is evidence of the 
hemolytic action of the extract. 
An eighth tube should be included, containing 0.1 c.c. diluted com- 
plement, two units of amboceptor, and 1 c.c. of the corpuscle suspension. 
This is the hemolytic control and should show complete hemolysis. Fig. 142.—Titration of Antihuman Hemolytic Amboceptor. 
Shows complete hemolysis with 4 mm. of paper ( unit). 
Fig. 143.—Noguchi Modification of the Wassermann Reaction. 
Positive reactions in tubes 1 and 3. NOGUCHI METHOD 
499 
2. Antigenic Titration.—Since the extract is likely to have a high anti- 
genic value, it is necessary to dilute the antigen still further by placing 
0.5 c.c. of the foregoing emulsion in a test-tube and adding 4.5 c.c. of saline 
solution (1 : 100 dilution of the antigen). Into a series of six test-tubes 
place increasing amounts of this emulsion as follows: 0.05, 0.1, 0.2, 0.3, 
0.4, 0.5 c.c. To each tube add 4 drops (0.08 c.c.) of inactivated or 1 drop 
(0.02 c.c.) of fresh active syphilitic serum; also 0.1 c.c. (5 capillary drops) 
of complement (40 per cent.) and 1 c.c. of 1 per cent, corpuscle suspension. 
Then add sufficient salt solution to bring the total up to 2 c.c. 
Two controls should be included: (1) The serum control, containing the 
dose of serum, 0.1 c.c. of the complement, 1 c.c. of corpuscle suspension, 
and saline solution; (2) the hemolytic control, containing at this time 0.1 c.c. 
of complement, 1 c.c. of corpuscle suspension, and sufficient saline solution. 
All tubes are incubated for one hour at 37° C. (one-half hour in water- 
bath), after which two units of amboceptor are added to each tube. The 
tubes are then shaken gently and re-incubated for two hours (one hour in 
water-bath). At the end of this time the two controls should be completely 
hemolyzed, and in the series proper that tube showing just complete inhibi- 
tion of hemolysis contains one antigenic unit. Usually the first and second 
tubes show some inhibition of hemolysis, and in the third and other tubes 
hemolysis is completely inhibited. In this case 0.2 c.c. of this emulsion 
would be one antigenic unit (= 0.02 c.c. undiluted antigen); five times this 
amount equals 0.1 c.c. of the first emulsion (1 : 10), which is the amount to 
be used in making the main tests. 
Unless the antigen shows signs of deterioration, these titrations need be 
made only about once a month. 
If paper antigen is employed, both titrations are conducted in exactly 
the same manner by adding increasing lengths of a strip of dried paper 5 
mm. in width, impregnated with the antigen. 
5. Fluid to Be Tested.—If active serum is used, it should be fresh, free 
from hemoglobin, and preferably not over twenty-four hours old. The dose 
is 0.02 c.c., or 1 capillary drop; inactivated serums are used in doses of 
0.08 c.c., or 4 capillary drops. Cerebrospinal fluid is used unheated in doses 
of 0.2 c.c., or 10 capillary drops. Sufficient blood for this test may be col- 
lected in a Wright capsule. (See p. 17.) 
6. The Test.—The complement, amboceptor, antigen, and serums may 
be conveniently measured by drops from a capillary pipet (Fig. 2). In 
placing a drop the pipet should be held uniformly at an angle of 45 degrees, 
or else the size of the drop will differ, depending on whether the pipet is held 
vertically or horizontally. 
Arrange four pairs of small test-tubes (10 by 1 cm.) in a rack containing 
two rows of holes. Into each of the tubes on the front row place 5 drops 
(0.1 c.c.) of antigen emulsion (alcoholic solution, 1 part, with saline solution, 
9 parts); then add 5 drops (0.1 c.c.) of complement (40 per cent.) to all the 
tubes. Into each of the first pair of tubes place 1 drop (0.02 c.c.) of active 
or 4 drops (0.08 c.c.) of inactivated patient’s serum, and mark the front 
tube with the patient’s name. To each of the second pair of tubes add an 
equal amount of syphilitic serum known to give a positive reaction (positive 
control), and to each of the third pair add normal serum known to give a 
negative reaction (negative control). Mark the tubes in the front row of 
each pair respectively. The front tube of the fourth pair is the antigen 
control, and the rear tube the hemolytic control, and each should be so 
labeled. Into each tube place 1 c.c. of the 1 per cent, corpuscle suspension 
and 1 c.c. of saline solution, making the total volume in each tube about 500 
COMPLEMENT FIXATION IN SYPHILIS 
2 c.c. Shake each tube and incubate at 37° C. for one hour (half an hour in 
the water-bath). At the end of this time add two units of amboceptor to 
each tube, shake gently, and re-incubate for two hours (one hour in the 
water-bath). During this time the tubes should be shaken gently once or 
twice to break up any masses of agglutinated corpuscles. 
The following chart, after Noguchi, illustrates the various steps to 
be taken in making the test with one patient’s serum. Of course, any 
number of serums may be examined with the same controls (Fig. 143). 
Set tor 
Diagnosis. 
Positive 
Control Set. 
Negative 
Control Set. 
Antigen and Hem- 
olytic Controls. 
.2 
M 
P 
QJ 
X 
X 
c3 
2. 
4. 
6. 
8. 
o 
X* 
P 
■+-> 
u, 
QJ 
Unknown serum: 
Positive serum: 
Normal serum: 
Hemolytic con- 
U 
d 
1 drop.1 
1 drop. 
1 drop. 
trol: 
-c 
QJ 
Complement: 5 
Complement: 5 
Complement: 5 
Complement: 5 
C 
drops. 
drops. 
drops. 
drops. 
o 
*-i 
O 
3 
1 per cent, cor- 
1 per cent, cor- 
1 per cent, cor- 
1 per cent, cor- 
cL 
X 
puscle suspen- 
puscle suspen- 
puscle suspen- 
puscle suspen- 
u 
CJ 
QJ 
C 
sion: 1 c.c. 
sion: 1 c.c. 
sion: 1 c.c. 
sion: 1 c.c. 
js-c 
J§ 
o 
Salt solution (q. 
Salt solution (q. 
Salt solution (q. 
Salt solution (q. 
d 
6 
s. 2 c.c.). 
s. 2 c.c.). 
s. 2 c.c.). 
s. 2 c.c.). 
O QJ 
fl 
u 
1. 
3. 
5. 
7. 
ih a) 
£ £ 
2 
p 
o 
u- 
CO 
Antigen: 5 drops. 
Antigen: 5 drops. 
Antigen: 5 drops. 
Antigen control: 
Unknown serum: 
Positive serum: 
Normal serum: 
Antigen: 5 drops. 
0 
c 
cn 
1 drop. 
1 drop. 
1 drop. 
o 
P 
Complement: 5 
Complement: 5 
Complement: 5 
Complement: 5 
CO 
o 
X 
drops. 
drops. 
drops. 
drops. 
O 
1 per cent, cor- 
1 per cent, cor- 
1 per cent, cor- 
1 per cent, cor- 
d 
3 
+-> 
puscle suspen- 
puscle suspen- 
puscle suspen- 
puscle suspen- 
QJ 
CN 
QJ 
sion: 1 c.c. 
sion: 1 c.c. 
sion: 1 c.c. 
sion: 1 c.c. 
d 
X 
X? 
d 
X 
Salt solution (q. 
Salt solution (q. 
Salt solution (q. 
Salt solution (q. 
P 
< 
P 
s. 2 c.c.). 
s. 2 c.c.). 
s. 2 c c.). 
s. 2 c.c.). 
c 
hH 
c 
hH 
Noguchi Modification of the Wassermann Reaction 
At the end of the second incubation, or after two hours more at room 
temperature, the tubes are inspected. The antigen and hemolytic system 
controls, as well as all the rear tubes or serum controls, should be completely 
hemolyzed. The first tube containing a known syphilitic serum shows 
inhibition of hemolysis; the front tube containing normal serum is completely 
hemolyzed; the front tube containing the patient’s serum shows complete 
inhibition of hemolysis (strong positive), varying degrees of inhibition 
(moderately or weakly positive), or is completely hemolyzed (negative). 
The results may be recorded and reported after the same manner described 
on p, 472. 
Craig2 employs a technic differing from the original Wassermann reac- 
tion in the use of a human hemolytic system in place of a sheep hemolytic 
system, and in a proportional reduction in the quantities of reagents used. 
Alcoholic extract of syphilitic liver is generally employed as antigen, although 
cholesterinized extracts of normal heart muscle have been found equally 
satisfactory.3 Vedder4 has also used Craig’s method in the military ser- 
Modification of Craig 
1 Four drops if serum is inactivated. 
2 War Department, Bulletin No. 3. 
3 Amer. Jour. Med. Sci., 1915, cxlix, 41. 
4 War Department, Bulletin No. 8. MODIFICATION OF CRAIG 
501 
vice, and it is widely used in the Army laboratories with satisfactory 
results. 
Corpuscles.—A 5 per cent, suspension of washed human corpuscles 
preferably those belonging to Group IV (Moss classification) as recom- 
mended by Williams.1 Dose, 0.1 c.c. 
Hemolysin.—Rabbit antihuman serum dried in paper. In a series of 
six small test-tubes place increasing amounts of paper as follows, measured 
in millimeters: 5x1,5x2, 5x3, 5x4, 5x5, and 5x6. To each tube add 
0.1 c.c. of 1 : 5 complement, 0.1 c.c. of 5 per cent, corpuscle suspension and 
0.9 c.c. salt solution. Incubate in a water-bath at 37° C. for one hour. A 
satisfactory paper should show complete hemolysis in a piece 5x5 mm. 
or less. 
Complement.—This is titrated each time tests are conducted. Dilute 
guinea-pig serum complement with 1| parts of normal salt solution and 
place the following amounts in a series of nine test-tubes: 0.02, 0.03, 0.04, 
0.05, 0.06, 0.07, 0.08, 0.09, and 0.1 c.c. Add two units of hemolysin paper, 
0.1 c.c. of 5 per cent, suspension of corpuscles, and 0.9 c.c. salt solution to 
each tube. Incubate in a water-bath at 37° C. for one hour. 
Antigens.—Alcoholic extract of syphilitic liver and alcoholic extract of 
normal human heart reinforced with 0.4 per cent, cholesterin are recom- 
mended. Each antigen diluted 1 : 10. 
Antigens are titrated for hemolytic activity in amounts of 0.05, 0.1, 
0.15, and 0.2 c.c. of 1 : 10 dilutions in a series of test-tubes with two units of 
complement, 0.1 c.c. of 5 per cent, suspension of corpuscles and 0.9 c.c. salt 
solution; water-bath incubation for one hour. None of the tubes should 
show hemolysis. 
The anticomplementary titrations are conducted by placing 0.05, 0.1, 
0.15, and 0.2 c.c. of 1 : 10 dilutions in a series of test-tubes with two units 
of complement and 0.9 c.c. of salt solution; water-bath incubation for one 
hour. Two units of hemolysin and 0.1 c.c. of 5 per cent, corpuscle suspen- 
sion are then added and the tubes re-incubated for one hour. There should 
be complete hemolysis in all tubes. 
The antigenic titrations are conducted in the same manner except that 
0.1 c.c. of inactivated luetic serum is placed in all tubes before the primary 
incubation. A satisfactory antigen is one giving complete inhibition of 
hemolysis in all tubes. A serum control should be included. 
Hemolytic system and corpuscle controls should be included in all 
titrations. 
If 0.05 c.c. of 1 : 10 antigen gives complete inhibition of hemolysis this 
amount is employed in the main tests; if 0.1 c.c. is the smallest amount 
giving complete inhibition this amount may be employed providing it is 
not anticomplementary in dose of 0.2 c.c. 
Serum.—Heated at 55° C. for one-half hour; dose, 0.1 c.c. 
Tests.—For each serum arrange three test-tubes. Place required amount 
of antigen of alcoholic syphilitic liver in the first tube, and the cholesterolized 
antigen in the second; the third tube is the serum control. Place 0.1 c.c. 
heated serum in each of the three tubes followed by two units of comple- 
ment and 0.9 c.c. of salt solution. Water-bath incubation for one hour 
followed by the addition of two units of hemolysin and 0.1 c.c. of 5 per cent, 
corpuscle suspension. Re-incubate for one hour, followed by settling of the 
corpuscles in a refrigerator overnight. 
Include controls with known positive and negative sera; also antigen, 
hemolytic system and corpuscle controls. 
1 Jour. Exper. Med., 1920, 32, 159. 502 
COMPLEMENT FIXATION IN SYPHILIS 
Readings are made as follows: 
+ + = complete inhibition of hemolysis. 
+ = anything between complete and 50 per cent, inhibition of 
hemolysis. 
=t = anything between 50 per cent, inhibition and complete he- 
molysis. 
— .= complete hemolysis. 
Modification of Hecht-Weinberg-Gradwohl 
In conducting the syphilitic reaction Hecht1 utilizes not only the natural 
antisheep amboceptor in human serum but also the native hemolytic com- 
plement. The serum must be perfectly fresh and, of course, is used un- 
heated. This modification has been said to be more delicate than the 
Wassermann reaction because none of the syphilis antibody is destroyed or 
complementoids produced, as presumably will occur during inactivation 
(heating) of a serum. In my experience this test has proved quite delicate, 
but is open to the same error that may occur whenever an active serum is 
used with a crude alcoholic organic extract as antigen—i. e., the appearance 
of false positive or proteotropic reactions. As with the Noguchi reaction, 
using active serum, a negative Hecht-Weinberg test has considerable diag- 
nostic value in excluding syphilis; a positive reaction must be, however, 
carefully controlled. In performing the test I always use an extract of 
acetone-insoluble lipoids as antigen. 
Since in the original Hecht-Weinberg2 test there is no way of determining 
beforehand the amount of sheep corpuscles a serum may hemolyze, Grad- 
wohl3 has modified the technic so that the hemolytic index of each serum is 
determined before the corpuscles are added to the main tubes. I conduct 
this test as follows4: Nine small sterile test-tubes (10 x 1 cm.) are arranged 
for each serum and properly labeled; into each is placed 0.1 c.c. of fresh 
unheated serum. To the first five tubes are added respectively 0.1, 0.2, 
0.3, 0.4, and 0.5 c.c. of a 5 per cent, suspension of washed sheep cells; to the 
sixth, seventh, and eighth tubes are added 1, 2, and 3 units of antigen re- 
spectively, as determined by titration. The last or ninth tube serves as the 
serum control. Sufficient normal salt solution is added to each tube to bring 
the total volume to 1 c.c. 
After one hour’s incubation at 37° C. the hemolytic index of each serum 
is read in the first five tubes of each series (that is, the largest dose of cor- 
puscles just completely hemolyzed by each serum) and one-half the indicated 
doses of corpuscles added to the remaining four tubes of each series. After 
re-incubation for one-half to one hour, according to the hemolysis of the 
controls, the results are read and recorded as in the Wassermann reaction. 
The antigen titrations are very important because the activity of human 
serum is quite sensitive to the inactivating influence of tissue extracts. No 
antigen should be employed in this test without preliminary titration with 
human serum to determine its anticomplementary and antigenic units. The 
anticomplementary titration is conducted by placing in a series of eight test- 
tubes the following amounts of antigen of acetone-insoluble lipoids diluted 
1 : 100: 0.2, 0.3, 0.4, 0.6, 0.8, 1.0,1.2, 1.5 c.c., and 0.1 c.c. of the fresh and 
mixed sera of two non-syphilitic persons; normal salt solution is added to 
2 c.c. and the tubes incubated for an hour at 37° C., when one-half the index 
1 Wien. klin. Wchn., 1909, xxii, 265. 
2 Wien. klin. Wchn., 1908, 21, 1742; ibid., 1907, 22, 265; ibid., 1909, 22, 338. 
3 Jour. Amer. Med. Assoc., 1914, 63, 240. 
4 Jour. Immunol., 1916, 2, 23. THE WASSERMANN REACTION IN NON-SYPHILITIC CONDITIONS 
503 
of cells for the serum is added to each tube. After a second incubation of 
an hour the anticomplementary unit is read as the smallest dose of antigen 
producing inhibition of hemolysis. The antigenic titration is conducted in 
the same manner with the serum of one or two syphilitic persons and the 
following doses of antigen (1 : 100): 0.05, 0.1, 0.15, 0.2, 0.25, and 0.3 c.c. 
The antigenic unit is the smallest amount giving complete inhibition of 
hemolysis. 
The Clinical Significance of the Wassermann Reaction 
The Necessity for a Proper Understanding of the Wassermann Reaction.— 
In their attitude toward the Wassermann test physicians may be divided 
into those who have considerable confidence in the reliability of the test 
and especially in the significance of a positive reaction both in the diagnosis 
of syphilis and as a guide to treatment, those who have little or no faith 
in the test in either a positive or negative way and those who have found it 
generally satisfactory providing weakly positive reactions are disregarded. 
Probably there is no other laboratory test the subject of as much disagree- 
ment between clinician and laboratorian; this is to be particularly regretted 
because no other strictly laboratory test possesses an equal importance or 
greater possibilities for good as a diagnostic means of a disease which is 
everywhere on earth among all classes of human beings, irrespective of sex 
and age. The disagreement and misunderstanding of the true value of the 
Wassermann reaction are due to biologic and technic sources of error, a 
correct understanding of which by both clinician and laboratorian, is essential 
for the proper practice and interpretation of this valuable test. 
The Biologically Non-specific Nature of the Wassermann Reaction.—In 
the first place and as discussed in the preceding chapter on the nature of the 
antibody concerned, the complement-fixation test in syphilis as ordinarily 
practised is not biologically specific as are other complement-fixation tests. 
The reason is that the so-called “antigens” are not true antigens, that is, 
need not and usually are not, prepared of syphilitic tissues or the Treponema 
pallida. In all other complement-fixation tests the antigen employed must 
be an extract of the microparasite or other protein producing the disease. 
Instead of using an extract or suspension of Treponema pallida for this 
test, the so-called “antigen” is nothing more than extract of tissue lipoids, 
and while these may be secured from either syphilitic or healthy tissues, the 
best “antigens” are probably prepared by extracting non-syphilitic tissues, 
as beef heart muscle. 
The reaction in syphilis depends upon the peculiar nature of the so- 
called antibody in serum and spinal fluid which is endowed with the property 
of fixing complement in the test-tube in the presence of these lipoids. 
Therefore, since the reaction in syphilis is not biologically specific be- 
cause the antigen is not, the question arises: Does the reaction possess prac- 
tical specificity for syphilis? The answer depends upon whether the peculiar 
lipotropic antibody-reagin is found in diseases other than syphilis. 
The Occurrence of the Wassermann Reaction in Non-syphilitic Conditions? 
the Practical Specificity of the Wassermann Reaction 
Unfortunately, the reaction is beset by so many technical errors that a 
review of the literature, and especially of the early literature, shows results 
that are quite confusing and contradictory. Following the original com- 
munications of Wassermann and Detre, and especially after it was demon- 
strated that the antigen need not be biologically specific, the subject was 504 
COMPLEMENT FIXATION IN SYPHILIS 
extensively investigated by various observers, who reported securing posi- 
tive reactions in many different diseases, results that we now know' must 
have been due largely to technical errors. At present it is known that 
positive Wassermann reactions may occur in a few diseases other than 
syphilis, but not to the extent that earlier investigators would have us 
believe. In most of the diseases yielding positive reactions the clinical 
symptoms are so marked that they may readily be differentiated from 
syphilis, and accordingly the Wassermann reaction is of unequaled and in- 
calculable diagnostic value. 
Positive reactions have been reported in frambesia (yaws), in which 
the causal micro-organism, the Spirochasta pertenuis, is morphologically 
almost indistinguishable from Spirochseta pallidum. In leprosy of the 
tuberous type positive reactions are frequently found, but in my experi- 
ence these occur only occasionally and only in those lepers who are likewise 
syphilitic. Positive reactions have been reported in cases of malaria during 
the febrile stage, when parasites are present, although the majority of cases 
react negatively. In my own series of 11 cases all the reactions were nega- 
tive. Positive reactions have also been found in some cases of relapsing 
fever and trypanosomiasis. 
In scarlet fever the Wassermann reaction is uniformly negative. Owing 
to the original communication of Much and Eichelberg, however, in the 
minds of many this disease is prominently associated with a positive reaction. 
While it is true that a positive reaction is very rarely found, it is almost 
impossible entirely to exclude a diagnosis of congenital lues, at least in some 
of these cases. In my own series of 250 cases examined by the Wassermann 
and Noguchi methods, with antigens of alcoholic extract of syphilitic liver 
and acetone-insoluble lipoids, the reactions were positive in 5 cases, or 2 
per cent. Similar results have been secured by Boas, Browning and Mac- 
kenzie, and others, so that it may be said that the reaction in scarlet fever is 
uniformly negative. 
Normal cerebrospinal fluid or the fluid from persons with ordinary 
non-syphilitic diseases reacts negatively. Positive reactions occur in 
frambesia. 
Positive reactions have also been reported in tuberculosis of the lungs 
and especially with cholesterolized antigens. This is certainly not my 
experience. Wassermann tests conducted with a large series of cases are 
almost sure to show a small percentage of positive reactions, but these 
occur among syphilitic individuals with tuberculosis. 
Positive reactions have also been reported in late pregnancy, that is, the 
serum of a woman reacts positively during the late months of pregnancy 
and negatively after delivery. I believe this may occur occasionally with 
antigens containing 0.4 per cent, cholesterol, but not with plain antigens 
or those containing 0.1 or 0.2 per cent, cholesterol. In so far as my owm ex- 
perience is concerned, pregnancy of itself does not yield a falsely positive 
reaction at any stage. It is possible and, indeed, probable that pregnancy 
may stimulate latent foci of syphilitic infection into activity, resulting in 
the serum yielding positive Wassermann reactions, but these generally 
persist for several weeks after delivery before subsiding again into latency 
or for longer periods of time and are not to be interpreted as falsely posi- 
tive. In my opinion a positive reaction in pregnancy wrhen the test has 
been technically correct indicates the presence of syphilis. 
Falsely positive reactions have also been reported as occurring in non- 
syphilitic persons with diabetes mellitus, jaundice, and uremia. In these 
conditions anticomplementary substances may be present in the serum THE WASSERMANN REACTION IN NON-SYPHILITIC CONDITIONS 
505 
increasing the chances of securing falsely positive reactions, but in so far as 
my own experience is concerned the reactions are uniformly negative in the 
absence of syphilis. 
Biologic and Technical Reasons for Positive Reactions in Non-syph- 
ilitic Diseases.—In frambesia tropica (yaws) the positive Wassermann reac- 
tions are apparently due to the same kind of lipotropic “reagin” as occurs in 
syphilis; this is to be expected because the Treponema pertenue causing this 
disease is biologically closely related to Treponema pallida and subject to 
the destructive influence of arsphenamin to even greater degree. It is 
possible that the positive reactions occurring in relapsing fever and try- 
panosomiasis are due to the presence of similar “reagins” and it is to be 
expected that positive reactions may occur in other spirochetic infections. 
In other words, there are biological reasons for positive Wassermann reactions 
in these non-syphilitic diseases, but in localities where frambesia and relapsing 
fever do not occur at all or but seldom, the practical specificity of the Wassermann 
reaction for syphilis is well established. 
The complement-fixation test for syphilis is so complicated that there 
are numerous sources of technical error, but these can be reduced to a 
minimum and rendered negligible by experience and skill. Probably the use 
of cholesterolized “antigens” are most important in this connection. It 
cannot be denied that alcoholic extracts saturated with cholesterin and 
employed in relatively large amounts may yield falsely positive reactions 
with a small percentage of normal sera. But I am convinced that these 
results can be prevented by proper technical procedures. It is not necessary 
to use more than 0.2 per cent, cholesterol; the extracts should be frequently 
titrated and none should be used unless the dose of two or more antigenic 
units employed is at least one-fifth and preferably one-tenth of the anti- 
complementary unit. 
Biologic and Technical Reasons for Falsely Negative Reactions in 
Syphilis.—It is readily understood that since the Wassermann reaction is 
due to the presence of “reagin” in the blood or spinal fluid, or both, that 
sufficient of this substance must be present before true positive reactions 
may be obtained. Therefore, in the primary stage of syphilis the reaction 
may be negative until the body cells are sufficiently stimulated to produce 
demonstrable amounts of this “reagin.” As stated in the preceding chapter 
I believe this “reagin” is produced by the tissue-cells surrounding the spiro- 
chetes. wherever they may be located and for this reason the “reagin” is to 
be found in the secretions of the chancre before it may be detected in the 
blood. For the same reason the “reagin” may be found in the spinal fluid 
and not in the blood of certain cases of syphilis. 
Likewise in cases of latent syphilis either acquired or congenital, the 
spirochetes may be so quiescent that demonstrable amounts of “reagin” 
are not being produced. A certain amount of this substance must be pres- 
ent for positive reactions and in some cases of syphilis it would appear that 
the infection is so mild and latent that too small amounts are produced for 
detection in the complement-fixation test. 
Of course falsely negative reactions may be due to faulty technic, that 
is, the technic may not be sufficiently sensitive. Every serologist endeavors 
to avoid falsely positive reactions and arranges his technic with this in 
mind; all will agree that it is better to miss the occasional case of syphilis 
than to secure a falsely positive reaction, but this may be carried too far. 
It is possible to render the complement-fixation test for syphilis very sensitive 
within the bounds of specificity and this should be the aim of all who con- 
duct the test. 506 
COMPLEMENT FIXATION IN SYPHILIS 
Negative Wassermann Reactions Do Not Exclude the Possibility of Syph- 
ilis.—Since there may be a biologic reason for negative Wassermann 
reactions in syphilis and numerous technical reasons, too much emphasis 
cannot be laid upon a negative reaction and especially a single negative 
result, in excluding the disease. This is especially true during the primary 
and latent stages (both acquired and congenital infections) of the disease when 
“reagin” production has not occurred to sufficient degree to permit its detection 
in the blood and spinal fluid by complement-fixation tests. The syphilis com- 
plement-fixation test as ordinarily practised is not too sensitive; rather it 
is not sensitive enough, always bearing in mind that falsely positive re- 
actions due to faulty technic are to be avoided. 
In my experience the reaction is but seldom negative in the presence 
of active syphilis after the primary stage. For this reason consistently 
negative reactions in the presence of an active pathologic process indicates 
the non-syphilitic nature of the latter with a great degree of accuracy pro- 
viding always that the technic is sufficiently sensitive and correct. 
Positive Spinal Fluid and Negative Blood Reactions.—It is now well 
established that in some cases of syphilis and especially those with involve- 
ment of the brain and spinal cord, the Wassermann test with the blood-serum 
may yield a negative reaction while the spinal fluid reacts positively. In 
long-standing and latent infections and especially in cases presenting clinical 
evidences of involvement of these tissues, a negative serum reaction possesses 
little or no value in excluding syphilitic infection. The spinal fluid requires 
examination and should be subjected not only to the complement-fixation 
test, but likewise to the colloidal gold test, protein determinations, and the 
counting of cells. 
Furthermore, in many cases of treated syphilis the serum reaction may 
be negative when the spinal fluid reacts positively; for this reason an ex- 
amination of the latter should be included before conclusion can be reached 
on the cure of syphilis in so far as this opinion is based upon serologic re- 
actions. 
The reasons for the positively reacting spinal fluid and negatively reacting 
serum are not understood. Since spinal fluid is used in the Wassermann 
test unheated and in amounts four to five times more than is permissible 
with serum, I believe that the chances for detecting the “reagin” in the 
spinal fluid are much greater and constitute the reasons explaining the 
reactions in some cases. Furthermore, it may be that “reagin” production 
occurs largely in the tissues of the brain and cord in syphilis of the central 
nervous system, and that it is thrown off into the spinal fluid rather than the 
blood, and that absorption of the “reagin” from the spinal fluid into the blood 
is reduced by pathologic processes. 
A Positive Wassermann Reaction is Not Always an Indication that a 
Particular Lesion is Syphilitic.—Simply because an individual is known or 
suspected as being syphilitic does not exclude the possibility of the presence 
of other diseases. Not at all infrequently a non-syphilitic tumor or ulcer 
occurring in a syphilitic is regarded a priori as luetic; I have known this to 
occur in several cases of tuberculous ulcers of the larynx in syphilitic indi- 
viduals. A positive Wassermann reaction, therefore, is an indication of 
syphilis, but does not necessarily mean that a particular lesion is syphilitic. 
Syphilitic individuals are especially liable to attribute every ache and 
pain to syphilis; the “mental scars” of this disease are frequently incurable, 
but the physician must guard against the error of interpreting every dis- 
ease process in a syphilitic as caused by Treponema pallidum and more 
especially guard against reaching such a conclusion on the basis of a positive THE WASSERMANN REACTION IN VARIOUS STAGES OF SYPHILIS 
507 
Wassermann reaction. Syphilis does not confer an immunity against other 
bacterial infections and pathologic processes; rather it may predispose to 
carcinoma, tuberculosis, and other diseases. 
Biologic Reasons for the Unexpectedly Positive Wassermann Reac- 
tion; the Significance of Weakly Positive Reactions.—Not infrequently 
when the Wassermann test is conducted as a matter of routine the clinician 
is greatly surprised by a totally unexpected positive reaction. This may 
be caused by technical error, but not infrequently is due to an error on the part 
of the physician, because syphilis may escape clinical detection, but yield a true 
positive Wassermann reaction. 
Cases of congenital and acquired syphilis insufficiently treated or not at 
all, but in the latent stages with indefinite lesions and symptoms or prac- 
tically none at all, may occur in the practice of every physician irrespective 
of his special field of work and yield true positive Wassermann reactions. 
It is now so well known that syphilis may manifest itself in so many different 
ways that it may escape clinical detection. As Osier is so frequently quoted: 
■“Know syphilis in all its manifestations and relations and all other things 
clinical will be added unto you.” I am mentioning this phase of the sub- 
ject in order to advise the physician against lightly brushing aside the 
unexpected or weakly positive reaction reported by a good laboratory, 
regardless of the sex or respectability of his patient; these results are not 
always due to technical errors in the laboratory and not infrequently indi- 
cate the presence of syphilis. 
I believe that as clinical and pathologic knowledge of syphilis is developed 
more and more emphasis will be placed upon the significance of weakly positive 
and unexpectedly positive reactions. ■ Furthermore, in the present state of 
our highly perfected technic every laboratory conducting the test should 
have enough faith in its results to place significance and confidence upon 
its weakly positive reactions. 
Technical Sources of Error in the Wassermann Test and Variation in 
Results in Different Laboratories.—Confidence in the Wassermann reaction 
has been destroyed for many physicians by reason of securing varying re- 
ports from different laboratories with portions of the blood of the same 
person. This is very much to be regretted and constitutes an important 
reason for the adoption of a standardized test of superlative and proved 
merit. 
These results, however, are not unexpected by serologists. The influence 
of technic upon the reactions is very great and especially in reference to the 
kind and amount of antigen employed, the hemolytic system and kind and 
duration of primary incubation, not to mention individual variations in 
skill and experience on the part of different laboratorians. 
These discrepancies in reactions from different laboratories do not under- 
mine the real value o] the complement-fixation test in syphilis; they are largely 
the result of numerous modifications of technic being employed. To the 
average physician a Wassermann test is a Wassermann test irrespective by 
whom or by which method it is conducted; this is far from being true, and 
the physician should exercise care in the selection of the laboratory con- 
ducting these examinations. 
The Wassermann Reaction in the Various Stages of Syphilis 
In general terms a sensitive complement-fixation test may be expected 
in my experience to yield approximately the following percentages of posi- 
tive reactions with serum in the different stages of syphilis: 508 
COMPLEMENT FIXATION IN SYPHILIS 
Per cent. 
Primary stage 92 
Primary latent period (healing chancre; pre-eruptive stage) 92+ 
Secondary stage 98-100 
Secondary latent stage (after secondary lesions have subsided; usually some treat- 
ment had been received) 86+ 
Tertiary stage with active lesions (exclusive of nervous system) 96+ 
Tertiary stage with lesions of the central nervous system: 
(a) Paresis 96-100 
(ib) Tabes dorsalis 90+ 
Active prenatal (first year) 100 
Latent prenatal (after second year) 80 
1. In Primary Syphilis.—As would be expected, a certain degree of tissue 
change must occur before syphilis reagin appears in the blood. Even with 
the best technic there is a limit to the sensitiveness of the Wassermann reac- 
tion, so that while the reagin may be produced at the very onset of an 
infection, time and further tissue changes are required before sufficient 
reagin is produced to yield a complement-fixation reaction. As shown by 
Klauder and the writer,1 however, tests conducted with chancre secretions 
are sometimes positive when the serum reactions are negative. It is pos- 
sible that tests of this kind may possess diagnostic value, although there is 
usually some difficulty in obtaining sufficient fluid from the chancre for the 
test. 
Since, therefore, the results of the Wassermann reaction in primary 
syphilis are dependent upon the virulence of the infection, the time at 
which the reaction is made, and the delicacy of the technic, it is not sur- 
prising that the results of different investigators vary in the proportion of 
positive reactions obtained. While positive reactions have been said to 
have been secured before the appearance of the initial lesion, these are 
rare, and there is always the likelihood that an earlier infection was over- 
looked. A careful review of our own work and the literature upon this 
subject establishes the following: 
(a) A positive reaction may be secured during the first week after the 
appearance of the chancre. Craig has reported a positive reaction occurring 
five days after the appearance of the initial lesion. Levaditi, Laroche and 
Yamanouchi, and others have recorded many positive reactions occurring 
in ten days or more after the chancre made its appearance. 
(b) In general, in primary syphilis the Wassermann reaction will be 
positive in about 80 to 92 per cent, of cases; where cholesterinized extracts 
are used as antigens, or with the Noguchi system, using active serum, the 
reactions are secured earlier and in a larger percentage of cases. Craig2 
has reported 34 per cent, positive reactions during the first week after the 
appearance of the chancre; 57 per cent, during the second week; 67 per cent, 
during the third week; 76 per cent, during the fourth week, and 80 per cent, 
during the fifth week. 
(c) The cerebrospinal fluid of persons in the primary stage of syphilis 
has always reacted negatively (Boas). 
Microscopic vs. Serum Tests.—It is generally agreed that a diagnosis 
should be made as early as possible, and vigorous treatment instituted. A 
Wassermann reaction may be performed, and if it shows a positive result, 
this indicates the presence of syphilis, even if the lesion under suspicion is 
not specific, the reaction being due to a previous infection. A negative 
reaction, however, does not exclude syphilis, and if it is at all possible, a 
1 Arch. Dermat. and Syph., 1922, 5, 566. 
2 Amer. Jour. Med. Sci., 1915, cxlix, 41. THE WASSERMANN REACTION IN VARIOUS STAGES OF SYPHILIS 
509 
microscopic examination, using the dark-ground illuminator, should be made 
for the treponema. In primary syphilis a microscopic examination of the 
secretions of the lesion by a competent person is usually more valuable than 
the serum test; as a general rule, both examinations should be made, espe- 
cially with patients in whom the chancre is almost healed or atypical. 
Klauder1 has observed a series of cases especially valuable for bringing 
out the comparative values of microscopic and complement-fixation tests in 
primary syphilis; it is to be noted that the microscopic examinations were 
made with the dark-field illuminator and not by staining methods, which are 
of inferior value. His results with 115 cases were as follows: 
Duration of chancre, 
Per cent, positive 
Per cent, positive 
days. 
dark field. 
Wassermann. 
One to ten 
93.9 
36.0 
Ten to twenty 
52.9 
64.7 
Twenty to thirty 
50.0 
70.0 
Thirty to forty 
60.0 
100.0 
Over forty 
30.0 
100.0 
It is worthy of note that the earlier the lesion, the more valuable the 
dark-field examination for diagnosis; as the lesion heals the Wassermann 
test increases in diagnostic value. The combined examinations will diag- 
nose practically all cases. Repeatedly negative dark-field examinations 
with repeatedly negative Wassermann reactions exclude the syphilitic 
nature of a sore with great accuracy. Due care must be exercised in the 
microscopic examination of sores on the lips and in the mouth, as mouth 
spirochetes may be mistaken for Treponema pallida. 
2. In Secondary Syphilis.—It is in untreated cases of secondary syphilis 
that the remarkable specificity of the Wassermann reaction is so well demon- 
strated. The initial lesion may have been inconspicuous and hence have 
been overlooked, and the secondary lesions may be quite mild and incon- 
clusive; in either case the Wassermann reaction will usually establish the 
diagnosis. 
(a) In untreated secondary syphilis the reaction is positive in from 
98 to 100 per cent, of cases. In the examination of 437 serums from untreated 
cases Boas has never had a negative reaction, and my own experience has 
been the same. Craig reports 96 per cent, positive reactions. 
(b) With the serums of patients who have received some treatment 
the percentage of positive reactions will be slightly lower. Of 310 such 
cases examined by Boas, 97.6 reacted positively. The influence of treat- 
ment upon the reaction is to be remembered, and a single negative reaction 
does not by any means exclude the possibility of syphilis. 
(c) The intensity of the reaction does not bear any direct relation to 
the severity of the infection: a mild infection with indefinite signs may 
react quite strongly and absorb a large number of units of complement, 
whereas a severe case may react quite mildly. 
(d) In secondary syphilis without cerebral symptoms the cerebrospinal 
fluid is practically always negative (Plaut, Boas, and Lind); conversely, 
cases showing cerebral involvement usually react positively. More recent 
work has shown that the cerebrospinal system is involved early and in a 
relatively large number of cases (Craig and Collins2). Udo J. Wile has 
found that about 30 per cent, of secondary syphilitics give a positive reac- 
tion with cerebrospinal fluid. 
1 Jour. Amer. Med. Assoc., 1919, lxxii, 694. 
2 Jour. Amer. Med. Assoc., 1914, lxii, 1955. 510 
COMPLEMENT FIXATION IN SYPHILIS 
3. In Tertiary Syphilis.—It is probably in tertiary syphilis that the 
Wassermann reaction has its greatest value. Lues is so diverse in char- 
acter, and may be responsible for so many diverse clinical conditions, that 
the reaction has become well-nigh indispensable as a diagnostic aid. There 
is no limit to the time following infection in which positive reactions may 
not be found. 
(a) In cases of untreated and active tertiary syphilis the reaction is 
positive in about 96 per cent, of cases. 
(b) In cases receiving more or less antispecific treatment the reactions 
are positive in about 75 per cent. In genetal, therefore, a positive reaction 
in tertiary syphilis may be expected in from 80 to 95 per cent, of cases. 
(c) In a large percentage of cases of syphilitic aortitis, aortic aneurysm, 
aortic insufficiency, gummas of various organs, etc., the reaction is positive 
and possesses great diagnostic value. 
(d) The Wassermann reaction has been especially valuable in the study 
of the so-called parasyphilitic diseases. In general paralysis or paresis the 
serum reacts positively in about 100 per cent, of cases, and the cerebrospinal 
fluid reacts positively in from 96 to 100 per cent. The final and conclusive 
proof of the syphilitic nature of this disease has been furnished by Noguchi 
and Moore, who found the Treponema pallidum in sections of the brain. In 
certain cases of general paralysis the blood-serum may react negatively, 
whereas with the cerebrospinal fluid the reaction is positive. 
The fact that the blood-serum of a patient with a nervous disease reacts 
positively does not necessarily indicate that the nervous disease is of syph- 
ilitic origin, as the reaction may be due to specific infection of some other 
structure; if, however, the cerebrospinal fluid also reacts positively, then it 
is almost certain that syphilitic infection of the central nervous system is 
present. 
In untreated and active cases of tabes dorsalis the blood-serum reacts 
positively in from 96 to 100 per cent, of cases. In treated cases the number 
of positive reactions drops to about 50 to 80 per cent.; in general, a positive 
reaction with the serums of tabetics may be expected in 90 per cent, of cases. 
With the cerebrospinal fluid the percentage of positive reactions is some- 
what lower, being about 85 per cent. The positive Wassermann reaction 
is less constant in locomotor ataxia than in general paralysis, due piobably 
to the fact that the former is more chronic and that intercurrent periods of 
arrest are more prone to occur. 
In cerebral syphilis the blood-serum, and particularly the cerebrospinal 
fluid, will react positively less frequently than in general paralysis. In 
some instances a positive reaction is found with the cerebrospinal fluid and 
not with the serum, a matter difficult to explain and believed to be due to 
the confining of the reacting substances in the subarachnoid space. On the 
other hand, the lesions are probably not brought in direct contact with the 
spinal fluid. 
There is much evidence to indicate that localization of syphilis in the 
nervous system is dependent upon a particular strain of Treponema pallidum; 
other strains appear to possess a special affinity for the visceral organs, 
bones, etc. 
(e) In tertiary syphilis not accompanied by lesions of the central nervous 
system the Wassermann reaction with cerebrospinal fluid may be positive 
in a relatively large percentage of cases. 
4. In Latent Syphilis.—In cases of latent syphilis the Wassermann reac- 
tion may constitute the only evidence of the existence of the disease, and 
prompt institution of treatment may prevent the development of tertiary THE WASSERMANN REACTION IN VARIOUS STAGES OF SYPHILIS 
lesions, which are so likely to follow. When the spirochetes are few in 
number and are dormant, there is little tissue destruction or alteration, 
and, as a result, so little reagin is frequently present in the body fluids that 
the Wassermann reaction will fail to detect the disease. 
(a) In 363 cases of early latent syphilis, or those included within a 
period of three years after infection, Boas found positive reactions in about 
40 per cent.; in latent cases of long standing, or in those following manifest 
tertiary lesions, the same investigator found 22 per cent, of positive reac- 
tions among those who had received proper treatment; of those receiving 
indifferent treatment, 74 per cent, reacted positively, giving a general aver- 
age of about 48 per cent. Craig has found 67 per cent, positive reactions in 
latent syphilis; Vedder, 80.7 per cent. 
(b) The reaction with cerebrospinal fluid depends upon whether or not 
the central nervous system is involved in the syphilitic process. Of 104 
latent cases of syphilis in whom the spinal fluid was examined by Altman 
and Dreyfus, positive reactions were found in about 10 per cent. 
5. Prenatal or Congenital Syphilis.—The Wassermann reaction has 
thrown considerable light upon the subject of congenital syphilis. While, in 
general, the majority of cases react positively, the results are largely de- 
pendent upon the time when the examinations are made, a fact brought 
out by highly instructive and systematic investigations of Boas and Thom- 
sen. These investigators divided their cases into three groups: (1) New- 
born children and their mothers; (2) two-year-old children; (3) older chil- 
dren with congenital syphilis. 
(a) Of 88 children born of syphilitic mothers and examined at birth, 
the reaction was positive in 31 and negative in 57 cases. Of the 31 positive 
cases, 4 showed no symptoms of syphilis for a period of observation covering 
from three to nine months, and it is possible that the syphilis reagin, and not 
the spirochetes, from the blood of the mother, passed into the circulation 
of the child; on the other hand, all 4 cases may have been examples of 
retarded congenital syphilis. The remaining 27 cases either developed 
symptoms of syphilis or died later with syphilitic manifestations in various 
organs. 
Of the 57 children reacting negatively at birth, 42 showed no symptoms 
of syphilis during a period of three months of observation; 2 died with evi- 
dences of syphilis in the internal organs; 13 developed symptoms after birth 
and gave positive reactions. 
It may therefore be stated that a Wassermann reaction of the mother 
and of the child at the time of birth in cases where syphilis of the mother 
is suspected has considerable prognostic value. A large majority of children 
reacting positively develop symptoms of syphilis; on the other hand, the 
majority reacting negatively remain healthy. While an examination of 
the mother alone does not warrant an absolutely definite prognosis for the 
child, in general it may be said that a positive reaction does not constitute 
a favorable prognostic sign for the child. 
(.b) The Wassermann reaction has also shed new light upon the inter- 
pretation of Colles’ law. Since the “apparently healthy mother of a syph- 
ilitic child could suckle the child without being infected, whereas the child 
is capable of giving syphilis to others,” the most logical conclusion to draw 
is that the mother was gradually immunized against syphilis during preg- 
nancy, whereas we now know that the majority of mothers show positive 
serum reactions and are really latent syphilitics; in not a few such instances 
tertiary lesions have developed at a later date. 
It is possible, however, for a syphilitic mother showing a positive Wasser- 
511 COMPLEMENT FIXATION IN SYPHILIS 
512 
mann reaction to give birth to a healthy child. Of 46 mothers whose 
children showed no evidences of syphilis over a period of obser\ ation o 
three months, 17 reacted positively. Of 81 mothers giving birth to syphilitic 
children, 61 reacted positively, and many of these would naturally, in former 
years, have been regarded as examples of Colles’ immunity and considered 
free of syphilis. In many instances the apparently healthy child of a 
syphilitic mother that could not be infected by the mother (Profeta s law) 
has been shown by the Wassermann reaction to be in reality a. case of re- 
tarded congenital syphilis, and that such children are nop immunized, 
during intra-uterine life, either passively or by means of pallidum toxins, 
against syphilis, as has been so generally believed in past years. In other 
words, there appears to be no lasting passive immunity in syphilis; it is 
doubtful if the toxins of pallidum can pass between mother and child and 
immunize one or the other without actual infection with the spirochetes 
themselves taking place; that most examples of so-called immunity in syph- 
ilis in both the mother (Colles’ law) and the child (Profeta’s law) are due to 
the actual presence of pallidum in the tissues and are really latent infections. 
(c) In manifest untreated congenital syphilis of children one year or o\ er 
in age the Wassermann reaction is positive in from 97 to 100 per cent, of 
cases. The clinical manifestations may be quite varied and clinically ill 
defined, so that the serum reaction possesses considerable diagnostic value. 
In most instances the reactions are quite strong, and while active treatment 
may improve local lesions, it is very difficult, indeed, to secure negatn e 
reactions 
(,d) In congenital mental deficiency and epilepsy the Wassermann reaction 
shows that syphilis plays a larger part in the etiology of this condition than 
is generally supposed. A not inconsiderable proportion of cases are of 
infectious origin, and that infection is syphilis. In Little s disease, which 
is regarded as due to meningeal hemorrhage incidental to. injury recei\ ed 
during labor, the serum reactions have shown that not infrequently the 
hemorrhage has a syphilitic origin. 
Citron originally observed that during the mercurial treatment of syph- 
ilis the Wassermann reaction gradually became weaker, and finally dis- 
appeared. He also found that treatment was best governed by the serum 
reaction, and that it should be persisted in until a negative reaction was 
secured. His observations have in the main been abundantly confirmed 
by various observers the world over, although the extensive series of ob- 
servations now on record have given us a fuller understanding ol its prin- 
ciples. 
The Wassermann reaction is the most constant and delicate single symp- 
tom of syphilis, usually the last to disappear under treatment and the first to 
reappear if complete sterilization has not been accomplished. It is now quite 
generally believed that a persistently positive reaction indicates the pres- 
ence of living spirochetes, and that treatment should be continued until the 
blood reacts negatively. The reports of observers from all parts of the world 
indicate quite clearly and conclusively that the schematic, symptomatic, 
intermittent, and hard-and-fast rules of treatment of former days are not 
sufficient. They would also tend to show that the Wassermann reaction 
is the most delicate symptom and the last to disappear, and that treatment 
should be continued until this reaction disappears entirely and permanently. 
It has been abundantly proved, however, that in syphilis a single negative ieac- 
The Effect of Treatment Upon the Wassermann Reaction THE EFFECT OF TREATMENT UPON THE WASSERMANN REACTION 
513 
tion is not sufficient or definite evidence that a cure has been effected, for the 
disease may recur after treatment is discontinued, at least to the extent that the 
Wassermann reaction reappears, followed by clinical manifestations. It is 
necessary, therefore, that successive examinations be made during a period of 
at least two years, and off and on during the remainder of life. Recent work 
indicates that certain strains of Spirochseta pallida have an apparent selective 
affinity for the tissues of the central nervous system; the Wassermann reac- 
tion with blood-serum may be negative, whereas with the cerebrospinal fluid 
it may be positive. In cases, therefore, of tertiary syphilis, at least, it is 
advisable to examine the spinal fluid and continue treatment in case it shows 
a positive Wassermann reaction. 
It should be the object of treatment, in every case, not only to dis- 
sipate the external and obvious lesions of the disease, but to produce a 
condition of the blood in which the Wassermann reaction is permanently 
negative. It is quite generally agreed that the older methods of treatment, 
consisting of the administration of mercury and the iodids over fixed and 
arbitrary periods of time, or until all manifest symptoms have disappeared, 
are insufficient, and that the criteria by which the effects of treatment can 
best be judged are: (1) Continued absence of symptoms, and (2) per- 
manent negative Wassermann reactions. 
It is to be remembered, therefore, that while a single negative reaction 
is a satisfactory indication of the progress of treatment, it does not signify 
that a permanent cure has been effected. The Wassermann reaction can- 
not be regarded as sufficiently delicate to indicate that a single negative 
reaction means that a patient is totally free from all spirochetes, for in some 
instances the reaction and the clinical symptoms may recur after the treat- 
ment has been suspended, but the reaction is the first symptom to reappear 
and the earliest indication of an impending lesion. For all practical pur- 
poses the occurrence of a negative reaction after treatment indicates either 
complete destruction of all the spirochetes, or at least that the parasites are 
being held in abeyance and rendered potentially harmless. 
It is, accordingly, reasonable to regard the Wassermann reaction as the 
most delicate indicator of generalized spirochetal infection or the assumption 
of spirochetal activity. A positive reaction indicates that serious effects 
and gross local lesions are likely to occur at any time, and that treatment 
should be continued. For all practical purposes a continued absence of 
symptoms and a permanently negative reaction are strong presumptive 
evidences that a cure has been effected. 
The serum should be tested every four months during the treatment, and 
at periods of at least six months to a year after treatment has been dis- 
continued for several years. Persistently positive reactions during treat- 
ment would indicate that more active measures or a change in therapy are 
needed. The occurrence of a positive reaction after treatment has been 
discontinued is an indication for its resumption. 
For a control on treatment the Wassermann reaction should be made 
as delicate as possible, for while more prolonged treatment may be some- 
what irksome to the patient, it is clearly indicated as a preventive of serious 
after-effects, especially of involvement of the central nervous system. It is 
in this branch of the work I have found that the use of sensitive cholesterin- 
ized extracts as antigens in making the Wassermann reaction, of great value 
as the most delicate indicators. 
One fact is to be clearly emphasized, namely, that the earlier energetic treat- 
ment is begun, the more likely it is that a permanent cure will be effected. Ener- 
getic treatment with mercurials or salvarsan, or, better, with a combination 514 
COMPLEMENT FIXATION IN SYPHILIS 
of both, begun early and continued long, will in the majority of cases restore 
the serum to its normal condition. In general, the greater the interval of 
time allowed to elapse between infection and institution of treatment, the 
more difficult it is to restore the serum to normal. Tertiary cases are cured 
only as the result of most persistent treatment, and not infrequently in con- 
genital syphilis, locomotor ataxia, and general paralysis all one can hope to 
accomplish is to check the progress of the disease. The most favorable cases 
are those in which early diagnosis is made possible by clinical manifestations, 
preferably confirmed by a demonstration of pallidum, and in which treatment 
is undertaken before the serum has begun to react positively, and in which the 
reaction remains negative throughout. 
Treatment will, however, at least influence the Wassermann reaction in 
practically all stages of syphilis. In a series of 435 cases of syphilis in all 
stages reported by Boas, a negative Wassermann reaction was secured in 
no less than 80 per cent., and all but one of the remaining cases showed a 
weaker reaction. The figures of different observers are not all so favor- 
able as these, a factor dependent to some extent, at least, upon differences 
in the technic of the reaction. In general, however, Boas’ observations have 
been confirmed by other competent workers. 
The effect of any treatment is greatly influenced by the individuality of 
the host, certain persons possessing tissues more amenable to the effects of 
the therapeutic agent than those of others. The therapeutic effect is also 
dependent upon the virulence of the parasite and the apparent selective 
affinity of certain strains of pallidum for particular organs, and upon the 
method of treatment selected. 
The influence of salvarsan and neosalvarsan as agents in the treatment 
of syphilis is considered elsewhere. My experience has shown that the 
earlier belief in the complete sterilization of the human patient by a single 
dose was generally unfounded, and that repeated smaller doses of the drug, 
used in conjunction with mercurials, are necessary. Potassium iodid alone 
may favorably influence the clinical symptoms and weaken the Wassermann 
reaction in a small percentage of cases, and the same result has been observed 
with such arsenical preparations as Fowler’s solution, atoxyl, arsacetin, and 
arsenophenylglycin. 
It is to be remembered that, during or immediately after active treat- 
ment with salvarsan or mercury, the Wassermann reaction may be nega- 
tive, even though the patient is not cured. As a general rule, a negative 
reaction under these conditions should not be considered of value unless all 
treatment has been omitted for at least two weeks; even then the test, if 
negative, should be repeated a month or so later. Craig has recently drawn 
attention to the fact that in frank untreated cases the degree of the reaction 
may vary within wide limits, and this is especially true if the patients are 
receiving active treatment. 
Provocatory Stimulation.—Paradoxic as it would at first appear, anti- 
syphilitic treatment may convert a negatively reacting serum into a positive 
one. In not a few cases of latent syphilis reacting negatively the adminis- 
tration of a specific spirillicidal agent, such as mercury or salvarsan, is fol- 
lowed by positive reactions, due probably to the liberation of endotoxins 
from destroyed spirochetes or to a stimulation of the spirochetes by a dose 
of drug that did not suffice to kill them. This condition is analogous to the 
Herxheimer reaction, or the aggravation of skin lesions sometimes observed 
to follow the administration of mercury or salvarsan. The fact possesses 
practical value, for in cases where lues is known to have been present or is 
strongly suspected, and the Wassermann reaction is indefinite or negative, PRACTICAL VALUE OF THE WASSERMANN REACTION 
515 
the administration of 0.3 to 0.4 gm. of arsphenamin or neo-arsphenamin, 
followed by a Wassermann reaction twenty-four, forty-eight, and seventy- 
two hours later, may now show a positive reaction and thus indicate a 
latent syphilis requiring further treatment. Pollitzer and Spiegel1 have not 
found this method of any value and believe that it may be misleading. 
Stokes and O’Leary2 believe that it possesses value in certain cases. 
As previously stated, the Wassermann reaction serves two important 
purposes: (1) As an invaluable aid in the diagnosis and (2) as a guide in the 
treatment of syphilis. 
The reaction may be of great value in determining the diagnosis of extra- 
genital sores and of atypical lesions in all stages of syphilis. A negative 
reaction, however, has less value than a positive one, and whenever pos- 
sible, a microscopic examination of the secretions with the dark-field il- 
luminator should be made in order to confirm the diagnosis. In early latent 
syphilis, after the initial lesion has healed, and before the secondary eruption 
appears, the Wassermann reaction is frequently the only means of making 
the diagnosis, especially if the chancre has been small, atypical, and prac- 
tically neglected. 
Indefinite symptoms and clinical unrecognizable cases constitute a 
considerable proportion of cases of syphilis, and, as is true in all other infec- 
tions, this class constitutes the greatest menace to public health. Many 
patients are sincere in denying knowledge of infection and early symptoms 
may be overlooked, the Wassermann reaction being the sole means of 
diagnosis and serving in this connection as an invaluable aid. 
Usually the symptoms of syphilis are so well marked in the secondary 
stage that the reaction is in most instances but confirmatory evidence. 
However, in doubtful cases a negative reaction excludes syphilis with 
almost absolute certainty, especially if the reaction is repeatedly negative. 
In the late latent and tertiary stages of syphilis the Wassermann reaction 
may be the only available basis on which to establish a diagnosis. When 
one remembers how varied are the clinical manifestations of chronic syph- 
ilis, how wide-spread is the disease, and how frequently the reaction estab- 
lishes the true diagnosis, the reaction must be regarded as being of great 
value and as an indispensable diagnostic aid. It must not be forgotten 
that patients showing an early involvement of the central nervous system, 
and even those showing no such symptoms, may react negatively with blood- 
serum and positively with spinal fluid; in all such cases the spinal fluid should 
be examined whenever possible. 
A positive reaction occurring in aborting women is an indication for 
treatment and may protect the fetus. Similarly a positive reaction in 
either parent of a seemingly healthy infant is an indication for treatment 
of the child, especially if the mother reacts positively. 
In this connection, however, one point is worthy of special emphasis, 
namely, that although a positive reaction indicates that the patient is 
luetic, it does not necessarily mean that a particular lesion is syphilitic. 
For example, a person may be luetic and yet have a cancerous ulceration 
of the larynx. The mere fact that the lesion does not improve under anti- 
syphilitic treatment does not detract from the value of the Wassermann 
reaction, but is a warning that more care is required in making the clinical 
Practical Value of the Wassermann Reaction 
1 Amer. Jour. Syph., 1919, 3, 252. 
2 Archiv. Dermat. and Syph., 1920, 2, 348. 516 
COMPLEMENT FIXATION IN SYPHILIS 
examination. I have seen a number of such cases in which a positive Was- 
sermann reaction was held a priori as evidence of the syphilitic nature of a 
lesion that later proved to be either malignant or tuberculous. A weak 
positive reaction, associated with an active ulcerating lesion, very frequently 
indicates that the lesion is not syphilitic, for active lesions usually yield 
strongly positive reactions. 
In this connection may also be mentioned the growing importance the 
Wassermann reaction has assumed in life-insurance examinations. Statistics 
show that from one-tenth to one-third of all persons infected with syphilis 
die as the results of the disease, and the death-rate among 5000 syphilitics 
accepted for insurance was one-third over expectation (Brockbank). 
An important question, especially from the standpoint of therapeutics, 
is: Does a positive reaction invariably indicate the presence of living spiro- 
chetes? May the reaction remain positive for an indefinite time after the 
patient has been cured, just as agglutinins and antitoxins may persist in the 
blood for some time after recovery from typhoid fever and diphtheria has 
taken place? The sum total of the experience of investigators from all 
parts of the world would indicate that a persistently positive reaction means 
the presence of living spirochetes somewhere in the body. The lesions may 
not be active; the patient, while clinically healthy, may be infective, and is 
always subject to possible recurrences of clinical syphilis. 
Although gummas are slightly infectious, it is now known that they 
contain living spirochetes, and the former view, which regarded them as 
sequels, rather than as actual active lesions of syphilis, is no longer tenable. 
Just how long the reaction may remain positive after the patient is 
actually cured and all spirochetes are dead is, of course, difficult to state, 
but experimental studies on the lower animals has shown that the reagin 
disappears somewhat quickly under these conditions. 
Although a persistently negative reaction is of good prognostic impor- 
tance, it is not so conclusive in the information it yields as is a positive 
reaction. In other words, an occasional active lues may react negatively, 
and not infrequently active syphilitic lesions are found at autopsy in persons 
whose blood reacted negatively during life. While it is true that great 
harm may result from a false positive diagnosis due to faulty technic, yet it 
must be admitted that the Wassermann reaction is not too delicate, and that 
we are just as prone to err on the side of securing too many negative reac- 
tions. Every effort should be made to render the test as delicate as is pos- 
sible with specificity. 
While the value and dependability of the Wassermann reaction are 
based upon skilful technic that will eventually limit the performance of the 
test to specially trained persons in central laboratories, every effort should 
be made to render accessible to all persons this valuable diagnostic test of 
a disease that has such great social and economic importance. At present 
many persons are unable to afford the expense of a number of tests, or even 
of one test, as required in the modern treatment of this disease. This de- 
ficiency should be corrected, and the test made available in all free dis- 
pensaries, especially those under the supervision of a Social Service De- 
partment. CHAPTER XXIV 
Owing to the complexity of the complement-fixation reaction for syph- 
ilis, the several sources of error, and the time required for conducting the 
test it is not surprising that early and numerous attempts have been made 
to simplify the serum diagnosis of this disease. The investigations of Mor- 
eschi, having shown that the phenomenon of complement fixation may be 
associated with precipitation, it was to be expected that similar flocculation 
tests should be applied in the serum diagnosis of syphilis, and especially 
since a considerable amount of evidence has accumulated in support of the 
hypothesis that complement fixation in syphilis is a phenomenon of floc- 
culation of lipoids in colloidal suspension with fixation or absorption of 
complement. Michaelis1 was among the first to note precipitation when 
heated syphilitic serum was added to diluted alcoholic extract of syphilitic 
liver; he erroneously regarded the phenomenon as a specific precipitin 
reaction. Since then a large number of similar tests have been described 
and compared wfith the complement-fixation test as diagnostic reactions for 
syphilis. 
Classification.—These tests may be classified as follows: 
1. Specific precipitin reactions, as the reaction of Fornet and Schere- 
schewsky,2 which was briefly described in Chapter XVII. This reaction 
was based upon the precipitation of pallida proteins in syphilitic sera by 
antisyphilitic sera secured from long-standing cases of syphilis, as individuals 
with paresis and tabes. The reaction is of interest from the standpoint 
of production of specific precipitins for pallida proteins in syphilis, but 
occurs infrequently and possesses no diagnostic value. 
2. Coagulating reactions by chemical agents, as the butyric acid test of 
Noguchi, the nitric acid test of Bruck, the “gel” test of McDonagh, and the 
formol test of Gate and Papacostas. 
3. Colloidal flocculation reactions, as the distilled water reaction of Klausner 
and the reactions of Hirschfeld and Klinger, Porges and Meier, Herman 
and Perutz, Meinicke, Vernes, Sachs and Georgi, Dreyer and Ward, Kahn, 
and others. 
PRECIPITATION REACTIONS IN SYPHILIS 
CHEMICAL COAGULATING REACTIONS 
These reactions are mainly based upon the detection of protein changes 
in the serum and spinal fluid in syphilis. As shown by Rowe,3 and con- 
firmed by Tokuda4 in my laboratory, refractometric studies have indicated 
that in syphilis there is an increase of the serum globulins. These changes, 
however, are not characteristic of syphilis, but have been found in other 
infectious diseases. In the spinal fluid an increase of protein, and especially 
of globulins, is known to occur in acute and chronic meningitis of bacterial 
origin as well as in some forms of neurosyphilis. The following tests applied 
to the diagnosis of syphilis have not proved specific for this disease because 
they are unable to measure quantitative changes; that of Noguchi, however, 
has proved a valuable means for the detection of increased protein of 
syphilitic or bacterial origin in the spinal fluid. 
1 Berl. klin. Wchn., 1907, xliv, 1477. 
2 Berl. klin. Wchn., 1908, 18, 874. 
3 Arch. Int. Med., 1917, 19, 354. 
4 Arch. Dermat. and Syph., 1921, 4, 512. 
517 518 
PRECIPITATION REACTIONS IN SYPHILIS 
Noguchi’s Butyric Acid Reaction.—Noguchi’s test1 depends upon the 
coagulation of proteins by butyric acid and is employed for the detection of 
an increase of these in cerebrospinal fluid. It was first employed for the 
detection of an increase of proteins in the spinal fluid in paresis, but has 
since proved to be generally useful for the examination of spinal fluids in 
other inflammatory conditions of the meninges. 
In my experience this test has proved of particular value in establishing 
the differential diagnosis between serous and tuberculous meningitis, being 
negative in the former and positive in the latter, whereas in both the fluid 
may be clear, the cytology may be indefinite, and tubercle bacilli may 
escape detection. Serous meningitis is not a true infection, but a reflex 
vasomotor disturbance of the cerebral vessels, causing an outpouring of 
serum that leads to various pressure symptoms closely resembling those of 
a true meningitis. This condition is particularly common during child- 
Fig. 144.—The Noguchi Butyric Acid Test for Globulins. 
The tube on the extreme left shows the formation of flocculi within a few minutes after adding 
NaOH; the middle tube shows a strongly positive reaction after standing several hours (supernatant 
fluid quite clear); the tube on the extreme right shows a very slight opalescence, but no flocculi (within 
the limits of normal). 
hood, and the general symptoms, the increased pressure of the cerebrospinal 
fluid, and its clear, watery character, are features that resemble those of 
tuberculous meningitis. It is just in such cases—and they are frequent— 
that I have found this protein reaction of considerable value. A positive 
reaction practically always means a true meningitis; a negative reaction 
usually means “serous meningitis,” with a much better prognosis if the 
underlying cause is corrected. 
Noguchi has found the test positive in about 90 per cent, of cases of 
general paralysis and in 60 per cent, of cases of locomotor ataxia or cere- 
bral or spinal syphilis. In the diagnosis of syphilis the Wassermann reac- 
tion with cerebrospinal fluid has greater value than the protein reaction. 
1 Jour. Exper. Med., 1909, 11, 92. CHEMICAL COAGULATING REACTIONS 
519 
However, the best results in diagnosis are usually secured by a Wassermann 
test, butyric acid test, and total and differential cell counts. In a case 
where the diagnosis rests between tuberculous meningitis and syphilis, a 
positive butyric acid test and a negative Wassermann reaction would decide 
in favor of the former. 
The test is extremely simple. Into a small, thin-walled test-tube place 
0.2 c.c. of cerebrospinal fluid (which must be clear and free from blood); 
add 1 c.c. of a 10 per cent, solution of butyric acid in normal salt solution; 
heat over a low flame and boil for a short period. Then add quickly 0.2 
c.c. of a normal solution of sodium hydroxid and boil once more for a few 
seconds. The presence of an increased content of protein is indicated by 
the appearance of a granular or flocculent precipitate, which gradually 
settles to the bottom of the tube, under a clear supernatant fluid (Fig. 144). 
The velocity and intensity of the reaction vary with the quantity of 
the protein contained in a given specimen. The granular precipitate 
appears within a few minutes in a specimen containing a considerable in- 
crease in protein, whereas one hour may be required to obtain a distinct 
reaction in specimens weaker in protein. In obtaining the reaction the 
time period should not be greater than two hours. A faint opalescence with- 
out the formation of a distinct precipitate is to be regarded as within the limits 
of the normal. 
Brack’s Nitric Acid Reaction.—Brack1 has described a simple test for 
syphilis based on the observation that the precipitate formed from 0.5 c.c. 
of serum after the addition of 0.3 c.c. of a 25 per cent, dilution of nitric 
acid (specific gravity of 1.149) is not dissolved when 16 c.c. of distilled 
water are added ten minutes later, the tube inverted three times and stood 
aside for half an hour. This reaction has been investigated by Smith and 
Solomon,2 Stillians,3 Terada,4 and others. Toyama and myself5 found that 
there was some difficulty in reading borderline reactions, that the test agreed 
with a sensitive Wassermann test in only 70 per cent, of cases, and that it 
yielded about 8 per cent, falsely positive reactions. Our conclusion was 
that this reaction was not as reliable as the Wassermann reaction. 
McDonagh “Gel” Reaction.—McDonagh6 has advocated what he 
has designated a “gel” test for the diagnosis of syphilis. Clear and 
blood-free sera are employed, and it is necessary to include a known posi- 
tive and negative serum each time the tests are conducted. Into each 
of three dry tubes place 2, 3, and 4 drops of serum, respectively; add 0.1 
c.c. of acetic anhydrid and 1 c.c. of glacial acetic acid; the tubes are now 
well shaken and 1 drop of a saturated watery solution of ammonium sul- 
phate added. Instead of acetic anyhdrid and solution of ammonium sul- 
phate the test may be conducted with 0.2 c.c. of a saturated solution of 
lanthanum sulphate, thorium sulphate, or thorium nitrate in glacial acetic 
acid. A preliminary reading may be made and a final reading after the 
tubes have stood over night. “In all the tubes at first crystals form, but 
in the tubes containing normal serum they often disappear in six to twenty- 
four hours, while they remain in the tubes with syphilitic serum.” Strickler 
has used this test in my laboratory, and my observations of his results leads 
me to the conclusion that it is frequently difficult to interpret the reactions, 
1 Munch, med. Wchn., 1917, 64, 25. 
2 Boston Med. and Surg. Jour., 1917, 177, 321. 
3 Jour. Amer. Med. Assoc., 1917, 69, 2014. 
4 Kitasato, Archiv. Exper. Med., 1919, 3, 123. 
3 Jour. Cutan. Dis., 1918, 36, 429. 
6 Brit. Jour, of Dermat., 1916, April-June, 114. 520 
PRECIPITATION REACTIONS IN SYPHILIS 
and that the test has not by any means the practical value of the Wasser- 
mann reaction. 
Formol Reactions.—Gate and Papacostas1 have recently described a 
test for syphilis consisting of the addition of 2 drops of commercial formalin 
to 1 c.c. of clear serum. The mixture is allowed to stand for twenty-four 
to thirty hours at room temperature; coagulation is supposed to occur with 
syphilitic sera, but not with the sera of non-syphilitic individuals. Ecker2 
found that the test was of no value because of its failure to react in syphilis 
and the occurrence of positive reactions with the sera of non-syphilitic in- 
dividuals. Burke,3 in a study employing both sera and spinal fluids, found 
the test unreliable and especially from the standpoint of yielding too many 
negative reactions. Spinal fluids from cases of syphilis were found reg- 
ularly to yield negative reactions. 
Suffern4 has described a similar test which in my laboratory yielded re- 
sults quite similar to those reported by Ecker. 
COLLOIDAL PRECIPITATION REACTIONS 
Since the discovery by Michaelis that in mixtures of syphilitic serum and 
alcoholic extracts of tissues colloidal precipitation may occur, a large number 
of tests for syphilis have been devised on this principle. It is now gen- 
erally accepted that suitable extracts for the Wassermann test depend 
upon the size of the colloidal molecules and their capacity or incapacity 
for flocculation. It is highly probable that in mixtures of serum, 
complement, and antigen that colloidal flocculation occurs which entangles 
or removes the hemolytic activity of the complement and thereby reduces 
the degree of hemolysis following upon the addition of hemolysin and 
corpuscles. 
This colloidal flocculation by syphilitic serum (and by normal serum 
as well under certain quantitative conditions) in mixtures with alcoholic 
tissue extracts or synthetic extracts of bile salts or other lipoidal substances, 
can be detected either by (a) direct inspection macroscopically or micro- 
scopically or (b) indirectly measured by the degree of inhibition of hemolysis. 
In the latter instance a hemolytic serum must be added to the mixtures of 
syphilitic serum and antigen in order that the degree of hemolytic activity 
shall be reduced in proportion to the degree of colloidal flocculation; this 
principle is the basis of Vernes’ indirect method. 
Klausner’s Water Reaction.—Klausner5 has described a test consisting 
of the addition of 0.6 c.c. distilled water to 0.2 c.c. of fresh, clear, and un- 
heated serum. In seven to fifteen hours a flocculent precipitate forms 
which is more marked in syphilitic than in non-syphilitic sera. While the 
reaction was originally regarded as due to an increase of serum globulins, 
Klausner6 now believes that it is caused by lipoids in syphilitic sera and that 
heating an extraction with ether removes the reacting substances. The re- 
action, however, has not proved of practical value in the serum diagnosis 
of syphilis. 
Hirschfeld and Klinger’s Coagulo Reaction.—This test7 is based upon 
the hypothesis that coagulation of the blood is due to the formation of 
1 Compt. rend. Soc. de biol., 1920, 83, 1432. 
2 Jour. Infect. Dis., 1921, 29, 359. 
3 Arch. Dermat. and Syph., 1922, 5, 469. 
4 Lancet, 1921 2, 1107. 
5 Wien. klin. Wchn., 1908, 21, 214, 363. 
6 Biochem. Ztschr., 1912, xlvii, 36. 
7 Deut med. Wchn., 1914, xl, 1607. COLLOIDAL PRECIPITATION REACTIONS 
521 
fibrin and fibrin is produced from fibrinogen of the plasma by the action of 
thrombin (fibrin-ferment). The thrombin itself consists of two substances, 
namely, (1) the serozyme or thrombogen, which is a protein constituent of 
plasma and (2) the cytozyme or thrombokinase, which belongs to the group 
of lipoids, particularly the lecithins, and in the blood is believed to be de- 
rived chiefly from disintegrated blood platelets. Ionized calcium must be 
present before these substances unite to produce thrombin, but the latter 
can precipitate fibrinogen into fibrin and cause coagulation in the absence 
of calcium ions, for example, in oxalate plasma. 
While the cytozyme is derived from disintegrated platelets during the 
coagulation of blood, it may be secured in vitro by extraction from almost 
any cells or tissues by means of alcohol; in Hirschfeld and Klinger’s test it 
is supplied by the ordinary alcoholic extracts of normal tissue used as anti- 
gens in the Wassermann reaction. The cytozyme or lipoid material is 
essential in the formation of thrombin and hence in the phenomenon of 
coagulation; anything tending to interfere with its activity will delay or 
inhibit coagulation. Hirschfeld and Klinger’s reaction is based upon the 
observation that syphilitic serum when mixed with cytozyme interferes 
with its activity to a greater extent than normal serum and thereby delays 
or prevents thrombin production and coagulation. No adequate explana- 
tion has been made of the substance in syphilitic serum responsible for this 
action upon the cytozyme or of the mechanism of its interference; at present 
syphilitic serum is regarded as possessing the property of destroying cytozyme 
or rendering it inactive by enmeshing the lipoid particles in the globulins of 
the serum. Because the test is based upon the inhibition of coagulation by 
interference with thrombin production, the reaction has been named by 
Hirschfeld and Klinger the “coagulo reaction.” 
In conducting the test the first phase consists in mixing 0.1 or 0.2 c.c. 
of treated serum with 0.1 c.c. of varying dilutions of alcoholic extract of 
tissue (the cytozyme) and allowing the mixtures to stand for one-half to 
one hour to permit the inactivation or enmeshing of the lipoid particles of 
the cytozyme by the serum, and particularly if the serum is derived from 
a syphilitic; calcium chlorid and serocym (fresh plasma) are then added as 
the second phase and the tube stood aside for fifteen minutes to permit the 
production of thrombin providing cytozyme is available, the amount of 
thrombin produced bearing a direct ratio to the amount of cytozyme present. 
The third phase consists in testing for the presence and amount of thrombin 
by adding a solution of fibrinogen and timing the reaction to determine 
when fibrin formation or coagulation has occurred. The controls gen- 
erally coagulate within a few minutes; normal serum may delay coagulation 
a few minutes longer, while syphilitic serum delays coagulation for a longer 
period or indefinitely; the reaction, therefore, is a quantitative one. Since 
solutions of fibrinogen are unstable, a weak solution of oxalate plasma is 
employed in the last step of the test to measure the amount of thrombin 
present. The employment of oxalate plasma by Bordet and Delange1 has 
not only the advantage of keeping for a long time, but it also prevents any 
further formation of thrombin from the moment at which the plasma is 
added, because the sodium oxalate precipitates the calcium in the form of 
insoluble calcium oxalate, and the consequent lack of calcium ions renders 
impossible the further formation of thrombin. Details of the technic will 
be found in the paper by the writer and Toyama2 upon this reaction. We 
have found the reaction highly delicate and constant in syphilis, although, 
1 Ann. d. l’lnst. Pasteur, 1912, 26, 657, 737; ibid., 1913, 27, 341. 
2 Amer. Jour. Syph., 1918, 2, 505. 522 
PRECIPITATION REACTIONS IN SYPHILIS 
in our experience, slightly less sensitive than the Wassermann reaction. 
Frankel and Thiele,1 Cole and Chiu,2 Nemura,3 and others have reported 
favorable results with this reaction. 
Porges-Meier Reaction.—Porges and Meier4 observed that luetic serums 
are capable of producing flocculent precipitates from solutions of lecithin 
and similar salts. Two-tenths of a cubic centimeter of a 1 per cent, solution 
of Merck’s sodium glycocholate in distilled water is placed in narrow test- 
tubes, and an equal amount of the patient’s serum, which must be abso- 
lutely clear and inactivated by heating at 56° C. for thirty minutes, is added. 
This mixture and the known normal and luetic controls are kept at room 
temperature for from eighteen to twenty-four hours. A positive reaction is 
marked by the appearance of distinct coarse flocculi, mere turbidity or faint 
precipitation being regarded as negative. 
In this connection it may be mentioned that other investigators have 
used other substances in similar tests, as salts of bile acids, cholesterol, 
vaselin, courin, palmatin, stearin, etc., but without establishing tests of 
practical value. 
Jacobsthal5 has shown that by adding syphilitic serum to alcoholic ex- 
tracts of tissue employed in the Wassermann test, that precipitates are 
formed which may be demonstrated by means of the dark-field illuminator. 
Bruck and Hidaka6 obtained similar results by macroscopic tests. 
Herman-Perutz Reaction.—More recently Herman and Perutz7 have 
devised a similar test requiring the following two solutions: Solution 1 
(stock solution, diluted 1 : 20 with distilled water before use) consists of: 
Sodium glycocholate, 2 gm.; cholesterol, 0.4 gm.; 95 per cent, alcohol, 100 
c.c. Solution 2 (freshly prepared before use) is a 2 per cent, solution of 
sodium glycocholate in distilled water. The test is performed by adding 
to 0.4 c.c. of clear inactive serum (heated at 56° C. for half an hour) in a 
small test-tube 0.2 c.c. of solution 1 and 0.2 c.c. of solution 2. The tubes 
are tightly plugged with cotton and set aside at room temperature for 
twenty-four hours, after which the presence or absence of precipitation is 
noted. It is well in this test, as in all immunologic reactions, to prepare 
controls with known normal and luetic serums and with distilled water. 
Meinicke’s Reactions.—Meinicke8 has described three reactions based 
upon the hypothesis that the colloids of alcoholic extract of tissues dis- 
turb the isotonicity of saline solution permitting the union of serum globulins 
and lipoids. This reaction is greater in syphilitic than in non-syphilitic sera. 
Meinicke has described a water method, a salt solution method, and a third 
modification, the latter being mostly employed at the present time; the 
technic is as follows9: 
The antigen is prepared from horse heart by grinding fat-free muscle 
and drying at 50° to 55° C. Ether is added to the powder in the proportion 
of 9 parts to 1 and shaken for one hour. The ether is then filtered off and 
the material dried at 37° C. Alcohol (96 per cent.) is then added to the 
powder in the proportion of 9 to 1, shaken from time to time for a day, and 
filtered. The filtrate is permitted to stand for several days, when it is ready 
for titration with alcohol and distilled water for determining the optimum 
combination of extract and alcohol to give the proper opalescence for 
tests: 
1 Munch, med. Wchn., 1914, lxi, 2095. 
2 Archiv. Int. Med., 1915, 16, 880. 
3 Amer. Jour. Med. Sci., 1917, 154, 533. 
4 Wien. klin. Wchn., 1908, 31, 831. 
5 Munch, med. Wchn., 1910, 13, 41. 
6Ztschr. f. Immunitatsf., 1910-11, 8, 476. 
7 Med. Klin., 1911, 2, 60. 
8 Munch, med. Wchn., 1918, xlv, xlix, li. 
9 Munch, med. Wchn., 1919, No. 33, 932. COLLOIDAL PRECIPITATION REACTIONS 
523 
Tube. 
Antigen, C.c. 
96 Per Cent. 
Alcohol, C.c. 
Distilled 
Water, C.c. 
After One Hour Add 3.5 C.c. Water 
to Each Tube. 
1 
0.4 
0.1 
0.25 
Precipitation. 
Very slight precipitation. 
Opalescent. 
Faintly opalescent. 
2 
0.3 
0.2 
0.25 
3 
0.2 
0.3 
0.25 
4 
0.1 
0.4 
0.25 
In the above titration Tube No. 3 showed the correct combination with 
the particular extract employed. 
After this titration seven times the quantity of 2 per cent, sodium 
chlorid solution is added to the mixture of extract, alcohol and water, as, 
for example, 0.2 c.c. extract + 0.3 c.c. alcohol + 2.5 c.c. water, and after 
one hour +21 c.c. of 2 per cent, saline solution. 
In conducting the test 0.8 c.c. of the diluted antigen and 0.2 c.c. of 
serum heated at 55° C. for twenty minutes, are mixed and left at 37.5° C. 
for twenty-four hours, when the readings are made. A check reading may be 
made after the tubes have stood an additional twenty-four hours at room 
temperature. In my laboratory Strumia found that this additional period 
of incubation improved the results by removing some doubtful reactions. 
Levinson1 states that the third modification reaction of Meinicke agrees 
with the Wassermann reaction in about 88.8 per cent, of sera; similar results 
have been reported by Stern.2 Not infrequently, however, the reactions 
are very weak and the prolonged incubations and numerous bacterial con- 
taminations render the readings undependable. 
The Vemes Reactions.—The studies of Vernes3 upon flocculation of 
colloids by sera have been particularly thorough and fruitful, two tests 
having been elaborated for the serum diagnosis of syphilis. This investi- 
gator first worked with colloidal suspensions of inorganic substances and 
particularly of ferric hydrate, finding that in syphilis the serum acquires 
an enhanced power of flocculation and precipitation. This property of the 
serum in syphilis was found to fluctuate, but repeated tests at intervals 
generally showed a curve of flocculation higher than that shown by the 
serum of a non-syphilitic individual. Later on Vernes adopted a colloidal 
suspension of organic substances for his test in the form of a specially pre- 
pared extract of dried horse heart muscle designated as “perethynol.” This 
extract is now employed for conducting his direct method and a diaphano- 
metric scale is provided for the more accurate reading of the degree of floc- 
culation. 
The Vernes indirect method is a later development based upon a meas- 
ure of the amount of flocculation according to the degree of inhibition of 
hemolysis of sheep corpuscles by swine-serum. According to Vernes, fresh 
swine-serum contains a substance inhibiting flocculation by syphilitic serum 
and a second hemolytic complex for sheep corpuscles (presumably a natural 
hemolysin and complement); these two agents are supposed to be bound 
together so that when one is exhausted the second or hemolytic substance 
will also be exhausted. Therefore, in mixtures of perethynol, syphilitic serum 
and swine-serum colloidal flocculation by the former is inhibited by the anti- 
flocculating substance in the latter, but the exhaustion of this substance also 
1 Amer. Jour. Syph., 1921, 5, 414. 
2 Ztschr. f. Immunitatsf., 1921, 32, 167. 
3 Compt. rend. Acad. d. Sci., 1917, 165, 769; ibid., 1918, 166, 575; ibid., 1919, 167, 500; 
Atlas de Syphilimetrie, Boll, Paris, 1920. 524 
PRECIPITATION REACTIONS IN SYPHILIS 
removes the hemolytic activity of the swine-serum. The end-result, therefore, 
is an inhibition of hemolysis as occurs in the complement-fixation reaction, 
and Vernes believes that the degree of inhibition of hemolysis affords a more 
accurate measure of the flocculating power of the serum in syphilis than is 
possible in the direct method. By an extensive series of investigations 
Vernes has worked out the intricate quantitative relationships and devised 
a flocculation test by which the degree of flocculation by normal serum is 
measured; by the same technic the degree of flocculation by syphilitic serum 
Fig. 145.—A Sohxlet Extraction Apparatus with Vacuum. 
is also measured and plotted into curves (syphilimetry). He noted that nor- 
mal sera gave a horizontal line, but in syphilis the curve of flocculation 
oscillates up and down during the course of weeks or months. 
Antigen.—This is the same for both methods and is prepared by suc- 
cessive distillations under negative pressure with the perchlorid of ethylene 
and alcohol; it is named “perethynol” and is prepared as follows1: 
1 Bull. d. Sc. Pharmacol., 1918, 25, 321. COLLOIDAL PRECIPITATION REACTIONS 
525 
1. Secure fresh horse heart and grind the muscle to a pulp. Dehydrate 
by adding several volumes of 95 per cent, alcohol and allow to stand for one 
hour; express the alcohol and repeat. Express the alcohol, spread the 
material on glass plates, and dry at 37° C. for twenty-four hours. Grind the 
material into a fine powder. 
2. To 30 gm. of powdered muscle in a 500 c.c. distilling flask add 60 gm. 
of washed and dried sand and 250 c.c. of ethylene perchlorid with a boiling- 
point of 115° to 121° C. Connect with a Sohxlet extracting and condensing 
apparatus (Fig. 145) in such way that the distillation is conducted under a 
partial vacuum by means of an air or water pump. The distilling flask is 
heated in a water-bath at a temperature of 60° to 65° C. and with a pressure 
of 4 cm. of mercury, so that the temperature in the distilling flask does not 
exceed 35° C. This requires from six to seven hours. 
3. The distillate is discarded, the residue again dried at 37° C., and again 
extracted for five hours with 200 c.c. of absolute ethyl alcohol in the same 
apparatus at a pressure of 5t o 6 cm. of mercury. The temperature of the 
water-bath should be 60° to 65° C. and the contents of the flask about 30° C. 
4. The residue is discarded. The distillate is allowed to stand for 
twenty-four hours and filtered. About 25 c.c. are dried in a weighed dish 
at 60° C. and weighed. On the basis of this calculation the alcoholic extract 
is adjusted so that it contains 0.15 gm. dried extract per 1000 c.c. If it con- 
tains more than this amount, add the necessary amount of absolute ethyl 
alcohol; if less, the extract must be concentrated to the necessary degree by 
evaporation in a vacuum at 30° C. 
In the direct method the antigen (perethynol) is diluted by adding 1 part 
of antigen (perethynol) drop by drop to 6.5 parts of water. The serum is 
heated thirty minutes at 55° C. and 0.8 c.c. added to 0.4 c.c. of the antigen. 
This mixture is allowed to stand for twenty-four hours at room temperature 
(19° to 22° C.) and centrifuged. The superfluid is removed and the sedi- 
ment shaken up in 2.4 c.c. of doubly distilled water slightly charged with 
carbonic acid and the amount of flocculation recorded by comparison with 
a diaphanometric scale prepared by a mixture of water and tincture of 
benzoin. Spontaneous precipitation is prevented by the use of glycerin and 
tincture of quillaja as disseminating agents. The readings are made over a 
black inclined background with diffuse light or electric arc. 
The scale is prepared as follows: Mix 250 c.c. of tincture of benzoin, 
125 c.c. tincture of quillaja, and 250 c.c. of 80 per cent, alcohol. 
Add 10 c.c. of this mixture drop by drop to 50 c.c. of glycerin-water 
(glycerin 30 c.c. plus water 70 c.c.), while constantly stirring. This suspen- 
sion is further diluted one-fourth with the 30 per cent, glycerin-water and 
distributed in the following amounts in a series of 10 test-tubes of uniform 
caliber (13 mm. outside) each containing 2 c.c. of 50 per cent, glycerin- 
water: 0.01 (No. 1), 0.015, 0.0225, 0.0337, 0.0405, 0.0606, 0.0909, 6.1363, 
0.2194, and 0.3291 c.c. (No. 10). Each tube contains 1.5 times the dose 
in the preceding tube. 
The indirect method, which is more generally employed, is conducted as 
follows1: 
The extract (perethynol) is diluted 1 : 40 with 0.9 per cent, saline solu- 
tion. The required amount of saline is placed in a beaker and while being 
mechanically stirred by means of a small glass propeller attached to an 
electric motor (Fig. 146) at the rate of about 300 revolutions per minute, 
the extract is added drop by drop in such manner as to avoid touching the 
side of the beaker. The suspension should be freshly prepared just previous 
1 Compt. rend. Acad. d. Sci., 1918, 167, 385, 500; ibid., 1919, 168, 247. 526 
PRECIPITATION REACTIONS IN SYPHILIS 
to the titration and main tests and kept constantly stirred until distributed 
in the tubes. The suspension should correspond to tube 5 of the diaphano- 
metric scale described above. 
Fig. 146.—An Electrically Driven Mixer for Preparing Perethynol Suspension for Vernes’ 
Test. 
The sheep corpuscles are washed three times with hypertonic saline 
solution prepared as follows: 
Sodium chlorid 9.5 gm. 
Sodium bicarbonate 0.15 “ 
Potassium chlorid 0.42 “ 
Calcium chlorid 0.125 “ 
Distilled water 1000 c.c. 
After the last washing the packed cells are made up into a 50 per cent, 
suspension by adding an equal volume of the hypertonic saline solution. 
The suspension is then titrated by placing the following amounts in 
six test-tubes: 0.025, 0.05, 0.075, 0.1, 0.125, and 0.15 c.c.; distilled water 
is added to make the total volume in each tube exactly 2.6 c.c. as follows: 
2.575, 2.55, 2.525, 2.5, 2.475, and 2.45 c.c. The proper dose of cells for the 
tests is the amount corresponding to the color of this mixture: COLLOIDAL PRECIPITATION REACTIONS 
527 
Acid fuchsin (Grubler) 0.1 per cent, in distilled water 1.0 c.c. 
Picric acid 1 per cent, in distilled water 1.0 “ 
Glacial acetic acid 0.45 “ 
Formaldehyd solution (40 per cent.) ... 0.25 “ 
Distilled water 10.0 
The color given by this solution corresponds to Tint 8 and is called 
Solution 8; other solutions are prepared from it by diluting with this solu- 
tion (glacial acetic acid, 4.5 c.c.; formaldehyd solution (40 per cent.), 2.5 
c.c., and distilled water 100 c.c.) as follows in test-tubes: 
No. 7: 2 c.c. of No. 8 + 2 c.c. dilutent (Tint 7). 
No. 6: 1 c.c. of No. 8+2 c.c. diluent (Tint 6). 
No. 5: 2 c.c. of No. 8+7 c.c. diluent (Tint 5). 
No. 4: 0.4 c.c. of No. 8 + 2.3 c.c. diluent (Tint 4). 
No. 3: 0.8 c.c. of No. 8 + 7.3 c.c. diluent (Tint 3). 
No. 2: 1.6 c.c. of No. 8 + 22.7 c.c. diluent (Tint 2). 
No. 1: 0.32 c.c. of No. 8 + 6.97 c.c. diluent (Tint 1). 
No. 0: 0.1 c.c. of No. 8 + 6.4 c.c. diluent (Tint 0). 
About 3 c.c. of each of these solutions should be placed in scrupulously 
cleaned and dry test-tubes of non-sol-glass having a uniform caliber (13 mm. 
outside), stoppered with cork, properly labelled 8 to 0, and kept in a dark 
place. 
Sufficient suspension for the tests to be conducted is made up with the 
hypertonic saline in such manner that the dose is contained in 0.6 c.c.; 
example: 
Dose of cells, 0.075. 
Number of doses required, 200; prepared by diluting 15 c.c. of the 50 
per cent, suspension with 105 c.c. of the hypertonic saline solution; dose, 
0.6 c.c. 
The swine-serum may be obtained by collecting blood in a vessel in an 
abattoir and allowing the serum to separate. It must be fresh in order to 
preserve the complement and natural antisheep hemolysin and must be 
titrated each time for (a) hemolytic activity as well as for (b) the inhibiting 
influence of perethynol and (c) the proper adjustment of the albumin con- 
tent. Not all swine-sera are satisfactory; some are lacking in sufficient com- 
plement, hemolysin, or both. A mixture of several should be used and blood 
may be obtained from an abattoir. The titrations are conducted as follows: 
(a) Arrange two series of 11 test-tubes of regulation size. 
(b) Place 5 c.c. swine-serum in a small beaker and dilute 1 : 3.2 by adding 
11 c.c. of saline solution; mix well. Remove and discard exactly 1.5 c.c., 
leaving a balance of 10 c.c. 
(c) Place 0.8 c.c. of this dilution in the first tube of each series. Place 
0.8 c.c. of saline solution in the beaker and mix well (avoid bubbling). Then 
place 0.8 c.c. in the second tube of each series. Place 0.8 c.c. of saline solu- 
tion in the beaker and mix well. Then place 0.8 c.c. in the third tube of 
each series and continue in this manner until each of the 11 tubes of the 
two series have received 0.8 c.c. of diluted swine-serum. 
(d) In each tube of the first series (A) place 1.2 c.c. of 0.9 per cent, saline 
solution and 0.6 c.c. (the dose) of sheep cell suspension. Mix -well and in- 
cubate in a thermostat at 38° C. for twenty-five minutes. Centrifuge the 
tubes showing hemolysis above Tint 5. 
(e) In each tube of the second series (B) place 1.2 c.c. of the perethynol 
suspension. Mix well and incubate at 38° C. for forty-five minutes. Add 
0.6 c.c. of sheep cell suspension, mix, and reincubate for twenty-five minutes;, 
centrifuge the tubes showing hemolysis above Tint 5. 528 
PRECIPITATION REACTIONS IN SYPHILIS 
(/) It is necessary to accurately determine the points of complete he- 
molysis in both series, and for this reason all tubes showing hemolysis above 
Tint 5 are centrifuged. Pour off the superfluids carefully and completely 
and add 2.4 c.c. of hypertonic saline to each, mix well, and centrifuge. Pour 
off the superfluids and add 2.4 c.c. of distilled water to each tube; mix well. 
Hemolysis wall now occur and thereby show the presence of corpuscles not 
completely hemolyzed in the titrations. In this manner the smallest 
amounts of pig-serum required for complete hemolysis in both series may 
be accurately determined. 
(g) According to the method of dilution employed the tubes of each 
series contain the following amounts of undiluted swine-serum: 
Tube 1 = 0.250 c.c. 
“ 2 = 0.228 “ 
“ 3 = 0.206 “ 
“ 4 = 0.185 “ 
“ 5 = 0.163 “ 
“ 6 = 0.161 “ 
“ 7 = 0.120 “ 
“ 8 = 0.098 “ 
“ 9 = 0.054 “ 
“ 10 = 0.033 “ 
“ 11 = 0.003 “ 
The dose of serum required for the test is obtained by dividing the 
smallest completely hemolytic dose in B by the smallest completely hemo- 
lytic dose in A, and is usually between 0.15 and 0.25 c.c. 
If the dose is less than 0.18 c.c. add sufficient pig-serum heated at 55° C. 
for thirty minutes to make the dose 0.2 c.c. This is for the purpose of main- 
taining the albumin content at an optimum point. If below this point 
flocculation is considerably increased, thereby utilizing to a greater extent 
the antiflocculent property of pig-serum and consequently displacing the 
syphilitic index toward the positive phase (Tint 0). It is unnecessary to 
make any correction for the slight difference between 0.18 and 0.2 c.c. 
0h) All of the manipulations in this titration and in the main tests 
should be made within a constant time limit and at a uniform temperature. 
The optimum temperature is 20° C. If the temperature of the laboratory 
is above or below 20° C., corrections in the dose of complement should be 
made as follows: 
16° C. subtract 0.02 c.c. from the dose. 
17° C. substract 0.015 c.c. from the dose. 
18° C. subtract 0.01 c.c. from the dose. 
19° C. subtract 0.005 c.c. from the dose. 
21° C. add 0.005 c.c. to the dose. 
22° C. add 0.01 c.c. to the dose. 
23° C. add 0.015 c.c. to the dose. 
24° C. add 0.02 c.c. to the dose. 
The swine-serum is now diluted with 0.9 per cent, saline solution so that 
the dose is 0.8 c.c. Example: 
Hemolytic unit of Series B = 0.287 c.c. (Tube 3). 
Hemolytic unit of Series A = 0.137 c.c. (Tube 8). 
„ 0.2064 . oin 
D0Se " “ °'210cc' 
Temperature of laboratory 22° C. 
Corrected dose: 0.210 + 0.01 = 0.220 c.c. 
200 doses of 0.8 c.c. required = 43.8 c.c. serum + 116.2 c.c. saline. COLLOIDAL PRECIPITATION REACTIONS 
529 
Main Test.—(a) The sera should be fresh, free of corpuscles, and heated 
to 55° C. for twenty minutes. 
(ft) For each serum to be tested arrange 2 regulation test-tubes and 
place 0.2 c.c. serum in each. 
(c) Spinal fluids are used unheated; dose in each tube, 1.6 c.c. 
(d) To the first tube of each set add 0.8 c.c. of perethynol diluted 1 : 40 
with 0.9 per cent, saline solution; to the second tube add 0.8 c.c. of saline 
solution (controls). 
(e) To each tube carrying serum add 0.8 c.c. diluted swine-serum carry- 
ing the proper dose. With spinal fluids use proper dose of undiluted swine- 
serum in order not to increase the volume. 
(/) Place the tubes in a thermostat at 38° C. for seventy-five minutes 
if the temperature of the laboratory is 12° to 15° C., or for sixty minutes if 
the temperature is 18° to 22° C. 
(g) Add 0.6 c.c. of sheep cells in appropriate dilution; mix and reincubate 
for twenty to twenty-five minutes. When the serum controls are he- 
molyzed, the front tubes containing perethynol are centrifuged and the 
tints of supernatant fluids compared with the color scale and recorded. 
Readings.—A normal non-syphilitic serum should give Tint 8 of com- 
plete hemolysis. Any hemolysis less than this indicates a positive reaction. 
A complete positive would be Tint 0 (no hemolysis) and partial hemolysis 
is represented by the intermediate tints. 
Vernes claims that these methods possess a high degree of practical 
value in the diagnosis of syphilis and as a serologic guide to therapy. Owing 
principally to difficulties in technic it has not been extensively employed 
by others. Cornwall1 found that in cases of syphilis under treatment the 
spinal fluid gave positive reactions with the Vernes test in about 12 per 
cent, more cases than with the Wassermann test. With sera, however, the 
Wassermann reaction was positive in about 19 per cent, more cases than 
the Vernes reaction. 
Vernes regards the reaction a better guide to the treatment of syphilis 
than the Wassermann reaction. While it is claimed to be a different mechan- 
ism, yet the indirect method appears to be a complement-fixation reaction 
in which the complement and hemolysin is supplied by active swine-serum. 
Very probably the Wassermann and Vernes reactions are identical colloidal 
phenomena, and in my experience the latter reaction has not yielded better 
results either in the diagnosis or treatment of syphilis than a quantitative 
complement-fixation test. 
Sachs-Georgi Reaction.—At the present time considerable interest is 
being given a flocculation reaction described by Sachs and Georgi2 employ- 
ing cholesterolized alcoholic extract of heart muscle. 
The preparation of the extract appears to be an important element in 
the test and already numerous modifications have appeared. Beef heart 
muscle is passed through a meat grinder and 25 gm. ground with sand and 
extracted with 125 c.c. of 95 per cent, alcohol by vigorous shaking in a 
machine for six hours. It is then placed in an incubator over night, filtered 
through paper and stored in a refrigerator for several days followed by 
refiltration. To 100 c.c. of the filtrate add 200 c.c. of 95 per cent, alcohol 
and 18 c.c. of a 1 per cent, solution of pure cholesterol in alcohol (0.6 per 
cent, cholesterol). 
Sachs and Georgi state that most of their extracts have been satisfactory 
1 Arch. Dermat. and Syph., 1922, 5, 433. 
2 Med. Klinik., 1918, No. 33, 805; Munch, med. Wchn., 1920, 67, 66 530 
PRECIPITATION REACTIONS IN SYPHILIS 
with the addition of 0.045 to 0.06 per cent, cholesterol. Parker and Haigh1- 
have used 0.06 to 0.07 per cent, solutions and advise a preliminary titration 
to determine the maximum amount to employ. This may be accomplished 
by placing 3 c.c. of the alcoholic antigen (100 c.c. extract + 200 c.c. alcohol) 
in each of 5 test-tubes and adding the following amounts of 1 per cent, 
alcoholic solution of cholesterol: 0.135, 0.15, 0.18, 0.21, and 0.24 c.c., the 
percentages being respectively 0.045, 0.05, 0.06, 0.07, and 0.08. Each of 
these five mixtures are then employed in tests with normal and syphilitic sera 
as described below, and if possible with duplicate tests set up with an extract 
of known properties. 
The serum should be fresh, free of corpuscles, heated to 55° C. for half 
an hour, and allowed to cool for three hours, as Munster has shown that sera 
used immediately after heating may yield doubtful reactions. Spinal fluid 
is used unheated. 
The test is conducted as follows: The extract is diluted 1 : 5 and the method 
of dilution is an important matter. The required amount of extract is 
placed in a small Erlenmeyer flask and an equal amount of saline solution 
rapidly added. The mixture is gently shaken and allowed to stand ten min- 
utes, when the balance of the 4 volumes of saline is rapidly added. The 
mixture is again shaken gently and is ready for use. Example: 10 c.c. 
extract + 10 c.c. saline solution; mix, stand ten minutes, add 30 c.c. saline 
solution, and mix. Craig and Williams2 have found that much better results 
were observed when the saline solution was added to the extract drop by 
drop and the mixture allowed to stand for two hours before use. 
The test is conducted as follows: 
(a) For each serum and spinal fluid arrange two small test-tubes (out- 
side diameter 13 mm.). 
(,b) Place 0.1 c.c. serum and 0.9 c.c. saline solution in each tube. With 
spinal fluids place 1.5 c.c. undiluted in each tube. 
(c) To the first or front tubes of each serum test add 0.5 c.c. of the 
diluted extract; in spinal fluid tests add 0.75 c.c. of extract to each front 
tube. 
(<d) To the second or rear tubes add 0.5 c.c. of a 1 : 5 dilution of 95 per 
cent, alcohol in saline solution. These are the serum controls, and if pre- 
cipitates form the serum is unsuitable. 
(,e) In an extra tube place 0.5 c.c. of the diluted extract and 1 c.c. saline 
solution (antigen control). 
(/) All tubes are thoroughly shaken and placed in an incubator at 37.5° 
C. for two hours and then allowed to stand at room temperature for 
eighteen to twenty hours, when the reaction may be read. Neykirch3 ad- 
vises twenty hours at 37.5° C. followed by twenty hours in a refrigerator in 
order to bring out the weakly positive reactions and permit non-specific 
precipitates to disappear. 
(g) Meyer4 has shown that the reactions may be read by centrifuging 
the tubes;'Hull and Faught5 state that if the tests are centrifuged immedi- 
ately after adding the extract, they can be read as well as after standing 
twenty-four hours. Centrifuging is often of aid in weakly positive reactions 
followed by shaking when compact flocculi rise from the bottom of the 
tube. 
1 Arch. Dermat. and Syph., 1921, 4, 67. 
2 Tour. Amer. Med. Assoc., 1922, 79, 1597. 
3 Arb. a. d. Inst. f. Exper. Ther. u. G. S. H., 1920, 10, 49. 
4 Berl. klin. Wchn., 1919, No. 14, 331. 
6 Jour. Immunology, 1920, 5, 521. COLLOIDAL PRECIPITATION REACTIONS 
531 
(h) Readings may be made with the naked eye or with the aid of a 
reading glass against a slanted black background and recorded as follows: 
+ + + + (Strongly positive): heavy, flocculent precipitate and clear 
supernatant fluid. 
+ + + (Moderately positive): diffuse, finely flocculent precipitate; 
supernatant fluid contains flocculi. 
+ + (Weakly positive): slight amount of precipitate. 
— (Negative): no flocculation. A slight grayish sediment may be 
found, which, on shaking, is dispersed in the supernatant fluid. 
A large literature has already accumulated upon the results observed 
with this test as compared with the Wassermann reaction. Sachs and 
Georgi,1 in a review of the literature covering 12,124 tests, found an agree- 
ment in 92.44 per cent. In 3 per cent, the Wassermann reaction was posi- 
tive or doubtful and the Sachs-Georgi reaction negative; in approximately 
7 per cent, the Wassermann reaction was negative and the Sachs-Georgi 
reaction positive. Messerschmidt2 reported 1100 comparative Wassermann 
and Sachs-Georgi reactions with an agreement in 85 per cent., Kirschner 
and Segall3 on over 1000 similar tests, with an agreement of 80 per cent., 
and Baumgartel1 on 7000 tests, with 90 per cent, agreement. Favorable 
reports have also been made by Levinson and Petersen,5 Gaehtjens,6 D’An- 
noy,7 Harryman,8 Logan,9 Parker and Haigh,10 and others. My own experi- 
ence is closely similar to that of Kilduff11 and Craig and Williams,12 who 
found the reactions often difficult to read, and prone to error on the negative 
side. I have not found the test with either serum or spinal fluid as sensi- 
tive and reliable as the complement-fixation test either for the diagnosis 
of syphilis or as a serologic guide to treatment. 
Sigma Reaction of Dreyer and Ward.—A similar reaction has recently 
been described by Dreyer and Ward13 except that it employs nine different 
amounts of serum and aims to yield a quantitative reaction. 
The antigen is a mixture of an acetone-free alcohol soluble heart extract 
and cholesterol, originally described by Bordet and Ruelens.14 The antigen 
is carefully diluted with saline solution to give a standard opacity and a 
fixed amount used with varying amounts of heated serum using a Dreyer 
pipet and Dreyer tubes, to give dilutions 1 : 1.25, 1 : 2.5, 1 : 5.2, 1 : 13.1, 
1 : 26.4, 1 :46, 1 : 96, 1 : 232, and 1 :462. 
The tubes are incubated in a water-bath at 37° C. for seven hours and 
read with the aid of a hand lens. Rook16 has reported the reaction with sera 
as good, or even slightly better than the results obtained by the Wassermann 
reaction; with spinal fluids the results were less satisfactory. 
The Kahn Reaction.—Kahn16 has endeavored to improve the Sachs- 
Georgi test by a new extract containing more lipoids and by shortening 
the period of incubation; both test-tube and microscopic methods have 
been described.17 
The extract is prepared of beef or guinea-pig heart after the method of 
Neymann and Gager18: a weighed amount of finely ground and dried heart 
1 Arb. a. d. Inst. f. Exper. Therapie, 1920, 10, 5. 
2 Deut. med. Wchn., 1920, No. 6, ISO. 9 Lancet, 1921, 1, 14. 
* Wien. klin. Wchn., 1920, No. 18, 377. 10 Arch. Dermat. and Syph., 1921, 4, 67. 
4 Munch, med. Wchn., 1920, No. 15, 421. 11 Arch. Dermat. and Syph., 1921, 3, 415. 
6 Arch. Dermat. and Syph., 1921, 3, 286. 12 Jour. Amer. Med. Assoc., 1922, 79, 1597. 
6 Arch. Dermat. and Syph., 1921, 129, 2. 13 Lancet, 1921, 1, 956. 
7 Jour. Med. Research, 1921, xlii, 339. 14 Compt. rend. Soc. de biol., 1919. 
8 Arch. Dermat. and Syph., 1921, 4, 299. 15 Lancet, 1922, 1, 118. 
16 Arch. Dermat. and Syph., 1922, 5, 570; Proc. Soc. Exper. Biol, and Med., 1922, 19, 182. 
17 Arch. Dermat. and Syph., 1922, 6, 332. 18 Jour. Immunology, 1917, 2, 573. 532 
PRECIPITATION REACTIONS IN SYPHILIS 
muscle is extracted with several liberal amounts of ether in a refrigerator 
until supernatant ether is free from coloring matter (usually three to five 
days with daily change of ether). The ether is filtered off after each ex- 
traction and discarded. The heart muscle is then spread on filter-paper 
and dried until free from ether odor. Five cubic centimeters of alcohol 
(absolute or 95 per cent.) are added to each grain of powder and extracted 
for nine days in a refrigerator and one day at room temperature with 
frequent shakings. The alcohol is filtered off and a portion used as such 
for plain antigen; the remainder is cholesterolized as follows: to each 100 c.c. 
add 0.4 gm. pure cholesterol. Dissolve by warming in a water-bath with gentle 
rotation. Filter through several layers of filter-paper to remove impurities 
as well as undissolved cholesterol and allow to stand at room temperature 
for twenty-four hours before use. 
For use the plain extract is diluted 1 : 2 with saline solution and the 
cholesterolized extract 1 : 3, the latter being prepared as follows: Measure 
5 c.c. of antigen into a dry 25 c.c. cylinder. Place 10 c.c. of physiologic 
saline solution into another 25 c.c. cylinder. Pour the saline into the 
antigen in about the same manner as one might add physiologic saline 
solution to red corpuscles. Invert the mixture back and forth several times. 
This mixture is opalescent and somewhat milky and may be kept in an 
incubator for several weeks without loss of activity. As a general rule the 
extract is employed about one-half hour after its dilution. 
The serum must be fresh, clear, and free of corpuscles. The presence or 
absence of natural antisheep hemolysin has no influence. Serum is heated 
at 56° C. for thirty minutes. The test is conducted as follows: 
(a) It is recommended that each serum be tested with two or three 
dilutions of the same extract or with two or three different extracts. 
(b) In small test-tubes place 0.3 c.c. serum and add 0.05 c.c. diluted 
extract. Spinal fluids are used in dose of 1 c.c. 
(c) Mix the contents of each tube vigorously for two or three minutes 
and incubate in a thermostat at 38° C. for sixteen to eighteen hours, or 
for four hours in a water-bath followed by overnight in a refrigerator 
(d) “The results are read after about eighteen hours’ incubation. The 
tubes should not be shaken again before the first reading. The specific 
precipitates are lipoidal in character and are suspended in the medium. 
The strongly positive will be recognized without difficulty. The + + + + 
reactions will show either one or several clumps. The + + + reactions 
will show comparatively large flocculi or granules. The ++ reactions will 
show clumps or granules of a lesser size, but large enough to be unmistakable. 
The + and =t reactions are best seen by slanting the tubes and observing 
the upper point of contact between the fluid and tube wall. The slanted 
tube should be several inches above the level of the eye and held in front 
of a window, focusing on some dark object, such as a window shade or frame. 
One will then see, by looking up, a thin layer of fluid with a precipitate 
floating in it.” 
The cholesterolized extracts are generally more sensitive than the plain 
alcoholic extracts, and with the eighteen-hour period of incubation great 
care must be exercised in the interpretation of weakly positive reactions. 
With about 80 per cent, of strongly positive Wassermann reactions a 
precipitate may appear within five minutes after the addition of antigen 
to the serum followed by vigorous shaking for several minutes. Kahn has 
also described a micro-method employing 0.03 c.c. of serum and 0.005 c.c. 
of diluted antigen. Herrold1 has recently described a modification embrac- 
1 Jour. Amer. Med. Assoc., 1922, 79, 957 MECHANISM OF PRECIPITATION IN SYPHILIS 
533 
ing a ring reaction at the line of contact between serum and antigen. In 
my laboratory this method proved inferior to the test described above. 
The true precipitate is whitish and floats in the fluid and is best seen by 
merely slanting the tubes and observing the upper layer of fluid. Shaking 
may bring out dust particles and spontaneous serum precipitates and sug- 
gests doubtful reactions. By adding 0.8 c.c. saline solution to such tubes, 
true precipitates are visible, while pseudo or serum precipitates disappear. 
Of 1119 comparative tests carried out with the precipitation and Was- 
sermann reactions, Kahn found 98 per cent, to agree in results, and he hopes 
that this test will form an important supplement to the Wassermann test. 
Keim and Wile1 have employed the test in the study of 350 sera with 
satisfactory results, believing that its simplicity, rapidity of reading and 
reduction of sources of error constitute advantages over the Wassermann 
reaction. Ide and Smith2 have employed the test with 2165 sera and believe 
that it is of value as a check on the Wassermann reaction. They have also 
reported that the test yields satisfactory results with sera proving anti- 
complementary in the complement-fixation reaction. In my laboratory 
Strumia has studied this and other precipitation reactions in syphilis, and 
found that with sera yielding strongly positive or negative complement- 
fixation reactions the reactions with both tests were closely parallel, but with 
sera containing small amounts of syphilis “reagin,” as in latent tertiary cases 
of syphilis and cases under treatment, that the complement-fixation test 
was generally superior in sensitiveness and practical value. 
MECHANISM OF PRECIPITATION IN SYPHILIS AND RELATION TO 
COMPLEMENT FIXATION 
Numerous investigations upon the mechanism of the various flocculation 
reactions in syphilis have shown that chemical changes occur mainly involv- 
ing the globulins of the serum and that flocculation is apparently a physico- 
chemical phenomenon involving the interaction of substances in colloidal 
suspension. The important relation of electrolytes and particularly of 
sodium chlorid to the phenomenon as shown, especially by Meinicke, have 
indicated the colloidal nature of these reactions, as is true of serum aggluti- 
nation and precipitation in other diseases as well. 
Reference has already been made to the increase of serum globulins oc- 
curring in syphilis, probably a result of irritation of body cells by the prod- 
ucts of Treponema pallidum. The water reaction of Klausner, the nitric and 
lactic acid tests of Bruck with serum, and the butyric acid test of Noguchi 
with spinal fluid are based upon this primary change, but at the present time 
none of these or similar tests evolved by others have proved sufficiently 
quantitative or differential for diagnostic purposes. Furthermore, the mere 
increase of serum globulins in syphilis does not necessarily explain or bear 
a relation to the colloidal flocculation and complement-fixation reactions, 
but simply represents an additional physical change apart from the chemical 
changes concerned in these reactions. 
Normal human serum is capable of flocculating various inorganic and 
organic substances in colloidal suspensions as shown by Lange, of colloidal 
iron as shown by Vernes, and of mastic, benzoin, and various lipoidal sub- 
stances in colloidal suspension as shown by the numerous flocculation tests 
devised for the serum diagnosis of syphilis beginning with the work of 
Michaelis over fifteen years ago. But in syphilis this power of serum and 
spinal fluid for precipitating inorganic and organic colloids is increased, and 
1 Jour. Amer. Med. Assoc., 1922, 79, 870. 
2 Arch. Dermat. and Syph., 1922, 6, 770. 534 
PRECIPITATION REACTIONS IN SYPHILIS 
usually in a measurable degree; or the serum proteins and especially the 
globulins acquire in syphilis an increased sensitiveness to flocculation by 
various organic and inorganic colloids. 
Chemical studies of the flocculi or precipitates produced by various floccu- 
lation tests have generally shown the presence of both serum proteins (espe- 
cially globulins) and substrate. Two theories or hypotheses may be advanced, 
therefore, in explanation of these colloidal precipitating reactions as follows: 
(a) That in syphilis the colloidal serum proteins and especially the glob- 
ulins acquire an increased sensitiveness to flocculation or precipitation by 
other colloids and especially lipoidal substances contained in the “antigen” 
(alcoholic tissue extracts) or, 
(b) That in syphilis a new and foreign antibody-like substance is pro- 
duced, associated or linked with the globulins of the serum, possessing a 
greater power for flocculating organic (tissue lipoids) and inorganic (iron, 
gold, etc.) colloids in suspension, than possessed by the serum proteins of 
healthy individuals. 
Available data indicate that the latter or second hypothesis is more 
tenable than the former, and that an antibody-like substance or cellular 
reactionary agent is produced and linked or associated with the globulins 
of the serum as is true of antibodies in general. Because this “reagin” is 
capable of flocculating lipoids in colloidal suspension it is commonly believed 
that it may be a lipoid-protein compound, but its exact chemical nature has 
not been determined. 
There can be no doubt, however, that the flocculating substance pro- 
duced in syphilis and present in the blood and spinal fluid has a special 
effect upon lipoidal substances in colloidal suspension, and for this reason 
the substrate for most flocculating tests is an alcoholic extract of some tissue. 
The methods of Herman and Perutz, Porges, and Meier, Meinicke, Sachs 
and Georgi, Vernes, Kahn, etc., employ substrates of this kind or solutions of 
isolated lipoids, and the occurrence, sensitiveness, and practical value of the 
reactions in syphilis depend to a remarkable degree upon the method of 
preparing and diluting the extracts, as is true of the “antigens” or extracts 
employed in the Bordet-Wassermann reaction. In the coagulation test of 
Hirschfeld and Klinger, a tissue lipoidal extract (the cytozyme) is regarded 
as essential for the production of thrombin by interaction with a serozyme 
(an albuminoid) and a calcium ion, and the reaction is based upon the obser- 
vations that syphilitic serum inhibits coagulation by its effect upon the 
lipoidal extract (cytozyme), probably one of inactivation by flocculation. 
Why this syphilis “reagin” or antibody-like substance has this flocculating 
effect upon lipoids in colloidal suspension is unknown. Some investigators 
believe that it is an antibody for lipoids produced in syphilis; that in this 
disease as well as in frambesia, a foreign lipoidal antigen is engendered 
through the action of the products of spirochetes upon body cells which is 
capable of calling forth a peculiar lipodophilic antibody or reagin. The 
weight of evidence is against the assumption that protein free lipoids can 
engender the production of antibodies, but the stimulating substance may be 
a lipo-protein compound produced by cells irritated by these pathogenic 
spirochetes and capable of engendering the production of an antibody or 
reactionary product of unknown chemical nature causing the flocculation or 
precipitation of organic and inorganic colloidt in the test-tube under favor- 
able conditions. 
In my opinion the fundamental mechanism of complement fixation in 
syphilis is identical with these macroscopic colloidal flocculation reactions. 
There is no important or essential reason for believing otherwise. The PRACTICAL VALUE OF PRECIPITATION REACTIONS IN SYPHILIS 
535 
agglutination and precipitin reactions with formed elements and proteins in 
solution are colloidal reactions even though the chemical nature of the anti- 
body in the serum is unknown. It is entirely reasonable to assume that in 
the Bordet-Wassermann reaction the same “reagin” or antibody-like sub- 
stance is operative as in the various flocculation reactions; that in the 
former flocculation occurs which is invisible to the naked eye, but sometimes 
visible microscopically by the dark-field illumination method. In a general 
manner the tissue extracts yielding best results in the complement-fixation 
reaction also prove most sensitive in the macroscopic flocculation reaction, 
e. g., the cholesterolized alcoholic extracts. The degree of dilution, how- 
ever, varies for the two methods, but the manner of diluting the extracts 
with saline solution is essentially the same for both. 
Just what happens to the complement or alexin is unknown. It is 
probably fixed or absorbed by the precipitate and its hemolytic activity 
inactivated. However, it is possible that the complement serum has two 
qualities closely linked, one that is hemolytic and the other antifloc- 
culating; that in mixtures of syphilitic serum and lipoidal extracts the anti- 
flocculating substance is consumed in counteracting the flocculation of 
proteins and lipoids in colloidal suspension, and that this removes at the 
same time the hemolytic complex. Vernes has advanced this theory in 
explanation of the role of fresh swine-serum in his indirect method, but 
whatever may be the true explanation, I believe that the indirect method 
of Vernes is nothing more than a complement-fixation reaction utilizing the 
complement and natural antisheep hemolysin of swine-serum. At any 
rate, I have conducted his test with equally good results employing the usual 
guinea-pig complement and rabbit antisheep hemolysin. 
SPECIFICITY AND PRACTICAL VALUE OF PRECIPITATION REACTIONS 
IN SYPHILIS 
Originally studied in relation to the mechanism of complement fixation 
in syphilis, flocculation or precipitation tests have since been advocated as 
practical means for the serum diagnosis of this disease. Methods evolved 
from time to time and compared with the Wassermann reaction were soon 
found to be inferior as practical diagnostic methods, but during the past 
few years the methods of Meinicke, Vernes, Sachs and Georgi, and Kahn 
have commanded much attention with the accumulation of a large literature, 
especially upon the Sachs-Georgi reaction. 
Advantages of Precipitation Tests.—As compared with the technic of the 
complement-fixation test, the precipitation methods are simpler, and this 
constitutes their main advantage. They are likewise more economical, since 
sheep corpuscles, hemolysin, and complement are not employed. They do 
not require preliminary titrations of complement and hemolysin, but, on 
the other hand, the preparation and titration of the extract employed as 
“antigen” requires as much attention as in the complement-fixation test. 
Contrary to current opinion the results of flocculation reactions are greatly 
influenced by technical conditions and especially the manner of diluting and 
using the extract, temperature and duration of incubation of the mixtures 
of extract and serum, etc.; furthermore, considerable experience is required 
in the reading and interpretation of the reactions, probably more than re- 
quired for complement-fixation tests. Owing to the simplicity of the 
technic when extract is available, it would appear off hand that one of 
these methods may prove of value as a substitute for the complement-fixation 
test for the serum diagnosis of syphilis on ships and under similar conditions 
in which the latter test cannot be conducted, but in my opinion they cannot 536 
PRECIPITATION REACTIONS IN SYPHILIS 
be relied upon alone for the serum diagnosis of clinically obscure or doubtful 
cases of syphilis or as a serologic guide to treatment. 
Disadvantages of the Precipitation Tests.—From the technical stand- 
point, many reactions are difficult to read and interpret. With serums yield- 
ing + + + + complement-fixation reactions, flocculation is usually well 
marked, but with serums containing less “reagin” the degree of precipitation 
is frequently slight and readily overlooked or confused with non-specific 
reactions. My experience in this regard is similar to that of Craig and 
Williams1: “The difficulty of reading slight Sachs-Georgi reactions and 
differentiating them from precipitation due to other causes in the serum is 
so great, in many instances, that no reliable conclusion can be reached, thus 
leaving the test open to so much individual interpretation that it destroys 
the scientific value of the results.” 
Considerable experience is required for reading the reactions—more, in 
my estimation, than required by a complement-fixation reaction. 
From the biologic standpoint precipitation methods are less sensitive 
than acceptable complement-fixation tests and probably more subject to the 
error or falsely positive reactions, despite their simpler technic. 
Sensitiveness of the Precipitation Reactions.—Most investigators have 
based their opinion on the value of flocculation reactions in syphilis accord- 
ing to comparative results with the Wassermann reaction or one of its modi- 
fications, and needless to state the sensitiveness and syphilitic specificity 
of the latter have influenced the percentages of positive and negative 
reactions and the opinions based upon these. My analysis of the literature 
covering over 24,000 comparative tests employing the Meinicke (third 
modification), Sachs-Georgi, and Kahn reactions, have shown that the posi- 
tive and negative results agreed with complement-fixation reactions in from 
80 to 95 per cent, of serums with a general average of about 89 to 90 per cent. 
About 8 per cent, of the differences were due to positive complement-fixation 
and negative flocculation reactions and 3 per cent, to negative complement- 
fixation and positive flocculation reactions. 
The results observed by Strumia, in a comparative study of several floc- 
culation tests wdth my new complement-fixation test for syphilis based upon 
studies in the standardization of technic, have shown the following results: 
COMPARISON OF COMPLEMENT-FIXATION AND FLOCCULATION REAC- 
TIONS IN SYPHILIS 
Flocculation Tests. 
Sera 
Tested. 
Complement-fixa- 
tion Reactions. 
Flocculation Reactions. 
Agree- 
ment. 
Per cent, 
positive. 
Per cent, 
negative. 
Per cent, 
positive. 
Per cent, 
negative. 
Per cent, 
doubtful. 
Meinicke (third modification).. . 
705 
40.4 
59.6 
39.5 
58.7 
1.7 
89.2 
Sachs-Georgi 
645 
40.3 
59.7 
47.8 
50.3 
1.9 
85.3 
Kahn (cholesterolized antigen). . 
624 
41 
59 
49.8 
48.2 
2 
83.4 
Kahn (plain antigen) 
566 
40 
60 
30.4 
65.5 
4 
82.7 
Vernes (direct method) 
503 
37.6 
62.5 
44 1 
55.5 ' 
0.4 
89.3 
Vernes (indirect method) 
500 
38.4 
61.6 
41.6 
56.6 
1.8 
90 
With the serums of the majority of syphilitic individuals the flocculation 
reactions are positive and practically specific, but with about 8 to 15 per 
1 Jour. Amer. Med. Assoc., 1922, 79, 1597. PRACTICAL VALUE OF PRECIPITATION REACTIONS IN SYP'HILIS 
537 
cent, of serums from individuals with latent and tertiary syphilis and con- 
taining small amounts of “reagin” the tests yield falsely negative reactions. 
Our aim should be to improve the sensitiveness of complement-fixation 
and flocculation reactions, because even under the best of conditions the 
reactions are still too frequently negative in syphilis. As the matter stands 
at present a good, sensitive complement-fixation reaction is our best means 
for the serum diagnosis of syphilis, and especially syphilis in its clinically 
difficult and unrecognizable stages. As stated by Craig and Williams, no 
less than 33.3 per cent, of the positive Wassermann reactions were missed 
by the Sachs-Georgi test in known cases of syphilis, while 26.5 per cent, of 
the positive Sachs-Georgi reactions were not confirmed by the Wassermann 
test nor by clinical findings. 
Others have observed better results—with the Sachs-Georgi and Kahn 
methods the reactions have agreed with the complement-fixation reaction 
in as high as 95 to 98 per cent, of serums, but the significance of these figures 
is reduced unless the complement-fixation test is known to be sensitive and 
reliable and the numerous modifications of the Bordet-Wassermann reaction 
in use are known to vary in these regards. 
As is to be expected in any large series of comparative tests, the floc- 
culation methods sometimes have yielded positive reactions in primary syph- 
ilis before the complement-fixation test has become positive, and likewise 
have yielded positive reactions in treated and latent cases of syphilis when 
the complement-fixation tests were negative, but in a broad and general 
manner it may be stated that the Meinicke, Sachs-Georgi, Vernes, direct and 
Kahn flocculation reactions are less sensitive than acceptable complement- 
fixation reactions conducted with acceptable reagents by individuals pos- 
sessing the experience required and justified by the importance of the subject. 
Specificity of the Precipitation Reactions.—As previously stated, normal 
serums possess the property of flocculating organic and inorganic colloids 
under certain conditions. By reason of the apparent technical simplicity of 
the flocculation tests it may be assumed that they are less subject to the 
biologic error of falsely positive reactions. But they are subject to this 
error in just the same manner as the Wassermann reaction and, indeed, in 
my opinion, to even greater degree because it is necessary to render the 
amount of flocculation as great as possible which tends to elicit non-specific 
reactions. Extracts employed as “antigens” in the flocculation tests are 
just as important from this standpoint as in the complement-fixation reaction, 
and particular care must be exercised with cholesterolized extracts. 
Flocculation reactions also occur with the serums of individuals with 
frambesia (yaws) as would be expected. Whether or not reactions occur 
in the febrile stages of malaria, pneumonia and other acute infections, in 
diabetes, leprosy, and other diseases in which falsely positive Wassermann 
reactions are said to occur, cannot be stated. In my opinion we must revise 
our ideas concerning falsely positive Wassermann reactions because so many 
have been based upon technical errors, and I believe that the immunologic 
changes responsible for the complement-fixation and flocculation reactions 
occur only in spirochetic infections, notably syphilis and frambesia. 
Practical Value of the Precipitation Reactions.—Summing up the sub- 
ject it may be stated that the precipitation reactions are not as sensitive 
as the complement-fixation reactions conducted with a technically correct 
and acceptable technic; that they are more subject to misinterpretation and 
should not be relied upon alone for the serum diagnosis of syphilis. They may, 
however, be useful as controls on the complement-fixation test, and for this 
purpose the Meinicke (third modification), and more especially the Sachs- 
Georgi and Kahn reactions, are to be recommended. CHAPTER XXV 
COMPLEMENT FIXATION IN BACTERIAL INFECTIONS AND FOR 
THE DIFFERENTIATION OF PROTEINS 
Specific Complement Fixation in Bacterial Diseases.—As has been 
stated elsewhere, the first complement-fixation tests were performed by 
Bordet with bacterial antigens and antiserums (pest and typhoid). Fol- 
lowing the application of the principles of complement fixation in the serum 
diagnosis of syphilis, it was but natural that the possibilities of this method 
as a general means of diagnosis soon became appreciated, and in a short 
time numerous infections were studied. 
Probably in no disease has complement fixation proved so constant 
or so valuable a diagnostic procedure as in syphilis. In this condition the 
peculiar lipodophilic reagin is largely responsible for the marked fixation of 
complement, and from our present knowledge on the subject we learn that 
this phenomenon has practically no analogy in any other disease except 
frambesia. 
With few exceptions bacterial antigens are likely to yield weaker and more 
inconstant reactions. This is due to the fact either that our antigen lacks a 
more available and specific antigenic principle, or that the amount of com- 
plement-fixing bodies is small and variable. For these reasons it becomes 
apparent that the preparation of antigen and sensitiveness of technic are 
highly important factors. 
Complement Fixation in the Differentiation of Microparasites.—Several 
investigators have applied the technic of complement fixation in a study 
of the relationship and differentiation of closely related bacteria, spiro- 
chetes, trypanosomes, and other microparasites. These studies have usually 
been made by immunizing rabbits with a particular strain and using the 
immune serum in complement-fixation tests with antigens of the other 
microparasites under study. As a general rule complement fixation is most 
marked with homologous antigen and antiserum; relationship is studied 
according to whether or not complement fixation occurs with the other 
antigens. Besredka,1 Foix and Mallein,2 Swift and Thro3 have reported 
that immune amboceptors specific for different strains of streptococci can 
be demonstrated by means of the complement-fixation test. In my own 
work with five strains of streptococci,4 with the specific purpose of studying 
the relationship of the streptococcus commonly found associated with 
scarlet fever to the group of streptococci, I found that differentiation among 
these was possible when using high dilution of the immune sera, but in 
lower dilutions differentiation was not found by means of complement- 
fixation tests, the results indicating either that this scarlet fever strepto- 
coccus belonged to the common group of streptococci or that the comple- 
ment-fixation test was a group reaction. In a similar study of the diph- 
theria group of bacilli I found that the pseudodiphtheria bacillus could 
not be differentiated from other members of the group by complement- 
fixation reactions5 indicating its relationship to the true diphtheria bacillus; 
similar studies made by Williams, Raiziss, and myself8 with the typhoid 
1 Ann. de l’Inst. Pasteur, 1904, 28, 363. 
2 Presse m6dicale, 1907, 15, 777. 
3 Arch. Int. Med., 1911, 7, 24. 
4 Arch. Int. Med., 1912, 9, 220. 
6 Jour. Infect. Dis., 1912, 11, 44. 
6 Jour. Infect. Dis., 1913, 13, 321. 
538 TECHNIC OF BACTERIAL COMPLEMENT-FIXATION TESTS 
539 
colon group of bacilli, by Craig and Nickols1 with several strains of spiro- 
chetes, by Cooke2 with acid-fast bacilli, by Olmstead3 and Wollstein4 with 
meningococci, by Hooker5 with typhoid bacilli, and by Hitchens and Brown6 
with streptococci indicate that the bacteriolytic amboceptors are highly 
specific and yield highly specific reactions with proper technic, a,nd par- 
ticularly with good and sensitive antigens, and under these conditions the 
complement-fixation test may be more delicate in differentiation than ag- 
glutination or other immunity recations. As a practical procedure, how- 
ever, the complement-fixation technic is too intricate and time consuming. 
In this connection I desire to direct particular attention to the possibility of 
the sera of normal rabbits yielding positive complement-fixation reactions with 
various bacterial and lipoidal extracts as discussed on page 438; in conducting 
complement-fixation tests with rabbit- or dog-sera the serum of each animal 
should first be tested with the antigen and immunization conducted only in case 
the animal’s serum is found to react negatively; subsequent complement-fixing 
properties of the serum may then be ascribed to the presence of specific ambo- 
ceptors. 
Preparation of Bacterial Antigens.—Either the endotoxins or whole 
bacterial body may constitute the main portion of an antigen. Most re- 
cent efforts have aimed thoroughly to disorganize the bacterial cell in order 
to liberate the endotoxic substances that pass into solution and constitute 
the antigen. Experience has frequently shown, however, that the protein 
substances of the bacterial cell itself possess antigenic properties, and, ac- 
cordingly, I have generally found that antigens composed of bacterial cells 
and the products of their activity are usually more satisfactory than those 
prepared of the endotoxic substances alone. 
As a general rule, bacterial antigens should be polyvalent—i. e., made up 
of a number of different strains of the same micro-organism. Recent re- 
searches in bacteriology tend to show that different strains of the same micro- 
organism have particular and more or less individual pathogenic and some- 
times biologic characteristics, and it is reasonable to assume that the anti- 
body will likewise show individual properties and a special affinity for its 
particular antigen. When, therefore, one antigen is being used in com- 
plement-fixation work, the results are more likely to be satisfactory if a 
large number of different strains are included in the antigen, with the hope 
that at least one of them will show a particular affinity for the antibody in 
the patient’s serum. 
Bacterial antigens may be prepared in various ways. 
Suspension 'in Broth or Saline Solution.—Cultures are grown in a suitable 
fluid medium, such as plain bouillon, for forty-eight hours, or upon a solid 
medium, and washed off with a suitable quantity of normal saline solution. 
The culture or emulsion is shaken for an hour or so to break up the clumps, 
and then heated to 60° C. for an hour. It is preserved by the addition of 
0.25 per cent, of phenol or tricresol. 
This constitutes the simplest bacterial antigen. It is composed of 
both bacterial cells and the products of bacterial activity, and frequently 
yields uniform and satisfactory results. 
Schwartz and McNeil Method.—Cultures are grown on a suitable solid 
medium for from twenty-four to forty-eight hours. Growths are removed 
TECHNIC OF BACTERIAL COMPLEMENT-FIXATION TESTS 
1 Jour. Exper. Med., 1912, 16, 336. 
2 Jour. Amer. Med. Assoc., 1917, lxviii, 1504. 
3 Jour. Immunology, 1916, 1, 307. 
4 Jour. Exper. Med., 1914, 20, 201. 
6 Jour. Immunology, 1916, 1, 1. 
6 Personal communication. 540 
COMPLEMENT FIXATION IN BACTERIAL INFECTIONS 
by adding sufficient distilled water or normal saline solution to yield a 
milky suspension. The emulsion is heated to 56° C. in a water-bath for one 
hour, followed by one hour at 80° C., and shaken mechanically with glass 
beads for twenty-four hours, to facilitate disintegration. It is then filtered 
through paper pulp and a sterile Berkefeld filter with a neutral reaction, or 
thoroughly centrifugalized; the filtrate is then heated to 56° C. for one-half 
hour on each of three successive days to sterilize, and preserved with 0.25 
per cent, phenol and used as antigen after being diluted and made isotonic 
with 1 part of 9 per cent, salt solution to 9 parts of the antigen. 
This antigen is composed essentially of endotoxic substances, and is 
the one usually employed in the preparation of gonococcus antigen. 
Antigens of Dried Bacteria.—Cultures are grown on a solid medium, 
washed off with normal saline solution, and the emulsion centrifuged thor- 
oughly. The sediment is dried over sulphuric acid or calcium chlorid, and 
the dried material thoroughly ground with crystals of sodium chlorid. 
Sufficient distilled water is then added to render the solution isotonic, and 
so that it will contain about 0.5 gm. of dried material in each cubic centi- 
meter. This emulsion is then shaken for twenty-four hours, filtered or 
centrifuged, the filtrate preserved with 0.5 per cent, of phenol, and used as 
antigen. Antigen may be prepared also as follows: 
Cultures are grown on a solid medium and washed off with normal saline 
solution. Saline suspension is then precipitated with an equal quantity of 
absolute alcohol and centrifugalized. The sediment is dried in vacuo over 
sulphuric acid, weighed, and ground into a fine powder with sufficient crystals 
of sodium chlorid to make a 2 per cent, suspension of dried material in 
isotonic saline solution. This stock suspension is not filtered or centrifuged, 
but is further diluted with saline solution, and constitutes the antigen 
(method of Besredka, modified by Gay). The actual amounts of dry anti- 
genic substance contained in 1 c.c. of various dilutions are as follows: 
1 c.c. of 1 : 40 dilution =0.5 mg. 
1 c.c. of 1 : 80 dilution =0.25 mg. 
1 c.c. of 1 : 160 dilution = 0.125 mg. 
1 c.c. of 1 : 320 dilution = 0.062 mg. 
1 c.c. of 1 : 640 dilution = 0.031 mg. 
1 c.c. of 1 : 1280 dilution = 1.0155 mg., etc. 
Hitchens' and Hansen Method.—Young cultures of bacteria grown on 
solid media are suspended in distilled water (about 10 c.c. being used for 
each 20 square inches). To this suspension is added an equal volume of 
95 per cent, ethyl alcohol followed immediately by centrifugalization. The 
precipitate is again suspended in alcohol and centrifuged; the precipitate is 
now suspended in ether, the original volume of the suspension being reduced 
to one-half each time. After removal of the supernatant ether the last traces 
of ether adhering to the precipitate are removed by vacuum. The bacterial 
mass is now dried over phosphorus pentoxid in vacuo for three days and 
stored in amounts of about 0.05 gm. in “H” tubes with phosphorus pent- 
oxid in vacuo. Suspensions are prepared in saline soluton as required. 
Small's Method.—Twenty-four-hour growths of bacteria on solid medium 
are removed and dried in sterile Petri dishes at 53° to 56° C. The scaly mass 
is ground in a mortar while being moistened with chloroform followed by 
small additions of ether. The grinding is maintained until the last addition 
of ether has evaporated and a fine buff colored powder is obtained. This is 
now suspended in a mixture of equal parts of chloroform and ether and 
shaken for four to six hours. After settling the supernatant fluid is re- Fig. 147.—Anticomplementary Titration of a Gonococcus Antigen. 
The tube containing 0.6 c.c. shows slight inhibition of hemolysis, and this amount is the anticom- 
plementary unit. TECHNIC OF COMPLEMENT-FIXATION TESTS 
541 
moved and the residue shaken up three or four times with ether. After 
decanting or filtering away the last ether the powder is dried in an incubator 
at 56° C.; likewise the residue on the filter-paper which is now recovered. 
The powder is now placed in sterile saline solution containing 0.5 per cent, 
phenol (25 c.c. for each 0.5 gm. of residue at the start), shaken for twelve 
to eighteen hours and used as antigen. This method was described by Small 
for the preparation of antigens of typhoid, paratyphoid and dysentery 
bacilli, pneumococci, and meningococci. 
Special methods for the preparation of antigens of tubercle bacilli and 
other micro-organisms will be described in the special sections of this chapter 
devoted to these subjects. 
TECHNIC OF COMPLEMENT-FIXATION TESTS 
Author's First Method 
This test is conducted in exactly the same manner as the author’s 
modification of the Wassermann test employing multiple antigens described 
on pages 473 to 478. It is applicable for the gonococcus, tuberculosis, 
typhoid, or any other bacterial complement-fixation test. 
Hemolytic System.—Complement is collected as described, diluted 1 : 20 
and used in dose of 1 c.c. Sheep corpuscles are collected and washed as 
previously described, prepared in a 2.5 per cent, suspension, and used in 
dose of 1 c.c. Antisheep hemolysin is titrated as described on page 473 and 
used in dose of 2 units. 
Titration of Bacterial Antigen.—After an antigen has been prepared 
it is standardized by determining the anticomplementary unit—i. e., the 
amount of antigen that just begins to show inhibition of hemolysis due to 
non-specific complement fixation. This unit is easily determined by adding 
increasing amounts of antigen to a series of test-tubes with a constant dose 
of complement in each. As a general rule, it is well to add to each tube a 
constant dose of fresh normal inactivated serum, e. g., as 0.1 to 0.2 c.c., 
when the anticomplementary action of serum alone is allowed for. I would 
emphasize the necessity of doing this in experimental work with rabbit-, 
dog-, or any other animal serum. After incubating for one hour, 2 units 
of hemolytic amboceptor and 1 c.c. of corpuscle suspension are added to 
each tube, and the tubes are reincubated for an hour or two and the reading 
made (Fig. 147). In the main test one-quarter to one-half the anticomple- 
mentary unit may be used, as this amount is known to be free from any power 
of non-specific complement fixation. The former dose is, of course, safer than 
the latter. 
Anticomplementary Titration of a Bacterial Antigen (Fig. 147) 
Tube. 
Antigen, 
1 : 10, C.c. 
Comple- 
ment, 1 : 
20, C.c. 
« x) , \ 
3 si o 
i-M flJ , . 
b 
c “ o 
Antisheep 
Ambo- 
ceptor, 
Units. 
Sheep’s 
Corpuscles 
(2.5 Per 
Cent.). C.c. 
-d 
30 
CJ 
Results 
1.. . 
0.2 
1 
2 
1 
r—3 ro 
Complete hemolysis. 
Complete hemolysis. 
Complete hemolysis. 
Slight inhibition of 
hemolysis {unit). 
Marked inhibition of 
2 
0.4 
1 
tu 
2 
1 
cj cj - 
3....... . 
0.6 
1 
O C 
2 
1 
c h 
<v d 
4 
0.8 
1 * 
SjS 2 
2 
1 
r* O 
c3 -C 
5 
1.0 
1 
o 
0) . V 
2 
1 
'S «j 
6 
0 
1 
rj CJ 
.S w 
3 
2 
1 
a u 
cn O 
CD 
hemolysis. 
Hemolytic control. 
Complete hemolysis. 
o 
d 
H 542 
COMPLEMENT FIXATION IN BACTERIAL INFECTIONS 
The titration may be completed by determining the antigenic dose of 
the antigen by titrating with a suitable and constant dose of specific im- 
mune serum. This titration is conducted by placing in a series of test-tubes 
increasing doses of antigen with a constant dose of heated immune serum 
(usually 0.1 c.c.) and a constant dose of complement. After an hour the 
proper dose of hemolytic amboceptor and corpuscles is added. The read- 
ings may be made an hour or two later, or after the tubes have been allowed 
to settle in a refrigerator. That tube showing just complete inhibition of 
hemolysis contains the antigenic unit. For the main test it is well to use 
double this amount, providing this dose is not more, and preferably less, 
than half the anticomplementary unit. 
The antigenic titration is not always satisfactory, for when an artificial 
immune serum is used the concentration of antibodies may yield a much 
stronger reaction than one would expect in testing human serums. Further 
than this, the antibody content in antiserums varies considerably, so that 
the antigenic unit fluctuates according to the particular serum used in making 
the titration. In general, therefore, it is sufficient to determine the anti- 
complementary unit and to use half or quarter this amount in performing the 
main test. 
After antigens are prepared they may require further dilution with 
saline solution. This can be determined only by experience and as the 
result of a trial titration. 
As watery extracts are prone to deteriorate, it should be made a rule 
that the anticomplementary dose be determined each time before the main test is 
conducted. 
It has quite generally been proved that alcoholic extracts of bacteria 
do not yield satisfactory antigens, despite the advantage to be gained 
because of their stability. 
The Test.—The serums should be fresh and clear, and heated to 56° C. 
for one-half hour. For each serum use four test-tubes (12 by 1 cm.) ar- 
ranged in a row. Into each of the first three place the dose of antigen 
and increasing doses of serum—0.05 c.c., 0.1 c.c., 0.2 c.c.; the fourth tube 
is the serum control, and into this is placed the maximum dose of serum 
(0.2 c.c.), but no antigen; 1 c.c. of complement diluted 1 : 20 is added to each 
tube. The following controls are included: 
1. A positive control with an immune serum or with the serum of a 
patient who reacted positively on a former occasion. 
2. A negative control with the serum of a healthy person. 
Both of these controls may be set up with but the maximum dose of 
serum (0.2 c.c.). 
3. The serum control of each serum is conducted in the fourth tube of 
each series. At the completion of the test this tube should show complete 
hemolysis and thereby indicate that the serum was not anticomplementary. 
4. The antigen control at this time includes the dose of antigen and 
complement. 
5. The hemolytic system control at this time receives the dose of com- 
plement. 
6. The corpuscle control receives 1 c.c. of the corpuscle suspension. 
To each tube sufficient saline solution is added to bring the total volume 
up to about 2 c.c. The tubes are shaken and incubated for one hour at 
37° C. in the thermostat or in a water-bath (not less than one hour), when 
2 units of antisheep amboceptor and 1 c.c. of sheep corpuscle suspension 
are added to each tube except the corpuscle control. The tubes are gently 
shaken'again and reincubated for one hour or longer, depending upon the Fig. 148.—Gonococcus Complement-fixation Reaction. 
Shows a + reaction with 0.05 c.c. serum; a ++ with 0.1 c.c., and a + + + + with 0.2 c.c., a strongly 
positive reaction. AUTHOR'S STUDIES IN STANDARDIZATION OF TECHNIC 
543 
hemolysis of the controls, after which the results are recorded. This sec- 
ondary incubation may be omitted and the tubes placed in a refrigerator 
overnight and the results read the next morning. Under these conditions 
hemolysis occurs slowly, and, according to some workers in this field, the 
reaction becomes more delicate. 
The following table is an example of a gonococcus fixation test with the 
serum of a case of gonorrheal arthritis (Fig. 148). 
Gonococcus Complement-fixation Test 
Patient’s Serum, 
C.c. 
Antigen, 
1 : 10, C.c. 
Comple- 
ment, 
1 :20, C.c. 
"d 
o3 
C 
<D 
Antisheep 
Ambo- 
ceptor, 
Units. 
Results After One and 
(2 5 Per a Half Hours’ Incuba- 
Cent.), C.c. tion- 
0.05 
0.2 
1 
rd 
C/5 
2 
1 Slight inhibition of he- 
>o 
molysis. 
0.1 
0.2 
1 
2 
1 Marked inhibition of 
hemolysis. 
0.2 
0.2 
1 
p a 
2 
1 Complete inhibition of 
tn § 
hemolysis. 
0.2 
0 
1 
4J 2 
2 
1 Serum control: hemol- 
Positive serum, 
SJ 
ysis. 
0.2 
0.2 
1 
o 
2 
1 Complete inhibition of 
H o 
hemolysis. 
0.2 
0 
1 
2 
1 Serum control: hemol- 
Negative serum, 
<u 
ysis. 
0.2 
0.2 
1 
2 
1 Hemolysis. 
0.2 
0 
1 
cr J3 
C2 9 
2 
1 Hemolysis. 
0 
0.2 
1 
J.s 
2 
1 Antigen control: hemol- 
4-> 
d 
ysis. 
0 
0 
1 
*3 
2 
1 Hemolytic control: he- 
<D 
molysis. 
0 
0 
0 
0 
1 Corpuscle control: no 
in 
hemolysis. 
In reading the results the controls are first examined and should show 
complete hemolysis; the test is reported as negative if all tubes are he- 
molyzed, weakly positive if the largest dose only of serum (0.2 c.c.) shows 
inhibition of hemolysis, moderately positive if the 0.1 and 0.2 c.c. doses of 
serum show inhibition of hemolysis, and strongly positive if the 0.05, 0.1, 
and 0.2 c.c. doses of serum react positively. 
The test may be conducted with but one dose of serum, namely, 0.2 c.c. 
with the antigen and 0.2 or 0.3 c.c. in the serum control tube. When conduct- 
ing the Wassermann and gonococcus tests with the same serum an extra tube 
is simply added to the Wassermann series with three antigens as described on 
pages 473 to 470, making a five tube test for conducting both Wassermann 
and gonococcus tests at the same time. In this case the readings are made 
after the usual +, ++, + ++, and + + + + method (see page 472). 
Author’s Method Based Upon Studies in the Standardization of Technic 
The test is conducted in exactly the same manner as the syphilis re- 
action described on page 478. Human serum seldom contains as much 
complement-fixing antibody in gonorrhea, tuberculosis, typhoid fever, and 
other bacterial infections as may be present in syphilis, and for this reason 
the amounts of serum employed should not vary too greatly. The qualita- 
tive test is conducted with two tubes in exactly the same manner as des- 
cribed. 544 
COMPLEMENT FIXATION IN BACTERIAL INFECTIONS 
The test has yielded much more sensitive reactions than the method described 
above; one important reason for this is because the primary incubation is 
eighteen hours at 6° to 8° C. instead of one hour in a water-bath. 
The antigen must be titrated in exactly the same manner as described 
for the titration of syphilis antigens. As a general rule the anticomple- 
mentary titration only is conducted. If it is desired to conduct this titra- 
tion just before the main tests are set up, the primary incubation may be 
one hour in a water-bath at 37° C., but in this case one-quarter the anti- 
complementary unit of antigen should be employed. If the antigen is 
titrated with the period of cold primary incubation it may be used in the 
main tests in one-third of the anticomplementary unit. 
COMPLEMENT FIXATION IN GONOCOCCUS INFECTIONS 
This was one of the first infections to be studied by means of the com- 
plement-fixation technic, but the results secured were not generally satis- 
factory until it was shown that the antigen must be polyvalent. 
Historic.—In 1906 Muller and Oppenheim1 applied the complement-fixation test to the 
diagnosis of gonorrheal arthritis, using a culture of the gonococcus as antigen. To these 
observers, therefore, belongs the credit of being the first to record a complement-fixation test 
in a gonococcus infection. A little later in the same year Carl Bruch2 applied the reaction 
to 3 cases of gonorrhea, using the serum of immunized rabbits, and reported favorable 
results. In 1907 Meakins3 reported having secured positive reactions in 3 cases of 
gonorrheal arthritis, which was the first report in America published on this subject. Th. 
Vannod4 studied the specificity of the reaction with the serums of rabbits immunized with 
gonococcus protein and one of a meningococcus, and reported that the meningococcus immune 
serum did not show complement fixation with gonococcus antigen, and, vice versa, that 
gonococcus amboceptor was not bound by meningococcus antigen. Wollstein5 (1907), in a 
study of the biologic relationship of the gonococcus and meningococcus, reported findings 
differing from those of Vannod. The former observer found that bacteriolytic amboceptors 
in the serums of rabbits immunized with these cultures were closely related and yielded fixa- 
tion of complement with either antigen. Teaque and Torrey6 in 1907 issued a very impor- 
tant communication showing that the differences in results of previous investigators were 
probably due in part to the use of single strains of the organisms in the preparation of antigens 
and immune serums. They emphasized the fact that the gonococcus belongs to a heterogene- 
ous family, and that in attempting to formulate a diagnosis of gonorrheal infection by the 
complement-fixation method the extracts of several different strains should be used. Naz 
Vannod and later Watabiki7 found that the gonococcus and meningococcus antibodies were 
quite specific for their homologous antigens in complement-fixation reactions. 
Particular attention was drawn to the gonococcus complement-fixation test by the work 
of Schwartz and McNeal.8 These investigators emphasized the necessity of using polyvalent 
antigens, and their encouraging reports have stimulated renewed interest in this subject. 
They found that if the infection is confined to the anterior urethra, a positive reaction is 
not obtained; that a strong reaction is not to be expected before the fourth week of the in- 
fection, and then only in acute cases with complications. They regard a positive reaction as 
indicating the presence or recent activity in the body of a focus of living gonococci, although 
a negative reaction does not exclude gonococcus infection. The test, therefore, has a more 
positive than a negative value. With Flexner’s antimeningococcus serum positive reactions 
resulted with their gonococcus antigen; with serums from cases of cerebrospinal meningitis 
(meningococcic) the results were negative. 
In the succeeding years numerous investigators, including Swinburne, Gradwohl, O’Neil, 
Gardner and Clowes, Thomas and Ivy, Kolmer and Brown, Smith and Wilson, Dixon, Kil- 
duffe, Lailey, and Cruickshank have reported favorably upon the practical value of the 
gonococcus complement-fixation test, particularly as an aid in determining whether or not a 
patient is cured of the infection. 
1 Wien. klin. Wchn., 1906, 19, 894. 2 Deutsch. med. Wchn., 1906, 70, 36. 
3 Johns Hopkins Medical Bull., 1907, 18, 255. 
4 Ztschr. f. Bakter., 1907, 44, 10. 6 Jour. Exp. Med., 1907, 9, 588. 
6 Jour. Med. Research, 1907, 17, 223. 7 Jour. Infect. Dis., 1910, 7, 159. 
8 Amer. Jour. Med. Sci., May, 1911; ibid., September, 1912; ibid., December, 1912. COMPLEMENT FIXATION IN GONOCOCCUS INFECTIONS 
545 
Antigen.—This constitutes the most important ingredient of the test 
As Teague and Torrey1 and Schwartz and McNeal2 have emphasized, the 
antigen should be prepared of many different strains of gonococci. The 
difficulty of isolating this organism and the constant care required in sub- 
culturing and keeping a large number of strains alive render it practically 
impossible for many persons to prepare a gonococcus antigen. Therefore 
until simpler methods are devised, this antigen is best prepared in large 
central laboratories, where the cultures are handled and preserved bv 
specially trained persons. 
Torrey and Buckell,3 however, have recently described a single culture- 
medium for the gonococcus whereby strains are readily carried along so 
that antigen may be freshly prepared as required. As a result of very ex- 
tensive immunologic studies with gonococci they have been able to select 
two strains for this purpose, each covering a large number of substrains. 
The gonococci are well grown on a salt-free veal agar, neutral in reac- 
tion to phenolphthalein, and to which a few drops of sterile hydrocele fluid 
may be added. After culturing for from twenty-four to forty-eight hours 
the growths are washed off with distilled water, and the emulsion is heated 
in a water-bath for two hours at 56° C. It is then heated at 80° C. for one 
hour, filtered through paper pulp or centrifugalized, and passed through a 
Berkefeld filter which is reserved for this purpose alone and is neutral in 
reaction. A small amount of preservative, as, e. g., 0.1 c.c. of a 1 : 100 
dilution of phenol to each cubic centimeter of antigen, may be added. The 
antigen is then well preserved in small amounts in ampules that are sealed 
and heated to 56° C. for half an hour on three successive days. Just before 
being used the antigen is made isotonic by adding 1 part of a 10 per cent, 
salt solution to 9 parts of antigen. I preserve the antigen in ampules con- 
taining 1 c.c., and after removing the antigen from the ampule to a large 
test-tube, add 1 c.c. of 10 per cent, salt solution, and dilute the whole 1 : 10 
with the addition of 8 c.c. of normal salt solution, after which the anti- 
complementary titration is made. 
In this method of preparing antigen the endotoxins constitute the main 
antigenic principle. Brown and myself,4 after an experimental study of 
the various antigens, found that a simple suspension of gonococci in salt 
solution yielded slightly better results. The various strains are grown for 
from forty-eight to seventy-two hours, and are then washed off with sterile 
saline solution, observing particular care not to include portions of the 
culture-medium. The suspension is then shaken to break up clumps, and 
heated to 56° C. for one hour. A small amount of preservative is now 
added, and the antigen stored in 1 c.c. ampules. Before using it is diluted 
1 : 10 or 1 ; 20, and titrated for the anticomplementary dose. 
Alcoholic extracts of gonococci have very little practical value, as alcohol 
is not satisfactory for extracting the antigenic principles of bacteria. Warden,5 
however, has reported good results with a new lipoid antigen. 
Specificity of the Gonococcus Complement-fixation Test.—Viewed from 
a practical standpoint, the reaction is highly specific. While complement- 
fixation experiments with antigens of gonococci and meningococci and their 
respective immune serums have demonstrated a biologic relationship be- 
tween these micro-organisms, yet practically with human serums an antigen 
of pure cultures of gonococci will fix complement only with the gonococcus 
antibody (amboceptor). In this technic a specific antigen is employed, and 
1 Jour. Med. Research, 1907, 17, 223. 3 Jour. Immunology, 1922, 7, 305. 
2 Amer. Jour. Med. Sci., 1911, 141, 693. 4 Jour. Infect. Dis., 1914, 15, 6. 
5 Jour. Amer. Med. Assoc., 1915, lxv, 2080; Jour. Lab. and Clin. Med., 1916, i, 333. 546 
COMPLEMENT FIXATION IN BACTERIAL INFECTIONS 
it is, therefore, a true application of the Bordet-Gengou reaction of com- 
plement fixation by specific antigen and specific antibody (amboceptor). 
Obtained under proper technical conditions, a positive reaction is invariably 
reliable, and indicates the presence of a focus of living gonococci. 
Practical Value of the Gonococcus Complement-fixation Test.—From 
our present knowledge of this reaction it may be stated: 
1. That the difficulty of isolating and preserving a sufficient number of 
cultures of true gonococci in order to prepare a satisfactory polyvalent 
antigen constitutes a weighty drawback to the practical use of the test. 
2. Because the gonococcus antibody, unless complicated by wide-spread 
gonococcal metastases, is produced in small amount in the majority of cases, 
the degree of complement fixation is usually much less than that which occurs 
in the syphilis reaction, and accordingly the reactions are usually weaker 
and often indefinite. A negative reaction does not exclude gonorrhea. 
3. The reaction is seldom positive during the first four to six weeks of 
an acute anterior or posterior urethritis, in the absence of complications. 
In acute exacerbations of a chronic urethritis the reaction is positive in 
about 80 per cent, of cases. In ordinary chronic urethritis with mild infec- 
tion of the prostate gland the reaction is positive in from 30 to 40 per cent, 
of cases. In chronic urethritis complicated by marked involvement of the 
prostate gland and epididymitis the reactions are frequently positive, 
occurring in from 50 to over 80 per cent, of cases. 
The test possesses considerable value in determining the fitness of an 
applicant for a marriage license, and will, no doubt, be employed for this 
purpose quite extensively, as a positive reaction is now generally regarded 
as indicating the presence of a focus of active gonococcal infection. A 
negative reaction, however, does not exclude the possibility of latent infec- 
tion in the urethra and accessory organs. 
4. During the course of an acute or a subacute urethritis the occurrence 
of an acute complication, such as prostatitis, epididymitis, etc., is likely to 
result in a positive fixation test. 
5. In gonorrheal iritis a positive reaction occurs in over 80 per cent, of 
cases. 
6. A positive reaction may persist for several weeks after the patient 
is clinically cured. Torrey1 has shown experimentally that the antibody 
persists in the blood of rabbits artificially immunized for from ten days to 
six or seven weeks. Usually, under proper treatment, the reaction in ordinary 
cases of urethritis disappears in from two to three weeks; if, however, a posi- 
tive reaction persists, a focus of infection is probably present, and the patient 
should be kept under further observation and the treatment persisted in. 
7. In women the reaction is seldom positive until the infection has reached 
the cervical canal. In the case of little children, however, we have known 
positive reactions to occur in acute and chronic vulvovaginitis, indicating 
either that the disease is more severe in children, with more antibody forma- 
tion, or that it may reach the cervical canal. 
The reaction is positive in about 60 per cent, of cases of pyosalpingitis, 
and the test may prove of value in making the differentiation of inflam- 
matory lesions from certain cystic and neoplastic conditions, and in estab- 
lishing the gonorrheal basis of many of these infections. 
In a recent thorough study by Torrey, Wilson and Buckell2 of the com- 
parative value of smears, cultures, and complement-fixation tests in the 
diagnosis of chronic gonorrhea of women the following results were observed 
in 56 cases as taken from their report: 
1 Jour. Med. Research, 1910, i, 95. 
2 Jour. Infect. Dis., 1922, 31, 148. COMPLEMENT FIXATION IN GLANDERS 
547 
Comparison of Smear, Culture, and Complement Fixation in Various Stages of 
Gonorrhea in Women 
Diagnosis. 
Number 
of Cases. 
f 
Percentage Positive Diagnoses. 
Smear. 
Culture. 
Complement Fixa- 
tion. 
Acute gonorrhea 
1 
Not stated. 
Positive. 
Positive. 
Subacute gonorrhea.... 
8 
50 per cent. 
50 per cent. 
50.0 per cent. 
Chronic gonorrhea 
33 
12 “ 
20 “ 
69.5 
Doubtful gonorrhea.... 
14 
14 “ 
28 “ 
71.0 
The authors concluded that the complement-fixation test was of par- 
ticular value in the diagnosis of chronic gonorrhea, although smear and 
cultural methods may be better criteria of cure. 
8. Cases of gonorrheal arthritis yield from 80 to 100 per cent, of positive 
reactions, and the complement-fixation test has considerable value in 
establishing the diagnosis of these infections. 
9. The administration of gonococcus vaccine and antigonococcus serum is 
likely to be followed by positive reactions. Just how long the antibodies may 
persist in the blood after a clinical cure has been effected it is difficult to 
state; at least from six to twelve weeks’ time should be given for them to 
disappear. 
10. In medicolegal cases the courts may not accept the usual evidence 
offered by a bacteriologic diagnosis based upon stained smears of a secre- 
tion, and cultures are frequently differentiated from other Gram-negative 
diplococci only with difficulty. Conducted with the proper technic the 
gonococcus fixation test is highly specific and much less difficult to perform. 
11. Finally, it must be emphasized that the reaction has a far more 
positive than negative value. The reaction is highly specific, but there is a 
limit to its delicacy, so that a negative reaction in urethritis does not exclude 
the possibility of gonococcal infection. 
The complement-fixation test is used extensively by veterinarians in 
making a laboratory diagnosis of glanders. The test has been found very 
reliable, and is usually more delicate than the agglutination test and the 
Strauss guinea-pig test. It has also been used successfully in the diagnosis 
of human glanders. 
Preparation and Standardization of Antigen.—The antigen should be 
polyvalent, and composed of at least several different strains. Cultures 
of Bacillus mallei are grown on slants of glycerin agar (1.6 per cent, acid) 
for from forty-eight to seventy-two hours. The growths are then removed, 
and sufficient distilled water added to give a milky suspension. This 
suspension is sterilized by heating the tubes to 60 C.° for two hours. They 
are then shaken mechanically with glass beads for a few hours on two suc- 
cessive days. Enough sodium chlorid is added to make the solution isotonic, 
and the whole is preserved with 0.25 per cent, phenol and stored in a dark, 
cold place, where it will keep for many months. Antigen may also be pre- 
pared in the same manner as gonococcus antigen as described on page 545 
and employed by Wade.1 
COMPLEMENT FIXATION IN GLANDERS 
1 Jour. Infect. Dis., 1913, 12, 7. 548 
COMPLEMENT FIXATION IN BACTERIAL INFECTIONS 
A simpler antigen is prepared by growing the bacillus in glycerin bouillon 
for seventy-two hours, sterilizing by heating to 60° C. for two hours, and 
preserving with the addition of 0.25 per cent, of phenol. 
It is now generally conceded among veterinarians that the Bacillus 
abortus of Bang is the specific cause of contagious abortion of cows. 
Evidence is gradually accumulating to show that an organism belonging 
to the paratyphoid group is frequently the cause of a similar condition 
among mares (Kilbourne and Smith,1 Liguierer,2 Liguierer and Zabala, 
Good,3 Van Neelsbergen,4 de Jong,5 Meyer and Boerner6). Meyer and 
Boerner, who have studied this bacillus with particular care, classify it with 
the paratyphoid enteritidis group (Bacillus aborti equi). 
Veterinarians are generally agreed that in contagious abortion of cows 
the complement-fixation test is highly specific; frequently of considerable 
value in establishing a diagnosis (Meyer and Hardenburgh and others). 
Meyer and Boerner have found fixation of complement to occur in con- 
tagious abortion of mares with an antigen of Bacillus abortus equi, and recom- 
mend the test as diagnostic aid in this infection. 
Preparation of Antigen.—The antigen of Bacillus abortus (Bang) for 
use in the complement-fixation test in contagious abortion of cows is pre- 
pared by cultivating a number of strains of the bacillus, which have been 
trained to grow aerobically, upon slants of glycerin agar for seventy-two 
hours. The growths are then washed off with sufficient normal saline 
solution containing 2 per cent, phenol to yield a cloudy emulsion. Shake 
briskly in order to break up the clumps of bacilli, and filter through paper. 
Place in a refrigerator for several days to complete the sterilization, and 
titrate the anticomplementary dose each time before the main test is con- 
ducted. 
The antigen may also be prepared by cultivating a number of strains 
in glycerin-serum bouillon for five or six weeks. Centrifuge thoroughly 
and wash the bacilli once or twice with normal saline solution to remove 
all traces of serum. Dilute the wrashed bacilli with sufficient normal saline 
solution to give an emulsion equal in density to a twenty-four-hour bouillon 
culture of Bacillus coli, and add 0.4 per cent, of phenol as a preservative. 
The antigen of Bacillus abortus equi for making the complement-fixation 
diagnosis of contagious abortion of mares is prepared of eighteen- to twenty- 
hour-old glycerin bouillon cultures, with an addition of 0.5 per cent, of 
phenol. These antigens are less anticomplementary than shake extracts, 
and keep their titer unaltered for many weeks (Meyer and Boerner). They 
may also be used for making the macroscopic agglutination test. These 
antigens may also be prepared after the method used in the preparation of 
gonococcus antigen (page 545). 
COMPLEMENT FIXATION IN CONTAGIOUS ABORTION 
COMPLEMENT FIXATION IN DOURINE 
Dourine, or horse syphilis, is a specific infectious disease of the horse 
and ass, transmitted from animal to animal by the act of copulation, and 
caused by the Trypanosoma equiperdum. It is characterized by an irregular 
1 United States Department of Agriculture, Bulletin No. 3, 1893, 49, 53. 
2 Rec. Med. veterinaire, lxxxii, 1905. 
3 Kentucky Agriculture Exper. Station Bulletin No. 165, 1912. 
4 Tijdschrift, v. Veeartsenijk., xxiv, 1912. 
5 Archiv. f. Wissenshaftl. u. prak. Tierheilkunde, xxv, 1900. 
6 Jour. Med. Research, 1913, xxix, No. 2, 325. COMPLEMENT FIXATION IN TYPHOID AND PARATYPHOID FEVER 
549 
incubation period, the localization of the early symptoms to the genital or- 
gans, and, finally, by complete paralysis of the posterior extremities, a fatal 
termination ensuing in from six months to two years. 
The disease is especially prevalent among horses in the northwestern 
states, and may occur in such various and atypical forms as to render clinical 
diagnosis difficult. 
Complement-fixation methods of diagnosis have been tried by Pavlosvici, 
Winkler and Wyschelersky, Moller, Watson, Brown, and in a large series of 
cases with good results by Moller, Eichhorn, and Buck.1 These last-named 
investigators examined 8657 specimens of blood from horses in Montana and 
North and South Dakota, and of these, 1076 yielded positive reactions. 
In most of these experiments the results were corroborated by clinical 
and pathologic findings, and the investigators conclude that the comple- 
ment-fixation test is of great value, especially in countries where only one 
of these protozoan diseases exists. 
Preparation of Antigen.—This is the most difficult part of the technic, 
because the trypanosome is not readily grown on artificial culture-media. 
Watery, alcoholic, and acetone extracts of various organs of horses dead of 
the disease do not yield satisfactory antigens. Since the reaction is a group 
reaction, and dourine is the only trypanosome infection in this country, 
Moller, Eichhorn, and Buck selected the surra organism for the preparation 
of antigen. After infecting a dog and at the height of infection withdrawing 
200 c.c. of blood into potassium citrate and hemolyzing with 0.5 gm. of 
saponin, the trypanosomes were secured after thorough centrifugalization 
and washed three times. After the last washing the trypanosomes were 
emulsified in 50 c.c. of salt solution and preserved with phenol. This antigen 
yielded highly satisfactory results, but the difficulty of preparing it, and the 
small quantity secured, made it necessary that another method be used. 
An extract of the spleen of a rat just dead of surra was found to yield 
a satisfactory antigen. The extract does not keep well, and must be pre- 
pared freshly every few days and carefully standardized. Gray or white 
rats are infected with surra by injecting 0.2 c.c. of blood from a seed rat or 
rabbit with this disease. If a large number of tests are to be made, the rats 
should be so infected that one or two are available each day for the prepara- 
tion of the antigen. 
The spleen from a rat with a small amount of salt solution added is 
ground in a mortar until a pulpy mass results. More of the salt solution 
is added from time to time, and the suspension thus obtained is filtered 
twice through a double layer of gauze and diluted with salt solution to 40 c.c. 
I have secured better preparations by using the liver instead of the spleen; 
such antigens show the presence of more trypanosomes. 
The anticomplementary and antigenic units are then determined, and 
the extract used in double the antigenic unit, providing that this amount 
is not more than half the anticomplementary unit. If a positive serum from 
an infected horse is not available, the anticomplementary unit may be 
determined and half this amount used in conducting the main test. 
COMPLEMENT FIXATION IN TYPHOID AND PARATYPHOID FEVER 
Typhoid fever was one of the original diseases in which Bordet and 
Gengou first demonstrated the occurrence of complement fixation. Widal 
and Lesourd attempted to make practical application of this method in 
the diagnosis of the disease, but their results were indifferent, and since 
1 Amer. Jour. Veter. Med., 1913, viii, 581. 550 
COMPLEMENT FIXATION IN BACTERIAL INFECTIONS 
then numerous writers have expressed various opinions as to the value of 
the test. Garbat1 has secured uniform and reliable reactions with a 
polyvalent antigen, and emphasizes the importance of this factor. He2 
has found that practically all individuals with typhoid fever develop comple- 
ment-fixing antibodies sooner or later during the course of the disease and 
that the test possesses diagnostic value. He also observed that people 
inoculated with the triple typhoid and para typhoid vaccine gave positive 
reactions only occasionally and for a very short time. For this reason 
Garbat believes that the test becomes of special value as a differential 
diagnostic aid in patients who are suspected of having typhoid fever, and 
give a positive Widal reaction due to previous vaccinaion, but in whom 
no typhoid bacilli can be found in the blood or excretions. Felke3 has 
arrived at the same conclusion, and Hage and KorfE-Petersen4 found that 
the serum of vaccinated individuals may yield positive reactions for at 
least two months. 
In my experience the typhoid complement-fixation test with polyvalent 
antigens has proved a valuable diagnostic aid for typhoid fever. Not 
infrequently the tests have yielded positive reactions earlier than the Widal 
test, employing the usual 1 :40 and 1 :80 dilutions of serum. Likewise 
the serums of rabbits immunized with living and dead typhoid bacilli have 
usually shown that complement-fixing antibody is produced very promptly 
and usually several days before the agglutinins are detectable. Hadjo- 
poulos5 has also found that complement fixation is one of the earliest and 
most constant reactions in typhoid fever. 
Vaccination with the usual three doses of triple vaccine results in the 
production of complement-fixing antibody in practically all individuals. 
As found by Garbat, these antibodies usually disappear after a month or two, 
but Berge and myself6 have observed positive reactions for as long as four 
months, and in one instance, for about two years. In the serum diagnosis 
of typhoid fever in an individual previously vaccinated, I believe it is ad- 
visable to conduct the tests with varying amounts of serum at intervals of 
two or three days. If typhoid infection is present a rise in complement- 
fixing antibody is observed; if typhoid is not present the complement-fixing 
titer of the serum remains fairly constant in much the same manner as the 
agglutinin curve determined by the Dreyer method. 
The complement-fixation test will probably prove of value in the diag- 
nosis of paratyphoid fever, employing antigens of paratyphoid bacilli. I 
have had no actual experience with the test for this purpose employing 
human serums, but with experimental infections in rabbits the complement- 
fixation reaction has proved more sensitive than the agglutination reaction 
for detecting and differentiating paratyphoid infections. It is necessary, of 
course, to use varying amounts of serum when conducting the tests in order 
to avoid group reactions. 
Preparation of Antigen.—The antigen is prepared of numerous strains of 
typhoid bacilli—the more, the better. In Garbat’s method cultures are grown 
on slants of agar for twenty-four hours, washed off with small quantities 
of sterile distilled water (about 1 c.c. to an agar slant), heated to 60° C. for 
twenty-four hours, shaken mechanically for twenty-four hours, and thor- 
1 Amer. Jour. Med. Sci., 1914, 148, 84; Jour. Immunology, 1916, 1, 391. 
2 Monograph No. 16 of the Rockefeller Institute, May 10, 1922, 82. 
3 Munch, med. Wchn., 1915, 52, 578. 
4 Deutsch. med. Wchn., 1915, 41, 1328. 
5 Jour. Infect. Dis., 1922, 31, 226. 
6 Jour. Immunology, 1916, 1, 409. COMPLEMENT FIXATION IN TUBERCULOSIS 
551 
oughly centrifugalized for four to eight hours until the supernatant fluid is 
absolutely clear. The filtrate is preserved with 0.25 per cent, phenol and 
used as antigen. The antigen is titrated for anticomplementary activity 
and used in one-half this amount. 
I have secured good results by removing the growths with small amounts 
of normal salt solution, and placing them in a shaking flask, and shaking for 
one hour to break up the clumps. After heating to 60° C. for one hour, 
1 per cent, glycerin and 0.25 phenol are added as preservatives, and the 
mixture stored away in ampules containing 1 c.c. each. The emulsion 
should be slightly milky in appearance. A comparative study of nine methods 
for preparing typhoid antigens by Motsumoto,1 in my laboratory, showed 
that these simple suspensions utilizing the broken up bodies of the bacilli 
were more satisfactory than filtrates or centrifugates of autolyzed bacilli. 
COMPLEMENT FIXATION IN TUBERCULOSIS 
It was the original studies in complement fixation in tuberculosis made 
by Wassermann and Bruch that later induced these workers, in co-operation 
with Neisser, to apply the method to the diagnosis of syphilis. 
The first application of complement-fixation in tuberculosis was made 
by Widal and LeSourd2 in 1901. They obtained deviation of complement 
in certain cases of tuberculosis, using as antigen homogeneous emulsions of 
tubercle bacilli of the Arloing-Courmont strain. In 1903 Bordet and Gengou3 
demonstrated the presence of antibody capable of uniting with tubercle 
bacilli and fixing complement in the sera of tuberculous animals. Wasser- 
mann and Bruch4 in 1906 demonstrated the presence of an antibody to tu- 
berculin in patients treated with tuberculin, but they examined only 13 
cases of pulmonary tuberculosis. Caulfield5 in 1911 examined 104 cases of 
pulmonary tuberculosis with bacillary emulsion as antigen and obtained 33 
per cent. Turban I cases, 70 per cent. Turban II, and 62 per cent. Turban 
III positive results. Laird6 (1912), out of 84 tests in 34 cases, obtained 24 
positives in 4 cases, using watery emulsion of tubercle bacilli (which he does 
not describe); his results were inconclusive. Hammer,7 using O. T. and 
extracted tuberculous nodules, obtained 97 per cent, positive results in 46 
tuberculous cases. Calmette and Massol,8 using preparations made from 
tubercle bacilli by extracting with water and peptone, obtained in 134 cases 
92.5 per cent, fixation. Fraser9 (1913), testing a large variety of antigens, 
found that living bacilli gave no fixation in 96.6 per cent, of normal indi- 
viduals, but gave positive reactions in 42.3 per cent, of tuberculous indi- 
viduals. She states that the most reliable antigen is prepared from living 
human bacilli, and that diagnostically the complement-fixation test with 
living bacilli is of more value from the standpoint of positive results than 
any other reaction discovered to date. She believes the absence of anti- 
bodies accounts for the low percentage of results obtained. Dudgeon. 
Meek, and Weir10 also tested a large number of antigens, and in 102 untreated 
cases obtained 86 positive results, while all cases which had been treated 
with tuberculin gave positive results. Products of the bacilli themselves 
were found to be the most satisfactory as antigen. With an alcoholic 
1 Jour. Immunology, 1920, 5, 111. 
2 Cited by Shennan and Miller, Edinburgh Med. Jour., 1913, 10, 81. 
3 Compt. rend. Acad, de Sci., 1903, 137, 351. 
4 Deutsch. med. Wchn., 1906, 32, 449. 
6 Jour. Med. Research, 1911, 24, 122. 8 Compt. rend. Soc. de biol., 1912, 73,120. 
6 Jour. Med. Research, 1912, 27, 163. 9Ztschr. f. Immunitatsf., 1913, 20, 291. 
7 Munch, med. Wchn., 1912, 59, 1750. 10 The Lancet, 1913, 184, 19. 552 
COMPLEMENT FIXATION IN BACTERIAL INFECTIONS 
antigen1 prepared from tubercle bacilli they obtained from a total of 234 
cases, 209 (89.3 per cent.) positives, 194 of these on first examination, 11 
(of the 15 negative) on second examination, and 4 more on third examination. 
Besredka2 (1913) prepared an antigen by growing tubercle bacilli on egg 
broth, heating it, and filtering. With this antigen Bronfenbrenner3 (1914) 
obtained a very high percentage of positive results, 93.8 per cent, in active 
cases, and 55.5 per cent, in convalescents, while suspected cases gave 75 
per cent, and syphilitic sera 24 per cent, positive reactions. Inman4 and 
Kuss, Leredde and Rubenstein5 found this antigen non-specific. McIntosh, 
Fildes, and Radcliffe6 (1914) also justly criticized Besredka’s antigen, and 
concluded, after testing a large number of antigens, that the living bacillary 
emulsion was best, yielding 76.7 per cent, positive results in 43 definite cases 
of phthisis, 80.7 per cent, in surgical tuberculosis, and 37.5 per cent, in gland- 
ular tuberculosis. Of 87 normal individuals, only 3 gave positive reactions 
(2 of these were lepers and 1 had Addison’s disease). Negative reactions 
were obtained in 18 syphilitic patients. They look upon a positive reaction 
as indicative of active tuberculosis. Stimson7 (1915), who gives a fairly 
exhaustive table of the recent literature, reports a small number of cases in 
which a variety of antigens were used, but his results were inconclusive. 
Craig8 (1915) reports the results of examination of 166 cases of pulmonary 
tuberculosis in which he employed as antigen an alcoholic extract of several 
strains of human tubercle bacilli which had been grown on a liquid medium 
of alkaline broth containing egg; 96.2 per cent, positive results were obtained 
in active cases and 66.1 per cent, positive in inactive cases. One hundred 
and fifty cases of syphilis gave only 2 positive reactions, and these, on 
further examination, revealed lesions in the lungs. One hundred other 
diseases examined all gave negative results. In a later report Craig, using 
his same antigen of a filtrate of an alcoholic extract of several strains of 
human tubercle bacillus plus the culture-medium in which the bacilli had 
been grown, has reported further favorable results in complement fixation in 
tuberculosis. Miller and Zinsser9 have described a simple antigen prepared 
by grinding living or dead tubercle bacilli with dry table salt, and then 
adding distilled water up to isotonicity. With this antigen they have 
reported excellent results; Miller10 found the reaction practically always 
positive in active tuberculosis with negative reactions in non-tuberculous 
syphilitic and normal persons. In my own studies in co-operation with 
Montgomery, antigen prepared after Miller’s method, yielded positive 
reactions only with the sera of tuberculous persons, but the percentage of 
positive reactions was much less than reported. Positive reactions were 
also occasionally observed with the serum of actively syphilitic individuals 
presenting no clinical evidence of tuberculosis. Corber,11 employing an au- 
tolysate of tubercle bacilli as antigen, found 30 per cent, positive reactions 
with clinically definite cases of tuberculosis both active and inactive. Active 
cases gave a higher percentage of positive results than inactive cases. Eich- 
horn and Blumberg,12 using Besredka’s antigen and the sera of tuberculous 
persons and the lower animals, found the complement-fixation test less 
reliable than the subcutaneous tuberculin test in the diagnosis of tuberculosis 
1 Jour. Hyg., 1914, 14, 52, 72. 4 Compt. rend. Soc. de biol., 1914, 76, 251. 
2 Compt. rend. Acad. d. sc., 1913, 156, 1633. 5 Compt. rend. Soc. de biol., 1914, 76, 244. 
3 Arch. Int. Med., 1914, 14, 786. 6 The Lancet, 1914, 185, 485. 
7 Bull. Hyg. Lab. U. S. P. H. and M. H. S., 1915, No. 101, 7. 
8 Amer. Jour. Med. Sci., 1915, 150, 781; Jour. Amer. Med. Assoc., 1917, lxviii, 773. 
9 Proc. Soc. Exper. Biol, and Med., 1916, xiii, 134. 
)0 Jour. Amer. Med. Assoc., 1916, lxvii, 1519. 
11 Jour. Infect. Dis., 1916, 19, 315. 12 Jour. Agr. Research, 1917, viii, 1. COMPLEMENT FIXATION IN TUBERCULOSIS 
553 
in cattle and not practical for general diagnostic purposes. Lucke1 found 
that the wax of tubercle'bacilli possessed no antigenic properties; Wood, 
Bushnell, and Maddux2 have observed a large percentage of apparently 
specific reactions with the partial antigens of Deyke and Much.3 Bron- 
fenbrenner4 has reported very favorably upon the diagnostic value of the 
test employing Besredka’s antigen; also von Wedel5 employing Wilson’s 
antigen. Moon,6 McCaskey,7 and others have made favorable reports. 
Preparation of Antigen.—It is apparent, therefore, that the question of 
proper antigen is of paramount importance in complement fixation in tuber- 
culosis. The main objections to the bacillary emulsion are the small differ- 
ence between the antigenic and anticomplementary units, the turbidity, 
and fairly high percentage of non-specific reactions. Eichhorn and Blum- 
berg have given the following directions for the preparation of a modified 
Besredka antigen: 
Twenty c.c. each of the white and the yolk of an egg are thoroughly 
beaten in an automatic egg-beater, and to the whipped material a solution 
of Liebig’s meat extract (3 : 1000) in distilled water is gradually added while 
the mixture is continuously beaten. A sufficient meat-extract infusion is 
used to make up the whole to 1000 c.c. The emulsion is strained through 
cotton and heated to the boiling-point, then strained again, and after the 
addition of 0.5 per cent, of sodium chlorid is carefully neutralized, heated 
again and strained, and neutralized if necessary. To the neutral medium 
sufficient normal sodium hydroxid solution is added to make the medium of 
0.2 alkalinity. This medium, to which neither peptone nor glycerin is 
added, is then autoclaved at 115° C. for twenty minutes, and after cooling 
is kept at 37° C. for forty-eight hours for observation as to sterility. 
Miller-Zinsser Antigen.—Young cultures are removed from a solid me- 
dium with saline solution autoclaved and centrifuged. To each 0.020 gm. 
of bacterial residue in a centrifuge tube add 0.090 gm. of sodium chlorid 
and stir with a glass rod for one hour; add 10 c.c. of water.8 
Wilson Antigen.—Broth cultures (tubercle bacilli grown for three weeks 
in glycerin-broth) are sterilized by heating in an Arnold sterilizer for one 
hour followed by filtering through paper. To each volume of residue is 
added ten volumes of alcohol, shaken by hand and extracted in a refrigerator 
for two weeks. This emulsion is then filtered through paper and the filtrate 
discarded, the residue being washed in absolute alcohol, secured by centrif- 
ugalization and washed twice with ether. The residue is now dried in the 
centrifuge tube overnight in an incubator and each gram emulsified in 200 
c.c. of saline solution, heated for one hour at 80° C. and kept in a refrigerator 
as stock antigen.9 
Petrof’s Antigen.—Washed bacteria are dried and thoroughly pul- 
verized. One gram of this residue is triturated in a mortar with 100 c.c. of 
25 per cent, glycerin in water and slowly boiled in a flask having a reflex 
condenser. The material is now centrifuged or set aside for a few hours 
and the supernatant fluid used as antigen. This is one of the methods 
described by Petroff10 for the preparation of an antigen. 
1 Jour. Immunology, 1916, 1, 457. 2 Jour. Immunology, 1917, 2, 301. 
3 Med. Klinik., 1908, 4, 1540; Munch, med. Wchnschr., 1909, 56, 1895; ibid., 1909, 56, 
1825; Beitr. z. klin. Tuberk., 1910, 15, 277; ibid., 1911, 20, 343. 
4 Jour. Lab. and Clin. Med., 1917, 3, 51. 
5 Jour. Immunology, 1920, 5, 159 (also gives a good review of the literature). 
6 Jour. Amer. Med. Assoc., 1918, 71, 1127. 
7 Amer. Jour. Med. Sci., 1917, 154, 648. 
8 Proc. Soc. Exper. Biol, and Med., 1916, 13, 134. 
9 Jour. Immunology, 1918, 3, 345. 10 Amer. Review of Tuberculosis, 1918, 2, 523. 554 
COMPLEMENT FIXATION IN BACTERIAL INFECTIONS 
Fleischer-Ives Antigen.—Young cultures are removed from a solid 
medium with saline and centrifuged or filtered through paper. Partially 
dried bacterial residue is weighed and gradually rubbed up in a mortar 
with sufficient saline solution to yield a 0.5 per cent, suspension; sterilized 
and preserved by adding one-tenth volume of 5 per cent, phenol.1 
Non-specific Reactions in Syphilis.—Since tubercle bacilli contain un- 
usually large amounts of lipoids, it is possible that non-specific reactions 
may occur with the serums of actively syphilitic individuals yielding strongly 
positive Wassermann reactions, but showing no clinical evidence of tuber- 
culosis. I have studied all of the above described antigens for this property 
and found that all of them are capable of yielding weakly positive reactions 
with the serums of syphilitics yielding strongly positive Wassermann reac- 
tions. This is especially true if the antigen is employed in an amount 
corresponding to one-half of the anticomplementary unit. For this reason 
I thoroughly extract the bacilli with fat solvents before preparing antigen 
and do not use more than one-quarter of the anticomplementary amount 
for conducting the test; under these conditions non-specific reactions caused 
by syphilis reagin are avoided. 
Practical Value of the Complement-fixation Test in Tuberculosis.—It is 
still too early to express an accurate opinion of the practical value of this 
test; an analysis of numerous reports and my own studies apparently war- 
rants the following statements: 
1. A definitely positive reaction taken in conjunction with other findings 
makes the diagnosis of tuberculosis certain and may be of distinct aid in the 
diagnosis of early tuberculosis. 
2. From the standpoint of differential diagnosis the test is of value 
when positive as indicating tuberculosis against other diseases of the lungs, 
as carcinoma, syphilis, abscess, empyema, etc. 
3. As the reaction is likely to be positive only in active cases, the com- 
plement-fixation test is superior to other biologic tests for active tubercu- 
losis, as, for example, the skin test, which is likely to be positive in persons 
with quiescent lesions. 
4. A persistently and strongly positive reaction probably indicates the 
presence of active tuberculosis. 
5. The strength of a positive reaction appears to bear some relation 
to the severity of the disease, and reactions becoming gradually weaker 
until negative have been frequently noted with clinical improvement and 
“cure.” 
COMPLEMENT FIXATION IN PERTUSSIS 
Investigations on complement fixation with the bacillus of pertussis has 
given varied results, perhaps due in part to a variation in the cultures em- 
ployed and the methods used to prepare the antigens. 
Bordet and Gengou2 used suspensions in salt solution of growths on 
solid media. Arnheim3 found negative results with antigens of both the 
Bordet-Gengou bacillus of pertussis and the influenza bacillus. Wollstein4 
used three forms of antigen: suspensions of the bacilli in salt solution; 
extracts of bacilli made by suspending the growths of three blood-agar 
slants in 5 c.c. of salt solution and shaking for twenty-four hours in thermo- 
stat; and extracts of tissue obtained from patients dying from pertussis. 
1 Jour. Lab. and Clin. Med., 1918, 3, 302. 
2 Ann. de l’Inst. Pasteur, 1906, 20, 731. 
3 Berl. klin. Wchn., 1908, xlv, 1453; Arch. f. Kinderh., 1909, 1, 295. 
4 Jour. Exper. Med., 1909, 11, p. 41. COMPLEMENT FIXATION IN PERTUSSIS 
555 
Friedlander and Wagner1 used live bacteria and fresh serum, and considered 
this innovation of great importance. The different hemolytic systems em- 
ployed may also have had some bearing on the failure to obtain identical 
results. 
Bordet and Gengou obtained complement fixation in all of their cases of 
pertussis. They used 0.1 c.c. to 0.3 of the human serum heated to 56° 
C., and 0.05 c.c. to 0.01 of fresh guinea-pig serum as complement. The 
amboceptor was the serum of rabbits previously injected with sheep corpus- 
cles. The antigen was an emulsion of the bacillus of pertussis in salt solu- 
tion, the growth being twenty-four hours old. 
Arnheim2 obtained complement fixation in 6 to 12 cases of pertussis. 
Wollstein examined the serum from 9 patients with pertussis and in no 
instance did she obtain complement fixation, using the three forms of anti- 
gen just mentioned in all cases. The quantities of complement and ambo- 
ceptor were about the same as those used by Bordet and Gengou. 
Gengou and Brunard3 describe 3 cases in which they determined the 
specific pertussis character of the infection by complement fixation. 
In 1911 Bordet and Gengou4 reported certain atypical cases diagnosed 
as pertussis by means of complement fixation. A little later Bordet5 con- 
cluded that the power to fix complement is not found early, and does not 
become marked until near the end of the disease. 
St. Bacher and Menschikoff6 report 27 cases of pertussis in all stages in 
which attempts were made to obtain complement fixation, without success 
in a single case. Only after vaccines of pure cultures of the Bacillus per- 
tussis were given did fixation occur. Their antigen was an emulsion of the 
bacillus of pertussis in salt solution, while 0.4 c.c. of a 1 :10 dilution of 
guinea-pig serum served as complement, and 0.5 c.c. of a 1 :150 dilution of 
serum of a rabbit previously injected with sheep corpuscles served as am- 
boceptor. 
Delcourt7 obtained complement fixation in 6 cases of pertussis. Poleff8 
gives a resume of the results of five investigators. In 5 cases only was 
there complement fixation and in 31 cases no fixation was obtained. He 
himself reports 10 cases in which he did not obtain fixation. 
Hess9 tested 10 cases and concluded ‘ ‘that results would seem to show that 
this reaction is present for some months after cessation of all symptoms.” 
Recently Anna Wessels Williams10 came to the conclusion that the test 
with serum of human beings is not as clear cut as it seems to be with serum 
of animals which have been injected with the bacillus of pertussis. 
Renaux,11 using the Bacillus pertussis for antigen, examined 73 sera for 
complement fixation, 32 cases of which were known cases of pertussis. Of 
the cases of pertussis, he obtained complement fixation in 23. Of the 9 
negative cases, the serum had been obtained early during the disease, and 
3 of these gave positive complement fixation when the attack had lasted 
about four weeks. No further examination was made on the remaining 6 
1 Amer. Jour. Dis. of Children, 1914, 8, p. 134. 
2 Berl. klin. Wchn., 1908, 14, p. 1453. 
3 Bull. Acad, de med. Belg., 1910, 24, p. 329. 
4 Centralbl. f. Bakteriol., I. O., 1911, 58, p. 573. 
5 Centralbl. f. Bakteriol., Orig., 1912, 66, p. 275. 
6 Centralbl. f. Bakteriol., Orig., 61, p. 218. 
7 Presse med. Belg., 1912, 64, p. 19. 
8 Centralbl. f. Bakteriol., I. O., 1913, 69, p. 23. 
9 Jour. Amer. Med. Assoc., 1914, 63, p. 1007. 
10 Arch. Pediat., 1914, 31, p. 567. 
11 Centralbl. f. Bakteriol., I. O., 1914, 75, p. 197. 556 
COMPLEMENT FIXATION IN BACTERIAL INFECTIONS 
cases. His results seem to show that fixation appears about three or four 
weeks after the appearance of the whoop. 
In 18 cases of pertussis Friedlander and Wagner1 obtained complement 
fixation in each case. They used fresh serum of pertussis patients and 
living bacteria. The Noguchi hemolytic system was used throughout their 
work on account of the small quantity of serum necessary. They claim that 
the diagnosis of pertussis can be made with certainty in the catarrhal stage 
by means of complement fixation. In a more recent article Freidlander2 
reports further results. He obtained fixation in 13 of 14 cases in the catar- 
rhal stage before the characteristic whoop had appeared. In one case there 
was fixation three weeks before the first whoop. Winholt2 has reported 
positive reactions in pertussis with polyvalent antigens prepared by culti- 
vating the bacilli on potato-glycerin blood-agar, suspending in normal salt 
solution, and heating to 56° C. for thirty minutes. Olmstead and Povitsky4 
have been able to differentiate between Bacillus pertussis and B. influenza 
by means of the complement-fixation tests employing rabbit immune sera. 
Preparation of the Antigen.—Antigens may be prepared in the same man- 
ner as gonococcus antigen (page 545) and should be polyvalent. The 
antigen should always be titrated for the anticomplementary unit as de- 
scribed in the gonococcus complement-fixation test, and used in a dose 
corresponding to one-third the anticomplementary unit. 
Practical Value of the Test.—Positive reactions are not usually obtained 
until the paroxysmal stage is reached, when about 50 per cent, of sera react 
positively (Park). For this reason the complement-fixation test possesses 
little value in early diagnosis. It is at times of value in the diagnosis of 
atypical cases. 
With rabbit immune sera the test has proved of value in the study and 
differentiation of micro-organisms resembling Bacillus pertussis. 
COMPLEMENT FIXATION IN OTHER DISEASES 
Hog-cholera.—Healy and Smith5 have observed positive reactions em- 
ploying a salt solution extract of mesenteric glands of cholera hogs as antigen; 
King and Drake6 have reported a large percentage of apparently specific 
reactions employing antigens of Spirocheta hyos. At the present time the 
• value of complement-fixation tests in the diagnosis of hog-cholera cannot 
be stated and the subject requires further investigation. 
Acute Anterior Poliomyelitis.—Freese and myself7 have observed a small 
percentage of positive reactions with sera employing salt solution extracts 
of poliomyelitic brain and spinal cord for antigens. Neustaedter8 has 
reported as high as approximately 90 per cent, of sera of acute cases of polio- 
myelitis yielding positive reactions with an antigen prepared by digesting 
an extract of poliomyelitic monkey brain and spinal cord. The subject 
requires further investigation and may be placed on a practical diagnostic 
basis. 
Vaccinia, Smallpox, and Chickenpox.—In 1906 Jobling9 reported posi- 
tive complement-fixation reactions using as antigen a suspension of finely 
ground calf vaccine pulp and the sera' of vaccinated calves. As far as I 
am aware this work constitutes the first investigation of its kind conducted 
in this field and, indeed, was in progress at the time of the original comple- 
1 Amer. Jour. Dis. of Children, 1914, 8, p. 134. 
2 The Lancet-Clinic, 1915, 113, p. 8. 6 Jour. Infect. Dis., 1916, 19, 46. 
3 Jour. Infect. Dis., 1915, 16, 389. 7 Jour. Immunology, 1917, 2, 327. 
4 Jour. Med. Research, 1916, 33, 379. 8 New York State Jour. Med., 1918, 18, 328. 
5 Jour. Infect. Dis., 1915, 17, 213. 9 Jour. Exper. Med., 1906, 8, 707. COMPLEMENT FIXATION IN OTHER DISEASES 
557 
ment-fixation work of Wassermann and his co-workers and Detre in syphilis. 
In 1909 Beintker1 reported indifferent results in which he used as antigen a 
salt solution extract of calf lymph and the sera of vaccinated rabbits and 
calves as well as the sera of persons suffering with smallpox. In the same 
year Sugai,2 using as antigens salt solution extracts of the contents of small- 
pox pustules and calf lymph, reported positive reactions with the sera of 6 
smallpox patients and 5 persons who had been vaccinated with calf lymph. 
He found no evidence of agglutination in an extract of calf virus material 
by these sera. Dalm3 reported positive results with the sera of 10 persons 
suffering with smallpox and an antigen of calf lymph; he also used watery 
extracts of the liver and spleen of a two-year-old child who had succumbed 
to smallpox, but with negative or indifferent results; this investigator also 
found agglutination of calf lymph by the sera of smallpox patients to be 
weak or entirely absent. Kryloff,4 using a salt solution extract of variola 
matter, reported positive reactions in variola and varioloid and considered 
the complement-fixation test as possessing diagnostic value. Aqueous and 
alcoholic extracts of variolous scabs and the liver and spleen of persons dead 
of variola as well as salt solution extracts of calf lymph yielded negative or 
weaker reactions. Bermbach5 tested the sera of animals before and after 
vaccination and the sera of vaccinated and revaccinated persons with a 
salt solution extract of lymph and reported generally positive reactions. 
Xylander6 tested 31 sera of persons before and after vaccination with 
an alcoholic extract of lymph in an effort to study the specificity of the 
Wassermann reaction in syphilis. Of these, the sera of 8 reacted weakly 
positive with the lymph antigens, and 7 of these reacted in the same 
manner with the alcoholic extract of heart; after vaccination 18 re- 
acted weakly positive with the lymph antigen and these included the sera of 
the same 7 persons who reacted weakly positive with the alcoholic 
heart antigen prior to vaccination. Teisser and Gastinell,7 using as antigen 
a salt solution extract of calf lymph, reported positive reactions with the 
sera of 39 persons suffering with variola, the reactions becoming 
positive about the tenth day of the disease and remaining so for at least 
thirty days; more prolonged observations were not made. Positive reac- 
tions were also reported with the sera of vaccinated rabbits, the reactions 
becoming positive about the seventh day after vaccination. Recently 
Klein8 has also reported favorable results with the complement-fixation test 
in variola especially with antigens prepared of the contents of the vesicles 
and pustules of variola and believes the test to possess practical diagnostic 
value. Konschegg9 has also reported positive reactions in variola with salt 
solution extracts of the contents of variolous lesions. Chang and Chen,10 
however, in a study employing the serums of 50 small-pox patients and 
antigens of variolous lymph, observed negative reactions with forty-five 
serums. 
In my work11 antigens for the complement-fixation tests were prepared 
of the contents of vesicles and pustules and the scabs of vaccinia and vac- 
cine pulp from calves and from the contents of vesicles and pustules and 
scabs of smallpox patients. With these antigens complement-fixation reac- 
tions were conducted with homologous antigens and sera and the immuno- 
1 Centralbl. f. Bakteriol., 1909, 48, 500. 4 Centralbl. f. Bakteriol., 1911, 60, 651. 
2 Centralbl. f. Bakteriol., 1909, 49, 650. 5 Centralbl. f. Bakteriol., 1909, 48, 618. 
3 Centralbl. f. Bakteriol, 1911, 51, 136. 6 Centralbl. f. Bakteriol., 1909, 51, 290. 
7 Centralbl. f. Bakteriol., 1912-13, ref., 55, 555. 
8 Munch, med. Wchn., 1914, 61, 2270. 10 Ztschr. f. Immunitatsf., 1921, 31, 18. 
9 Munch, med. Wchn., 1915, 62, 4. 11 Jour. Immunology, 1916, 1, 59. 558 
COMPLEMENT FIXATION IN BACTERIAL INFECTIONS 
logic relationship between vaccinia and variola studied by crossing the 
antigens and sera in complement-fixation experiments; the results were as 
follows: 
1. The sera of rabbits inoculated with cowpox virus yielded positive com- 
plement-fixation reactions with salt solution antigens of cowpox and small- 
pox viruses in seven to eight days after vaccination. 
2. The antibody of cowpox virus in the sera of vaccinated animals 
showed a distinct and close biologic relationship to the antigen of variola 
in complement-fixation experiments. 
3. Of 13 persons vaccinated with cowpox virus from seven days to ten 
years previously, and whose sera yielded negative Wassermann reactions, 
positive reactions with salt solution antigens of cowpox virus were observed 
with 4, or 22 per cent. The sera of 1 of these 4 persons yielding positive 
reactions (vaccinated eight, twenty-one, twenty-one, and twenty-four days 
previously) with cowpox virus, reacted positively with a salt solution anti- 
gen of variolous material. The sera of unvaccinated persons reacted 
negatively with all antigens. 
4. Of 17 persons suffering with mild smallpox, the sera of 9, or about 60 
per cent., yielded positive complement-fixation reactions with salt solution 
antigens of variolous and cowpox viruses. While the degree of comple- 
ment absorption was relatively weak in all instances, the reactions were 
generally stronger with the variolous antigens than with the cowpox antigens. 
5. Alcoholic extracts of variolous and cowpox viruses possessed little or 
no antigenic sensitiveness. 
6. These complement-fixation reactions have demonstrated the close 
biologic relationship between the antibodies of vaccinia and variola; it is 
probable that complement-fixation reactions with salt solution antigens of 
the contents of smallpox lesions or fresh cowpox virus will prove of some 
value in the diagnosis of smallpox. 
Girard1 has observed specific reactions in varicella and believes that the 
complement-fixation test possesses diagnostic value for differentiating be- 
tween this disease and smallpox. 
Echinococcus (Hydatid) Disease and Intestinal Helminthiasis.—Several 
investigators have employed a complement-fixation technic in the diagnosis 
of echinococcus disease of the liver. Ghedini2 found this test of value, and 
he also reports specific reactions with the sera of persons infested with 
Ankylostoma duodenales and Ascaris lumbricoides.3 Weinberg,4 Jiani,6 
Israel,6 and Henius7 report favorable and specific reactions in echinococcus 
disease with aqueous or alcoholic extracts of cyst fluid, or both. Kurt 
Meyer8 found these reactions non-specific, in that the serum of a person 
infested with echinococcus showed complement absorption with antigens of 
senia solium and T. saginata, and vice versa. Branes9 found that alcoholic 
extracts of echinococcus cyst fluid showed complement absorption with the 
syphilis antibody as well as with that of echinococcus disease, and this 
1 La Reaction de Deviation Dans La Varicelle, 1918, Jouve & Co., Paris. 
2 Gaz. degli Ospedali e delle Cliniche, 1906, 27, p. 1616; 1907, 28, p. 53. 
3 Ibid., 1907, 28, p. 476. 
4 Ann. de l’Inst. Pasteur, 1909, 23, p. 472 (in which references are given to his previous 
work in this field). 
6 Wien. klin. Wchnschr., 1909, 22, p. 1439 (in which the author gives a good bibliography 
of earlier literature). 
8 Ztschr. f. Hyg. u. Infektionskrankh., 1910, 66, p. 487. 
7 Deutsch. med. Wchnschr., 1911, 37, p. 1212. 
8 Berl. klin. Wchnschr., 1910, 47, p. 1316. 
8 Munch, med. Wchnschr., 1911, 58, p. 1073. COMPLEMENT FIXATION IN OTHER DISEASES 
559 
observation has been generally confirmed. Thomsen and Magnusson,1 in 
a study of 12 cases of echinococcus disease, found that the sera of 10 reacted 
positively; the sera of 55 control cases (32 of which reacted positively to the 
Wassermann reaction) all were negative except one. These authors also 
report favorably on the specificity of the reaction: the sera of 10 persons 
infected with senia saginata, of 2 with T. solium, and of 1 with Bothrio- 
cephalus latus, all were negative with echinococcus antigen. Zapelloni2 
has made a particularly exhaustive study compiling from literature and his 
own experience the results observed with about 700 cases; about 93 per cent, 
of sera gave positive reactions, while controls including the sera of individuals 
with tapeworm reacted negatively. 
In so far as echinococcus disease of the liver is concerned, a review of 
the literature shows a general consensus of opinion that antibodies are 
present in the sera of the majority of diseased persons and animals and that 
these may be detected by means of a complement-fixation test. There is, 
however, a division of opinion in regard to the specificity of the reaction and 
its practical value in diagnosis. 
Much less work has been reported on complement fixation with sera of 
persons infested with such parasites as Tsenia saginata, Ascaris lumbricoides, 
etc. Miss Trist and myself3 have observed positive reactions with sera of 
dogs infested with various parasites and salt solution extracts of the re- 
spective worms. The method is worthy of trial in the diagnosis of infest- 
ments of persons with the various intestinal parasites. Positive reactions 
have also been reported by Jerlov4 and Fairley5 in human cases. 
The antigen is best prepared of the fresh fluid of an echinococcus cyst of 
man or sheep or by extracting the thoroughly ground-up parasites with 
saline solution for other forms of helminthiasis. It should be filtered, if 
necessary, preserved with 0.5 per cent, phenol, and kept constantly at a 
low temperature. It is highly important not to use the antigen in an anti- 
complementary dose, hence it should be titrated before each test is made. 
I have found that these antigens react positively with syphilitic sera; hence 
they should be freed of lipoids or used very cautiously if the patient is 
luetic. 
Sporotrichosis and Blastomycosis.—Cases of transmission of sporotri- 
chosis of horses to persons have been recorded, the slowly growing granulo- 
mata ending in degeneration being often mistaken for tuberculosis or syph- 
ilis; the disease is particularly prevalent among horses in the western states. 
Several strains of pathogenic sporotricha have been described, but are 
grouped by Meyer and Aird6 as one species, designated as Sporothrix 
schenckii-beurmanni. Moore and Davis7 have recently reported favorably 
upon the specificity and diagnostic value of agglutination and complement- 
fixation tests, and point out that while antibodies are readily produced in 
sporotrichosis, they are not generally at all or with difficulty in blastomycosis, 
which is a closely related disease. 
Basedow’s Disease.—Recently Berkeley8 has reported positive com- 
plement-fixation reactions with the serums of individuals with toxic thyroid 
disease and an antigen prepared of normal thyroid glands of dogs prepared 
1 Berl. klin. Wchnschr., 1912, 49, p. 1183. 
2 Policlinico, Rome, 1915, 22, 748, 1569. 
3 Jour. Infect. Dis., 1916, 18, 88. 
4 Ztschr. f. Immunitatsf., 1919, 28, 489. 
6 Med. Jour. Australia, 1921, 1, 205. 
6 Jour. Infect. Dis., 1915, 16, 399. 
7 Jour. Infect. Dis., 1918, 23, 252. 
8 Proc. New York Path. Soc., 1921, 21, 51. 560 
COMPLEMENT FIXATION IN BACTERIAL INFECTIONS 
according to a method described by Koopman.1 Berkeley believes that the 
reaction may prove of aid in differential diagnosis. 
Cancer.—None of the various complement-fixation methods that have 
been advocated from time to time in the diagnosis of cancer have proved of 
practical value, von Dungern2 has advocated a method which he claims 
has yielded 90 per cent, of positive reactions in known cases of cancer. 
Positive reactions have also occurred in tuberculosis and syphilis, and 
the reports of various other investigators are somewhat contradictory. 
Whitman, in a study of 30 cases, found the method highly satisfactory. 
Lucke3 has found the test of no value at all. 
The Complement-fixation Test in the Standardization of Immune 
Serums.—The technic of complement fixation has also been employed as 
one means in effecting the standardization of antimeningococcic and anti- 
gonococcic serums. Since, however, the amount of complement-fixing 
amboceptors in a serum is no index to its therapeutic and prophylactic value, 
a measure of this one factor is not a reliable standard. 
The technic consists in preparing the antigen and in determining its 
anticomplementary dose. Whatever this is, one-half to one-quarter this 
amount is added to increasing quantities of heated immune serum, ranging 
from 0.001 to 0.1 c.c. Complement and saline solution are added, and 
after incubating one hour at 370,C. the amount of complement fixation is 
determined by adding hemolytic amboceptors and corpuscles. 
While this titration is one measure of the reaction of the animal used in 
the immunization, better evidence of the therapeutic value of the serum 
is obtained by determining the content in bacteriotropins, by testing the 
serum with the antigen in susceptible animals, or by a combination of all 
methods. 
PROTEIN DIFFERENTIATION BY COMPLEMENT FIXATION 
The Determination of An Antigen by Complement Fixation.—In the 
tests hitherto considered the antigens were known, and the suspected anti- 
bodies sought for in the blood-serum or other body fluid. In making the 
reactions it was necessary to bring the serum to be tested into contact with 
the antigen specific for the suspected antibody, in the presence of com- 
plement, and at a suitable temperature. At the end of an hour the mix- 
ture was tested for free complement by adding hemolytic amboceptor and 
red blood-corpuscles. This order may be reversed, and with a known 
antibody the suspected antigen may be detected. The antigen to be de- 
tected, as in a solution of blood or bacterial extract, is brought into contact 
with its specific antibody in the presence of complement. At the end of an 
hour, at a suitable temperature, the mixture is tested as previously for free 
complement, by adding corpuscles and hemolytic amboceptors. Under 
proper conditions complement fixation would indicate a specific reaction 
between the antibody and its antigen, and thus serve to identify the latter. 
This method has clinical applications similar to those in which the pre- 
cipitin reaction is used: 
1. In the differentiation of blood-stains a solution of the stain con- 
stitutes the unknown antigen. By furnishing a known antiserum the anti- 
gen is detected, i. e., the animal from which the blood was derived is ascer- 
tained. 
2. In the recognition and differentiation of meats. 
1 Proc. New York Path. Soc., 1921, 21, 56. 
2 Munch, med. Wchn., 1912, 59, 65, 1098, 2854. 
3 Personal communication. COMPLEMENT FIXATION FOR IDENTIFICATION OF BLOOD-STAINS 
561 
3. In the detection of bacterial antigens in the blood-serum of patients, 
or with highly immune serums an unknown bacterial antigen may be iden- 
tified and the test employed as a means of differentiation among bacterial 
species. 
4. Similar applications of the test may be made in the differentiation of 
milks, seminal stains, and other albuminous substances. 
The complement-fixation test, however, shows biologic relationships in the 
same manner as the precipitin reaction, only to a finer degree, being a much more 
sensitive reaction. As in the precipitin reaction the complement-fixation 
reaction will show the species of animal from which an albuminous substance 
was derived, but does not differentiate among such products from animals of the 
same species. For example, the reaction cannot differentiate between an 
extract of human blood, human seminal stain, or human albuminous urine; 
it is necessary to identify the stain as blood or as semen or as urine by other 
means. The complement-fixation test will show that it is a human product. 
As compared with precipitin reactions, the complement-fixation test is 
probably more delicate and reliable and easier of interpretation. The 
technic of the latter method is, however, more complicated, and the liability 
to error is greater unless the principles of complement fixation in general 
are thoroughly understood and the importance of quantitative factors is 
appreciated. 
COMPLEMENT FIXATION FOR THE IDENTIFICATION OF BLOOD-STAINS 
The application of the technic of complement fixation to the determina- 
tion of specific protein antigen, such as human or animal blood, was 
demonstrated by Gengou in 1902. The principles worked out by him were 
extensively studied and practically applied by Neisser and Sachs1 in the 
forensic differentiation of animal proteins. 
Author’s First Method.—Hemolytic System.—Complement is furnised by 
the fresh serum of a guinea-pig diluted 1 : 20 and used in dose of 1 c.c. 
(= 0.05 c.c. serum); washed sheep's corpuscles are made up in a 2.5 per 
cent, suspension and used in dose of 1 c.c.; antisheep amboceptor should be 
highly potent, and is titrated after the method previously given (p. 473). 
In the following titrations and in conducting the main test the hemolytic 
amboceptor is used in an amount equal to 2 units. 
Specific Antiserum.—This is obtained from a rabbit immunized with 
the protein for which the test is to be made, namely, human or animal 
blood-serum. In forensic tests it may be necessary to prepare a number of 
these antiserums with the serums of man and the ordinary domestic animals. 
The technic of immunization is the same as that employed for the prepara- 
tion of precipitins (p. 323). An antiserum for forensic tests must be suffi- 
ciently potent to fix complement with 0.001 c.c. of its antigen. This is 
determined by a process of titration. If, for example, an antihuman serum 
is to be titrated, the method of procedure is as follows: 
Secure 0.1 c.c. of fresh human serum and dilute 1 : 100 by adding 9.9 
c.c. of normal saline solution. Of this dilution, 0.1 c.c. is equivalent to the 
standard dose of 0.001 c.c. of undiluted serum. The antiserum is heated 
from 60° to 62° C. for half an hour and diluted 1 : 20 (1 c.c. immune serum 
plus 19 c.c. of saline solution). Decreasing doses of immune serum are 
mixed with a constant dose of antigen and complement. At the same time 
the anticomplementary titration of the immune serum is made by sub- 
stituting salt solution for antigen. The doses to employ and the results of 
an actual titration are shown in the following table: 
1 Berl. klin. Wchn., 1906, xliii, 67. 562 
COMPLEMENT FIXATION IN BACTERIAL INFECTIONS 
Tube. 
Anti- 
serum, 
1 : 20, C.c. 
Serum 
Antigen, 
1 : 100, 
C.c. 
Comple- 
ment, 
1:20, C.c. 
.s 
a 
c3 
CJ 
a 
Anti- 
sheep 
Ambo- 
ceptor, 
Units 
Sheep’s 
Corpus- 
cles (2.S 
Per 
Cent.), 
C.c. 
Results After Incubation for 
One and One-half Hours. 
Inhibition of hemolysis. 
1.... 
1.0 
0.1 
1 
2 
1 
2.... 
0.9 
0.1 
1 
4-> CO 
2 
1 
Inhibition of hemolysis. 
3.... 
0.8 
0.1 
1 
2 
1 
Inhibition of hemolysis. 
4.... 
0.7 
0.1 
1 
tn b 
2 
1 
Inhibition of hemolysis. 
5.... 
0.6 
0.1 
1 
<5 5 
2 
2 
1 
Inhibition of hemolysis. 
6.... 
0.5 
0.1 
1 
2 
1 
Inhibition of hemolysis. 
7.... 
0.4 
0.1 
1 
s 
2 
1 
Inhibition of hemolysis. 
8.... 
0.3 
0.1 
1 
CJ o 
2 
1 
Inhibition of hemolysis; unit. 
9.... 
0.2 
0.1 
1 
° o 
CN *+-« 
2 
1 
Partial inhibition of hemol- 
tn 
ysis. 
10.... 
0.1 
0.1 
1 
2 
1 
Slight inhibition of hemolysis. 
11.... 
1.0 
0 
1 
2 
1 
Very slight inhibition of he- 
12.... 
.O 
molysis. 
0.8 
0 
1 
d 
2 
1 
Complete hemolysis. 
13.... 
0.4 
0 
1 
o 
2 
1 
Complete hemolysis. 
14.... 
0.2 
0 
1 
d> 
2 
1 
Complete hemolysis. 
15.... 
0 
0.1 
1 
.s 
2 
1 
Complete hemolysis. 
16.... 
0 
0 
1 
oj 
cn 
2 
1 
Complete hemolysis. 
Titration of an Immune Serum 
Tube 16 is the hemolytic system control, and shows complete hemol- 
ysis; tube 15 is the antigen control, and shows complete hemolysis, as the 
quantity of serum is too small to exert an anticomplementary influence; 
tubes 11 to 14 are the tests for anticomplementary action of the antiserum. 
In the present instance the serum was several months old and the maximum 
dose of 1 c.c. (= 0.05 c.c. undiluted serum) was very slightly anticomple- 
mentary. A fresh serum is practically never anticomplementary in this 
dosage, but these tubes should, nevertheless, be included in each titration. 
Tubes 1 to 10 include the antigenic titration, and show that the antiserum 
is perfectly antigenic in dose of 0.3 c.c. of this dilution (= 0.015 c.c. undiluted 
serum). In performing the main test double this quantity, or 0.6 c.c., would 
be used. 
Caution.—Unless an immune serum is perfectly antigenic in at least 
0.02 to 0.03 c.c. it should not be used in this test because 2 units must be 
used and amounts of rabbit serum greater than 0.06 to 0.08 c.c. may yield 
non-specific complement-fixation reactions. This is the greatest drawback 
to the use of the complement-fixation test for medicolegal purposes. I 
always test the sera of rabbits for non-specific complement fixation before 
immunization is begun. These tests are conducted with rabbit serum 
heated to 60° to 62° C. for one-half hour and antigen of fresh or heated 
human serum in dose of 0.1 c.c. of 1 : 100 dilution. If undiluted rabbit 
serum in dose of 0.2 or 0.1 c.c. yields a positive reaction this animal is not 
used for the preparation of an immune serum. 
Each antiserum is tested in a similar manner. In forensic blood tests 
an antihuman serum is, of course, employed first; if this is negative and|it 
is desirable to determine the source of the blood, other antiserums, as that of 
the ox, horse, dog, etc., are prepared, titrated, and tested with a solution 
of the blood-stain. 
The Blood-stain.—It is first necessary to ascertain that the stain is of 
blood; this is done by performing the hemin crystal or an oxydase test 
(p. 321). The stain is then extracted in normal saline solution, as COMPLEMENT-FIXATION METHOD FOR IDENTIFICATION OF MEATS 
563 
described on p. 321. A 1 : 1000 dilution is made approximately by so 
diluting the extract that it just gives a slight opalescence when boiled with 
a few drops of acetic acid, and a slight foam persists after shaking. Unless 
it is perfectly clear, it should be filtered. 
The Test.—Into a series of six small test-tubes place increasing doses 
of extract of the blood-stain (antigen), as follows: 0.1, 0.2, 0.4, 0.6, 0.8. 
and 1 c.c.; add double the titrated dose of antiserum and 1 c.c. of 
complement (1 : 20), with sufficient salt solution to bring the total volume 
in each tube up to 3 c.c. 
The following controls are included: 
1. Antigen control: 1.0 c.c. of the blood extract plus 1 c.c. of diluted com- 
plement and salt solution. 
2. Antiserum control: double the titrated dose plus 1 c.c. of diluted 
complement and salt solution. 
3. Hemolytic control: at this time 1 c.c. of diluted complement and salt 
solution. 
4. Corpuscle control: 1 c.c. of corpuscle suspension and salt solution. 
The tube should be plugged with 'cotton. 
Shake all the tubes gently and incubate for one hour at 37° C. in a water- 
bath. Add 2 units of hemolytic amboceptor and 1 c.c. of corpuscle suspen- 
sion to each tube except the corpuscle control. Shake gently and reincubate 
for from one to two hours in a water-bath, depending upon the degree of 
hemolysis present in the controls. 
The readings are made at once, and again after the tubes have been 
allowed to settle in the refrigerator overnight. Inhibition of hemolysis with 
the smallest dose of blood extract—0.1 c.c. (= approximately 0.0001 c.c. of 
blood)—indicates that the blood extract is most certainly the antigen for 
the antiserum employed. Even with the maximum dose of extract—1 c.c. 
(= approximately 0.001 c.c. of blood)—inhibition of hemolysis serves to 
show the nature of the blood. With an antihuman serum, for instance, a 
similar specific reaction would be possible only with bloods of the higher apes. 
In making blood tests for medicolegal purposes the antiserum should 
not only be standardized with a definite dilution of human serum, but 
the whole test should first be conducted with a known dried human blood- 
stain, and it must be borne in mind that extreme accuracy in all manipula- 
tions is essential. 
I prefer this complement-fixation test to the precipitin reaction in the 
differentiation of proteins, as the readings are sharper and more definite. 
This test is fully as reliable as the precipitin test, and there is less danger of 
group reaction. 
COMPLEMENT-FIXATION METHOD FOR THE IDENTIFICATION OF MEATS 
The technic is essentially similar to that used in the foregoing test. Anti- 
serums are prepared by immunizing rabbits with the serums of various ani- 
mals, as the ox, horse, dog, cat, or any other animal the prseence of whose flesh 
is to be identified in sausages, bologna, etc. It is not necessary to immunize 
with an extract of these meats themselves, as the blood or blood-serums 
will suffice. The technic of immunization is the same as that employed in 
the preparation of precipitin serums. Each antiserum is titrated with its 
antigen, as previously described, and is used in double the titrated dose in 
conducting the main test. 
An extract of the flesh to be examined is prepared as described on p. 329. 
The test is then conducted exactly the same as previously described. 564 
COMPLEMENT FIXATION IN BACTERIAL INFECTIONS 
In the following table are shown the method and the results of an actual 
test, using a dried human blood-stain and the same antiserum as previously 
directed. 
Forensic Blood Test 
Tube. 
Extract 
of 
Blood- 
stain, 
1 : 1000 
C.c. 
Anti- 
serum 
1 : 20, C.c. 
Comple- 
ment, 
1 : 20, C.c. 
"d 
-M 
Cj 
.a 
d 
V 
.5 
Anti- 
sheep 
Ambo- 
ceptor. 
Units. 
Corpus- 
cles (2.5 
Per 
Cent.), 
C.c. 
Results After One and One- 
Half Hours’ Incubation. 
1.... 
0.1 
0.6 
1 
'd 
d 
c3 
2 
1 
Marked inhibition of hemol- 
2.... 
0.2 
0.6 
1 
d 
M 
2 
1 
ysis. 
Marked inhibition of hemol- 
3.... 
0.4 
0.6 
1 
rd 
cn . 
t/j 
2 
1 
ysis. 
Complete inhibition of hemol- 
4.... 
0.6 
0.6 
1 
n o 
a; 
2 
1 
ysis. 
Complete inhibition of hemol- 
5.... 
0.8 
0.6 
1 
u o 
<J 
2 
1 
ysis. 
Complete inhibition of hemol- 
6.... 
1.0 
0.6 
1 
o 
2 
1 
ysis. 
Complete inhibition of hemol- 
7.... 
1.0 
0 
1 
.2 
*d 
d 
2 
1 
ysis. 
Antigen control: hemolysis. 
8.... 
0 
0.6 
1 
"^o 
2 
1 
Antiserum control: hemolysis. 
9.... 
0 
0 
1 
2 
1 
Hemolytic control: hemolysis. 
10.... 
0 
0 
0 
0 
1 
Corpuscle control: no hemol- 
zn 
ysis. 
COMPLEMENT-FIXATION METHOD FOR THE IDENTIFICATION OF 
BACTERIAL ANTIGENS 
As a means of diagnosis this test has very limited practical value. It 
aims to detect, by means of complement fixation with a known antiserum, 
a soluble bacterial antigen in the blood-serum of a patient. For example, in 
typhoid fever the patient’s serum is mixed with a potent antityphoid serum 
in the presence of complement. After incubating for one hour at 37° C., 
amboceptor and corpuscles are added to test for free complement. An ab- 
sence of hemolysis indicates that complement has been fixed by the anti- 
serum and soluble typhoid antigen in the serum of the patient. As a rule, 
and for purposes of diagnosis, this order of procedure is reversed: the antigen 
is furnished and then sought for in the patient’s serum, as in the gonococcus 
fixation test, syphilis reaction, etc. 
An immune serum is prepared by immunizing rabbits with increasing 
doses of an emulsion of the bacteria wdiich we wish to test for in the patient’s 
serum. 
The bacterial extract is prepared as follows: Make cultures of the 
bacteria on slants of agar; wash off a sufficient number with normal saline 
solution until 20 or 30 c.c. of a heavy emulsion are secured; add 0.4 per cent, 
of phenol, and shake mochanically with glass beads for twenty-four hours; 
then heat to 60° C. for four hours, and either centrifuge thoroughly or filter 
through a Berkefeld filter. The clear filtrate should be preserved in a tightly 
stoppered bottle in an ice-chest. It is well to titrate this extract for its 
anticomplementary dose. As a rule, these extracts are free from anti- 
complementary action until relatively large doses are employed. 
The antiserum is heated to 60° C. for half an hour and titrated with 
0.01 c.c. of the bacterial extract (0.1 c.c. of a 1 : 10 dilution) in ten doses, 
ranging from 0.1 to 1.0 c.c. Double the dose giving complete fixation of 
complement is used in testing for the bacterial antigen in human serum. AUTHOR’S NEW COMPLEMENT-FIXATION METHOD 
565 
In conducting this test the patient’s serum is heated to 55° C. for half 
an hour, and decreasing doses, ranging from 0.5 to 0.01 c.c., are placed 
in a series of test-tubes together with double the titrated dose of antiserum. 
Complement and salt solution are now added, and after incubating for ane 
hour at 37° C., amboceptor and corpuscles are added and the tubes rein- 
cubated. The general technic and controls are the same as those previ- 
ously described. 
The test has some value in special research work, but for practical use 
it has given way to the agglutination reactions and complement-fixation tests 
for the detection of antibody with a known antigen. 
Author’s New Complement-fixation Method for the Identification 
of Blood-stains, Bacteria, and Other Proteins 
The complement-fixation method based upon studies in the standardiza- 
tion of the Wassermann reaction has been applied to the differentiation of 
proteins as the identification of blood-stains, with considerable success. 
The method has proved much more sensitive than the precipitin test and 
superior to the complement-fixation test described above. 
Great care, however, must be exercised in the choice of rabbits for the 
preparation of immune sera because the test is so sensitive that the majority 
of normal rabbit sera yields non-specific reactions unless amounts smaller 
than 0.02 c.c. are employed. For this reason I have worked out a method 
for preparing immune sera by the immunization of guinea-pigs. 
The technic embraces the same principles as described above. Owing 
to lack of space the details based upon extensive investigations covering the 
selection and immunization of rabbits and guinea-pigs, preservation and 
titration of immune sera, preparation and titration of antigens (sera, meats, 
seminal fluid, etc.) are to be given in the author’s separate monograph upon 
the complement-fixation reaction. CHAPTER XXVI 
CYTOTOXINS 
In Chapter XVIII the cytolysins in general were considered, especially 
their theoretic structure and the mechanism of their action. 
It will be remembered that the general name cytolysin is applied to 
an amboceptor or antibody of the third order of receptors that is capable 
of preparing its antigen for the disintegrative or lytic action of a comple- 
ment. The two best known and most important members of this group of 
antibodies have been considered, namely, the hemolysins and the bacterio- 
lysins. 
Following the discovery of the hemolysins and the bacteriolysins and of 
the mechanism of their action, it was not long before similar studies were 
undertaken with other cells, with the result that attempts have been made to 
prepare immune cytolytic serums for practically every organ of the body. 
This outcome was but natural, in view of the enormous theoretic importance 
of specific cytolysins, not only from the additional light that may be thrown 
upon physiologic and pathologic processes in general but also from the 
standpoint of specific therapeutics. 
Nomenclature.—While actual lysis or solution of erythrocytes and bac- 
teria may be brought about by antibodies of this order, yet actual solution 
is not apparent with most other body cells, although a distinct toxic action 
may be observed. For instance, an antispermatozoa serum will cause these 
cells to lose their motility, but does not actually dissolve them. Hence the 
name cytotoxin has been applied to these immune serums. This is probably 
a better term than cytolysin; but it is to be remembered that, so far as is 
now known, both cytolysins and cytotoxins are antibodies that possess the 
same nature and structure, except that in the former group the process is 
complete and ends in actual lysis of the cell. The term “cytolysin” is, there- 
fore, more appropriately applied to the bacteriolysin and hemolysins; whereas 
the term cytotoxin is reserved for those immune serums that injure their cells 
without complete lysis (a toxic action), such as nephrotoxin, hepatotoxin, etc. 
This chapter is mainly concerned with the latter group. 
Nature and General Properties of Cytotoxins.—As previously stated, 
cytotoxins are amboceptors or antibodies of the third order, and possess the 
same general properties as the bacteriolysins and hemolysins, except that, 
as will be pointed out later, they do not possess the same specificity. With- 
out the presence of a complement they are inactive. They are thermostabile, 
possess the same general affinity for their antigen, and may be removed from 
a serum by saturation with the antigen in a manner similar to that used for 
the removal of a hemolysin. 
Natural Cytotoxins.—Human serum as well as the sera of- the lower ani- 
mals may contain small amounts of cytotoxins for a wide variety of cells. 
Flexner and Noguchi1 have found cytotoxins in the sera of warm and cold- 
blooded invertebrates for cells of the kidney, liver, and testes. By absorption 
tests these were found to possess a well-defined, but not an absolute speci- 
ficity. 
Preparation of Cytotoxins.—Cytotoxic serums are prepared by im- 
munizing an alien animal with fine suspensions of cells of the particular 
1 Jour. Med. Research, 1903, 9, 257. 
566 METHODS OF STUDYING CYTOTOXINS 
567 
organ being studied. Every effort should be made to remove all traces 
of blood, and to secure as pure an emulsion of the same cells and to work 
as aseptically as possible. Injections are best given intraperitoneally, rabbits 
being well adapted for the preparation of these serums. 
Beebe1 has made the statement that more specific serums are obtained 
if the nucleoproteins are isolated and used in the process of immunization, 
than if the cells themselves are used. Wells,2 however, believes that the 
nucleoproteins of cells are not specific in character, and Pearce, Karsner, 
and Eisenbrey3 found that nephrotoxic and hepatoxic serums prepared by 
the injection of the nucleoproteins of these cells were no more toxic than 
serums prepared from the globulins and albumins of the same organs. 
Methods of Studying Cytotoxins.—While the phenomena of hemolysis 
and bacteriolysis may readily be observed in experiments in vitro, the in- 
fluence of other cytotoxins on the particular cells used as antigens is more 
difficult to determine. 
Microscopic Method.—Suspensions of cells may be obtained by passing 
the tissue through a meat grinder and shaking the pulp very vigorously in 
a large test-tube with sterile saline solution. On standing, the emulsion 
separates into three layers: a lower layer of large fragments, a medium of 
separated cells, and an upper in which fragments of cells and blood-corpuscles 
are contained. The middle layer is removed, again shaken with saline solu- 
tion, allowed to stand and the supernatant opaque suspension removed, 
filtered through gauze, and employed. 
The serum must be fresh and should be used undiluted; 1 part of sus- 
pension of cells may be mixed with 9 parts of serum and incubated at 38° 
C. for two hours followed by placing in a refrigerator for several hours. 
The effects of cytolysis may be apparent by partial clearing and by micro- 
scopic examination to detect evidences of cytolysis. Controls in which 
physiologic saline solution replaces serum should always be included. 
Animal Inoculation Method.—The technic generally employed consists 
in making subcutaneous, intraperitoneal, or intravenous injections of the 
immune serum into the animal, or into the arteries leading to particular 
organs. Functional disturbances and delicate histologic changes in various 
organs have served as criteria for determining the degree of specificity that 
exists. The loss of some manifestation of vitality on the part of the cell, 
as a loss of motility (spermatozoa) or an inability to proliferate, may aid in 
studying the effect of these serums. 
Complement-fixation and Other Methods.—More recently several other 
methods have been used, especially the complement-fixation and the epi- 
phanin reactions; Fleischer, Hall, and Arnstein4 have employed a comple- 
ment-fixation reaction with success. Rabbits were immunized with the 
liver and kidney of the guinea-pig and these sera employed with antigens 
of the guinea-pig organs prepared by different methods. By means of these 
reactions and absorption tests they were able to show that there exists a 
definite relationship between the antiorgan sera and the homologous antigens. 
Lambert5 has suggested that a method of cultivation of tissues outside 
of the body may constitute an ideal technic. He has observed that mouse 
sarcoma, which grows vigorously in the plasma of normal rats, shows little 
or no activity in the plasma of rats immunized by mouse sarcoma injec- 
1 Jour. Exper. Med., 1905, vii, 733. 
2 Chemical Pathology, 1914, second edition, W. B. Saunders Co. 
3 Jour. Exper. Med., 1911, xiv, 44. 
4 Jour. Immunology, 1920, 5, 437; ibid., 1921, 6, 223. 
5 Jour. Exper. Med., 1911, 14, 453. 568 
CYTOTOXINS 
tions. Rat sarcoma, readily cultivated in the plasma of normal guinea- 
pigs, remained inactive or presented a feeble growth in the plasma of guinea- 
pigs previously treated with rat tissues. Lambert has attributed these 
results to the action of cytotoxins. 
Specificity of Cytotoxins.—As has been stated elsewhere, the hemolysins 
and the bacteriolysins are highly specific, especially the former group. With 
the cytotoxins, however, this specificity is not observed. Most cytotoxic 
serums are also hemolytic, notwithstanding the fact that careful precau- 
tions have been taken to remove, so far as possible, all traces of blood from 
the inoculum during the process of immunization. Metchnikoff found a 
spermatotoxic serum to be also hemolytic, but he believed that this property 
could be removed by treating the immune serum with the corresponding 
corpuscles, and in this manner dissolve out the hemolysin. Numerous other 
investigators have found, however, that cytotoxic serums may attack the 
cells of other organs, as well as those that have been used as their antigens. 
The subject has been very carefully investigated by Pearce.1 The 
injection of an antidog nephrotoxic serum prepared by immunizing rab- 
bits with washed dog kidney is followed by the development of a tubular 
nephritis, with albuminuria and occasionally hemoglobinuria, and accom- 
panied by granular degeneration of the liver. These serums are usually 
hemolytic in vitro. Similarly, in a study of hepatotoxic serums, Pearce 
found that the most striking lesions were referable to the hemagglutinating 
and hemolytic properties of the serum, causing thrombosis, embolism, and 
hemorrhages, whereas secondary necroses may be caused by a direct toxic 
action of the serum on certain parenchymatous cells. 
Pearce, Karsner, and Eisenbrey found that the serums of rabbits in- 
jected repeatedly with the nucleoproteins, globulins, and albumins of the 
liver and kidney of the dog gave no evidence of organ specificity in vitro 
or in vivo experiments. These investigators were not able to support the 
view put forward that nucleoproteins play an important part in the pro- 
duction of cytotoxic immune serums. 
Lambert2 has recently studied the subject with cultures of rat sarcoma 
and rat embryo skin and their immune serums, and found that these cyto- 
toxins were not specific for the tissue injected. 
These results are not surprising when it is remembered that all the 
body cells have a common origin, and that, although the cells of various 
organs may differ considerably in morphologic and functional characters, 
they have certain receptors in common, and, as Pearce originally main- 
tained, it is hardly conceivable that specific somatogenic cytotoxins can 
be produced. 
That some degree of specificity may exist is, however, to be considered 
as a possibility. By means of the complement-fixation reaction Fleischer 
and Arnstein3 have recently observed that it is possible to demonstrate some 
tissue specificity in liver, kidney, spleen, brain, muscle, and testicle, their 
experiments being conducted by injecting guinea-pig tissues into rabbits. 
In some cases they observed absolute specificity, while in others specificity 
was only relative. The difficulty in demonstrating specificity was found in 
the extreme complexity of the biologic composition of the tissues, and pos- 
sibly in the interrelationship existing between various tissues. 
Autocytotoxins.—While these toxins possess some theoretic interest, 
they are of very rare occurrence in experimental work. Certainly cells of 
1 Jour. Exper. Med., 1914, xix, 277 (includes a good review of the literature). 
2 Univ. Penna. Med. Bull.. 1903, xvi, 217; Jour. Med. Research, 1904, xii, 1. 
3 Jour. Immunology,. 1921, 6, 223. VARIETIES OF CYTOTOXINS 
569 
the kidney, liver, and other organs are constantly dying and being replaced 
by new cells; the receptors of these cells are thereby set free, and are capable 
of forming a union with the receptors of other cells, and the possibility 
for the formation of autocytotoxins is established. According to Ehrlich, 
however, the side arms anchoring the receptors of the dead cells are sessile 
in nature, and are unlikely to cause overproduction, as in the case of anti- 
bodies for bacterial substances or for the cells of other species. On the 
other hand, a simultaneous production of anti-autotoxins that counteract 
the autotoxins and preserve a delicate physiologic equilibrium may occur, 
the whole subject being, however, still in the experimental stage. 
Of most interest in this connection are the theoretic autonephrotoxins. 
These may be produced when part of a kidney becomes disorganized in the 
living body, as by means of a toxin. Theoretic autotoxins may then be 
produced, which, acting upon other kidney cells, institute a vicious cycle. 
Acting upon these assumptions, Ascoli and Figari1 and Lindeman2 have 
proposed a new theory as to the pathogenesis of certain of the nephritides. 
These observers would account for the cardiac hypertrophy of nephritis by 
attributing it to the action of the nephrotoxic serum in causing contraction 
of the peripheral vessels, with consequent increase of blood-pressure; the 
nephritic nervous symptoms, they believe, are due to the fact that the 
serum contains a neurotoxic constituent. 
Lindeman has produced a toxic nephritis in dogs by giving them injec- 
tions of potassium bichromate, and found that the serum, although free 
from the chromate, was toxic to other dogs, a finding he believed due to the 
presence of autonephrotoxins produced in the first dog as a result of the 
destruction of kidney cells. According to this view, the original toxic cause 
of a degenerative nephritis would be less responsible for the continuance of 
the process than would the formation of an autonephrotoxin. While these 
conclusions are somewhat far-reaching, they serve to indicate that the same 
processes operative in bacterial infection and immunity may have an impor- 
tant relation to other pathologic conditions. 
It is not rarely observed that in large tumors and similar lesions certain 
groups of cells may undergo digestion, but in these the lysis is commonly 
ascribed to the action of ferments liberated upon the death of cells. 
Isocytotoxins have been produced experimentally, as, for example, by 
Ehrlich, who produced isohemolysins by injecting goats with goat blood, 
and by Metchnikoff, who prepared isospermatoxic serums. 
Anticytotoxic serums have likewise been prepared by careful immuniza- 
tion with cytotoxic serums. They were first discovered by Camus and 
Gley3 and independently of them, by Kossel4 in connection with a study of 
the toxic power of blood-serum of eels. Animals treated with eel’s serum 
acquire an antitoxic property which protects their corpuscles against the 
hemolytic action of ichthyotoxin, or the toxic substance of- the blood of eels. 
Varieties of Cytotoxins.5—As previously stated, attempts have been 
made to prepare cytotoxic serums for practically all the organs and tissues. 
Since none of these has been found to be absolutely specific, and hence since 
they possess little or no practical value, they will receive here but brief 
consideration. 
1. Spermatotoxin.—This serum was prepared simultaneously by Metch- 
1 Berl. klin. Wchn., 1902, xxxix, 634. 
2 Ann. de l’Inst. Pasteur, 1900, xiv, 49. 
3 Archiv. internat. d. Pharmacol., 1898, 3 and 4. 
4 Berl. klin. Wchn., 1898, 152. 
6 For literature see Sachs, Biochem.-Centralbl., 1903, 1, 573, 613, 653, 693. 570 
CYTOTOXINS 
nikoff1 and Landsteiner in 1899, and was one of the earliest cytotoxins 
to be studied. It is a hemolytic serum, and causes spermatozoa to lose 
their motility. As shown by Metchnikoff2 the preparation of these sera is 
frequently difficult. Furthermore, their full activity was only apparent 
after the sera had been heated to 56° C. It would seem to affect also the 
vitality of the spermatozoa in vivo, inasmuch as De Lester, by the injection 
of this serum, rendered male mice sterile for from sixteen to twenty days. 
Dittler3 has recently renewed interest in this subject by finding that 
immunization of female rabbits with rabbit sperm renders them sterile for 
a few months. Immunization with sperm from other species failed to pro- 
duce sterility, and as the serum of the rabbits injected with rabbit sperm 
was spermatotoxic and agglutinative, Dittler believes that the temporary 
sterilization depends on immunization against the sperm of the homologous 
species. These and similar observations have aroused considerable specula- 
tion regarding their possible relation to sterility. 
2. Epitheliotoxin.—A cytotoxic serum for the ciliated epithelium of 
the trachea was prepared by von Dungern.4 The cells became disintegrated 
in the peritoneal cavity of the immunized animal, but not in that of the 
normal animal. This serum also proved to be hemolytic. 
Similar serums have been prepared with cancer cells, in the hope of estab- 
lishing a specific serum therapy, but all efforts have thus far proved futile. 
3. Leukotoxins.—This serum was first prepared by Metchnikoff5 and 
Besredka6 by injecting the spleen of rats into guinea-pigs. The serums 
have also been prepared by effecting immunization with exudates rich in 
leukocytes or with the emulsion of lymphoid organs (Flexner and Ricketts). 
They are usually hemolytic, and also attack endothelial cells. Their action 
may be observed in vitro when the leukocytes lose their ameboid motility 
and the protoplasm swells, clears, and may disintegrate, leaving the nucleus. 
Lindstroem7 has recently prepared a serum by injecting a sheep intra- 
venously with rabbit leukocytes and reports that the administration of this 
immune to 3 cases of myeloid leukemia had a profound influence upon the 
leukocytes and produced temporary benefit. 
4. Nephrotoxin.—This serum is best adapted for experimental studies 
of the cytotoxins. As shown by Pearce, with the aid of this serum physio- 
logic and anatomic alterations and lesions are readily studied. The injec- 
tion of a nephrotoxic serum produces a tubular nephritis, with albuminuria 
and possibly hemoglobinuria. The serums are usually hemolytic, and 
frequently cause degenerative lesions in the liver, due in part to hemolysis 
and hemagglutination of red corpuscles. 
5. Hepatotoxin.—Delezenne8 was the first to work with the so-called 
hepatotoxin, which, he claimed, possessed absolute organ specificity. Sub- 
sequent investigations, however, have brought forth contradictory findings. 
Pearce found that hyaline thrombi, formed of agglutinated red corpuscles, 
are primarily responsible for the areas of necrosis and hemorrhage, with 
secondary effects, which may be ascribed to a cytotoxin liberated chiefly 
by the cells in the thrombus, and acting on the liver cells. These findings 
and views have recently received support from the investigations of Karsner 
and Aub.9 
6. Gastrotoxin.—Gastrotoxic serums have been studied by Bolton,10 who 
1 Ann. de l’Inst. Pasteur, 1900, xiv, 369. 
9 Ibid., 1900, 14, 5. 
s Munch, med. Wchn., 1920, 67, 1495. 
4 Mtinch. med. Wchn., 1899, xlvi, 1228. 
5 Ann. de l’Inst. Pasteur, 1899, xiii, 737. 
6 Ann. del ’Inst. Pasteur, 1900, xiv, 402. 
7 Arch. d. mal. du coeur, 1921, 14, 145. 
3 Semaine med., 1900, xx, 290. 
9 Jour. Med. Research, 1913, xxviii, 377. 
10 Proc. Roy. Soc., lxxvii, 426, and lxxix, 533. VARIETIES OF CYTOTOXINS 
571 
immunized rabbits with emulsions of the mucosa of the stomach of the guinea- 
pig. The injection of this serum into guinea-pigs was followed by the 
development of areas of hemorrhage, necrosis, and ulcer formation that 
resembled peptic ulcers. According to Bolton, if the gastric secretions were 
neutralized with large quantities of alkali, the ulcers did not develop, indi- 
cating that the peptic ferments may be operative in the digestion of the 
cells after their destruction by the immune serum. Gastrotoxic serums were 
found to produce precipitates with clear filtrates of gastric cells, and were 
also shown to be hemolytic. 
7. Synocytotoxin.—This serum has been produced experimentally by 
immunization with an emulsion of placental cells. According to Liepmann,1 
it produces a precipitate with a filtrate of placenta cells, and at one time it 
was believed that it might constitute a diagnostic test for pregnancy. 
Syncytotoxins are interesting as considered in reference to eclampsia 
and other toxemias of pregnancy. As is well known, placental cells may 
become detached and, gaining entrance to the circulation, become lodged 
in remote organs (Schmorl). This has given rise to the theory that a pla- 
centotoxin is developed that produces the nephritis of pregnancy and 
necrotic lesions in the liver. Weichardt asserts that, by digesting placenta 
in vitro with an active placentotoxic serum and injecting the digestate, he 
produced symptoms resembling eclampsia in the lower animals. It was 
hoped that an anticytotoxic serum might be prepared to combat the effects 
of the placentotoxin, but this hope has not been realized. Renewed interest 
in this particular subject has been manifested by the recent studies of Abder- 
halden in ferments as applied to the diagnosis of pregnancy (p. 231). 
8. Neurotoxin.—This toxin has been prepared and studied by Dele- 
zenne,2 Centanni, Delille, and others by immunization experiments with 
emulsions of cerebrum, cerebellum, and spinal cord. When injected into 
the brain direct these serums may cause profound intoxication of the nerve- 
centers, with torpor or convulsions, subnormal temperature, and death. 
When injected directly into the veins they are usually without effect. In 
addition to their neurotoxic action they are generally hemolytic, and fre- 
quently endotheliotoxic and leukotoxic. 
9. Thyrotoxins.—This serum is prepared by immunizing animals with 
emulsions of thyroid gland. Thyrotoxins were quite prominently before 
the profession a few years ago, owing to the work of Beebe,3 who advocated 
their use in the treatment of various goiters. They have not fulfilled their 
expectations, however, since they may also produce degenerative changes 
in the various organs, as the liver, spleen, and kidneys. 
Adrenotoxins.—In 1901 Bigart and Bernard4 found that the serum of 
ducks which had been injected with the adrenals of guinea-pigs was capable 
of killing these animals and that the medullary portion of the adrenals was 
sometimes diffluent and gelatinous. Abbott,5 Pearce,6 and others, however, 
have failed to produce adrenotoxic sera by the immunization of rabbits. 
Ritchie7 has reported the successful preparation of a serum by immuniza- 
tion of ducks, which was adrenophilic, but not adrenolytic. This serum 
was not strictly specific, but there was no evidence of it being hemolytic. 
Thymotoxins.—Ritchie8 has immunized ducks with suspensions of cells 
1 Deutsch. med. Wchn., 1902, xxviii, 911. 4 Compt. rend. Soc. de biol., 1901, liii, 161. 
2 Ann. de l’lnst. Pasteur, 1900, 14, 686. 5 Jour. med. Res., 1903, 49, 329. 
3 Jour. Exper. Med., 1905, 7, 732. 6 Jour. med. Res., 1904, 12, 1. 
7 Jour. Path, and Bacteriol., 1908, 12, 140 (gives a good bibliography of the literature on 
the cytotoxins in general). 
8 Jour. Path, and Bact., 1908, 12, 140. 572 
CYTOTOXINS 
from the thymus glands of guinea-pigs. These sera were leukophilic, but 
not hemolytic. Structural changes were produced by injecting these sera 
into guinea-pigs, but these were ascribed to its leukolytic action and were 
regarded as non-specific. . ■ _ . , 
10. Infusoriotoxins.—Rossle1 and more particularly Takenouchi- ha\ e 
described toxins for various infusoria in normal and immune sera. 1 ake- 
nouchi has studied human, horse, sheep, hog, beef, guinea-pig, rabbit, pigeon, 
frog and turtle sera with paramecia finding the evidences of toxic action 
expressed by depression of motility, discharge of trichocysts, swelling, and 
finally disintegration. Toxic activity was found to disappear when serum 
was heated. Curiously, the sera of cold-blooded animals were found much 
more toxic than the sera of warm-blooded animals. 
ROLE OF CYTOTOXINS IN IMMUNITY 
It is apparent that, according to our present knowedge, the cytotoxins 
proper, although they possess great theoretic importance from their pos- 
sible relationship to the removal and disposal of enfeebled and dead cells, 
occupy a subsidiary place in the processes of immunity. The processes 
governing these changes are finally and delicately balanced, and although 
obscure, they offer an intricate but fascinating field for research. . 
If the ferments concerned in Abderhalden’s studies are related in an\ 
way to the cytolysins, the subject becomes of great interest, and a new field, 
with immense possibilities, is opened for further study. 
Practical Applications—(1) In therapeutics the cytotoxins have not 
established a place for themselves and their use has been disappointing. 
As previously stated, the use of thyrotoxic serums has not met with con- 
siderable success; epitheliotoxic serums have likewise not been efficient in 
the treatment of cancer. They possess theoretic interest, however, from the 
possibility of their so injuring glands that their functions may be studied; 
theoretically, the use of minute and carefully graded doses of hemolytic 
serums may effect the production of antihemolysins and aid in the treatment 
of certain anemias. , , „ , 
(2) In diagnosis cytotoxic reactions have been employed by I reunci 
and Kaminer. These observers used the cytotoxins as a diagnostic aid 
in cancer, but with indifferent success. 
CYTOTOXIC REACTIONS 
Cytolytic Cancer Diagnosis of Freund and Kaminer.3—This reaction is 
based upon the observation that while normal serum has the power to dis- 
solve cancer cells, the serum of cancerous persons lacks this property, and 
has the power to inhibit the destruction of such cells by normal serum. 
The same authors have also observed that when cancer serum is nuxe 
with an extract of cancer cells a precipitate forms. They claim to have 
secured 88 per cent, of positive reactions in 113 cases examined, and believe 
that the reaction occurs early enough and is sufficiently specific to render 
it of practical value. These observations, however, have not been suffi- 
ciently confirmed. . , , , 
An emulsion of cancer cells is prepared by grinding the undegenerated 
portions of a tumor, freed as much as possible from fat and fibrous tissue, 
in a mortar and adding about five volumes of 1 per cent, sodium biphos- 
1 Arch. f. Hyg., 1905, 54, 1. 
3 1910, 26,^12; Wien. klin. Wchn., 1910, 23, 378, 1221; ibid., 1911, 
24, 1759. PRODUCTION OF CYTOTOXINS 
573 
phate. The suspension is filtered through several layers of gauze, and 
after the cells have become precipitated, the supernatant fluid is decanted. 
The residue of cells is washed with 0.6 per cent, sodium chlorid and allowed 
to settle again, the supernatant fluid is decanted, and the residue covered 
with 1 per cent, sodium fluorid. The last-named fluid must first be neu- 
tralized against alizarin until only a trace of the violet color remains. The 
emulsion will keep for several weeks in an ice-chest. 
Serum.—The patient’s serum should be collected just a few hours (not 
over twenty-four) before the test is to be made, and must be clear and free 
from cellular elements. 
The Test.—To 10 drops of the patient’s serum add 1 drop of 0.5 per 
cent, solution of sodium fluorid. Then add 1 drop of the cancer-cell emulsion 
so diluted that wdien 1 drop of the mixture is placed in a blood-counting 
chamber, about 10 to 20 cancer cells will be found in a field of four large 
squares. Close the counting chamber carefully, ring with vaselin to prevent 
evaporation, and place in the incubator for twenty-four hours. 
A second slide is prepared from a mixture composed of one volume each 
of normal serum, cancer serum, and 0.6 per cent, sodium chlorid and suffi- 
cient cell emulsion. 
A third slide is prepared with a fresh normal serum in the same manner 
as when the patient’s serum is used. 
All slides are incubated for twenty-four hours at 37° C. and the cells 
counted. A material reduction in the number of cells with the normal 
serum will be noted; if the patient has carcinoma, the first and second slides 
will not show this reduction, whereas if the patient is free from cancer, 
similar reductions will be found in all three slides. 
The authors recommend that both the cytolytic and the precipitin test 
be conducted when enough serum is available for both. Herly1 has recently 
applied this test to a study of the serums of normal rats and rats inoculated 
with the Flexner-Jobbing rat carcinoma. Normal rat serum was found 
without any deleterious effects upon the cells of this tumor; likewise the 
serums of tumor rats did not show any effects. 
PRODUCTION OF CYTOTOXINS 
The term “cytotoxins” is usually applied to cell toxins other than he- 
molysins, such as nephrotoxins, spermatotoxins, etc., and although “lysin” 
is frequently used, the term “toxin” is better, being descriptive of the 
changes produced by all cell toxins except the hemolysins and the bac- 
teriolysins. 
Cytotoxic serums can be made theoretically for any cell, but only the 
hemolysins possess much practical value. The cytotoxins are prepared with 
some difficulty by injecting emulsions of cells from one animal into another. 
Immunization should always be conducted by means of intraperitoneal or 
subcutaneous injection unless the material is so finely divided that intra- 
venous injection is a safe procedure. 
Antispermatozoa serum is prepared by injecting rabbits intravenously 
with 1 c.c. semen diluted with 4 c.c. sterile saline solution every five days; 
after five or six injections the rabbit-serum is usually found to contain large 
amounts of precipitin, complement-fixing and cytolytic amboceptors, suit- 
able for the detection of seminal stains. The animals should be bled about 
ten days after the last injection. 
For the purpose of studying the action of cytotoxic serum, nephrotoxic 
1 Jour. Cancer Research, 1921, 6, 337. 574 
CYTOTOXINS 
serum is preferably to be used, as the effects, e. g., the production of albu- 
minuria, may be observed. A series of two or three animals should be car- 
ried along at the same time, as many die after the third injection. So far 
as possible an aseptic technic should be carried out. 
Practically all cytotoxic serums are hemolytic partly because of the 
great difficulty of removing all traces of blood from the organ used in pre- 
paring the emulsion. This difficulty may be reduced to a minimum by 
thoroughly washing the organ or tissue prior to preparing an emulsion for 
injection. 
The following method, after Pearce, illustrates the mode of preparing 
a nephrotoxic serum by immunizing rabbits with dog kidney: 
1. Anesthetize a dog with ether, open the abdomen, wash the blood 
out of the kidneys by inserting a cannula high up in the abdominal aorta, 
and flushing with from 6 to 10 liters of salt solution; open the vena cava. 
2. Remove the kidney as aseptically as possible, and grind it in a meat- 
grinder; rub through fine-meshed wire gauze, and wash the residue in several 
changes of sterile salt solution. The fat should be rejected and only the 
cortex used. 
3. Weigh and suspend 10 gm. of the substance in 30 c.c. of sterile salt 
solution. 
4. -Inject the rabbit intraperitoneally with 10 c.c. of the emulsion every 
seven days. 
5. After the third dose has been given test the serum, as subsequent 
doses increase the danger of losing the animal. 
6. The animal is bled one week after receiving the last injection. 
7. The injection of this serum into dogs is usually followed by albu- 
minuria and possibly hemoglobinuria. This subject is further considered 
under the head of Practical Exercises with Cytotoxins. 
Some investigators have asserted that by effecting immunization with 
the nucleoproteins of an organ more specific cytotoxic serums are secured. 
These claims have not been confirmed by Pearce, Wells, and others. 
Nucleoproteins may be secured as follows: Grind the organ or tissue in a meat-grinder, 
and finally rub up with sand with a pestle in a mortar; add two volumes of normal salt solu- 
tion, and pass through a meat press; collect the effluent, place in a refrigerator for twenty- 
four hours, and then filter through gauze and centrifuge the filtrate; to the supernatant fluid 
add acetic acid to remove the nucleoproteins. Place in the refrigerator for eighteen hours 
and centrifuge. Collect the sediment and wash several times aith normal salt solution. 
Dissolve the sediment in normal salt solution containing 0.5 per cent, sodium carbonate. 
Reprecipitate with acetic acid, wash, and redissolve in the alkaline solution. CHAPTER XXVII 
THE RELATION OF COLLOIDS AND LIPOIDS TO IMMUNITY 
While at the present time Ehrlich’s side-chain theory best explains 
the specificity and mode of action of various antibodies, there is a growing 
tendency to explain many of these reactions on a physicochemical and 
colloidal basis. 
From the fact that, without exception, antigens are colloids, and that 
antibodies also are colloid in their chemical characters, it is advisable to 
review briefly some of the main facts and theories concerning these bodies 
and their reactions. 
Varieties of Colloids.—Colloids may be composed of two different classes 
of substances: 
1. Organic substances, as, e. g., all forms of proteins and also gums, 
starch, glycogen, tannin, chondrin, the greater number of organic dyes, and 
probably the enzymes. 
2. Inorganic substances, as, for example, the inorganic colloids, such as 
silicic acid, ferric hydroxid, arsenic sulphid, and many other similar com- 
pounds. 
Since the living tissues and fluids are, without exception, colloids and col- 
loidal solutions, the properties of the cells are largely the properties of colloids. 
Nature and Properties of Colloids.—Since Graham,1 in 1861, studied 
the differences between the substances that did or did not diffuse readily 
through animal or parchment membranes, soluble substances have been 
classified in two main groups: (a) Colloids, or those substances that were 
dissolved to the extent of showing no visible particles in suspension, but 
that did not pass through diffusion membranes at all, or did so very slowly 
indeed, and (h) crystalloids, or solutions that diffuse through membranes 
quite readily. 
Since Graham’s dicoveries investigations have shown that any and 
all substances may be either colloid or crystalloid, depending upon the 
treatment they receive; thus albumin may be crystallized and common 
table salt obtained in a state of colloidal solution.2 Furthermore, there is 
much evidence indicating that all colloid systems are unstable and never 
in equilibrium; conditions which determine the appearance of a body in the 
colloid or crystalline form appear to indicate that bodies always separate 
from solution in the amorphous or colloidal condition and that ell crystal- 
lization is a secondary phenomenon. 
On the other hand, we may have substances that are quite insoluble 
when aggregated in masses, but when derived in pure form by mechanical 
means can be suspended and uniformly distributed through a fluid without 
showing any marked tendency to precipitate. Such suspensions or emul- 
sions contain particles that are visible under the microscope; they usually 
appear turbid, do not transmit electricity, and are not diffusible. Colloids 
occupy a place between the true solutions of crystalloids and the emulsions. 
Sharp boundaries cannot usually be drawn between any of the members of 
1 Phil. Trans., 1861, 183. 
2 V. Weimarn, Ztsch. f. Chem. Ind. Koll., 1908, 326; ibid., 9110, 7, 92. 
575 576 
THE RELATION OF COLLOIDS AND LIPOIDS TO IMMUNITY 
the series. They differ quantitatively in some manner from the true solu- 
tions and the emulsions, but may approach them closely, and sometimes 
resemble them so strongly as to be almost indistinguishable from them. 
For the most part, however, they show decided characteristics that will 
differentiate them from the crystalloids, on the one hand, and the suspensions, 
on the other. 
Those colloids that closely resemble the true solution have been desig- 
nated colloidal solutions, and those resembling more closely the suspen- 
sions, colloidal suspensions. Of the two types, the colloidal solutions are 
far more important biologically, since the colloidal suspensions are usually 
prepared artificially and seldom occur in nature. 
Colloids, therefore, appear to be suspensions of masses of molecules, or 
perhaps of very large single molecules. When these aggregations are suf- 
ficiently large, we have an ordinary suspension. 
When colloids occur in a highly dispersed state they are called sols; when 
in an undispersed or but slightly dispersed state they are spoken of as gels. 
A colloid substance may be converted from a sol state into a gel state and 
back again, when it is called a reversible colloid or emulsion; a colloid which 
refuses to redisperse is an irreversible colloid or suspension. 
1. Colloids are usually amorphous in character, and with few exceptions 
do not present a typical structure; they are not crystalline under any visible 
condition. This, however, is not invariably the case, for we may have a 
protein, like hemoglobin, which resembles a typical colloid in every respect, 
and may yet form crystals readily and abundantly. 
2. Colloids do not form true solutions, but the solvent is probably an 
important factor in determining whether or not a substance is colloidal in 
nature; e. g., soaps form true solutions in alcohol and colloidal solutions 
in water; rubber forms colloidal solutions in ether, but not in water. The 
term “colloidal solution” does not, therefore, refer to a true solution in the 
sense of a crystalloid, but to a colloidal state of suspension (the so-called 
colloidal solution). 
3. Colloids are non-dif usible, or lack the power of passing through animal 
and parchment membranes. Not all colloids possess the same rate of dif- 
fusion, this property being relative rather than absolute; however, solu- 
tions of salts (crystalloids) pass through so readily that they are easily 
separated from proteins (colloids) by dialyzation, a process that is in con- 
stant practical use. 
4. Colloids have an extremely small osmotic pressure. They may, to a 
very slight degree, exert some influence upon osmotic pressure, the freezing- 
and boiling-points of fluids, but in all cellular processes in which mani- 
festations of osmotic pressure or diffusion are present the crystalloids may 
be considered as almost entirely responsible for these. 
5. The colloids exhibit surface tension to a high degree—in other words, 
colloid fluids possess the force that strives to reduce its free surface to a 
minimum. As partial expressions of this force, the formation of emulsions 
when oil and water are mixed and the ameboid movements of the ameba 
and leukocytes may be mentioned as examples. 
6. Colloids do not separate freely into ions when dissolved, and accord- 
ingly do not conduct electricity to an appreciable extent. When an electric 
current is passed through a colloidal fluid, most of the colloids move toward 
the anode; this phenomenon, known as cataphoresis, is also generally ex- 
hibited by suspensions, and in this particular the colloids resemble sus- 
pensions. 
7. Colloids are usually easily precipitable and coagulable, and this is NATURE AND PROPERTIES OF COLLOIDS 
577 
readily understood when the slender margin that exists between many of 
the colloids and the suspensions is borne in mind. Relatively slight changes, 
such as exposure, gentle heat, the presence of large quantities of crystalloids, 
the action of enzymes, etc., may throw an organic colloid out of solution, 
and when once precipitated, it is often incapable of again dissolving in the 
same solvent. Colloids are also precipitated by many electrolytes, appa- 
rently through the formation of true ion compounds. 
8. The physical structure and size of colloids. This subject has been 
studied extensively by Hardy.1 Cells contain but one type of colloids, 
the proteins that form non-reversible coagula. So long as a colloid is in 
solution it is structureless; but such solutions may become solid as the 
result of changes of temperature and other physical means and from ad- 
mixture with certain chemical fixing agents. The structure of the coagula 
varies according to the concentration of the colloidal solution and the nature 
of the coagulant, but in general the figures obtained in the solidification of 
protein solutions by such fixing agents as mercury bichlorid and formalin 
bear a striking resemblance to the finer structure of protoplasm as described 
by cytologists. These facts, no doubt, have an important bearing upon the 
various “foam,” “reticular,” and “pseudo-alveolar” structures of the proto- 
plasm of cells described by Biitschli, Fromann, Arnold, Reinke, and others, 
and may indicate the effect of fixatives upon colloid solutions, explaining 
the usual time-worn objections to theories of protoplasmic structure as based 
upon artificial conditions not present in the normal living cell, and variously 
interpreted according to the fixative employed. 
Studies in the size of colloidal particles have been greatly facilitated 
by dark-field illumination of the microscopic field or the so-called ultra- 
microscope devised by Siedentopf and Zsigmondy. Various other methods 
have been devised for studying the size of colloidal particles, as ultrafiltra- 
tion by Bechhod2; the weight of a dispersed substance in a given volume 
by chemical or other analysis; by studies in the density of the dispersed sub- 
stance and other methods. These studies have indicated that colloidal 
particles are usually round or at times ovoid, as indicating beginning crystal- 
lization; that the main difference betwreen suspensions and colloidal solu- 
tions is one of dispersion and not relative size of particles, and that in a true 
colloidal solution particles of widely differing sizes may be found side by 
side. 
9. Colloids may he precipitated by electrolytes of opposite sign, as well as 
by colloids. In a colloidal solution surface tension constantly tends to make 
the particles of colloid approach one another, so that the surface may become 
as small as possible, and in this manner brings about precipitation or coag- 
ulation. 
In a stable solution this action is counterbalanced by a force of electric 
repulsion. Pure colloids do not carry an electric charge and are not con- 
veyed by an electric current; their apparent charge depends upon the nature 
of electrolytes that may be present. Traces of acid and of acid salts give 
it a positive charge, whereas alkalies and alkaline salts do the opposite 
(Pauli). 
The process of coagulation of proteins, therefore, must depend upon 
the neutralization of their electric charge, and, as previously stated, this 
can be accomplished either by electrolytes or by colloids: 
(a) Precipitation by electrolytes is best illustrated by the action of a 
1 A good general outline of the subject of colloids may be found in Pauli’s Physical Chem- 
istry in the Service of Medicine, 1907, translated by Fischer (Chapman and Hall). 
2 Ztschr. f. Chem. Ind. Roll., 1907, 2, 3. 578 
THE RELATION OF COLLOIDS AND LIPOIDS TO IMMUNITY 
strong acid on an albuminous solution. The negatively charged particles 
attract to themselves the positively charged hydrogen ions; their charge 
is now neutralized, and the force of attraction due to their surface tension 
is no longer counterbalanced by an electric repulsion. The particles are 
drawn together, form larger and larger masses, which finally come under 
the influence of gravity and precipitation takes place. 
(b) Precipitation of colloids by colloids is illustrated by the precipita- 
tion of albumin by acetic and ferrocyanic acids. The colloid must be of 
opposite sign. As a result of the acid the particles acquire a positive charge, 
if they are not so charged already. This charge is then neutralized by the 
colloidal ferrocyanic acid of negative sign; the surface tension is no longer 
neutralized by an electric repulsion, and particles come together to form 
larger masses that are finally deposited as a precipitate. 
Instead of precipitating the other, an excess of one colloid may act in a 
reverse manner. For example, as Neisser and Friedmann have shown, a 
suspension of particles of mastic in water (made by dropping an alcoholic 
solution in water) takes on a negative charge, and can be precipitated by 
positive colloids or ions, such as ferric chlorid. If the dose of ferric chlorid 
is increased gradually, the precipitate becomes more and more abundant, 
until an excess of ferric chlorid is present, when the reaction ceases and the 
precipitate may be redissolved. This has been explained on the assumption 
that when two colloids of opposite sign are mixed, they tend to fuse and form 
masses; the addition of an excess of either colloid tends to electrify the masses, 
causing mutual repulsion and possibly resolution of the masses. Hence the 
precipitate is soluble in an excess of both substances, just as a precipitate 
is soluble in an excess either of precipitin or of its antigen. 
10. Absorption is the taking up of dissolved or volatile substances by 
finely divided or colloidal bodies. It is a combination between two sub- 
stances dependent on physical attraction rather than on chemical affinity, 
and taking place in variable ratios, rather than in simple and constant ones, 
as occur in a true chemical union. It is believed by many that the two 
substances entering into the phenomenon of absorption exist as such side 
by side in the compound, which is to be regarded as an intimate admixture 
of the two rather than as a new compound. 
The lack of definite ratios by which colloids are absorbed has been 
shown by Bordet in the amount of hemolytic immune body that can be 
taken up by a given volume of corpuscles—i. e., the amount varies according 
to whether the corpuscles are added at once or in successive small portions. 
Thus, in one example, 0.4 c.c. of a hemolytic serum dissolved 0.5 c.c. of 
corpuscles if added at once; but if 0.2 c.c. of corpuscles was added first and 
successive amounts of 0.1 c.c. then put in, no solution took place after the 
one that followed the addition of the first portion. This was explained by 
Bordet according to the principles of absorption, this observer comparing 
it with the absorption of a dye by filter-paper. While other explanations 
are possible, yet exactly analogous phenomena may be seen in the mutual 
absorption of colloids of opposite sign. Thus, as we have previously stated, 
the addition of a solution of an electropositive colloid to a solution of an 
electronegative one tends to repel the particles, with the formation of masses 
for the purpose of self-protection, and in this manner the process of ag- 
glutination and precipitation is begun. But if a small amount of a second 
colloid is added to the same volume of the others, new aggregates of the two 
are formed that are less favorable to precipitation and require more of the 
second colloid to bring about complete precipitation. REACTIONS OF IMMUNITY AND COLLOIDAL CHEMISTRY 
ANALOGY BETWEEN THE REACTIONS OF IMMUNITY AND COLLOIDAL 
CHEMISTRY 
579 
With these few brief remarks on the properties and nature of colloids 
and the close resemblance of cellular protoplasm and fluids to colloids, we 
may consider briefly the apparent similarity that exists between the col- 
loidal reactions and some of the reactions of immunity. This is especially 
pertinent for several reasons: it has been shown that cellular protoplasm is 
colloidal in nature; that antigens are certainly colloidal, and that anti- 
bodies, while they may or may not be solutions of colloids, are, in the final 
analysis, products of cellular activity, and therefore derived from colloidal 
solutions. Too much emphasis, however, cannot be placed upon the value 
of interpretation of immunologic phenomena in the test-tube and living ani- 
mal on the basis of colloidal reactions. As stated by Krogh1 we must con- 
sider colloidal chemical laws in relation to immunologic researches, but our 
present knowledge of colloidal chemistry is so elemental that only a small 
part of immunology is explainable by colloidal reactions of alternations and 
absorptions. A thorough review of the subjects of colloids and catalysis in 
relation to antigens and antibodies has been recently made by Nicolle and 
Cesari.2 
1. Antitoxins.—The side-chain theory of Ehrlich was first applied in 
explanation of the principles of immunity as affording an explanation of the 
action of toxins, the formation of antitoxin, and the interaction between 
these. Ehrlich has placed the various phenomena of immunity upon a 
chemical basis, bringing forward new theories to explain the various dis- 
crepancies that were found. For example, it was soon found that both 
toxin and antitoxin were unstable, and that neutralization of a toxin by the 
addition of antitoxin was not a simple process, like the neutralization of an 
acid by an alkali, but, on the contrary, was likely to be exceedingly com- 
plicated. This was explained as being due to the degeneration of toxin into 
various toxoids, which were able to neutralize antitoxin without being in 
themselves toxic when in a free state. They were likewise found to have 
a greater affinity for antitoxin than the toxin itself, so that when a toxin 
was tested and its toxicity determined, it was discovered that more anti- 
toxin was needed to neutralize the mixture than was originally calculated, 
because the toxoids took no part in testing the toxin, but were active in unit- 
ing with antitoxin, and in this manner leaving true toxin unneutralized, and 
therefore toxic, unless an excess of antitoxin was used. 
Analogous conditions may be observed among colloidal solutions. Thus, 
Danysz has shown that more toxin is neutralized if antitoxin is added at 
once than when it is added in successive doses. As stated elsewhere, this- 
is explained by Ehrlich upon the assumption that time is allowed for the 
degeneration of toxin into toxoids to take place, the latter having a greater 
affinity for the antitoxin. It has been shown, however, that in some cases 
the addition of a small amount of a second colloid of opposite sign to a col- 
loidal solution may render the solution more stable and protect it from 
precipitation by an excess of the second substance. Similarly, the amount 
of colloid necessary to precipitate a constant amount of another colloid 
is reduced to a minimum if the addition is made at once, and is rendered 
much greater if the colloid added is made slowly in small amounts, an in- 
terval being allowed to elapse after each addition. This is closely analogous 
to the Danysz reaction, and explains the latter as being due, when antitoxin 
is added slowly to toxin, to the formation of transitional compounds of toxin 
1 Jour. Infect. Dis., 1916. 19, 452. 
2 Ann. de l’Inst. Pasteur, 1922, 36, 463. THE RELATION OF COLLOIDS AND LIPOIDS TO IMMUNITY 
580 
and antitoxin of diverse nature, requiring more antitoxin for complete 
neutralization of the toxin than if the antitoxin were added at once and in 
one dose. , , 
The difference between the L0 and L+ dose of a toxin also has an an- 
alogy in the reaction of simple colloidal substances. Thus, Bilty has used 
ferric hydroxid, which neutralizes arsenic trioxid (the antidote for acute 
arsenical poisoning), and found that the addition of one lethal dose of 
arsenic to a neutral mixture of the two did not render the mixture toxic, 
but that several lethal doses were required, just as it is necessary to add 
several instead of one lethal dose of diphtheria toxin to the L0 dose. 
Thus it would appear that the neutralization of a toxin by an anti- 
toxin has analogies among the known and simple colloidal reactions. One 
objection to placing the toxin-antitoxin reaction upon a colloidal basis is 
that both have the same electric charge, i. e., both move toward the cathode, 
and, as we have seen, for the neutralization and precipitation of colloids 
the solution of colloids should be of opposite sign. It must be remembered, 
however, that toxins and antitoxins react in very complex fluids containing 
other substances consisting of both colloids and electrolytes, and until the 
electric charge of these in pure form is determined, the apparent similar 
electric charge of toxin and antitoxin can hardly outweigh the otherwise 
remarkable analogy it bears to colloidal reactions. The neutralization of 
toxin by antitoxin is not, however, according to Arrhenius and Krogh,1 a 
purely colloidal phenomenon of absorption. L.rogh found that toxin 
absorbed to an organic suspended substance quite as harmful to guinea- 
pigs as before, the only difference being that it was absorbed more slovly, 
absorption may explain the union, but does not explain the neutralization. 
2. Agglutinins and Precipitins.—Various theories in explanation of the 
phenomenon of agglutination have been described in a previous chapter. 
The theory of Bordet appears to be best, and is based upon certain prin- 
ciples of colloidal chemistry. When bacteria are suspended in a fluid free 
from salt agglutination does not take place because the bacteria carry a 
similar negative charge of electricity. When, however, ions of positive 
charge are added, as, c. g., sodium chlorid, the bacteria or other cells are 
repelled and coalesce to form masses, according to the laws of surface ten- 
sion, in an effort to protect themselves. Larger masses may be formed that 
finally come within the influence of gravity and are deposited at the bottom 
of the test-tube. According to the same laws the addition of agglutinin 
removes the negative charge of bacteria or other cells, with the consequent 
formation of clumps and masses. Similar phenomena may be observed in 
the precipitation of colloidal suspensions of clay in distilled vater by the 
addition of a salt. ... 
Solutions of inorganic colloids, as, for example, that of silicic acid, may 
agglutinate red corpuscles; bacteria, such as suspensions of typhoid and 
colon bacilli, may be agglutinated by solutions of the ferric salts. 
Just as an excess of one colloid solution will charge masses of the other, 
resulting in a repelling action and breaking up of the agglutinated clumps, 
so the addition of an excess of agglutinin is found to prevent agglutination 
or to give but a slight reaction. This phenomenon has been explained, 
according to Ehrlich’s side-chain theory, as due to the presence of agglutin- 
oids that have a great affinity for the bacteria and unite with them without 
being active in the free state, owing to a loss of the agglutinophore portion 
of the molecule. Each cell united with an agglutinoid is one cell less to 
undergo agglutination by agglutinin, and accordingly in weak dilutions of 
1 Jour. Infect. Dis., 1916, 19, 452. COMPLEMENT FIXATION AS A COLLOIDAL REACTION 
581 
serum agglutination is feeble or absent, whereas in higher dilutions the 
phenomenon may be clearly observed. 
Other explanations of the action of agglutinins and precipitins, based 
upon colloidal reactions, have been advanced. Thus Neisser and Fried- 
mann have shown that suspensions of mastic may be “protected” against 
the precipitating action of ferric hydroxid by the addition of a small amount 
of organic colloid, such as serum, leech extract, or extract of typhoid bacilli, 
regardless of whether this colloid is charged positively or negatively or is 
neutral. The aforenamed observers believe that normal bacteria may be 
surrounded by a similar protective envelope that prevents the agglutinating 
action of substances of opposite sign. The action of agglutinin, therefore, 
would be to remove this layer, so that the ions of opposite electric charge 
can unite with the bacteria and bring about their agglutination. This may 
be an explanation of the role of salts in the phenomenon of agglutination, 
the agglutinins removing the protecting envelopes and the salt furnishing the 
ions of opposite charge that bring about agglutination. 
Owing to the fact that a discrepancy arises here for the reason that 
emulsions of red corpuscles are agglutinated by both positive and negative 
colloids (ferric hydroxid and cuprum ferrocyanid), Girard, Mangin, and Henri 
have given the following explanation of agglutination: When a red corpuscle 
is suspended in a fluid, various salts, especially the sulphates of magnesium 
and calcium, are diffused, which tends to facilitate the precipitation of 
negative and positive colloids, so that each corpuscle comes to be surrounded 
by a layer of precipitated colloid material. This zone of precipitated col- 
loids of either negative or positive charge determines agglutination in the 
presence of a colloid solution of opposite charge, such as agglutinin or in- 
organic colloids (silicic acid, etc.). 
3. Hemolysins.—Reference has been made elsewhere to the original 
observations of Bordet, showing that red corpuscles may absorb much more 
hemolytic antibody than is necessary to bring about their lysis, and that 
this absorption is analogous to colloidal absorption. 
Inorganic colloidal solutions, such as that of silicic acid, may produce 
hemolysis of red blood-corpuscles, e. g., those of the rabbit. Its action is 
manifested in extremely small doses. It is rendered inert by heat, and 
gradually deteriorates at room temperature. Furthermore, this inorganic 
colloid possesses some of the properties of a serum hemolysin; thus mice red 
corpuscles that have been agglutinated by colloidal silicic acid are dis- 
solved by traces of lecithin or of fresh serum, but not by serum that has been 
heated to 60° C. (inactivated). An excess of silicic acid tends to prevent 
hemolysis, which is another example of the action of an excess of one col- 
loidal solution upon another of opposite sign. 
Probably saponin hemolysis and the influence of fatty substances, 
such as lecithin and the fatty acids, upon the phenomenon of hemolysis, 
are closely related to, or to be explained by, the action of organic colloidal 
solutions. 
The fixation of hemolysin to corpuscles apparently follows the simple 
partition law rather than being purely a phenomenon of absorption, because 
the velocity is such as may be readily measured; the colloidal alternation 
law adequately explains the quantitative features of hemolysis. 
The intimate relationship of precipitation to complement fixation has 
suggested that the phenomenon may be a colloidal reaction and principally 
of absorption. This subject is discussed more fully on page 432. 
COMPLEMENT FIXATION AS A COLLOIDAL REACTION 582 
THE RELATION OF COLLOIDS AND LIPOIDS TO IMMUNITY 
THE RELATION OF LIPOIDS TO IMMUNITY 
It is becoming more and more evident that lipoids bear an important 
relation to various immunologic processes, especially to certain cytolytic 
phenomena. This is especially true if the cell walls are composed chiefly 
of lipoids as regarded by several investigators; these substances would thereby 
play an important part in protecting the fixed and wandering body cells as 
well as bacteria. 
As the relation of lipoids to various immunologic processes1 has fre- 
quently been described in earlier chapters, as, e. g., where the role of lipoids 
in venom hemolysis, in the Wassermann syphilis reaction, and in the various 
precipitin reactions in syphilis were considered, a brief resume may be of 
service in directing attention to this important and particular phase of 
immunity. 
Lipoids as Antigens.—This important subject has been discussed on pages 
147 and 426. It would appear to be the consensus of opinion that pure lipoids 
are not antigenic and when injected into animals do not produce antibodies. 
If lipoids are combined with protein, however, the complex may be antigenic 
and engender the production of various antibodies. This factor probably 
explains the discrepancies in the reports of various investigators. As 
previously stated, Jobling and Bull2 observed an increase of serum lipase in 
the sera of rabbits immunized with chicken corpuscles indicated by in- 
creased hydrolysis of ethyl butyrate; it may be stated, however, that accord- 
ing to Thiele and Embleton3 hydrolysis of ethyl butyrate cannot be accepted 
as evidence of lipolytic activity of serum. 
Relation of Lipoids to Ferments, Antiferments, and Immunity.—The 
important investigations of Jobling and Petersen regarding the unsaturated 
fatty acids of serum being antitryptic have been discussed in the chapter 
on Ferments and Antiferments. According to these investigators various 
substances, including the halogens, kaolin, and bacteria, may bring about 
saturation of these and render the serum proteases active. This may be 
followed by the digestion of serum or other proteins with the production of 
toxic substances. 
Relation of Lipoids to Opsonins and Phagocytosis.—These have been 
studied by Walbum,4 Stuber,5 Arkin,6 Dewey and Nuzum.7 According to 
the results observed by Walbum, the administration of cholesterol increases 
phagocytosis, but Stuber and Arkin found this substance to depress phago- 
cytic activity. This was confirmed by Dewey and Nuzum, who observed 
that the depressing effect is chiefly on the leukocytes. Graham8 has ob- 
served that the opsonins are decreased following ether anesthesia, which sug- 
gests that these effects may be due to the activity of lipoidal substances. 
According to Muller9 the bacterial lipoids do not increase or decrease phago- 
cytosis and are unimportant in this connection. 
The available date bearing upon this subject would indicate, therefore, 
that certain lipoids in serum, and notably cholesterol, may inhibit phago- 
cytosis, but that the lipoids of the phagocytosed cells are themselves inactive 
in the process. 
1 Bibliography on Lipoids and Immunity given by Landsteiner, Kolle and Wasser- 
mann’s Handbuch, 1913, 2, 1240. Also review of literature by Landsteider, Jahresb. Im- 
munitatsf., 1910, 6, 209. 
2 Jour. Exper. Med., 1913, 17, 61. 
3 Jour. Path, and Bacteriol... 1914, 19, 349. 
4 Ztschr. f. Immunitatsf., 1910, 7, 544. 
5 Biochem. Ztschr., 1913, 51, 211; ibid., 1913, 53, 493.' 
8 Jour. Infect. Dis., 1913, 13, 408. 8 Jour. Infect. Dis., 1911, 8, 147. 
7 Jour. Infect. Dis., 1914, 15, 472. 9 Ztschr. f. Immunitatsf., 1909, Ref., 1, 61. THE RELATION OF LIPOIDS TO IMMUNITY 
583 
Relation of Lipoids to Agglutinins.—Apparently very little investigation 
has been devoted to this subject. From1 found that corpuscles free of lipoids 
produce agglutinins, but not hemolysins, when injected into animals. Thiele 
and Embleton2 have likewise found that the lipoids play no part in the pro- 
duction of agglutinins. According to Stuber,3 however, immunization of 
rabbits with fat-free typhoid bacilli results in diminished agglutinin produc- 
tion. He believes that the agglutinins are produced as a result of the 
stimulus afforded by the fats liberated after destruction of the bacteria. 
This subject requires investigation. At the present time the role of 
lipoids in the production of agglutinins, that is, as agglutininogens, is doubt- 
ful, and nothing has been done to show that they are concerned in the mechan- 
ism of the agglutination reaction. 
Relation of Lipoids to Hemolysis.—(a) From the standpoint of im- 
munity, venom hemolysis is of peculiar interest as indicating the possible 
important relation of lipoids to hemolytic complement. Granting that venom 
contains a hemolytic amboceptor (Flexner and Noguchi), the complementing 
substance must be derived from the corpuscles, and, according to Kyes, 
this complementary agent is represented in lecithin. Kyes was able to pro- 
duce what he considers are compounds of the hemolysin with lecithin, 
namely, “lecithids.” Whether these “lecithids” are true compounds of 
hemolysins and corpuscular lecithin or simply the active hemolytic products 
of the cleavage of lecithin by ferments contained in the venom, is at present 
unknown. Noguchi and Lieberman have shown that not only lecithin, but 
soap as well, especially unsaturated fatty acids, and probably protein com- 
pounds of soaps and lecithin, may act as the hemolytic complement and 
activate the hemolysin of the venom. Lipoids from bacteria and trypano- 
somes have been found to possess similar properties. Hemolytic lipoids 
have been secured from serum, and the complementary activity of a fresh 
normal serum may be destroyed by fat solvents, e. g., ether. While other 
investigators have not been able to confirm Noguchi’s attempts to produce 
an artificial complement of fatty substances of exactly the same properties 
as serum complement, this work indicates most strongly the close relation of 
serum complement to lipoids or protein-lipoid compounds. 
(b) The hemotoxic activity of various toxins is probably dependent largely 
upon their action on the lipoids of red corpuscles. The saponin substances,4 
a group closely related to glucosids, and found in at least 46 different families 
of plants, are strongly hemolytic. Ransom5 has found that an ethereal 
extract of red corpuscles contains a substance that inhibits saponin he- 
molysis. This substance consists largely of cholesterin, and it is the pres- 
ence of cholesterin in normal serum that inhibits saponin hemolysis. This 
may be demonstrated experimentally by adding cholesterin to a solution of 
a saponin. Noguchi6 has shown that lecithin does not possess the same 
antihemolytic action on saponin. It would appear, therefore, that saponin 
causes hemolysis by combining with, altering, or dissolving the lipoids of 
the stroma of corpuscles. The resistance of corpuscles to saponin hemol- 
ysis varies in certain diseases, being especially low in jaundice (McNeil7). 
While saponins, solanins, phallin, and other vegetable poisons are of 
relatively simple chemical composition and quite unlike proteins, enzymes, 
1 Compt. rend. Soc. de biol., 1912, 72, 154. 
2Ztschr. f. Immunitatsf., 1913, 16, 161. 
3 Munch, med. Wchn., 1915, 1173. 
4 Complete literature on saponin, see Kobert, Die Saponinsubstanzen, Stuttgart, 1904. 
5 Deut. med. Wchn., 1901, 27, 194. 
6 Univ. of Penna. Med. Bull., 1902, 15, 327. 
7 Jour. Path, and Bact., 1910, 15, 56. 584 
THE RELATION OF COLLOIDS AND LIPOIDS TO IMMUNITY 
or toxins, it is possible that bacterial and vegetable hemotoxins, such as 
tetanolysin, abrin, ricin, crotin, and robin, may produce their effects by a 
similar action on the lipoids of the erythrocytes. Noguchi has shown that 
cholesterin inhibits the action of tetanolysin. Landsteiner and Bottori have 
found that protagon, a brain lipoid, possesses the property of binding tetanus 
toxin, which indicates that this toxin may produce its effects by some action 
upon the lipoids of nerve-cells. 
(c) Whether serum hemolysins are lipoidal constituents is doubtful. 
Jobling and Bull1 have claimed that immunization of rabbits with hen cor- 
puscles results in an increase of lipase based upon the serum acquiring the 
property of hydrolyzing ethyl butyrate, but Thiele and Embleton,2 claim 
that the hemolytic power of a serum bears no relationship to its lipolytic 
power and serum hemolysin is not a lipase. 
Relation of Lipoids to the Wassermann Reaction and Precipitation.— 
The important relation of lipoids to the Wassermann reaction and certain 
precipitin or floccule-forming reactions (Klausner, Porges-Meier, Hermann- 
Perutz) has been mentioned repeatedly. Just what role the lipoids play in 
these phenomena is not known. While the globulins of syphilitic serums 
are strongly suspected of being concerned in these processes, their relation 
is not clear. Klausner3 now believes that the precipitate that forms when 
distilled water is added to syphilitic serum is due to the high lipoid content.. 
I have previously discussed (page 435) the important relation serum 
lipoids may bear toward the anticomplementary activity of serum and to 
non-specific complement fixation by the normal sera of rabbits, dogs, and 
mules. Removal of these substances by lipoid solvents tends to decrease 
these effects. 
Mention has also been previously made (page 429) to the probable role 
of lipoids in the complement-fixation reaction, and more especially to the 
lipoidal nature of alexofixagens in syphilis and other infections. The 
syphilis “reagin” is especially important in this connection because it may 
be dissipated by such lipoid solvents as ether and alcohol, while chloroform 
narcosis is said to render the sera of normal individuals Wassermann positive. 
Furthermore, the serum lipoids may be increased in syphilis as claimed by 
Citron and Reicher,4 Peritz,5 and others, indicating without much doubt 
that lipoids play an important role in the phenomena of serum hemolysis, 
antihemolysis, non-specific and probably specific complement fixation. 
Relation of Lipoids to Anaphylaxis.—This subject is considered in more 
detail in the chapters devoted to. Anaphylaxis. In regard to sensitization it 
is now well proved that pure lipoids, that is, protein-free lipoids, are not 
able to act as anaphylactogens. This does not mean, however, that lipoids 
may not play some part in anaphylactic phenomena. For example, Saula6 
has observed an increase in soap and fatty acids subsequent to sensitization, 
and Jobling and Petersen7 have noted symptoms and lesions resembling 
anaphylaxis following the intravenous injection of guinea-pigs with soap. 
According to these investigators when the antitryptic power of the serum 
of guinea-pigs is raised (increase of unsaturated fatty acids) or, when soaps 
are added to the intoxicating dose of protein, the animals are able to resist 
several times the amount of specific protein fatal for the controls. 
1 Jour. Exper. Med., 1913, 17, 61. 
2 Jour. Path, and Bact., 1914, 19, 349. 
3 Biochem. Ztschr., 1912, 47, 36. 
4 Berl. klin. Wchn., 1908, 1398. 
5 Deut. med. Wchn., 1910, 36, 481. 
6 Compt. rend. Soc. de biol., 1913, 15, 273. 
7 Jour. Exper. Med., 1914, 20, 468. TECHNIC OF COLLOIDAL REACTIONS 
585 
TECHNIC OF COLLOIDAL REACTIONS 
Lange’s Colloidal Gold Reaction 
Principles.—According to the exhaustive studies of Zsigmondy1 on 
metallic colloids a solution of a protein will precipitate colloidal gold in the 
absence of an electrolyte. An electrolyte, as sodium chlorid, will in certain 
concentrations precipitate the colloidal gold itself; if proteins are present 
this precipitation is inhibited. The degree of protection afforded has been 
found specific for each protein, and is expressed in terms of milligrams of 
the protein capable of protecting 5 c.c. of colloidal gold against 0.5 c.c. of a 
10 per cent, solution of sodium chlorid. 
Lange2 endeavored to distinguish between normal and luetic sera by 
this means on the basis of disturbances in syphilis, but failed; he then 
sought to measure the protein content of cerebrospinal fluid by the degree 
of precipitation of gold, but failed again, because he used distilled water as a 
diluent, thereby throwing the protein, particularly the globulins, out of 
solution, and rendering them inert. When, however, he used a 0.4 per cent, 
solution of sodium chlorid as a diluent it was found that the proteins were 
not precipitated and that this amount of salt was too weak to precipitate 
the colloidal gold. 
It was now possible, according to Lange, to measure the protein con- 
tent of spinal fluid according to the degree of precipitation, and very inter- 
esting and practical results have been secured, although it cannot be stated 
as proved that the precipitation is due entirely to protein, particularly since 
it has been shown that the globulins may actually protect colloidal gold 
against precipitation. Fischer3 has recently, however, disproved the pro- 
tective value of the globulins and found that fibrinogen plus fibrinoglobulin- 
euglobulin, and pseudoglobulins possess a flocculating effect upon the col, 
loidal gold solution. Zaloziecki4 regards the reaction as a form of immu- 
nity reaction; Jaeger and Goldstein5 consider it purely physical and probably 
of an electric nature. 
Preparation of Colloidal Gold.—The preparation of colloidal gold is 
frequently an exceedingly troublesome procedure, and the success of the 
test depends upon a satisfactory preparation. The method which I shall 
briefly describe here is after that of Miller, Brush, Hammers, and Felton,6 
which I have found to yield fairly constant and satisfactory products. 
The glassware (beakers, pipets, and test-tubes) must be absolutely clean. 
They may be washed in hot water with ivory soap; rinsed in tap-water for 
five minutes; placed in hot bichromate cleaner for half an hour; rinsed in 
tap- and, finally, triple distilled water. The beakers should be used at once; 
the pipets and test-tubes are to be dried in a hot-air oven. Thermometers 
should be cleansed in a similar manner. 
It is necessary to use water triply distilled in an apparatus free of rubber 
connections. The reagents should be freshly prepared of the best products 
obtainable. 
1. Heat 1000 c.c. of triply distilled water over a good Bunsen burner in 
a prepared beaker with a thermometer. 
1 Ztschr. f. Anal. Chem., 1901, xl, 697. 
2Berl. klin. Wchn., 1912, xlix, 897; Ztschr. f. Chemotherap., 1913, 1, 44. 
3 Ztschr. f. d. ges. Exper. Med., 1921, 14, 60. 
4 Deutsch. Ztschr. f. Nervenh., 1913, xlvii, 783. 
5 Ztschr. f. d. ges. Neur. u. Psych., 1913, xvi, 219. 
6 Bull. Johns Hopkins Hosp., 1915, xxvi. 391. 586 
THE RELATION OF LIPOIDS AND COLLOIDS TO IMMUNITY 
2. At 60° C. add 10 c.c. of a 1 per cent, solution of Merck’s gold chlorid 
crystals in triply distilled water and 7 c.c. of a 2 per cent, solution of Merck’s 
blue label potassium carbonate in triply distilled water. 
3. At 80° C., while stirring briskly, add 10 c.c. of a 1 per cent, solu- 
tion of Merck’s blue label oxalic acid crystals in triply distilled water. 
4. At 90° C. remove the burner and, while stirring, add 5 c.c. of a solu- 
tion of 1 c.c. of Merck’s highest purity formaldehyd in 40 c.c. of triply dis- 
tilled water, or enough to produce an initial pink color. 
5. The solution must be neutral in reaction when used, and for this 
purpose is tested with a 1 per cent, solution of alizarin red in 50 per cent, 
alcohol. With this indicator the neutral point is a brownish-red tint; an 
acid solution gives a lemon-yellow, and an alkaline solution a purplish-red 
color. 
To 10 c.c. of the colloidal gold in a clean beaker add 2 drops of indicator. 
If it is acid, titrate to the neutral point with n/50 NaOH; if alkaline, with 
n/50 HC1. Calculate the amount to be added for the amount of colloidal gold 
solution at hand and neutralize with normal or decinormal solutions of acid 
or alkali as required. 
The following are suitable standards for a good colloidal gold product: 
{a) Absolutely transparent and of a brilliant red orange or salmon-red 
color. 
(b) Five c.c. of the solution must be completely precipitated in one 
hour by 1.7 c.c. of a 1 per cent, solution of sodium chlorid in distilled water. 
(c) The solution must be neutral in reaction when used. 
(d) The solution must not produce a reaction greater than a No. 1 
with normal cerebrospinal fluid, and must give a typical reaction curve 
with a known paretic fluid. 
The Test.—1. Arrange eleven clean, dry test-tubes in a row; put 1.8 c.c. 
of fresh, sterile 0.4 per cent. NaCl solution into the first tube and 1 c.c. in the 
following ten. 
2. With a clean, dry pipet add 0.2 c.c. of a blood-free cerebrospinal fluid 
to the first tube and mix; transfer 1 c.c. from the first to the second tube, 
mix, and proceed in this manner up to and including the tenth tube, from 
which 1 c.c. is discarded. The eleventh tube is the control and contains 
no cerebrospinal fluid. The dilutions now range from 1 to 10 to 1 to 
5120. 
3. Add to each tube 5 c.c. of colloidal gold; mix and stand aside at room 
temperature over night; the readings are made the next day and recorded 
by numbers according to the following scheme: 
5 = complete precipitation (water clear). 
4 = pale blue. 
3 = blue. 
2 = lilac or purple. 
1 = red-blue. 
0 = no change. 
Types of Reactions.—1. Normal fluid produces no changes at all or, at 
most, a No. 1 change in the first tube of the series. 
2. The typical reaction is observed in general paresis, giving complete 
precipitation in the first four to eight tubes of the series, with changes of 
color in most of the remaining ones, as, for example, 555554210 0, and 
constituting the “paretic curve” of Miller and Levy.1 A similar curve is 
usually found in taboparesis. (See following table and Fig. 149.) 
1 Bull. Johns Hopkins Hosp., 1914, xxv, 123. Fig 149 —Colloidal Gold Reactions. 
The upper set shows a reaction with the cerebrospinal fluid of a paretic (“paretic curve”); the lower set shows a reaction 
with the cerebrospinal fluid of a tabetic (“luetic curve”). TECHNIC OF COLLOIDAL REACTIONS 
587 
Color Reactions 
Numbers used, 
for designating 
Color Reactions 
Dilutions of 5 
pin a! Fluid with OF 
oer cent Nad. 
7:10 
7:20 
7-M 
1:80 
1:160 
1:320 
l:6h0 
1S280 
1Z560 
1:5/20 
Complete deco/oraaTion 
5 
\ 
,A 
Pale blue 
4- 
\ 
\ 
—/ 
/ 
\ 
\ 
Blue 
3 
\ 
/ 
A, 
\ 
\ 
Lilac or purple 
Z 
. \ 
\ \ 
—\\ 
\ 
Bed— blue 
7 
/ 
\ 
\ 
V 
Brilliant Red orange 
0 
— 
—2 
_ No reaction; normal fluid. - Luetic Reaction Type (Fig. 149) 
CFig. 149). (Verschiebung nach oben.) 
Showing the Four Common Types of Colloidal Reactions 
3. The cerebrospinal fluid in tabes dorsalis usually produces the “luetic 
zone” curve of a No. 4 intensity, as, for example, 444554200 0; pre- 
cipitation being partial in the first two or three tubes, then becoming com- 
plete, and gradually returning to normal through the balance of the series. 
The changes, however, are not constant or characteristic. 
4. The fluids in cerebrospinal syphilis usually yield weak reactions of 
the “luetic zone” type. In tertiary syphilis without symptoms referable to 
the central nervous system similar reactions are frequently observed. 
5. Fluids from cases of purulent or tuberculous meningitis may give 
reactions which are usually maximal in the higher dilutions, the “meningitic 
zone” type and so-called “Verschiebung nach oben.” In acute anterior 
poliomyelitis the cerebrospinal fluid frequently yields reactions similar to 
the “luetic” and “meningitic” zone types of reactions. 
Practical Value of the Colloidal Gold Test.—The reports of a very large 
number of investigators, including Lange,1 de Cruris and Frank,2 de Cruris 
and Eberhardt,3 Grulee and Moody,4 Eicke, Jaeger5 and Goldstein,6 Klien- 
berger,7 Zaloziecki,8 Kaplan,9 Miller and Levy,10 Swalm and Mann,11 Lee and 
Hinton,12 Weston, Darling and Newcomb,13 Solomon and Welles,14 Miller, 
Brush, Hammers and Felton,15 and others show that under proper condi- 
tions the colloidal gold test is highly specific and of value in the diagnosis of 
general paresis. 
A paretic colloidal gold reaction with the cerebrospinal fluid of a luetic 
person may be the first sign of an incipient paresis. 
A paretic reaction with the cerebrospinal fluid of a luetic individual 
should be regarded as of grave import; while positive symptoms of paresis 
may not be present, intensive antiluetic treatment may serve to arrest the 
disease. 
The colloidal gold reactions in tabes dorsalis and cerebrospinal syphilis 
are not characteristic of these diseases, and at most may yield the luetic zone 
type of reaction, and thereby prove of value in confirming a doubtful diag- 
nosis. 
1 Ztschr. f. Chem. u. v. Gehete., 1913, 1, 44; Berl. klin. Wchn., 1912, xlix, 897. 
2 Munch, med. Wchn., 1915, lxi, 1209. 6 Loc. cit. 
3 Miinch. med. Wchn., 1914, 61, 1216. 7 Arch. f. Psych., 1911, xlviii, 264. 
4 Amer. Jour. Dis. Child., 1915, ix, 19. 8 Loc. cit. 
5 Miinch. med. Wchn., 1913, lx, 2713. 
9 Ztschr. f. d. ges. Neurol, u. Psych., 1915, xxvii; Jour. Amer. Med. Assoc., 1914, lxii, 511, 
246. 
10 Bull. Johns Hopkins Hosp., 1914, 25, 123. 12 Amer. Jour. Med. Sci., 1914. cxlviii, 33. 
11 New York Med. Jour., 1915, xi, 719. 13 Amer. Jour. Insan., 1915; lxxi, 773. 
14 Boston Med. and Surg. Jour., 1914, clxxi, 886; ibid., 1915, clxxii, 398. 
15 Bull. Johns Hopkins Hosp., 1915, 26, 391. 588 
THE RELATION OF COLLOIDS AND LIPOIDS TO IMMUNITY 
Principle.—This reaction is based upon the observation made by Weich- 
ardt1 in 1908; he found that diffusion is accelerated when differently colored 
solutions of antigen and its specific antibody are brought together. Changes 
in diffusion are associated with changes in the surface tension, both of which 
depend on a change in the osmotic pressure. This is the principle made use 
of by Ascoli in his miostagmin reaction, which will be described further on. 
Later Weichardt made the reaction more accessible to practical use by 
introducing into the solution of serums and antigen a system composed 
of sulphuric acid and barium hydroxid, together with certain catalytic agents. 
Using phenolphthalein as an indicator, he could show’ that fresh serums in 
high dilutions alter the surface tension of the finely divided barium sulphate 
particles by their colloidal action, so as to increase the absorption of H-ions, 
thus rendering the solution more alkaline. 
This phenomenon has been utilized by Weichardt, under the name 
of “epiphanin reaction,” to determine the occurrence of such interaction of 
antigen and antibody. The reaction probably depends upon physico- 
chemical principles of absorption, but the exact nature of the change is not 
yet understood. The reaction is based upon the following generalizations: 
1. Solutions containing colloids—i. e., antigen alone, antiserum alone, 
or antigen plus non-specific antiserum in certain dilutions—act in the fore- 
going system by shifting the phenolphthalein end-point (the point of neu- 
tralization when acid and alkali are brought together in the presence of this 
indicator) in the sense of increased OH-ions (pink color). 
2. Specific antigens can inhibit the activity of their specific antiserums, 
the specific antigen-antibody combination then becoming evident in vitro 
by a shift of the end-point in the sense of increased H-ion concentration 
(light color). 
Specificity.—The specificity of the reaction has been confirmed by a 
number of investigators who used the test for the identification of a host of 
antigen-antibody combinations in vitro. The underlying principles have 
been confirmed by Kraus and Amiradzibi,2 Schroen,3 Seifert,4 Mosbacher,0 
and others. The reaction has been applied to a study of various antigens 
and their antibodies, such as diphtheria toxin, tetanus toxin, typhoid and 
tubercle bacilli, tumor extracts, and placenta extracts by Weichardt; extracts 
of syphilitic livers and serums of syphilitic patients by Seifert, Keidel, and 
Hurwitz.6 
Technic.—The technic of this reaction has been modified from time to 
time. The method here given is essentially the latest given by Weichardt/ 
slightly modified by Keidel and Hurwitz. 
Five constituents enter into the test: 
1. The Antigen.—This is an alcoholic extract of syphilitic liver, prepared 
in exactly the same manner as for performing the Wassermann reaction. 
High dilutions of the antigen, ranging from 1 : 100 to 1 : 10,000, are pre- 
pared with normal salt solution. As in the Wassermann reaction, not every 
antigen is satisfactory, a point that can be determined only by making pre- 
liminary tests. 
2. The patient’s serum should be fresh, unheated, and highly diluted, 
the dilutions ranging from 1 : 100 to 1 : 10,000,000. Usually it is better to 
The Epiphanin Reaction 
i Berl. klin. Wchn., 1908, No. 20; Centralbl. f. Bakteriol., xliii, 143; ibid., xlvii, 39; 
Ztschr. f. Immunitatsf., 1910, vi, 651; Deutsch. med. Wchn., 1911, No. 4, 154. 
2Ztschr. f. Immunitatsf., 1910, vi, 16. 8 Deutsch. med. Wchn., 1911, 22, 1021. 
3 Miinch. med. Wchn., 1910, 38, 1981. 6 Jour. Amer. Med. Assoc., 1912, lix. 1257. 
4 Deutsch. med. Wchn., 1910, 50, 2333. 7 Berl. klin. Wchn., 1911, 43, 1935. THE EPIPHANIN REACTION 
589 
use higher than lower dilutions. When too concentrated solutions of serums 
and of antigen are used, erroneous results are likely to be obtained. 
3. A normal solution of sulphuric acid. 
4. A saturated solution of barium hydroxid made equivalent to the normal 
solution of sulphuric acid. In the use of the barium hydroxid it is im- 
perative to prevent its exposure to the air. A solution that has become 
cloudy, owing to the entrance of carbon dioxid, should not be used. In 
carrying out the test it is best to pour out the amount of barium hydroxid 
needed for the test into a rubber-stoppered bottle or test-tube, so as not to 
contaminate the stock solution. 
5. A 1 per cent, alcoholic solution of phenolphthalein containing 1 per 
cent, of a 10 per cent, solution of strontium chlorid. The strontium chlorid 
has been found to catalyze the reaction. 
The Test.—This is conducted as follows: A number of clean beakers of 
about 50 c.c. capacity are used. For each dilution of the serum a separate 
beaker is required. One beaker is used for an antigen control, and another 
to control the system of barium hydroxid and sulphuric acid. Five beakers 
may be used, Nos. 1, 2, and 3 constituting the main test, No. 4 the antigen 
control, and No. 5 the system control. 
The reagents are added by means of overflow pipets. To each of the 
first four beakers is added 1 c.c. of the dilute antigen to be used in the test 
(about 1 : 10,000). To beaker 5 is added 1 c.c. of the salt solution used in 
making the dilutions of the antigen and the serums. Now 0.1 c.c. of the 
dilute serum to be tested is added to each of the first three beakers, each 
beaker, however, containing the same serum in a different dilution. To 
beaker 5 the same quantity of salt solution is added, but to beaker 4—the 
antigen control—no serum or salt solution is added. 
To each of the five beakers the system of sulphuric acid and barium 
hydroxid and phenolphthalein is now added carefully. First, 2 c.c. of 
the normal sulphuric acid solution are added to each; then 2 c.c. of the 
barium hydroxid, and finally 0.1 c.c. of the phenolphthalein strontium 
chlorid mixture. 
It will be seen that beaker 4—the antigen control—contains all the 
constituents of the test-beakers 1 to 3 except serum. To make beaker 4 
qualitatively as well as quantitatively equal to beakers 1 to 3, 0.1 c.c. of the 
dilute serum (the average of the dilutions of serum which are used in the 
test) is now added to beaker 4, the reaction having already taken place. 
The addition of the sulphuric acid and barium hydroxid requires great 
care. Since the reaction depends on small differences in acidity or alka- 
linity, it is obvious that slight errors will vitiate the results. For the acid 
and the alkali separate pipets are used. After emptying the pipet of its 
content of acid or alkali, the last traces adhering to the inside of the pipet 
are removed by washing. These washings are later added to the beakers 
to which they belong. In filling the pipets with acid or alkali, the latter 
should first be drawn up into the pipet at least once, and then emptied 
again before the pipet is finally filled for delivery into the next test-beaker. 
Only by careful attention to these points in the technic can reliable results 
be obtained. 
Reading the Results.—If beakers 1 to 3 contained the antiserum to the 
antigen used, a positive epiphanin reaction will be obtained, and if the 
barium hydroxid and sulphuric acid were previously carefully adjusted 
to each other, it will be found that beakers 1 to 3 will be lighter than the 
antigen control, beaker 4. The presence of a specific antigen-antibody com- 
bination has shifted the phenolphthalein end-point in the sense of increased 590 
THE RELATION OF COLLOIDS AND LIPOIDS TO IMMUNITY 
H-ion concentration. The exact differences in the alkalinity between beak- 
ers 1 to 3 and beaker 4 can be quantitatively determined by titration with 
n/100 sulphuric acid, and the results expressed as a curve. 
If the antigen and antibody were not specific, the epiphanin reaction 
will be negative. Beakers 1 to 3 will be more alkaline than the antigen 
control, beaker 4, because, as previously pointed out, serums alone or antigen 
with non-specific serums shift the phenolphthalein end-point in the sense of 
increased OH-ion concentration. 
The results may be plotted as curves. The titration values in n/100 
sulphuric acid are placed on the ordinates, and the serum dilutions on the 
abscissae. The positive values are plotted above the line and the negative 
values below the line. No reaction is regarded as positive unless it gives a 
titration value of at least 0.05 c.c. n/100 sulphuric acid. Values below 0.05 
c.c. are easily within the limits of error. 
The following method, employed by Seifert, is much simpler, but is open 
to the error on account of using the antigen and serum in too concentrated 
a state. In a small test-tube place 0.1 c.c. of a 1 : 10 solution of the serum 
in normal salt solution, and add 0.1 c.c. of an alcoholic extract of syphilitic 
liver. To this slowly add 1 c.c. of decinormal sulphuric acid and 1 c.c. of a 
solution of barium hydroxid of the exact concentration needed to neutralize 
the sulphuric acid solution. On the addition of the drop of the phenol- 
phthalein solution the fluid turns red when the serum is from a syphilitic, 
whereas no change in tint occurs with non-syphilitic serum. 
Practical Value.—The reaction appears to be of considerable value in 
the diagnosis of syphilis. With serums and antigen in proper dilutions, the 
results closely parallel those secured by the Wassermann reaction. Keidel 
and Kurwitz report positive reactions with luetic serums in about 75 per 
cent, of their cases. The reaction was found highly specific in that syphilitic 
extracts gave negative reactions with serums of non-syphilitic persons and 
patients suffering from malignant disease. Extracts of normal fetal liver 
and beef heart gave negative reactions with serums of syphilitic persons. 
Positive reactions have also been found in malignant disease, as with the 
antigens of carcinoma and sarcoma. Keidel and Hurwitz obtained 16 posi- 
tive reactions in a series of 24 serums of persons suffering with definite or 
suspected malignant disease. Burmeister did not find the reaction of value 
in cancer. 
The epiphanin reaction has also been used in the diagnosis of pregnancy, 
but sufficient work has not been done to render an expression as to its merits 
of value at this time. 
Among the very large number of immunity reactions employed in 
attempts to secure a diagnostic test for cancer, the “miostagmin reaction” 
of Ascoli and Izar1 is the only one thus far devised that claims the serious 
attention of the clinician. 
Principles.—This reaction is founded on the fact, noted by Ascoli, that 
by the mixing of an antigen and its corresponding antibody there results a 
reduction of the surface tension of the liquid containing these, which may be 
demonstrated by counting the number of drops of the fluid in a given volume 
(usually 1 c.c.), under constant conditions. Normal serum diluted with 
salt solution is first tested, and the number of drops found in a cubic centi- 
meter determined with a specially devised instrument known as Traube’s 
The Miostagmin Reaction 
1 Munch, med. Wchn., 1910, 57, 62, 182, 403. THE MIOSTAGMIN REACTION 
591 
stalagmometer. The antigen is so diluted that when mixed with this normal 
serum it does not increase the number of drops more than 1 in a cubic 
centimeter. When properly diluted patient’s serum and antigen are mixed 
it may be found that the number of drops is increased from 2 to 8 in a cubic 
centimeter. This constitutes a positive reaction. The reaction is appa- 
rently due to the lowering of surface tension, so that more and smaller drops 
are found; hence the term “miostagmin” has been applied to the test, the 
word being derived from the Greek, meaning “small drop.” 
The reaction is said to be sharply specific and very delicate, so that 
antigens diluted up to 1 : 100,000,000 or higher may be detected. The 
technic requires considerable practice and experience or erroneous results 
are quite likely to occur. 
The exact nature of the reaction is not known. The antigens are soluble 
in alcohol, but their nature is obscure. The antibody involved in the 
reaction is referred to as the miostagmin, but its relation to other anti- 
bodies is also unknown. It is probably a physicochemical or colloidal 
reaction, and for this reason it has been placed in this chapter. 
Technic.—The antigen is most difficult to prepare. A recent method 
described by Ascoli is as follows: 
1. Cut non-degenerated portions of malignant tumor (cancer or sar- 
coma) into small pieces and dry in vacuo or spread out in a thin layer on clean 
glass plates and keep at a temperature of 37° C. 
2. Pulverize the dried substance and extract with pure methyl alcohol 
(in the proportion of 5 gm. to 25 c.c.) for twenty-four hours at 50° C. in closed 
vessels, and shake occasionally. 
3. Filter while still hot, and allow the filtrate to cool, and then filter 
again through Schleicher and Schull’s filter-paper No. 590. 
4. It is now necessary to titrate the antigen and to determine in what 
dilution it should be employed. Various dilutions of the antigen are made 
with distilled water, as, e. g., 1 : 10, 1 : 25, 1 : 50, 1 : 100, 1 : 150, 1 : 200, 
etc. A fresh normal serum is diluted 1 : 20 with normal salt solution, and 
9 c.c. of this are mixed with 1 c.c. of the various antigen dilution. Into an- 
other tube place 9 c.c. of the diluted serum and 1 c.c. of distilled water. 
All test-tubes, pipets, and other glassware used must be perfectly dry. 
The tubes are gently shaken and placed in an incubator at 37° C. for 
two hours. The drop number for each fluid is then estimated by Traube’s 
stalagmometer. This instrument is merely a finely and elaborately grad- 
uated pipet with a central bulbous reservoir. The dropping end of the 
instrument ends in a flattened ground base, thus insuring uniformity in the 
size of the drops. The instrument is so graduated that a fraction of a drop 
can be estimated. That antigen is to be chosen that does not alter the drop 
number for normal serum by more than 1 drop in a cubic centimeter—the strong- 
est dilution that fulfils this condition being chosen. 
The Test.—The patient’s serum is diluted 1 : 20 with normal salt solu- 
tion and its drop number determined. Then take two tubes, and into one 
place 9 c.c. of diluted serum plus 1 c.c. of antigen dilution; into the other 
place 9 c.c. of diluted serum plus 1 c.c. of distilled water. A third tube 
may be prepared, which should contain 9 c.c. of normal serum (1 : 20) plus 
1 c.c. of the same antigen dilution. A fourth tube contains 9 c.c. of a known 
positive serum (1 : 20) from a case of cancer and 1 c.c. of the antigen dilution. 
All tubes should be carefully labeled, their drop numbers determined, 
and then placed in an incubator at 37° C. for two hours or in the water- 
bath at 50° C. for one hour. At the end of this time they are removed, al- 
lowed to cool, and the drop number of each is determined. 592 
THE RELATION OF COLLOIDS AND LIPOIDS TO IMMUNITY 
The controls are first examined to show that the antigen has not under- 
gone any change. Variations of the number above 1.5 or 2 drops (as com- 
pared with the control containing distilled water instead of antigen) are regarded 
as positive reactions. The increase in drops is seldom greater than 8. 
Miostagmin Reaction in Cancer 
9 C.c. Serum (1 : 20) Plus 1 C.c. 
Antigen (1 : 200). 
Drops Per C.c. 
After Incuba- 
tion at 37° C. 
for Two Hours. 
Drops Per C.c. 
Controls 9 C.c. 
Serum (1 : 20) 
Plus 1 C.c. 
Distilled Water 
After Incuba- 
tion. 
Results. 
Normal serum 
56.0 
56.4 
Negative. 
Known cancer serum 
62 4 
59 2 
Unknown serum for diagnosis 
61.2 
60.0 
Positive. 
Unknown serum for diagnosis 
Unknown serum for diagnosis 
57.2 
62.4 
57.0 
59.8 
Negative. 
Positive. 
Other Methods for Preparing Antigens.—Various methods for preparing 
antigen and conducting the test are to be found in the literature, and it is 
extremely difficult to arrive at a correct conclusion as to which is the best 
method for preparing antigen. Among these methods for the preparation 
of antigen other than those previously described are the following: 
1. After securing the alcoholic extract described elsewhere evaporate it 
to dryness. Again extract in methyl alcohol and evaporate. Extract with 
warm ether, renewed several times during the course of twenty-four hours. 
Dry, and repeat the extraction several times until the alcohol remains color- 
less. Evaporate the alcoholic and ethereal extracts at 50° and 37° C. 
respectively. A yellowish-red, sticky mass results. Dissolve this in a large 
amount of water-free ether. Filter, and evaporate at room temperature 
until a slight powdery precipitate is deposited. This solution constitutes 
the stock antigen, which, however, may require still further concentration. 
2. A synthetic cancer antigen may be prepared by grinding up 0.5 gm. 
of lecithin (ovolecithin Merck, or lecithin Richter), and extract it with 
50 c.c. of acetone for twenty-four hours at 50° C. Filter through Schleicher 
and SchulPs filter-paper No. 590 until clear. Just before it is to be used it 
should be diluted with water in such amount that 1 c.c. will contain the 
largest amount that does not cause a marked reduction of surface tension 
in normal serum. As a rule, this dilution is between 1 : 50 and 1 : 100 
(Kohler and Luger1). 
3. A syphilis antigen may be prepared by extracting 0.5 gm. of dried and 
powdered syphilitic liver with 50 c.c. of absolute alcohol for two hours at 
37° C. with frequent shaking. Filter, and concentrate to 10 c.c. 
4. A bacterial antigen, as e. g., one of typhoid bacilli, may be prepared 
as follows: Wash off five forty-eight-hour agar cultures of typhoid bacilli 
with 5 c.c. of normal salt solution for each tube. Cover the emulsion with 
toluol, and shake vigorously for several hours. Place in an incubator at 
37° C. for forty-eight hours, and filter through a sterile Berkefeld filter. This 
filtrate may be used as antigen, or it may be used in preparing an alcoholic 
extract in the following manner: To the original aqueous filtrate add 50 c.c. 
of absolute alcohol. Allow the mixture to stand for one-half hour, shake, 
1 Wien. klin. Wchn., 1912, 25, 1114. THE MIOSTACMIN REACTION 
593 
centrifugate, and then mix the sediment with 20 c.c. of absolute alcohol. 
Shake thoroughly once more, and again centrifugate. Combine the two 
extracts, and concentrate on the water-bath to about 20 c.c. 
Practical Value.—This test is quite delicate, and errors due to faulty 
technic are quite likely to creep in. Unless all precautions are rigidly 
observed, the results are worthless. Although an extensive literature has 
accumulated upon this test, the method has not come into general use and 
it is not possible to express an opinion of its value in the study of disease. 
Ascoli and Izar especially have advocated the test in the diagnosis of 
cancer. In 100 cases of malignant tumors they obtained 93 positive reac- 
tions; in 103 cases of other diseases they obtained only one positive reaction. 
Tedesko, Stabilini, Leitch, Kelling, and others have reported favorably upon 
the practical value of the test in the diagnosis of cancer, a good review of 
the literature up to 1911 being given by Bernstein and Simons.1 Burmeister2 
has found that a negative reaction has some value in excluding cancer, and 
is of more value in arriving at a diagnosis than a positive reaction, i. e., it 
has a higher negative than a positive value. Gouwens,3 however, has 
recently reported that the reaction failed with Bacillus paratyphosus im- 
mune rabbit-serum of high titer regardless of the methods employed for 
preparing antigen and conducting the tests. 
The test has also been used in the diagnosis of typhoid fever, para- 
typhoid fever, syphilis, tuberculosis (positive only in active cases), echino- 
coccus disease, etc. Obviously, other methods of diagnosis, such as the 
agglutination reaction and the Wassermann reaction, have superseded this 
test in practical diagnosis. The method possesses, however, considerable 
theoretic interest and is worthy of further investigation. 
1 Amer. Jour. Med. Sci., 1911, 142, 852. 
2 Jour. Infect. Dis., 1913, 12, 459. 
3 Jour. Infect. Dis., 1922, 31, 237. CHAPTER XXVIII 
ALLERGY. ANAPHYLAXIS AND HYPERSENSITIVENESS 
It is generally believed that when an animal previously injected with 
an antigenic substance is subsequently reinjected with the same substance, 
the antibodies induced by the first injection are reinforced, and that a con- 
tinuation of the process of immunization will eventually lead to a high 
degree of immunity. Under certain circumstances, however, this is not the 
case, because severe and even fatal symptoms, as well as other manifesta- 
tions, may set in after the second injection, indicating that, instead of being 
immune, the animal is indeed hypersusceptible or hypersensitive to the 
affects of the antigenic substance. 
Before experimental investigation of this subject was undertaken not 
a few observations were made and described by the early workers in the 
fields of bacteriology and immunity that correspond exactly with the phe- 
nomenon of anaphylaxis, as we understand it today, although the true 
explanation of their unexpected results was not suspected, and they were 
modestly ascribed to faulty technic, embolism, toxicity of the inoculum, etc. 
A typical example is clearly recorded by Flexner1 in 1894: “Animals that 
had withstood one dose of dog-serum would succumb to a second dose given 
after the lapse of some days or weeks, even when this dose was sublethal for 
a control animal.” Rabbits were being employed and the reactions were 
perfect examples of anaphylaxis in these animals to dog-serum. From this 
it followed that the discovery that the experimental injection of such ordi- 
nary innocuous substances as normal serum and milk may produce violent 
symptoms and death gave rise to much surprise and incredulity, since 
scientists had long been accustomed to regard the reaction of an animal 
to an injection as a process of immunization, or diminished sensitiveness, 
instead of one of increased sensitiveness. Here, as Besredka remarked, the 
rules of immunity are “standing on their heads.” 
In this chapter will be presented the known facts regarding allergy and 
the theories that have been advanced in explanation of its nature and 
mechanism, the consideration of allergy in its practical application to 
medicine being left for the following chapter. 
Historic.—The first observation of allergy as it occurs in an infectious 
disease was probably made by Jenner in 1798. This investigator observed 
the sudden appearance of an “efflorescence of a palish red color” about the 
parts where variolous matter had been injected into a woman who had had 
cowpox thirty-one years before. 
In 1839 Magendi found that rabbits that had been injected with egg- 
albumen died after a repetition of the injection, a phenomenon strikingly 
similar to that observed sixty-five years later by Theobald Smith following 
injections of horse-serum. This phenomenon was subsequently studied 
thoroughly by Rosenau and Anderson and Otto. 
While the effects of diphtheria and tetanus antitoxins were being studied, 
peculiar and apparently paradoxic results were occasionally observed during 
immunization of animals with the bacterial toxins. Thus in 1895 Brieger2 
reported the case of a goat that was highly immunized against tetanus and 
1 Medical News, 1894, 65, 116. 
2 Ztschr. f. Hyg., 1895, 101. 
594 HISTORIC 
595 
yet was subject to tetanus. In 1901 von Behring and Kitashima1 reported 
similar findings with diphtheria in horse immunized against that infection. 
At this time it was shown that the results could not be due to the cumulative 
effect of the toxin, and the explanation offered aimed to show that the proc- 
ess was purely histogenetic, and based upon the assumption that receptors 
attached to the body cells had a closer affinity for toxin than the free (anti- 
toxin) receptors in the blood-stream. At the present time toxin hyper- 
susceptibility is held by some to be a true anaphylactic reaction brought 
about by protein substances in the toxin filtrate; others regard von Behring’s 
explanation as satisfactory. Further reference to this subject will be made 
later on in this chapter. 
Richet’s Studies.—The fundamental observations upon which our pres- 
ent knowledge of anaphylaxis is based were made in 1898 by Hericourt and 
Richet.2 These observers found that repeated injections of eel-serum into 
dogs gave rise to an increased susceptibility to this substance, instead of 
immunizing the dogs against the serum. These studies were continued by 
Richet and Portier3 with extracts of the tentacles of certain sea anemones. 
These studies showed that a second injection of the poison into dogs, given 
after an interval of several days, is followed by greater and more intense 
activity than marked the first injection. If the animal survives, however, 
the disease is conquered more readily after the second than after the first 
injection. As previously stated, Richet coined the word “anaphylaxis,” 
meaning “without protection,” and indicating that the first injection de- 
stroyed any natural resistance that the animal might possess against the 
poison (actionocongestin). From these studies he concluded that two dif- 
ferent substances are contained in eel-serum and in the tentacles of actiniens, 
one concerned in establishing an immunity, and the other in calling forth a 
hypersensitiveness; thus far, however, the separate existence of these two 
hypothetic substances has not been proved. 
Arthus Phenomenon.—In 1903 Arthus,4 at the instigation of Richet, 
showed that similar results may be obtained with non-toxic substances, 
like serum and milk. On injecting rabbits at definite intervals with normal 
horse-serum, he found that the first two or three doses were absorbed, 
whereas subsequent injections, given subcutaneously, led to increasingly 
severe local reactions (Arthus phenomenon). If the animals, however, 
were first injected subcutaneously and later intravenously, or intraper- 
itoneally, serious symptoms of dyspnea, convulsions, and diarrhea, and 
even death resulted. At this time the specificity of anaphylaxis was dis- 
covered; as stated by Arthus: “the rabbit sensitized by and for serum is not 
so for milk, and vice versa.” 
Von Pirquet’s Early Studies.—For a long time urticarial eruptions 
were occasionally observed to follow transfusion of blood from lambs and 
other lower animals to persons suffering from anemia and similar condi- 
tions. Soon after diphtheria antitoxin was discovered the medical pro- 
fession was shocked to learn of the sudden death of the healthy child of an 
eminent German professor following a prophylactic injection of the serum. 
In 1902 von Pirquet began the study of these clinical manifestations with 
a child in Escherich’s clinic who, after receiving a second dose of horse- 
1 Berl. klin. Wchn., 1901, xxxviii, 157. 
2 Compt. rend. Soc. de biol., 1898, 53. 
3 Compt. rend. Soc. de biol., 1902, liv, 170; 1903, lv, 246; 1904, lvi, 302; 1905, lviii, 112; 
1907, lxii, 358, 643; 1909, lxvi, 763; 1909, lxvi, 810; 1909, lxvi, 1005. Ann. de l’Inst. Pasteur, 
1907, xxi, 497; 1908, xxii, 465. 
4 Compt. rend. Soc. de biol., 1903, lv, 817; 1906, lx, 1143. Archiv. internat. de physiol., 
1908-09, vii, 472. 596 
ALLERGY. ANAPHYLAXIS AND HYPERSENSITIVENESS 
serum ten days after the first, on the same day developed symptoms of fever 
and a rash. On the basis of this observation von Pirquet1 reached the con- 
clusion that the prevailing views regarding the length of the incubation period 
of an infectious disease could not be correct. He therefore propounded 
the theory that the organism concerned in the etiology of disease calls forth 
symptoms only when it has been altered by antibodies, the period of incuba- 
tion representing the interval necessary for the formation of these anti- 
bodies. 
In conjunction with Schick, von Pirquet endeavored to study all infec- 
tious diseases from the same point of view, especially smallpox, measles, 
recurrent fever, streptococcus infections, and the reactions to cowpox virus, 
tuberculin, and mallein. Later these same observers2 studied the symp- 
toms following injection and reinjection of horse-serum, designating the 
train of symptoms serum sickness. They emphasized that a single injec- 
tion of serum may suffice to bring about the symptoms, and that this im- 
mediate reactivity possesses diagnostic value in so far as it enables us to 
decide whether a previous infection has occurred. How near the astute 
Jenner came to reaching the same conclusion is shown in the following 
abstract from his report in 1798: 
“It is remarkable that variolous matter, when the system is disposed 
to reject it, should excite inflammation on the part to which it is applied 
more speedily than when it produces the smallpox. Indeed, it becomes 
almost a criterion by which we can determine whether the infection will be 
received or not (italics ours). It seems as if a change, which endures through 
life, had been produced in the action, or disposition to action, in the vessels 
of the skin; and it is remarkable, too, that whether this change has been 
effected by the smallpox or the cowpox, that the disposition to sudden 
cuticular inflammation is the same on the application of variolous matter.” 
Von Pirquet at this time proposed the term “allergy,” from ergeia, re- 
activity, and alios, altered, meaning altered energy or a changed reactivity, 
as a clinical conception expressing a truth without binding any one to a theory 
based upon bacteriologic, pathologic, or biologic findings. 
Theobald Smith Phenomenon.—While von Pirquet was making these 
studies, great impetus was given the experimental study of anaphylaxis by 
the observation of Theobald Smith, who found that guinea-pigs that were 
used for standardizing the strength of diphtheria antitoxin after a second 
injection of serum frequently presented symptoms of a serious character, 
such as great restlessness, dyspnea, itching of the skin, and violent con- 
vulsive seizures. In fully 50 per cent, of the animals death occurred within 
half an hour. 
Simultaneously Rosenau and Anderson3 in this country and Otto4 in 
Germany undertook the study of this phenomenon. The first-named in- 
vestigators showed most conclusively, by a thorough series of experiments, 
the action of horse-serum and other substances in guinea-pigs, and proved 
that serum sickness was due to some constituent of the serum independent 
1 Zur Theorie der Infectionskrankheiten (Vorlaufige Mitteilung), April 2, 1903. Zur 
Theorie der Vaksination, Verhandl. d. Gesellsch. f. Kinderh., Kassel, 1903. 
2 Wien. klin. Wchn., 1903, xvi, 758, 1244; 1905, xviii, 531. Die Serumkrankheit, Leipsic, 
Deuticke, 1905; Munch, med. Wchn., 1906, liii, 66. For a full bibliography on this subject 
of allergy up to 1910 see von Pirquet, Archiv. Int. Med., 1911, vii, 259, 383. 
3 Bull. 29, Hyg. Lab., U. S. P. H. and M. H. S., 1906; Bull. 36, Hyg. Lab., April, 1907; 
Jour. Infect. Dis., 1907, iv, 552; Bull. 45, Hyg. Lab., June, 1908; Bull. 50, Hyg. Lab., 1909; 
Jour. Amer. Med. Assoc.. 1906, xlvii, 1007; Archiv. Int. Med., 1909, iii, 519. 
4 von Leuthold Gedenkschrift, 1905, i; Munch, med. Wchn., 1907, liv, 1665; Kolle and 
Wassermann, 1908, ii, 255. CLASSIFICATION OF PHENOMENA 
597 
of the antitoxic antibodies, as normal horse-serum yielded exactly similar 
results. 
Among the earlier studies of anaphylaxis of importance were those of 
Weichardt.1 These were made with extracts of placental cells, and later 
with the proteins of pollen, in relation to hay-fever. Wolff-Eisner2 wrote a 
treatise that had as its fundamental idea the belief that hypersensibility was 
due to endotoxins liberated by a lysin formed as a result of the first injec- 
tion. Also among the earliest and most valuable studies upon the nature 
of anaphylaxis, and showing the important relation of proteins to the proc- 
ess, are those of Vaughan3 and his co-workers; indeed, the studies of Smith, 
Rosenau and Anderson, Vaughan and Wheeler, Gay and Southard, Auer 
and Lewis, and others have gained for America a prominent part in the de- 
velopment of this important subject. 
Practical Importance of Hypersensitiveness.—Since these original dis- 
coveries made a large amount of clinical and laboratory research has 
been devoted to a study of hypersensitiveness in man and the lower animals. 
Developed along with the subject of serum therapy it was but natural 
that most of the earlier work should have dealt with the sensitization of 
guinea-pigs with horse-serum, and curiously enough this combination has 
since proved the most typical example of anaphylaxis. 
A natural result of these discoveries was a closer study of hypersen- 
sitiveness in man so that the tuberculin reaction, food and drug idiosyn- 
crasies were soon placed in the category of anaphylactic phenomena. Not 
a few physicians know in a general way of the fatal character of serum 
anaphylaxis in the guinea-pig and a fear of eliciting similar reactions in man 
by the administration of sera for prophylactic and curative purposes is 
wide-spread among the profession and shared by many of the laity. 
Few subjects in immunology are as important and as perplexing as 
hypersensitiveness. Despite a very large amount of investigation we are 
still ignorant of much that is fundamental, and the introduction of numerous 
theories and new terms and the tendency for drawing conclusions on the 
basis of analogy with observations on the lower animals has tended to con- 
fuse the subject as applied to man. It is extraordinary how the phenomena 
and lesions of hypersensitiveness vary in different species of animals, and the 
subject in a broad and general way is one of great importance in man in rela- 
tion to disease and immunity, serum, vaccine and chemotherapy, and various 
idiosyncrasies to foods, drugs, and other substances. 
Classification of Phenomena.—As previously stated, the word “ana- 
phylaxis” was coined to express a condition of absence of protection and the 
antithesis of prophylaxis and immunity. It has since been widely adopted 
for designating all conditions of hypersensitiveness in man and the lower 
animals. 
Typical anaphylaxis, as occurring in guinea-pigs sensitized with horse- 
serum, is known to be caused by protein antigens interacting with specific 
antibodies which may be passively transferred by injecting the serum of a 
sensitive animal into a normal animal. Coca4 has recently advocated that 
the word “anaphylaxis” be reserved for phenomena of this kind and that 
“allergy” be used for designating all similar phenomena which have not 
been clearly and definitely proved to be antigen-antibody reactions. 
1 Berl. klin. Wchn., 1903, No. 1. 
2Ztschr. f. Bakteriol., 1904; Berl. klin. Wchn., 1904; xli, 1105, 1131, 1156, 1273; 
Munch, med. Wchn., 1906, liii, 217. 
3 Summarized in Protein Split Products, Lea & Febiger, 1913. 
4 Tice’s Practice of Medicine, W. F. Prior Company, New York, 1920, 110. 598 
ALLERGY. ANAPHYLAXIS AND IIYPERSENS1T1VENESS 
Coca’s classification is, therefore, as follows: 
Hypersensitiveness. 
Allergy. 
Anaphylaxis. 
In my opinion Coca’s insistence upon the demonstration of antibodies 
in the serum of a sensitized animal by passive transfer is too rigid; anti- 
bodies may be developed in a sensitized animal, including man, without being 
successfully transferred by injecting the serum into another animal. 
Throughout the study of hypersensitiveness in man and the lower animals 
it would appear that “altered reactivity” of the cells of certain organs and 
tissues expressed as an exaggerated susceptibility to the exciting substance stands 
out as the predominating and constant departure from the normal, and for this 
reason we agree with Doerr on the suitability of adopting von Pirquet’s 
term allergy for designating hypersusceptible states. There is apparently 
so much difference in the nature of the exciting causes of allergy with such 
substances as serum, milk, and egg white at one end and drugs, as quinin 
and mercury, at the other, that in the interests of orderly classification the 
subject of allergy may be divided according to the nature of the exciting 
causes. In our opinion it is doubtful whether we are justified in making the 
division on the basis of being able to demonstrate antibody production by the 
exciting substances. Our technic for discovering antibodies is too imper- 
fect for warranting such a classification; it is possible that antibodies may be 
present in cells when their presence in the blood may not be demonstrable.1 
The proposed classification is, therefore, as follows: 
Allergy. 
A. Hypersentivenessj^P^ 
(Exciting agents are non-protein sub- 
stances) . 
B. Anaphylaxis^^ 
(Exciting agents are proteins or protein 
derivatives). 
Definitions.—Allergy may be defined as a condition of unusual or ex- 
aggerated specific susceptibility to a substance which is harmless in similar 
amounts for the majority of members of the same species. 
The exciting substance may or may not be primarily a protein and may 
or may not engender the production of free antibodies demonstrable in the 
serum. The lesions and symptoms are uniform in any one species of animal 
for different exciting substances, but different from those produced in 
animals of another species and different from the normal or physiologic 
effects of the substances. 
Anaphylaxis may be defined as a state of allergy characterized by “un- 
usual or exaggerated susceptibility of the organism to foreign proteins” 
(Rosenau). In my classification I have made the term embrace protein 
derivative products. For example, Koch’s old tuberculin may be a proteose 
and we regard the tuberculin reaction as an anaphylactic phenomenon. 
Natural anaphylaxis designates a condition in which the time of sen- 
sitization is unknown; practically it covers those cases of immediate reac- 
tion to injections of horse-serum in children or adults who have never re- 
ceived previous injections. It also includes those cases of food intolerance 
encountered in children. 
Acquired anaphylaxis designates the usual type in which sensitization is 
known to have been produced by the administration of a foreign protein. 
1 Handbuch. d. path. org. Kolle-Wassermann, 2d ed., 112, 947. SYMPTOMS AND PATHOLOGY OF ALLERGY 
599 
It embraces the majority of anaphylactic states; however, the time of sensi- 
tization may not be discoverable in anaphylaxis to pollens, bacteria, foods, 
and various effluvia of the lower animals. 
Hypersensitiveness may be defined as a state of allergy characterized by 
unusual or exaggerated susceptibility to non-protein substances as drugs. 
Wolf-Eisner has advanced the theory that these substances (crystalloids) 
may combine with plasma proteins with the production of an altered protein 
sufficiently foreign to prove antigenic. If this is ultimately shown to be 
true despite the negative evidence at present, or, if it is shown that drugs 
excite hypersensitiveness by union with cells with the production of foreign 
intracellular antigens, drug idiosyncrasies or hypersensitiveness may be 
classified as anaphylactic. 
Natural hypersensitiveness designates those cases of intolerance to drugs 
exhibited by individuals who have never before taken the drug. Not in- 
frequently it remains undiscovered until adult age. 
Acquired hypersensitiveness designates those cases of intolerance to a 
drug developing during the course of its administration; apparently one type 
of the arsphenamin reaction is attributable to acquired hypersensitiveness. 
Terminology.—The substance producing the allergic state has been des- 
ignated anaphylactogen by Richet; sensibiligen or sensibilisinogen by Besredka, 
and allergen by von Pirquet and Schick. The term “anaphylactogen” is most 
commonly employed but, in our opinion, should be reserved for designating 
the exciting agents of anaphylaxis. The best general term for embracing all 
states of hypersusceptibility is one carrying the idea of “sensitizing” the body 
cells; for this purpose sensitinogen appears expressive. 
In the artificial production of allergy the first injection of sensitinogen is 
called the sensitizing dose and the process is commonly known as sensitization^ 
The antibody regarded as responsible for the allergic reaction has been 
called toxogen by Richet, sensibilisin by Besredka, sensitizin by Weil, reac- 
tion body by Otto, albuminolysin by Nicolle, and allergin by von Pirquet. 
The term most frequently employed is anaphylactin, but I am uncertain 
regarding who coined this word. 
The dose of sensitinogen (anaphylactogen), bringing on the allergic reac- 
tion, is commonly spoken of as the intoxicating dose. In the typical guinea- 
pig-horse-serum anaphylactic reaction Besredka designates the first injec- 
tion of serum as the “sensitizing injection” and the second injection ten to 
twelve days later as the “toxic or exciting injection.” 
SYMPTOMS AND PATHOLOGY OF ALLERGY 
In the lower animals the allergic reactions which have received most 
study have been those of anaphylaxis induced by the injection of foreign sera, 
egg-albumen, and vegetable proteins in guinea-pigs, rabbits, and dogs. 
The pathologic changes and symptoms of anaphylaxis vary considerably 
in different animals, but in any species are the same for different anaphyl- 
actogens regardless how widely separated in biologic origin or chemical char- 
acters the latter may be. While the same may be true in man this has not 
been proved and for the obvious reason that our data is accumulating more 
slowly. 
Man.—In man the symptoms and pathologic changes of allergy are quite 
varied. Acute fatal anaphylaxis following the administration of horse-serum 
is fortunately rare and such accidents have usually occurred among indi- 
viduals subject to “horse asthma” and sensitive to horse proteins to an 
extreme degree. The usual form of anaphylaxis in man is either an im- 600 
ALLERGY. ANAPHYLAXIS AND HYPERSENSITIVENESS 
mediate syncopal attack following the intravenous injection of serum or the 
development of what has been called “serum sickness” characterized by 
urticarial rashes, joint pains, adenitis, and high fever. 
Allergy to the effluvia of the lower animals, as the dandruff of horses and 
cats, and to pollens is usually expressed by coryza, sneezing, lacrimation, 
and ‘asthma.” Likewise allergy to bacterial proteins derived from infec- 
tions of the respiratory tract are generally expressed as a form of “asthma.” 
Allergy to foods and drugs may be manifest by acute symptoms of 
swelling of the lips and tongue, vomiting, diarrhea, and violent intestinal 
pain; chronic allergy to foods is generally expressed by various skin eruptions, 
of which urticaria and eczema are prominent. 
. With many of these sensitinogens, including the products of certain 
micro-organisms (tuberculin, mallein, etc.), local allergic reactions may be 
produced by application to abrasions of the skin, intracutaneous and sub- 
cutaneous injection. 
In all of the allergic reactions of man the main lesion is vasomotor paral- 
ysis characterized by dilatation of blood-vessels and serous exudation. 
With this brief statement we shall pass to a consideration of anaphylaxis 
in the lower animals, experimentation having given us some insight into the 
mechanism of the process. Allergy in man and the relation it bears to im- 
munity and disease will be discussed again in the following chapter. 
Guinea-pig.—This animal gives the most constant and the most intense 
symptoms. According to Doerr, guinea-pigs are four hundred times as 
sensitive an anaphylactic reagent as the rabbit. 
Horse-serum, when injected into normal guinea-pigs, gives rise to no 
symptoms. As much as 20 c.c. may be injected into the peritoneal cavity, 
and small amounts may even be injected into the brain, without causing any 
untoward symptoms. 
When a small dose of serum is injected intravenously, intraperitoneally, 
or subcutaneously, and ten days later a second injection is made, the animal 
develops symptoms of acute anaphylactic asphyxia, which, in the majority 
of instances, terminates fatally. “In five or ten minutes after injection the 
pig becomes restless and then manifests indications of respiratory embar- 
rassment by scratching at the mouth, coughing, and sometimes of spasmodic, 
rapid, or irregular breathing; the pig becomes agitated, and there is a dis- 
charge of urine and feces. This stage of exhilaration is soon followed by one 
of paresis or complete paralysis, with arrest of breathing. The pig is unable 
to stand, or if it attempts to move, falls upon its side; wdien taken up it is 
limp; spasmodic, jerky, and convulsive movements now supervene. This 
chain of symptoms is very characteristic, although they do not always follow' 
in the order given. Pigs in the state of complete paralysis may fully recover, 
but usually convulsions appear, and are almost invariably a forerunner of 
death. Symptoms appear about ten minutes after the injection has been 
given; occasionally in pigs not very susceptible they are delayed thirty to 
forty-five minutes. Animals developing late symptoms are not very sus- 
ceptible and do not die. Death usually occurs within an hour, and fre- 
quently in less than thirty minutes. If the second injection be made directly 
into the brain or circulation, the symptoms are manifested with explosive 
violence, the animal frequently dying within two or three minutes” 
(Rosenau.) 
Anaphylaxis in the guinea-pig may be prolonged, milder, and finally dis- 
appear, leaving the animal in an apparently normal state. These symptoms 
embrace coughing or sneezing, discharge of urine and feces, restlessness, 
erection of the hair, and scratching. Sometimes acute symptoms develop- SYMPTOMS AND PATHOLOGY OF ALLERGY 
601 
ing immediately after an injection are followed by depression, characterized 
by somnolence, in which the reflexes are maintained. 
H. Pfeiffer1 has shown that a depression of the temperature is a con- 
stant finding in the severer forms of anaphylaxis in the guinea-pig. In 
fatal cases this decrease may be as much as from 7° to 13° C. Some relation 
exists between the extent and the duration of the fall of temperature and the 
severity of the symptoms. During acute anaphylaxis the blood shows a 
leukopenia—a diminution in complement—and, as shown by Friedberger,2 
a delay in or a loss of coagulability. The most striking change observed 
after death is permanent distention of the lungs, resembling emphysema, 
described by Gay and Southard, and particularly by Auer and Lewis.3 The 
lungs do not collapse, but remain fully distended, forming a cast of the 
Fig. 150.—Section of Anaphylactic Lung of Guinea-pig Showing Emphysema; Also Infolding 
of Bronchial Mucosa. 
pleural cavities. The alveoli are distended and, in some instances, the walls 
may be ruptured (Fig. 150). The walls of the secondary and tertiary 
bronchi are contracted, with infoldings of the normally thick mucosa, due to 
contraction of the smooth muscle by peripheral action, death really resulting 
from inspiratory immobilization of the lungs. The heart continues to beat 
long after respiration has ceased, and as shown by the electrocardiographic 
studies of Konigsfeld and Oppenheimer,4 the disturbances of this organ are 
accounted for as occurring in consequence to asphyxia. Rosenau5 and Gay 
and Southard6 have also described minute hemorrhages in various organs 
and mucous membranes. 
1 Wien. klin. Wchn., 1909, 14, 989, 1227, 1375. 
2 Ztschr. f. Immunitatsf., l’orig., 1909-10, 8, 636. 
3 Jour. Exp. Med., 1910, xii, 172. 
4 Ztschr. f. d. ges. exper. med., 1922, 28, 106. 
5 Bull. No. 32 of the Hyg. Lab., 1906. 
6 Jour. Med. Research, 1908, xix, 1, 5, 17. 602 
ALLERGY. ANAPHYLAXIS AND HYPERSENSITIVENESS 
In non-fatal or protracted anaphylactic shock of the guinea-pig the 
lungs are less inflated than in acute shock. Auer, however, has described 
partial inflation and it is not improbable that the symptoms are caused by 
slow asphyxia instead of by liver changes as suggested by Dale. A marked 
fall in body temperature is characteristic and has been advocated by Pfeiffer 
and others as a quantitative criterion of the degree of shock, particularly in 
animals that present few other symptoms and ultimately recover. Accord- 
ing to Friedberger and Mita1 the injection of smaller amounts of antigen 
may result in an elevation of temperature, which is an important observa- 
tion in relation to bacterial anaphylaxis in infection. 
Rabbit.—Reference has been made elsewhere to the pioneer work of 
Arthus, who first described the local anaphylactic reaction about the site 
of subcutaneous injection. He also described objectively the most impor- 
tant symptoms of acme anaphvlactic death in the rabbit, as well as the 
more ordinary type, which ends in recovery. 
In acute and fatal anaphylactic shock in the rabbit Auer2 found slow 
respiration, weak or absent heart action, general prostration, the sudden 
Fig 151.—Anaphylactic Contraction op Excised Sensitized Guinea-pig Uterus (Schultz- 
Dale Method). 
falling of the animal on its side, a short clonic convulsion, increased per- 
istalsis, and expulsion in feces and urine. Immediately after the injection 
the animal remains quiet for a brief period and then suddenly begins to run 
about violently and aimlesslv, when it falls over with the head thrown 
backward and still making running movements. Or this period of initial 
excitement may be lacking and its place taken by a somnolent state inter- 
rupted by muscular twitchings. Finally the animal succumbs with a few 
short gasps and with the eyes in exopthalmos. The blood pressure rises 
at first to be followed by gradual reduction as the heart beat becomes pro- 
gressively slower. Loewit found the initial rise to be of central origin, as it 
is lacking when the spinal cord has been severed. In protracted or mild 
anaphylaxis the animal presents an increased rate of respiration, muscular 
weakness, and after a few days progressive loss of weight. Death is ascribed 
to a vascular or cardiac shock or to a failure of the heart action of peripheral 
origin, mostly affecting the right side, and due to a form of chemical rigor. 
The muscle may be gray, stiff, very tough to the finger-nail, and non- 
1 Ztschr. f. Immunitatsf., 1911, 10, 216. 
2 Jour. Exp. Med., 1911, xiv, 476. SYMPTOMS AND PATHOLOGY OF ALLERGY 
603 
irritable. Further evidence of the importance of heart failure in anaphylaxis 
in the rabbit is furnished by the electrocardiographic study of Auer and 
Robinson.1 Blood coagulability is delayed. 
Anaphylactic shock of the rabbit is, therefore, much different from the 
symptoms and lesions presented by the guinea-pig. Instead of the lungs 
being inflated they are usually collapsed. Coca2 has recently shown that 
there is marked obstruction to the pulmonary circulation apparently due to 
tetanic contraction of the muscular coat of the pulmonary arterioles. It is 
possible that a similar vascular constriction of the vessels in the neighbor- 
hood of a subcutaneous injection of serum following the initial dilation, may 
be responsible for the necrosis of the Arthur reaction. The liver may 
participate in the anaphylactic reaction of the rabbit, being usually the seat 
of congestion, as shown by Weil.3 
Cats.—Anaphylaxis in the cat has been studied especially by Schultz,4 
who observed that cardiac disturbances followed. Horse-serum, however, 
was found markedly toxic in effect, even in the unsensitized animal, as little 
as 0.25 c.c. per kilo killing young, normal cats, and 0.1 c.c. per kilo causing 
a fall in blood-pressure both in the normal and in the sensitized animals. 
Dogs.—In these animals the most constant symptom of anaphylaxis is 
an initial and transitory rise in blood-pressure, followed by a prompt fall of 
from 80 to 100 mm. of mercury. This was first described by Biedl and 
Kraus,5 and subsequently by Eisenbrey and Pearce,6 Robinson and Auer.7 
The general symptoms are not as violent as are those that occur in the 
guinea-pig and death is infrequent. Following intravenous injection of the 
intoxicating dose of serum there may be great restlessness, marked prostra- 
tion, and vomiting, tenesmus, and involuntary discharge of feces and urine. 
If death does not occur, a condition of hemorrhagic inflammation in both 
the large and the small intestine may develop, called by Richet “chronic 
anaphylaxis,” and by Schittenhelm and Weichardt, “enteritis anaphylactia.”* 
Robinson and Auer, by an electrocardiographic study, detected cardiac 
changes consisting of disturbance of the heart impulses, abnormalities in 
ventricular contractions, and other interferences with the mechanism of the 
heart, due probably to the effect of horse-serum on the peripheral cardiac 
tissue, and independent of the drop in blood-pressure or any effect upon 
the central nervous system. The heart changes do not appear to exert a 
primary influence on the blood-pressure, which is due to an effect upon the 
splanchnics, and is probably a secondary factor in anaphylaxis or vascular 
shock of the dog. In many instances there is leukopenia, with loss of mono- 
nuclear cells. Coagulation of the blood is delayed, a condition first described 
by Biedl and Kraus,8 who believed it to be due, probably, to a decrease in 
thromboplastin or an excess of antithrombin. Pepper and Krumbharr9 
have shown that, by adding small amounts of thromboplastin to the non- 
coagulating, post-anaphylactic, oxalated plasma, the coagulability of the 
blood will be restored. 
As shown by Man waring10 and confirmed by the studies of Voegtlin and 
Bernheim,11 Den eke,12 and others, the liver plays an important role in ana- 
1 Jour. Exp. Med., 1913, xviii, 450. 2 Proc. Soc. Biol, and Med., 1919, 16, 47 
3 Jour. Immunology, 1917, 2, 525. 
4 Jour. Phar. and Exper. Therap., 1911-12, 3, 302. 
5 Wien. klin. Wchn., 1909, xxii, 365. 
6 Jour. Pharmacol, and Exper. Therap., 1912, iv, 27. 
7 Jour. Exper. Med., 1913, xviii, 556. 9 Jour. Infect. Dis., 1914, xiv, 476. 
8 Wien. klin. Wchn., 1909, ii, 363. 10Ztschr. f. Immunitatsf., 1910, 8, 1. 
11 Jour. Pharm. and Exper. Therap., 1911, 2, 507. 
12 Ztschr. f. Immunitatsf., 1914, 20, 501. 604 
ALLERGY. ANAPHYLAXIS AND HYPERSENSITIVENESS 
phylaxis in the dog. These investigators have shown that if the liver is 
excluded by Eck fistula or other means anaphylaxis does not develop. The 
predominant change is congestion; Weil1 has calculated that the liver may 
contain as much as 60 per cent, of the blood during shock. Pearce and 
Eisenbrey2 also noted congestion of the liver and abdominal veins and 
doubtless these liver changes are primarily responsible for the reduction in 
blood-pressure and decreased coagulability of the blood. 
Mice and Rats.—White mice and rats, while they may not develop 
acute anaphylactic asphyxia, such as is observed in guinea-pigs, do react 
to horse-serum, as was shown by Schultz and Jordan. This reaction is 
evidenced by restlessness, marked irritability of the skin, involuntary pass- 
age of urine and feces, and temperature and blood-pressure changes. 
Horses; Sheep; Cattle; Hogs.—Hoskins3 has described the following 
symptoms produced in these animals by intravenous injection: 
Horses show uneasiness, an anxious expression, dyspnea, and trembling 
of certain groups of muscles, sweating, colicky pains and retching, tenesmus, 
frequent defecation and micturition, and nasal discharge. 
Sheep show rapid and labored respiration and cyanosis, lacrimation, 
nasal discharge and frothing, protrusion of the tongue and swallowing 
movements, muscular weakness, trembling, and collapse. 
Hogs show similar symptoms of restlessness, muscular weakness, and 
collapse, dyspnea and cyanosis, vomiting, micturition, and defecation. 
Cattle usually show immediate trembling, muscular weakness, and col- 
lapse if the irijection is continued; lacrimation, dribbling, and frequent 
micturition and defecation. 
The pathologic changes do not appear to have been studied with sufficient 
detail. 
THE PATHOLOGIC PHYSIOLOGY OF ALLERGY 
Whereas the lesions and symptoms of anaphylactic shock here described 
in different species of animals are those commonly observed with serum 
proteins, they vary in no essential when any protein agent is used when the 
conditions of dosage and administration are the same. It is evident, how-, 
ever, that no one symptom, or group of symptoms, can be regarded as char- 
acteristic of anaphylaxis in all animals. The various species present widely 
differing pictures with the same protein substance, and these differences 
are best explained on the ground of changes in the anatomic structure and 
physiologic reaction of different animals. Thus, Schultz has shown that 
serum anaphylaxis is essentially a matter of hypersensitization of smooth 
muscle in general, and that, during anaphylactic shock, all smooth muscle 
contracts. In the guinea-pig this effect is most evident in the bronchi, 
owing to the peculiar, though normal, anatomic structure of the mucosa, 
which is relatively thick as compared with the lumen, so that contraction 
of the smooth muscle throws it into folds that completely occlude the bronchi, 
causing death from inspiratory asphyxia. The bronchial mucosa of dogs, 
rabbits, and rats, however, is relatively thin and poor in smooth muscle 
tissue, which may account for an entire absence of transitory respiratory 
difficulties during anaphylactic shock in these animals. In the dog the 
most marked effect is apparent upon the smooth muscle of the gastro- 
intestinal tract, contraction resulting in setting up vigorous intestinal 
peristalsis, vomiting, and involuntary emptying of the urinary bladder. 
1 Jour. Immunology, 1917, 2, 542. 
2 Jour. Infect. Dis., 1910, 7, 565. 
3 Vet. Alumni Quarterly (Ohio University), 1919, 7, No. 1. THE PATHOLOGIC PHYSIOLOGY OF ALLERGY 
605 
The characteristic initial rise in blood-pressure may be due to constriction 
of the splanchnic, pulmonary, coronary, and systemic arteries, followed 
by a condition of paresis and a fall in blood-pressure. The cardiac muscle 
is also involved, particularly on the right side, as shown by Robinson and 
Auer, and this favors a venous accumulation of blood. In the rabbit a 
similar effect is noted upon the smooth muscle of the blood-vessels, and 
particularly on the heart, as well as upon the gastro-intestinal tract. Our 
present knowledge would ascribe these effects, therefore, to a local or per- 
ipheral action of the protein upon smooth muscle, and not primarily on the 
central nervous tissues, as was originally believed. 
The fall in blood-pressure, therefore, appears to be a most constant 
and primary factor. In guinea-pig the heart continues to beat after respira- 
tion ceases, but this phenomenon may be due to mechanical and other 
factors dependent upon the extreme pulmonary emphysema. Fall in blood- 
pressure and congestion of the splanchnic area may produce cerebral anemia, 
and be responsible in some measure for the respiratory disturbances, the 
retching, the involuntary expulsion of urine and feces, the great depression 
and muscular weakness, and the speedy recovery when death does not 
result. 
In man the marked urticarial and other rashes and the inspiratory 
asthma of those peculiarly sensitive to a protein due to a narrowing of the 
bronchi, the latter being analogous to the condition observed in the guinea- 
pig, the diarrhea, and the secondary drop in blood-pressure, all indicate a 
similar action on smooth muscle. This also provides an adequate pharma- 
cologic explanation of the action of atropin, sedatives, and anesthetics in 
alleviating or masking the symptoms of acute anaphylaxis. 
Aside from the severe fall in blood-pressure and temperature, other effects 
of anaphylaxis are leukopenia, local and general eosinophilia (Vaughan,1 
Moschowitz,2 Schlecht and Schwenket3), and reduced coagulability of the 
blood. Pfeiffer4 found poisonous substances in the urine during anaphyl- 
actic intoxication, and Hirschfeld5 detected a pressor substance in the serum 
of intoxicated guinea-pigs. 
The single prominent feature of anaphylaxis in all species of animals is 
the effect upon non-striated muscle. The lesions and symptoms in animals 
of different species are readily explained upon differences in the distribution 
of involuntary muscle. The exciting substance, whatever it may be, but 
commonly designated as “anaphylatoxin,” produces immediate and tran- 
sitory or protracted effects due to contraction which may be followed by 
relaxation. Such structural changes in the cells of various organs as have 
been described are readily explainable on the basis of vascular occlusion or 
dilatation with transudation. 
It may well be that tissues other than non-striated muscle are involved; 
changes in the involuntary muscles have received particular attention, 
because they are so readily detected and graphically recorded by Dale’s 
method. As will be discussed in the mechanism of allergy nothing is known 
of what actually occurs in the cell during the allergic reaction. 
Apparently some changes occur in the antigen participating in the 
anaphylactic reaction, as shown by Manwaring and Crowe,6 who observed 
a loss of toxicity of antigen diffused through the livers of sensitized animals. 
Apparently there is an increase of animo-acids resulting from proteolysis 
in the tissues but, if this occurs at all, it is not large enough to be of sig- 
1 Ztschr. f. Immunitatsf., 1911, 9, 458. 
2 New York Med. Jour., January 7, 1911. 
3 Arch. exp. Path. u. Pharm., 1912, 68, 163. 
4 Ztschr. f. Immunitatsf., 1911, 10, 550. 
8 Ztschr. f. Immunitatsf., 1912, 14, 466. 
6 Jour. Immunology, 1917, 2, 517. 606 
ALLERGY. ANAPHYLAXIS AND HYPERSENSITIVENESS 
nificance. For example, Auer and Van Slyke1 were unable to demonstrate 
an increase of free aminonitrogen in the lungs of anaphylactized guinea- 
pigs and Barger and Dale2 failed to find an increase of non-coagulable nitro- 
gen in the liver. However, in the blood of anaphylactized guinea-pigs there 
may be an increase in non-coagulable and urea nitrogen, but it is not known 
whether this comes from the tissues or from the antigen-antibody reaction 
in the blood. 
ANAPHYLACTOID PHENOMENA 
Criteria of True Anaphylaxis.—In allergy due to protein substances 
(anaphylaxis) the following criteria should be capable of demonstration: 
1. The toxicity of the antigen must depend upon previous sensitization; 
that is, the substance must not produce similar symptoms in non-sensitized 
animals of the same species. 
2. The symptoms and lesions must be those characteristic of anaphyl- 
axis for the species of animal under study. 
3. After recovery from shock there should be exhibited some degree of 
desensitization under proper conditions. 
4. The possibility of the symptoms being caused by thrombosis or 
embolism must be excluded. 
The demonstration of passive sensitization cannot be insisted upon. 
Anaphylactoid Reactions.—-The literature upon anaphylaxis abounds 
with descriptions of the reactions produced in guinea-pigs, rabbits, dogs, 
and other animals, following the intravenous injection of serum that has 
been acted upon by such substances as agar, inulin, kaolin, starch,. etc. 
These investigations were devoted to the production in vitro of poisons 
believed by many to be responsible for anaphylactic shock. These sub- 
stances may be protein in character, decomposition products of proteins, or 
non-protein colloids. The injection of these may result in embolism or 
thrombosis with serious circulatory and respiratory disturbances.. Man- 
waring and Crowre have observed embolic lesions in the lungs and designated 
the reaction as “pseudo-anaphylaxis.” 
The subject has been extensively studied by Karsner and Hanzlik3 with 
thirty colloidal agents and a variety of methods, including intravenous 
injection, histologic studies of perfused organs, protection by atropin and 
epinephrin, as well as test-tube studies on hemolysis and hemagglutination. 
Many of the agents caused serious circulatory and respiratory disturbances, 
but both clinically and pathologically were generally, different from true 
anaphylaxis. These investigators regard anaphylactoid phenomena as of 
a colloidal nature and refer to them as “colloid shock.” Peptone was found 
to produce symptoms and lesions more nearly like true anaphylaxis, but there 
is a greater tendency for thrombosis, hemorrhage, and edema of the lungs. 
The toxicity of the protein decomposition products was found quite similar 
to that of histamin and the effects produced by these substances together 
with the primary toxic action of beef- and eel-sera for guinea-pigs have been 
classed as anaphylactoid. 
Sensitinogens or allergens are substances capable of exciting or inducing a 
condition of specific hypersensitiveness or allergy. It is a general term and 
embraces both the protein and non-protein substances. 
SENSITINOGENS (ALLERGENS) 
1 Jour. Exper. Med., 1913, 18, 210. 
2 Biochem. Jour., 1914, 8, 670. 
3 Jour. Pharmacol, and Exper. Therap., 1919, 14, 229, 379, 425, 449, 463, 479. PROTEIN SENSITINOGENS (ANAPHYLACTOGENS) 
607 
Kinds of Sensitinogens or Allergens.—A large number of substances are 
known to excite allergy or hypersusceptibility in man; the list for the lower 
animals is much smaller because our knowledge of allergy occurring among 
them is practically confined to anaphylaxis induced by the injection of such 
foreign substances as horse-serum, milk, egg-albumen, bacteria, and vege- 
table proteins. Some of the lower animals suffer from hypersensitiveness 
to certain foods and even to drugs, but these subjects do not appear to 
have received much attention. The immediate death of healthy and unused 
laboratory animals (rabbits, dogs, and guinea-pigs), sometimes following the 
first injection of a foreign serum in subtoxic amounts, suggests that examples 
of natural anaphylaxis among them may occur, although such incidents are 
usually designated as anaphylactoid and ascribed to embolism and throm- 
bosis. 
According to the proposed classification of allergy the exciting sub- 
stances are divided into those known to be proteins and those generally 
regarded as non-proteins. The former is the larger group, responsible for 
anaphylaxis to proteins of both animal and plant origin and are designated 
as anaphylactogens. This term is most generally employed. The latter 
embrace the allergic principles of certain plants commonly regarded as 
glucosids and responsible for hypersensitiveness to poison ivy, oak, sumac, 
and other plants; also certain drugs, as the alkaloids of quinin bark, opium, 
belladonna leaves, and the like, and some of the elements, as mercury, iron, 
arsenic, and other substances, which will be discussed later. 
PROTEIN SENSITINOGENS (ANAPHYLACTOGENS) 
The group of proteins known to produce anaphylaxis in man is very 
large and derived from both animal and vegetable sources. 
Practically any foreign soluble protein will produce sensitization and 
intoxication of susceptible animals. Bacterial substances, extracts of plant 
tissues, purified vegetable proteins, and proteins derived from invertebrates 
and cold-blooded vertebrates have all been found capable of acting as ana- 
phylactogens when introduced in a soluble and unaltered condition into an 
animal. 
The proteins concerned must be foreign to the circulating blood of the 
injected animal, but they may be tissue proteins of the same animal— 
e. g.7 syncytial cells—that are not normally present in the blood. Indeed, 
Uhlenhuth and Haendel1 claimed to have sensitized a guinea-pig with the 
dissolved lens of one eye so that it reacted to a subsequent injection of the 
lens of the other eye. Proteins in solution are more active than those in 
suspension or in partial solution, and in general tissue proteins are less 
active than proteins in the blood, lymph, and secretions, but even keratins 
may produce anaphylaxis when dissolved (Krusins2); Uhlenhuth3 has obtained 
positive results with proteins from mummies. As previously stated, the 
altered protein of an animal may be reinjected again into the animal and 
induce an anaphylactic reaction. Recently Richet4 has directed attention 
to this phenomenon, which he calls indirect anaphylaxis, through observing 
an intense leukocytosis in a dog which reached the maximum on the eighth 
day following a second chloroformization. 
In many cases the means or mechanism of sensitization is unknown; in 
1 Ztschr. f. Immunitatsf., 1910, 4, 761. 
2 Arch. f. Augenheilk., Suppl., 1910, 47, 47. 
3 Ztschr. f. Immunitatsf., 1910, 4, 774. 
4 Quoted in Jour. Amer. Med. Assoc., 1914, lxii, 711. 608 
ALLERGY. ANAPHYLAXIS AND HYPERSENSITIVENESS 
some instances the hypersusceptible state appears to be inherited, but the 
majority are apparently acquired. As previously stated, instances of natural 
anaphylaxis in the lower animals to these proteins are almost unknown, 
but some of them have been employed for inducing anaphylaxis experiment- 
ally in guinea-pigs, rabbits, and dogs: 
Zoosensitinogens 
(anaphylactogens) 
1. Alien sera and corpuscles. 
2. Certain foods, as eggs, milk, and meats. 
3. Effluvia as dandruff, and expired organic matter of alien species. 
4. Products of animal parasites. 
Phytosensitinogens 
(anaphylactogens) 
1. Foods of vegetables and plant origin. 
2. Pollens (hay-fever). 
3. Products of bacteria and fungi. 
Serum Anaphylactogens.—Since the pioneer discoveries in anaphylaxis 
were made with horse-serum injected into guinea-pigs and rabbits followed 
up and stimulated by the development of serum therapy in the prophylaxis 
and treatment of disease, it was but natural that most of our knowledge 
of anaphylaxis has been developed by investigations employing serum. 
Curiously enough the combination of horse-serum and guinea-pigs employed 
by Rosenau and Anderson, Otto, and others in a study of “Theobald Smith’s 
phenomenon” has since proved the ideal anaphvlactogen and test animal for 
eliciting the typical anaphylactic reaction. 
Serum of course is a mixture of different simple proteins capable of acting 
as anaphylactogens. These will be discussed shortly in a consideration of 
the chemistry of these sensitinogens. Practically the sera of all warm and 
cold-blooded animals have been found capable of acting as anaphylactogens 
when introduced in a soluble and unaltered condition into animals of dif- 
ferent species. Not infrequently, however, the immediate reactions some- 
times observed following the first intravenous injection of serum are ana- 
phylactoid rather than anaphylactic, due to toxicity of the sera or to em- 
bolism and thrombosis. 
Food Anaphylactogens.—Allergy of some persons to various foods of 
animal and plant origin has been known for many years as “idiosyncrasies.” 
Apparently it has been observed only in human beings, but probably may 
occur among the lower animals as well. The proteins of these foods are 
regarded as the anaphylactogenic substances although in fruits (straw- 
berry) the exciting agent may be a glucosid and lacking in antigenic activ- 
ity. A large variety of foods are now knowm in this connection, notably 
buckwheat, eggs, milk (casein), meats, strawberries, etc. The mechanism 
of sensitization is not definitely known although experiments by Wells and 
others, discussed under Active Sensitization, indicate that absorption may 
take place from the intestinal canal. The lesions and symptoms produced 
are frequently immediate and striking and referable to the gastro-intestinal 
and cutaneous system; on the other hand, the manifestations may be chronic 
and the subject will be discussed in greater detail in the following chapter. 
Hair and Other Zoo-anaphylactogens.—The acute coryza and asthmatic- 
like attacks of some persons brought on by inspiration of the effluvia from 
the skins, hair, and expired air of the horse, cat, dog, rabbit, guinea-pig, 
and other domestic animals, including the feathers of the goose and other 
fowls, are regarded as anaphylactic reactions excited by the proteins con- 
tained in these substances. Here again the phenomenon appears to be en- 
countered only among human beings; at least there does not appear to be 
instances recorded as occurring among the lower animals. 
Pollen Anaphylactogens.—Hay-fever of both the spring and autumnal PROTEIN SENSITINOGENS (ANAPHYLACTOGENS) 
609 
types is commonly regarded as allergic reactions to the proteins of pollens. 
Direct contact with the plant is not necessary for producing a reaction, the 
pollens carried in air currents being sufficient when inspired. Why some 
persons are allergic and the majority are not, and why the allergy appears 
to be confined to the human race are unknown; doubtless a fundamental 
cause is inherited in some instances at least and Coca believes that this is 
always the case. 
In addition to the pollens some persons are susceptible to principles con- 
tained in the leaves of such plants as poison ivy, poison oak, poison sumac, 
and other plants. The lesions are almost solely cutaneous and the exact 
chemical nature of the exciting substances is unknown. At present they 
are regarded as glucosids and for this reason I have classed them among the 
non-protein sensitinogens, but it is possible that traces of protein may be 
present and constitute the sensitinogen. 
Bacterial and Protozoan Anaphylactogens.—Sensitization of both man 
and the lower animals may be brought about by the proteins of various 
bacteria and protozoa, either during infection or by means of active sen- 
sitization with vaccines. The best known examples in this connection are 
sensitizations to tubercle, typhoid, glanders, and contagious abortion bacilli, 
the gonococcus, pneumococcus and meningococcus, Treponema pallidum and 
T. echinococcus. 
Sensitization in bacterial infections appears to be of two kinds. One 
is apparently due to poisons belonging to the proteoses produced only 
by the living bacterium in the tissues and doubtfully in culture-media; 
tuberculin is the best known example. According to Kraus and others 
tuberculin hypersensitiveness is found in man and the lower animals only 
in the presence of active tuberculosis. It is doubtful whether this kind of 
allergy can be induced by injections of Koch’s old tuberculin; the exciting 
agent or anaphylactogen is produced only in the tissues as a secretory prod- 
uct of the bacillus or as a derived protein due to action of bacterial poisons 
upon the cellular or plasma proteins in the focus of infection. 
The second variety of anaphylaxis to bacteria and protozoa appears to 
be due to the proteins (nucleoproteins) of the microparasites which may be 
prepared from artificial cultures. Instances of this kind of anaphylaxis are 
not lacking; they are encountered during active immunization with vaccines 
and are apparently responsible for a type of bronchial asthma due to sen- 
sitization by bacteria upon the mucosa of the upper and lower respiratory 
tract. Probably the earliest clinical observation on anaphylaxis was that 
by Jenner, who noted the occurrence of skin reactions following coWpox 
vaccination among those immune and anaphylactic to the smallpox virus; 
Rosenau and Anderson likewise conducted some of their earliest experiments 
with the proteins of typhoid, colon, and other bacilli, being among the first 
to see the relation between serum and bacterial anaphylaxis. 
Toxin Anaphylactogens.—This brings up the interesting question as to 
whether toxins are anaphylactogens, a subject previously mentioned in the 
historic review of this subject. Instances of hypersensitiveness to diph- 
theria and tetanus toxins were early observed in attempts to immunize 
horses in the production of antitoxins. As it is extremely difficult, if not 
impossible, to isolate a toxin free from other constituents of the medium 
into which it was excreted by micro-organisms, this question cannot be 
answered in a definite manner. There is no direct proof, however, that 
toxins sensitize, although the protein in the toxin filtrate may serve to do so. 
In 1902 Vaughan and Gelston1 showed that the poison contained in the cellu- 
1 Trans. Assoc. Amer. Phys., 1902, 17, 308. 610 
ALLERGY. ANAPHYLAXIS AND HYPERSENSITIVENESS 
lar substance of the diphtheria bacillus is an entirely different one from the 
toxin elaborated by the same micro-organism, results that were confirmed 
in 1911 by Friedberger and Reiter,1 working with the dysentery bacillus. 
Thus hypersensitiveness to toxins is probably not an anaphylactic phenomenon, 
but is due to a greater affinity of the body cells for the toxin. This explains 
the so-called paradox of Kretz, who found that while the injection of an 
accurately neutralized toxin-antitoxin mixture produces no bad results in 
a normal animal, in one that has been previously actively immunized with 
toxin the reverse occurs. Apparently the sessile receptors have a stronger 
affinity for toxin than have the free receptors, and, accordingly, the toxin 
becomes dissociated and combines with the cells. 
This view is also substantiated by the observation that the symptoms 
of intoxication caused by the toxin used for immunization are not those of 
anaphylaxis, which, for a certain animal, are the same regardless of the 
source of the protein. Toxin hypersensitiveness does not seem to be trans- 
missible to normal animals, whereas in anaphylaxis the condition may be 
transmitted (passive anaphylaxis). 
Whether or not endotoxins act as anaphylactogens cannot be definitely 
stated. If they do, their action and effects are intimately connected with 
those ascribed to the protein contained in the bacterial cell. It is unlikely, 
however, that they play any role in inducing hypersusceptibility, as their 
toxicity is usually apparent soon after injection, and before sensitization has 
occurred. 
Physical State of Anaphylactogens.—The results of experiments all 
tend to support the theory that proteins in solution are most powerful in 
producing anaphylaxis, because they are able to come into intimate con- 
tact with body cells, and cell permeation is probably necessary for the most 
complete sensitization. This explains in part the conflicting statements 
concerning the effect of heat on the sensitizing properties of blood-serum. 
Rosenau and Anderson found that animals could not be sensitized with serum 
that has been heated at 100° C., whereas Doerr and Russ placed the point 
at 80° C. Besredka showed that the sensitizing properties are in part at 
least dependent upon the physical condition of the protein, and that heating 
undiluted blood-serum coagulates the protein and leads to a decrease of its 
anaphylactogenic properties. Similarly, Vaughna found that proteins that 
were insoluble in water, as, for example, edestin, sensitize more readily when 
dissolved in salt solution. The same factors are operative with the protein 
used for intoxication, the physical state of the protein substance having a 
direct bearing on the rapidity with which shock is produced. Tempera- 
tures high enough to disrupt and destroy proteins are, however, equally 
destructive to their sensitizing properties. 
There are, however, a few proteins that are not made insoluble by 
heat, and these, except gelatin, have been found by Wells to be antigenic 
despite boiling; examples are casein,2 ovomucoid,3 so-called proteoses of 
seeds,4 and beta-nucleoproteins.8 
Rosenau and Amos6 have demonstrated that proteins in a volatile 
state, as in the exhaled breath of men, when condensed and injected into 
guinea-pigs will sensitize these animals to subsequent injections of human 
‘Ztschr. f. Immunitatsf., 1911, 11, 493. 
2 Jour. Infect. Dis., 1908, 5, 449. 
3 Ibid., 1909. 6, 596; 1911, 9, 147. 
4 Ibid., 1915, 17, 259. 
s Ztschr. f. Immunitatsf., 1913, 19, 599. 
6 Jour. Med. Research, 1911, xxv, 35. PROTEIN SENSITINOGENS (ANAPHYLACTOGENS) 
611 
serum. While it is doubtful if the complex molecule possesses the power 
of passing into the air in a gaseous form, it may probably exist in colloidal 
solution. Rosenau was also able, by keeping guinea-pigs in stables together 
with horses, to sensitize them to horse-serum. These experiments are of 
fundamental importance in explaining instances of human anaphylactic 
phenomena among those sensitive to horse protein, as, e. g., persons seized 
with sneezing and asthma when they come near horses—and also tend to 
show how minute may be the quantity of protein capable of sensitizing and 
intoxicating body cells. 
Chemistry of Protein Anaphylactogens.—The purest known proteins 
act as anaphylactogens or sensitizers; in fact, the purer the protein, the 
more thoroughly it sensitizes the animal and the smaller is the dose neces- 
sary to produce intoxication. The crystallized proteins of hemoglobin, 
egg-albumen, and such pure vegetable proteins as edestin and excelsin are 
powerful sensitizers. According to Wells,1 nothing less than an entire pro- 
tein molecule will suffice to produce anaphylaxis, although Zunz2 claims to 
have observed typical reactions with the proteoses of fibrin, and Abderhalden3 
obtained one with a synthetic polypeptid. It is not necessary, however, for 
a protein, in order to be active, to contain all the known amino-acids of 
proteins, for certain vegetable proteins, e. g., hordein and gliadin, which lack 
one or more amino-acids, such as glycocoll or tryptophane, may produce 
typical reactions. Presumably, the inability of pseudoproteins, such as 
gelatin, to act as anaphylactogens depends upon their deficiency in aromatie 
radicals. 
Wells has obtained negative results with purified nucleoproteins, as well 
as with the isolated components of nucleins, such as histon and nucleic acid. 
The Structural Basis of Anaphylactogenic Activity and Specificity.— 
While, therefore, it is probable, although it has not been definitely proved, 
that nothing less than the entire protein molecule is capable of producing 
the typical reaction, the questions arise whether the whole molecule, or only 
a certain group thereof, determines the specificity, and whether the whole 
molecule, or only a portion, is concerned as the sensitizing agent. It is 
now generally accepted that both the sensitizing and the intoxicating agents 
are one and the same protein, and the older view, which held that in a mixed 
protein substance, such as blood-serum, corpuscles, egg-albumen, etc., one 
protein is present that sensitizes and another that intoxicates, is probably 
erroneous. Besredka, for instance, finds that when a protein used to pro- 
duce intoxication is heated it is less likely to prove fatal, and he concludes 
that proteins contain a thermostabile sensitizing and a thermolabile intoxi- 
cating portion. Doerr and Russ, however, have shown by carefully con- 
ducted experiments, that heat affects both properties of proteins to the 
same degree. Since pure proteins, as, e. g., highly purified edestin, which is 
believed to be a chemical unit, act as exquisite sensitizers and intoxicants, 
it seems reasonable to believe that the sensitizing and poisonous group are 
constituents of the same protein substance. Whether or not both sensitizing 
and intoxicating groups are contained in each single molecule of a pure pro- 
tein is a question that cannot be answered until we can be certain that ab- 
solutely pure proteins are secured to start with, and until our methods of 
effecting its cleavage have been perfected. Vaughan and his co-workers 
have long maintained that a sensitizing non-poisonous and a non-sensitizing 
toxic portion are groups of the same molecule, which they are able to obtain 
1 Jour. Infect. Dis., 1913, xii, 341. 
2 Ztschr. f. Immunitatsf., 1913, 60, 580. 
3Ztschr. physiol. Chem., 1912, 81, 314. 612 
ALLERGY. ANAPHYLAXIS AND HYPERSENSITIVENESS 
in vitro from animal, bacterial, and vegetable proteins by a method of splitting 
with sodium hydroxid in absolute alcohol, as described in the chapter on 
Infection. The toxic intramolecular group is regarded as non-specific, and 
the same for all proteins, which explains the identity of the symptoms of 
anaphylactic shock whatever the protein by which it is induced. The 
non-tbxic sensitizing group, however, is specific, although it may not itself 
be a protein, or at least a biuret body. Whether or not all proteins contain 
a sensitizing group has not been determined. In keeping with his theory of 
the role of the toxic moiety of a split protein molecule in the production of 
disease, Vaughan believes that when proteins are introduced parenterally 
into animals, the non-toxic portion stimulates the body cells to elaborate 
specific ferments, constituting the phase of sensitization, so that when this 
protein is subsequently introduced, digestion rapidly takes place with the 
liberation of the toxic substance responsible for the characteristic symptoms, 
which may terminate in death. 
The correctness of this theory as relating to anaphylaxis has been ques- 
tioned by Coca,1 who believes that the process employed may not result in 
the splitting of all molecules and that mere traces of the original protein 
in the non-toxic residue sufficient for sensitization, but insufficient for intoxi- 
cation or desensitization, may explain the experimental results. Zunz2 
has apparently fallen into the same kind of error in experiments with pro- 
teoses of ox-serum from which he concluded that the amino-acid group re- 
sponsible for sensitization is not the same as that producing shock. In the 
experiments of Gay and Robertson3 upon this question, globin was found 
unable to engender the production of anaphylactic and complement-fixing 
antibodies; a compound of globin caseinate, however, engendered the pro- 
duction of complement-fixing antibody for globin, but did not sensitize to 
globin. Gay and Robertson have interpreted these results as indicating that 
the property of specificity is exercised independently by chemical groups in 
the protein molecules, but their experiments do not indicate that this holds 
good in anaphylaxis. 
The evidence supporting the theory that in a protein substance one part 
acts as anaphylactogen and other as intoxicating agent or producer of 
the anaphylactic reaction, is incomplete; very probably both functions are 
exercised by the same protein molecule. 
The investigations of Wells and Osborne4 in their study of the anaphylac- 
togenic activity of purified vegetable proteins, either chemically related, 
but derived from different sources (gliadin of wheat and rye and hordein or 
barley) or chemically different, but derived from the same source (gliadin 
and glutenin of wheat), has led them to the opinion that specificity is deter- 
mined by the chemical structure of the reacting proteins rather than by 
their biologic origin. They say, however, “Until we have some means 
whereby the chemical individuality of a protein can be established the pos- 
sibility will remain that our so-called pure preparations of proteins consist 
of mixtures, or combinations of proteins which have thus far resisted all 
efforts to separate them.” 
These investigators likewise believe that the entire protein molecule is 
not involved in the specific portion of the anaphylactic reaction, but that 
this is dependent upon certain groups only and that one and the same 
protein molecule may contain two or more such groups. Here again, how- 
ever, the question of the purity of the proteins employed in the experiments 
may be questioned, and the subject requires further investigation. 
1 Tice’s Practice of Medicine, 1920, 113, 114. 
2 Ztschr. f. Immunitatsf., 1912-13, 16, 580. 
3 Jour. Exper. Med., 1913, 17, 535. 
4 Jour. Infect. Dis., 1913, 12, 341. PROTEIN SENSITINOGENS (ANAPIIYLACTOGENS) 
613 
Anaphylactogenic Activity of Protein Derivatives, Racemized Proteins, 
etc.—As previously mentioned, whole proteins possess the greatest degree 
of antigenic activity; this function is greatly reduced by even the earliest 
digestion products. Until recently the proteoses (albuminoses) and peptones 
have been regarded as lacking in antigenic activity; Fink,1 who gives a good 
review of the literature, has found some slight evidences of antigenic activ- 
ity of proteoses, and it is probable that tuberculin anaphylaxis is due to 
proteose sensitization. As mentioned above, Abderhalden2 claims to have 
sensitized with a synthesized polypeptid containing fourteen molecules of 
leucin and glycin, and Zunz3 claims success with simpler polypeptids. A 
large amount of the work conducted with proteoses and peptones (poly- 
peptids) is based, however, upon reactions in the lower animals which may 
have been produced by the toxicity of the substances themselves, that is, 
anaphylactoid reactions, to which reference will be made later in more detail. 
On the other hand, while the evidence of anaphylactogenic activity of proteoses 
and peptones is very slight as based upon experiments with the lower animals, 
it is probable that they may be more active in man, who is more susceptible to 
anaphylaxis, but with different manifestations than those exhibited by ihe lower 
animals. 
The anaphylactogenic activity of proteins is likewise reduced or entirely 
suppressed for guinea-pigs and dogs by other chemical methods and phys- 
ical agents. For example, Wells4 has found acid albuminate antigenic, but 
alkali albuminate non-antigenic; Busson5 found osmic acid to reduce the anti- 
genic activity of proteins. Ultraviolet rays have been found by Doerr 
and Moldovan6 to reduce antigenic activity. Broeck7 found that the 
racemized proteins of Dakin were non-anaphylactogenic. As a general rule 
the modified proteins have shown a progressive weakening in antigenic activ- 
ity, but have preserved biologic specificity; in the investigations of Schitten- 
helm and Stroebel,8 however, with iodized serum and iodized egg-white, it 
was found that while biologic specificity was preserved, the animals also 
exhibited an acquired iodin specificity. We shall discuss this phase in more 
detail in a consideration of the drug sensitinogens. 
Non-protein Anaphylactogens.—As with other immunologic reactions, 
observations have been made that are interpreted as indicating that non- 
protein substances are capable of producing anaphylaxis; thus Pick and 
Yamanouchi9 sought to demonstrate the antigenic properties of alcohol- 
soluble constituents of horse- and beef-serum, but conservatively concluded 
that their results may have been due to a combined action of protein and 
fat combinations. Similar conclusions were also drawn by Uhlenhuth and 
Haendel10 in their study of animal and vegetable oils and fats. Bogomolex11 
is less conservative, and believes that he has succeeded in producing lipoid 
anaphylaxis; these claims, however, could not be confirmed by Thiele and 
Embleton.12 Meyer13 was able to sensitize pigs with pure lipoids extracted 
1 Jour. Infect. Dis., 1919, 25, 97. 
2 Ztschr. f. physiol. Chem., 1912, lxxxi, 315. 
3 Archiv. internat. de physiol., 1919. 15, 179, 192. 
4 Jour. Infect. Dis., 1909, 6, 506. 
6 Ztschr. f. Immunitatsf., 1911-12, 12. 671. 
6 Biochem. Ztschr., 1912, xli, 27. 
7 Jour. Biol. Chem., 1914, 17, 369. 
8 Ztschr. f. exper. Pathol, u. Therap., 1912, 21, 102, 108. 
9 Ztschr. f. Immunitatsf., 1909, 5, 676. 
10 Ztschr. f. Immunitatsf., 1910, iv, 761. 
11 Ztschr. f. Immunitatsf., 1910, v, 121; ibid., 1910, vi, 332. 
12 Ztschr. f. Immunitatsf., 1913, xvi, 160. 
13 Folia Serologica, 1911, vii, 771; Ztschr. f. Immunitatsf., 1914, xxi, 654. 614 
ALLERGY. ANAPHYLAXIS AND HYPERSENSITIVENESS 
from tapeworms, but was unable to intoxicate them with the same extracts, 
results that may be understood, since White and Avery1 have shown that as 
little as 0.0001 milligram of edestin will serve to sensitize a pig, whereas 
larger amounts of protein—more than is contained in Meyer’s preparations 
—are necessary to produce intoxication. Finally, the studies of Wilson2 
and White3 leave no doubt as to the fact that pure lipoids cannot produce 
anaphylaxis. 
It is probable, therefore, that while all anaphylactogens are soluble pro- 
teins, not all soluble proteins are anaphylactogens. Some proteins are 
known definitely to be sensitizers; others are believed with equal sureness 
not to be, while others still are doubtful owing to the difficulties presented 
in their preparation in pure form. It is to be borne in mind, however, that our 
data is mainly based upon the results of experiments conducted with the lower 
animals, and in our opinion it has been so clearly proved that the success or 
failure of active sensitization and the phenomena of anaphylaxis vary according 
to the species of animal, that negative results observed in experiments with guinea- 
pigs or other of the lower animals are not necessarily applicable to man. It 
may be said that man is more susceptible to allergy in a broad sense than 
any other animal. All substances known to produce anaphylaxis in the 
lower animals appear to be anaphylactogenic for man; but the reverse is 
probably not true, that is, substances non-anti genic for the lower animals 
may still be sensitizers for man. 
I have attempted to summarize our information upon the anaphylacto- 
genic activity of various proteins for the lower animals in the following table; 
some of those grouped as non-antigenic for these animals may be antigenic 
for man, and with the simple proteins, primary and secondary protein deriva- 
tives, we believe it is justifiable to regard them anaphylactogenic for man 
until proved otherwise. 
Anaphylactogenic Activity of Proteins for the Lower Animals 
Albumins. 
Globulins. 
Glutelins. 
Prolamines. 
Albuminoids. 
A. Simple proteins ■ 
Mucins and mucoids. 
Hemocyanin. 
Casein; vitellin. 
Lipoproteins. 
Antigenic ■ 
B. Conjugated proteins 
f Acid albumin. 
C. Primary and secondary proteins 
Proteoses (albuminoses) 
[ Peptones (polypeptids) 
Histones. k 
Protamines. 
Gelatin. 
A. Simple proteins 
Apparently non- 
antigenic 
B. Conjugated proteins 
Alpha-nucleoproteins. 
Hemoglobin. 
Globin. 
C. Primary and secondary proteins 
Alkali albumin. 
Crystallizable amino-acids. 
1 Jour. Infect. Dis., 1913, xiii, 103. 
2 Jour. Path, and Bact., 1913, xviii, 163. 
3 Jour. Med. Res., 1914, xxx, 383. 615 
NON-PROTEIN SENSITINOGENS 
Relation of Precipitinogens to Anaphylactogens.—According to Doerr 
and Moldovan1 the same substances in a protein antigen produce precipitins 
and anaphylaxis antibody. Weil2 came to the same conclusion in experi- 
ments showing that precipitin and anaphylaxis antibody are identical, pro- 
posing the name “sensitizin’’ for the latter. Numerous experiments by 
others have shown a close parallelism between precipitin and sensitizin 
production in rabbits and guinea-pigs. While the same antigen appears to 
produce both antibodies, it cannot be said to be proved that the antibodies 
are identical, although most evidence points to this conclusion. For ex- 
ample, Burckhardt3 showed that a serum may lose its power for conferring 
passive anaphylaxis at a time when the precipitin content remains high and 
unchanged; Weil was able to differentiate between the two properties 
of immune serum by heating and other procedures, to which we will refer 
later in the discussion on the nature of the anaphylaxis antibody. 
NON-PROTEIN SENSITINOGENS 
Allergy to non-protein substances presents one of the most perplexing 
problems in immunology. Apparently the phenomenon is not confined to 
the human race. The superpurgation of horses sometimes following the 
administration of an ordinary laxative dose of aloes resembles an allergic 
reaction; likewise the intolerance of cattle to mercury. Auer4 failed to in- 
duce hypersensititiveness in guinea-pigs to salvarsan. When iodin was 
mixed with guinea-pig-serum Friedberger and Ito5 claim to have successfully 
sensitized guinea-pigs, and Swift6 obtained similar results with salvarsan. 
These results support the theory of Wolff-Eisner that drugs may combine 
with the blood or tissue proteins and form a protein sufficiently foreign to 
act as a sensitinogen. On the other hand, hypersensitive persons may react 
so quickly after swallowing the drug to which they are susceptible or the 
drug may produce a local skin reaction so promptly when applied to an 
abrasion, that it would appear that this explanation is inadequate. Further- 
more, as pointed out by Doerr,7 not all compounds are capable of entering into 
chemical combination with proteins, and even when they do, the specificity 
of the original protein appears to be maintained. 
Sensitinogens of this kind from plants (poison ivy, poison oak, poison 
sumac) have been generally regarded by Acree and Syme,8 Adelung,9 and 
others as glucosids; it is possible, however, that traces of protein may be 
present in these as well as in some preparations of quinin, belladonna, and 
the like, although all attempts to sensitize the lower animals with them or 
engender the production of antibodies have failed. 
Practically all of the known non-protein sensitinogens embracing the 
medicaments and non-medicaments, possess toxicity and more or less con- 
stant physiologic activity peculiar for each substance. But the allergic 
effects are different from the toxic effects and quite similar for all. 
The mechanism of sensitization with non-protein sensitinogens is, there- 
fore, still unknown. The process of sensitization and the allergic reaction 
is apparently cellular, as will be discussed later with the theories of the 
1 Ztschr. f. Immunitatsf., 1910, 5, 5, 125 and 161. 
2 Jour. Immunology, 1916, 1, 1. 
3 Ztschr. f. Immunitatsf., 1910, 8, 87. 
4 Jour. Exper. Med., 1911, 14, 497. 
5 Ztschr. f. Immunitatsf., 1912, 12, 241. 
6 Jour. Amer. Med. Assoc., 1912, lix, 1236. 
7 Hand. d. path. org. Kolle-Wassermann, 2d ed., 112, 947. 
8 Jour. Biol. Chem., 1907, 2, 547. 
9 Arch. Int. Med., 1913, 11, 148. 616 
ALLERGY. ANAPHYLAXIS AND HYPERSENSITIVENESS 
mechanism of allergy. Tt is possible that non-protein sensitinogens may 
produce their effects by alteration of cellular proteins of some organs, and notably 
the skin, with the production of a modified protein sufficiently foreign to engender 
the production of cellidar antibodies. At any rate the phenomenon apparently 
results as a reaction in the cells rather than in the body fluids; this altered, 
exaggerated activity toward the exciting substance aptly expressed as 
“allergy” is the characteristic change, although nothing is as yet known of 
the exact nature of this cellular change and how it is produced. 
The list of non-protein sensitinogens is quite large and includes the 
following: 
A. Glucosids of various plants1 as of the Anacardiaceae embracing Rhus 
diversiloba (poison oak), Rhus toxicodendron (poison ivy), Rhus venenata 
(poison sumac), etc. 
B. Alkaloids from opium, quinin bark, belladonna leaves, stramonium, 
hyoscyamus, etc. 
C. Essential oils and balsams (cubebs, copaiba, sandalwood), resins, tur- 
pentine, etc. 
D. Metals, as mercury, arsenic (including salvarsan and neosalvarsan), 
iron, etc. 
E. The halogens, as bromids and iodids. 
F. Coal-tar, benzol, and methane derivatives, as antipyrin, iodoform, 
salicylic acid, creosote, salol, etc. 
ACTIVE SENSITIZATION 
Active sensitization or allergization is the process of introduction of the 
exciting agents or sensitinogens into the body cells. 
The route of administration is usually parenteral, as by subcutaneous, 
intramuscular, intraperitoneal, intravenous, intracardiac, subdural, or intra- 
cerebral injection. Any of these routes of administration may be followed 
for the experimental production of active sensitization of the lower animals; 
probably the intraperitoneal route is mostly employed and the time required 
for bringing about sensitization varies according to the route of adminis- 
tration, as will be shortly discussed. Anaphylaxis of man is commonly 
acquired by the administration of horse antisera for prophylactic and thera- 
peutic purposes; sensitization may result from subcutaneous, intramuscular, 
intravenous, or subdural injections. 
Sensitization of guinea-pigs has also been effected by instillations of horse- 
serum into the conjunctival sac, as reported by Rosenau and Anderson,2 
and by repeated instillations into the nose, as reported by Sewall and Powell.s 
These results are of considerable importance in relation to the sensitization 
of human beings with pollens and other proteins through the mucous mem- 
branes of the respiratory tract. 
Sensitization may also result from the enteral introduction of sensitin- 
ogens, traces apparently being absorbed unchanged through the intact 
mucous membrane of the intestines or, more often, through injured mucosa. 
Wells and Osborne have reported successful sensitizations of guinea-pigs 
by feeding cow’s milk, sera, and egg-albumen, and similar experiments have 
been reported by others. This route of sensitization is of primary impor- 
tance in relation to allergies of foods and drugs encountered among human 
beings. Apparently sensitization by this route is especially apt to occur in 
the young when injuries to the gastro-intestinal tract are most likely to occur. 
1 For a description of plants producing dermatitis venenata consult White’s Dermatitis 
Venenata, Cupples & Hurd, Boston. 1887. 
2 Jour. Med. Research. 1908, 19, 37. 3 Arch. Int. Med., 1915, 16, 605. ACTIVE SENSITIZATION 
617 
The amounts of protein sensitinogens required to effect sensitization may 
be surprisingly small, but varies in some degree with the route of admin- 
istration and the species of animal. Guinea-pigs are apparently the most 
readily sensitized; rabbits, cats, dogs, and human beings require large 
amounts and usually several injections. Sensitization occurs more readily 
upon intracerebral and intravenous injections than upon subcutaneous, and 
the “physiologic” avenues of introduction, as the respiratory and gastro- 
intestinal tracts, are still more uncertain. 
Active Sensitization of Guinea-pigs.—The amount of horse-serum re- 
quired to sensitize guinea-pigs may be as small as 0.000,001 c.c. as reported 
by Rosenau and Anderson.1 Uniform sensitization, however, cannot be 
effected by such amounts. Besredka places the minimum amount neces- 
sary to secure uniform results at 0.001 c.c., whereas 0.0001 c.c. proved in- 
sufficient in a considerable percentage of animals. The sensitizing dose of 
horse-serum ordinarily employed in experiments upon guinea-pigs is 0.01 
c.c. by subcutaneous or intraperitoneal injection; amounts ranging from 
0.001 to 0.1 c.c. are ordinarily followed by an incubation period of ten to 
twenty days. Larger doses also sensitize, but may require a longer period 
of incubation. Wells2 has successfully sensitized guinea-pigs with as little 
as 0.000,001 gm. of crystallized egg-albumen and 0.000,000,1 gm. of edestin. 
The following table gives the average amounts of various protein sensitin- 
ogens for guinea-pigs of about 300 gm. weight: 
Sensitinogen. 
Sensitization. 
Incubation 
Intoxication. 
Route. 
Dose. 
(Days). 
Route. 
Dose. 
Serum 
Subcutaneous. 
0.1 C.C. 
16 
Intravenous. 
0.1 c.c. 
Serum 
Intraperitoneal. 
0.01 c.c. 
10-12 
Intravenous. 
0.1 
c.c. 
Serum 
Intracerebral. 
0.001 c.c. 
8-10 
Intracerebral. 
0.1 
c.c. 
Milk 
Intraperitoneal. 
1.0 c.c. 
20 
Intravenous. 
0.1 
c.c. 
Egg-white 
Intraperitoneal. 
0.01 c.c. 
21 
Intravenous. 
0.05 
c.c. 
Crystallized egg- 
white 
Intraperitoneal. 
0.005 gm. 
21 
Intraperitoneal. 
0.05 
gm. 
Purified vegetable 
protein 
Intraperitoneal. 
0.002 gm. 
21 
Intravenous. 
0.005 
gm. 
Purified vegetable 
protein 
Intraperitoneal. 
0.002 gm. 
.21 
Intraperitoneal. 
0.1 
gm. 
Active Sensitization of Rabbits, Cats, and Rats.—Rabbits are not as read- 
ily sensitized; as a general rule the amounts required are from 1000 to 10,000 
times as much of the same substances as suffice for guinea-pigs. Sensitiza- 
tion is usually affected by intravenous or intraperitoneal injections of 1 c.c. 
of serum per 1000 grams of weight; the former route is preferable. The period 
of incubation is at least ten days and may be four to six weeks. Much 
smaller amounts of serum may, however, suffice as occasionally shown by 
sensitization with traces of serum adhering to corpuscles injected in the 
preparation of immune hemolysins. Coca has given the following method 
for the sensitization of rabbits with horse-serum: “Two injections of 1 or 2 
c.c. are given by the subcutaneous or intraperitoneal route at an interval of 
five days. Three days after the second injection daily intraperitoneal or in- 
travenous injections of 0.2 c.c. are made over a period of two weeks or more, 
1 Hygienic Lab. Bull., 1906, No. 29. 
2 Jour. Infect. Dis., 1908, 5, 449. 618 
ALLERGY. ANAPHYLAXIS AND HYPERSENSITIVENESS 
and five days after the end of this period the test injection of 2 c.c. is made 
into the marginal vein of the ear.” The same method may be employed 
for cats, the amounts being the same; likewise for rats, the amounts being 
about one-tenth less. 
Active Sensitization of Dogs.—Dogs are likewise more difficult to sensi- 
tize than guinea-pigs. The injections may be given subcutaneously or in- 
traperitoneally in dose of 1 c.c. per 100 gm. of weight; the incubation is 
usually three to six weeks and anaphylactic shock is best produced by in- 
travenous injection. For producing anaphylaxis in the dog Weil has ad- 
vised two sensitizing injections of 5 c.c., the first being given subcutaneously, 
the second intravenously after an interval of several days; the test injection 
of 20 to 30 c.c. is given intravenously several weeks later. 
Incubation Period of Active Sensitization.—This refers to the time re- 
quired for sensitization to be effected after the sensitinogen has been intro- 
duced into the body. It varies with the species of animal, the amount of 
antigenic material employed and route of its administration, and method of 
introducing the intoxicating dose of antigen. The shortest period of in- 
cubation observed in the guinea-pig was five days; the average is eight to 
twenty days. In dogs and rabbits from ten to twenty-eight days are gen- 
erally required. 
In human beings the incubation after an injection of serum may be from 
a few hours to sixteen days. In about 20 per cent, of individuals a reaction 
(serum sickness) may appear within three days, but the usual period is eight 
to twelve days. This subject will be discussed in more detail under Serum 
Sickness. 
In drug hypersensitiveness the period of incubation required for sen- 
sitization is generally unknown. Hypersensitiveness to arsphenamin and 
neo-arsphenamin may develop in a few weeks during the administration of 
repeated doses. 
The simultaneous injection of more than one antigen in equal amounts 
does not influence the incubation periods; if, however, one antigen is given 
in relatively large amounts twenty-four hours previous to the usual sen- 
sitizing injection of a second antigen, the incubation period for the latter is 
markedly increased (Coca). 
Duration of Active Sensitization.—Guinea-pigs are likely to remain sen- 
sitized throughout the remainder of their lives, which may be a year or 
more; according to Coca1 and Scott2 rabbits may lose their hypersensitiveness 
in two to three weeks. Dogs are likewise apt to lose the hypersensitive state, 
while human beings are likely to remain hypersensitive to serum, foods, 
pollens, and drugs for years and sometimes throughout life. 
PASSIVE SENSITIZATION 
Passive sensitization consists in transferring the allergic condition to a 
perfectly normal animal by injecting it with the serum from an actively 
sensitized one. 
Historic.—The phenomenon was discovered almost simultaneously and 
independently by several investigators. Nicolle3 passively sensitized a 
normal rabbit with an injection of serum from a sensitized animal and pro- 
duced the Arthus local reaction in the former; Richet4 injected a normal dog 
with the blood of a dog sensitized to mytilocongestin and found that the 
former became sensitive one or two days later. Otto5 and Friedmann8 
1 Virchow’s Archiv., 1909, cxcvi, 92. 
2 Jour. Path, and Bacteriol., 1911, 15, 31. 
3 Ann. de l’lnst. Pasteur, 1907, 21, 128. 
4 Ann. de l’lnst. Pasteur, 1907, 21, 497. 
8 Munch, med. Wchn., 1907, 1665. 
* Miinch. med. Wchn., 1907, 2414. PASSIVE SENSITIZATION 
619 
succeeded in sensitizing normal guinea-pigs with injections of serum from 
sensitized animals. Gay and Southard1 also recorded possible instances of 
passive sensitization in their early experiments by injecting normal guinea- 
pigs with the serum of sensitized ones, and finding the former sensitive about 
ten days later. It is possible, however, that under these conditions the 
serum carried both antigen and anaphylaxis antibody and that the animals 
were actively rather than passively sensitized. 
Maternal Transmission.—Lewis2 showed that the newborn of a sensi- 
tized mother guinea-pig may be sensitive, which is an example of homol- 
ogous passive sensitization. This transmission was found to vary greatly 
in different young pigs and likewise the duration of the inherited sensitive- 
ness. Otto found that young pigs may remain sensitive for as long as forty- 
five days after birth. This function of transmitting the condition of sen- 
sitization is solely maternal; the male takes no part whatever in the trans- 
mission. 
Homologous and Heterologous Passive Sensitization.—Sensitization can 
be passively transferred to individuals of the same species (homologous) or 
to those of different species (heterologous). Doerr has published the follow- 
ing list of positive and negative results in attempted heterologous passive 
sensitization: 
(a) Guinea-pigs have been passively sensitized with injections of serum 
from man, monkey, rabbit, dog, cat, and horse sensitized to various proteins. 
The rabbit with injections of serum from sensitized guinea-pigs and the 
pigeon with the blood of the chicken. 
(b) Negative results have been observed in transferring the serum of 
birds to mammalia, and the reverse, mammalia to birds; likewise attempts 
to transfer the sensitiveness of rabbits and guinea-pigs to white mice have 
failed. 
The best example of passive sensitization is the injection of guinea-pigs with 
the serum of sensitized rabbits, and this combination is recommended for experi- 
mental work in this subject. Rabbits excel other laboratory animals in the 
production of antibodies and guinea-pigs are most satisfactory for passive 
transfer. Guinea-pigs of equal weight that have received identical amounts 
of a sensitizing serum passively acquire an almost identical degree of hyper- 
sensitiveness to the antigen. 
Passive Sensitization wtih Human Serum.—According to Coca allergies 
to serum, drugs, pollens, etc., among human beings cannot be passively 
transferred to guinea-pigs. Coca has termed as true anaphylaxis only such 
reactions as are due to protein antigens and an antibody capable of con- 
ferring passive sensitization. However, this criterion appears to be too 
rigid; even the serum of guinea-pigs highly sensitized to a foreign protein 
may contain too few antibodies to confer passive sensitization to other 
guinea-pigs, and it is possible that the cells of a human being may be highly 
sensitized without there being sufficient free antibodies in the serum to con- 
fer passive sensitization. 
Schultz and Larson3 claim to have passively sensitized guinea-pigs with 
injections of the sera of infants hypersensitive to cow’s milk, and Ramirez4 
has recorded a remarkable instance of accidental passive sensitization of a 
man by transfusion with the blood of an individual anaphylactic to horse 
protein. 
Bruck5 and Klausner6 were the first to claim success in the passive trans- 
1 Jour. med. Research, 1907, 16, 143. 
2 Jour. Exper. Med.; 1908, 10, 1. 
3 Archiv. of Pediatrics, 1918, 35, 705. 
4 Jour. Amer. Med. Assoc., 1919, lxxiii, 984. 
8 Berl. klin. Wchn., 1910, 517. 
6 Munch, med. Wchn., 1910, 1983. 620 
ALLERGY. ANAPHYLAXIS AND HYPERSENSITIVENESS 
fer of drug allergies to guinea-pigs, but these results have been questioned, 
and it is now the consensus of opinion that guinea-pigs cannot be passively 
sensitized with the sera of human beings hypersensitive to drugs and pollens. 
Mechanism of Passive Sensitization.—It is commonly believed that 
following the injection of serum carrying anaphylactic antibodies, that the 
latter combine with the body cells and apparently with those concerned in 
active sensitization. As a general rule a few hours must be allowed for this 
union with cells to occur, but in the rabbit passive sensitization takes place 
almost immediately after the intravenous injection of serum. Upon union 
of the antibodies with cells a similar disturbance in colloidal equilibrium 
apparently results as occurs in active sensitization so that when the antigen 
is suddenly brought into relation with the cells, the allergic reaction results. 
PREVENTION OF SENSITIZATION AND ALLERGY; ANTISENSITIZATION 
Several of the early investigators in allergy sought for means of pre- 
vention of sensitization and allergic shock or intoxication by reason of the 
importance of these subjects in the serum prophylaxis and treatment of 
disease. 
Prevention of Allergic Shock.—Efforts along this line were directed toward 
the neutralization of the toxic substance in horse-serum, considered at this 
time responsible for the allergic reaction to this substance. A large variety 
of substances was employed by Besredka, including alcohol, chloroform, fer- 
ments, and various salts, but with negative results; indeed, toxicity of the 
serum was sometimes increased (serotoxin) and presumably by the auto- 
digestion of serum proteins, following the removal of antiferment by these 
substances, as shown by later researches. Furthermore, it is now definitely 
known that there is no distinct or toxic fraction of serum responsible for the 
production of allergic shock. 
Rosenau and Anderson discovered that boiling horse-serum removed 
its capacity for eliciting the anaphylactic reaction in sensitized guinea-pigs; 
Besredka also found that serum diluted with 3 parts of distilled water 
and heated at 100° C. for twenty minutes was rendered harmless for sen- 
sitized guinea-pigs. Even moderate heating at 56° C. for one hour on each 
of four days in succession was found capable of reducing the shock-pro- 
ducing effects of serum, and this method is practised at the Pasteur Institute 
in the preparation of therapeutic sera. Higher temperatures are not per- 
missible owing to the destruction of antibodies. 
According to Besredka,1 ether and alcohol narcosis protects sensitized 
guinea-pigs against allergic shock; Rosenau and Anderson2 found that these 
substances may mask the reaction, but fail to prevent a fatal outcome. 
Morphin narcosis is without effect. Auer has found the administration of 
atropin of aid in avoiding severe reactions among sensitized guinea-pigs. 
Further discussion of this subject is reserved for a succeeding chapter in a 
consideration of desensitization and prevention of allergy in man. 
Prevention of Sensitization; Antisensitization.—As described by Julian 
Lewis3 if guinea-pigs are injected with large amounts of dog-serum (1 or 2 
c.c.) at the same time or within twenty-four hours of the time of injection 
of the usual sensitizing dose of horse-serum (0.1 c.c.), active sensitization to 
horse-serum either does not occur at all or to a much less degree than occurs 
among animals receiving only the horse-serum. 
1 Ztschr. f. Immunitatsf., 1909. 2, 591. 
2 Ann. de l’Inst. Pasteur, 1908, 22, 465. 
3 Jour. Infect. Dis., 1915, 17, 241. THE PRODUCTION OF ALLERGIC REACTIONS 
621 
Weil1 described instances of resistance to passive sensitization in guinea- 
pigs that had received previous injections of normal rabbit-serum and 
proposed the term “antisensitization” for the process. Coca has sum- 
marized the results of these investigations as follows: 
1. The previous injection of rabbit’s serum obstructs the passive sen- 
sitization with rabbit’s immune serum, but not with homologous (guinea- 
pig’s) immune serum. 
2. The interference is established after an incubation period, which is 
longer (eight days) after small injections (0.1 to 0.5 c.c. repeated), than after 
large injections (1 to 8 c.c.), that is, four days. 
3. The duration of the normal period of heterologous passive sensitiza- 
tion in guinea-pigs is six days. If the guinea-pigs have received 0.1 c.c. of 
normal rabbit’s serum two to eight days previously, the period of passive 
sensitization with heterologous serum is shortened to barely five days. 
This period may be shortened, also, by the injection of 0.6 c.c. of normal 
rabbit’s serum made on the day following the passively sensitizing injection. 
4. The refractory condition of antisensitization persists for at least 
sixty-eight days. 
5. Active sensitization is unaffected by the injection of large amounts of 
normal rabbit’s serum. 
Of considerable importance in relation to this subject of antisensitization 
are the feeding experiments conducted by Wells2 with young guinea-pigs. 
In these investigations it was found that animals raised on oat proteins can- 
not be sensitized to these proteins, but if raised without oats they may be 
sensitized. When fed on egg-albumen they are at first sensitized, but if 
feeding is kept up for a long enough period the animals become refractory and 
cannot be sensitized. Feeding for as long as ten weeks on bread and milk 
rendered the animals sensitive to milk, but did not result in desensitization. 
Weil explained the phenomenon on the basis of the development and 
activity of anti-antibodies; another explanation is to the effect that the 
number of receptors available for anchoring antigen or antibody to body cells 
as the first step toward active or passive immunization, is limited, and that 
the large dose of indifferent serum given before or with the injections of sen- 
sitizing antigen or antibody serum effectually blocks union of these to cells 
which otherwise would become actively or passively sensitized. It is highly 
probable, however, that both of these hypotheses are in error and that the 
phenomenon is one involving a disturbance of intracellular colloidal equilib- 
rium. 
The Production of Allergic Reactions 
The amount of sensitinogen required to produce allergic reactions in- 
cluding anaphylactic shock, may be infinitely small for human beings, but 
in experimental anaphylaxis of guinea-pigs and other of the lower animals, 
usually more is required than suffices for sensitization. Doerr and Russ3 
have stated that in general terms the intoxicating dose is about a thousand 
times greater than the sensitizing dose. The amount required varies, however, 
with the route of administration; with intracerebral, intravenous, and intra- 
cardiac injections the doses sufficing for the production of shock are much 
less than required by intraperitoneal and subcutaneous injections. As a 
general rule the intravenous and intracerebral (Besredka) routes are pre- 
ferred for the production of acute shock. Subcutaneous injections induce 
shock more slowly, but are said to yield more specific reactions. If the 
1 Ztschr. f. Immunitatsf., 1913, 20, 199; 1914, 23, 1. 
2 Jour. Infect. Dis., 1911, 9, 147. 
3 Ztschr. f. Immunitatsf., 1909, 2, 109. 622 
ALLERGY. ANAPHYLAXIS AND HYPERSENSITIVENESS 
sensitinogen is known to be toxic and capable of producing anaphylactoid 
phenomena upon intravenous injection due to intravascular agglutination and 
hemolysis, with thrombosis and embolism, the intraperitoneal route is to be 
preferred. 
Allergic reactions may be elicited by applying the sensitinogen to mucous 
membranes as in the conjunctival tuberculin reaction; likewise the inhalation 
of pollens and the effluvia of the lower animals suffices for inducing attacks 
of hay-fever and allergic asthma in sensitized individuals. The application 
of these substances to abrasions of the skin may elicit local reactions and 
indicate that the amounts of sensitinogens required to elicit local and general 
allergic reactions may be infinitely small. 
Allergic reactions may be elicited by absorption of the exciting substance 
from the intestinal tract, as shown by the food allergies. Indeed, in buck- 
wheat allergy a reaction may be elicited by dropping small amounts on the 
tongue. Apparently absorption does not occur from the stomach and 
reactions are likewise infrequent following rectal injections; reactions are 
likely to follow when the substance has reached the small intestine. 
THEORIES OF THE MECHANISM OF ALLERGY 
Having reviewed our knowledge bearing upon the nature of the sensi- 
tinogens or exciting substances of allergy, the symptoms, lesions, and patho- 
logic physiology of the allergic reaction, we now arrive at the most per- 
plexing portion of the subject, namely, a discussion of the nature of the 
mechanism of allergy. The subject has commanded the attention and 
efforts of numerous investigators and a great mass of data has been ac- 
cumulated, but the exact nature of allergy cannot be said to have been 
elucidated and, as is usual under these conditions, numerous theories have 
been proposed. I shall briefly present some of these theories and then 
proceed to discuss the more important data bearing upon this subject, which, 
I believe, shows that allergy is a cellular colloidal reaction and to be explained 
on a basis of physical chemistry. 
The Humoral Theories of Allergy.—These include the following: 
1. Richet1 held that the sensitizer, or anaphylactogen, contains a sub- 
stance which he called “congestin'’ (because he did his original work with 
extracts of the tentacles of sea anemones, which are toxic and produce 
congestion of the internal organs), and that this generates in the animal 
another substance, known as the “toxogenin.” The reaction between the 
latter and the homologous protein on reinjection sets free a poison, “apo- 
toxin,” which, because of its effect on the nervous system, produces the 
symptoms of anaphylaxis. This theory is practically the same as that 
generally accepted today, except that the antigen is not of necessity primar- 
ily toxic for the animal. 
2. Hamburger and Moro2 suggest that the first injection leads to the 
formation of precipitins, and that on reinjection precipitates are formed; 
these, they contend, may, by the formation of capillary emboli, produce 
acute anaphylaxis, or at least that precipitin formation runs parallel with 
the antibody formation. The symptoms of anaphylaxis, however, are not 
those of embolism, and there is no evidence to show that precipitation occurs 
in vivo, although, as Zinsser points out, precipitins may play the r61e of sen- 
sitizers of the antigen, preparing them for final lysis or cleavage by a com- 
plement. In other words, the precipitin would act as an amboceptor, dif- 
fering, however, from our general conception of the nature of amboceptors 
1 Ann. de l’Inst. Pasteur, 1908, 12, 465. 
2 Wien. klin. Wchn., 1903, 445. THEORIES OF THE MECHANISM OF ALLERGY 
623 
by being active in the absence of complement, unless precipitation is a 
secondary physical phenomenon in the nature of a colloidal reaction. 
3. Besredka1 taught that the sensitizer contains two substances— 
“sensibilisinogen” and aantisensibilisin.” When injected the first time the 
former develops in the body a substance called “sensibilisin,” and on re- 
injection the sensibilisin and antisensibilisin continue to form a poison that 
acts on the nervous system. 
4. Gay and Southard2 have stated that, as a result of the first injection 
of serum, there remains in the circulation a protein substance called “ana- 
phylactin,” which is slowly absorbed and continues to stimulate the cells, 
leading to an abnormal affinity for the homologous protein, which, on re- 
injection, leads to anaphylactic shock. 
5. Vaughan and Wheeler3 are of the opinion that, with the parenteral 
introduction of a foreign protein, the body cells are stimulated to produce 
a specific zymogen or ferment that digests it. The protein of the first injec- 
tion is so slowly digested that the effects are not recognizable. After the 
protein of the first injection has been disposed of, the new ferment continues 
to be formed in the cells, and on the second injection, after the proper inter- 
val has been allowed to elapse, this zymogen is activated and splits up 
the protein, which promptly and abundantly results in the production of 
the symptoms of anaphylactic shock. Vaughan believes that there is a 
non-specific poisonous group or moiety in each protein molecule which, 
when liberated by the ferment, is responsible for anaphylaxis. This poison- 
ous group is held as being the same in all proteins, and hence the similarity 
of lesions and symptoms of anaphylactic intoxication in animals regardless 
of whether the protein is of animal, vegetable, or bacterial origin. The 
nature of the ferment is not clear. In 1907 they regarded it as a zymogen— 
a theoretic labile chemical body resulting from intramolecular rearrange- 
ment in the protein molecules of the cell. Little is known of the action of 
these ferments except that in .some manner they cause cleavage of the pro- 
tein molecule and liberation of the toxic moiety. Later, Vaughan speaks of 
the ferment as consisting of an amboceptor and a complement, the ferment 
(presumably the complement portion) being inactivated by a temperature 
of 56° C. and reactivated on the addition of serum and organic extracts. 
Although Vaughan’s theory best explains the nature and source of the 
anaphylactic poison, that of Friedberger explains the production of the 
“ferment,” or rather the protein sensitizer (amboceptor), which, with a 
complement, digests the protein and sets free or produces the protein poison. 
6. Friedberger* has attempted to explain anaphylaxis on the basis of 
Ehrlich’s side-chain theory of the action of antigens and the production of 
antibodies similar to toxin-antitoxin immunity. This theory assumes that, 
on the first injection, the protein finds but few groups of cellular receptors 
with which it can combine, and for this reason it is not poisonous. During 
the period of incubation the animal cells develop receptors specific for the 
homologous protein; with a single small dose of protein most of these re- 
ceptors remain attached to the cell (sessile); on repeated injections, the newly 
formed receptors are in large part cast off into the blood, and constitute 
the precipitins. In this manner an animal relatively insusceptible to a foreign 
protein is rendered highly susceptible, and on the second injection the 
protein is anchored firmly to the cell, just as the cells of an animal may 
1 Compt. rend. Soc. de biol., 1907, lxiii, 294; 1909, 21, 384. 
2 Jour. Med. Research, 1907, 16, 143; 1908, 18, 407; 1908, 19, 1, 5, 17. 
3 Jour. Infect. Dis., 1907, 4, 476. 
4Ztschr. f. Immunitatsf., 1909, 2, 208; 1909, 3, 692; 1910, 4, 636. 624 
ALLERGY. ANAPHYLAXIS AND HYPERSENSITIVENESS 
anchor diphtheria toxin. One of the essential features of this theory is that 
it assumes that, ordinarily, the receptors are not preformed in sufficient 
numbers to anchor enough protein to injure the animal with the first injec- 
tion, regardless of the size of the dose. On the other hand, tetanus and 
diphtheria toxins find large numbers of sessile or cellular receptors, and are 
highly toxic on the first injection. As Qriginally evolved, the theory did not 
explain the nature of the toxic agent responsible for the lesions and symp- 
toms of anaphylaxis, and made no mention of the protein poison. Never- 
theless it affords the best explanation we have on the formation of the 
“ferment” or protein sensitizer (amboceptor). At one time Friedberger 
believed that anaphylaxis could be explained on the basis of a precipitin 
reaction. Anti-anaphylaxis was explained on the assumption that the 
protein of the reinjection uses up the sessile receptors already developed, 
and, accordingly, not enough are present at the end of the period of incuba- 
tion to prod\*ce anaphylactic shock. Passive anaphylaxis was explained on 
the ground that the free receptors in the blood of a sensitized animal be- 
come, on injection into a fresh animal, anchored to the cells, thus forming 
fixed or sessile receptors that anchor the protein on reinjection and lead to 
anaphylaxis. 
Nolf1 has proposed a theory of anaphylaxis that has come to be known 
as the “physical theory.” It assumes that the active constituent of proteins 
is a thromboplastic substance that disturbs the colloidal equilibrium of the 
blood and leads to the deposition, on the surface of the leukocytes and the 
endothelial cells of capillaries, of a delicate film of fibrin. Thus stimulated, 
the cells pour out an unusual amount of antithrombin. On account of the 
consumption of a part of the fibrinogen and the increased formation of anti- 
thrombin the blood fails to coagulate after anaphylactic shock or peptone 
poisoning. Owing to the coagulation deposits on the endothelial cells, the 
viscidity is increased and the leukocytes adhere to the vessel walls, thus 
accounting for the leukopenia observed after protein injection. The endo- 
thelial cells are injured, and the walls of the capillaries become more readily 
permeable, thus accounting for the local edema often seen in anaphylaxis. 
The fine capillaries of a given area may be occluded by thrombi, thus ex- 
plaining the necrosis characteristic of the Arthus phenomenon. The irrita- 
tion of the endothelial cells extends to the smooth muscle, leading to vaso- 
paralysis and the characteristic fall in blood-pressure. The affinity of the 
endothelial cells for the protein is stimulated by the first injection, and acts 
in a fulminating way on reinjection, thus explaining the suddenness of ana- 
phylactic shock. 
The theory, therefore, also assumes the formation of a ferment that 
acts primarily upon the proteins of the blood, leading to the formation of 
fibrin, which, as it were, mechanically induces the lesions and symptoms of 
anaphylaxis. While it offers a plausible explanation, the theory is not well 
supported, and at best may be regarded as a modification of Vaughan’s 
theory, demonstrating one way in which the protein poison may act. 
The Cellular Theory of Allergy.—Many investigators have attempted to 
explain anaphylaxis on the basis that the reaction is a cellular one, that is, 
that the antibody is within the cell and that the antigen-antibody reaction 
occurs in this position rather than in the blood-stream by means of free or 
circulating antibody and antigen. According to the “cellular theory,” if 
the serum of an immunized animal containing the anaphylactic antibody is 
injected into a normal animal (passive anaphylaxis) and is followed by an 
injection of the antigen, an anaphylactic reaction cannot occur before the 
1 Arch, internat. de physiol., 1910, 10, 37; Bull, de l’Acad. roy. de Belg., 1910. ANALYSIS OF DATA BEARING UPON THE MECHANISM OF ALLERGY 
625 
elapse of sufficient time for the antibody to become anchored to cells. The 
“humoral theory,” on the other hand, assumes that the antigen meets the 
antibody in the blood-stream and explains the time required between the 
injection of immune serum and antigen in passive anaphylaxis as due to a 
failure of rapid union between antigen and antibody unless qualitative rela- 
tions between the two are accidentally correct. 
The early theory of Friedberger,1 explaining anaphylaxis on the basis 
of “sessile receptors”; the experiments of Friedberger and Girgolaff,2 who 
passively sensitized normal animals by transplanting the thoroughly washed 
organs of a sensitized animal; the transfusion experiments of Pearce and 
Eisenbrev,3 who transferred the blood of a sensitized animal to a normal 
animal and the blood of a normal animal to a sensitized one, finding that 
the latter, but not the former, reacted when the antigen was injected as 
soon as the transfusions were completed; the work of Coca,4 who found 
that sensitized guinea-pigs would still react after being thoroughly bled and 
perfused writh salt solution; the investigations of Schultz,5 Dale,6 and par- 
ticularly of Weil,' showing that the excised and washed muscles of sensitized 
animals would react in vitro in a bath of Ringer’s solution when the antigen 
was added (Fig. 151), support most strongly the cellular theory of ana- 
phylaxis as emphasized also in the work and recent communications of 
Doerr.8 
In the opinion of Weil, Doerr, Bayliss, Coca, and others the “cellular” 
theory is the only tenable one today. According to this theory of ana- 
phylaxis, the antibody is in or on the body cells; upon union with the antigen 
the cells undergo a physical shock which has been likened to an electric 
shock, and this constitutes the basis of the anaphylactic reaction without 
the formation of any intermediate or chemical poison. 
As emphasized by Weil6 there is no direct evidence of the production 
of a chemical poison or anaphylatoxin during or after an anaphylactic reac- 
tion in the living animal. This poison has not been satisfactorily demon- 
strated in the blood and in animals recovering from a general anaphylactic 
reaction, the phenomena being more suggestive of a transitory “shock” 
than of an intoxication with a chemical poison. (See pages 632 and 633). 
The exciting agents or sensitinogens have been discussed and grouped 
into two broad classes: (a) the proteins embracing not only the whole mole- 
cule, but the primary and secondary derivatives down to and including the 
proteoses; (b) the non-protein agents. Some of the agents grouped in the 
latter, as the glucosids and drugs of vegetable and plant origin, may pos- 
sibly owe their sensitizing properties to traces of protein; others, and espe- 
cially synthetic compounds, are certainly not proteins, but whether or not 
they are capable of forming combinations with body proteins and altering 
these sufficiently to act as sensitinogens is still an open question. At any 
rate an acceptable explanation of the mechanism of allergy must, in the 
writer’s opinion, cover both kinds of sensitinogens. 
ANALYSIS OF DATA BEARING UPON THE MECHANISM OF ALLERGY 
1 Ztschr. f. Immunitatsf., orig., 1909, 11, 208. 
2 Ibid., 1911, ix, 575. 
3 Congr. Amer. Phys. and Surg., 1910, viii, 402. 
4 Ztschr. f. Immunitatsf., orig., 1914, xx, 622. 
5 Jour. Pharmacol, and Exper. Therap., 1910, 1, 549. 
6 Jour. Pharmacol, and Exper. Therap., 1913, iv, 167. 
7 Jour. Med. Research, 1914, xxx, 87. 
8 Ergebnisse der Immunitatsf., Reichhardt, Berlin, 1914, 1, 257. 
9 Proc. Soc. Exper. Biol, and Med., 1915, xiii, 23 (1087). 626 
ALLERGY. ANAPHYLAXIS AND 11YPERSENSITI YEN ESS 
The Relation of Antibodies to Allergy; Terminology.—Is an antibody 
concerned in the mechanism of the allergic reaction? The demonstration 
that sensitization may be passively transferred to animals of the same or 
different species (passive allergy), answers this question affirmatively. 
Failure to demonstrate passive sensitization may be due to technical errors 
or more especially to the absence of antibody from the blood or serum at 
the time of transfer or the presence of insufficient amounts, but does not 
necessarily indicate that antibodies are absent as they may be attached to 
the body cells. This must be true if we accept the contraction of the thor- 
oughly perfused sensitized uterus or lung of the guinea-pig in vitro as an al- 
lergic reaction. Furthermore, hypersensitiveness may persist for long periods 
of time, whereas antibodies in the serum tend rapidly to disappear; for this 
reason and because of frequent failure to demonstrate antibodies by the 
precipitin and complement-fixation reactions, Gay and Southard do not 
regard allergy as an antigen-antibody reaction. In other words, if there is 
an allergic antibody then the experiments of the cellularists show that it 
may be in part, at least, in certain cells or attached to them. For this reason 
failure of passive sensitization or detection of antibodies in vitro cannot be 
accepted as evidence against a phenomenon being anaphylactic. 
Various names have been proposed for this antibody. Richet named it 
toxogen, believing it responsible for the production of a poison. Otto spoke 
of it as reaction body, Nicolle as an albuminolysin, Besredka as sensibilisin, 
von Pirquet as alter gin, and Weil as sensibilisin. It is most commonly 
designated as anaphylactin. In my opinion the terms allergin or sen- 
sitizin are most appropriate because they convey the fundamental meaning 
of altered and exaggerated activity of cells and sensitization. 
This antibody is regarded as a precipitin by some investigators, as a 
“ferment” by others, and by others still as an albuminolysin or proteolysin, 
that is, a protein amboceptor requiring the co-operation of a complement. 
The Relation of Precipitin to Allergin (Sensitizin).—Several investigators, 
as Friedberger, Doerr, and Russ, Weil, Wells, and others, have shown that 
sensitization of guinea-pigs and other of the lower animals with various 
proteins of animal and vegetable origin is followed by the appearance of 
precipitins in the blood, together with the capacity for conferring passive 
sensitization. Although Weil found that heating immune serum at 70° C. 
for thirty minutes destroys its precipitating power while its capacity for 
passive sensitization persists, he did not interpret this as evidence of a 
separation of precipitin and sensitizin, but that the precipitoids so formed 
(thermostabile haptophore portions only) are essential to passive sensitiza- 
tion. As previously discussed in the chapter on Precipitins, Weil found that 
the precipitates formed by mixing immune serum and antigen served for 
active and passive sensitization, that is, the precipitate carries down both 
whole antigen capable of active sensitization and precipitin, regarded as the 
passively sensitizing agent. 
In serum allergy of human beings usually caused by injections of horse- 
serum precipitins may be found in the serum, although passive transfer of 
this kind of allergy has generally failed. Longcope and Rackeman1 believe 
that precipitins are always present in those subject to this allergic reaction 
and that the symptoms and lesions occur until all circulating antigen has 
been neutralized. C. W. Wells,2 however, in agreement with von Pirquet 
and Schick, was unable to show a constant coincidence of the onset of symp- 
toms of serum allergy with either the appearance of precipitin in the blood, 
its concentration, or disappearance. 
1 Jour. Exper. Med., 1918, 27, 341. 
2 Jour. Infect. Dis., 1915, 16, 63. ANALYSIS OF DATA BEARING UPON THE MECHANISM OF ALLERGY 
627 
With protein sensitinogens it would appear, therefore, that precipitin and 
sensitizin are closely associated even though they cannot be regarded as 
identical; especially is this true of anaphylaxis in the guinea-pig and rabbit. 
In serum allergy of human beings precipitin may be found in the serum, 
while passive sensitization fails, indicating in a general way that sensitizin 
is absent and the two antibodies separate and distinct. In the other aller- 
gies of human beings due to protein and non-protein sensitinogens, pre- 
cipitins have not been found and passive sensitization has uniformly failed. 
Finally, Longcope1 has recently found that in spite of the fact that the 
white rat could not be made anaphylatic to horse-serum, the tissues of 
these animals reacted with horse-serum to form precipitins in fair con- 
centration, and the antigens disappeared from the circulation soon after the 
precipitins reached their greatest concentration in the blood. These ex- 
periments indicated, therefore, that in the white rat anaphylaxis and pre- 
cipitin formation are independent and represent different types of im- 
munologic processes. 
The Relation of Albuminolysins or Proteolysins (Amboceptors and 
Complement) to Allergin (Sensitizin).—Friedmann2 observed the develop- 
ment of toxic substances in normal rabbit-serum brought in contact with 
sensitized ox-blood corpuscles before hemolysis occurred, which produced 
anaphylactiform symptoms upon intravenous injection into guinea-pigs. 
Since then Friedberger3 and others have sought to explain the anaphylactic 
reaction on the basis of the production of a poison designated as “ana- 
phylatoxin” from the antigen by the action of protein amboceptors (albu- 
minolysins or proteolysins) and a ferment (complement). This work is 
offered as one explanation for the humoral theory of anaphylaxis, the other 
explanation being being based upon a similar production of a protein poison 
by the activity of serum ferments (proteoses). 
If precipitins are identical with protein amboceptors, as maintained by 
Zinsser and others, the role of proteolysins in the production of anaphylactic 
shock is strengthened, but several objections have been raised against this 
theory of the mechanism: 
1. Weil4 and others have failed to demonstrate the poison in the blood of 
animals during or immediately after anaphylactic shock. 
2. Loewit and Bayer5 are said to have produced anaphylactic shock in 
animals previously deprived of complement by the injection of hypertonic 
saline solution (Hektoen); however, intracellular complement may have been 
available and the work requires more confirmation before being acceptable. 
3. The proof of the ferment activity of complement is almost totally 
lacking. 
4. If this theory is correct, the simultaneous injection of a normal animal 
with sensitinogen and the serum of a sensitized animal carrying sensitizin 
should result in immediate symptoms because the anaphylactic reaction 
may develop within a few seconds. As a matter of fact, such results have 
been claimed by Biedl and Kraus6 in guinea-pigs when they injected intra- 
venously mixtures of horse-serum and the serum of sensitized guinea-pigs; 
Briot7 and Gurd8 also obtained similar results, but such experiments have 
failed or yielded indefinite results in the hands of others. 
1 Jour. Exper. Med., 1922, 36, 627. 2 Jour. Immunology, 1917, 2, 399. 
3 Arch. f. exper. Path. u. Pharmacol., Suppl. Bd., 1908, 355. 
4 Jour. Immunology, 1917, 2, 399. 
5 Arch. f. exper. Path. u. Pharmacol., Suppl. Bd., 1908, 355. 
6Ztschr. f. Immunitatsf., 1910, 4, 115. 
7 Compt. rend. Soc. de biol., 1910, lxviii, 402. 
8 Jour. Med. Research, 1914, 31, 205. 628 
ALLERGY. ANAPHYLAXIS AND HYPERSENSITIVENESS 
5. A similar poison has been produced in vitro by Doerr and Russ1 and 
numerous other investigators by sera deprived of complement. 
6. Many of the experiments offered in confirmation of this theory have 
been based entirely upon the production in guinea-pigs and rabbits of symp- 
toms resembling anaphylaxis without excluding the possible effects of em- 
bolism and thrombosis capable of exciting anaphylactoid reactions. 
7. The immediate contraction of perfused sensitized guinea-pig uterus 
in vitro by the Schultz-Dale method, when antigen is added to the bath, 
cannot be reconciled with this theory that the exciting agent is a product 
of proteolysis. 
8. Similar poisons have been produced in vitro by normal serum and inert 
substances and in the absence of antibodies and antigen. Friedberger claims 
that under these conditions natural proteolysins (amboceptors) and traces of 
protein in the substitute antigen may be operative, but these claims cannot 
have a basis in fact with such a substance as kaolin. This is discussed in 
more detail in the following paragraphs. 
The Relation of Ferments to Allergin (Sensitizin).—The word “fer- 
ment” has been loosely employed in connection with immunologic phe- 
nomena in general and with anaphylaxis in particular. Vaughan and 
Wheeler believe that in anaphylaxis the first injection of antigen engenders 
the production of a “ferment” capable of digesting the antigen upon sub- 
sequent injection, with the production of a poison similar in effects to that 
produced in vitro by splitting with alcoholic potash. As mentioned above 
Friedberger and his school speak of the “ferments” in normal and immune 
sera and apparently refer to thermolabile complement and protein ambo- 
ceptors or sensitizers. 
Jobling and Petersen speak of a proteolytic ferment in serum as protease 
and this is the meaning adopted in this discussion. These investigators2 
have produced protein poisons in vitro similar to Friedberger’s anaphyla- 
toxin by chemical methods and in such manner as apparently rules out the 
influence of the thermolabile and easily destroyed complement or alexin. 
According to their views, serum complement and normal serum proteases 
are not identical and, whereas the former may be inactivated by heating at 
56° C. for half an hour, the latter are more resistant, the increased digestive 
power observed following the addition of fresh serum being ascribed to the 
amounts of protease thereby added. 
Therefore, while the role of complement in the production of poisons in 
serum may be ruled out, the question remains whether or not anaphylaxis 
is caused by an increase of proteases and if the latter constitute the sen- 
sitizins capable of conferring passive immunization? A great deal of work 
has been done on these problems, but very little of it is of use in the present 
discussion. For example, anaphylaxis is highly specific and the poison 
should be correspondingly specific, that is, its source should be the protein 
of the sensitinogen, although its properties may be the same irrespective of 
source since the phenomena of anaphylaxis in any given species are the 
same for different sensitinogens. But the experiments of Keysser and 
Wassermann,3 Bordet,4 Jobling and Petersen,5 Plaut,6 Peiper,7 Friedmann 
and Schonfeld,8 Bronfenbrenner,9 and others, tend to show that the mechan- 
1 Centralbl. f. Bakteriol., 1912, lxiii, 243. 
2 Jour. Exper. Med., 1914, 19, 459, 479. 
3 Folia Serolog., 1911, 7, 593. 
4 Compt. rend. Soc. de biol., 1913, lxxiv, 877. 
Loc. cit 
6 Munch, med. Wchn., 1914, lxi, 238. 
7 Deutsch. med. Wchn., 1914, xl, 1467. 
8 Berl. klin. Wchn., 1914, li, 348. 
9 Jour. Lab. and Clin. Med., 1916, 1, 573. ANALYSIS OF DATA BEARING UPON THE MECHANISM OF ALLERGY 
629 
ism is not quite so simple and direct, and that a protein poison may be 
produced in the absence of the specific antigen. They have demonstrated 
by experiments in vitro that such inert substances as kaolin, barium sulphate, 
agar or indifferent bacteria or precipitates may replace the antigen in the 
production of a protein poison. Since the presence of protein substances 
could be excluded, as with such inorganic substances as kaolin and barium 
sulphate, the conclusion was naturally drawn that the matrix of the poison 
was not the antigen or substrate, but the constituents of the serum itself. 
Jobling and Petersen advanced the theory that the proteolytic ferment wras 
held in check by antiferments (largely the unsaturated fatty acids in serum), 
and that these substances, as kaolin or barium sulphate, absorb the anti- 
ferments and thereby release the proteolytic ferments which proceed to 
digest the protein of the serum. For this reason these investigators have 
applied the name “serotoxin” to the protein poison as indicating its source. 
A very extensive and valuable investigation by Novy and De Kruif1 
have corroborated and extended these observations and shown, in addition, 
that informed substances, as peptone solutions and water and even mere 
coagulation of the blood, may render a serum toxic. 
Applying these observations and views to the subject under discussion, 
it wmuld appear necessary to infer that the antigen is non-specific and acts 
as it were in a purely mechanical manner; this does not at all agree with the 
great mass of data showing the highly specific nature of anaphylaxis, and 
Bronfenbrenner explains the specific production of anaphylatoxin by non- 
specific ferments as follows: 
Specific antibodies are produced, and the combining of these with the 
antigen causes a falling out or inactivation of the antiferments of the serum 
by a change in colloidal conditions resulting in the release of normal or non- 
specific proteolytic ferments which disrupt or digest the protein of the 
serum. In other words, specific antibodies are produced, but instead of 
these digesting the protein of the antigen or of the serum directly, they 
act by uniting with the antigen, and this union results in the removal of 
anti-enzyme, thereby releasing or rendering active the normal and non- 
specific enzymes of the serum which procluce a protein poison or anaphyla- 
toxin, by digesting the protein of the serum. 
However, practically the same objection listed above against accepting 
the production of a poisonous protein substance by the interaction of com- 
plement and specific protein amboceptors, can be offered against the pro- 
duction of anaphylaxis by these poisons produced by the activity of pro- 
teases. In other words, the poison has not been satisfactorily demonstrated 
in the blood of anaphylactized animals, the latent period of passive sen- 
sitization is not explained, the possibility of the production of anaphylac- 
toid symptoms by embolism and thrombosis has not usually been suffi- 
ciently controlled and the immediate production of anaphylactic con- 
traction of the perfused sensitized guinea-pig uterus by the Schultz-Dale 
method when the antigen is added to the bath, is far too rapid for prote- 
olysis. 
On the other hand, these proteases may be important and active in the 
production of pseudopositive skin reactions which I will discuss in relation 
to the local allergic reactions. 
The Relation of Humoral Poisons to Allergy.—Setting aside for the pres- 
ent the question of the nature of the antibody concerned in allergy, that is, 
whether it is a proteolysin of the nature of an amboceptor or a proteolytic 
ferment, the important question remains whether or not a poison of protein 
1 Jour. Infect. Dis., 1917, 20, 499-854. 630 
ALLERGY. ANAPHYLAXIS AND HYPERSENSITIVENESS 
origin or otherwise is produced in the blood responsible for the allergic 
reaction. According to the humoral theories such a poison is produced and 
is regarded as the direct cause of the allergic reaction; according to the cell- 
ular theory such poisons are not produced and have no essential relation 
to allergy. 
The evidence for and against the production of humoral poisons and 
their relation to allergy may be summarized as follows: 
1. The presence of poisons in the blood during and after anaphylactic 
shock has not been conclusively demonstrated. Novy and DeKruif found 
that the blood of guinea-pigs sensitized with large amounts of egg-white 
or horse-serum and shocked by the intravenous injection of the respective 
antigens, were more toxic than the blood of normal guinea-pigs transfused 
under similar conditions of transfer time. - The amounts of the antigens 
injected were many times larger than those ordinarily employed in the 
classical anaphylaxis experiment and the authors state as their opinion that 
“the antigen is in nowise the source of the anaphylatoxin which is brought 
into being in shock.” As previously stated, Weil obtained uniformly nega- 
tive results when normal dogs were bled and transfused with a volume of 
blood equal to or greater than the amount removed from dogs dying or 
dead in anaphylactic shock. 
This subject requires further investigation. It may be, as Dale1 says, 
that the poison is produced within the cells and need not be large enough to 
be detected with the methods at our disposal. 
2. The almost total lack of incubation period following the intravenous 
injection of antigen into sensitized animals and the development of the 
lesions and symptoms of shock, is evidence against the production of a 
poison, unless the amount required to produce a reaction is infinitely small. 
The same may be said of anaphylactic experiments in vitro with strips of 
uteri from sensitized guinea-pigs after the method of Schultz and Dale, 
when the addition of antigen to the bath is followed almost immediately 
by vigorous contraction. These experiments appear to rule out the possi- 
bility of even an intracellular digestion of a sensitized antigen by antibody 
or ferment. 
3. The production of symptoms of anaphylaxis following the simulta- 
neous intravenous injection of antigen and antiserum into normal animals 
supports the theory of production of humoral poisons. Results of this kind 
have been reported by Gurd and by Thiele and Embleton.2 In the experi- 
ments of Manwaring and Kusama3 employing guinea-pig’s lung, it was found 
that perfusion with a mixture of antigen and the blood of a sensitized animal 
caused a bronchial spasm. Weil also found that the simultaneous injection 
of antigen and antiserum was sometimes followed by mild symptoms of ana- 
phylaxis, but the significance to be attached to these results is questioned 
by reason of similar results being observed by him with intravenous injec- 
tions of antiserum and an unrelated antigen. 
4. Against the theory of production of poisons by humoral antibodies 
at least, are the experiments of Doerr and Pick, who observed fatal ana- 
phylactic shock in rabbits upon reinjection of the antigen at a time when 
circulating antibodies had disappeared from the blood. Weil also found 
that the delayed form of anaphylactic shock in guinea-pigs may be elicited 
by the intraperitoneal injection of antigen at a time when sensitizing anti- 
bodies could not be found in the blood. However, these experiments do not 
1 Jour. Pharm. and Exper. Ther., 1912-13, 4, 167. 
2 Jour. Immunology, 1916-17, 2, 109. 
3 Ztschr. f. Immunitatsf., 1913, 20, 159. ANALYSIS OF DATA BEARING UPON THE MECHANISM OF ALLERGY 
631 
prove that poisons may not be produced by antibodies in or upon the cells; 
they do indicate, however, that shock may be produced in the apparent 
absence of circulating antibodies and to this extent reduce the chances of 
poison production in the blood. 
5. The experiments of Weil indicating that the presence of antibodies 
in the blood of sensitized animals may actually protect against anaphylactic 
shock by preventing the injected antigen from reaching sensitized cells, is 
to be interpreted as evidence against the production of humoral poisons by 
antibodies. Further reference to these observations will be made in the 
sections dealing with anti-anaphylaxis and desensitization. 
6. The isolation experiments of Schultz, Dale, Weil, Coca, and others 
conducted with strips of uteri, lungs, or other organs from sensitized guinea- 
pigs, suspended in an oxygenated bath and made to undergo almost im- 
mediate and spectacular contraction upon the addition of antigen to the 
bath, appears to be almost indisputable evidence against the production 
and activity of humoral poisons, and especially so, since these experiments 
have been conducted by Dale and others with organs thoroughly perfused 
with saline solution and washed free of blood. 
7. The latent period of passive sensitization is evidence against the 
production of humoral poisons by proteolysins or ferments. When large 
amounts of serum from a sensitized rabbit are injected into normal guinea- 
pigs, a certain period must elapse before the injection of antigen elicits an 
anaphylactic response. Test injections undertaken during this period have 
shown that no amount of antibodies in the blood alone can bring about the 
reaction of anaphylactic shock. 
8. That anaphylactic shock may be produced by protein poisons is sup- 
ported by the fact that the intravenous injection of peptone and sera rendered 
toxic in vitro by admixture with various substances, is frequently followed 
by anaphylactiform lesions and symptoms. Too much importance, how- 
ever, cannot be placed upon these results for several reasons, as follows: 
(a) The symptoms may be anaphylactoid and due to intravascular agglutina- 
tion, hemolysis, or to embolism and thrombosis. (b) The production of these 
toxic sera by simple contact with filters, mixture with kaolin, starch, agar, 
and so forth, has no analogy to anaphylaxis which is a highly specific reac- 
tion in which the antigen or exciting agent is specific, (c) The amount of 
nitrogenous poisons produced in vitro required to engender anaphylactiform 
symptoms is many times greater than the maximum amounts that could 
possibly be produced in the classical anaphylaxis experiment, as, for ex- 
ample, the fatal outcome produced by the intravenous injection of sensi- 
tized guinea-pigs with 0.001 c.c. serum or 0.00005 gm. of crystalline egg- 
white. 
Of the toxic substances of protein cleavage producing anaphylactiform 
lesions and symptoms, histamin and methyl guanidin are the best examples. 
Histamin not only produces bronchial spasm in guinea-pigs, obstruction to 
pulmonary circulation in rabbits, and fall of blood-pressure in dogs, but 
causes marked urticaria when applied to the skin, resembling the local 
allergic reaction. Dale has found that it causes the contraction of de- 
sensitized strips of uteri and this is interpreted as indicating that it repre- 
sents a separate substance from that concerned in the anaphylaxis reaction, 
but this is not necessarily the case, inasmuch as the failure of desensitized 
tissue to react upon the addition of antigen may be due to exhaustion of 
antibodies in the cells. Histamin also does not produce the temperature 
reactions of anaphylaxis or the reduced coagulation time of the blood, but 
these effects in anaphylaxis may be due to other substances. 632 
ALLERGY. ANAPHYLAXIS AND HYPERSENSITIVENESS 
The Cellular Colloidal Nature of Allergy.—The sum total of the evi- 
dence in favor of humoral poisons being the cause of anaphylactic shock is 
very slight. That toxic products of protein cleavage may be produced in 
vitro which upon injection into animals produces lesions and symptoms 
closely resembling anaphylaxis is well established. The possibility of such 
toxic substances being produced in the living animal by disturbances in the 
equilibrium of plasma colloids is to be granted. It is entirely likely that 
such poisonous substances may be produced in the blood in anaphylaxis, 
but the evidence that anaphylactic shock is due primarily to their produc- 
tion and activity is inconclusive. The mere fact that their physiologic 
effects resemble the lesions and symptoms of anaphylaxis does not establish 
that they are identical. Such humoral poisons may be produced in ana- 
phylaxis and probably are to some extent under certain conditions, but their 
production is of secondary importance. Their effects may add to the lesions 
and symptoms of anaphylaxis as in the production of temperature changes, 
alterations in the corpuscular elements and coagulation time of the blood, 
the increase of amino nitrogen, the production of the non-specific phase of 
local allergic skin reactions, and so forth, but in the writer’s opinion there is 
no proof that these humoral poisons are responsible for the primary and 
essential lesion of anaphylaxis, namely, the contraction of smooth muscle in 
man, rabbit, and guinea-pig, changes in the liver cells in dogs, and so forth. 
That cellular poisons of like nature may be produced commands more 
attention, but the immediate and vigorous contraction of the washed-out 
and excised uterus of a sensitized guinea-pig in vitro renders this almost 
impossible of acceptance. The anaphylactiform effects noted upon the ex- 
cised lung of a guinea-pig perfused with a mixture of antigen and antiserum 
may be due to non-specific production of toxic substances by disturbance 
of plasma colloids; this is especially likely in view of experimental data 
showing how very slight and insignificant manipulations of the blood may 
render it toxic. 
Just what occurs in sensitized cells responsible for the allergic reaction 
is unknown. We know’ nothing of what sensitization consists of beyond 
that it may be accompanied by antibody production. These antibodies may 
be found in the blood and for protein antigens may be precipitins; on the 
other hand, they may be purely cellular as shown so conclusively by Block 
in the grafting of a patch of skin from a person hypersensitive to iodoform 
to a normal individual. In many forms of allergy the antibody is entirely 
cellular, as in allergies to tuberculin, other bacteria, foods, and drugs. 
Nothing is definitely known of the exact nature of this antibody. With 
protein antigens as serum and egg-albumen, it is produced in the lower ani- 
mals in such abundance as to be found free in the blood and capable of 
engendering passive sensitization. In man sensitized to serum, free anti- 
body is generally not found in the serum, its presence in the blood appar- 
ently varying according to the species. Whether it is an intracellular and 
extracellular precipitin is not definitely knowm. At any rate the antibody 
is not to be found in the blood or serum of all allergies; it is essentially and 
primarily located in the tissue-cells and principally in the cells of involun- 
tary muscle. 
The allergic reaction with both protein and non-protein sensitinogens is 
cellular and apparently caused by alterations in colloidal equilibrium within 
the protoplasm of the cells carrying the antibody (sensitized cells). As 
stated by Dale, the reaction can probably be accounted for by the antibody 
bearing such a relation to the antigen that when they meet a disturbance 
of the conditions of colloidal equilibrium is set up in the sensitized cells. ANALYSIS OF DATA BEARING UPON THE MECHANISM OF ALLERGY 
633 
As discussed in the chapter on Precipitins the phenomenon of precipita- 
tion is apparently a colloidal reaction. Under certain conditions actual in- 
tracellular precipitation may initiate allergic shock by changes in the state 
of aggregation of colloidal particles, but it is not necessary to assume that the 
quantitative relations of antigen and antibody must be such as to yield 
visible precipitation. 
Briefly, in the writer’s opinion, allergy embracing hypersensitiveness to 
both protein and presumably non-protein substances is a cellular rather than 
a humoral reaction; the exact nature of the cellular disturbances responsible 
for sensitization and the production of acute and chronic allergic shock or 
reactions is unknown, but very probably a phenomenon in the domain of 
colloidal chemistry. 
Lumiere1 likewise subscribes to the colloidal theory of allergy, stating 
that the phenomenon is one of colloidal flocculation of serum and that the 
flocculates irritate the endothelium of the cerebral capillaries producing 
vasodilation, which is transmitted by reflex action to the splanchnic and 
other capillaries and responsible for the vascular phenomena. 
The Production of Passive Anaphylaxis.—An important phase of this 
subject over which there has been considerable difference of opinion refers 
to the question whether some time must elapse between the injection of 
immune serum and anaphylactogen before anaphylaxis is produced, or 
whether intoxication may follow the simultaneous injection of both anti- 
body and antigen. Thus Gay and Southard, in their early studies, found 
their recipients first sensitized on the fourteenth day after injection of 
immune serum; these results, however, may have been due to active sen- 
sitization by antigen carried over in the serum. Otto and Friedmann observed 
shock twenty-four hours after injecting the anaphjdactic serum subcuta- 
neously and antigen intraperitoneally- By injecting both serums intra- 
venously and simultaneously, Doerr and Russ finally succeeded in producing 
acute anaphylaxis and almost immediate death. Weil,2 however, believes 
that the simultaneous injection of antigen and of antiserum into opposite 
jugular veins in the guinea-pig never produces an anaphylactic reaction, 
in spite of the use of wide quantitative variations on both substances. On 
the other hand, if the antigen were injected a few hours after the anti- 
serum, in the same quantitative variations, the reaction occurred regularly. 
In the rabbit and dog, on the other hand, passive hypersensitiveness occurs 
almost at the moment of the injection of the antiserum. From this it was 
concluded that the body cells anchor the antibody during the latent period, 
and that anaphylaxis is the result of an interaction between the cellular 
antibodies and the antigen. 
It would appear that passive anaphylaxis is not wholly determined by 
the amounts of antigen or antibody, but by the proportion that exists be- 
tween the two. For example, Friedmann,3 in his studies on passive homol- 
ogous anaphylaxis in rabbits, found that, by employing 2.5 c.c. of anti- 
serum with from 2.5 to 0.25 c.c. of antigen, no results were observed, whereas 
positive reactions were obtained when the amount of antigen was reduced 
from 0.025 to 0.0025 c.c. 
Ordinarily, normal guinea-pigs may be passively sensitized by 0.1 to 
0.5 c.c. of serum injected intraperitoneally, and anaphylactized one or 
two days later by an intravenous injection (0.1 to 0.5 c.c.) of the antigen. 
The immune serum may be prepared by injecting rabbits with horse-serum, 
1 Paris med., 1921, 11, 445; Presse med., 1921, 29, 960. 
2 Jour. Med. Research, 1914, 30, No. 2, 87. 
3 Jahr. u. d. Ergeb. Immunitatsf., 1910, vi, 67. 634 
ALLERGY. ANAPHYLAXIS AND HYPERSENSITIVENESS 
after the methods for the production of precipitins described in Chapter 
XVII. 
DESENSITIZATION AND ANTIANAPHYLAXIS 
Desensitization.—The state of allergy to various exciting substances due 
to active or passive sensitization may be decreased or entirely removed for 
varying periods of time by certain procedures. For this refractory condi- 
tion Besredka and Steinhardt1 originally proposed the terms “antianaphy- 
laxis” and “desensitization”; since then the word “desensitization” has come 
into general use for designating this particular kind of refractory state or 
insensibility to further injections that may follow recovery from anaphy- 
lactic shock or be induced by one or several administrations of the antigen 
in subintoxicating amounts. 
Desensitization may be accomplished by the administration of non- 
shocking amounts of the specific sensitinogen (specific desensitization) or, to 
some extent, by the administration of various unrelated substances (non- 
specific desensitization). 
Antianaphylaxis.—This is the refractory state due to the presence of 
sufficient antibodies in the blood to neutralize the antigen and thereby pro- 
tect the sensitized cells. It is a specific phenomenon and is commonly 
designated as antianaphylaxis by antibody protection. 
The phenomenon of desensitization was described by the early investi- 
gators in anaphylaxis. For example, Rosenau and Anderson observed that 
guinea-pigs reinjected with horse-serum before the end of the primary in- 
cubation for sensitization, did not become hypersensitive for some time 
later. They also gave guinea-pigs a series of three or four injections of large 
amounts of horse-serum at intervals of six days, and finding them refractory 
to anaphylactic shock, believed that the animals had become immunized 
against the supposed toxic portion of serum regarded at that time as re- 
sponsible for the production of acute anaphylactic shock. Similar observa- 
tions were made by Otto and Besredka. These results are now known to 
have been due not to immunization against some constituent in the serum, 
but to desensitization. 
Production of Specific Desensitization.—If a guinea-pig that has been 
actively or passively sensitized be given by subcutaneous or intraperitoneal 
injection a non-shocking quantity of the respective anaphylactogen, the 
animal will be found, after an interval of several hours, to have lost its 
previous hypersensitiveness, so that now it may withstand larger injections 
in the same manner as a normal guinea-pig. The isolated muscle strips 
(uteri) from such animals exhibit the same desensitization. 
This desensitization is highly specific; if a guinea-pig has been sensitized 
to two different proteins and recovers from the shock induced by one of 
them to which it thus becomes desensitized, it still remains sensitive to the 
other, but to a somewhat lesser degree. This is apparently due in part to 
non-specific desensitization. 
The desensitizing dose of antigen may be given subcutaneously, intra- 
peritoneally, intravenously, or intracerebrally, but the amount injected 
should be less (five times at least) than that required to produce shock, by 
these different routes of administration. According to Besredka, desensitiza- 
tion of guinea-pigs follows on an average four to five hours after subcuta- 
neous injection, one or two hours after intraperitoneal or intraspinal, and 
almost instantaneously after intravenous and intracerebral injection. De- 
sensitization may also be effected by oral and rectal administrations of the 
1 Ann. d. l’lnst. Pasteur, 1907, 21, 117, 384. DESENSITIZATION AND ANTIANAPHYLAXIS 
635 
antigen, but the time required is usually forty-eight to seventy-two hours 
and the degree of desensitization is less complete. The rectal route is some- 
what more rapid than the oral. 
A guinea-pig sensitized to horse-serum is usually desensitized partly or 
completely by the injection of 0.05 c.c. serum subcutaneously, 0.02 c.c. in- 
traperitoneally, and 0.01 c.c. intravenously. A series of injections desen- 
sitizes more efficiently than a single injection. For example, the intravenous 
injection of 0.01 c.c. serum may desensitize against only double the shock- 
producing dose, but if this injection is followed at intervals of five minutes 
by the injection of increasing amounts, as 0.02, 0.05, and finally 1 c.c., the 
animal may be desensitized against twenty or more times the shock-pro- 
ducing dose of antigen. 
Besredka has also given the following example of desensitization of 
guinea-pigs actively sensitized to egg-albumen to such degree that after an 
incubation period of eighteen days the intravenous injection of 0.002 c.c. was 
fatal within a minute. For desensitization 0.0005 c.c. egg-albumen was 
given intravenously; two minutes later 0.002 c.c. was injected without symp- 
toms, followed at ten-minute intervals by the injection of 0.02,0.2, and finally 
2 c.c. Within a period of forty-five minutes, therefore, an animal was 
desensitized to the extent of being able to survive one thousand times the 
calculated fatal amount of this substance. 
Weil1 has found that the degree of desensitization does not bear a quanti- 
tative relationship to the amount of antigen injected. The antibody meas- 
ured in terms of precipitin unites with its antigen in the test-tube in con- 
stant proportions to form precipitate, but when attached to the cells it 
appears to combine with antigen in varying proportions so that antigen 
injected into guinea-pigs passively sensitized “does not depress the reac- 
tivity of cellular antibody in regular quantitative ratios.” Coca2 has sug- 
gested that the antibodies may display varying combining power for antigen 
so that the partially desensitizing injections have the effect of neutralizing 
only or chiefly the antibodies of the greater avidity. 
Another form of transitory desensitization or, rather, unusual delay in 
development of active sensitization, is observed when guinea-pigs are given 
large sensitizing injections, as 2 c.c. of horse-serum on each of three suc- 
cessive days. When given the test injection after the usual period of in- 
cubation of eight to twelve days they may be found refractory, although 
the blood does not contain sufficient antibodies to explain the results on 
the basis of antianaphylaxis due to antibody protection. Weil has ex- 
plained this on the basis of coexistence of antigen and antibody in the cells, 
but a simpler explanation is on the basis of the persistence of antigen in the 
blood over a long period of time which gradually unites with the cellular 
antibodies as they are formed, and thereby desensitize in a constant and 
mild manner until the excess of antigen has been consumed in this manner. 
The subject of desensitization is of great importance in connection with 
serum therapy of man and the lower animals; also in relation to the treat- 
ment of hay-fever, drug, and bacterial allergies. For this reason it will be 
considered in greater detail in relation to these subjects in a following 
chapter. 
Mechanism of Specific Desensitization.—Desensitization is commonly 
regarded as due to the specific neutralization, exhaustion or saturation of 
the sensitizin or allergic antibodies situated in the cells of the sensitized 
tissues. Apparently when the sensitinogen is introduced slowly and in the 
1 Jour. Immunology, 1917, 2, 469. 
2 Tice’s Practice of Medicine, 1920, 140. 636 
ALLERGY. ANAPHYLAXIS AND HYPERSENSITIVENESS 
first dose in less than the shock-producing amount, the effects of union with 
the sensitized cells are not injurious or of minor importance; on the con- 
trary, shock is produced by the sudden access of sensitinogen to the sen- 
sitized cells. Desensitization may be a process of restoration of colloidal 
equilibrium within sensitized cells. The colloidal nature of phenomenon is 
also suggested by non-specific desensitization. 
Non-specific Desensitization.—Transitory desensitization of partial de- 
gree of guinea-pigs can sometimes be effected by the intravenous injection 
of numerous substances a short time previous to the test injection. In this 
connection sodium chlorid and other inorganic salts, peptone, trypsin, urine, 
foreign serum, or other protein substances, have been employed. 
Biedl and Kraus1 have asserted that animals sensitized to sera can be 
desensitized with such substances as peptone, and the oral administration 
of this substance has been advised by others for desensitization in food al- 
lergy. Kopaczevski and Vohram2 found that sodium oleate reduced ana- 
phylactic shock, which they attributed to a lowering of surface tension of 
the blood. Sodium chlorid has received special attention in this connection; 
Richet3 ascribes its effects to an influence upon nerve-cells, while other 
investigators attribute its inhibitory effect to a modification of the colloidal 
state of fluids and cells. 
In relation to non-specific desensitization of serum anaphylaxis, the 
injection of beef-serum has lately commanded considerable attention. 
Penna4 has observed in the Argentina that the administration of beef-serum 
in the treatment of anthrax has rarely produced serum sickness. Pfeiffer 
and Mita5 have protected guinea-pigs sensitized to horse-serum against 
anaphylactic shock by giving previous injections of beef-serum; Penna, 
Cuenca, and Kraus6 have prevented serum sickness in man by injections of 
beef-serum. 
Non-specific desensitization is not quite the same as antisensitization; 
for example, in the experiments of Julian Lewis7 sensitization may be pre- 
vented when horse-serum is first mixed with dog-serum before injection into 
the lower animals. According to this author, as much as 0.1 c.c. horse- 
serum in 10 c.c. of dog-serum would not sensitize, even though 0.000,001 c.c. 
of horse-serum alone produced a high degree of sensitization. Other sera 
(human, cat, and beef) added to horse-serum had a similar effect. 
Karsner and Ecker8 have recently reported an extensive series of ex- 
periments upon guinea-pigs actively sensitized with horse-, human, and 
beef-sera; the sera of the horse, beef, goat, swine, dog, cat, rabbit, and 
human were employed for non-specific desensitization. The authors found 
that guinea-pigs sensitized to serum may be desensitized to a certain degree 
by injections of heterologous sera, but never as completely as when homol- 
ogous serum was employed. They also found that heterologous desen- 
sitization was best accomplished by giving the serum intravenously, as 
has been generally found true in specific desensitization with homologous 
sera. In general, non-specific desensitization with heterologous sera was of 
shorter duration than observed with specific desensitization. 
This non-specific refractory state is not absolute; shock in full measure 
1 Wien. klin. Wchn., 1909, 22, 363. 
2 Compt. rend. Acad. d. Sci., 1919, 169, 250. 
3 Compt. rend. Acad. d. Sci., 1919, 169, 9. 
4 Jour. Amer. Med. Assoc., 1918, 71, 587. 
5 Ztschr. f. Immunitatsf., 1910, 4, 410. 
6 Rev. de l’Inst. Bacteriol., 1918, 1, 405. 
7 Jour. Amer. Med. Assoc., 1921, 76, 1342. 
8 Jour. Infect. Dis., 1922, 30, 333. LOCAL ALLERGY 
637 
can be elicited by the administration of larger amounts of the specific 
antigen. 
The mechanism is not understood, but, as referred to above, is appar- 
ently referable to colloidal phenomena within sensitized cells. 
The Mechanism of Antianaphylaxis.—As stated above, antianaphylaxis 
is applied to the refractory state due to the presence of sufficient sensitizin 
or antibody in the blood to effectually neutralize small amounts of the 
antigen before the latter reaches sensitized cells with the production of 
shock. It is not absolute, since Weil1 has shown that it can be overcome with 
relatively large intravenous injections of the antigen; Weil has proposed 
the phrase “antibody protection” to the phenomenon. 
Weil found that if small desensitizing quantities of antigen are injected 
subcutaneously into passively sensitized guinea-pigs that had been given 
large protective injections of antiserum, the circulatory antibodies are found 
to be “neutralized,” although the uterine muscle still remains sensitive. 
Final proof of the correctness of these observations was furnished by Man- 
waring and Kusama, who found that in lung perfusion experiments conducted 
with sensitized guinea-pigs that the tissues were protected against reaction 
by the presence of blood. 
The phenomenon of antianaphylaxis in this restricted meaning of the 
term is probably of rare occurrence in human beings inasmuch as the pres- 
ence of free sensitizin (antibody) in the blood as determined by the success 
of passive transfer to the lower animals, is relatively uncommon. 
LOCAL ALLERGY 
Among the earliest allergic phenomena to be described was the local 
reaction of edema and necrosis observed by Arthus2 in rabbits sensitized to 
horse-serum following the subcutaneous injection of the antigen. Since then 
it has become well established that local allergic reactions may occur among 
human beings not only to serum, but to other sensitinogens including vari- 
ous proteins of animal and vegetable origin as well as drugs and other pro- 
tein substances. 
Nicolle3 and Lewis4 have observed the Arthus reaction among guinea- 
pigs actively sensitized with horse-serum, but similar reactions have not been 
reported occurring in dogs. 
In man local allergic reactions are elicited by the application on or the 
injection of the sensitinogen into the skin or mucous membranes in indi- 
viduals hypersensitive to foods, pollens, bacteria, and drugs as well as to 
horse-serum and to the proteins of the skin and hair of the horse and other 
animals and the feathers of fowl. Of considerable interest in this connection 
are the observations of Auer5 to the effect that injury to the skin of a sen- 
sitized animal at a time when the antigen is present in the blood may result 
in a deposit of the antigen in the injured tissues by extravasation from the 
blood, with the production of a local allergic reaction. 
Passive Sensitization of the Skin and Mucous Membranes.—Passive 
sensitization of the skin and underlying tissues has been accomplished by 
Nicolle, who injected 50 to 60 c.c. of antihorse rabbit-serum into each of 
5 normal rabbits and found that twenty-four hours later the subcutaneous 
injection of horse-serum elicited a local Arthus reaction in some animals so 
1 Jour. Immunol., 1917, 2. 157. 
2 Compt. rend. Soc. de biol., 1903. lv, 817; 1906, lx, 1143. 
3 Ann. d. l’lnst. Pasteur, 1907, 21, 128. 
4 Jour. Exper. Med., 1908, 10, 1. 
6 Jour. Exper. Med., 1920, 32, 427. 638 
ALLERGY. ANAPHYLAXIS AND HYPERSENSITIVENESS 
treated. Attempts to confer passive sensitization of rabbits and guinea- 
pigs with injections of the sera of human beings hypersensitive to pollens, 
bacteria, foods, and drugs have generally failed, and for this reason Coca 
regards the Arthus phenomenon as the only true anaphylactic (antigen- 
antibody) reaction. As previously stated, this does not appear .to be a war- 
rantable conclusion; even the sera of highly sensitized guinea-pigs may con- 
tain too few circulating antibodies to passively sensitize other pigs, and the 
cells of man may readily be sensitized to a protein without there being 
sufficient free antibodies in the blood to confer passive heterologous sen- 
sitization. . • • j 
Schultz and Larson,1 for example, claim to have passively sensitized 
guinea-pigs with the blood of infants with exudative diathesis apparently 
due to milk allergy, and Ramirez2 has recorded a striking example of passive 
transfer of allergy to horse protein in a human being. A man who had 
never had asthma, hay-fever, urticaria, and the like received a transfusion 
of 600 c.c. of blood from a man with typical horse asthma who gave a cuta- 
neous reaction to horse dandruff in 1 : 50,000 dilution. Two weeks later the 
transfused patient went for a carriage ride and within five minutes had a 
typical attack of asthma, and a skin test yielded a positive reaction to horse 
dandruff diluted 1 : 20,000, but not to other proteins. 
Tuberculin hypersensitiveness apparently has been passively trans- 
ferred from tuberculous to normal animals by Helmholz3 and Austrian, 
but against these positive results is a long list of negative results. 
Active Sensitization of the Skin and Mucous Membranes.—Sensitization 
of the skin and mucous membranes develops more actively among human 
beings than among the lower animals. However, in serum allergy the skin 
and mucous membranes may not demonstrate sensitiveness when tested by 
the usual methods, whereas the administration of serum may elicit a general 
allergic reaction. . 
During active immunization of man and the lower animals with bac- 
terial vaccines skin hypersensitiveness sometimes develops as demonstrated 
by unusually severe local reactions following subcutaneous injections. 
Hypersensitiveness in man and the lower animals to tuberculin, how- 
ever, apparently only occurs in active tuberculosis, as shown so clearly.in 
the valuable researches of Kraus.5 While it is easily possible to sensitize 
guinea-pigs to tuberculoprotein so that they will yield general or systemic 
allergic reactions, such animals do not display hypersensitiveness of the skin 
and mucous membranes to either tuberculoprotein or tuberculin. Skin 
hypersensitiveness to tuberculin is apparently, engendered only by. pro- 
ducing an actual tuberculous infection or by the injection of animals with an 
emulsion of tuberculous tissue as described by Bail6 and confirmed by Onaka. 
On the other hand, acquired hypersensitiveness of human beings to tuber- 
culin has been described, the allergy developing in both tuberculous and 
non-tuberculous individuals after repeated administrations and character- 
ized by various skin eruptions, sneezing, lacrimination, or general symptoms. 
According to Fleischner, Meyer, and Shaw8 the immunization with dead 
typhoid and contagious abortion bacilli may produce systemic allergic sen- 
1 Arch. Pediat., 1918, 35, 705. 
2 Jour. Amer. Med. Assoc., 1919, lxxiii, 984. 
3 Ztschr. f. Immunitatsf., 1909, 3, 371. 
4 Jour. Exper. Med., 1912, 15, 149. 
5 Amer. Rev. Tuberc., 1917, 1, 65. 
6 Ztschr. f. Immunitatsf., 1909, 4, 470; 1912, 12, 451. 
7 Ztschr. f. Immunitatsf., 1910, 7, 507. 
8 Amer. Jour. Dis. Child., 1919, 18, 577. LOCAL ALLERGY 
639 
sitization, but not hypersensitiveness of the skin, the latter being engendered 
only by living bacilli as in induced skin hypersensitiveness to tuberculin. 
As previously discussed under sensitinogens, these investigations suggest 
that bacterial sensitinogens of this kind are products of living bacteria and 
not merely the protein of the bacterial cells. 
The administration of horse-serum may result in the production of skin 
hypersensitiveness, but the most striking examples of this condition are 
met with among horse asthmatics, who develop extreme hypersensitiveness 
in unknown ways. 
Pollen hypersensitiveness is apparently acquired by inhalation of the 
pollen, but the subcutaneous injection of pollens does not appear to actively 
sensitize, according to Cooke, Flood, and Coca.1 According to these in- 
vestigators sensitization to pollens and foods are established spontaneously 
and never by immunologic processes. 
The Mechanism of the Skin Allergic Reaction.—Very little is definitely 
known of this subject; by analogy it is regarded as fundamentally the same 
as the general or systemic reaction. 
It is to be assumed that the cells of the part are sensitized and contain 
antibodies, and this refers not only to the smooth muscles of the blood- 
vessels in the subcutaneous and submucous tissues, but to the connective 
tissue and even the epithelial cells of the parts yielding a reaction. Mere 
installation of the antigen into the conjunctival sac or injection into epi- 
dermis may elicit a reaction and thereby demonstrate sensitization of the 
cells with which the antigen is brought into relation. 
Positive reactions have been explained by the humoralists on the basis 
of the production of a poison (anaphylatoxin) either by the interaction of 
antigen, specific amboceptors and complement, or by the digestive activity 
of proteolytic ferments. Both of these hypotheses have been previously 
discussed and found inadequate. The writer, who now subscribes to the 
cellular theory of allergy, is of the opinion that the specific allergic skin 
reaction is due to an intracellular disturbance of colloidal equilibrium in the 
same manner as the general allergic reaction. 
However, the local lesion may not be entirely one of allergy, and espe- 
cially if the antigen has been injected intracutaneously or subcutaneously. 
Under these circumstances there is no reason to doubt that cellular pro- 
teolytic enzymes may bring about the digestion of a portion of the injected 
antigen in a purely non-specific manner, and that the resulting products 
may aid in the production of inflammatory changes. 
Furthermore, the lesion may be due in part to mechanical irritation by 
the injected substances and particularly the preservative, as well as to the 
trauma induced by the needle. For this reason a local skin reaction may be 
due to (a) a true allergic reaction plus (b) a non-specific protein digestion, 
and (c) trauma. 
The Symptoms and Lesions of Local Allergy.—The symptoms are largely 
due to vascular changes. Prominent symptoms are erythema and edema 
(urticaria). In hay-fever the nasal mucosa is markedly hyperemic and ede- 
matous, and these changes may involve the bronchial mucosa with the pro- 
duction of asthma. 
The same changes occur in the skin reactions to pollens, tuberculin, 
foods, drugs, etc. In extreme hypersensitiveness, as in horse asthma or 
pollen asthma, the mere application of the sensitinogen or exciting agent to 
an abrasion of the skin is followed in a few to fifteen minutes by the develop- 
ment of a wheal or urticarial lesion accompanied by a burning sensation. 
1 Jour. Immunology. 1917, 2, 217. 640 
ALLERGY. ANAPHYLAXIS AND IIYPERSENSITI YEN ESS 
It would appear that these changes are due primarily to vasomotor 
paralysis, that is, extreme relaxation of the vessel walls which may follow 
a primary contraction of the smooth muscle of the involved parts. This 
relaxation is quickly followed by extravasations of serum and leukocytes, 
and especially of eosinophils. Bandler and Kreibich1 and Daels2 have de- 
scribed histologic changes of tuberculosis in local tuberculin reactions; 
the writer has not found any changes of this kind, however, in excised 
lesions sixty hours after the cutaneous application of tuberculin. 
A more complete description of these lesions will be given in the suc- 
ceeding chapter. 
SPECIFICITY OF ALLERGY 
Extensive investigations on anaphylaxis of guinea-pigs sensitized with 
various proteins of animal and vegetable origin have indicated that ana- 
phylaxis is a highly specific phenomenon; the degree of specificity compares 
favorably with that observed with the precipitin and complement-fixation 
reactions. 
Drug allergies are known to be highly specific and the same is apparently 
true of allergies in man to bacteria and their products, serum, pollens, etc. 
In drug allergies the specificity is often referable to a certain element or 
chemical group. For example, in allergy or hypersensitiveness to mer- 
curial compounds the active portion is the element mercury; in the allergy 
to iodoform, the methyl group rather than the iodid or iodin is apparently 
the allergenic group. Iodin hypersensitiveness, however, does exist as 
shown by allergy to the iodids of certain metals and not to their 
chlorids. 
It is true that group reactions are sometimes observed, but, as stated by 
Wells, this is due to the fact that the substances (serum, egg-albumen, milk, 
etc.) commonly employed as antigens are very complex and contain a num- 
ber of sensitizing substances and especially when administered in large 
amounts. Experiments conducted with isolated proteins purified as much 
as possible, and especially with the proteins obtainable from seeds and nuts, 
many of which are crystallizable and soluble in alcohol, have indicated 
striking degrees of specificity, as tested by active sensitization and desen- 
sitization. 
The in vitro anaphylactic reaction conducted with strips of uteri from 
sensitized young virgin guinea-pigs after the Schultz-Dale method, fre- 
quently exhibits sharper and clearer evidences of specificity than reactions 
produced in the living animal. According to Doerr the subcutaneous injec- 
tion of antigen, while eliciting a slower response, may prove more specific 
than reactions produced by intravenous injections. 
Working with purified vegetable proteins Wells and Osborne3 found that 
the chemical composition and not biologic origin determines specificity. This 
is a very important conclusion and of fundamental importance. In other 
words, chemically similar proteins from seeds of different genera were found 
to react anaphylactically with one another, while chemically dissimilar pro- 
teins from the same seed failed to do so in many cases. This led to a second 
conclusion by these investigators, namely, that in the protein molecule 
there are only certain groups essentially concerned in the specific nature of 
the anaphylaxis reaction. 
As to what groupings in the protein molecule determine the specificity, 
1 Deutsch. med. Wchn., 1907, 1629. 
2 Med. Klin., 1908, 58. 
3 Jour. Infect. Dis., 1916, 19, 183. SPECIFICITY OF ALLERGY 
641 
little is known. On the basis of precipitin tests Obermayer and Pick believed 
that specificity was determined by the aromatic radicals, but this could not 
be corroborated by Wells in anaphylaxis experiments. When two protein 
derivatives (halogenized, azotized, etc.) react with each other, the position 
in the molecule of the substituted radicals is apparently identical or closely 
related and the cross reactions dependent on chemical relationships. 
While the anaphylactic reaction is one of the most specific of biologic 
phenomena, nevertheless group reactions occur as in the precipitin and com- 
plement-fixation reactions. At one time it was hoped that the reaction would 
differentiate one organ or tissue of an animal from another organ of the 
same animal. Ranzi1 succeeded in sensitizing guinea-pigs with extracts of 
various tissues, but did not observe any evidences of organ specificity; similar 
conclusions were arrived at by Minet and Bruyant.2 With the proteins of 
the lens of the eye, however, organ specificity has been observed. Kopsen- 
berg3 rendered guinea-pigs anaphylactic by giving large sensitizing injec- 
tions of guinea-pig lens; with the lens substance of other animals sensitiza- 
tion resulted with smaller amounts. Anaphylactic reactions were produced 
by injections of lens substance from any species, but not by the serum. 
In other words, anaphylaxis to lens protein is organ specific, but not species 
specific. 
The specificity of bacterial anaphylaxis has been the subject of con- 
siderable investigation and controversy. Kraus and Doerr found it rigidly 
specific; according to them guinea-pigs sensitized with typhoid bacilli re- 
spond only to a second injection of typhoid and remain unaffected by injec- 
tions of paratyphoid bacilli. Indeed, Kraus and Admiradzibi4 later claimed 
that the anaphylactic reaction differentiated among strains of the same 
bacterium. Delarroe5 was not able to confirm this degree of specificity and 
subsequent researches by Studinski6 and Nefedoff7 likewise show that the 
specificity of bacterial anaphylaxis while high, is not as rigid as believed by 
Kraus. 
Specific and Non-specific Focal Reactions.—The reaction of hyperemia, 
edema, pain, etc., developing around foci of tuberculosis after the injection 
of a sufficient amount of tuberculin and the “lighting up” of abscesses and 
other lesions after the injection of vaccines, are known as focal reactions, and 
in tuberculosis especially possess valuable diagnostic significance. These 
reactions are commonly regarded as specific allergic reactions due to the 
union of antigen and allergic antibody in or upon the sensitized cells. 
It is well known, however, that the injection of other non-specific agents, 
as typhoid vaccine and proteoses intravenously or milk intramuscularly, 
may elicit similar reactions regarded as due to increased cellular activity 
(the plasma activation of Weichardt) and increased permeability of lymphatic 
and capillary channels with an outpouring of lymph, plasma, leukocytes, 
enzymes, etc. The enzymes doubtless aid in the production of protein poi- 
sons which, being absorbed from the foci along with other poisons, con- 
tribute to the production of a general or constitutional reaction of fever, 
tachycardia, etc. This dual character of the focal reaction will be discussed 
in more detail in Chapter XXXIX. 
1 Ztschr. f. Immunitatsf., 1909, 2, 12. 
2 Compt. rend. Soc. de biol., 1911, lxxi, 166. 
3 Ztschr. f. Immunitatsf., 1912, 15, 518. 
4 Ztschr. f. Immunitatsf., 1910, 4, 607. 
6 Compt. rend. Soc. de biol., 1909, lxvi, 207, 252, 248, 389. 
6 Compt. rend. Soc. de biol., 1911, lxx, 173. 
7 Compt. rend. Soc. de biol., 1913, lxxiv, 672. 642 
ALLERGY. ANAPHYLAXIS AND HYPERSENSITIVENESS 
Differentiation of Proteins by the Anaphylaxis Reaction.—The specificity 
of the anaphylactic reaction has led to many attempts to apply it to the 
differentiation of proteins in a practical way and especially in the differen- 
tiation of blood-stains and meats for medicolegal purposes. 
Thomsen1 has used the method for the differentiation of blood-stains, as 
likewise Uhlenhuth2 and others. Guinea-pigs are sensitized by injections of 
extracts of the stain and after a suitable period of incubation the animals 
are reinjected with the sera of different animals to determine to which sen- 
sitization has been effected. I have conducted these tests by sensitizing a 
series of guinea-pigs weighing approximately 300 gm. with intraperitoneal 
injections of 5 c.c. of approximately 1 : 100 dilutions of the blood-stain. 
After an interval of twenty-one days the animals were given human serum 
intravenously in dose of 0.1 c.c. and control animals the same amount of 
serum from other sources, as the dog, chicken, horse, sheep, and ox. For 
the detection of human blood this test is fairly satisfactory and in medico- 
legal work may be employed in conjunction with the precipitin and com- 
plement-fixation tests when sufficient material is available. Uhlenhuth be- 
lieves that the test has value only when the precipitable properties of the 
unknown protein have been lost by preservation or decomposition. 
The test has also been employed for the detection of meat adulteration 
and especially for the detection of cooked dog and other meats in sausage. 
Guinea-pigs are sensitized with saline extracts of the material and the test 
injections conducted with the corresponding sera by intravenous admin- 
istration. 
Pfeiffer and Finsterer3 have attempted to employ the reaction for the 
diagnosis of cancer; guinea-pigs were injected with the sera of patients 
(passive sensitization) and injected forty-eight hours later with juices ex- 
pressed from tumors. They believed that in some instances positive and 
specific reactions were observed (mainly temperature changes), but Ranzi4 
and others have failed to confirm these observations. 
The anaphylaxis reaction has also been applied for the differentiation of 
bacteria, but the uncertainties of sensitization and other practical diffi- 
culties limit the usefulness of this method, and the results so far have been 
generally less satisfactory than those observed with the agglutination and 
complement-fixation reactions. 
Probably the employment of the Schultz-Dale in vitro anaphylaxis reac- 
tion conducted with excised uteri of young virgin guinea-pigs sensitized 
with these protein substances, and especially extracts of blood-stains, will 
yield better results than those observed with living animals. 
Diagnosis of Disease.—Anaphylactic skin reactions have been widely 
employed for the diagnosis of tuberculosis, syphilis, glanders, typhoid fever, 
etc.; these will be described in a subsequent chapter. 
PRACTICAL DIAGNOSTIC APPLICATIONS 
1 Ztschr. f. Immunitatsf., 1908-09. 1, 741. 
2 Ztschr. f. Immunitatsf., 1908-09, 1, 770; referate, 1909, 1, 525. 
3 Wien. klin. Wchn., 1909, No. 28. 
4 Ztschr. f. Immunitatsf., 1909, 2, 12. Part IV 
CHAPTER XXIX 
ALLERGY IN RELATION TO INFECTION AND IMMUNITY 
In the preceding chapter mention was made that the proteins of micro- 
parasites embracing both the bacteria and protozoa may bring about active 
sensitization of the cells of man and the lower animals. By reason of the 
importance of this subject to infection and immunity, it is worthy of special 
discussion, although all that has already been discussed of anaphylaxis to 
proteins in general, may be broadly applied to bacterial anaphylaxis. 
Active Sensitization by Bacteria and Protozoa.—Rosenau and Anderson,1 
in one of their earliest investigations in anaphylaxis, conceived that bac- 
terial proteins may sensitize body cells, and they succeeded in sensitizing 
guinea-pigs with extracts of colon, typhoid, subtilis, tubercle, and anthrax 
bacilli as well as with some yeasts. Relatively large amounts of these 
extracts were required for sensitization and for the production of ana- 
phylactic shock, but they showed that bacterial anaphylaxis and desen- 
sitization could be accomplished. 
These results were soon corroborated by Kraus and Doerr,2 Holobut,3 
and others. Delanoe4 questioned the specificity of bacterial anaphylactic 
reactions, but, as stated in the preceding chapter, it is now known that 
bacterial anaphylaxis possesses the same degree of specificity as character- 
izes anaphylaxis in general. 
For the active sensitization of guinea-pigs a single injection of the bac- 
terial substance may fail, and especially so, since the administration of a 
large amount may kill the animals by reason of toxic effects. Better results 
are usually obtained by suspending a loopful of the bacteria in 2 c.c. of salt 
solution, heating at 56° C. for about thirty minutes, and injecting 1 c.c. sub- 
cutaneously or intraperitoneally daily for at least ten days. The test injec- 
tion may be made about three weeks after the last dose, should be given 
intravenously and in rather large amount, as 2 c.c. of a heavy suspension. 
A non-sensitized control animal should always be included to make sure 
that the test dose is without anaphylactoid effects. 
The proteins of practically all bacterial cells may act as sensitinogens 
for guinea-pigs and presumably for the body cells of man. Owing, however, 
to the physical state of these anaphylactogens, being usually particulate 
bodies in suspension rather than solutions, sensitization is both tardier and 
less complete than is usually observed with the soluble proteins. 
Apparently sensitization of man is sometimes effected by repeated sub- 
cutaneous injections of bacterial suspensions, and not infrequently the 
unusually severe local reactions sometimes following these administrations 
are in part anaphylactic reactions. 
BACTERIAL ANAPHYLAXIS 
1 U. S. Public Health and M. H. S. Hyg. Lab. Bull., 1907, No. 36. 
2 Wien. klin. Wchn., 1908, No. 28. 
3 Ztschr. f. Immunitatsf., 1909, 3. 
4 Compt. rend. Soc. de biol., 1909, 66, 207, 252, 348, 389. 
643 644 
ALLERGY IN RELATION TO INFECTION AND IMMUNITY 
Sensitization of guinea-pigs has also been accomplished with extracts of 
pathogenic and non-pathogenic yeasts and fungi, as well as with such pro- 
tozoa as the cyst fluid of Taenia echinococcus and extracts of other helminths. 
Active Sensitization in Infectious Diseases.—Unquestionably active 
sensitization of the body cells may occur during the course of some bacterial 
and protozoan diseases and especially in those of a prolonged or chronic 
character. A striking example of bacterial sensitization of man is afforded 
by numerous cases of bronchial asthma, apparently due to sensitization by 
bacteria in the inflamed mucosa of the respiratory tract. 
This sensitization is apparently effected by both the proteins of the 
bacterial cells and possibly by excretory products. But the mechanism of 
sensitization of the skin in tuberculosis and possibly in other infectious dis- 
eases is obscure and unique and apparently engendered by the presence and 
activities of the living bacteria in the tissues. For example, sensitization and 
skin reactions to tuberculin can be acquired only by the stimulation afforded 
by foci of tuberculosis as shown so conclusively by Kraus1; numerous at- 
tempts to sensitize animals with Koch’s old tuberculin and similar products 
have failed. Apparently the sensitizing substance is similar to the tuber- 
culin produced in cultures, but the latter does not serve for sensitization 
while that produced in the tissues sensitizes the skin and tuberculous tissues 
to an exquisite degree. The sensitizing agent is certainly in the tissues 
of the tubercle and has been successively transmitted by Bail to normal 
animals by injections of suspensions of tuberculous tissue. Fleischner, 
Meyer, and Shaw2 have corroborated these observations on tuberculin hy- 
persensitiveness of the skin, and find that it is equally true for typhoid 
bacilli and Bacillus abortus. Similar results have been observed by Zinsser 
and Parker3 in experiments carried out with tubercle bacilli, pneumococci, 
staphylococci, influenza bacilli, and typhoid bacilli. Judging by the local or 
skin reactions of hypersensitiveness, sensitization in disease may, therefore, be 
produced by the bacterial proteins, on the one hand, and by a product of the living 
bacteria in the tissues on the other. Since the tuberculin skin reaction cannot 
be elicited by immunization of healthy animals with tuberculin, the phenom- 
enon is not considered as an allergic reaction by many observers, but I prefer to 
so list it in the present state of our knowledge pending further investigations, 
believing that in tuberculosis and probably in other bacterial infections as 
well, we are dealing with a peculiar kind of active sensitization by bacterial 
substances produced only in the living animal during the course of disease 
in addition to sensitization to the bacterial proteins (a true anaphylaxis) in 
at least some instances. 
Whether anaphylactic sensitization occurs in all infectious diseases of 
bacterial, mycotic, and protozoan origin and in infestments with helminths, 
cannot be stated. Our data is almost solely based upon the results of skin 
tests. Studies on the comparative sensitizing rates of the proteins from 
different microparasites do not appear to have been made. Based upon our 
general knowledge of sensitization, one may surmize that the sensitizing 
properties are equal when the respective proteins are rendered soluble in the 
body fluids, but that the rate, degree, and amount of solution in different 
diseases varies widely and that in diseases of short duration may not occur 
at all or to an insufficient degree. But that active sensitization may occur 
in those infectious diseases where the conditions of time and adequate 
amounts of sensitizing substances are met, would appear to be the case and 
1 Amer. Rev. Tuberculosis, 1917, 1, 65. 
2 Amer. Jour. Dis. Child.,1919, 18, 577. 
3 Jour. Exper. Med., 1921, 34, 495; ibid., 1923, 37, 275. BACTERIAL ANAPHYLAXIS 
645 
worthy of the closest study in relation to the course and treatment of infec- 
tious diseases. 
Active Sensitization by Toxins; Toxin Hypersensitiveness.—Among the 
earliest observations on hypersensitiveness are those of von Behring,1 who 
found that animals once injected with non-lethal amounts of both diphtheria 
and tetanus toxins occasionally became more sensitive to them subsequently 
than normal animals. These results were particularly puzzling because 
increased tolerance rather than intolerance was expected owing to the pro- 
duction of antitoxins. Repeated injections of toxin may induce such a state 
of hypersensitiveness than even 1/700 of the usual non-lethal dose may prove 
fatal. 
The condition is sometimes encountered in horses undergoing toxin im- 
munization, but is most easily and regularly induced in guinea-pigs by 
repeated injections at short intervals. Lowenstein2 claims to have pro- 
duced hypersensitiveness to tetanus toxin by a single injection and with 
an incubation period of at least twenty days; according to Kretz3 repeated 
injections of diphtheria toxin and ricin are required. 
Various explanations have been offered. Some investigators have theo- 
rized that it is due to the presence of an excess of toxin receptors attached to 
the body cells (sessile receptors) possessing an increased affinity for the 
toxins, and that as a result of this the toxins when injected are rapidly an- 
chored with disastrous effects. However, this explanation appears insuffi- 
cient because reactions may occur at a time when the blood contains specific 
antitoxins and it is difficult to understand how the injected toxin escapes 
sufficient neutralization by these to protect the cells even though these may 
have a greater affinity for the toxin than the free receptors (antitoxin). 
That the condition is not a mere summation of toxic effects is shown by 
Loewi and Meyer,4 who found with spaced injections that the total amount 
injected was less than a single minimal lethal dose. According to these 
observers hypersensitiveness to tetanus toxin is due to changes in the gan- 
glion cells of the spinal cord because of traces of toxin retained in them. 
Friedberger, Mita, and Kumagai5 have sought to explain the phenomenon 
on the basis of “anaphylatoxin” production by specific amboceptors and 
complement. They succeeded in rendering normal guinea-pig serum toxic 
by the addition of minute amounts of dried toxin, but these results are 
analogous to those observed in the treatment of serum with agar, kaolin, 
and the like, and, as discussed in the preceding chapter, are insufficient for 
explaining anaphylaxis. 
While Friedberger’s theory is not acceptable, yet in the writer’s opinion 
toxin hypersensitiveness is an allergic phenomenon and probably ana- 
phylactic. The exact chemical constitutions of tetanus and diphtheria toxins 
are unknown, but available data indicates that they are protein substances; 
even if proteoses, it is possible that they may act as sensitinogens. While 
the symptoms of tetanus toxin hypersensitiveness are said to be largely 
those of tetanus, it is not at all improbable that the ganglion cells of the 
central nervous system possessing a known affinity for this toxin are like- 
wise with other cells acutely sensitized. Under these conditions they would 
participate in the anaphylactic shock, and if so, symptoms of tetanus may 
be expected as a result of overstimulation of their physiologic functions. 
1 Deutsch. med. Wchn., 1893, xlviii, 1253. 
2 Ztschr. f. exper. Path. u. Therap., 1914, 15, 279. 
3 Ztschr. f. Heilk., 1902. 
4 Ztschr. f. exper. Path. u. Pharmacol. Suppl. Bd., 1908, 355. 
5 Ztschr. f. Immunitatsf., 1913, 17, 506. 646 
ALLERGY IN RELATION TO INFECTION AND IMMUNITY 
While it is true that precipitins have not been found in the blood of hyper- 
sensitive animals, and that all attempts to passively transfer the condition 
to normal animals have failed, these observations cannot carry much weight 
against the probable anaphylactic nature of toxin hypersensitiveness for 
reasons discussed in the preceding chapter. I regard toxin hypersensitive- 
ness as an allergic phenomenon due to disturbances of intracellular colloidal 
equilibrium of the same nature and induced in the same manner as allergy 
to other substances. 
Relation of Bacterial Anaphylaxis to Disease.—Since active sensitization 
may occur during some of the infectious diseases caused by bacteria, fungi 
(as ringworm, favus, sporotrichosis, actinomycosis, etc.), and protozoa, a 
list of these diseases being given in the succeeding chapter, the question 
naturally arises regarding the relation of anaphylaxis to the cause and 
course of infectious diseases. 
According to the humoral theories of anaphylaxis the relation is a very 
intimate one. It will be remembered that these theories are divisible into 
three kinds, briefly as follows: (a) That anaphylaxis is due to the production 
of poisons by the action of specific amboceptors (proteolysins) and comple- 
ment upon the antigen; (b) that anaphylaxis is due to the production of a 
poison from the antigen by the digestive activity of specific ferments, and 
(c) that anaphylaxis is due to the production of a poison by the digestion of 
the serum by serum ferments or proteases. While the theories differ in re- 
gard to the mechanism of the production of a protein poison, all are based 
upon the assumption that such a poison (anaphylatoxin, serotoxin) is pro- 
duced and primarily responsible for the phenomena of anaphylaxis. The 
writer has been until the last few years an adherent of the humoral theories, 
but cannot subscribe to them now in view of the mass of excellent experi- 
mental data accumulated by Weil, Doerr, Coca, and others in favor of the 
cellular theory. 
I am convinced that poisonous substances may be produced in vitro and 
presumably in the living animal, by the interaction of antigen, specific 
antibody, and complement. Even though such poisons have not been con- 
clusively shown in the blood during anaphylaxis, they may develop during 
the course of disease and doubtless contribute to the production of lesions 
and symptoms, but this is not anaphylaxis. The same may be admitted 
for the production of such poisons by the cellular and plasma proteases. 
The writer cannot subscribe to the broad and sweeping views of Vaughan, 
Friedberger, von Pirquet, and others that such poisons are the cause of 
anaphylaxis, although it is freely admitted that they are probably produced 
during disease and not only in those diseases caused by microparasites, but 
in non-infectious diseases as well where cellular death is prominent. Doubt- 
less they exert an important role in the production of symptoms and lesions, 
and for this reason I have included them in the chapters dealing with infec- 
tion and the production of infectious diseases. 
This leaves for consideration the probable r61e of anaphylaxis in disease 
in a strict sense and in terms of the cellular colloidal nature of the phenomenon 
discussed in the preceding chapter. 
The symptoms of allergy in man may be so diverse that many of the 
symptoms ordinarily attributed to bacterial toxins may be expressions of 
anaphylaxis. The essential anaphylactic lesion is apparently contraction of 
smooth muscle followed by relaxation. For this reason dilatation of vessels 
accompanied by serous exudation are characteristic changes of the allergic 
reaction in human beings. These effects are apparently responsible for the 
skin lesions and particularly of the urticaria in general allergy; swelling of BACTERIAL ANAPHYLAXIS 
647 
the lips, tongue and diarrhea in alimentary allergy and swelling and con- 
gestion of the mucosa of the respiratory tract in pollen and bacterial allergy. 
It may well be that the exanthemata of some of the acute infectious diseases 
are anaphylactic reactions, as so considered and grouped by von Pirquet. 
The allergic reaction generally first occurs at the place of introduction of 
the exciting substance. For example, the inhalation of pollen excites a local 
reaction of the nasal mucosa and if deeply inhaled may result in lesions in 
the trachea and bronchi. In food allergy, the lips and tongue may show 
the first vasomotor changes upon contact with the exciting substance. In 
bacterial anaphylactic asthma the changes are primarily in the bronchi. 
It is possible that the ordinary “common cold” is an anaphylactic reaction, 
the lesions occurring primarily in the nasal mucosa upon inhalation of the 
exciting agent. 
Probably a close analysis of many of the infectious diseases would show 
that some of the lesions and symptoms could reasonably be explained on the 
basis of anaphylaxis in a strict sense and as a cellular phenomenon without 
confusing the subject with the production of protein poisons. Doubtless 
the factors of antianaphylaxis by antibody protection, the effects of co- 
existence of antigen and sensitizin or allergic antibody in the cells and 
blood and desensitization would require consideration as modifying factors. 
At least the essential facts are apparent, namely, that the microparasites of 
disease contain proteins which may sensitize anaphylactically and some 
may produce a more active sensitinogen in the tissues. Under proper con- 
ditions it is reasonable to assume that anaphylactic shock or shocks may 
occur during the course of a given disease and the effects of these may 
contribute to the lesions and symptomatology in no small degree. 
The Relation of Anaphylaxis to Immunity.—Many adherents of the 
humoral theory of anaphylaxis believe that the same antibody or ferment 
mechanism capable of digesting the proteins of microparasites .with the 
production of the poisons regarded as responsible for anaphylaxis, may 
attack and “split” the proteins of the living parasites, bringing about the 
destruction of the latter and thereby indicating a very close relation of 
anaphylaxis to the phenomena of immunity and recovery from infectious 
disease. In other words, according to these views, either the anaphylactic 
antibody may be itself destructive of living microparasites or its production 
is accompanied by other antibodies having these properties. 
Romer1 and Sata,2 in experiments among cattle with Bacillus tuber- 
culosis, reached the conclusion that a state of hypersensitiveness meant a 
certain degree of resistance, while Krause3 and Austrian4 have expressed 
the opinion, based upon experiments, that sensitization of non-tuberculous 
animals with tubercle protein does not raise their resistance to experimental 
tuberculosis infection, and, indeed, may lower it. 
More recently Gay and Force5 have greatly renewed interest in this 
subject by advocating the anaphylactic skin test as a means of determining 
defensive activity following typhoid fever or active immunization by means 
of vaccines. Their first work was conducted with a “typhoidin” prepared 
in the same manner as Koch’s old tuberculin by cutaneous inoculation. 
Later Gay and Claypole6 prepared typhoidin by precipitating the solution 
1 Beitr. z. Klinik. d. Tuberk., 1908, xi, 79. 
2Ztschr. f. Tuberk., 1911, xviii, 1. 
3 Jour. Med. Research, 1911, xxiv, 361; ibid., 1916, 35, 1, 35. 
4 Bull. Johns Hopkins Hosp., 1913, xxiv, 11. 
5 Archiv. Int. Med., 1914, xiii, 471. 
6 Archiv. Int. Med., 1914, xiv, 671. 648 
ALLERGY IN RELATION TO INFECTION AND IMMUNITY 
with alcohol, washing the precipitate with alcohol and ether, drying in a 
vacuum, and suspending the resulting powder in phenolized normal salt 
solution, which was injected intracutaneously and applied cutaneously, a 
control powder being prepared from broth and used in the same manner. 
With this skin test Gay and his associates have studied the relative value 
of various vaccines and regard the anaphylactic reaction as indicative of a 
state of immunity. Nichols1 has questioned the value of the anaphylactic 
skin test as an index of immunity and regards the typhoidin reaction as 
indicating nothing more than sensitization to typhoid protein, which is 
apparently less lasting and less specific than the true immunity to this 
infection. 
The experiments of my associates and myself2 in this field have been 
largely tests in vitro for various antibodies, as agglutinins, bacteriolysins, 
and complement-fixing substances in the fresh sterile blood sera of persons 
and lower animals hypersensitive to various proteins (typhoid, syphilis, 
diphtheria, and canine distemper), and have shown that the state of hyper- 
sensitiveness to a particular bacterial protein bears no relation to the pres- 
ence or absence of demonstrable amounts of these antibodies. The experi- 
ments of Meyer3 have corroborated our results and conclusions, and have 
furthermore shown that a positive typhoidin skin reaction in a rabbit does 
not indicate the presence of resistance to an infection with Bacillus typhosus. 
The sum total of these studies indicate that although antibodies that 
may be regarded as possessing protective and curative properties toward a 
certain bacterium may be present in the body fluids of persons and animals 
hypersensitive to this particular protein, the condition of hypersensitiveness 
in itself is no direct evidence of their presence of resistance to a particular 
infection, although these antibodies are most likely to be present in the 
body fluids of those persons who are hypersensitive. The positive ana- 
phylactic skin test is, therefore, evidence of infection or sensitization to a 
particular protein probably without bearing any direct relation to resistance 
to infection or reinfection. 
This still leaves open the question, however, whether the anaphylactic 
antibody is capable of attacking and destroying living microparasites. It is 
well known that while a single or a few injections of bacteria may sensitize 
the cells of an animal that additional injections may render it anaphylac- 
tically immune. For this reason sensitization probably always occurs as an 
early step in immunization. The immunity apparently depends upon anti- 
bodies in the blood which neutralize the antigen before it reaches the cells. 
Manwaring and Kusama, for example, found that the lungs of a sensitized 
guinea-pig were protected against anaphylactic shock as long as they con- 
tained blood, but when thoroughly washed out they reacted upon the intro- 
duction of the antigen. This immunity of the cells is an example of anti- 
anaphylaxis by antibody protection; it is an immunity of the sensitized cells 
against the introduced protein. The question naturally arises: Are these 
antibodies responsible for immunity or protection against anaphylactic 
shock able to destroy or neutralize a living antigen in the form of a virulent 
microparasite or is this the function of the leukocytes and other defensive 
agencies, the anaphylactic antibody being powerless to kill, but capable of 
protecting the sensitized cells against the dead proteins? 
In a strict sense, therefore, the relation of anaphylaxis to the questions 
1 Jour. Exper. Med., 1915, xxii, 780. 
2 Jour. Immunology, 1916, 1, 409; ibid., 1916, 1, 429; ibid., 1916, 1, 443; ibid., 1916, 
1, 571. 
3 Jour. Infect. Dis., 1917, xx, 424. BACTERIAL ANAPHYLAXIS 
649 
of immunity and recovery from disease resolves itself into whether or not 
the sessile (cellular) or free anaphylactic antibody is capable of destroying 
living microparasites. Available data upon which to express a definite 
opinion is too meager, but the writer does not believe that such is the case. 
However, if the anaphylactic antibody is bactericidal, anaphylaxis and 
immunity bear a very intimate relationship. 
Curative Influence of Anaphylactic Shock.—Since protection of sen- 
sitized cells against anaphylactic shock appears to depend in part upon 
whether or not sufficient anaphylactic antibody is present in the blood for 
the neutralization of the antigen, it is apparent that shock or shocks may 
occur during the course of bacterial diseases when the antibody is not pres- 
ent or insufficient. The question then arises: Has anaphylactic shock any 
beneficial or curative effects? 
In all probability the reaction of congestion and exudation about foci of 
tuberculosis following the subcutaneous injection of a sufficient amount of 
tuberculin, is an anaphylactic reaction as well as in part a non-specific 
reaction (see Chapter XLI). Likewise similar reactions about foci of other 
bacterial infections following the administration of adequate amounts of 
bacterial vaccines. Are these reactions beneficial? Kraus has evidence that 
focal tuberculin reactions gradually result in an increase of fibrous tissue 
about the lesions and this is certainly desirable; it is commonly believed on 
the basis of clinical experience that focal reactions after the administration 
of bacterial vaccines are of good prognostic import. It would appear, there- 
fore, that when these reactions are mild and not too frequent that they 
are beneficial; when too violent they are doubtless very harmful as shown 
in the early days of tuberculin therapy with excessive doses, the lamentable 
days of the “tuberculin delirium.” 
Protein “shock” therapy may be under certain circumstances an ana- 
phylactic phenomenon when the antigen is specific; as ordinarily practised 
it is apparently not of an anaphylactic nature and will be further discussed 
in Chapter XXXIX with special reference to its therapeutic effects. CHAPTER XXX 
CLINICAL ALLERGY 
Importance and Frequency of Allergy in Man.—While in the early days 
of investigations in anaphylaxis it was believed that the manifestations in 
man were confined to those cases developing after the administration of horse- 
serum antitoxin, it is now known that man is subject to allergy by many 
different kinds of exciting agents and the subject has come to occupy a very 
important place in medicine. The manifestations of allergy are likewise 
many and diverse and in some instances unexplainable lesions and symp- 
toms of disease have been discovered to be allergic in nature. The list of 
exciting agents embraces not only the sera of the lower animals and espe- 
cially of the horse, but emanations from the skin, hair, and expired air of 
many of the lower animals; also different bacteria, fungi, and protozoa as well 
as pollens and juices of some plants, various foods of animal and vegetable 
origin, and drugs. Cooke and Vander Veer have estimated that the fre- 
quency of human sensitizations with clinical manifestations is probably not 
over 10 per cent.; the results of skin tests, however, show a larger percentage 
of individuals apparently hypersensitive to some protein or proteins although 
without symptoms, probably because the proteins as such never reach the 
sensitized cells. 
Natural and Acquired Allergy.—Is the allergic state or hypersensitiveness 
inheritable, that is, is there a variety of true natural allergy? Otto and 
Rosenau and Anderson have shown that mother guinea-pigs sensitized to 
horse-serum may transmit this anaphylaxis passively to their young, but in 
the latter the state of hypersensitiveness is of short duration and does not 
usually last more than six weeks. It has never been shown experimentally 
that a sensitized father guinea-pig transmits a sensitization. In other words, 
it would appear that a mother may passively transmit a transient hyper- 
sensitiveness by passage of the anaphylactic antibody through the placenta, 
but whether this occurs with human mothers is unknown. 
Whether a child may acquire active sensitization before birth is very 
doubtful; it would appear that placental transmission of the exciting agent 
resulting in prenatal active sensitization to a given substance is at least rare 
and unusual, although the occasional occurrence of hypersensitiveness to 
foods among infants suggests that this may occur. 
However, a tendency to acquire sensitization may be inherited and, indeed, 
constitutes a very important phase of human sensitizations, as shown by 
the investigations of Cooke and Vander Veer with special reference to hay- 
fever. When both father and mother had hay-fever, the children did not 
show any symptoms during the first few years constituting evidence against 
the actual transmission of the sensitization, but the transmission of the 
tendency to acquire sensitization is shown by the fact that 36.3 per cent, of 
such children showed symptoms before the fifth year of age. When one parent 
was hypersensitive about 14.3 per cent, of the group of children showed 
symptoms before the fifth year, while only 5 per cent, of a group in which the 
parents were healthy developed symptoms at this period. 
These studies have indicated that allergic individuals do not transmit 
to their offspring their own sensitizations, but an unusual capacity for acquiring 
sensitization to a variety of substances; this tendency is apparently inherited 
650 GENERAL LESIONS AND SYMPTOMS OF HUMAN ALLERGY 
651 
as a dominant characteristic. The nature of the inherited tendency is 
unknown; Cooke and Vander Veer have not found these individuals any 
easier to sensitize artificially than average individuals, so that the mechan- 
ism is dependent upon some other factor than the mere administration of a 
foreign protein. 
This subject may be summarized as follows: 
1. Passive sensitization by placental transmission of anaphylactic anti- 
body may occur among the lower animals and possibly among human beings; 
this sensitization is of short duration. 
2. Natural allergy by prenatal active sensitization does not occur at all, 
or, at least, is apparently a rare and unusual occurrence. 
3. The tendency to acquire sensitization is transmissible and especially 
if both parents are allergic to some substance or substances. 
4. Parents do not transmit their own hypersensitiveness, but a tendency 
to acquire hypersensitiveness to the same or different substances. 
5. The nature of the inherited tendency for acquiring hypersensitiveness 
is unknown. 
6. When the tendency for acquiring hypersensitiveness has been inherited, 
the individual may become allergic to many different substances during 
the course of a lifetime. 
The lesions of allergy in man are largely disturbances of the vasomotor 
system. Hyperemia with serous exudation and resulting edema are the 
most frequent, and according to the organ or organs involved, contribute 
most of the symptoms. 
The lesions and symptoms of allergy in man usually occur in the respira- 
tory, alimentary, and cutaneous systems. When the exciting agent is inhaled, 
as in pollen and dust allergies, or produced by bacteria upon the mucosa of 
the respiratory tract, the allergic lesions and symptoms are largely confined 
to these organs. Likewise in alimentary allergy swallowing of the exciting 
agent generally produces local lesions and symptoms, but may likewise elicit 
lesions and symptoms in the respiratory and cutaneous systems. 
Allergic Bronchial Asthma.—In 1910 Meltzer1 suggested that asthma 
may be in some cases an anaphylactic stenosis of the fine bronchi. Since 
then numerous investigators have shown this to be true. Goodale2 soon 
demonstrated the allergic nature of horse asthma by skin tests induced by 
the application of horse-serum to abrasions, and these observations were 
quickly followed by those of Talbot3 showing sensitization to eggs and milk 
as causes of asthma in children. The very extensive investigations of 
Walker soon followed and afforded conclusive proof of the allergic nature of 
some cases of asthma. 
Walker4 has found that the proteins of bacteria commonly found in the 
sputum of asthmatics may be the primary exciting agents, and notably 
Staphylococcus aureus. Matthews,5 Rackemann,6 and others have verified 
these observations demonstrating sensitization to the various strains of 
streptococci, staphylococci, pneumococci, and other bacteria by means of 
skin tests. 
Walker7 has also shown that asthma may be due to allergy to the pro- 
teins of horse hair and dandruff as well as to the hair, and emanations of the 
GENERAL LESIONS AND SYMPTOMS OF HUMAN ALLERGY 
1 Jour. Amer. Med. Assoc., 1910, 55, 1021. 
2 Boston Med. and Surg. Jour., 1914, 171, 695. 
3 Boston Med. and Surg. Jour., 1914, 171, 708. 
4 Jour. Med. Res., 1916-17, 35, 487. 
B Medical Record, 1913, 84, 507. 
6 Jour. Immunology, 1920, 5, 373. 
7 Jour. Med. Res., 1916, 35, 497. 652 
CLINICAL ALLERGY 
cat, dog, rabbit, mouse, and other of the lower animals, including the feathers 
of chickens and geese; Ratner1 and Larsen and Bell2 have reported asthmas 
in children due to the inhalation of rabbit hair. These individuals suffer with 
hay-fever-like symptoms when in close proximity to the animal to which sen- 
sitization exists, and in not a few cases persistent bronchitis and asthma have 
been found due to the use of pillows stuffed with the feathers or hair of ani- 
mals to which individuals were allergic. Gerdon3 has recently described in- 
stances of asthma among hair dyers which he ascribed to the inhalation of 
the dyes belonging to certain p-phenyldiamin derivatives. 
Walker4 has also observed occasional cases of asthma due to allergy to 
common foods including those of animal, vegetable, and fruit origin. Talbot 
and Schloss were among the first to describe cases of food allergy among chil- 
dren with asthmatic attacks and particularly in allergy to eggs. Indeed, 
the allergic asthmas of children are usually due to foods rather than to 
bacterial proteins. 
Probably the asthma due to pollen allergy is best known, and was des- 
cribed by Blakely and other pioneer investigators in hay-fever. This type 
of asthma is usually present during the spring and autumn hay-fever seasons; 
not all persons subject to pollen allergy suffer with asthma, this symptom 
being present only in those who are especially sensitive. 
Even drug allergy may be accompanied by asthma; I know of a phys- 
ician who suffers with eczema and asthma a few hours after handling ars- 
phenamin powder. 
Aside from these well known causes, cases of asthma are frequently en- 
countered apparently belonging to this group etiologically, but without 
the exciting agent being discoverable. For example, I know of several cases 
of asthma in which the attacks are brought on by the inhalation of coal 
smoke and gas, the exciting agent being either carbon particles or gaseous 
substances. Others are rendered asthmatic by street and house dust, etc. 
These cases are generally included in the group of dust asthmas and Cooke5 
has recently shown the presence of a substance in most house dusts bearing 
an important relation to asthmas of this origin; hay dust was also found cap- 
able of acting as an allergen and not solely a simple mechanical irritant. 
Bronchial asthma is, therefore, a common symptom of allergy in man. 
It is a type of dyspnea due to difficult inspiration or expiration, or both, and 
apparently caused not only by contraction of the smaller bronchi, but to 
hyperemia and edema of the mucosa as well. Of course not all asthmas are 
of allergic origin; asthma of renal, cardiac, and thymic causes, as well as to 
enlargement of the bronchial glands, are to be carefully differentiated. The 
causes of allergic asthma, however, are so many and diverse that this type of 
asthma should always be kept in mind; these may be divided into four 
groups as shown on page 653. 
xAmer. Jour. Dis. Child., 1922, 24, 346. 
2 Amer. Jour. Dis. Child., 1922, 24, 441. 
3 Ztschr. f. Gewerbehygiene, 1920, 8, 201. 
4 Jour. Med. Research, 1916, 35, 509; ibid., 1917, 36, 231. 
5 Jour. Immunology, 1922, 7, 147. GENERAL LESIONS AND SYMPTOMS OF HUMAN ALLERGY 
653 
(a) Pollens—Hay-fever. 
1. Dandruff and hair of horse, 
cat, dog, etc. 
2. Feathers. 
3. Flour dust. 
4. Orris root. 
5. Face powders and especially 
rice powder. 
6. House and hay dusts. 
I. Inspiratory Type 
due to: 
(b) Animal emanations and 
dusts. 
(c) Drugs, as arsphenamin dust. 
(a) Various meats. 
(b) Fish and shell-fish. 
(c) Milk and eggs. 
(d) Grains, as oats, wheat, rye, etc. 
(e) Vegetables, as peas, beans, potato, etc. 
(/) Fruits, as orange, banana, etc. 
(g) Nuts, as walnut, pecan, etc. 
(.h) Drugs. 
II. Alimentary Type 
due to foods: 
III. Bacterial Type 
due to: 
(a) Bacteria in the upper and lower respiratory tracts. 
(b) Foci of bacteria and mucoid secretions in the nose and accessory 
sinuses, etc. 
(a) Subcutaneous and intravenous injections of horse-, sheep-, or 
i other alien serum or blood. 
(b) Injections of drugs. 
IV. Parenteral Type 
due to: 
Allergic Rhinitis.—Closely allied to allergic bronchitis and asthma is 
perennial vasomotor rhinitis, in which the physical findings in the nose are 
quite similar to those during acute hay-fever. The condition occurs from 
day to day and week to week regardless of the environment or occupation of 
the patient. The symptoms are largely profuse watery discharge from the 
nose wdth respiratory obstruction due to hyperemia and edema of the 
mucosa. Walker1 has recently described 20 cases due to allergy to emana- 
tions from horses, 6 cases due to cats, and a number to feathers. He speaks 
of similar symptoms among laboratory workers due to rabbit and guinea- 
pig hair, likewise of symptoms in bakers and housewives due to cereal 
flours. I have seen 2 cases in women due to allergy to rice powder em- 
ployed as a cosmetic. 
Rackemann2 has recently reported that 45, or 22.7 per cent., of a group 
of 198 patients wfith this ailment yielded positive skin tests to various sub- 
stances including 13 sensitive to orris powder commonly employed in den- 
tifrices, 5 to feathers, 5 to one or more of the cereals, etc. 
Analogous to allergic rhinitis are cases of vasomotor conjunctivitis, of which 
vernal conjunctivitis is a type. The symptoms are mainly hyperemia with 
profuse lacrimation in which many eosinophil cells are present. The presence 
of the latter are always suggestive of an anaphylactic reaction., Vernal 
conjunctivitis is particularly likely to occur during the spring months of the 
year and may be an allergic lesion. 
Allergic Skin Lesions (Urticaria, Angioneurotic Edema, Eczema, Ery- 
thema Multiforme).—Allergy is one of several causes of urticaria and angio- 
neurotic edema. Probably the urticaria of serum allergy (serum sickness) is 
best known in this connection and has been thoroughly studied and de- 
scribed by von Pirquet and Schick and others. A common mistake, how- 
ever, is to place too much importance upon allergy as a cause of chronic 
urticaria. Schloss3 has recently reported that only 10 cases of a series of 50 
yielded positive skin reactions to food proteins, and 6 of these reacted to 
1 Jour. Amer. Med. Assoc., 1920, 75, 782. 
2 Amer. Jour. Med. Sci., 1922, clxiii, 87. 
3 Amer. Jour. Dis. Child., 1920, 19, 445. 654 
CLINICAL ALLERGY 
many proteins of the common foods; tests were not made with bacterial or 
pollen proteins or at different intervals. According to McBride and Schorer,1 
allergies to fish, tomatoes, and cheese are especially likely to show urticaria, 
while the cereals and pork produce erythemata in a considerable proportion 
of cases. 
Cases of angioneurotic edema (Quincke’s disease) are sometimes due to 
allergy and particularly to foods. Schloss found 3 cases in a series of 11 
yielding positive skin reactions, one to milk and wheat and the other to egg 
and beef. The causal relationship was demonstrated in both by dietetic 
experiment. Both patients suffered from recurring attacks of local edema 
of skin recurring every few weeks and lasting for a day or two. Highman 
and Michael2 have studied 14 cases of chronic urticaria and angioneurotic 
edema, and in a thorough and splendid paper have expressed their belief in 
allergy to different proteins, and especially foods, as one cause for these con- 
ditions. Phillip3 has recently described cases occurring in young dogs caused 
by food allergies. 
Next to urticaria probably eczema, and especially eczema of children, has 
attracted most attention as an allergic lesion. The relationship of eczema to 
allergy has been largely studied by means of skin tests, and a common error 
has been to conclude that the eczema was allergic when positive reactions 
were observed with the allergens of one or more foods. There can be no 
doubt that some cases are of an allergic nature, but not in as high propor- 
tion as commonly believed. 
In 1916 Strickler and Goldberg4 found positive skin reactions in a small 
group of cases to various food allergens. White5 reported 66 per cent, of 
positive reactions in a group of cases, and Talbot6 observed fourteen positive 
reactions to egg-white and other food proteins in a series of 16 cases of eczema 
in infants and children. Schloss7 observed that 77.4 per cent, of a series of 
cases in children under sixteen months of age yielded positive skin reactions 
to various food allergens; among children over sixteen months of age 41.6 
per cent, yielded positive reactions. O’Keefe8 applied the skin tests in 70 
cases of infantile eczema, and found that 29, or 41 per cent., gave a positive 
reaction to one or more foods, egg-white being most commonly observed. 
Blackfan9 has called attention to the fact that many asthmatic children 
suffer from eczema in infancy or early childhood. Of 43 patients without 
eczema, only one showed susceptibility to protein by skin tests, while of 27 
patients with eczema, 22 yielded positive reactions, and particularly to 
cow’s milk and woman’s milk. Usually positive reactions were observed to 
several proteins. Ramirez10 has recently reported that 38 of a series of 78 
cases of eczema yielded positive skin reactions to one or more proteins. 
These investigations leave no doubt but that some cases of eczema are 
due to food allergy, but as pointed out by Fox and Fisher,11 the significance 
of positive skin reactions in these cases is by no means clear; children may 
react in a positive manner to food allergens which produce no evidences of ali- 
1 Jour. Cutan. Dis., 1916, 34, 55. 
2 Archiv. Dermat. and Syph., 1920, 2, 544. 
3 Jour. Amer. Med. Assoc., 1916, 66, 249. 
4 Jour. Amer. Med. Assoc., 1922, 78, 497. 
6 Boston Med. and Surg. Jour., 1918, 175, 5. 
6 Med. Clinics North America, 1918, 985. 
7 Amer. Dis. Child., 1920, 19, 433. 
8 Boston Med. and Surg. Jour., 1920, 183, 569. 
8 Amer. Jour. Med. Sci., 1920, clx, 341. 
10 Archiv. Dermat. and Syph., 1920, 2, 365. 
11 Jour. Amer. Med. Assoc., 1920, 75, 907. GENERAL LESIONS AND SYMPTOMS OF HUMAN ALLERGY 
655 
mentary disturbance. Not infrequently and, indeed, usually, reactions occur 
to more than one food, and withdrawal of these articles from the diet is not 
always followed by improvement. Up to the present time allergic eczemas 
have been identified almost solely with food substances; in children doubt- 
less these are of most importance, but in adults bacterial and other allergens 
are to be kept in mind as bearing a possible relationship to the lesions and 
symptoms. Eczema in adults is sometimes caused by allergy to a drug; I 
know of a physician who suffers with eczema of the hands after preparing 
solutions of arsphenamin and neo-arsphenamin. 
Various erythemas may be caused by allergies to foods; sometimes serum 
allergy is accompanied by scarlatiniform rashes. Hazen1 has described a 
case of severe erythema multiforme due to allergy to the oyster. 
Skin rashes are especially common in drug allergies, and may be macular, 
papular, vesicular, urticarial, bullous, or hemorrhagic. 
Allergic Vomiting, Diarrhea, and Abdominal Pain.—In allergies to foods 
the ingestion of the allergen may produce almost immediately swelling of 
the lips and tongue and presumably of the stomach and intestines unless 
vomiting occurs. In less sensitive individuals symptoms of nausea, vomiting, 
flatulence, abdominal pain, and diarrhea may develop within a few hours, due, 
presumably, to hyperemia and edema of the gastro-intestinal mucosa. 
Richet and Girons2 have drawn particular attention to these cases of ali- 
mentary anaphylaxis, and Duke3 has recently described a series of cases in 
which abdominal pain was apparently caused by intestinal allergy. 
Allergy to Serum (Serum Sickness).—With the advent to antitoxin 
treatment of diphtheria in 1894 the injection of immune horse-serum quickly 
became a common practice in the treatment of disease; almost immediately 
cases of untoward effects were observed and reported, although occasional 
cases were previously recorded following transfusion with lamb’s blood. 
Among the earliest cases of serum disease following the injection of diphtheria 
antitoxin was the immediate and fatal attack of a son of Professor Langer- 
hans. This case created a profound impression upon the medical pro- 
fession and at the present time many physicians, fearing anaphylaxis, hesi- 
tate to administer diphtheria antitoxin and other sera for prophylactic and 
therapeutic purposes, these fears being shared by many of the laity. 
The Nature of Serum Disease or Serum Sickness.—Inasmuch as the 
reactions followed the injection of horse-serum, the condition became known 
as “serum disease.” 
This name was applied by von Pirquet and Schick4 to the various clinical 
manifestations, such as eruptions, fever, edema, and pain in the joints, follow- 
ing the injection of horse-serum. These symptoms are due to the horse-serum 
itself, for, as was early shown by Johannessen,5 Bokay,6 and others, they 
may manifest themselves after the injection of sterile normal horse-serum. 
The serum, moreover, of certain horses appears to be more likely than that 
of others to cause these symptoms, thus accounting for the fact that one 
lot of antitoxin will cause a higher percentage of serum sickness than will 
another. A concentrated serum is not so likely to produce serum sickness 
as whole serum, owing partly to the fact that smaller doses of it are given. 
According to Rolleston and Ker, the frequency of serum sickness is, as a 
1 Jour. Amer. Med. Assoc., 1914, lxii, 695. 
2 Bull. d. l’Acad. d. Med., 1920, 34, 625. 
3 Arch. Int. Med., 1921, 28, 151. 
4 Die Serumkrankheit, Leipsic, Deuticke, 1905. 
5 Deutsch. med. Wchn., 1895, 21, 855. 
6 Jahresb. f. Kinderh., 1897, xliv, 133. 656 
CLINICAL ALLERGY 
rule, in direct proportion to the amount of serum given, and in inverse 
ratio to the severity of the attack; in other words, we may expect to en- 
counter it most often in mild and moderately severe cases that have received 
very liberal dosages of serum. 
Serum sickness is regarded as a true anaphylactic phenomenon. We are 
prone to call the severe, fatal, and rare instances of death following serum 
injection examples of anaphylaxis, and to regard serum sickness as a differ- 
ent condition. Both are fundamentally the same, except that in the first 
instance the body cells are for some unknown reason unduly and highly 
anaphylactic. Fortunately, this undue hypersensitiveness is frequently fore- 
shadowed by the asthmatic or hay-fever-like attacks which the susceptible person 
may exhibit when he enters stables or is otherwise around horses. It goes 
without saying that horse-serum should never be given to such persons. 
While serum sickness is usually due to horse-serum for the reason that 
the horse is so commonly employed in the preparation of various curative 
serums, the serum of the ox, rabbit, and other animals may induce the 
same train of symptoms in addition, in some instances, to producing a 
direct toxic effect. 
Types of Sero-anaphylactic Reactions.—An immediate reaction rarely 
follows the first injection of serum unless the patient is one of those unfor- 
tunate but rare persons who in some manner have been rendered highly 
sensitive to horse protein. In the majority of instances symptoms do not 
develop for from eight to twelve days, during which time the antibody is 
being produced; this is the usual delayed reaction. When antibody forma- 
tion has reached a certain point it reacts upon any of the horse-serum 
that may persist in the cells or circulation, producing the anaphylactic 
reaction. If the dose of serum has been small, antibody formation goes 
on as usual, but the serum may not persist in the circulation. Hence symp- 
toms do not develop in such a person, although if reinjected subsequently 
symptoms will appear. This is one reason wThy a concentrated serum is 
not so likely to produce serum sickness, since a smaller quantity of it is 
injected. 
If, however, the patient has received an injection of serum some months 
previously, a reinjection is likely to be followed by an immediate reaction, 
that is, the symptoms appear within from a few minutes to twenty-four to 
forty-eight hours. If the first injection had been given a year or more pre- 
viously, no antibody may be present in the blood, so that an immediate 
reaction does not occur. If, however, the cells once stimulated are “keyed 
up” indefinitely, and, accordingly, antibody formation is quite rapid, so 
that we find symptoms developing in from four to seven days after injec- 
tion—the accelerated reaction. Or a small amount of antibody may be 
present wdiich gives a mild reaction around the site of serum injection, fol- 
lowed in from four to seven days by a general reaction, this being an im- 
mediate followed by the accelerated reaction. 
As was previously stated, anaphylaxis is specific—that is, a person 
receiving ox-serum in the first injection would not be affected subsequently 
by an injection of horse-serum, but only by ox-serum. For this reason it 
has been recommended that diphtheria antitoxin to be used for prophylactic 
purposes should be prepared by immunizing cattle, reserving the horse-serum 
for treatment if the disease should be contracted subsequently. 
Relation of Route of Administration of Serum to Anaphylactic Reac- 
tions.—The route by which §erum is administered to man bears a very 
important relation to both the severity and time of appearance of the symp- 
toms of anaphylaxis. GENERAL lesions and symptoms of human ALLERGY 
657 
Acute anaphylactic shock of the lower animals is produced by the sudden 
introduction of serum into sensitized animals by intravenous injections. 
The same is true of human beings. The intravenous and subarachnoid in- 
jections of serum produce anaphylactic effects more quickly than intra- 
muscular and subcutaneous injections, while interal administration by 
mouth or colonic injection are least likely of all to produce these effects. 
With those exceptional human beings who are very highly sensitized to 
horse protein as the “horse asthmatics,” even the subcutaneous injection of 
serum may result in an immediate reaction developing within a few minutes, 
but with the great majority of persons the subcutaneous or intramuscular 
injection of horse-serum is followed by a period of several days to two weeks 
or more before the onset of symptoms. 
Relation of Amount of Serum Administered to Anaphylactic Reactions.— 
Ordinarily the administration of purified or concentrated serum does not 
produce as high an incidence of serum sickness as whole serum. Horse- 
serum contains at least three sensitinogens: euglobulin, pseudoglobulin, 
and albumin. Concentrated antitoxin sera prepared by the Gibson-Banzhof 
method is composed largely of the pseudoglobulins which carry most of the 
antitoxin; the employment of these concentrated sera means that there is 
a reduction of the total protein in relation to the antitoxic value and the 
elimination of all or most of the euglobulins and albumins lessens very ma- 
terially the chances of inducing serum sickness and the production of recur- 
rent eruptions. 
Even with whole serum, smaller amounts, injected subcutaneously or in- 
tramuscularly, are less likely to produce serum sickness than larger amounts. 
With intravenous injections there is less relation and among “horse asth- 
matics” a few centimeters by any route may induce as much reaction as 
larger amounts. 
Danger of Anaphylactic Reactions.—The occurrence of immediate reac- 
tions and fatalities after the administration of horse immune sera employed 
for prophylactic and curative purposes, has unduly impressed many phys- 
icians and engendered a fear of producing reactions which is shared by many 
of the laity; one result has been hesitation in the administration of diphtheria 
and tetanus antitoxins for the prophylaxis and treatment of these infections. 
For this reason a general and brief statement on the dangers of anaphylaxis 
in man may be in order: 
1. Bearing in mind that thousands upon thousands of injections of 
horse immune sera and especially diphtheria and tetanus antitoxins are ad- 
ministered throughout the world each year, the incidence of fatal anaphy- 
lactic reactions is very small (probably about 1 in 50,000 to 100,000 injec- 
tions). The exact incidence cannot be stated because not all cases are re- 
corded, but several investigators have reported the absence of fatal reactions 
in series as large as 200,000 injections. Furthermore, some of the fatalities 
attributed to the anaphylactic effects of the serum may have been caused 
by other factors. Approximately 50 fatalities after the administration of 
serum have been recorded. 
The incidence of severe reactions during or immediately after the intra- 
venous or subarachnoid injections of serum is much larger, but recovery is 
the rule. Likewise the incidence of mild anaphylaxis designated as “serum 
sickness” or “serum disease” is high, but this reaction is without danger. 
2. Individuals, including children as well as adults, who are uncomfort- 
able around horses and are seized with sneezing, coughing, “asthma” and 
other hay-fever-like symptoms, are extremely sensitive to horse protein 
and should not receive injections of horse-serum by any route. Apparently 658 
CLINICAL ALLERGY 
some of the small group of fatalities following the administration of serum 
have occurred among these persons. A physician called upon to administer 
serum should always first inquire upon this point; if the history is sug- 
gestive a skin test may be conducted, as described in a succeeding chapter. 
Desensitization may likewise be tried as described later, but in a typical 
“horse asthmatic” the administration of horse-serum is dangerous despite 
these precautions. 
Children with status-thymo-lymphaticus are likewise bad risks; this 
condition is regarded by some investigators as responsible for some of the 
fatalities following the administration of serum. 
3. Individuals who have received injections of serum upon a previous 
occasion are almost sure to develop “serum sickness” after reinjection, but 
this is not dangerous, although distressing, and should not be a contradic- 
tion to the use of serum. Serum may be administered by subcutaneous or 
intramuscular injection with safety. 
When serum is given intravenously to such individuals, preliminary 
desensitization should be practised. This is likewise advisable before the 
intraspinal injection of serum. Methods are given in a subsequent chapter. 
Symptoms of Immediate and Severe Anaphylactic Shock.—These are 
characterized by sudden onset—often within a few minutes after or during 
the intravenous injection of serum. The patient becomes restless, anxious, 
and may give an outcry; there is marked pallor, perspiration, rapid and feeble 
pulse. Respirations become deep, labored and sometimes rapid, and the 
patient may become unconscious. Muscular twitchings, rigors, and con- 
vulsions may occur with micturition and defecation. In rare instances 
death occurs during coma with respiratory failure, while the heart is still 
beating. 
Some of these symptoms may develop during the intravenous injection 
of serum, but subside as soon as the injection is stopped. Death does not 
occur except in those rare cases of extreme sensitizations to horse protein; 
in these even the subcutaneous injection of serum may bring on the reaction 
with a fatal outcome. 
Not infrequently a reaction of fever, chills, rapid pulse, and prostration 
develop an hour or two after the intravenous injection of serum; this reaction 
is probably not anaphylactic, but protein shock. 
Symptoms of the Usual Reaction (Serum Disease).—The most typical 
of these symptoms are rash, fever, and prostration, besides joint and muscle 
pains, edema, and adenitis. 
These symptoms and lesions may develop within a few hours to two 
weeks or so after the intravenous, intraspinal, intramuscular, or subcutaneous 
injection of serum. In the majority of cases the reaction develops eight to 
twelve days after the injection, although it may appear within the first 
few days and especially if the individual had received an injection of serum 
on a previous occasion. 
Frequency.—The frequency of these reactions varies greatly in the dif- 
ferent recorded series of cases. It would appear that the incidence in Ger- 
many (8.1 per cent.) and France is much lower than in America. The inci- 
dence in this country has varied with the sera of different horses and has 
been generally lower with concentrated (purified) sera than with whole 
sera. 
Reactions are slightly more frequent after intrathecal injections of 
serum than after subcutaneous injections. The following table gives the 
approximate incidence from different parts of the world: Fig. 152.—Urticarial Rash of Serum Sickness. 
Case of laryngeal diphtheria; had received 40,000 units of antitoxin eight days previously; urticarial 
rash first appeared about thirty hours before this drawing was made. GENERAL LESIONS AND SYMPTOMS OF HUMAN ALLERGY 
659 
INCIDENCE OF SERUM DISEASE 
Reported by 
Route of Administration. 
Serum. 
Incidence, Per Cent. 
Flexner 
Intrathecal 
Whole 
70 
Rolleston 
ti 
it 
41 to 60 
Drought 
U 
a 
55 
Goodall 
Subcutaneous 
It 
35 to 40 
Hartung 
it 
a 
11.4 
Rolleston 
U 
u 
44 
Dant 
a 
a 
11.4 
Coca 
it 
Purified 
8.5 
Kolmer 
it 
10 
In a general way it may be stated that the administration of whole horse- 
serum by subcutaneous, intramuscular, intravenous, and intrathecal injec- 
tion is followed by serum disease in 40 to 60 per cent, of cases; the subcuta- 
neous or intramuscular injection of purified horse antitoxin serum is followed 
by serum disease in approximately 10 per cent, of cases. 
Incubation Period.—Symptoms may appear in fifteen minutes after the 
intravenous injection of whole serum. In about 10 per cent, of cases de- 
veloping serum sickness the symptoms appear within the first four days; the 
majority (about 60 per cent.) of cases develop symptoms in from six to ten 
days after injection and these are the critical days. 
Fever.—After the intravenous injection of serum a thermal reaction may 
develop within one hour or two, but this is usually due to non-specific pro- 
tein shock rather than to anaphylaxis. In serum disease the temperature 
usually rises one or two days before the rash or on the same day just as the 
rash appears; it is well for the physician to keep this in mind in order to 
guard against concluding that this fever is due to reinfection or exacerba- 
tion of infection. Sometimes fever is absent and especially in mild cases. 
Eruption.—The most obvious and important of the symptoms are un- 
doubtedly the various forms of rash. In 1000 consecutive cases of diph- 
theria treated with antitoxin in the Philadelphia Hospital for Contagious 
Disease, a rash developed in 430, or 43 per cent. The time of appearance 
of the eruption depends, as was just stated, upon whether or not the patient 
has been injected on a previous occasion, and if so, the length of the interval 
between the first and subsequent injection, and to a lesser extent upon the 
amount of the first injection, these factors influencing the quantity of anti- 
body present at the time of reinjection. Of the 430 cases just mentioned, 
the time of appearance of the rash, in days, after subcutaneous injection of 
antitoxin was as shown in table on page 660. 
In that table are included some cases of reinjection, as, e. g., scarlet 
fever patients who received a routine immunizing dose of antitoxin upon 
admission, and another after having contracted diphtheria; also cases of 
diphtheria that became reinfected within a few months after their discharge 
from the hospital. As will be seen, about 63 per cent, of cases develop a 
rash between the sixth and the ninth day after the injection of antitoxin. 
Three main types of rashes are generally recognized: 
1. Urticarial Rashes.—If we include in this group all eruptions that 
present a resemblance to urticaria, these rashes are the most common, 
constituting from 70 to 90 per cent, of all eruptions. They usually appear 
after the seventh day, becoming manifest first about the site of injection. 
Large, irregularly shaped, and scattered blotches appear, frequently with 
true wheals in the center (Fig. 152), accompanied by intense itching and 660 
CLINICAL ALLERGY 
TIME OF APPEARANCE OF SERUM RASH IN 430 CASES OF SERUM DISEASE 
Day Upon Which the Rash Appeared After Injection of 
Diphtheria Antitoxin. 
Total Number 
Showing Rash. 
Percentage. 
First 
4 
0.9 
Second 
6 
1.4 
Third 
7 
1.6 
Fourth 
30 
6.9 
Fifth 
35 
8.1 
Sixth 
75 
17.6 
Seventh 
65 
15.1 
Eighth 
85 
19.8 
Ninth 
44 
10.2 
Tenth 
39 
9 0 
Eleventh 
22 
5.1 
Twelfth 
5 
1.1 
Thirteenth 
6 
1.4 
Fourteenth 
5 
1.1 
Fifteenth 
1 
0.2 
Sixteenth 
1 
0 2 
irritation. Sometimes true wheals do not appear. The rash is often very 
profuse, and fresh blotches may continue to appear for two or three days. 
Occasionally, the rash is quite sparse and mild, and may disappear within 
twenty-four hours. 
2. Multiform Rashes.—This type of rash is quite common. It is often 
circinate in its arrangement, or occurs in large blotches mixed with a scat- 
tered morbilliform or measly form of rash (Fig. 153). Different parts of 
the body may present different appearances at the same time. This rash 
may occasionally closely simulate true measles, especially since it involves 
the face, and as the conjunctivae are likely to be congested in almost any 
variety of serum sickness. It is differentiated from measles by the fact 
that Koplik’s spots are absent, there is no prodromal rise in temperature, 
the papules are not elevated above the skin as much as in measles, and that 
the eruption frequently starts from the site of injection, instead of on the face. 
3. Scarlatiniform Rashes.—This type of rash occasionally occurs, and 
may bear so close a resemblance to true scarlet fever as to be indistinguish- 
able from it. The eruption usually appears early, that is, in from the first 
to the sixth day after injection, and may vary from a uniform erythema 
which first appears about the site of injection, to a true punctate scarlatinal 
rash. The differentiation of these rashes from the true scarlet fever erup- 
tion is one of the greatest sources of trouble in hospital practice, especially 
when they develop in the diphtheria wards, where occasional cases of true 
scarlet fever are always likely to appear from time to time. The absence of 
the following symptoms, or their occurrence only in mild degree, would favor 
a diagnosis of serum disease: Fever, or, at least, the presence of but a mild 
pyrexia, the vomiting, the typically furred tongue, the angina, or at least 
but a mild throat involvement, and the leukocytic inclusion bodies—all 
forming the symptom-complex of true scarlatina. In many instances, how- 
ever, it is necessary to isolate the patient, when speedy recovery, absence of 
complications, and less definite desquamation, which does not involve the 
palms and soles, indicate that the patient was suffering from serum disease 
and not from scarlet fever. 
As previously stated, the eruption is at first usually local about the site of 
injection of serum and several days may elapse before it becomes generalized. 
In some mild cases the eruption remains localized. Recurrent eruptions Fig. 153.—Multiform Rash of Serum Sickness. 
Child with laryngeal diphtheria on the sixth day after receiving 65,000 units of antitoxin. 661 
sometimes occur. Occasionally there may be considerable edema present 
with the skin rash, or it may occur independently. It is most common about 
the face and neck and may be so extensive as to close the eyes; this usually 
lasts only a few hours. 
Arthritis; Neuritis; Myalgia.—Painful and tender joints occur in about 
2 per cent, of cases. The process seems to be extra-articular; there is seldom 
any fluid in the joints, nor is the skin hot or reddened. There may be slight 
periarticular edema. Boots and Swift,1 however, have recently stated that 
fluid was found in the joints of four of six cases studied and that there were 
cytological changes of acute arthritis. Furthermore, horse-serum was found 
in two fluids, suggesting “that the irritation of the joint may be due to the 
presence of this foreign protein in an allergic tissue.” All the joints may be 
attacked or only one or two; the temporomandibular and metacarpophalan- 
geal joints appear to be more frequently affected than other joints. Neuritis, 
myalgia, and headache may also occur. 
Adenitis.—The regional lymph glands and especially those nearest the 
site of subcutaneous injection of serum frequently become swollen and 
tender. These lesions are among the first to appear and first to disappear. 
Generalized adenitis may develop. Suppuration does not occur. Occa- 
sionally the cervical glands beneath the sternomastoid muscle may become 
so enlarged and tender as to suggest a pyogenic infection, but this usually 
disappears in two to three days. Occasionally the spleen becomes palpable. 
Leukocyte Changes.—In the first few hours of serum disease there is 
likely to be a slight leukocytosis with an increase in the relative number 
of lymphocytes (30 to 40 per cent); this is followed by leukopenia. Fre- 
quently there is a slight increase of the eosinophil cells (3 to 7 per cent.). 
General Symptoms.—Tachycardia may occur even when the temperature 
is approximately normal. Diarrhea may develop and prove a troublesome 
symptom for several days. There is rarely nausea and vomiting. Marked 
prostration of the neuromuscular system may occur followed by an asthenia 
of many days or weeks. 
Mild nephritis with slight albuminuria is sometimes present. Longcope 
found about 10 per cent, of cases showed albuminuria and casts. There is 
little change in the coefficient of urea excretion or in the phthalein output. 
The excretion of water and chlorids are rapidly diminished at the onset, but 
these renal changes are transitory and the return to normal is usually rapid. 
Delahet2 has recently described meningeal symptoms of anaphylactic 
nature following repeated intrathecal injections of serum. 
Recurrent Serum Disease.—Mention has already been made of the “double 
reactions” of von Pirquet and Schick, the “accelerated” following the “im- 
mediate” reaction. A number of observers have reported second eruptions 
following in three to twenty-one days after the first eruption. Goodall3 
has described 2 cases of triple eruptions; in the first case the eruptions ap- 
peared on the first, third, and seventh days, and in the second case they oc- 
curred on the second, seventh, and tenth days following reinjection. Axenow4 
reports that of 683 cases of serum disease, 10 per cent, developed a triple erup- 
tion and 2.3 per cent, a quadruple eruption. 
These recurrent eruptions appear to be caused by more than one sensi- 
tinogen in the serum; as reported by Coca they do not occur with the use of 
purified (concentrated) antitoxin sera. 
GENERAL LESIONS AND SYMPTOMS OF HUMAN ALLERGY 
1 Jour. Amer. Med. Assoc., 1923, 80, 12. 
2 Bull. d. 1. Soc. Med. d. H6p., 1920, 44, 1272. 
3 Jour. Hyg., 1907, 7, 607. 
4 Jahrb. f. Kinderh., 1913, Ixxviii, 565. 662 
CLINICAL ALLERGY 
The Detection, Prevention, and Treatment of Serum Disease.—Anaphy- 
lactic phenomena can usually be expected to occur if serum is given within 
a year or so following a previous injection. Persons who experience discom- 
fort when about horses should always be closely questioned. By means of 
skin tests the state of allergy to serum may be detected and especially in those 
cases of extreme hypersensitiveness to horse protein. These preliminary tests 
are particularly indicated before the intravenous and intrathecal injection 
of serum when the patient is known to have had an injection of serum upon 
a former occasion. When serum is to be given subcutaneously or intra- 
muscularly they are not necessary unless the physician suspects that the 
patient is hypersensitive to horse protein. The technic and a fuller dis- 
cussion are given in the succeeeding chapter. 
Sero-anaphylactic reactions may be prevented or at least greatly reduced 
in severity by preliminary desensitization before serum is administered. This 
is particularly indicated before the intravenous and intrathecal injection of 
serum. Methods are described in Chapter XXXII. 
For the immediate reaction sometimes developing within fifteen minutes 
after the administration of serum and especially by intravenous and intra- 
thecal injection, adrenalin chlorid (1 : 1000) should be injected intramusT 
cularly in dose of 1 c.c. (rtjjxvj). Atropin sulphate should also be injected 
subcutaneously in dose of 0.5 mg. (1/120 grain). Whenever serum is given 
intravenously it is a good routine practice to have these two medicaments ready 
in syringes for immediate use; usually they are not required, but if needed, 
no time is lost in the administrations. If the pulse is particularly soft and 
rapid stropanthin may be given intramuscularly in dose of 0.5 mg. (1/120 
grain) or an equivalent preparation of digitalis. 
In the usual case of serum sickness, the urticarial rash is the most dis- 
tressing symptom. A brisk cathartic should be given. The subcutaneous 
injection of adrenalin chlorid 1 : 1000 in dose of \ to 1 c.c. (irj)vij to xvj) 
often affords instant relief which may persist for several hours. I usually 
give an injection late in the evening in order that the patient may have a 
few hours of sleep. Calcium chlorid and lactate, in doses of from 3 to 5 
grains, have been advocated as a prophylactic and cure, but they are of 
doubtful utility. Sometimes the administration of a sedative dose of 
morphin sulphate is indicated along with an injection of adrenalin. A sooth- 
ing lotion should be applied to the skin, the following well-known calamin 
lotion being efficacious: 
Phenol 2 c.c. 
Calamin 4 gm. 
Zinc oxid 8 gm. 
Glycerol 12 c.c. 
Lime-water 16 c.c. 
Water up to 120 c.c.—M. 
Aspirin may be administered for headache, myalgia, and neuritis. Aspirin 
or sodium salicylate combined with sodium bicarbonate may be given every 
four to six hours for the joint pains. 
ALLERGY TO POLLENS (HAY-FEVER) 
The hypersensitiveness displayed by some individuals to the pollens of 
various plants is one of the earliest and best known examples of human 
allergy. 
Historic.—In an excellent review of the early literature upon this sub- 
ject by Hitchens and Brown,1 mention is made of clinical observations by 
Botallus as early as 1565, who wrote: “For there are many who are attacked 
1 Jour. Lab. and Clin. Med., 1915-16, 1, 457. ALLERGY TO POLLENS (HAY-FEVER) 
663 
with sneezing, by the slightest thing whatsoever, others by merely smelling 
a rose.” These authors have stated that there is evidence that hay-fever 
has existed for centuries before its recognition as a specific disease, and all 
attempts to estimate its antiquity are futile. 
In 1819 Bostock1 contributed a very complete clinical description of 
hay-fever and expressed the belief that it was caused by the heat of summer. 
Excellent statistical studies were made by Elliotson2 and Phoebus,3 and of 
special interest in relation to the hereditary tendency for acquiring the dis- 
ease was the opinion expressed by Perrie4 in 1867, that it was of nervous origin. 
The most notable contributions were made by Blackley5 in 1873. Him- 
self a sufferer with hay-fever, he conducted a remarkable series of experi- 
ments constituting one of the most complete researches in the history of 
experimental medicine. Blackley tested on himself the pollens of the 
grasses and plants belonging to thirty-five orders, and was first to employ 
skin and mucous membrane tests for sensitization. He also showed that 
pollen may be carried for long distances in the air. 
The next notable contributions on the etiology of hay-fever were made 
by Dunbar,6 who confirmed the observations of Blackley and expressed the 
belief that the condition was caused by a pollen toxin. Since then a very 
large literature has accumulated upon treatment with Dunbar’s supposedly 
antitoxin serum designated as “Pollantin,” and more especially with extracts 
of pollens, as first employed by Curtis7 and Noon.8 
The Nature of Hay-fever.—Hay-fever is now commonly regarded as a 
condition of hypersensitiveness or allergy to pollens. Some of the earlier 
investigators considered the disease as a bacterial infection and endeavored 
to explain the relation of pollens to the production of attacks by assuming 
that they acted as carriers of bacteria. As a matter of fact, it is now well 
known that perennial hay-fever-like attacks or vasomotor rhinitis may be 
due solely to bacterial sensitization and that some cases of hay-fever are 
apparently caused by sensitizations to both pollens and bacteria. 
The older theory of Dunbar that the disease was caused by a pollen toxin 
analogous to bacterial toxins has been thoroughly disproved. The theory 
of the allergic nature of pollen disease first proposed by Wolff-Eisner9 is 
now generally accepted, although his explanation of the production of the 
symptoms and lesions by an “anaphylatoxin” from the digestion of the 
pollen protein by specific antibody and complement, is not acceptable. 
There can be no doubt that the pollens contain protein substances accord- 
ing to the analyses reported by Kammann10 and Koessler11; Heyl12 has recently 
obtained an albumin, a proteose, and a glutelin from the pollen of ragweed. 
However, it appears that the sensitinogens in pollens are but feebly anti- 
genic. Ulrich13 and Koessler14 have reported successful sensitization of guinea- 
1 Medico-Chi. Trans., 1819, 10, 161; ibid., 1828, 14, 437. 
2 Lancet, 1830-31, 2, 370. 
3 Gressen, Ricker, 1862, 136. 
4 Med. Times and Gaz., 1867, 2, 2, 30. 
5 See Blackley: Hay-fever, Its Causes, Treatment and Effective Prevention, 2d ed., 
London, 1880. 
6 Berl. klin. Wchn., 1905, Nos. 26, 28, 30. 
7 New York Med. News, 1900, 37, 16. 
8 Lancet, 1911, 1, 1572. 
9 Wolff-Eisner: Das Heufieber, Miinchen., 1906. 
10 Biochem. Ztschr., 1912, 46, 151. 
11 Jour. Amer. Chem. Soc., 1917, 39, 1470; 1919, 41, 670. 
12 Jour. Amer. Chem. Soc., 1919, 41, 670. 
13 Jour. Immunology, 1918, 3, 455. 
14 Forchheimer’s Therapeusis of Internal Diseases, 1914, 5, 686. 664 
CLINICAL ALLERGY 
pigs with pollens, but Cooke, Flood and Coca,1 and Smith2 were unable to 
sensitize guinea-pigs with them. Koessler also claims to have successfully 
sensitized guinea-pigs in a passive manner with the sera of sensitized human 
beings. Clowes3 observed that subcutaneous injection into a hay-fever 
subject of a pollen protein to which the individual is not naturally hyper- 
sensitive does not induce clinical hypersensitiveness to that protein. 
Furthermore, only a small percentage of human beings acquire hay-fever 
(10 per cent, in the United States, according to Cooke and Vander Veer). 
This indicates that some predisposing factor must be present rendering the 
body cells of some individuals more susceptible of sensitization with pollens 
than is the case of most persons and the lower animals. 
This predisposition to sensitization is apparently inherited. Some of the 
earliest writers on hay-fever observed family predisposition, and finally Cooke 
and Vander Veer were recently able to show quite conclusively the hereditary 
transmission of the tendency to become hypersensitive to various protein 
substances including the pollens. Actual sensitization is not inherited, but 
the predisposition to become sensitized is, the subject having been dis- 
cussed at more length in a preceding chapter. 
Pollens Causing Hay-fever.—These are broadly divisible into two main 
groups: (a) Those causing early or spring fever, running from the middle of 
May to the first or middle of July, and (b) those causing late or autumnal 
fever, the attacks usually beginning in the middle of August and running 
to the first frost. Cooke and Vander Veer have given the following list as 
commonly producing the spring or early hay-fever: 
Common Name. 
Timothy 
Red-top 
June grass 
Sweet vernal 
Low spear 
Orchard grass 
Rye 
Wheat 
Quickgrass 
Locust 
Chestnut 
Maple 
Daisy 
Rose 
Honeysuckle 
Privet 
Botanical Name. 
Phleum pratense 
Agrostis alba 
Poa pratensis 
Anthoxantheum odoratum 
Poa annua 
Dactylis glomerata 
Secale cereale 
Triticum sativum 
Agropyrum repens 
Robenia pseudo-acacia 
Castanea dentata 
Acerrubrum 
Chrysanthemum leucanthemum 
Rosa 
Lonisera caprifolium 
Ligustrum vulgare 
The following is a list of those pollens usually responsible for the late or 
autumnal cases: 
Common Name. 
Ragweed 
Ragweed 
Goldenrod 
Goldenrod 
Chrysanthemum 
Dahlia 
Zenia 
Clematis 
Marsh grass 
Aster 
Botanical Name. 
Ambrosia trifida 
Ambrosia artenisiaefolia 
Solidago canadensis 
Solidago nemoralis 
Chrysanthemum 
Dahlia 
Zenia 
Clematis Virginiana 
Spartina stricta 
Aster 
1 Jour. Immunology, 1917, 2, 217. 
2 Jour. Immunology, 1918, 3, 325 (disc.). 
3 Jour. Immunology, 1918, 3, 325 (disc.). ALLERGY TO POLLENS (HAY-FEVER) 
665 
Cooke and Vander Veer found that with few exceptions individuals sen- 
sitive to one of the grasses reacted to all, which indicates a biologic relation- 
ship of the proteins derived from the pollens of the graminaceae. Further- 
more, desensitization effected with the pollen of one grass usually reduces 
hypersensitiveness to all, as will be discussed in more detail in the suc- 
ceeding chapter. 
Likewise individuals who are hypersensitive to ragweed react to the pol- 
lens of the two varieties, and the same is true of individuals hypersensitive 
to goldenrod reacting to both varieties of this plant, indicating that these 
pollens are biologically identical while botannically different. But indi- 
viduals who are sensitive to ragweed may not be sensitive to goldenrod, and 
the reverse. 
Symptoms of Hay-fever.—The description of a case by Bostock in 1819 
adequately summarizes the usual symptoms and lesions of hay-fever and 
especially of the late or autumnal type which is apt to be more severe than 
the spring variety and frequently accompanied with asthma. “A sensation 
of heat and fulness is experienced in the eyes, first along the edges of the lids, 
and especially in the inner angles, but after some time over the whole of the 
ball. At the commencement the external appearance of the eye is little 
affected, except that there is a slight degree of redness and a discharge of 
tears. This state gradually increases, until the sensation becomes converted 
into what may be characterized as a combination of the most acute itching 
and smarting, accompanied with a feeling of small points striking upon or 
darting into the ball, at the same time that the eyes become extremely 
inflamed and discharge very copiously a thick mucous fluid. 
“After this state of the eyes has subsisted for a week or ten days, a 
general fulness is experienced in the head, and particularly about the fore- 
part; to this succeeds irritation of the nose, producing sneezing, which occurs 
in fits of extreme violence coming on at uncertain intervals. To the sneezings 
are added a further sensation of tightness of the chest, and a difficulty of 
breathing, with a general sensation of the fauces and trachea. There is no 
absolute pain in any part of the chest, but a feeling of want of room to receive 
the air necessary for respiration, a huskiness of the voice, and an incapacity 
of speaking aloud for any time without inconvenience. To these local symp- 
toms are at length added a degree of general indisposition, a great degree of 
languor, an incapacity for muscular exertion, loss of appetite, emaciation, 
restless nights, often attended with profuse perspirations, the extremities, 
however, being generally cold. The pulse is permanently quickened, from 
80, the average standard, to about 100, and upon any considerable exertion 
it arises to 120 or more.” 
Diagnosis and Specific Treatment of Hay-fever.—The history and symp- 
toms are usually so characteristic that diagnosis is a simple matter. Some- 
times the patient knows from experience the plant or plants to which hyper- 
sensitiveness exists, and especially in the late or autumnal type in which the 
pollens of ragweed or goldenrod, or both, are usually the causes. 
Skin tests, however, are readily applied and have proved quite satisfactory 
for determining the pollens to which sensitization exists. These are de- 
scribed in a succeeding chapter on page 674. 
Specific treatment consists in desensitization by means of injections of 
pollen extracts. This method of treatment is described in Chapter XXXII. 666 
CLINICAL ALLERGY 
It is now well known that some individuals suffer with asthma, vaso- 
motor rhinitis, and other hay-fever-like symptoms when in contact with cer- 
tain of the lower animals, notably the horse. 
Meltzer1 suggested in 1910 that this type of asthma may be an ana- 
phylactic phenomenon, and Walker, Rackeman, and other investigators have 
since shown this to be the case. 
Allergies of this kind may be acquired to the hair, dandruff, and expired 
air of the horse, rabbit, guinea-pig, and other of the lower animals; likewise 
to goose feathers. In the latter instance sleeping upon goose feather pillows 
may elicit attacks of vasomotor rhinitis and asthma. 
The degree of sensitization may be so extreme that allergic attacks are 
elicited by proximity to the animal sensitinogen, without actual contact. 
Not infrequently the exciting agent is carried in dust and constitutes one 
form of allergic dust asthma. 
The tendency to acquire these sensitizations is sometimes apparently 
inherited. Injections of horse-serum rarely result in the production of such 
extreme degrees of hypersensitiveness. 
The diagnosis of this type of allergy is sometimes quite easy, as the 
patient frequently knows the animal or animals to be avoided. Skin tests 
have proved of considerable diagnostic value and are described in the suc- 
ceeding chapter. Individuals may be hypersensitive to the hair of an 
animal and not to its serum; as a general rule, however, hypersensitiveness 
to the horse includes sensitization to hair, dandruff, and serum, and for this 
reason the administration of horse-serum in the prophylaxis and treatment 
of disease may be very dangerous for such persons. 
To be included in the category of allergies to animal substances are the 
excessive effects experienced by some individuals to the stings of insects. 
Longcope2 has reported one individual in whom the sting of the mosquito 
produces enormous areas of edema similar to the so-called angioneurotic 
form, and in another case a sting of a wasp was followed by symptoms ex- 
actly like the immediate reaction in serum disease. 
ALLERGY TO ANIMAL SUBSTANCES 
ALLERGY TO PLANT SUBSTANCES (DERMATITIS VENENATA) 
The extreme hypersensitiveness of some individuals to poison ivy (Rhus 
toxicodendron), poison sumac or dogwood (Rhus venenata), poison oak 
(Rhus diversiloba), and the primrose (Primula obconica) strongly suggests 
that allergies to these plants may develop. Recently Simmons and Bolin3 
have described a similar condition among some individuals to the mango 
(Mangifera indica). 
The exact nature of the exciting agents are unknown; toxicodendric acid 
and various glucocidal substances are commonly mentioned in this connec- 
tion. Simpson4 has found that protein from primrose did not elicit a reac- 
tion, whereas the active principle was found soluble in absolute ethyl 
alcohol. Simmons and Bolin likewise found that the proteins of the mango 
did not engender skin reactions, but that a non-protein skin irritant was 
obtained from the stem sap of either ripe or green mangoes. 
Doubtless the juices of these plants contain irritants capable of producing 
dermatitis by direct irritation and without allergic sensitization. It is com- 
mon experience, however, that only certain individuals are susceptible and 
that some of these are so extremely hypersensitive, that attacks of derma- 
1 Jour. Amer. Med. Assoc., 1910, 55, 1021. 
2 Amer. Jour. Med. Sci., 1916, 152, 637. 
3 Amer. Jour. Trop. Med., 1921, 1, 351. 
4 Jour. Amer. Med. Assoc., 1917, 69, 95. ALLERGY TO BACTERIA 
667 
titis are produced by mere proximity to the plant without actual contact. 
In these respects dermatitis venenata bears a close analogy to pollen hay- 
fever and may be an allergic phenomenon, even though the exact nature of 
the sensitinogens and the mechanism of sensitization and reaction are un- 
known. Doubtless a tendency to acquire sensitization to these plants may 
be inherited, as has been found true of other examples of human sensitiza- 
tions, and especially so, since the skins of many susceptible individuals ap- 
pear to be sensitive to other substances. Further evidence of the allergic 
nature of dermatitis to poison ivy is afforded by the observation of Scham- 
berg that the internal administration of the tincture or fluidextract may 
bring about rapid but temporary amelioration of the symptoms, suggestive 
of a process of desensitization. As shown by Strickler1 and Anderson and 
Pruett2 similar results are obtained with the intramuscular injection of the 
extract. 
McNair5 explains the tolerance of most individuals for lobinol (the active 
principle of Rhus diversiloba) on the basis of non-specific factors, and espe- 
cially the particular structure of the sebaceous and sudoriparous glands and 
the chemical nature and abundance of their secretions. This tolerance, 
however, could be overcome by the application to the skin of strong extracts, 
showing that it was relative and not absolute; similar results were observed 
by Simmons and Bolin with the sap of the mango. These investigators do 
not subscribe to the view that dermatitis venenata may be an allergic phe- 
nomenon, but that tolerance is dependent upon non-specific factors relating 
to the skin and that susceptibility is due to a modification or absence of these 
factors. 
Spain4 has recently studied a large group of cases of dermatitis venenata 
and found that the typical vesicular lesion can be produced by applying 
alcohol and chloroform extracts to the skin, but that these reactions could 
not be produced by intracutaneous injections. Infants were found to be 
insusceptible. 
ALLERGY TO BACTERIA 
Our knowledge of clinical allergy to bacteria is practically confined to a 
type of asthma occurring with chronic bacterial infections of the upper res- 
piratory tract. These have been the subject of special studies by Walker 
and Rackemann with the aid of skin tests conducted with the proteins of 
various bacteria recovered in cultures from the discharges. 
Chronic bacterial infections of the accessory sinuses of the upper respira- 
tory tract may be likewise associated with asthma, but this is the only clinical 
condition clearly identified with bacterial anaphylaxis at the present time. 
Bacterial allergic asthma is not encountered in children, and is apparently 
acquired in the later years of life and frequently as a result of previous 
chronic bronchitis. 
It is possible that perennial rhinitis, commonly designated as a vasomotor 
rhinitis which persists throughout the year, may be an expression of bac- 
terial allergy; likewise some cases of eczema. Skin tests with different bac- 
terial proteins not infrequently yield positive reactions without clinical 
evidences of the sensitization. 
1 Jour. Amer. Med. Assoc., 1921, 77, 910. 
2 Californ. State Jour. Med., 1921, 19, 188. 
3 Jour. Infect. Dis., 1916, 19, 419. 
4 Jour. Immunology, 1922, 7, 179. 668 
CLINICAL ALLERGY 
The intolerance or idiosyncrasy of certain individuals for certain foods 
has been known clinically for many years, as attested to by the aphorism: 
“One man’s meat is another man’s poison.” That is, the immediate symp- 
toms of vomiting, diarrhea, and abdominal pain have been long associated 
with food idiosyncrasy for the very evident reason that the average adult 
patient was able to associate cause with effect. 
It has been only within the last ten or twelve years, however, that our 
knowledge of this subject has been greatly extended and principally by the 
pediatricians and dermatologists, to include one kind of asthma and other 
untoward effects from foods in children including the development of vari- 
ous skin lesions, notably eczema and urticaria. These studies were given 
great impetus by the application of skin tests by Smith1 in the diagnosis of 
allergy to buckwheat, and Schloss,2 Goodale,3 Talbot,4 and others in the 
diagnosis of allergies to raw eggs, milk, and other foods. 
Sensitization in Food Allergy.—Among the allergies of children, that 
for food substances is probably the most common. Possibly active sensi- 
tization may be inherited inasmuch as Talbot and others have described 
instances of allergy to cow’s milk in very young infants, the administration 
of small amounts being followed by immediate vomiting, but this is im- 
probable. Since inherited sensitization is of the passive type and of very 
short duration in experimental animals, it is more likely that a predisposi- 
tion to sensitization is inherited and that actual sensitization may result 
from the passage of foods, eaten by the mother, in her milk. This possibility 
is shown by the observations and experiments of O’Keefe5 and Shannon.6 
The former found that some nursing infants with eczema gave positive skin 
reactions to proteins they had never eaten, and the latter observed the 
withdrawal of certain foods from the diet of a mother to which she was hy- 
persensitive, to be followed by the disappearance of an urticarial eruption 
on her nursing infant. 
Several investigators have shown that young guinea-pigs may be sen- 
sitized by feeding. Reference to the experiments of Wells was made in the 
preceding chapter and Kassowitz7 and others have also reported successful 
sensitizations by feeding. Numerous investigators, including Moro,8 Lust,9 
Hahn,10 and others, claim to have found precipitins in the sera ascribed to 
the permeability of the intestinal mucous membrane and the antigenic 
effects of absorbed foreign proteins. 
It would appear, therefore, that sensitization to foods is acquired, al- 
thought a predisposition to sensitization may be inherited. If active sen- 
sitization is not inheritable, then it must be assumed that infants become 
sensitized by food substances in the mother’s milk in order to explain specific 
sensitization to foods which a child has never eaten. Doubtless inflam- 
matory changes in the gastro-intestinal mucosa aid in sensitization by 
facilitating the absorption of food sensitinogens. Since proteins digested 
to the peptone and amino-acid stages are doubtfully allergenic, it is prob- 
ALLERGY TO FOODS 
1 Arch. Int. Med., 1909, 3, 358. 
2 Amer. Jour. Dis. Child., 1912, 3, 341. 
3 Boston Med. and Surg. Jour., 1916, 175, 181. 
4 Boston Med. and Surg. Jour., 1914, 171, 708. 
5 Boston Med. and Surg. Jour., 1920, 182, 569. 
6 Amer. Jour. Dis. Child., 1921, 22, 223. 
7 Ztschr. f. Kinderh., 1912-13, 5, 75. 
8 Munch, med. Wchn., 1906, 49. 
9 Jahrb. f. Kinderh., 1913, lxxvii, 383. 
10 Jahrb. f. Kinderh., 1913, lxxvii, 405. ALLERGY TO DRUGS 
669 
able that sensitization can result only from the absorption of undigested or 
partially digested foods. 
Owing to the chemical relationships among certain foods, it is highly 
probable that sensitization with one may result in sensitization to several. 
For example, in sensitization to wheat, Goodale has shown that reactions may 
be produced by oat, barley, and other members of the grass family. 
Foods Commonly Producing Allergy.—These include foods of both ani- 
mal and vegetable origin including some of the fruits, notably the strawberry 
and gooseberry. Practically every common food has been implicated, and 
especially buckwheat, pork, oysters, clams, lobsters, cheese, etc. In chil- 
dren the foods mostly involved are cow’s milk, eggs, oatmeal, potato, peas, 
rice, beef juice, and chicken. Raw eggs and uncooked milk have been found 
especially important in connection with allergic asthma and “exudative 
diathesis” of children. 
Symptoms of Food Allergy.—These may arise from the alimentary, res- 
piratory, and cutaneous systems. 
In cases of extreme allergy the alimentary lesions and symptoms are 
quite marked, and may include swelling of the lips and tongue, nausea, eruc- 
tations, vomiting, abdominal pain, and diarrhea. In buckwheat allergy I 
have seen extensive swelling of the tongue, lips, and face occur within a few 
minutes after the ingestion of a minute amount of buckwheat flour. Infants 
subject to milk allergy are likely to vomit almost immediately after milk is 
swallowed. 
Asthma may be a symptom and especially in children. Talbot has de- 
scribed cases of croup in children due to allergy to eggs, as likewise re- 
current bronchitis. Asthma and vasomotor rhinitis may also occur among 
adults. 
Mention has already been made of edema of the lips and tongue, which 
may involve the pharynx and larynx in cases of extreme hypersensitiveness. 
Urticarial rashes resembling serum sickness are of frequent occurrence, as 
likewise angioneurotic edemas. These lesions may develop within a few 
minutes to a few hours after ingestion of food. Eczema, erythema multi- 
forme, and other cutaneous lesions may develop as more chronic manifesta- 
tions of food allergy. 
Diagnosis and Treatment.—The diagnosis of acute allergic reactions by 
foods in adults is usually an easy matter and especially if the symptoms are 
gastro-intestinal and accompanied by the development of urticaria or angio- 
neurotic edema. The diagnosis of allergic asthma, eczema, and other 
chronic lesions may require the aid of skin tests, which are described in the 
succeeding chapter. 
Treatment has usually consisted in the total or partial withdrawal of the 
allergic food or foods from the diet. Desensitization has also been employed 
and will be described in a succeeding chapter. 
Drug allergy is a condition of hypersensitiveness in which a quantity 
of the substance non-toxic for most individuals elicits an unusual but 
characteristic reaction. 
A large number of medicinal substances are known to be active in these 
allergies, a list being given on page 616. Drug allergies are highly specific; 
this phase, together with the nature and mechanism of the reaction, are 
discussed in Chapter XXVIII. 
Not all skin eruptions following the administration of drugs can be 
ALLERGY TO DRUGS 670 
CLINICAL ALLERGY 
designated as allergic; they may be caused by excessive cutaneous elimina- 
tion and other factors. Allergy is characterized by the development of symp- 
toms with small and non-toxic amounts borne by most individuals without 
untoward effects. 
Incubation Period.—In most cases of drug allergy the symptoms appear 
within a few minutes to a few hours, depending upon the route of adminis- 
tration. In such cases the time of sensitization is usually unknown. In 
many individuals the symptoms of allergy develop after several administra- 
tions covering a period of ten to twenty days, as has been especially observed 
with arsphenamin. 
Drug allergy is sometimes present in very young children and even at 
birth, but recorded instances of inherited allergy are very few. It is likely, 
however, that a tendency to acquire drug hypersensitiveness is inheritable, 
as indicated by the studies of Cooke and Vander Veer in “human sensiti- 
zations.” 
Symptoms.—The most marked and characteristic symptom is the skin 
eruption which may be urticarial as in serum allergy, erythematous, papular, 
vesicular, hemorrhagic, or a combination of these. Eruptions are especially 
developed in the allergies to quinin, arsenic, belladonna, copaiba balsam, 
antipyrin, mercury, salicylicates, and iodids. The same individual may 
react to the same drug at different times in a different manner. Itching 
frequently accompanies the rash or may exist without the eruption. 
A peculiar form of eruption sometimes found in the allergies to antipyrin, 
phenacetin, and salipyrin is the “fixed pigmented erythema.” 
Fever, sometimes preceded by a chill, is one of the most frequent symp- 
toms of drug allergy and may reach as high as 104° to 106° F. It may be 
present without the skin eruption or the eruption may be present without 
fever, but as a general rule they are both present. 
Edema and especially of the face is frequently found in the allergic re- 
actions to arsphenamin, chloral, antipyrin, opium, and digitalis. Arthritic 
and muscular pains and adenitis may develop as in serum disease, as like- 
wise leukopenia and lowering of blood-pressure. Not infrequently the al- 
lergic reaction to a drug is unusual and characteristic for a given indi- 
vidual; for example, I know of an individual whose reaction to a minute 
dose of quinin consisted of a severe balanitis. 
Skin reactions are frequently elicited in drug allergies by application of 
the drug to abrasions, and tests of this kind are employed for diagnostic 
purposes. They are described in the succeeding chapter. CHAPTER XXXI 
CLINICAL ALLERGY (Continued) 
The Reactivity of the Skin; Specific and Non-specific Reactions; the 
Influence of Age, Disease, and Non-specific Factors.—It is highly probable 
that the skin plays a more important role in the mechanism of recovery of 
disease of the internal organs than is commonly surmised. Certainly the 
skin and mucous membranes are among the most important factors in 
natural immunity to microbic infection. As discussed in Chapter IX, the 
skin may possess an important function in the destruction of microbes and 
toxic products during elimination, and it is not surprising that the skin and 
mucous membranes have been found to become sensitized during the course 
of many diseases, some of which are not cutaneous diseases or characterized 
by any special cutaneous manifestations. The skin appears to become 
sensitized or allergic by reason of the intimate part it appears to play in the 
general processes of infection and immunity. 
Allergic skin tests are conducted by applying the agent to abrasions or 
injecting it very superficially (intradermically). If the cells are sensitized, a 
local allergic reaction occurs. In addition to this specific reaction, however, 
a non-specific reaction may occur. Practically any substance injected 
intracutaneously will elicit a non-specific reaction if the amount injected is 
sufficiently large. 
This non-specific reaction may be due in part to trauma and in part to 
the production of toxic split products by the ferments of the skin, recently 
studied by Sexsmith and Petersen.1 The possibility of its occurrence is so 
important that in conducting skin tests a control fluid should always be 
included, if possible, in order to guard against error. Otherwise the amount 
of allergen applied or injected should be such as is known on the basis of 
numerous trials with normal individuals, to be less than the amount giving 
non-specific reactions. 
The reactivity of the skin varies with age. Roily,2 working with a variety 
of bacterial toxins, found that infants did not react, but that reactions oc- 
curred with advancing age. Tezner,3 using Witte’s peptone, colon bacilli, 
and tuberculin, observed that at about the time that the skin of children 
became increasingly sensitive to tuberculin, a corresponding sensitiveness 
was manifest toward the other substances, a more or less general “sen- 
sitization” or “Umstimmung.” Sexsmith and Petersen believe that this 
phenomenon is related to the alteration in the ferments of the skin, the skin 
of the young containing peptidase in a considerable amount, while the 
adult skin seems to be without this ferment activity. 
The reactivity of the skin may be decreased in cachexia, advanced tuber- 
culosis, pregnancy, and other diseases. From the standpoint of allergy 
this depression is probably due to desensitization by circulating antigens; 
from the standpoint of the non-specific or enzyme reaction it may be due to 
an increase in the titer of the serum antiferment. Both processes may be 
ALLERGIC SKIN REACTIONS 
1 Jour. Exper. Med., 1918, 27, 273. 
2 Miinch. med. Wchn., 1911, lxiii, 1285. 
3 Monatschr. f. Kinderh., 1911, 10, 131. 
671 672 
CLINICAL ALLERGY 
suppressed by a decrease of permeability of the capillary endothelium, as 
studied especially by Luithlen, von den Velden, and others. 
An increased non-specific reactivity of the skin has been found by Hoke1 
in leukemia and thyroid feeding; Sherrick has found that the administration 
of potassium iodid has the same effect by accelerating the rate of digestion 
with the production of split products capable of exciting acute inflammation 
and commonly resulting in the production of a pustule. 
It is apparent, therefore, thait the subject of allergic skin reactions is 
quite complicated and full of possibilities for error. The occurrence of 
non-specific reactions, differences in the reactivity of the skin according to 
age, the influence of disease processes in increasing or decreasing reactivity, 
and the important influence of non-specific agents upon reactivity, are all 
to be kept in mind when conducting and interpreting skin reactions. 
Mechanism of Specific Reactions.—The application of skin and mucous 
membrane tests for allergy to various substances was given great impetus 
by von Pirquet’s studies with the cutaneous tuberculin test. Since then 
tests of this kind have been developed for the diagnosis of allergies to other 
microparasites, serum, pollens, foods, drugs, and other substances, and have 
generally proved reliable and of considerable diagnostic aid. 
The use of these tests is based upon the assumption that in allergy the 
cells of the skin and mucous membranes share in the sensitization. In 
tuberculosis this occurs to an exquisite degree; likewise in the majority of 
cases of hay-fever and other allergies. 
When the allergen in suitable form is applied to a mucous membrane, 
an abrasion of the skin, or injected intradermally, a local reaction of allergic 
shock is produced. With tuberculin an interval of twelve hours or more 
usually elapses before a reaction appears, but with pollens and food allergens 
a reaction usually develops within fifteen minutes. Why there is this dif- 
ference in the time factor is unknown. 
The reaction apparently occurs in or upon the sensitized cells. Cook 
and Smith2 have shown in a conclusive manner by perfusion experiments, 
employing pieces of excised skin by the Schultz-Dale method, that the anti- 
body occurs in or upon the cells as is apparently true of allergy in general. 
It is probable that the mechanism of the local reaction is identical with the 
general reaction and in the nature of an intracellular colloidal phenomenon. 
This reaction is expressed clinically by the development of vasomotor 
changes with hyperemia and edema. The irregularly outlined urticarial- 
like lesion is the most typical; hyperemia alone is less significant. When the 
allergen is placed upon the conjunctival or nasal mucous membranes similar 
changes occur, that is, acute hyperemia, serous exudation, and profuse watery 
discharge. 
Non-specific Skin Reactions.—It is well known that the skins of some 
individuals are more sensitive than others to trauma of cutaneous tests. 
This is especially true of persons subject to hay-fever and other allergies. 
In syphilis the “Unstimmung” of the skin is generally known and may occur 
in other diseases. An abrasion and application of some ordinarily harmless 
material, as a drop of decinormal sodium hydroxid solution, may elicit a 
well-defined wheal. This tendency introduces a source of error and must 
be checked in every case by a control scarification or injection. 
Intradermal injections are especially likely to produce non-specific reac- 
tions. These may be due to (a) trauma and the direct injection of an irri- 
tant used as a preservative for the material, as phenol, or, to a preformed 
1 Wien. klin. Wchn., 1920, 33, 41. 
2 Jour. Immunology, 1916, 2, 415. ALLERGIC SKIN REACTIONS 
673 
toxic and irritant substance, as a bacterial toxin; (b) to the production of an 
irritant by the action of non-specific ferments in the serum or derived from 
injured cells, upon the protein of the patient’s serum, devitalized cells, 
injected protein, or all three. 
According to Petersen the enzymes of the skin vary somewhat according 
to age. The young skin contains more ereptase (erepsin or peptidase) 
and little lytic protease; the adult skin, on the other hand, little ereptase 
and more protease. Neither type of skin contains much antienzyme. 
Allergic Skin Reactions as Indices of Infection and Hypersusceptibility.— 
As is well known, allergic skin reactions may be elicited in various bacterial 
and protozoan diseases with allergens prepared from the protein of the 
respective microparasites, particularly in tuberculosis, glanders, typhoid 
fever, and syphilis. Well-marked and specific reactions have also been found 
by Amberg1 and Kolmer and Strickler in ringworm and favus, and isolated 
reports show that they may be elicited in various other diseases with the 
proper preparations. 
In these conditions, however, the skin reactions are not always elicited, 
and especially during the early and acute stages. An interval of time is 
required for the purpose of sensitization, and during the acute stages antigen 
and antibody may both be present in the cells and body fluids, as shown by 
Weil2 and Denzer,3 with continual interaction expressed in the symptom- 
complex of the infection, and thereby giving no response when the protein 
is applied or injected into the skin. Furthermore, the chemical nature of 
the protein of our allergen may have been altered in the course of prepara- 
tion to a sufficient extent to fail to elicit an allergic reaction. These and 
other factors not understood diminish the practical value of a skin test. At 
present, however, it may be stated that they possess a diagnostic value which 
is particularly high in chronic infections, and that no other satisfactory 
clinical or laboratory test for the state of allergy to a particular protein and 
for the allergic antibody has been discovered except that by which the 
protein is actually brought into relation with the body cells, as in the paren- 
teral introduction of the protein. Of great interest in this connection is the 
relation of the intensity of the allergic skin reaction to the extent of the 
infection. Krause4 has recently studied in a thorough and excellent manner 
the tuberculin skin reaction in relation to experimental tuberculosis, finding 
that cutaneous hypersensitiveness to tubercle protein is inaugurated by the 
establishment of infection and the development of the initial focus; that the 
skin reaction increases with progressive disease; is diminished with healing 
and increased by reinfection. 
The clinical significance and practical value of skin reactions are largely 
of a diagnostic nature for the detection of hypersensitiveness to a protein or 
proteins, which may, when introduced into the organism, produce various 
acute or chronic lesions and symptoms of disease. 
Cutaneous Versus Intracutaneous Tests.—Mucous membrane tests are 
not generally employed except the ophthalmic tuberculin and mallein tests 
by veterinarians among the lower animals. 
Cutaneous tests for allergy were first employed by Blackley in 1873 in 
hay-fever by the application of pollens to abrasions of the skin. In 1909 
Cole conducted a skin test for allergy to buckwheat, but the extensive pioneer 
work of Schloss since 1912 finally established the value of the cutaneous or 
“scratch” test for allergic sensitization. 
1 Jour. Exper. Med., 1910, xii, 435. 2 Proc. Soc. Exper. Med. and Biol., 1914, xii, 37. 
3 Jour. Infect. Dis., 1916, xviii, 631. 
4 Jour. Med. Research, 1911, xxiv, 361; ibid., 1916, 35, 1, 25. 674 
CLINICAL ALLERGY 
The intracutaneous test has been developed largely by investigations of 
Cooke in 1911 and employed and advocated by him up to the present time. 
The cutaneous test, consisting in the application of the allergen to a 
superficial cut or abrasion of the skin, is generally employed, and especially 
in tests for allergies to pollens and foods. Walker and Adkinson1 have com- 
pared the cutaneous and intracutaneous methods with different proteins in 
allergic asthmas and found the former specific, sufficiently sensitive, easier 
to do, and more agreeable to the patient. The intradermic test was found 
too sensitive and because of the tendency for non-specific reactions, does 
not always separate closely related proteins; furthermore, it is more diffi- 
cult to do, causes the patient considerable annoyance and discomfort, and 
is not as practical when many substances must be tested. In other words, 
the cutaneous tests while easier, less painful, more prompt in results, and 
less likely to error by yielding pseudopositive reactions, are less sensitive 
than the intradermal tests. Therefore, they may fail to indicate an allergy, 
and when cutaneous tests are negative Blackfan2 has recommended that 
intracutaneous tests are advisable. Due care must be exercised in reading 
and interpreting the latter to guard against the traumatic and non-specific 
reactions being regarded as specific allergic reactions. 
Brown3 has likewise found the intracutaneous test superior to the cuta- 
neous. In series of 78 patients clinically sensitive to proteins he found posi- 
tive reactions in all with the intradermal test, 82 per cent, positive with a 
cutaneous test employing fluid allergens, and only about 50 per cent, posi- 
tive with the cutaneous test conducted with the dried allergens commonly 
employed. Brown injects only about 0.01 c.c. which reduces the trauma 
and tendency to non-specific reactions. 
Intracutaneous tests are more likely than cutaneous tests to elicit gen- 
eral reactions by reason of the greater chances for absorption. In cases of 
extreme sensitization, as allergic asthmas to horse protein, egg-white, and 
pollens, the cutaneous tests are certainly to be preferred, and if intracuta- 
neous tests are employed, the amounts injected must be very small,*as 
0.1 c.c. of 1 : 100 or 1 : 1000 dilutions. 
Cooke4 has studied with particular care these constitutional effects, 
sometimes following the application of intracutaneous tests and subcutane- 
ous injections for desensitization. He records one death from allergic 
asphyxia in a child three years of age attributed to skin tests in extreme 
allergy to glue. Cooke has advised the use of diluted extracts for these 
tests in order to avoid constitutional effects, warns against frequently re- 
peated tests, and states that not more than six to eight intracutaneous tests 
should be conducted at one time. 
When general symptoms develop after intracutaneous injections, as 
extensive wheals, erythema, short paroxysmal coughs, or dyspnea, Cooke 
advises the application of a tourniquet and the injection of adrenalin 1 : 1000 
in dose of 1 c.c. for adults and 0.4 to 0.6 c.c. for children. Also the intra- 
venous injection of 0.001 gm. strophanthin for adults if there is cardiac 
dilatation. 
In veterinary practice cutaneous tests are not employed, preference 
being expressed for the intradermal tests employing the skin at the root of 
the tail, the nose, or mucosa of the eye. 
Technic of the Cutaneous Test.—The skin of the arm or forearm is 
cleansed with alcohol and dried. Abrasions are made for each substance to 
be tested, with one or two extra for controls. Walker makes a series of linear 
1 Jour. Med. Research, 1917-18, 37, 287. 
2 Amer. Jour. Med. Sci., 1920, 160, 341. 
* Jour. Immunology, 1922, 7, 97. 
4 Jour. Immunology, 1922, 7, 119. ALLERGIC SKIN REACTIONS 
675 
cuts about | inch long (never longer than f inch) with a sharp scalpel on the 
flexor surface of the forearm. These cuts are only deep enough to penetrate 
the outer layers of the skin, care being taken not to draw blood. Abrasions 
may also be made with a needle as commonly practised in cowpox vaccina- 
tion; the von Pirquet skin bore is also very useful (Fig. 160). Care must 
be exercised against drawing blood; a slight oozing of serum is desirable. 
The instrument should be wiped with alcohol. 
The cuts or abrasions should be at least § inch apart in order that posi- 
tive reactions may be clearly defined; when only a few tests are to be ap- 
plied, they should be 1 inch or more apart (Fig. 154). 
When ten or more tests are to be conducted at one time, Walker’s method 
is very good, as two rows of incisions are easily accommodated on the forearm. 
On each cut or abrasion is placed a different protein. If the protein is 
in powder form a minute amount is placed on the cut, followed by the addi- 
tion of 1 drop of N/20 sodium hydroxid to dissolve the protein and hasten 
Fig. 154.—Method oe Making Scarifications for Cutaneous Allergic Tests with 
Daland Lancet. 
its absorption. It is well to rub the protein in for a few seconds with an 
applicator needle or scarifier, but care should be exercised against causing 
bleeding. I use small round wooden applicators like toothpicks (Fig. 155). 
After making the abrasions, a toothpick is dipped into N/10 sodium hy- 
droxid solution and applied to a cut which transfers sufficient of this solu- 
tion; the powder is next applied and rubbed in for a few seconds with the 
same toothpick, which is then discarded and a fresh one used for each sub- 
stance to be tested. 
The control receives a drop of hydroxid solution, followed by the same 
amount of rubbing in exactly the same manner except that no protein is 
used. If hydroxid is employed in the tests, it is essential to use it in the 
control, as some skins are quite sensitive to this substance. 
One-half hour later the proteins are washed off and the results noted. 
If the protein causes a marked reaction which may be accompanied by con- 
siderable itching, it is usually removed before a half-hour in order to avoid 
discomfort and the absorption of too much of the protein. 676 
CLINICAL ALLERGY 
As a general rule pure proteins are employed, but with individuals who 
are extremely hypersensitive due care must be exercised against the pro- 
duction of general reactions by the application of too much protein. In 
practice this refers especially to allergic asthmas to horse protein and eggs. 
In these cases the first tests may be conducted with 1 : 10 or 1 : 100 dilu- 
tions of horse-serum and egg-white; if reactions do not result, undiluted 
allergens may be employed. 
The Cutaneous Reactions.—In reading the results the control is first 
inspected and the other reactions compared with it. Sometimes the con- 
trol shows the presence of a small wheal surrounded by erythema, and in this 
case reactions with proteins are interpreted as positive when larger and 
better defined. 
The reactions vary considerably with different proteins and with dif- 
ferent individuals. Walker has described four main types, as follows, with 
proteins of bacteria, pollens, hair, etc.: 
Fig. 155.—Method of Applying Allergens to Cutaneous Abrasions in Allergic Skin Tests 
(a) An elevated white urticarial wheal which measures about 1 cm. in 
diameter and is surrounded by a pinkish, red areola. 
(b) A smaller wheal, but a larger area of erythema which often measures 
1 or 2 cm. in diameter; this reaction looks not unlike a large mosquito bite 
which has been irritated by scratching. 
(c) A large area of erythema with little or no elevation about the cut. 
(d) A doubtful reaction consisting of a small area of erythema measuring 
cm. or less in diameter. If the control shows no reaction or much less than 
this degree of erythema, this reaction may be regarded as doubtful or wreakly 
positive and should be repeated at a later date. 
A fifth type of reaction is described by Walker and occasionally seen. 
At the end of a half-hour it is negative or nearly so, but next day the skin 
about the cut is hot, very red, and slightly elevated. The cut may contain 
pus which has always proved to be sterile. This delayed reaction may not 
be allergic and its significance is doubtful. Fig 156 Positive Allergic Reactions to Pollens ALLERGIC SKIN REACTIONS 
677 
Itching and a burning sensation may accompany these reactions. A 
negative reaction is one that shows practically no difference from the con- 
trol cut or abrasion. 
As a general rule the reactions begin to fade after a half-hour, but may 
persist for two or three hours. The cuts or abrasions heal rapidly and 
without infection, but it is advisable to keep them clean and free of gross 
contamination. 
The cutaneous tuberculin reactions are described later; they develop 
much more slowly and present a different appearance in that a wheal or ur- 
ticarial lesion is not usually produced. 
Fig. 157.—Method of Administering an Intradermal Injection 
The skin has been cleansed with alcohol and pinched up between the thumb and index-finger of 
the left hand; the needle (No. 26) has been entered into the epidermis and 0.1 c.c. of fluid injected. 
Note the anemic area, indicating that the injection has been intradermic. 
Technic of Intracutaneous Tests.—The skin of the outer side of the arm 
or flexor surface of the forearm is generally employed and should be cleansed 
with alcohol. 
The injections are made with a 1 c.c. perfectly working syringe graduated 
into 0.1 or 0.01 c.c. amounts and fitted with a fine needle. I prefer the 
Record or Luer syringe graduated in 0.1 c.c. and fitted with a No. 26 needle. 
The syringe and needle should be sterilized beforehand and must be thor- 
oughly cleansed for each protein. For this reason a series of tests requires 
considerable time. 
The injections are given intracutaneously and not subcutaneously. The 
skin is pinched up between the forefinger and thumb and the needle entered 
eye upward until the eye is completely in the epidermal layer of the skin 
(Fig. 157). The injection is accompanied by the production of a whitish 678 
CLINICAL ALLERGY 
elevated area (Fig. 158); if this is not seen, the injection is too deep and apt 
to be unsatisfactory. A slight stinging hot pain is experienced. 
Not more than 0.2 c.c. should be injected; it is better to inject less, as 
0.1 c.c. or even 0.05 c.c., as these reduce the degree of trauma and pain. 
The solution of protein should not be more than 1 or 2 per cent; in cases of 
extreme sensitiveness, as in allergic asthmas due to pollens, serum or foods, 
the dilution should be less, for example, as 0.1 per cent. (1 : 1000) and only 
0.05 c.c. injected in order to avoid general reactions. 
When the tests are conducted with bacteria isolated from the sputum 
in asthma or from some other focus of infection, I inject 0.1 c.c. of a sus- 
pension in saline solution of approximately 500,000,000 per cubic centimeter. 
Fig. 158.— Showing the Small Anemic Area Immediately After an Intracutaneous Injection. 
The injections should be at least 2 inches apart. By starting near the 
shoulder a series of six to eight injections can be readily given an adult on 
one side. 
A control injection is always advisable. Since the allergens are gen- 
erally dissolved in saline solution and preserved with 0.2 to 0.3 per cent, 
phenol or tricresol, I inject 0.1 c.c. of a control fluid of sterile saline solu- 
tion containing 0.25 per cent, of the same preservative as the solutions of 
allergens. 
Intracutaneous Reactions.—Within a few minutes after the injection of 
the protein solution an area of edema and erythema develops similar to the 
positive cutaneous reactions. The control and negatively reacting sites 
also show some erythema, but as a general rule the true positive reactions ALLERGIC SKIN REACTIONS 
679 
are readily differentiated from these in fifteen to thirty minutes by a larger 
area of erythema and a wheal-like lesion. 
Positive reactions usually persist for twenty-four to forty-eight hours 
as a papule with slight edema and erythema. As a rule the traumatic and 
non-specific protein reactions subside in this time and permit more accurate 
readings to be made. 
The Effect of Drugs Upon Skin Reactions.—It is now generally recognized 
that the oral administration of iodids and hromids may influence intracuta- 
neous skin reactions to the extent of increasing their degree and severity or 
producing well-marked papular or pustular reactions in the skins of persons 
who did not react to these injections in preliminary tests, and who are not allergic. 
Ambery and Knox1 found that the intravenous administration of neutral 
sodium ortho-iodoxybenzoate and sodium iodosobenzoate inhibited the 
development of reactions following the intracutaneous injection of serum 
in sensitized rabbits; sodium iodobenzoate and sodium benzoate in equal 
molecular concentrations had no inhibitory effect on the local reaction. The 
authors believed that the inhibitory influence of these drugs was attributable 
to their effect on oxidative processes. Opposite results have been reported 
by Sherrick,2 who found that the administration of potassium iodid either 
simultaneously or shortly before or after the intracutaneous injection of luetin 
and agar, resulted in the production of pustular or nodular reactions in healthy 
non-syphilitic persons. Kolmer, Matsunami, and Broadwell3 confirmed 
these observations among healthy non-syphilitic persons, persons sick with 
diseases other than syphilis, and normal rabbits and guinea-pigs. The oral 
administration of potassium iodid may cause the site of a former intra- 
cutaneous injection of luetin or agar to “light up” and present well-defined 
inflammatory changes sometimes leading to pustulation as pointed out by 
Sherrick; we also found that a purer form of luetin, prepared of washed spiro- 
chetes and free of all culture-media, was influenced in a much less degree by 
potassium iodid. 
In a further study of this subject by Kolmer, Immerman, Matsunami, 
and Montgomery,4 the influence of iodids, bromids, chlorids, ether, and 
chloroform upon cutaneous and intracutaneous tests with luetin, tuberculin, 
and other substances were studied. 
Potassium iodid administered orally in amounts of 10 gm. per day for 
two or three days was found to have the most influence upon skin reac- 
tions; the bromids exerted a similar but less marked influence, and the 
chlorids almost none. Ether and chloroform anesthesias exerted no demon- 
strable effects. 
Intracutaneous tests were mostly influenced; cutaneous tests were 
sometimes influenced and particularly by potassium iodid, while conjunctival 
tests were not influenced at all. • 
Lyons,5 Cole and Paryzek6 have also noted the influence of potassium 
iodid upon intracutaneous reactions with luetin. Farr7 has noted an 
apparently similar influence upon the luetin reaction by mercury admin- 
istered by inunction. 
I am of the opinion that the influence of these drugs upon skin reaction 
by increasing their severity and a tendency to papulation and pustulation 
is exerted upon the traumatic and non-specific phases. It would appear 
probable that the iodids, bromids, and to a less extent the chlorids, are cap- 
1 Jour. Amer. Med. Assoc., 1912, lix, 1598. 
2 Jour. Amer. Med. Assoc., 1915, Ixv, 404. 
3 Jour. Amer. Med. Assoc., 1916, lxvii, 718. 
4 Jour. Lab. and Clin. Med., 1917, 2, 401. 
5 Southern Med. Jour., 1916, 9, 483. 
6 Jour. Amer. Med. Assoc., 1917, 68, 1089. 
7 Jour. Amer. Med. Assoc., 1915, 64, 850. 680 
CLINICAL ALLERGY 
able of increasing tryptic activity by removal of antiferment, after the 
hypothesis of Jobling and Petersen,1 which is followed by the increased local 
digestion of proteins by the tissue proteases with the production of diffusible 
irritants. 
At any rate it is well for the physician to keep in mind the influence of 
these drugs upon intracutaneous reactions, and especially the influence of 
potassium iodid upon the luetin reaction, in order to avoid erroneous results. 
If the individual is known or suspected to be extremely sensitive to the 
serum, cutaneous tests should be done; otherwise the intracutaneous test is 
more sensitive and to be preferred when serum is to be given intravenously 
for therapeutic purposes. 
For conducting the cutaneous test either raw normal or immune serum 
(not purified or concentrated antitoxins) or dried serum proteins may be 
employed. Walker2 prefers the latter, which is prepared as follows: 
The proteins are precipitated by adding to the serum several volumes 
of acetone and centrifuging. The precipitate is well mixed with absolute 
alcohol, centrifugalized, and the alcohol decanted; this process is repeated. 
The protein is then wrashed twice with ether in the same manner. The 
material is then dried and the powder employed for cutaneous tests accord- 
ing to the technic previously described. 
Reactions develop within fifteen minutes as described; Fig. 159 shows 
the urticarial-like lesion that develops on the arm of one of my colleagues 
following scarification and the thorough application of a drop of horse-serum. 
This man is also susceptible, to a less extent, to guinea-pig- and rabbit-serum, 
and is seized with sneezing and distressing dyspnea after entering a house 
where these animals are kept. 
For intracutaneous tests sterile raw normal or immune serum (not purified 
or concentrated serum) should be diluted 1 : 50 with sterile saline solution 
and 0.1 c.c. (= 0.002 c.c. undiluted serum) injected intracutaneously; or, 
the dose injected may be larger, as 0.1 c.c. of 1 : 5 dilution (0.02 c.c. undiluted 
serum), if only a slight degree of hypersensitiveness is suspected. A small 
white wheal showing the little depressions of the hair-follicles is produced. 
An injection of 0.1 c.c. salt solution is made as a control. 
If the test is negative the wheals produced by both serum and saline 
solution disappear in a few minutes. If sensitiveness, exists, a genuine 
urticarial wheal develops in five to ten minutes, which reaches its maximum 
within one hour and fades away within a few hours. General symptoms of 
flushing of the face, tachycardia, and difficult breathing develop, but rarely 
and only in cases of extreme hypersensitiveness. 
SKIN TESTS FOR SERUM ALLERGY 
SKIN TESTS FOR POLLEN (HAY-FEVER) AND OTHER PLANT ALLERGIES 
Cutaneous tests are now employed almost solely, the conjunctival, 
nasal, and intracutaneous tests formerly used being practically abandoned. 
The allergens are prepared of a very large variety of pollens and grasses 
by different manufacturing laboratories and are available for the profession 
in convenient and generally satisfactory outfits. 
The pollens are mechanically isolated and macerated in 10 per cent, 
sodium chlorid solution, which has been rendered faintly alkaline with 
sodium hydroxid. The filtered extract is dialyzed until free from salts and 
1 Jour. Exper. Med., 1914, 19, 480. 
2 Jour. Med. Research, 1916—17, 35, 498. Fig. 159.—Local Serum Anaphylactic Reactions. 
Dr. W. P., anaphylactic reactions fifteen minutes after application of serum; rabbit-serum in the 
upper; horse-serum in the lower. Subject to asthmatic attacks when in an animal house or stable 
where rabbits, horses, and guinea-pigs are kept. SKIN TESTS FOR ALLERGIES TO HAIRS, DANDRUFFS, FEATHERS 
681 
the dialyzate precipitated with acetone and the pollen protein thus obtained 
dried with anhydrous acetone. The products are odorless and tasteless 
powders, insoluble in water or dilute acids, but readily soluble in physio- 
logic salt solution and dilute alkalies. Wodehouse1 has described a simple 
method employing carbontetrachlorid, which was found more rapid and to 
yield larger amounts of pollen. 
Coca2 has recently described methods for the collection of various pollens 
and prepares extracts as follows: The dried pollen is repeatedly extracted 
with three volumes of ether until the decanted ether shows only a slight 
yellow color. Four or five extractions suffice for ragweed pollens; fewer for 
grass pollens because they contain less oil. The ether is finally removed 
by immersing the beaker in hot water (50 to 60° C.). Each gram of pollen 
is now extracted with 300 c.c. of the following fluid: 
Sodium chlorid 0.5 per cent. 
Sodium bicarbonate in such concentration that 10 c.c. of 
the final flui(l equals about 3 c.c. of alkali. 
Carbolic acid in final concentration 0.4 “ 
The mixture is covered with toluol and extraction continued for four 
days at room temperature, the sediment being shaken up once or twice daily. 
The supernatant fluid is now decanted and filtered through a sterile Berke- 
feld filter. Each cubic centimeter usually contains about 0.3 to 0.4 mgm. 
of nitrogen. 
In spring hay-fever, perennial vasomotor rhinitis, and vernal conjunc- 
tivitis it is usually necessary to test with many different allergens (see list 
on page 664). In autumnal hay-fever tests with goldenrod and ragweed 
usually suffice. 
The tests are conducted as previously described and the results usually 
read within half an hour. Figure 156 shows the appearance of positive and 
negative reactions. 
Similar tests may be conducted for the allergies to rhus (dermatitis ven- 
enata), although the history is usually so definite that these tests are not 
required unless there is doubt regarding the kind of plant to which the 
patient is hypersensitive, as poison ivy, poison oak, or poison sumac. 
SKIN TESTS FOR ALLERGIES TO HAIRS, DANDRUFFS, AND FEATHERS 
These are conducted by the cutaneous method. The allergens are avail- 
able as prepared by different manufacturing biologic laboratories. These 
tests are especially useful in the diagnosis of the allergic asthmas and peren- 
nial vasomotor rhinitis. The history of the patient usually suggests the 
allergens, but if not, a series prepared of the hairs and dandruffs of the 
horse, cat, dog, and other domestic animals should be employed as well as 
allergens of goose and chicken feathers. 
For preliminary tests allergens of hairs and dandruffs may be prepared, 
according to Walker, by placing a handful of uncleaned hair or a few grams 
of dandruff in a small bottle, cover with 14 per cent, alcohol and extract 
the proteins, for several days, with frequent shaking, followed by filtration. 
A drop of this alcohol extract is used in the skin test. Purified alkaline and 
acid metaproteins from the hair of the cat and dog and dandruff of the 
horse may be employed, the immochemistry and preparation of these being 
described by Wodehouse.3 
1 Boston Med. and Surg. Jour., 1916, clxxiv, 430. 
2 Jour. Immunology, 1922, 7, 166. 3 Jour. Immunology, 1917, 2, 227, 237, 243. 682 
CLINICAL ALLERGY 
Since the yield of protein powders is usually very small, Walker pre- 
pares them in 1 : 100 dilution in N/100 sodium hydroxid or weak acid or 
alcohol, as necessity ihay require in each particular protein, and standard 
dilutions are made from this stock solution, as 1 : 1000, 1 : 10,000, 1 : 100,000, 
and 1 : 1,000,000. These various dilutions are used for skin tests, so that it 
is possible to determine the degree of sensitiveness. The dry undiluted pro- 
teins are kept in stock, from which these dilutions are made as required. 
Coca1 has described a method for preparing extracts of dandruff, hair, 
wool, and feathers which is essentially as follows: The oils are removed by 
mixing the material with three volumes of ether and after an hour or two 
the ether is decanted and discarded and the last traces completely removed 
by immersing the beaker in water at 50° to 60° C. Each gram of dried 
material is now mixed with 10 c.c. of toluol and then with 100 c.c. of the 
sodium bicarbonate extracting fluid described above for the extraction of 
pollens. The mixture is stoppered and extracted for twenty-four hours 
followed by centrifuging or paper filtration. This fluid is then passed 
through a sterile Berkefeld filter and usually contains about 0.5 mgm. of 
nitrogen per cubic centimeter. 
SKIN TESTS FOR FOOD ALLERGIES 
These are generally conducted by the cutaneous method, but as pre- 
viously stated, the intracutaneous method is more delicate, but likewise more 
subject to error in interpretation. Probably a good rule for practice is to 
first conduct cutaneous tests, and if the results are negative, intracutaneous 
tests may be performed. 
A very large number of allergens of vegetables, grains, fruits, meats, 
fishes, and other animal products, as cow’s milk and eggs, are now available 
in convenient form for these tests, being prepared by different manufacturing 
firms. Allergens of foods eaten after cooking may be prepared of these 
heated foods; this eliminates the heat-coagulable proteins unless they hap- 
pened to be partly hydrolyzed by the heating. Methods for the preparation 
of allergens of vegetable and animal foods have been fully described by 
Wodehouse and Olmstead.2 Shorter methods consist in the removal of vege- 
table or animal fats if any are present by extraction of the ground-material 
with petroleum ether or centrifuging in the case of milk, extracting with 
faintly alkalinized 10 per cent, solution of sodium hydroxid, removing the 
salts by dialyzation and precipitating the proteins with acetone, and drying 
into powders with anhydrous acetone. 
Coca3 has described a method for preparing food allergens which is 
essentially as follows; Dried materials, as whole wheal flour, rice meal, ground 
beans, etc., are weighed, and each kilogram thoroughly mixed with 150 c.c. 
of toluol and then with 2200 c.c. of the sodium bicarbonate extracting fluid 
described on p. 681 for the extraction of pollens. Extraction is continued at 
room temperature for twenty-four hours, being thoroughly shaken up at 
least once in this period, when the fluid is decanted and set aside. The 
material is now extracted again for twenty-four hours with 2000 c.c. of 
extracting fluid followed by decantation. The two fluids are now mixed, 
covered with toluol, and set aside for a few days for the collection of any 
sediment that may form. The supernatant fluid is now decanted or the 
sediment removed by centrifuging or paper filtration, the material being 
finally passed through a sterile Berkefeld filter. 
1 Jour. Immunology, 1922, 7, 166. 
2 Boston Med. and Surg. Jour., 1916, clxxvi, 467, 468; ibid., 1917, clxxvii, 85. 
3 Jour. Immunology, 1922, 7, 166. SKIN TESTS FOR FOOD ALLERGIES 
683 
Moist materials, from which little or no juice can be expressed, such as 
meats, sweet potatoes, green pea, turnip, cauliflower, siring bean, celery, cabbage, 
lettuce, spinach, cucumber, potatoes, lima bean, etc., are passed twice through 
a meat chopper, mixed with extracting fluid, and covered with a thin layer 
of toluol. The extraction is generally interrupted after twenty-four hours 
and the fluid is obtained by pressing the mixture in a stout towel by hand. 
The amount of extracting fluid varies with the nitrogen content. For the 
meat three or four volumes of the fluid (figured on the weight of the material) 
are used. For the vegetables, one or two volumes. 
Some of the shell fish yield a juice in addition to the meat. To the juice 
one-quarter volume of the following “preserving fluid” is added: 
Sodium chlorid 2.5 per cent. 
Sodium bicarbonate 1.5 “ 
Carbolic acid in final concentration 2.0 “ 
The meat is mixed with three or four volumes of the extracting fluid 
previously described and the two mixtures thrown together, covered with 
toluol and extracted for at least two weeks. The material is now decanted, 
filtered through paper, and finally a sterile Berkefeld filter. 
With such fruits and vegetables that can be peeled, as peach, tomato, 
orange, lemon, and grape-fruit, it is advantageous to make separate extractions 
of juice squeezed from the pulp and the ground up peelings. The juice 
may be extracted with one-quarter volume of “preserving fluid” and the 
ground up pulp and peelings with two or three volumes of “extracting fluid.” 
The two mixtures may- now be thrown together, covered with toluol, ex- 
tracted for two weeks at room temperature, and finally filtered through 
paper and sterile Berkefeld filters. 
Milk casein is prepared by adding acetic acid to milk and collecting the 
precipitate in a towel. This material is now washed by immersing the 
towel in several changes of water. The casein is now forced through a wire 
kitchen strainer, while kept moistened with distilled water. To this sus- 
pension is now slowly added 4 per cent, sodium hydroxid solution until 
nearly all of the casein is dissolved. Stirring is continued some time beyond 
this point, and if some casein still remains undissolved the solution will be 
neutral to litmus. The mixture is now filtered through paper and the 
filtrate kept under toluol. Before filtration through a Berkefeld filter the 
solution is further diluted with 0.4 per cent, carbolic acid in 1 per cent, 
sodium chlorid to a nitrogen content of 2 mgm. per cubic centimeter. After 
filtration the nitrogen content per cubic centimeter is further reduced to 
1 mgm. 
From the original acetic acid filtrates the other proteins may be pre- 
cipitated by adding volumes of saturated ammonium sulphate and the 
precipitate collected on paper freed of the ammonium salt by dialysis against 
changes of 1 per cent, sodium chlorid. 
Skin tests for allergy to eggs may be conducted by rubbing a little egg 
white and yolk into abrasions or cuts; if intracutaneous tests are conducted 
the amount injected may be 0.1 c.c. of a 1 : 100 dilution. 
Intracutaneous tests are conducted by dissolving the dried proteins in 
physiologic saline solution; usually the injection of 0.1 c.c. of a 1 : 100 dilu- 
tion is sufficient. 684 
CLINICAL ALLERGY 
SKIN TESTS FOR ALLERGIES TO DRUGS 
These are usually conducted by the cutaneous method. In tests for 
allergy to quinin both Boerner1 and Malley and Richey2 have observed 
almost immediate reactions following the application of 1 or 2 drops of a 
10 per cent, solution of quinin bisulphate to abrasions. In allergies to 
procain, Lane3 has elicited reactions by the application of a 2 per cent, solu- 
tion and shows the appearance of positive reactions. The tests may be con- 
ducted by adding a minute amount of the drug to an abrasion and dissolving 
in the exuding serum aided by the addition of 1 drop of saline solution. A 
control with saline solution should always be included. 
Tests for allergies to arsphenamin, mercury, and other metals may be 
conducted in the same manner. With the vegetable drugs, as belladonna, 
stramonium, etc., fluidextracts may be employed. Extracts are readily 
prepared by macerating the leaves and extracting with 14 per cent, alcohol 
for several days, the tests being conducted by applying 1 drop of the extract 
to an abrasion of the skin. 
If intracutaneous tests are conducted, the amount injected should not 
be greater than 0.1 c.c. of 1 : 100 dilution. 
SKIN TESTS FOR ALLERGIES TO BACTERIA, FUNGI, AND PROTOZOA 
In Allergic Asthma, Bronchitis, and Rhinitis.—Cutaneous or intracuta- 
neous tests may be conducted. Stock allergens of different bacteria are 
available in dried powder form for cutaneous tests in exactly the same 
manner as the hay-fever tests. I prefer intracutaneous tests conducted 
with separate autogenous antigens prepared of the different bacteria re- 
covered from cultures of the sputum, nose, or other focus. 
Stock dried allergens are prepared by Walker4 by cultivating the bac- 
teria on an appropriate medium and securing them in a sediment by remov- 
ing the growths with saline solution containing 0.5 per cent, phenol and cen- 
trifuging. The sediment is now thoroughly mixed with phenolized saline 
solution, centrifugalized, and the saline decanted. The sediment is now 
mixed with absolute alcohol containing 0.5 per cent, phenol, centrifugalized, 
and the alcohol decanted. This process is repeated once more with alcohol, 
twice with ether, and the sediment dried into a powder. The method of 
conducting the tests has been described. 
For intracutaneous tests with autogenous antigens I proceed as follows: 
1. Cultures are made on Petri plates of hormone blood-agar and the 
different bacteria isolated and identified. 
2. Pure cultures of these are grown on appropriate solid media for twenty- 
four to forty-eight hours and removed with sufficient saline solution to give 
dense suspensions of at least 2,000,000,000 per cubic centimeter, or the 
bacteria may be grown in a fluid medium and secured by centrifugalization. 
If any medium has been carried into the suspension it is filtered through 
sterile paper and the bacteria are washed once with saline solution con- 
taining 0.5 per cent, phenol before being resuspended in the saline solution. 
3. Each suspension is heated in a water-bath at 60° C. for one hour, 
cultured for sterility, and preserved 'with sufficient phenol to make 0.5 per 
cent. 
4. For the skin tests a small amount of each suspension is diluted with 
sterile saline solution to give approximately 500,000,000 per cubic centi- 
1 Jour. Amer. Med. Assoc., 1917, lxviii, 907. 
2 Arch. Int. Med., 1919, 24, 378. 
3 Arch. Dermat. and Syph., 1921, 3, 235. 
4 Jour. Med. Research, 1916-17, 35, 488. TUBERCULIN REACTION 
685 
meter; 0.1 c.c. of each is injected intracutaneously. A control injection of 
saline solution is made at the same time. 
5. The reactions are read approximately one and twenty-four hours 
later. 
6. Those bacteria causing positive reactions are then incorporated into 
an autogenous vaccine prepared from the stock suspensions. 
As a general rule the bacteria recovered from cases of rhinitis, bronchitis, 
and asthma are of the following: staphylococci (albus and aureus), strep- 
tococci (hemolytic and non-hemolytic), pneumococci (usually Type IV), 
Micrococcus catarrhalis, and Bacillus pseudodiphtheria. Tests are not 
conducted with the fungi and yeasts. Great care must be exercised in 
conducting these intracutaneous tests with bacteria, as some, and especially 
M. catarrhalis, are irritating and likely to produce “positive reactions” in 
nearly every case (Rackemann). For this reason Rackemann prefers 
injecting 0.1 to 0.25 c.c. of the bacterial suspensions (1,000,000,000 per 
cubic centimeter) subcutaneously and selecting for treatment those produc- 
ing local reactions. 
Allergic Skin Tests in Focal Infections and Mixed Infections.—I have 
also used this technic for determining whether or not sensitization exists to 
bacteria recovered from the apical abscesses, tonsillar cultures and other 
foci in relation to focal infection; also with the different bacteria recovered 
in cultures from chronic suppurative lesions, as otitis media. Organisms 
producing positive reactions are incorporated into autogenous vaccines for 
treatment. 
TUBERCULIN REACTION 
An account of Koch’s discovery of tuberculin in 1891 is given in the 
chapter on Tuberculin Therapy. Suffice it to say here that Koch was 
most interested in the curative properties of tuberculin, and while he has 
accurately and clearly described the classic picture of the systematic tuber- 
culin reaction, he failed to appreciate the true significance of the reaction at 
the site of injection, although its occurrence is carefully noted. 
The Tuberculin Reaction.—The reaction to tuberculin is characterized 
by three essential features: 
1. A constitutional reaction, consisting of fever and the accompanying 
general symptoms of lassitude, anorexia, and rapid pulse, varying in severity 
with the intensity of the reaction. 
2. A local reaction at the site of administration, varying in intensity 
from slight tenderness and redness to severe inflammation with adenitis. 
3. A focal reaction about the tuberculous lesion. 
These reactions do not by any means run parallel. An intense local 
reaction may occur, with no or but slight constitutional disturbance. Not 
infrequently, and particularly in slight pulmonary lesions, signs indicating 
a focal reaction may not be elicited. 
Nature of the Tuberculin Reaction.—Koch believed that the tuberculin 
reaction was due to a summation of the effects of the injected toxin and the 
toxic bodies formed by the tubercle bacillus within the infected host. Koehler 
and Westphal in 1891 suggested that, by a union of the tuberculin with the 
products of the tubercle bacillus, a third new body was formed in the tuber- 
culous focus. Marmorek in 1894 suggested that the tuberculin stimulated 
the tubercle bacilli to secrete a fever-producing substance. 
Finally, in 1903 von Pirquet and Schick explained the reaction as due to 
a “vital antibody reaction,” and this explanation is the one most generally 
accepted today. Tuberculin is not toxic for non-tuberculous individuals 686 
CLINICAL ALLERGY 
and produces no reactions upon cutaneous application, intracutaneous or 
subcutaneous injection. In tuberculous individuals, however, surprisingly 
small amounts produce marked reactions. The local reaction on the skin 
at the point of application or injection and the focal reaction of hyperemia 
and exudation of the tissues around the foci of disease are regarded as 
allergic reactions. The general symptoms of fever, tachycardia, lassitude, 
etc., are due to the absorption of toxic substances from the foci of disease, 
according to Kraus,1 absorption being stimulated and facilitated by the focal 
allergic shock. Furthermore, there is increased capillary permeability due 
in part to the . non-specific phase (see Chapter XXXIX) which results in an 
outpouring of plasma and enzymes and enhanced proteolytic digestion and 
production of toxic substances. 
The focal tuberculin reaction is primarily an allergic reaction. Evi- 
dently the phenomenon is one of cellular allergy, the antibody being in or 
upon the cells and the reaction being an intracellular colloidal phenomenon. 
Evidence of the cellular allergic nature of the reaction has been produced 
by Weil2 and more recently by Smith,3 with the Schultz-Dale method em- 
ploying the uteri of sensitized guinea-pigs. As discussed in a previous 
chapter, the humoral theory of anaphylaxis is inadequate for explaining the 
reaction on the basis of the production of a poison (anaphylatoxin) by either 
specific amboceptors and complement or by a ferment; one millionth of a 
cubic centimeter of Koch’s old tuberculin may elicit a reaction in the tuber- 
culous and it is impossible to understand how sufficient anaphylatoxin could 
be produced from so small an amount. 
In addition to being an allergic phenomenon the focal tuberculin reac- 
tion is probably also in part non-specific due to the stimulation of cells 
(plasma-activation of Weichardt) by proteins in the tuberculin. Further- 
more, the injection of other non-specific agents as vaccines and proteoses 
intravenously and milk intramuscularly, may elicit reactions around foci of 
tuberculosis analogous to those engendered by tuberculin. These reactions 
are regarded as due to stimulation of cellular activity with increased capil- 
lary and lymphatic permeability and exudation of serum, leukocytes, 
enzymes, etc. It is to be emphasized, however, that in comparison to 
tuberculin relatively large amounts of these non-specific agents are required 
to elicit focal reactions. Autogenous sputum vaccines are especially likely 
to produce these reactions. 
How sensitization of the skin, mucous membranes, and other organs of 
the body to tuberculin is brought about in tuberculosis is not known; experi- 
mental attempts to sensitize the skin of normal animals with injections of 
tuberculin have usually failed, although the skin has been sensitized by 
injections of the proteins of tubercle bacilli. The chemical nature of 0. T. 
is not definitely known, but it probably contains proteoses and sensitization 
of animals with proteoses is generally weak and irregular. It is highly prob- 
able that the sensitinogen is a product of living bacilli in the tissues or a 
new combination substance with the tissues, inasmuch as sensitization can 
only be produced regularly by foci of active tuberculosis. 
Specificity of the Tuberculin Reaction.—The tuberculin reaction is 
highly specific. This does not mean that every case of tuberculosis will 
give a tuberculin reaction, and positive reactions are occasionally found in 
apparently healthy persons and cattle. Furthermore, the focal tuberculin 
reaction may be partly non-specific and capable of being engendered by non- 
1 Jour. Med. Research, 1916, 35, 43. 
2 Jour. Amer. Med. Assoc., 1917, 68, 972. 
3 Jour. Immunology, 1922, 7, 47. TUBERCULIN REACTION 
687 
specific agents, as previously mentioned. The conditions under which a 
negative reaction may occur in the presence of tuberculosis are, however, 
fairly well understood, and physicians should be thoroughly acquainted with 
these. Likewise most instances in which a positive reaction was observed 
in the apparent absence of tuberculosis have usually narrowed down to the 
fact that the lesion was so small or so situated as to escape detection, and, 
indeed, this has been shown so conclusively by autopsies that, in the presence 
of a tuberculin reaction, on the autopsy must rest the burden of proof and 
blame. When we realize how small a lesion may produce hypersensitiveness, 
it will readily be understood how easily the clinician and pathologist may 
fail to detect the lesion. 
A large part of our knowledge regarding the specificity of the tuberculin 
reaction has been gained from veterinary practice, as the results of a test in 
an animal could immediately be controlled by the autopsy findings. Thus 
Fraenkel1 collected from the literature 8000 carefully observed instances, 
and found only from 2 to 3 per cent, of differences between the result of the 
tuberculin test and of the autopsy. Voges,2 in 7327 instances, noted 2.7 
per cent, of contradictions. Kuhnau,3 Bang,4 and von Behring5 speak of 
similar experiences. 
It has long been known that the prevalence of tuberculous findings 
anatomically far exceeded the number of cases recognized clinically. Among 
cattle, anatomic tuberculosis is found in from 12 to 25 per cent., and about 
80 per cent, or more react to tuberculin. In many of the latter, however, 
the disease does not progress, but, on the contrary, tends to recede. 
Similar conditions exist in human pathology. That tuberculosis is very 
frequent among adults is now well known. The figures of Nageli6 and 
Burkhardt7 showed that the increasing frequency of tuberculous infection 
with advancing years reached over 90 per cent, among those past the eigh- 
teenth year; these figures are now well corroborated. Hamburger,8 in the 
published results of an analysis of 848 autopsies on children, showed that 
tuberculosis was in-evidence in 40 per cent., increasing from 4 per cent, among 
infants under three months of age to 70 per cent, among children from 
eleven to fourteen years. This explains, in part at least, the relatively high 
resistance of children to tuberculin, the difficulty there is said to be in 
eliciting reactions, and the necessity that exists for using large doses. 
Usually, when children fail to react, it is because they are not tuberculous 
or because the lesion is too small, whereas in later years, until adult life is 
reached, the reaction is observed with increasing frequency and with smaller 
doses, because the incidence of infection increases progressively from 5 per 
cent, during infancy to 90 per cent, and over in adult life. 
The prevalence of tuberculosis, however, by no means indicates that 
the infected individual suffers ill health or will succumb to the infection. 
An individual may be enjoying excellent health, and still harbor a tuber- 
culous lesion, and display a marked degree of hypersensitiveness to tuber- 
culin. Such a person is not usually regarded as tuberculous until there are 
tangible symptoms referable to its existence. It is important to remember 
that tuberculin is an index of tuberculous infection, and not of disease in a clinical 
1Zeitschr. f. Tuberk., 1900, i, 291. 
2 Tuberculin und Organismus, Jena, 1905, 77. 
3 Berl. tierarztl. Wchn., 1899, 78. 
4 Sixth International Congress on Tuberculosis, 1908, 211. 
8 Beit. z. exper. Therap., 1905, x, 1. 
6 Virch. Arch. f. path. Anat., 1900, clx, 426. 
7 Zeitschr. f. Hyg. u. Infectionsk., 1906, liii, 139. 
* Wien. klin. Wchn., 1907, xx, 1070. 688 
CLINICAL ALLERGY 
sense. Numbers of persons and cattle reacting to tuberculin remain healthy 
and do not develop symptoms of disease, the autopsy disclosing the presence of 
inactive or regressing lesions. 
In former years it was considered possible to obtain false positive reac- 
tions in convalescents and patients in an enfeebled condition who were 
non-tuberculous, and also in other diseases, such as syphilis, leprosy, and 
actinomycosis. More accurate anatomic statistics and careful studies of 
the tuberculin test administered to a large number of individuals, healthy, 
tuberculous, and sufferers from other diseases, have gradually changed the 
attitude of the profession and served to establish the high specificity of the 
tuberculin reaction. 
Sources of Error in the Tuberculin Reaction.—From the foregoing it 
will readily be understood that most errors in the tuberculin reaction refer 
to false negative rather than to false positive reactions. 
False positive reactions may be observed in leprosy, where the bacillus 
bears such close morphologic and biologic resemblance to the tubercle bacil- 
lus, and it is likewise true that massive doses of tuberculin injected sub- 
cutaneously may produce a toxic fever in debilitated individuals, but posi- 
tive reactions in healthy individuals can usually be ascribed—(a) to a small 
hidden tuberculous lesion or (b) to a healed tuberculous lesion. As just stated, 
tuberculin simply indicates hypersensitiveness to the tubercle protein, and 
this may exist with a very small unimportant lesion, or persist after a lesion 
has been “healed” to the extent of encapsulation. 
False negative reactions are much more likely to occur, and the various 
conditions that may be responsible for these should be well understood and 
remembered. 
1. In the final stage of tuberculosis, especially in miliary tuberculosis 
and in tuberculous cachexia, as in the third stage of pulmonary tubercu- 
losis, the tuberculin reaction may be negative, or be attained only after the 
injection of very large doses. There is a lessened cutaneous reactivity 
(cachectic reaction), marked by the appearance of colorless or pinkish spots, 
instead of an intense papillary eruption. Koch and Ehrlich have explained 
this by assuming that the tissues had become too thoroughly saturated with 
tuberculin produced at the infected area to respond to further artificial 
additions. This condition may be regarded as analogous to a state of anti- 
anaphylaxis in which we may consider the free and sessile receptors united 
with the tubercle protein or the cells loaded with the antigen, with depression 
of cellular activity. 
2. In the first stage of infection. At this period the antibody has not 
been formed in sufficient amounts, different authors obtaining such findings 
in nurslings. 
3. In small, completely healed lesions, especially in those showing 
nothing but scar tissue, as in the apex of a lung. In these the antibody 
and cellular hypersensitiveness have disappeared, and while the lesion 
may have been tuberculous, one cannot tell anatomically, in a given instance, 
whether or not hypersensitiveness should have been present. 
4. During continued treatment with tuberculin, when the reaction may 
be negative owing to a condition of anti-anaphylaxis. 
5. During measles. As von Pirquet and Preisich have demonstrated, 
the cutaneous reactivity disappears during the first days after the erup- 
tion, reappearing after about a week. Greuner showed that the Stich- 
reaktion which occurs after large doses of tuberculin did not disappear 
entirely, indicating that in measles there is a lessened activity rather than 
a total disappearance. TUBERCULIN REACTION 
689 
6. Finally, according to von Pirquet, there are a few cases in which 
we have a minimal reactivity and to which none of the former explanations 
can be applied. Some cases of active tuberculosis may show only a slight 
hypersensitiveness, although they are not cachectic. 
Of course, it can readily be understood that the use of weak or an other- 
wise unsatisfactory solution of tuberculin and an improper dosage and tech- 
nic will greatly influence the results. Likewise errors in interpreting what 
constitutes a positive tuberculin reaction are to be guarded against. This 
applies especially to veterinary practice. For example, cattle brought from 
the fields and confined in a stall for the purpose of making a tuberculin test 
may exhibit a fever for several days without apparent cause. On the other 
hand, dishonest dealers may force cattle to drink cold water or have given 
cold water irrigations just before the temperature is taken, preventing the 
registration of a febrile reaction. 
Methods of Conducting the Tuberculin Test.—The object of the tuber- 
culin test is to introduce sufficient tubercle protein to react with the tuber- 
cular antibody, which shows its presence and effects by a general, a local, 
or a focal reaction, or by a combination of these. Various methods have 
been proposed, of which the following are best known: 
1. The subcutaneous tuberculin test is the oldest test of its kind, having 
been discovered by Koch1 in 1891. It consists in the subcutaneous injec- 
tion of old tuberculin. A positive reaction manifests itself in a constitu- 
tional disturbance, accompanied by fever, a local reaction at the site of 
injection (“Stichreaktion”), and frequently a focal reaction at the site of 
tuberculous disease. 
2. The cutaneous tuberculin test of von Pirquet,2 consisting in the local 
application of old tuberculin to a superficial abrasion of the skin. A positive 
reaction is indicated by redness, edema, and other inflammatory phenomena. 
3. The conjunctival tuberculin test of Wolff-Eisner3 and Calmette,4 con- 
sisting in the local application to the conjunctiva of one eye of a drop of 1 
per cent, solution of old tuberculin or purified tuberculin. A positive reac- 
tion is indicated by congestion and lacrimation. 
4. The percutaneous tuberculin test of Moro and Doganoff,5 consisting 
in the application of tuberculin ointment prepared by mixing equal parts 
of old tuberculin and anhydrous lanolin and applying it to the skin over the 
upper portion of the abdomen or about the nipple. A positive reaction is 
indicated by an efflorescence of papules upon the anointed skin. 
Analogous to this test is that of Ligniere, who rubs the skin vigorously 
with alcohol and xylol until slightly reddened. After drying, 5 or 6 drops of 
undiluted tuberculin are applied and rubbed in with a rubber-coated finger 
for one or two minutes. 
5. The intracutaneous tuberculin test of Mendel6 and Mantoux,7 con- 
sisting in injecting into the superficial layers of the skin 0.05 c.c. of diluted 
old tuberculin. A positive reaction is denoted by infiltration and hyperemia 
about the site of injection, similar to the reaction to the cutaneous test. 
Comparative Delicacy and Relation of the Various Tuberculin Tests.— 
In judging of the comparative delicacy of the various tuberculin tests by a 
review of the literature, it must be remembered that results will vary accord- 
1 Deut. med. Wchn., 1891, xvii, 101. 
2 Berl. klin. Wchn., 1907, xliv, 699. 
3 Discussion, Berl. klin. Wchn., 1907, xliv, 70. 
4 Presse medicale, 1907, xv, 388. 
8 Wien. klin. Wchn., xx, 933. 
6 Med. Klin., 1908, iv, 402. 
7 Munch, med. Wchn., 1908, No. 40. 690 
CLINICAL ALLERGY 
ing to the portion of the body inoculated, as sensitiveness of the cells varies 
in different parts of the body, and there are individual differences in various 
persons that are difficult or impossible to explain. 
By comparing the figures obtained as the result of different tests upon 
the same person with the relative frequency of individual tests, Hamman 
and Wolman1 have drawn the following conclusions: 
“1. The intracutaneous and subcutaneous local tests are the most deli- 
cate we possess. They reveal practically the full percentage of tuberculosis- 
infected individuals. 
“2. In the order of their sensitiveness, the tests arrange themselves as 
follows: 
Intracutaneous Test. 
Subcutaneous Local Test. 
Cutaneous Test. 
Subcutaneous Test. 
Percutaneous Test. 
Conjunctival Test. 
“3. There is a definite but not a constant relation between the vari- 
ous tests. An individual reacting to the conjunctival test will, as a rule, 
give all the others, but not always. The cutaneous or the subcutaneous 
tests may be negative when the conjunctival is positive. The subcutaneous 
positive when the cutaneous is negative, etc. Some of the unusual varia- 
tions may, no doubt, depend upon faulty technic in performing the tests, 
but all can certainly not be thus explained. Local changes in sensitiveness 
and variation in the facility of absorption are probably factors, but the 
exact conditions are not understood. 
“4. We have been unsuccessful in an attempt to make the cutaneous test 
with different strengths of tuberculin equivalent to the conjunctival test.” 
Although the subcutaneous test may be dangerous on account of the 
harm that may result from too severe focal reactions, yet, when carefully 
conducted, it is frequently the method of choice, especially in the diag- 
nosis of an obscure pulmonary lesion, and in bone, joint, skin, and other 
local infections, where the focal reaction may be detected by direct ex- 
amination. It is to be emphasized, however, that the absence of focal 
changes during a constitutional reaction does not exclude the tuberculous 
nature of a suspected lesion. In children, as shown by Hamill, Carpenter, 
and Cope,2 the various tuberculin reactions are likely to yield results that 
are quite similar. 
The Value of Tuberculin in Diagnosis.—As previously stated, a reaction 
to tuberculin means essentially that the individual reacting has a tuberculous 
infection, and in itself means nothing more. Since tuberculin tests disclose 
inactive and relatively benign tuberculous infections, it has little value, in 
doubtful cases, in aiding us to decide whether or not the individual has active 
disease, which clinically is the type of infection about which we are most 
concerned. Lack of critical discernment in the interpretation of the reac- 
tion and its apparent indefiniteness have contributed toward diminishing 
its diagnostic value. 
A positive tuberculin reaction is to be regarded as a symptom, or as another 
link in the chain of clinical evidence, but is not in itself indisputable evidence 
that a certain lesion is tuberculous, for it can never decide with certainty an 
otherwise doubtful diagnosis. A similar example is that of a positive Wasser- 
mann reaction in a patient with a lesion in the throat; such a reaction does 
1 Tuberculin in Diagnosis and Treatment, 1912, Appleton & Co., 167. 
2 Archiv. Int. Medicine, 1908, ii, 405. TUBERCULIN REACTION 
691 
not necessarily mean that the lesion is syphilitic, for the lesion itself may 
be cancerous, although, coincidentally, a latent syphilitic infection may be 
present. 
If tuberculin could differentiate between active and inactive lesions 
according to the degree of reaction, its value would be greatly increased. 
While the studies of Krompecker and Romer upon animals indicate that 
the more virulent the infection the greater the degree of hypersensitiveness, 
no such fixed relation exists in man. All that may be said is that, in general, 
the severer the reaction, the more acute the infection. On the other hand, 
acute miliary tuberculosis or chronic cachectic cases may react negatively. 
The conditions under which a negative reaction may he obtained in a tuber- 
culous person are to be carefully borne in mind, for if these can be excluded, a 
negative tubercidin reaction precludes, in all probability, an active or clinically 
important tuberculous lesion. Tubercidin has, therefore, a higher negative than 
a positive value in diagnosis. 
While it is obviously beyond the scope of this volume to discuss the 
diagnostic value of tuberculin in tuberculous infection of the different 
organs, I may briefly refer to a few of the more important conclusions reached 
by individual investigators of large experience in this particular field: 
1. In the diagnosis of pulmonary tuberculosis, while a positive constitu- 
tional or local tuberculin reaction is never conclusive evidence that a defi- 
nite pulmonary lesion is tuberculous, a focal reaction, on the other hand, 
tells definitely of the presence of the disease, and shows, in some measure 
at least, its extent. In these questionable cases, therefore, the subcuta- 
neous injection of tuberculin finds its most important application, since a 
definite focal is of more value than a local reaction. 
Tuberculin may also be of service when the symptoms suggest the 
presence of a pulmonary tuberculous lesion, but the physical signs are in- 
definite. Here the focal reaction is likely to be slight and to escape detec- 
tion, so that one of the cutaneous tests are usually employed. 
2. In the diagnosis of bone, joint, and glandular tuberculosis the sub- 
cutaneous test is likely to be most valuable, on account of the focal reac- 
tion of hyperemia, swelling, heat, and pain about the lesions. The cutaneous 
test is also valuable, but since this may react on account of the presence of a 
lesion situated elsewhere, the focal reaction is more conclusive. According 
to Hamman and Wolman, (a) a focal reaction confirms the diagnosis of tuber- 
culous bone or joint disease; (b) an absence of reaction to the subcutaneous 
test excludes, with the highest probability, the presence of tuberculosis. 
3. In the diagnosis of genito-urinary and pelvic tuberculosis a positive 
tuberculin reaction is of little value unless the subcutaneous method is 
employed and the physician is sure of his ability to detect a focal reaction, a 
proceeding that may be very difficult or indeed impossible. A positive 
cutaneous test indicates the presence of a lesion somewhere in the body, 
without disclosing the nature of the renal or pelvic lesion. A negative reac- 
tion, however, is of more value, especially when the physician bears in mind 
the conditions under which a falsely negative result may occur. 
4. In the diagnosis of tuberculosis of the eye, ear, and larynx the tuber- 
culin reaction usually has a limited value, because the nature of the disease 
can be so readily detected by direct inspection. In tuberculosis of the 
larynx a focal reaction may be dangerous on account of edema. Similarly 
in advanced tuberculosis of the ear a focal reaction may lead to extension of 
the process to the meninges. In these instances, therefore, a cutaneous test 
should first be made, and if it is found to be negative or inconclusive, the 
subcutaneous test should be applied with extra caution. 692 
CLINICAL ALLERGY 
5. In the diagnosis of tuberculosis of the skin tuberculin may occasion- 
ally prove of value—especially the focal reaction following subcutaneous 
injection or a direct local application upon the lesion with a weak dilution 
such as a 1 per cent, solution of old tuberculin. 
6. In the diagnosis of tuberculosis of a serous membrane tuberculin usually 
possesses a limited value. In tuberculous meningitis the subcutaneous test 
is contraindicated, as a focal reaction may do harm. Owing to the acute 
infection the cutaneous test may be negative, and even if positive, would 
not aid greatly in the diagnosis because the meningeal condition is always 
secondary to a primary focus. Tuberculous pleurises, dry or with effusion, 
and unaccompanied by evident pulmonary disease, are frequently associated 
with a low-grade tuberculin hypersensitiveness. According to Hamman 
and Wolman, in a large proportion of cases of pleurisy with effusion the con- 
junctival test is negative and the cutaneous test but mildly positive. Ban- 
delier and Roepke assert that in a dry pleurisy increased pain and more pro- 
nounced and extensive friction may occur during a constitutional reaction 
to the subcutaneous test and indicate a focal reaction. In tuberculous 
peritonitis the tuberculin test is frequently negative. A positive reaction has 
far more value, especially in virulent types of the disease, which come on 
insidiously with little or no constitutional disturbance. 
The Value of Tuberculin in Prognosis.—As previously stated, tuber- 
culin may not react in the very early and very late cases of tuberculosis. 
In patients with rapidly advancing lesions the power to react tends to 
decrease and frequently is absent. But this condition is apparent without 
the aid of tuberculin. While Wolff-Eisner and Stadelmann1 laid some 
stress upon the conjunctival reaction in prognosis, others have been unable 
to confirm the results, and, as stated by Hamman and Wolman,2 tuberculin 
fails to yield us information of prognostic value that other methods of clinical 
observation do not bestow. 
The Dangers of Tuberculin.—Practically, the only danger lies in the 
subcutaneous and conjunctival tests. With the subcutaneous test, the 
greatest danger in pulmonary tuberculosis is the possibility of overdosage, 
with the production of an extensive focal reaction which may bring on hem- 
orrhage or lead to local extension of the lesion. Because of its very impor- 
tant bearing on tuberculin treatment, the subject will be discussed in the 
next chapter. The same danger of excessive focal reaction holds for tuber- 
culous meningitis (increased intracranial pressure), in tuberculous laryngitis 
(edema), and in tuberculosis of the ear and nasal accessory sinuses (extension 
to meninges). 
The cutaneous, intracutaneous, and percutaneous reactions are practi- 
cally devoid of danger, providing that a careful technic is observed. 
Regarding the dangers of the conjunctival test, opinions differ. There 
can be no doubt, however, but that distressing sequelae, such as severe 
recurring conjunctivitis, phlyctenular conjunctivitis, and corneal ulcers 
with permanent opacities have resulted from the use of this test. Most of 
the unfavorable results have followed instillation in already diseased eyes, 
or of too strong solutions, but this is not true of all cases. As will be pointed 
out further on, in the first instillation not over 1 per cent, of old tuberculin 
should be used, and a second instillation should never be made in the same 
eye for at least several years. 
Since old people are especially prone to conjunctival inflammation 
and corneal ulceration, it is probably better to exclude them from the test. 
1 Deutsch. med. Wchn., 1908, xxxiv, 180. 
2 Tuberculin in Diagnosis and Treatment, 1912, 179. THE SUBCUTANEOUS TUBERCULIN REACTION 
693 
Another drawback to this test is the possibility of a “flare up” in the eye 
following subsequent subcutaneous administration of tuberculin, either for 
diagnostic or therapeutic purposes. Hamman and Wolman, however, do 
not consider this dangerous, and have observed but 2 cases in an extensive 
experience. 
As has been stated repeatedly, the subcutaneous injection of tuber- 
culin is resorted to at the present time mainly for the purpose of eliciting a 
focal reaction, and, as a rule, this is more easily appreciable when the gen- 
eral reaction is well marked. If tuberculin is used simply to establish whether 
or not a person is hypersensitive, this fact may be demonstrated by employing 
much smaller doses, as by the cutaneous, intracutaneous, or conjunctival tests, 
the patient being spared the discomfort of the constitutional symptoms. 
Variety of Tuberculin.—Koch’s old tuberculin is now used almost exclu- 
sively. The technic of its preparation is given in the chapter on Tuberculin 
Therapy. 
Manufacturing chemists usually market this tuberculin in a series of 
dilutions, labeled and accompanied by explicit directions, so that the phys- 
ician may administer practically any dose desired by injecting so many 
minims of such or such dilution. Otherwise a series of dilutions are readily 
prepared in the physician’s office or dispensary, according to the method 
followed by Hamman and Wolman: 
(a) Seven wide-mouthed bottles of about from 12 to 15 c.c. capacity, 
and fitted with ground-glass or rubber stoppers, are sterilized, labeled, and 
numbered from 2 to 8. 
(b) The diluent is sterile 0.8 per cent, salt solution with 0.25 per cent, 
pure phenol. This is readily prepared by adding 8 gm. of pure sodium chlorid 
and 2.5 c.c. of pure phenol to 1000 c.c. of distilled water. Dissolve, and filter 
into one large Erlenmeyer flask or, preferably, into ten smaller flasks. Steril- 
ize in the Arnold sterilizer for one hour, or in the autoclave for twenty 
minutes, or by gently boiling for fifteen minutes on each of two consecutive 
days. 
(c) Into each bottle place 9 c.c. of the diluent with a graduated and 
sterile pipet. Bottle 1 contains pure tuberculin. To bottle 2 add 1 c.c. 
of tuberculin and shake carefully; to bottle 3, 1 c.c. from bottle 2 and shake; 
to bottle 4, 1 c.c. from bottle 3 and shake; continue in this manner to bottle 
8, from which 1 c.c. is discarded. 
(d) We now have the following dilutions: 
The Subcutaneous Tuberculin Reaction 
No. 1—pure tuberculin. 
No. 2—0.1 c.c. tuberculin in each cubic centimeter (100 mg. tuberculin.) 
No. 3—0.01 c.c. “ “ “ “ “ ( 10 
No. 4—0.001 c.c. “ “ “ “ “ ( 1 
No. 5—0.0001 c.c. “ “ “ “ “ (0.1 
No. 6—0.00001 c.c. “ “ “ “ “ ( 0.01 “ 
No. 7—0.000001 c.c. “ “ “ “ “ ( 0.001 “ 
No. 8—0.0000001 C:C. “ “ “ “ “ ( 0.0001 “ 
(e) These dilutions are usually prepared every two weeks. When not 
in use, the bottles are kept in a cool, dark place. It may not be necessary 
to prepare all dilutions. For example, dilutions Nos. 3 and 4 are suffi- 
cient for diagnostic purposes, as 0.1 c.c. of No. 4 equals 0.1 mg. of tuberculin 
and 1 c.c. of No. 3 equals 10 mg., thus affording an ample range of dosage. 
Method of Conducting the Test.—1. The patient’s temperature and 
pulse-rate should be taken every two hours for from four to seven days. 694 
CLINICAL ALLERGY 
This is easily accomplished in a hospital; ambulatory patients can usually 
be readily trained to take their own temperature. All observations should 
be recorded in writing, and preferably on a temperature chart. The pa- 
tient’s temperature must be constantly below 99° F. before beginning the 
test, and, if necessary, prolonged rest in bed should be enforced to overcome 
any existing fever. The test may be given in spite of a daily rise of not over 
100° F., but tuberculin by subcutaneous injection should be given only 
■exceptionally to febrile patients. 
2. A very careful physical examination should be made, and the results 
recorded just before and just after the test in order to detect a focal reaction. 
This is extremely important, for it is our main justification for injecting the 
tuberculin. 
3. Injections are made subcutaneously in the region of the back, below 
the angle of the scapula, or in the arm. The skin needs no preparation other 
than to be rubbed with alcohol. The injections are best given during the 
late evening hours or early in the morning, in order that the temperature 
and pulse changes may be observed, especially twelve hours after the injec- 
tion. Records of temperature and pulse should be made every two hours 
during the day and night for forty-eight hours following an injection. 
Hamman and Wolman recommend the “Tuberculin Sub 2 Syringe,” made 
by the Randall, Faichney Co. Each syringe should be sterilized by boiling 
prior to use, and it is recommended that a separate syringe be provided for 
each dilution. 
4. Considerable controversy has arisen over the size of the doses to be 
•employed. Koch’s directions called for 1 mgm. at first, then for 5, then 
for 10, and if no reaction followed this dose, it was repeated. There is a 
decided tendency, however, to use smaller doses. If the object is to estab- 
lish the presence or absence of tuberculin hypersensitiveness, mild reactions 
will suffice, and for this purpose small doses repeated or in gradually increas- 
ing amounts may be given. If one aims to produce a focal reaction, larger 
doses and more rapid increase are desirable. Hamman and Wolman advise 
the following plan for adults: For the first dose, mg. is given. If there is 
a slight febrile reaction of about one-half a degree, and especially if this is 
accompanied by a local reaction at the point of injection, the second dose, 
which is the same as the first, is given at the end of forty-eight hours. The 
reaction will now most likely be more conclusive. If there is no appreciable 
reaction after the first dose, the second, consisting of 1 mg., is given at 
the end of forty-eight hours, and the third dose, if one is necessary, con- 
sists of 5 mg. Occasionally the third dose must be repeated, or even 10 
mgm. given if the negative result is at variance with the clinical impressions. 
If the temperature shows any irregularities, the intervals between injec- 
tions should be prolonged to three or four days or more. 
Roth-Schultz advise the following doses: 0.5 mg.; 1.25 mg., and 2.5 mg. 
as the terminal dose. 
For children under fifteen years of age smaller doses are indicated—an 
initial dose of 0.1 mg. and a terminal dose of 1 mg., with one or two 
intervening doses. Baldwin has.advised 0.05, 0.2, 0.5, and 1 mg. 
The Reaction.—The constitutional reaction is quite variable. Fever is 
the most delicate indicator. A rise of 1° F. or more above the previous 
maximum is considered positive, especially if it is accompanied by a local 
and a focal reaction. A definite febrile reaction due to tuberculin is rare 
without the presence of a local reaction to the same or the preceding doses. 
General symptoms of headache, muscle pains, anorexia, nausea, etc., may 
accompany the reactions. The local reaction consists of redness and pain at THE INTRACUTANEOUS TUBERCULIN REACTION 
695 
the site of injection, with tenderness of the neighboring lymph-glands, and 
is absolutely specific of tuberculin hypersensitiveness. The focal reaction 
consists of an inflammatory reaction with the production of rales, change 
in breath sounds, etc. 
The Intracutaneous Tuberculin Reaction 
Variety of Tuberculin.—Koch’s old tuberculin is used either in one dose 
of 0.005 mg., or preferably in three different doses injected simultaneously; 
these will be described further on. 
Method of Conducting the Test.—The skin of the forearm is cleansed 
with alcohol and then dried. A small glass syringe fitted with a fine needle 
is used. A separate syringe is used for the control fluid, consisting of sterile 
salt solution, and three others for each of the different dilutions used. In 
performing this test Hamman and Wolman make four simultaneous injec- 
tions : 
First: 0.05 c.c. of sterile normal salt solution (control). 
Second: 0.05 c.c. of a 1 : 1,000,000 dilution of old tuberculin, which equals 
0.00005 mg. This equals a dose of 0.05 c.c. of dilution No. 4, just 
described in the preceding test. 
Third: 0.05 c.c. of a 1 : 100,000 dilution, or 0.05 c.c.-of dilution No. 3, 
equivalent to 0.0005 mg. 
Fourth: 0.05 c.c. of a 1 : 10,000 dilution, or 0.05 c.c. of dilution No. 2, 
equivalent to 0.0005 mg. 
Mantoux uses the last or fourth dose only. The injections are given 
with the skin held taut or pinched up in a fold between the index-finger 
and thumb. The needle is inserted superficially, with the aperture directed 
toward the outer surface of the skin. If the point of the needle is in the skin, 
a white elevation occurs immediately upon the introduction of the solu- 
tion; if it is in the subcutaneous tissue, no infiltration is apparent. Cohen1 
injects T. R., the first doses being one ten-millionth, one millionth, and 
one hundred thousandth of a milligram. If these injections provoke no re- 
actions, he repeats the tests with doses ranging as high as 10 mg. 
The reaction consists of infiltration and hyperemia about the site of 
injection similar to the reaction in the cutaneous test. It appears in from 
six to eight hours, reaches its maximum intensity in from twenty-four to 
forty-eight hours, and usually disappears in from six to ten days. The 
salt solution generally produces a traumatic reaction, similar to a mild 
tuberculin reaction, which subsides in forty-eight hours. 
The reactions are best recorded after twenty-four hours, and the simplest 
method of recording the results is to measure the width of the area of infiltra- 
tion of each reacting point. 
Wildbolz Intracutaneous Urine Test.—Wildbolz2 reports that his re- 
search during the last year and a half and experiences with more than 200 
persons have demonstrated that when there is an active process of tubercu- 
losis the urine contains an antigen which injected by the Mantoux intra- 
dermal technic induces infiltration and redness. According to Wildbolz this 
does not occur with urine from healthy persons or in urine from persons 
with healed tuberculous processes. It is said never to occur unless the per- 
son gives a positive response to injection of 1 : 10,000 tuberculin, but it 
seems to occur whether the urine is from the person being tested or not, so 
1 Jour. Infect. Dis., 1917, xx, 233. 
2 Corresp.-Blatt. f. Schw. Aertze, 1919, 49, 793. 696 
CLINICAL ALLERGY 
long as he has an active tuberculous process anywhere in the body, in 
glands, peritoneum, lung, bones, or elsewhere. Wildbolz evaporates morn- 
ing urine to 1 : 10 passed once or twrice through a paper filter impregnated 
with 2 per cent, phenol, and then makes three sets of two injections on the 
arm, the two upper with 1 : 1000 tuberculin; 3 or 4 cm. below this, two 
with 1 : 10,000 tuberculin, and, the same distance below, two with a minute 
amount of the 1 : 10 evaporated urine. The response with an active tuber- 
culous process is the same with the urine as with the diluted tuberculin, but 
the tuberculin response persists unmodified after the process has healed, 
while the urine response fades out completely. A similar response was never 
obtained in the non-tuberculous, not even in syphilis, influenza, etc., with 
the single exception that urine containing large amounts of staphylococci 
induced a reaction, so that the findings are not pathognomonic in certain 
cases of nephritis. With this exception, this biologic reaction, Wildbolz 
states, may be depended on to reveal the tuberculous or non-tuberculous 
nature of lesions, and will also disclose when they are healed. If the urine 
reaction persists after the clinical healing of the known process we can be 
confident that there is some other active process elsewhere. 
The specific nature of the urine reaction is demonstrated still more 
conclusively by the fact that, after subsidence of the urine reaction, if an 
injection of 1 : 1000 tuberculin is made nearby, the apparently extinct urine 
reaction flares up anew, the infiltration and redness becoming distinct again. 
Eliasberg and Schiff,1 Imhof,2 Schmidt,3 and Gibson and Carrol4 have 
reported favorably upon this test; the latter states that the salt content of 
the urine may modify the reaction and that the urine should be made free 
of salts before being applied. 
The Cutaneous Tuberculin Reaction (von Pirquet) 
Variety of Tuberculin.—Undiluted old tuberculin is now used almost 
exclusively in conducting the test. 
Method of Conducting the Test.—1. The flexor surface of the forearm 
is chiefly used for making the applications, but it should be remembered 
that tests performed on different portions of the body are not strictly com- 
parable. 
2. The skin is cleansed lightly with alcohol and dried. Three abrasions 
are made, about 1| or 2 inches apart, with a von Pirquet skin borer (Fig. 160) 
or with a needle, small lancet, or blood sticker. The object is to scarify the 
superficial layers of the skin, avoiding as much as possible bleeding, although 
a few small points of blood should appear. To the upper and lower abrasions 
add 1 drop of tuberculin; after ten minutes wipe away the excess with a bit 
of cotton. No shield or protective dressings are required. The middle 
abrasion is the control, and shows the amount of traumatic reaction following 
the scarifying process. Due precautions should, of course, be observed that 
none of the tuberculin flows down the arm and reaches this spot. 
3. The tests are inspected at the end of twenty-four hours. 
The Reaction.—The traumatic reaction as shown in the control area 
may present an inflammatory areola with, at times, slight infiltration. 
Before a test may be considered positive its areola should be at least 5 mm. 
wider than the control area. The reactions are usually designated as follows l 
1 Monatschr f. Kinderh., 1920, 19, 5. 
2 Schweiz, med. Wchn., 1920, 50, 1033. 
3 Schweiz, med. Wchn., 1921, 51, 996. 
4 Jour. Amer. Med. Assoc., 1921, 76, 1381. Fig. 161.—A Positive Cutaneous Tuberculin Reaction (von Pirquet). 
Child with incipient pulmonary tuberculosis; a + reaction. The control scarification is barely to be 
seen, and is midway between the tuberculin reactions. THE CUTANEOUS TUBERCULIN REACTION 
697 
1. Negative Reaction.—No appreciable difference between the tuber- 
culin areas and the control. 
2. Slight Reaction.—Definite but slight redness with some infiltration. 
3. + Reaction.—A wider area of redness, with definitely raised centers. 
4. ++ Reaction.—Wider area of redness, with more marked infiltration 
than + (Fig. 161). 
5. + + + Reaction.—Unusual redness and a wide area of infiltration, all 
cases which go on to vesiculation. 
The abrasion is being made over the insertion of the deltoid muscle. The borer is held firmly and 
perpendicular to the arm. A quick rotatory motion serves to remove a circular area of epidermis. 
The borer is shown in the upper right-hand corner. 
Fig. 160.—Method of Performing a von Pirquet Tuberculin Test. 
The usual or normal reaction begins to appear in from four to six hours, 
reaches its maximum intensity in from twenty-four to forty-eight hours, 
and then fades rapidly, although the infiltration may persist for some days. 
Special types of the reaction have been described as follows: 
1. The premature reaction, characterized by a rapid course and slight 
intensity. This type is supposed to occur in patients with manifest tuber- 
culosis who are not doing well. 
2. The persisting reaction, which reaches its maximum intensity about 
the second day and persists for a week or longer. 698 
CLINICAL ALLERGY 
3. The late reaction, which appears after twenty-four hours and de- 
velops and recedes slowly. These last two types are believed to occur in 
patients having inactive lesions. 
4. The cachectic reaction, which is characterized by infiltration with 
little or no redness. This type is common in the late stages of tuberculosis. 
5. The scrofulous reaction, which is characterized by numerous small 
elevated nodules, which may also appear on the extremities and trunk. 
This reaction is peculiar to children and rare in adults. 
The Conjunctival Tuberculin Reaction (Calmette) 
Variety of Tuberculin Used.—A 1 per cent, solution of Koch’s old tuber- 
culin is now generally used, as it is quite reliable, least expensive, and the 
results that follow its use are more regular than those obtained with purified 
tuberculin (0.5 to 2 per cent, aqueous solution). 
Method of Conducting the Test.—The conjunctive of both eyes are 
inspected to ascertain if there is any evidence of disease and if they are 
strictly comparable in color. The lower lid of one eye is drawn forward to 
form a little pouch, and the patient is directed to look upward; 1 drop of a 
1 per cent, dilution of old tuberculin is then applied from an ordinary eye- 
dropper to the lid at the inner canthus. Profuse lacrimation impairs the 
test, and it is useless to attempt it with weeping or resisting children. If 
no reaction is apparent and it is still desired to further the test, 1 drop of a 
5 per cent, dilution may be placed in the opposite eye. 
The Reaction.—In a positive reaction the conjunctiva begins to redden 
in from six to eight hours, and reaches its maximum in from twenty-four to 
forty-eight hours, and then rapidly subsides and disappears in from four to 
six days. In mild reactions the inner canthus is the seat of the most marked 
changes. Positive reactions have been classified as follows: 
+ Reaction: Definite palpebral redness. 
-j—(- Reaction: More marked palpebral redness with secretion. 
+ -j- + Reaction: Palpebral and bulbar redness with subjective symp- 
toms and well-marked secretion. (See Fig. 162). 
Precautions.—On account of severe reactions and the danger of in- 
flicting permanent injury on the cornea there is a growing tendency to 
regard this reaction with disfavor. Hamman and Wolman, however, 
believe that, with proper precautions, these risks may be minimized, if not 
completely avoided; they believe, moreover, that the risk incurred is not great 
enough to warrant the abandonment of a procedure that has proved itself 
of such great value in diagnosis. Nevertheless, they emphasize the neces- 
sity for observing proper precautions, and give the following rules to be 
adopted: 
1. The conjunctival test should never be repeated in the same eye. 
2. A solution stronger than 1 per cent, original tuberculin should never 
be used for making the first instillation. 
3. Any existing inflammatory disease of the eye is an absolute contra- 
indication to the test. 
4. The test should not be given to manifestly scrofulous children. 
5. Skin diseases in which the lesions are situated upon the face, near 
the eye, especially when these are suspected of being tuberculous, preclude 
the application of the test. 
6. It is safest not to give the test to elderly persons, and particularly to 
arteriosclerotics, as they are unduly prone to develop corneal ulceration. Fig. 162. A Positive Conjunctival Tuberculin Reaction (Wolff-Eisnf.r-Calmette). 
Severe reaction ( + + +) in a tuberculous cow; appearance of the eye about fourteen hours after 
second instillation of tuberculin. Fig. 163.—A Positive Percutaneous Tuberculin Reaction (Moro). 
E. McK., adult male with arrested early pulmonary tuberculosis. Lesions about sixty hours after 
application of tuberculin ointment. STREPTOTRICHIN REACTION 
699 
Variety of Tuberculin Used.—This consists of 5 c.c. of old tuberculin 
and 5 grams of anhydrous lanolin thoroughly mixed. The ointment retains 
its potency for many months when preserved in a cold dark place. Manu- 
facturers market the preparation in small collapsible tubes containing 
sufficient for at least one or two tests. 
Method of Conducting the Test.—A piece of ointment about the size 
of a pea is thoroughly rubbed for at least a minute into an area of skin about 
2 inches in diameter over the upper abdomen or near a nipple. The patient 
may be instructed how to apply this ointment, or the physician may do it 
himself, protecting the finger with a rubber cot. 
The Reaction.—This may appear within twenty-four hours, or be de- 
layed for as long as from four to six days. It consists of an eruption of 
slightly elevated papules situated upon a hyperemic base, which vary in 
size from a pinhead to large areas of infiltration (Fig. 163). It subsides in 
from three to ten days. Moro described these grades of reaction as follows: 
1. Mild reactions: From 1 to 10 scattered papules. 
2. Moderate reactions: About 50 or more papules, partly discrete, partly 
confluent, and with a general reddening of the skin. There may be well- 
marked itching. 
3. Severe reactions: Numerous large, extremely red papules or vesicles 
up to 8 mm. in diameter, having an intensely reddened base. 
The cutaneous test of Ligniere is conducted by washing the skin with 
alcohol to produce redness; 6 drops of Koch’s old tuberculin is then applied 
and rubbed into the skin by the protected fingers. The reaction appears in 
twenty-four to forty-eight hours and is quite similar in appearance and 
interpretation to the Moro reaction. 
The Percutaneous Tuberculin Reaction (Moro) 
Among the diseases in man and animals that are produced by the strep- 
totrices may be mentioned madura foot, actinomycosis, nocardiosis, and 
pseudotuberculosis. These diseases are, as a rule, characterized as slow, 
progressive, localized processes which are, however, capable of an acute 
general pyemic or pneumonic course. Claypole1 has shown that they run 
from mycelial, non-acid-fast forms through non-acid-fast or partially acid- 
fast bacillary forms to strongly acid-fast forms that are closely allied to the 
tubercle bacillus. Not only are these gradual transitions in type demon- 
strable from one species to another, but they may actually occur in the 
cultivation of a given species. A given organism may from time to time be 
more or less acid-fast and bacillary, or again at one time almost wholly 
mycelial and later bacillary. 
In a considerable number of human cases of cervical adenitis and pul- 
monary disease that clinically are indistinguishable from tuberculosis some 
form of streptothrix has been found in pure culture by Flexner,2 Lubarsch,3 
Ophuls,4 Burnet,6 and Foulerton.6 That a distinct though unknown per- 
centage of glandular, pulmonary, and bone tuberculosis, so-called, is due to 
infection with a streptothrix is becoming evident from the work of Bridge7 
and Claypole. Even with the examination of discharges and sputum the 
diagnosis is overlooked. In the presence of a non-acid-fast streptothrix the 
Streptotrichin Reaction 
1 Jour. Exper. Med., 1913, 17, 99. 
2 Jour. Exper. Med., 1898, 3, 435. 
3 Ztschr. f. Hyg., 1899, 31, 185. 
4 Jour. Med. Research, 1902, 8, 242. 
6 Compt. rend. Soc. de biol., 1913, lxxiv, 674. 
6 Lancet, 1899, 2, 779; ibid., 1913, 1, 381. 
7 Jour. Amer. Med. Assoc., 1911, lvii, 1501. 700 
CLINICAL ALLERGY 
carbolfuchsin stain results in the report of “no tubercle bacilli found.” On 
the other hand, an acid-fast streptothrix, particularly if non-mycelial, is 
passed as Bacillus tuberculosis. 
Claypole1 prepared solutions from two types of streptothrix, in all re- 
spects similar to old tuberculin (concentrated glycerin-bouillon growths). 
Her cases were tested by the skin test of von Pirquet with old tuberculin,, 
with Streptothricin H. (Streptotricin Hominis, mycelial, non-acid-fast)r 
and with Streptothricin E. (S. eppingeri bacillar, partly acid fast). 
In 45 control cases which gave no clinical evidence of tuberculosis, 22 
reacted to old tuberculin, but none to either of the streptotrichins. In 55 
cases of suspected tuberculosis, comprising 42 lung, 11 gland, and 5 bone 
involvements, 37 reacted to tuberculin (67 per cent.); 13 of these cases re- 
acted to Streptothricin H.; 8 to Streptothricin E., and 6 gave reactions to 
two of the test solutions. 
Of the 13 Streptothricin H. reactions, 11 were lung cases. In 9 of these 
no tubercle bacilli were found, but in all of them thread-like, Gram-positive 
organisms resembling Streptothrix H., in the 2 others acid-fast bacilli were 
found as well as the mycelial organisms, and one reacted to tuberculin and 
one to Streptothricin E. (acid-fast), as well as to Streptothricin H. The 
other two positive Streptothricin H. cases were glandular; both were nega- 
tive to tuberculin. In one of these cases Gram-positive segments and rods 
were found histologically. 
Eight cases reacted to streptothricin E., 6 with lung involvements and 2 
with bone lesions; 3 of the lung cases were negative to tuberculin and had 
no tubercle bacilli in the sputum, without doubt infections with a less acid- 
fast organism than Bacillus tuberculosis. Two of the 3 remaining cases 
reacted to tuberculin and to streptothricin E. In one of these tubercle 
bacilli were found in the sputum. The sixth case reacted to both strepto- 
tricins and had a mycelial Gram organism and an acid-fast organism in the 
sputum. The bone cases did not react to tuberculin and the infections were 
by inference due to the Streptothrix eppingeri group of organisms. 
The data from these cases indicates that the streptothricin reaction may 
be a means of differential diagnosis between mycelial streptothrix infectionf 
partly acid-fast streptothrix infections, and tuberculosis. 
Tuberculin Reactions Among the Lower Animals 
In veterinary practice tuberculin is used almost solely for diagnostic 
and only occasionally for therapeutic purposes. 
Four reactions are in common use: 
The Subcutaneous Test. 
The Conjunctival Test. 
The Cutaneous Test. 
The Intracutaneous Test. 
The technic for conducting these tests and the reactions secured are 
quite similar to those just described, and I would refer the veterinary surgeon 
to these respective descriptions. Differences in technic are confined prin- 
cipally to dosage. 
The Subcutaneous Tuberculin Test.—This is conducted with Koch’s 
old tuberculin. The method of preparation is given in the succeeding 
chapter, under Tuberculin Therapy. Concentrated tuberculin is prepared 
by concentrating the bouillon filtrate to one-tenth its original volume. 
Diluted tuberculin is the concentrated product diluted to its original volume 
1 Archiv. Int. Med., 1914, 14, 104. TUBERCULIN REACTIONS AMONG THE LOWER ANIMALS 
701 
by mixing 1 part of the dilution with 9 parts of 0.5 to 1 per cent, phenol in 
sterile normal salt solution or distilled water. 
The injections are always given subcutaneously in some convenient 
area, preferably around the shoulder, which has been shaved and cleaned 
beforehand with a solution of creolin. 
Cary1 has recently described the following method: 
“1. Secure the cattle in separate stalls or by chains, ropes or stanchions, 
so as to confine them in fixed places, and then number, tag, or mark them for 
temporary or permanent identification. 
“2. Take preinjection temperatures two hours apart, at least three times. 
Better take preinjection temperatures once every two hours, from 6 a. m. 
to 6 or 10 p. m. 
“3. In all normal animals inject 3 to 10 c.c. of standard subcutaneous 
tuberculin at 6 or 10 p. m. 
“4. Begin taking postinjection temperatures not later than eight hours 
after time of injection. For animals that have been repeatedly tested or 
have previously been given large doses of tuberculin it is better to begin 
taking temperatures not later than four hours after injection (for early 
reactors continue the regular two-hour temperatures until the eighteenth 
or twentieth hour after injection. When no regular reaction appears and 
the temperature rises to 104° F. or higher at the sixteenth or eighteenth hour 
postinjection, continue taking temperatures up to the twenty-fourth or 
twenty-eighth hour (for late reactors). 
“5. Give the same care, feed, and water both days, and avoid sudden 
changes of any kind. At no time give heavy feeds, large quantities of 
water, or produce excitement or exercise. 
“6. Interpret the clearly, carefully, and accurately recorded records as 
follows: 
“(a) A regular or typical reactor is one where the postinjection tem- 
peratures make a more or less definite curved rise of temperature for three 
or more readings, and the maximum of the curve is two or more degrees 
higher than the maximum preinjection temperature. 
“Example: Preinjection, (1), 101, (2) 101.4, (3) 101.8. Postinjection, 
(1) 102, (2) 103, (3) 104.2, (4) 105, (5) 104.1, (6) 103. 
“(b) When the temperatures are high and level or nearly equal for 3 
to 7 postinjection records. This may be regarded as a reactor. 
“Example: Preinjection, (1) 101.4, (2) 101.8, (3) 102. Post 1, (1) 104.4, 
(2) 104.8, (3) 104.8, (4) 104.4, (5) 104.6. Post 2, (1) 105.1, (2) 105, (3) 
105.4, (4) 105.2, (5) 106. 
“(c) An early reactor will begin rise in temperature in two to four hours 
after injection and will not always show a normal or big curve. 
“Example: (1) Postinjection, second hour, 104.5; fourth hour, 105; 
sixth hour, 105; eighth hour, 104; tenth hour, 102. (2) Postinjection, sec- 
ond hour, 104; fourth hour, 104.2; sixth hour, 104.1; eighth hour, 103.4; 
tenth hour, 101. 
“(d) A late reactor may start up at the fifth, sixth, or seventh reading 
(sixteenth, eighteenth, or twentieth hour postinjection). 
“Examples: Post-injection 1, (1) 101.2, (2) 101.4, (3) 101, (4) 102 (5) 
101.8, (6) 104, (7) 105, (8) 105.2, (9) 104.8, (10) 103. Postinjection 2, (1) 
101.4, (2) 101.6, (3) 102, (4) 101.8, (5) 102, (6) 103.5, (7) 104, (8) 105.1, (9) 
104.2, (10) 102.8. 
“(e) When there are up and down temperatures, irregular curves, the 
test may be more or less questionable and the animal is marked suspicious, 
1 Jour. Amer. Vet. Med. Assoc., 1921, lix, 746. 702 
CLINICAL ALLERGY 
unless cold drinks of water or sudden exercise or excitement may account for 
the sudden fluctuations of temperatures, and then it is usually best to mark 
them as suspicious. 
“Example: Preinjection, (1) 102, (2) 101.2, (3) 102.2. Postinjection, 
(1) 102, (2) 104, (3) 101.2, (4) 104.2, (5) 103, (6) 102, (7) 103. 
“(/) When the curve is low and does not have a maximum of two degrees 
above the maximum preinjection temperature, then make the animal a 
suspect, and give it an ophthalmic test or wait ninety or more days and re- 
test with intradermal and ophthalmic. 
“Example: Preinjection, (1) 101.6, (2) 102, (3) 101.4, Postinjection, 
(1) 101.5, (2) 101.6, (3) 102.8, (4) 103.2, (5) 103.6, (6) 102.2. 
“(g) Always note the physical changes in the reacting animals. Some 
have a brief chill, the hair stands up and the coat is dry and rough, and some 
get very thirsty and refuse feed. Some may have a cough, a discharge from 
the nose, and enlarged lymph glands. Some have tubercles in the udder. 
The local changes at point of injection must be observed. 
“The quantity of standard tuberculin to inject varies with the size, age, 
and number of the previous injections. If the cattle have never been in- 
jected, 4 to 6 c.c. is sufficient for the large cattle and 2 to 4 c.c. for the small 
and young cattle. There is no doubt about a tuberculous animal becoming 
tolerant or plugged by frequent small or large doses. A very large dose (10 
to 15 c.c.) of tuberculin will often bring forth a reaction in a plugged or toler- 
ant animal, but it is not always certain or positive. As a rule tolerant ani- 
mals react early and some react exceptionally late. In order to stop owmers 
from plugging animals it is important that all tuberculin production and 
sales shall be licensed and controlled by Federal and State authorities. 
“Temperatures taken before and after injection are very important. 
The Federal and State regulations specify at least three preinjection and 
at least six postinjection temperatures two hours apart running to the 
eighteenth hour postinjection and beginning not later than the eighth hour 
postinjection. Where every animal is normal and the course is typical 
this method will catch many or nearly all of the tuberculous animals, but it 
will not catch the early and late reactors so often found in herds that have 
been tested several times by the subcutaneous method. 
“To my mind we should go back to the old-time original method of 
taking temperatures. Begin at 4 or 6 a. m. and take preinjection tempera- 
tures every two hours until 6 or 10 p. m. Inject at 6 or 10 p. m., and begin 
at 8 p. m. or 12 midnight and take temperatures every two hours to the 
twentieth or twenty-eighth hour after injection. When this method is care- 
fully and accurately done, there will be very few mistakes and very few 
tuberculous animals missed in the test. It may take more time and atten- 
tion, but it pays in accuracy. The morning preinjection temperatures can 
be compared with the morning postinjection temperatures. The same is 
true with afternoon temperatures. Hence the common daily variations give 
opportunity for a more accurate comparison with same hours each day. 
“The good points about the subcutaneous method are: 
“1. The numerous data from which to draw conclusions, numerous 
temperatures, conditions of animals, etc. 
“2. In non-plugged animals it is very accurate. 
“3. It keeps the veterinarian on his job and gives him a chance to note 
and see clearly any physical reactions that appear. 
“4. It is easily carried on in combination with the ophthalmic method. 
“The objectionable points are: 
“1. It is more exacting in work and often takes day and night work. TUBERCULIN REACTIONS AMONG THE LOWER ANIMALS 
703 
“2. It should cost more on account of more work for the veterinarian and 
the use of more tuberculin. 
“3. It takes about as much time to test one animal as it does for ten. 
“4. It is more or less unreliable in plugged animals or animals that have 
been repeatedly tested by this method. 
“5. It cannot be applied to young calves or to wild cattle.” 
The Conjunctival Tuberculin Test.—For this test Koch’s concentrated 
tuberculin may be used. Preference is usually given to the purified product, 
prepared as follows: Mix 1 part of Koch’s concentrated tuberculin with 
20 parts of absolute alcohol. The precipitate that forms is filtered off and 
dried over sulphuric acid. This powder is then made up into a 4 and 8 per 
cent, solution in sterile distilled water. 
Two or 3 drops of the 4 per cent, dilution are placed in the inner canthus 
of one eye to sensitize the tissues. After twenty-four hours, unless a posi- 
tive reaction is present, 2 or 3 drops of the 8 per cent, solution are instilled 
in the same eye in the same manner. The reaction is usually apparent in 
from six to twelve hours (Fig. 162). 
For the types of reaction and precautions to be observed see the previous 
description. 
One advantage of this test is that the animal will give a reaction in cases 
where, prior to the test, dishonest dealers have injected tuberculin. 
DeFosset1 has recently described the following method: 
“Cases cannot be detected by the ophthalmic method unless the animals 
are properly prepared for treatment, which should be done by instilling 
ophthalmic tuberculin in one eye about seventy-two hours before giving 
the regular test dosage. This preparatory process is known as sensitization. 
For making the ophthalmic test I would recommend the special disk prepared 
by the United States Bureau of Animal Industry. There are tablets on the 
market prepared by the commercial firms which also are satisfactory. 
“Technic.—It is most convenient for a right-handed man to treat the left 
eye of an animal, and we have for this reason chosen the left eye for the test, 
and for uniformity in work it is necessary that a certain system be followed. 
An attendant should hold the head of the animal, which is not difficult to 
do and needs no special restraint, while the operator stands in front of the 
subject and raises the upper lid of the left eye by grasping it gently with the 
left forefinger and thumb. The tablet, which is held between the right 
forefinger and thumb, should be pushed up under the external lateral lid 
into the conjunctival fornix. After a little experience a deft and skilful 
operator can do this without injury or pain to the animal. When the 
tablet has been placed in position the fingers on the left hand should be 
released from the upper lid and brought down over both lids, holding them 
together firmly, yet gently, so as not to injure the eye until the tablet has 
dissolved in the closed conjunctival sac which has been forced by bringing 
the lids in apposition with each other. Great care must be taken that the 
tablet is not worked out of the eye before it has had time to dissolve. 
“The tablet should not be placed in the inner angle of the eye because 
the proper tissue structure cannot be treated there, also the membrana 
nictitans, whose function it is to remove foreign bodies, would remove much 
of the tuberculin before proper treatment has been effected. The tuberculin 
should be placed in the upper fornix of the lateral angle because that por- 
tion of the conjunctiva is covered with stratified cylindric epithelium 
containing goblet cells and tubular glands. These cells and glands are 
capable of secreting mucin when reacting to the toxins in the tuberculin. 
1 Jour. Amer. Vet. Med. Assoc., 1921, lix, 750. 704 
CLINICAL ALLERGY 
“Reaction.—The criterion of a reaction is a mucopurulent exudate in 
more or less profusion (Fig. 162). We cannot expect a reaction unless we 
apply the treatment to tissue capable of reacting. A reaction will usually 
be preceded by slight redness and lacrimation. A reaction usually does not 
follow the application of the first tablet, which is employed as the sensitizing 
agent. The best results follow the application of the second tablet, which 
should be given about seventy-two hours later. It is not wrell to hasten 
the time of sensitization, because the reaction may in some cases be very 
slight and tend to cause confusion. The best results may be obtained in 
from three to five days’ sensitization. The ophthalmic test should never 
be concluded by the application of only one treatment. Very careful 
sensitization is absolutely essential if one would expect satisfactory results. 
If either the sensitization or the subsequent treatment is carelessly or im- 
properly applied one may note a faint mucous discharge at the inner canthus, 
which may appear as a little wavy thread, and it is difficult to make an accu- 
rate diagnosis and differentitate between the mucopurulent exudate of a 
tuberculous reaction and a slight mucous discharge commonly seen in normal 
eyes following the introduction of some foreign substance. A man who 
would make a successful ophthalmic tuberculin test must be a student of 
the test and very observing of what takes place after treatment has been 
placed in the eye. 
“Observation should be made if possible in about ten hours after the first 
tablet has been introduced, and a note should be made of any changes 
observed and any deviation from the normal or control right eye, using the 
proper characters for future reference. After the introduction of the subse- 
quent treatment, observations should be made at hourly intervals or often- 
er, beginning not later than three hours after the instillation of the tablet. 
Some reactions take place within two hours after the second treatment has 
been applied. If the treatment has been properly applied, the reactions 
will be clear and well defined, manifesting themselves first by lacrimation, 
slight redness of conjunctiva, photophobia, and a free discharge of muco- 
purulent exudate. 
“If there is a tendency of the reaction being somewhat retarded or ill 
defined, an additional tablet may be given five hours after the second tablet, 
and this ofttimes tends to give conclusive results if the animal is tuberculous. 
If there still remains some doubt as to a positive reaction, the animal may 
be retested within one week by using two tablets without previously sen- 
sitizing. Most eyes remain sensitized a week or longer after the subsequent 
treatment. Very young animals or calves very frequently will give a 
reaction to one treatment of ophthalmic tuberculin. 
“I recommend the tablet form of tuberculin because of the ease with 
which it can be placed in the proper tissue structures. Only fresh disks 
which dissolve readily should be employed. 
“Recording Reactions.—It is necessary that we adopt the use of some 
character so that the degree of reaction can be noted as one would note the 
degree in a thermic reaction. Inasmuch as certain changes take place in 
the treated eye during reaction, a note for future reference should be made 
of these changes. These notes should be in such form that they may be 
recorded on a chart as one would the numerals in thermic reaction. The 
notations should be made in the order in which the change becomes manifest. 
For instance, if lacrimation is observed two hours after treatment, the letter 
L should be placed in that column. If exudation follows next, then the 
next column should show LX, and so on. The degree of exudation may be 
noted from X, meaning slight, to XX, well marked, and XXX, very exten- LUETIN REACTION 
705 
sive or copious exudation. When only a trace of exudation is observed, it 
may be noted by writing f or | X. A positive reaction may be LX or LXXX 
of varying degree, depending also upon the character of exudation. Hy- 
peremia or redness may be noted by H. 
“Precautions.—1. Don’t treat abnormal eyes or a herd when sore eyes 
are prevalent. 
“2. Don’t try to make an ophthalmic test in a barn or corral where the 
wind is sweeping through and blowing dust, cinders, or other foreign matter. 
“3. Don’t manipulate the eye of the subject unnecessarily and cause 
irritation from natural results. 
“4. Don’t try to put tablets into the eyes of animals before you pare 
your finger-nails carefully and cleanse the hands. 
“5. Don’t use irritating chemicals on your hands. 
“6. Don’t try to see something in the eye that isn’t present. 
“7. Don’t let the owner feel that you have no confidence in your work.” 
The Cutaneous Tuberculin Test.—In making this test some convenient 
area is shaved and scraped slightly until serum exudes. A small amount 
of Koch’s old tuberculin is applied to the prepared area. In a positive case 
a well-marked area of congestion and hyperemia appear at the end of twenty- 
four hours. This may also be accompanied by a rise in temperature. 
The Intracutaneous Tuberculin Test.—In performing this test from 
0.2 to 0.4 c.c. of Koch’s concentrated tuberculin are injected into the skin 
through a fine needle. The skin at the root of the tail is generally employed. 
A white swelling should appear while the injection is being given; if it does 
not appear, the injection is subcutaneous and unsatisfactory for this test. 
The appearance of hyperemia and redness with a rise in temperature indi- 
cates a positive reaction. 
LUETIN REACTION 
Stimulated by von Pirquet’s discovery of a specific cutaneous reaction 
for tuberculosis, a number of investigators (Finger and Landsteiner, Wolff- 
Eisner, Nobe, Ciuffo, Nicolas, Favre, and Gauthier) attempted to obtain a 
specific reaction for syphilis by applying extracts of syphilitic tissues— 
prepared from syphilitic fetal liver or chancre—to the skin of syphilitic 
patients. In spite of some encouraging effects tjieir results were, on the 
whole, contradictory. Further, Neisser and Bruck found that a reaction 
similar to that produced with syphilitic extract can be obtained also with a 
concentrated extract of normal liver. This peculiarity of the skin of syphili- 
tics is ascribed by Neisser to what he calls the state of “ Umstimmung” in the 
later stages of syphilis. Both Neisser and von Pirquet expressed the hope 
and belief that a reaction may be secured by employing an extract of pallida 
free from tissue constituents; this was apparently accomplished by Noguchi,1 
in 1911, first with syphilitic rabbits and then with human patients. Noguchi 
gave the appropriate name of “luetin” to the extract of pallida. Theoretic- 
ally, one should not expect to obtain an allergic reaction in syphilis so long 
as the activity of pallida is maintained at its maximum, or in the very early 
stages, before there is sufficient time for antibody formation. One can 
reasonably expect the appearance of the phenomenon when the activity of 
the microparasite begins to abate through a gradually acquired defensive 
power of the host, or under an effective therapeusis, as in the later stages 
of the disease and in hereditary syphilis. 
Preparation of Luetin.—At least six different strains of pallida in pure culture are 
being used by Noguchi in the preparation of luetin. These are cultivated in ascites-kidney 
1 Jour. Exper. Med., 1911, 14, 557; Jour. Amer. Med. Assoc., lviii, 1163. 706 
CLINICAL ALLERGY 
agar for periods of six, twelve, twenty-four, and fifty days, at 37° C., under anaerobic con- 
ditions. The tubes showing large numbers of spirochetes are then selected, the oil is poured 
off, the tube is cut, the agar column is removed, and the tissue then cut off. The ascites- 
agar cultures are then carefully ground in a sterile mortar, the resulting thick paste being 
gradually diluted with a fluid culture until a homogeneous liquid emulsion is secured. The 
preparation is next heated for an hour in a water-bath at 60° C., and then tricresol or phenol 
added to make 0.5 per cent. Cultures are made from this suspension, and rabbits inoculated 
intratesticularly; both, after suitable intervals, must show an absolutely sterile preparation. 
The luetin should be kept in the refrigerator when not in use. The 
isolation of Treponema pallidum in pure culture is a difficult procedure, and, 
obviously, a luetin must be prepared of pure cultures and from as many 
different strains as possible. While pallida quickly loses its pathogenicity 
in artificial culture-media and is also highly susceptible to the influence of 
germicides, its preparation, nevertheless, is an important matter requiring 
skilful supervision. 
A control fluid prepared of sterile agar and bouillon in exactly the same 
manner as luetin was originally advised by Noguchi, but recently he claims 
that its use is not necessary. 
Method of Application.—Luetin is not applied to an abrasion of the skin, 
but is injected iniracutaneously with a very fine needle and a sterile syringe 
(Fig. 156). According to original directions the luetin is to be well shaken, 
diluted with an equal part of sterile salt solution, and 0.07 c.c. injected 
(0.035 c.c. undiluted). A slightly smaller dose, as, e. g., 0.05 c.c., maybe 
used for children. The skin of the upper arm is usually selected as the site 
for inoculation. If a control fluid is used, the luetin may be injected into 
the skin of the left and the control fluid into the skin of the right arm, or 
both injections may be given in the same arm, about 2 inches apart, the 
control being above the luetin. As shown by Sherrick,1 and confirmed by 
my associates and myself,2 normal and non-syphilitic persons taking iodids 
or bromids at the time the skin test is made or while these drugs are still 
in the body fluids may yield well-marked non-specific reactions which the 
physician may readily interpret as a positive luetin reaction. 
After cleansing the skin with alcohol it is drawn taut or pinched up 
between forefinger and thumb and the needle introduced, with the aperture 
directed toward the outer surface of the skin. If the point of the needle is 
in the skin, a white elevation occurs immediately upon injecting the solution; 
if it is in the subcutaneous tissue, no filtration is apparent. 
Reactions.—Normal or Negative Reactions.—In the majority of normal 
persons the injection of luetin is followed by a very slight traumatic reaction, 
or a small erythematous area appears, after twenty-four hours, at and around 
the point of injection. No pain or itching sensation is experienced; the 
reaction recedes in forty-eight hours and leaves no induration. In certain 
individuals the reaction may reach a stage of small papule formation after 
from twenty-four to forty-eight hours, which subsides within seventy-two 
hours, leaving no induration. 
Positive reactions have been classified by Noguchi into three main varie- 
ties: 
(a) Papular Form.—“A large, raised, reddish, indurated papule, usually 
5 to 10 millimeters in diameter, makes its appearance in twenty-four to 
forty-eight hours. The papule may be surrounded by a diffuse zone of 
redness and show marked telangiectasis. The dimensions and the degree of 
induration slowly increase during the following three or four days, after 
1 Jour. Amer. Med. Assoc., 1915, July 31, 404. 
2 Jour. Amer. Med. Assoc., 1916, lxvii, 718; Jour. Lab. and Clin. Med., 1917, xi, 401. Fig. 164.—A Positive Luetin Reaction 
E. C., adult male; tertiary syphilis with detachment of the retina; papular lesion of moderate severity; 
about forty-eight hours after injection of 0.07 c.c. luetin. LUETIN REACTION 
707 
which the inflammatory processes begin to recede. The color of the papule 
gradually becomes dark bluish red. The induration disappears within one 
week, except in certain instances in which a trace of the reaction may persist 
for a longer period. The latter effect is usually met with among cases of 
secondary syphilis under regular mercurial treatment in which there are no 
manifest lesions at the time of making the skin test. Cases of congenital 
syphilis also show this reaction (Fig. 164). 
“(b) Pustular Form.—The beginning and course of this reaction resemble 
the papular form until about the fourth or fifth day, when the inflammatory 
processes commence to progress. The surface of the indurated round papule 
becomes mildly edematous, and multiple miliary vesicles occasionally form. 
At the same time a beginning central softening of the papule can be seen. 
Within the next twenty-four hours the papule changes into a vesicle, filled 
at first with a semi-opaque serum that later becomes definitely purulent. 
Soon after this the pustule ruptures spontaneously or after slight friction or 
pressure. The margin of the broken pustule remains indurated, while the 
defect caused by the escape of the pustular content becomes quickly covered 
by a crust that falls off within a few days. About this time the induration 
usually disappears, leaving almost no scar after healing. There is a wide 
range of variation in the degree of intensity of the reaction described in 
different cases, as some show rather small pustules, while in others the 
pustule is much larger. This reaction was found almost constantly in cases 
of tertiary syphilis, as well as in cases of secondary or hereditary syphilis 
which had been treated with salvarsan. 
“(c) Torpid Form.—In rare instances the injection sites fade away 
almost to invisible points within three or four days, so that they may be 
passed over as negative reactions. But sometimes these spots suddenly 
light up again after ten days or even longer and progress to small pustular 
formation. The course of this pustule is similar to that described for the 
preceding form. 
“This form of reaction has been observed in a case of primary syphilis, 
in one of hereditary syphilis, and in two cases of secondary syphilis, all being 
under mercurial treatment.” 
Aside from these three types of reactions, there have since been described 
several cases of the formation of a hemorrhagic exudate, the lesion usually 
rupturing spontaneously and not running a more severe or a longer course 
than the pustular. Two such reactions have been reported by Kilgore.1 
“Neither in syphilities nor in parasyphilitics did a marked constitutional 
effect follow the intradermic inoculation of luetin. In most positive cases 
a slight rise in temperature took place, lasting for one day. In three tertiary 
cases and in one hereditary case, however, general malaise, loss of appetite, 
and diarrhea were noted.” 
In view of the occasional instances in which the reactions are retarded 
a patient should be observed for two weeks before a reaction is regarded as 
negative. 
Results.—1. The reports of Noguchi and of a number of different ob- 
servers show that the luetin reaction is generally negative in the primary and 
secondary (untreated) stages of syphilis. 
2. In latent and tertiary syphilis Noguchi has reported positive reactions 
in from 80 to 95 per cent, respectively, and the reports of others have showed 
from 64 to 100 per cent, of positive reactions. 
3. In cerebrospinal syphilis positive reactions have been reported in 
from 42 to 80 per cent, of cases. 
1 Jour. Amer. Med. Assoc., lxii, 1236. 708 
CLINICAL ALLERGY 
4. In congenital syphilis the results have varied within wide limits— 
10 to 96 per cent, of positive reactions. In cases under one year of age 
Noguchi has reported about 23 per cent., and among later cases 96 per cent., 
of positive reactions. 
5. Second injections of luetin apparently do not give positive reactions 
in non-syphilitic cases. 
Non-specific Reactions and Practical Value.—A large literature has ac- 
cumulated on the use of the luetin reaction as a means for the diagnosis of 
syphilis, but not a few investigators have reported non-specific or doubtful 
reactions with the comment that control fluids of the culture-medium alone 
yielded reactions of the same or almost the same kind and degree. 
The most thorough investigation bearing upon the non-specific phase 
of the luetin reaction has been that of Stokes.1 This investigator has 
observed that the intracutaneous injection of 0.5 per cent, solutions of 
sterile agar and emulsions of normal skin elicit reactions in syphilis quite 
similar to those produced by luetin. Stokes has explained these reactions 
on the basis that agar and other substances are capable of absorbing antifer- 
ments and thereby liberate or render active proteolytic ferments capable 
of producing toxic or irritating substances by processes of proteolysis and 
autolysis of the individual’s serum and cells responsible for the inflammatory 
reactions of papulation and pustulation. Stokes has very clearly described 
this mechanism in the following words: “While the luetin reaction seems 
several steps removed from the plainer case of the agar reaction, a close 
examination of experimental evidence bearing on it gives ground for placing 
it, in large part at least, in the class of antiferment-absorbent or anaphyla- 
toxin reactions, and treating it as in the case of the reaction to agar, not as 
specific for syphilis, but as a measure, albeit, perhaps a sensitive one, of 
the ferment-antiferment balance, and of the amount or intensity of action 
of non-specific proteases in the body of the syphilitic.” 
Stokes believes that with luetin the processes are of a dual nature, that 
is, that antiferment is absorbed not only by the ascites agar, but likewise by 
the fragments of Spirocheta pallida in which case the injection of luetin may 
be expected to yield somewhat stronger reactions than the control fluid 
of culture-medium (ascites fluid and agar) alone, but that both processes 
are essentially non-specific. 
In my opinion studies of this kind interpreted in conjunction with the 
fundamental observations of Jobling, Eggstein, and Petersen upon ferments, 
antiferments, and their balance, offer satisfactory explanations for the 
production of reactions following the injection of various substances (agar, 
blood, kaolin, gelatin, tissue, and bacterial extracts, etc.) into the skins of 
healthy and diseased individuals; they do not, however, explain specific 
allergic skin reactions and notably the intracutaneous tuberculin reaction, 
which may be elicited by the injection of 0.1 c.c. of a 1 : 10,000 dilution of 
O. T. carrying an almost infinitesimal amount of glycerin-bouillon incapable 
of eliciting these non-specific reactions. In other words, the physical theory 
of anaphylatoxin production enlisted for explaining the agar reaction is not 
acceptable in explanation of specific allergic reactions which is very probably 
an intracellular colloidal phenomenon without the essential production of 
such toxic substances as “anaphylatoxins,” “proteotoxins,” and the like. 
Very probably the intracutaneous injection of luetin may elicit a dual 
reaction, one non-specific due to the absorption of antiferment by the agar, 
ascites fluid, and even fragments of Spirocheta pallida, and the other specific, 
a true allergic reaction to pallida protein. But the latter has not been proved 
1 Jour. Infect. Dis., 1916, 18, 402, 415, Jour. Amer. Med. Assoc., 1915, 65, 404. LUETIN REACTION 
709 
and the experiments of Greenbaum and the writer1 have questioned whether 
true cutaneous allergy to culture pallida actually exists. On the other hand, 
it cannot be questioned .that the intracutaneous injectionof luetin into 
syphilitcs has usually elicited more frequent and stronger reactions than 
observed among non-syphilitics; this difference may be ascribable, however, 
more to the development of non-specific and non-allergic hypersensitiveness 
of the skin in syphilis, an enhancement of the non-specific ferment-antifer- 
ment activities, than to the development of specific allergic sensitiveness to 
the proteins of Spirocheta pallida. It was for the purpose of shedding more 
light if possible upon the question of whether or not true cutaneous allergy 
developed in syphilis that our experiments were conducted. 
In our investigations three products were employed: 
(a) Ascites agar luetin prepared according to the original method of 
Noguchi of six or more strains of pallida originally isolated by Noguchi 
and, therefore, many years removed from syphilitic lesions. 
(b) Ascites agar culture-medium in exactly the same proportions as 
contained in the luetin and prepared in exactly the same manner. 
(c) A saline suspension of washed Spirocheta pallida. Naturally most 
interest centered in these preparations as being most likely to yield true 
information on the occurrence of intracutaneous reactions to the fragments 
and products of pallida alone and free of the constituents of culture-media. 
These vaccines of Spirocheta pallida were prepared by cultivating the 
same strains as used for preparing the regular luetin in fluid media and 
removing them by thorough centrifuging. The sediment of spirochetes 
was then washed at least once with sterile saline solution and resuspended 
in sufficient saline solution to give a turbid suspension showing twelve or 
more spirochetes in each dark field by microscopic examination. The 
number of spirochetes in each unit volume was, therefore, somewhat higher 
than present in ordinary Luetin. 
These suspensions were thoroughly shaken with beads, heated at 60° C. 
for one hour, cultured for sterility, and preserved with 0.3 per cent, tricresol. 
Each of the three products were injected intracutaneously into groups 
of syphilitic and non-syphilitic individuals, the skin of the forearms being 
employed. 
The amount injected was always 0.1 c.c. of each, but varying dilutions 
of the stock products were injected in order to elicit possible differences in 
the reactions. For example, the regular ascites agar luetin was injected in 
dose of 0.1 c.c. undiluted, 1 : 2, 1 : 4, 1 : 8, 1 : 16, etc., at the same sitting. 
The control fluid of ascites agar was injected at the same time in the other 
arm in similar amounts. The pure luetin or vaccine of washed spirochetes 
was injected in dose of 0.1 c.c. undiluted, 1 : 2 and 1 : 4, higher dilutions 
not being employed because of the complete absence of all reaction except 
that ascribable to trauma. 
By injecting the luetin and its control in serial dilutions and then measur- 
ing the reactions at the end of forty-eight hours it was hoped to elicit differ- 
ences ascribable to the presence of pallida and that these may be checked 
up by the reactions engendered by pallida alone in the washed suspension. 
Without detailing or tabulating the results of a large number of tests, 
it may be stated that the reactions engendered by the regular luetins and 
their controls were almost identical in degree and extent. Minor differences 
occurred in some instances as might be expected, but these were not constant, 
in that sometimes a given dilution of the control fluid gave a slightly stronger 
reaction than the corresponding dilution of the luetin and the reverse. 
1 Jour. Amer. Med. Assoc., 1922, 79, 2063. 710 
CLINICAL ALLERGY 
The intracutaneous injection of the pallida vaccine in dose of 0.1 c.c. 
sometimes produced well-defined areas of erythema and induration in 
syphilitics, but milder reactions of the same kind were observed among 
non-svphilitics, and in any set of tests conducted on the same day with both 
classes of patients reactions of similar kind and degree were observed in both. 
In general terms the syphilitic patients usually showed somewhat larger 
and better defined reactions which we ascribed to the increased non-specific 
ferment-antiferment cutaneous processes occurring in this disease. 
Our results indicate one of two possibilities: (a) either true cutaneous 
allergy to pallida protein and products does not occur in syphilis and that 
the ordinary luetin reaction is purely a non-specific physical reaction or, 
(b) some change has occurred in the culture pallida employed in our work 
and for the manufacture of luetin in general, reducing or removing their 
allergogenic or anaphylactogenic activities. 
That the latter is no idle possibility is indicated by the recent report of 
Alderson,1- who has compared the results obtained with luetin made at present 
by different commercial firms with those observed some years ago after 
luetin was first prepared by Noguchi. Owing to the frequency of negative 
reactions among syphilitic subjects Alderson has concluded that the luetin 
available at present may be inert and wisely cautions against its use for 
diagnostic purposes. 
In view of the fact that allergic reactions in tuberculosis and some other 
bacterial and mycotic infections may be elicited by preparations of organisms 
long in artificial cultivation it would appear improbable that Spirocheta 
pallida could lose to such a large extent in allergogenic qualities; future 
investigations conducted with freshly isolated pallida must be awaited to 
settle the question, but with the culture pallida not available our results 
indicate that either true and specific cutaneous allergy to pallida does not 
develop in syphilis at all or but to a slight degree, or that the culture pallida 
have actually lost in allergogenic activity, as indicated by the work of 
Alderson. 
MALLEIN REACTION 
Mallein is a glycerin extract containing the toxic principles of the Bacillus 
mallei, the microorganism causing glanders. It is used entirely as a diag- 
nostic agent in veterinary practice, but may also be used for the diagnosis of 
human glanders, the dosage being the same as that of old tuberculin. 
Two methods are commonly employed: (1) The subcutaneous injection 
and (2) instillation into the eye (ophthalmic test). 
Method of Preparing Mallein.—A pure culture of Bacillus mallei is usually obtained 
by injecting a male guinea-pig with infected material, and at the end of twenty-four or forty- 
eight hours isolating a pure culture from the testicle. 
The micro-organism is grown for from six to eight weeks in special bouillon containing 
5 per cent, glycerin, at 37° C., similar to tuberculin. Unlike the tubercle bacillus, glanders 
bacillus grows evenly through the bouillon instead of upon the surface. 
At the end of six or eight weeks the flasks are removed from the incubator and placed in 
a sterilizer at 100° C. for at least two hours. This process kills the bacilli and extracts the 
toxic principles. The entire solution is evaporated down to one-tenth of its volume, filtered 
in small bottles, and sterilized. This is called concentrated mallein, and is kept in this form 
until ready for use. 
Before using it is diluted to its original volume with 0.5 per cent, phenol solution, and 
passed through a porcelain filter. This process removes all the bacilli, and renders the 
solution aseptic and ready for use. 
Subcutaneous Mallein Reaction.—The dose of mallein for a horse is 
0.4 c.c. of concentrated mallein, or 4 c.c. or 1 dram of the diluted product. 
1 Arch. Dermat. and Syph., 1922, 5, 610. ABORTIN REACTION 
711 
The dose for a retest is 0.8 c.c. of concentrated or 8 c.c. or 2 drams of the 
diluted solution. Mallein is injected subcutaneously in some convenient 
area, such as around the shoulder, which has been shaved and cleaned. 
A positive reaction is based on the same principle as tuberculin, that is, 
a rise of temperature within twenty-four hours following the injection, with 
a local inflammatory reaction at the site of injection. 
Ophthalmic Mallein Reaction.—Ophthalmic mallein desiccated, first em- 
ployed by Meyer,1 is prepared by taking 1 part of concentrated mallein 
and adding 20 parts of absolute alcohol. This forms a precipitate that is 
filtered and dried in a desiccator over sulphuric acid. A 5 per cent, solution 
of the powder is made with sterile water. Two or 3 drops of the aqueous 
solution are dropped into the inner canthus of one eye. In case of a positive 
reaction this is followed by a marked conjunctivitis, associated with a 
purulent exudate extending from the inner canthus similar to the reaction 
shown in Fig. 162. In most cases there is also a rise of temperature. Only 
one dose is to be applied in the ophthalmic mallein test. It is not considered 
necessary to sensitize the eye, as in tuberculosis. The ophthalmic mallein 
test has the same advantages as the ophthalmic tuberculin test, that is, one 
can obtain a reaction when dishonest horse dealers have injected mallein 
prior to a subcutaneous mallein test, also in cases of far-advanced glanders, 
which at times give no reaction to a subcutaneous injection of mallein. 
Ferry2 has prepared a convenient tablet of dried mallein for the ophthalmic 
test similar to the tablet of tuberculin previously described. Each tablet 
contains the exact amount of mallein required for the test. Instead of 
dissolving the tablet in water prior to its application, as has previously been 
done with desiccated mallein, the tablet is placed directly into the conjunc- 
tival sac at the inner canthus of the eye and there allowed to remain. The 
tablet will soon (one to three minutes) dissolve without apparent discomfort 
or annoyance to the animal and without an irritating effect upon the con- 
junctiva. The mallein which is thus set free produces typical reactions 
similar to those recorded as the result of the instillation of the raw mallein, 
or the solution of the dried mallein. 
The ophthalmic mallein test is preferred to the subcutaneous and has 
been especially endorsed by the American Veterinary Association.3 The 
French veterinarians appear to prefer anintrapalpebral test, but this requires 
the use of a syringe and is more laborious. Mason4 and others have found 
the complement-fixation test of greatest value, the agglutination test second, 
and the mallein test third, for the diagnosis of glanders in horses and mules. 
ABORTIN REACTION 
Bang5 has observed that artificially infected animals reacted with a 
marked rise of temperature, loss of appetite, and slight diarrhea when in- 
jected subcutaneously with a culture of Bacillus abortus, which he had 
established as the course of infectious abortion of cattle. The English 
Commission6 prepared and used a glycerin extract of Bacillus abortus in 
the same manner as tuberculin, and with success; Meyer and Hardenbergh7 
have prepared a precipitated purified abortin with which highly specific 
1 Jour. Infect. Dis., 1913, 12, 170. 
2 Jour. Amer. Vet. Med. Assoc., October, 1916. 
3 Amer. Vet. Rev., 1913-14, 44, 218. 
4 Jour. Immunology, 1920, 5, 489. 
5 Ztschr. f. Tier., 1897, 1, 241; Archiv. f. Wissen. und Prakt. Tierhl., 1907, 33, 312. 
6 Report of Departmental Committee on Epizootid Abortion, London, 1909, Part I. 
7 Jour. Infect. Dis., 1813, 13, 35. 712 
CLINICAL ALLERGY 
reactions have been observed; Reichel and Harkins1 have employed an 
intradermal test with a suspension of heat-killed and washed bacilli. The 
results of these investigations indicate that the abortin test is highty specific; 
as with other anaphylactic reactions, this test does not serve to differentiate 
the actively infected from the recovered animals, but when applied to a 
herd will show whether or not Bang’s disease is or has been a source of 
infection among the animals. 
According to Fleischner and Meyer the test has been uniformly negative 
in infants, indicating that Bacillus abortus equinus is probably not pathogenic 
for human beings and that infection does not occur through the ingestion of 
milk. 
ALLERGIC REACTIONS IN TYPHOID FEVER 
Historic.—In 1907 Chantemesse2 observed characteristic inflammatory 
symptoms follow the installation of typhoid bacilli extract into the eye of 
patients suffering from typhoid fever. Kraus3 and his associates, repeating 
these experiments, could not convince themselves of the specificity of this 
reaction, stating that healthy individuals also give it to some extent, and 
that other bacterial extracts cause similar symptoms in typhoid-fever 
patients. In addition, he tried a cutaneous reaction, but without result. 
Zupnik,4 on the contrary, states that a cutaneous reaction is useful, while the 
ophthalmic reaction is not useful. Deehan5 obtained a weak to moderate 
reaction in 12 cases of typhoid fever, whereas 8 control cases showed none. 
Floyd and Barker6 obtained positive results in 19 out of 20 cases and none 
in 18 controls, including 2 cases of paratyphoid fever. Chaufford and 
Trosier7 reported unfavorably on the reaction. 
Austrian8 has reported very favorably upon an ophthalmic reaction in 
typhoid fever following the installation of “typho-protein” prepared by 
cultivating a large number-of different strains of typhoid bacilli, precipitating 
the protein with alcohol, drying the precipitate, and redissolving in 
water so that from to milligram is contained in each drop. In 
typhoid-fever patients reaction of the palpebral conjunctiva of the lower lid 
and of the caruncle appears on an average of two and a half hours later, 
reaching the maximum about the sixth hour, and usually subsiding within 
forty-eight hours. In 75 cases of typhoid fever this test was found positive 
in 71 and negative in 4. In 3 cases the eye test antedated the Widal reaction, 
and in only 23 per cent, was the Widal reaction positive at as early a date as 
the eye test. A study of 190 persons normal or ill with diseases other than 
typhoid has convinced Austrian of the specificity of the test, and he recom- 
mends it as an aid to diagnosis on account of its simplicity and the absence 
of any discomfort to the patient. 
Typhoidin Test of Gay and Force.—Gay and Force9 have reported 
favorably upon a cutaneous reaction indicative of immunity against typhoid 
fever. The preparation which they used, “typhoidin,” is prepared in the 
same manner as Koch’s old tuberculin: 250 c.c. of a 5 per cent, glycerin 
bouillon is inoculated with Bacillus typhosus and cultivated at 37° C. for 
1 Jour. Amer. Vet. Med. Assoc., March, 1917. 
2 Deut. med. Wchn., 1907, 33, 1572. 
3 Wien. klin. Wchn., 1907, 20, 1335. 
4 Munch, med. Wchn., 1908, 45, 148. 
5 Univ. Penn. Med. Bull., 1909, 22, 6. 
6 Amer. Jour. Med. Sci., 1909, 38, 188. 
7 Compt. rend. Soc. de biol., 1909, lxvi, 519. 
8 Bull. Johns Hopkins Hosp., 1912, 23, 1. 
9 Archiv. Int. Med., 1914, 13, 471. ALLERGIC REACTIONS IN TYPHOID FEVER 
713 
five days. It is then reduced, without filtration, to one-tenth of its original 
volume by evaporation over an acetone bath for about eight hours. A con- 
trol solution of sterile 5 per cent, glycerin bouillon is prepared in the same 
manner. 
The skin of the forearm is cleansed with alcohol, and two abrasions are 
made with the von Pirquet borer, as described under the cutaneous tuber- 
culin test. The “typhoidin” is applied to one cut and the control fluid to 
the other. The reactions are observed six and twenty-four hours later. 
Occasionally there is a traumatic reaction in the control, but a positive 
reaction may be detected by a wider areola and increased induration. 
Positive reactions were secured in 95 per cent, of cases that had recovered 
from typhoid fever, 2 of the cases having had the disease respectively forty- 
one and thirty-three years before. The reaction was found negative in 
85 per cent, of individuals not having typhoid fever. Of 15 persons immu- 
nized by the army method, from four and three-quarter years to eight months 
previously, nine gave a positive skin reaction. Twenty-four individuals 
immunized by a sensitized vaccine (Gay and Claypoole) for from one to 
eight months previously reacted positively. Later Gay and Claypoole1 
prepared typhoidin by precipitating the solution with alcohol, washing the 
precipitate with alcohol and ether, drying in a vacuum, and suspending the 
resulting powder in phenolized normal salt solution which was injected 
intracutaneously and applied cutaneously; a control powder was prepared 
from broth and used in the same manner. With this skin test Gay and his 
associates have studied the relative value of various vaccines and regard the 
anaphylactic reaction as indicative of a state of immunity. Nichols2 has 
questioned the value of the anaphylactic skin test as an index of immunity 
and regards the reaction as indicating nothing more than sensitization to 
typhoid protein, which is apparently less lasting and less specific than the 
true immunity to this infection. He bases this opinion on the fact that in 
his experience the typhoidin skin test gave fewer positive reactions (75 per 
cent.) than generally expected, as about 90 per cent, of persons who have 
had typhoid fever are immune for many years or even for the balance of life. 
Furthermore, according to Nichols, experience has shown that protection 
following typhoid fever is of longer duration than is indicated by the typhoid 
in test, and while a large percentage of persons who have had typhoid fever or 
have been immunized with typhoid vaccine react to paratvphoidin, recent 
experiences and statistics, particularly in Europe, have indicated that these 
persons are not immune to paratyphoid fever. Kilgore3 has reported favor- 
ably upon the value of the typhoidin cutaneous test; Austrian and Bloomfield4 
found that the test failed to furnish data by means of which it was possible 
to differentiate between those who had neither typhoid fever nor had re- 
ceived the vaccine and those who had either had the disease or had been 
immunized. 
In our own experience5 powTdered typhoidin and its control produced 
severe reactions when injected intracutaneously in doses of 0.0005 to 0.001 
mgm.; these reactions and particularly that produced by the control rendered 
the reading and interpretation of the test quite difficult and subject to much 
error. 
Force and Stevens6 later prepared a stable preparation of typhoidin 
which may be rapidly prepared by precipitating a concentrated broth 
culture of Bacillus typhosus with 95 per cent, alcohol, and subsequent 
1 Archiv. Int. Med., 1914, 14, 671. 
2 Jour. Exper. Med., 1915, 22, 780. 
3 Archiv. Int. Med., 1916, 17, 25. 
4 Archiv. Int. Med., 1916, 17, 663. 
5 Jour. Immunology, 1916, 1, 409. 
6 Archiv. Int. Med., 1917, 19, 440. 714 
CLINICAL ALLERGY 
dehydration with absolute alcohol and absolute ether. According to these 
investigators it is imperative that no account be taken of the appearance of 
the reaction at the end of twenty-four hours. A positive typhoid in reaction 
is indicated by the presence, forty-eight hours after the test, of a well-defined 
erythematous papule at least 5 mm. in diameter. 
Out of 18 normal persons, 17 gave negative reactions; out of 25 persons 
with a history of typhoid, 19 gave positive reactions, 1 gave a doubtful, and 
6 persons (4 with questionable typhoid histories) gave negative reactions. 
Out of 152 persons previously vaccinated against typhoid, 12 of 56 vaccinated 
during 1916, 3 of 11 vaccinated during 1915, 13 of 29 vaccinated during 1914, 
and 26 of 52 vaccinated during 1913 gave negative reactions. 
Meyer and Christiansen have likewise emphasized the importance of 
the method employed for the preparation of typhoidin in order to reduce 
the degree of non-specific irritation. 
Cutaneous anaphylaxis to typhoidin was found apparently to persist 
for a longer time among those who have had typhoid fever than among those 
actively immunized with the vaccine. Among the latter the highest per- 
centage of reactions was found during the first year following immunization. 
The test was advocated as a means of determining whether or not a person 
possesses immunity to typhoid fever either acquired by recovery from the 
disease or by artificial immunization; the sum total of researches by various 
investigators is to the effect that this skin test is an indication of hyper- 
sensitiveness to typhoid protein, but cannot be accepted at present as an 
indication of immunity. 
Berge and myself1 found no relation between cutaneous allergy and the 
presence of antibodies, and Meyer and Christiansen2 observed exactly similar 
results in experiments with rabbits. The latter observed that a positive 
typhoidin skin reaction in a rabbit does not indicate that this animal will 
resist a subsequent intravenous injection of living typhoid bacilli, or that 
the animal is so protected that it will not become a chronic carrier of bacilli 
in gall-bladder or liver. And again, a marked hypersensitiveness of a 
typhoid-immune rabbit to fowl typhoidin does not indicate the presence 
of a resistance to an infection with the same organisms. 
Force and Stevens,3 however, have maintained that the original assump- 
tion of Gay and Force is true that the typhoidin test may be used as a measure 
of protection against typhoid. They state that even granting that typhoid 
immunity is of longer duration than cutaneous sensitiveness to typhoid 
protein, the disappearance of this sensitiveness furnishes an indication for 
revaccination of the person, and still allows a margin of safety within the 
as yet indefinite limits of typhoid immunity. There is as yet no evidence 
that a positive typhoidin test is not indicative of protection against typhoid 
fever. In two instances at least, in their experience, a negative typhoidin 
test after vaccination was followed by typhoid infection. 
ALLERGIC REACTIONS IN OTHER DISEASES 
Gonococcus Infections.—In 1908 Irons4 reported general and local 
reactions in persons suffering from gonococcus infections following the 
subcutaneous injection of gonococcal vaccines. This reaction has been 
observed by Bruck5 in epididymitis, by Reiter6 in pelvic infections in women, 
and also by other observers in other conditions. 
1 Jour. Immunology, 1916, 1, 409. 
2 Jour. Infect. Dis., 1917, 20, 357^41. 
3 Archiv. Int. Med., 1917, 19, 440. 
4 Jour. Infect. Dis., 1908, v, 279. 
5 Deut. med. Wchn., 1909, xxxv, 470. 
6 Ztschr. f. Geburtsh. u. Kinderh., 1911, lxviii, 471. ALLERGIC REACTIONS IN OTHER DISEASES 
715 
Experiments with glycerin extracts of the gonococcus prepared from 
several strains, singly or combined in one preparation, have yielded Irons,1 
well-defined cutaneous reactions. These tests were conducted after the 
method of von Pirquet’s tuberculin test. 
Irons found that the cutaneous inoculation of glycerin extracts of au- 
tolyzed gonococci in patients infected by the gonococcus produces a well- 
defined reaction. This reaction is not usually obtained in normal persons, 
nor in those suffering from other infectious diseases. In persons recently 
infected, the reaction is negative and increases gradually during the course 
of the disease. In the more chronic forms of gonococcal infection, such as 
arthritis, the degree of the cutaneous reactivity varies from day to day, and 
these variations may be correlated with the changes in the clinical course 
of the disease. Cases of severe infection, such as extensive arthritis, may 
give negative reactions. Later, when improvement has occurred, the 
reaction becomes positive. In general, a positive reaction is obtained in 
patients with gonococcal infection at some time during the course of the 
disease. In normal persons the gonococcin prepared in the manner de- 
scribed gives a cutaneous reaction rarely more than 2 to 3 mm. in diameter. 
Occasionally in adults and somewhat more frequently in children fairly 
marked reactions are met with where previous gonococcal infection can be 
excluded. In these cases the normal antibodies may be increased to an 
unusual degree. The cutaneous reactions obtained with meningococcal 
and gonococcal antigens suggest that we are dealing with a group reaction. 
In diagnosis, a positive reaction is to be regarded as confirmatory evidence 
of gonococcal infection. Other infections, such as those by the meningo- 
coccus or Micrococcus catarrhalis, which may give rise to a group reaction, 
must be excluded. The clinical value of the reaction must be determined 
by further tests and its limitations defined by a study of many groups of 
cases. 
Diphtheria.—Schick has advocated the intracutaneous injection of a 
minute dose of diphtheria toxin as a test for antitoxin in the serum of an 
individual. If sufficient antitoxin is present, the toxin is neutralized and 
no local disturbances are apparent; otherwise local inflammatory areas 
may be observed. This test is not regarded as an allergic reaction; a further 
description of it will be found in Chapter XXXII. 
Recently Dr. Moshage and myself2 have applied an anaphylactic skin 
test in diphtheria with a polyvalent antigen designated diphtherin; positive 
reactions were observed in about 70 per cent, of children and 35 per cent, 
of adults, and the test was of practical interest mainly from the viewpoint 
that the anaphylactic reaction may be mistaken for a positive Schick reaction. 
Pneumonia.—Many attempts have been made to devise a skin test for 
pneumococcus pneumonia not only for possible differential diagnosis of 
types of pneumococci, but more especially as studies upon immunity in this 
disease and the mechanism of the crisis. 
Clough3 employed cutaneous and ophthalmic tests with pneumococcus 
protein secured by precipitating autolyzed pneumococci with alcohol, with 
practically negative results. 
Weil,4 employing suspensions of Type I and Type III pneumococci in water 
heated to 60° C. injected intracutaneously, observed no reactions during 
the course of the disease, but after crisis a considerable percentage yielded 
1 Jour. Amer. Med. Assoc., 1912, lviii, 931; Jour. Infect. Dis., 1912, 11, 77. 
2 Amer. Jour. Dis. Children, 1916, xii, 316. 
3 Johns Hopkins Hosp. Bull., 1913, 24, 295. 
4 Jour. Exper. Med., 1916, 23, 11. 716 
CLINICAL ALLERGY 
positive reactions; positive reactions were also observed in some controls 
which Weil believed may have been due to unrecognized sensitizations to 
pneumococcus proteins. 
Steinfield and myself1 prepared antigens of pneumococci of Types I, II, 
and III by suspending the cocci in saline solution and heating to 60° C.; 
the amount injected intracutaneouslv was 0.1 c.c. or 200,000,000 cocci. 
Of 19 cases of lobar pneumonia, 6 gave positive reactions with one or 
more antigens. But in none did the type or types of pneumococci found in 
the sputum exactly correspond with those shown by the skin reaction, and 
we concluded that the allergic reactions to pneumococcus protein are of a 
more general character than the agglutination reactions. The earliest 
reaction occurred three days after the crisis and a large number of controls 
reacted negatively. 
Weiss and myself2 have also prepared “pneumotoxin” by autolyzing 
young cultures of virulent Type I pneumococci in 2 per cent, solution of 
sodium oleate. This material represented a thermolabile toxin which had 
to be freshly prepared. Tests were conducted by injecting 0.1 c.c. intra- 
cutaneously, which represented one-twentieth the minimal lethal dose for 
guinea-pigs. A control fluid was prepared in the same manner. 
This toxin was analogous to the diphtheria toxin employed in the Schick 
test for immunity to diphtheria. Positive reactions were observed in 38 
cases of lobar pneumonia as early as the fifth day (the earliest under obser- 
vation) and as late as the thirteenth. Positive reactions were also observed 
in 5 of 16 cases after the crisis and convalescent; 23 controls yielded negative 
reactions. 
While wre considered these positive reactions due to sensitization to 
pneumotoxin, it is possible that the reactions were not altogether allergic, 
but the result of direct irritation of the skin just as diphtheria toxin is an 
irritant unless neutralized by antitoxin. Positive reactions may indicate 
the absence of neutralizing principles early in pneumonia, but these may be 
developed later in the disease and account for the high percentage of 
negative reactions after the crisis. 
Bigelow3 has studied the subject of allergic skin reactions in pneumonia 
employing ten different kinds of antigen of all four types of pneumococci. 
Of 104 cases of lobar pneumonia tested, 11 (10.5 per cent.) gave specific 
type reactions, 46 (42.3 per cent.) gave a common reaction, and 52 (50 per 
cent.) gave no such reaction. Of the 20 controls, none gave a specific type 
reaction, 9 gave a common reaction, and 11 gave no reaction. Best results 
were observed with an antigen prepared by autolyzing pneumococci in 
saline solution or distilled wrater. 
While apparently specific type intracutaneous reactions may be observed 
in pneumonia, these do not occur early enough to be of service in directing 
serum therapy; a second non-specific reaction, differing in time and character 
from the specific type reactions, may occur in cases of pneumonia and 
controls. 
Meningococcus Carriers.—Gay and Minaker4 have prepared menin- 
gococcin from six strains of normal and parameningococci by precipitating 
saline suspensions with alcohol, washing the sediment with ether, and drying.. 
This material was ground into a fine powder and suspended in physiologic 
saline solution containing 0.5 per cent, phenol in such proportions that 
1 Tour Infect T)is 1Q17 20 S4-4- 
2 Jour! Immunology, 1918, 3*, 395; Arch. Int. Med., 1923, 31, 263. 
3 Archiv. Int. Med., 1922, 29, 221. 
4 Jour. Amer. Med. Assoc., 1918, 70, 215. ALLERGIC REACTIONS IN OTHER DISEASES 
717 
0.05 c.c., the amount injected intracutaneously, contained 1/150 mg. of 
the powder. 
A reaction was interpreted as positive when presenting at the end of 
twenty-four hours an areola of 3 to 7 mm. with distinct induration. Of 
31 positive carriers, the test was positive in 20, or 64.5 per cent.; of 38 carriers, 
the test was positive in 10, or 26.4 per cent. 
The authors have drawn no conclusions as to the significance of these 
reactions, but suggest that this intradermal test may prove of aid in the 
detection of those carrying meningococci for sufficient time to effect sensiti- 
zation, and that positive reactions may indicate the acquisition of some 
degree of resistance. 
Pertussis.—Modigliani and Villa1 have recently described an intradermal 
test for whooping-cough. A loop of a culture of the Bordet-Gengou bacillus 
in 1 c.c. of distilled water containing a little of a 3 per cent, solution of 
toluene, was injected into the skin by the tuberculin intradermal technic. 
The dose for each child was 0.1 c.c. No response was obtained with various 
diseases other than whooping-cough, while an inflammatory reaction formed 
constantly in the 38 children with pertussis. The reaction was negative 
also in 10 children that had recovered from whooping-cough. The results 
were most instructive in 3 children who had been exposed to pertussis, but 
showed no signs of it at the time. They gave a pronounced positive response 
to the injection, and a few days later the symptoms of pertussis developed. 
Orgel2 has recently applied an intracutaneous test by injecting 2 minims 
of a vaccine of Bacillus pertussis containing 2,000,000,000 per cubic centi- 
meter. He states that specific reactions were observed that sensitization 
occurs early enough in the disease to offer hope that the skin test may 
prove of value in early diagnosis, at least before the paroxysmal stage. 
Sporotrichosis, Blastomycosis, and Coccidiosis.—Meyer3 and Moore and 
Davis4 have reported positive reactions conducted by the intracutaneous 
injection of sporotrichin and may prove of practical value in the diagnosis 
of sporotrichosis. 
Stober5 has reported negative skin and ophthalmic reactions in blastomy- 
cosis. 
Cooke6 has reported negative cutaneous and intracutaneous reactions 
in coccidiodal granuloma. 
Smallpox; Cowpox; Chickenpox.—As previously stated, Jenner in 
1798 first described skin reactions when reporting his failure to inoculate 
14 persons with cowpox who had previously had smallpox: 
“It is remarkable,” writes Jenner, “that variolous matter, when the 
system is disposed to reject it, should excite inflammation on the part to 
which it is applied more speedily than when it produces smallpox. Indeed, 
it becomes almost a criterion by which we can determine whether the infec- 
tion will be received or not. It seems as if a change which endures through 
life has been produced in the action or disposition to action in the vessels 
of the skin; and it is remarkable, too, that whether this change has been 
effected by smallpox or the cowpox that the disposition to sudden cuticular 
inflammation is the same on the application of variolous matter.” 
This “sudden cuticular inflammation” of Jenner is recognized by von 
Pirquet as “the immediate reaction” which follows attempts to vaccinate 
1 Pediatrica, Naples, 1921, 29, 337. 
2 Jour. Amer. Med. Assoc., 1922, 79, 1508. 
3 Jour. Amer. Med. Assoc., 1915, 65, 579. 
4 Jour. Infect. Dis., 1918, 23, 252. 
5 Arch. Int. Med., 1914, 13, 509. 
6 Arch. Int. Med., 1915, 15, 479. 718 
CLINICAL ALLERGY 
those how have recovered from smallpox or have been recently vaccinated. 
It indicates, in other words, that protection already exists in that individual, 
that antibodies are present which detroy the vaccine colony and prevent 
the evolution of the vaccine pustule. Force has suggested that this imme- 
diate or “immunity reaction” (see Fig. 173) should be looked for in twenty-four 
hours, as its presence is indicative of immunity to cowpox and smallpox 
and would obviate the repeated and vain attempts to produce true vaccinia 
in such an already protected person. 
Force and Beckwith1 have further employed a localized reaction of this 
type in the differential diagnosis of smallpox. They find that rabbits may 
be immunized against vaccinia or variola by subcutaneous injections of 
vaccine virus. Such animals show a specific local reaction on the intra- 
dermal injection of vaccine virus or of pus from a smallpox vesicle. They 
do not, however, react to varicella material. Force2 has recently been able 
to obtain an early diagnosis in 2 cases of suspected smallpox by this method. 
One drawback to the method, however, is the difficulty of giving rabbits 
intradermal injections on account of their thin skins; sometimes these injec- 
tions are facilitated by using the skin of the ears instead of the abdomen or 
back. 
Tieche3 has employed similar tests for the differential diagnosis between 
smallpox and chickenpox by inoculating himself and 2 other volunteers 
immune to smallpox with the material from the lesions of the patient. 
The lymph from the patient is first heated to 60-70° C. for five minutes and 
then applied to scratches on the arm; if the patient has smallpox a reaction 
occurs in a few hours, whereas in the case of chickenpox no local reaction 
is evident for several days. 
Echinococcus Disease.—In echinococcus disease of the liver and spleen 
Pontano4 and Luridiana5 have employed intradermal tests for diagnosis 
consisting of the injection of 0.2 to 0.3 c.c. of cyst fluid. They quote Casoni 
as having previously used this test successfully for diagnosis employing 
the fluid from cysts of the lower animals preserved with phenol. Pontano 
reports positive reactions in 84 per cent, of cases, suppuration of the cysts 
being responsible for the negative reactions sometimes obtained. Luridiana 
reports 5 positive reactions in 6 patients, the negative reaction occurring 
in a patient with cyst walls unusually fibrous, which probably interfered with 
sensitization. Serra6 reports positive skin reaction in 90 per cent, of cases 
of echinococcus disease of the liver and spleen. Negative reactions were 
abscribed to thickened cyst walls limiting sensitization and to processes of 
desensitization. Complement-fixation and skin reactions are believed capa- 
ble of correctly diagnosing the majority of cases. 
Ringworm and Favus.—Bloch and Massini7 have reported positive 
skin reactions to trichophytin among animals immunized with ringworm 
cultures. Amberg8 studied the test in a large group of individuals by apply- 
ing tricophytin to abrasions of the skin and reports a high percentage of 
positive reactions. Strickler and myself have prepared antigens of various 
cultures of ring-worm and favus by grinding the mycelia with sodium chlorid 
and suspending in sufficient water to render the solutions isotonic, followed 
1 Jour. Amer. Med. Assoc., 1915, 65, 588. 
2 Jour. Amer. Med. Assoc., 1916, 66, 1384. 
3 Corresp. Blatt. f. Schw. Aertze, 1913, 44, 1121. 
* Policlinico, 1920, 27, 1921, 28, 41, 405. 
6 Policlinico, 1921, 28, 41, 399. » 
6 Policlinico, 1921, 28, 35. 
1 Ztschr. f. Hyg., 1909, lxiii, 68. 
8 Jour. Exper. Med., 1910, 12, 435. ALLERGIC REACTIONS IN OTHER DISEASES 
719 
by sterilization with heat and preservation with phenol. The intracutaneous 
injection of 0.05 to 0.1 c.c. of this trichophytin into individuals with body 
ringworm was followed by positive reactions, whereas with favin, negative 
or weakly positive reactions occurred. With individuals with favus the 
intracutaneous injection of favin was followed by positive reactions, whereas 
the tricophytin yielded negative or much weaker reactions. 
Pregnancy.—From time to time the possible allergic nature and mechan- 
ism of labor is discussed in medical literature. Heide1 in 1911 advanced 
experiments in support of this theory, but in my own experiments2 I was 
unable to demonstrate any anaphylactic influence upon guinea-pigs at or 
near term by the intravenous injection of maternal pig sera collected just 
before or just after labor; nor by the injection of sera of newborn pigs and 
placental extracts. Thies3 believed that eclampsia may be an allergic 
phenomenon, but Johnstone4 secured no support for this hypothesis in his 
experiments. 
Engelhorn and Wintz5 and Esch6 claim to have observed positive skin 
reactions among pregnant women with preparations of placenta designated 
as placentins. Williams and myself' also observed reactions of this character 
among pregnant women and those who have borne children, but neither 
we nor Falls and Bartlett8 found them constant or of diagnostic value; 
they occur, however, in late pregnancy and are significant from the stand- 
point of possible sensitization with placental protein. 
Cancer.—Analogous to these attempts toward demonstrating allergic 
skin reactions in pregnancy are the skin tests in cancer with tumor juices, 
on the basis that the proteins of new cancer cells may effect sensitization of 
the host. 
Ransohoff9 has reported that guinea-pigs passively sensitized with 0.1 c.c. 
serum from individuals with advanced cancer can be thrown into an ana- 
phylactic reaction about ten days later by the intraperitoneal injection of 
3 to 5 c.c. of blood-serum from individuals with cancer. Presumably under 
these conditions serum in cancer carries both antibody and antigen. The 
author concluded that the test may prove of diagnostic value, inasmuch 
as it occurred regularly in cancer and had a margin of error of not more 
than 8 per cent. Pfeiffer10 and Ranzi11 had previously reported positive 
reactions employing guinea-pigs actively sensitized by injections with tumor 
juices. 
Further reports on work of this kind do not appear to have been made, 
and the methods of passive and active anaphylactic tests for cancer have 
not been definitely established. 
Skin tests of a different character have been employed by Elsberg, 
Neuhof, and Geist12 in the diagnosis of cancer, consisting of the subcutaneous 
or intracutaneous injection of a few drops of a 20 per cent, suspension of 
the washed corpuscles of a normal healthy person. This test is not allergic, 
but simply a test in vivo for the isohemolvsins sometimes developing in 
1 Munch, med. Wchn., 1911, lviii, 1705. 
2 Jour. Med. Research, 1914, 24, 425. 
3 Archiv. f. Gyn., 1911, 92. 
4 Jour. Obstet. and Gyn. of Brit. Empire, 1911, 19, 253. 
5 Munch, med. Wchn., 1914, lxi, 1115. 
6 Munch, med. Wchn., 1914, lxi, 1115. 
7 Amer. Jour. Obstet., 1915, 71, No. 6. 
8 Amer. Jour. Obstet., 1914, 70, 884. 
9 Jour. Amer. Med. Assoc., 1911, 57, 103; 1913, 61, 8. 
10 Ztschr. f. Immunitatsf., 1909-10, 4, 458. 
11 Ztschr. f. Immunitatsf., 1909, 2, 12. 
12 Amer. Jour. Med. Sci., 1910, 139, 264. CLINICAL ALLERGY 
720 
cancer. The authors reported 89.9 per cent, positive reactions among 
persons with cancer and about 6 per cent, reactions among controls. 
Warfield,1 however, observed only 32.4 per cent, positive reactions and 
concluded that the reaction was without value in the early diagnosis of 
cancer; similar results were observed by Risley.2 Lisser and Bloomfield3 
observed positive reactions in 62 cases of a series of 158 and negative reac- 
tions in about 90 per cent, of controls. 
1 Archiv. Int. Med., 1911, 7. 
2 Boston Med. and Surg. Jour., July, 27, 1911. 
3 Bull. Johns Hops. Hosp. Bull., 1913, 23, 356. CHAPTER XXXII 
TREATMENT OF HUMAN ALLERGIES 
Principles.—The prophylactic treatment of allergy embraces methods for 
the prevention of allergic reactions. This may be accomplished by (a) 
removing the intake of the allergen (exciting substance); (b) by removal of 
the allergic antibodies from the sensitized cells; (c) by the administration of 
drugs tending to depress allergic cells or the effects of allergic shock or by a 
combination of these. 
After the allergic reaction has occurred as in “serum sickness,” asthma, 
etc., remedial measures are directed to the treatment of allergic shock; this 
treatment is usually symptomatic, as the administration of adrenalin chlorid 
for the relief of urticaria. It may be, however, a form of specific treatment, 
as the administration of pollen extracts for the relief of hay-fever during an 
attack by the process of desensitization. 
The prevention of allergy and the specific treatment of allergic reactions 
depends upon the process of desensitization; this refers to the exhaustion of 
allergic antibody from sensitized cells by gradually and slowly bringing the 
allergen or exciting substance into contact with these cells. Probably a 
series of allergic shocks occur, but ill effects are avoided by rendering these 
of a mild character. Exhaustion of antibody results in anallergy or desensi- 
tization of the cells; the process generally requires considerable time and 
the administration of a large number of doses of allergen and particularly 
in cases of natural allergy, as hay-fever, the dust asthmas, and food 
allergies. 
This process of desensitization is generally a specific process, that is, the 
antibody is exhausted by the administration of its specific allergen. But 
partial desensitization can sometimes be brought about by the administration 
of a non-specific substance. For example, in allergy to horse-serum the 
injection of beef-serum may protect against a subsequent injection of horse- 
serum; in the opinion of some observers the injection of a pollen extract to 
which allergy does not exist may partially desensitize against the pollen to 
which sensitization does exist. 
Desensitization is not anti-anaphylaxis. The latter term as originally 
defined by Besredka and Steinhart meant desensitization, but, as discussed 
in Chapter XXVIII, we have adopted Weil’s definition to cover that state of 
anallergy in which the sensitized cells are protected against the exciting 
agent by the presence of antibody in the blood. These free antibodies are 
supposed to neutralize the allergen before it reaches the cells. According 
to this definition anti-anaphylaxis is really a state of extreme sensitization 
with an excess of antibody and probably is of exceptional occurrence in 
human beings, inasmuch as the demonstration of the presence of these free 
or excess antibodies by injecting human blood into a lower animal (passive 
sensitization) is only occasionally successful in serum allergy. 
The theoretic considerations of specific and non-specific desensitization 
and anti-anaphylaxis are given in Chapter XXVIII. In this chapter the 
practical aspects and methods of prophylaxis and treatment of human 
allergies will be considered. 
721 TREATMENT OF HUMAN ALLERGIES 
722 
PROPHYLACTIC TREATMENT OF SERUM ALLERGY 
This refers almost entirely to the treatment of allergy to horse-serum 
since immune horse-sera are widely employed for the prophylaxis and treat- 
ment of some of the infectious diseases, and notably diphtheria, tetanus, and 
meningococcus meningitis. 
Prevention of Sensitization.—No efficient method of preventing sensiti- 
zation of human beings by horse-serum has been discovered, although this 
is a very important problem worthy of investigation. 
Besredka1 states that heating therapeutic sera to 56° C. for an hour on 
four days in succession renders the serum less sensitizing. This method is 
practised at the Pasteur Institute, and Besredka is of the opinion that it 
“explains why in France serum accidents have always been relatively rare, 
and why in the small number of cases (13 per cent.) in which they occur 
they are not of such a serious nature as they are in countries where sera are 
not heated.” 
Cases Requiring Desensitization.—Not all persons who have received 
injections of normal or immune horse-serum become sensitized or require 
desensitization before a subsequent injection may be given. 
In the majority of instances serum is given subcutaneously or intra- 
muscularly for the prophylaxis and treatment of diphtheria or the prophy- 
laxis of tetanus, and while serum sickness may follow, desensitization is not 
absolutely required. When serum is given intravenously and especially to 
individuals who have received serum on a previous occasion or when serum is 
to be given by any route and even by simple subcutaneous injection to asthmatics 
or individuals known or suspected to be hypersensitive, desensitizalion is advis- 
able and necessary. 
The writer has formulated the following rules which have proved quite 
satisfactory in practice for guidance in this perplexing problem: 
1. Horse-serum should never be given to “horse asthmatics” by any 
route of parenteral injection—subcutaneous, intramuscular, intrathecal, or 
intravenous—without previous desensitization and very special precautions. 
Horse asthma is usually a natural allergy, that is, acquired by unknown 
means. There are some cases on record, however, where this extreme 
sensitization has apparently resulted from a previous injection of horse- 
serum. 
Some cases of horse asthma yield skin reactions to the proteins of horse 
hair and dandruff and not to serum, but the majority react to serum as well, 
and all are extremely bad risks. Desensitization of these persons to the 
point where an adequate dose of serum may be safely given is a tedious 
process and frequently impossible by our present methods. Horse-serum 
should be avoided and it is very much desired that our manufacturing 
laboratories render available diphtheria and tetanus antitoxins and anti- 
meningococcus serum by the immunization of cattle for use in these indi- 
viduals. 
If horse-serum is given at all, even after attempts to secure desensitiza- 
tion, the injections should be subcutaneous and the preliminary injection 
of 1/50 grain of atropin is advisable. After the injection the patient should 
be carefully watched and treated as described on p. 662 if a reaction occurs. 
All cases of asthma due to other proteins are risks and especially if serum is 
given intravenously and intrathecally (intraspinal injection); likewise cases of 
thymolymphaticus. At least skin tests should be done and desensitization 
1 Anaphylaxis and Antianaphylaxis. Translated by Gloyne, C. V. Mosly Co., St. Louis, 
1919. PROPHYLACTIC TREATMENT OF SERUM ALLERGY 
723 
practised if positive reactions are observed. Subcutaneous injections are the 
least dangerous if serum must be administered. 
2. A skin test should be made before serum is given intravenously even 
though it is the first injection of serum the patient has ever received. Even 
if the reaction is negative it is good practice to give 1 c.c. of the serum 
subcutaneously one hour before the intravenous injection. If the reaction 
is positive several preliminary subcutaneous injections should be given. 
Serum given intravenously is brought into sudden contact with the cells 
and conditions are perfect for eliciting shock if the cells are sensitized. It 
is to be remembered, however, that ill effects may be due to non-specific 
protein shock, and as these may not be allergic, desensitization may not 
prevent such symptoms as fever and chills. 
The same applies to the administration of serum by intraspinal injection, 
but the danger is less. 
3. When an individual has been injected with horse-serum some weeks, 
months, or years previously and serum is to be administered again by intra- 
venous or intraspinal injection, preliminary desensitization is always advis- 
able regardless whether or not skin tests are made. Skin tests, however, are 
advisable because if positive reactions are observed desensitization should 
be more thorough than if they are negative. 
4. When serum is to be given by subcutaneous or intramuscular injec- 
tion, skin tests and desensitization are not necessary if the patient is not 
asthmatic and if it is the first injection. “Serum disease” may follow five 
to twelve days after the injection, but is not dangerous. 
Occasionally severe allergic reactions of collapse, dyspnea, tachycardia, 
and fall in blood-pressure follow the subcutaneous injection of serum into 
individuals who have received an injection one to six or eight weeks pre- 
viously; for this reason a desensitizing injection of 1 c.c. of serum one hour 
previously is advisable before the administration of the balance of the 
serum, by subcutaneous or intramuscular injection. 
The following table summarizes this subject in a brief manner: 
DESENSITIZATION 
1. When patient is not asthmatic, has never before received serum, and when 
the injection is to be subcutaneous, intramuscular, or subthecal. 
2. When patient is not asthmatic, but has received serum before, and the 
injection is to be subcutaneous or intramuscular. 
Not required 
1. When serum is to be given for the first time intravenously to an individual 
who has not received an injection of serum on a previous occasion. 
2. When serum is to be given intravenously or intrathecally to an individual 
who has received serum on a previous occasion. 
3. WThen serum is to be given subcutaneously, intrathecally, or intravenously 
to an individual who is an asthmatic and regardless of whether or not 
serum has been previously administered. 
Advisable.... 
Methods for Specific Desensitization.—Unfortunately, methods proving 
satisfactory for the desensitization of guinea-pigs and other of the lower 
animals do not desensitize human beings equally well. Ordinarily four 
methods may be employed according to circumstances, as follows: 
(a) Precautionary Desensitization; Individual Not Asthmatic and Skin Test 
Negative.—Sterile horse-serum (either normal serum or the immune serum 
to be employed) should be injected subcutaneously in dose of 1 c.c. at least 
one hour and preferably two hours, before the balance of serum is adminis- 
tered. If the serum is to be given intravenously the first few cubic centi- 
meters should be injected very slowly. Desensitization persists for from 
six to eight hours to several days. 724 
TREATMENT OF HUMAN ALLERGIES 
The following methods are modified after those of Besredka when serum 
is to be injected subthecally: Method A, Give 2 c.c. by intraspinal injection 
and balance (20 to 30 c.c.) one or two hours later. Method B, Dilute 2 c.c. 
serum with 18 c.c. of sterile saline solution in a 20 c.c. syringe and give 1 c.c. 
( = 0.1 c.c. undiluted serum) intravenously; while the needle is still in the 
vein wait three minutes, and if there are no symptoms inject 5 c.c. Wait 
two minutes, and if there are no symptoms inject the balance of 14 c.c. 
Serum may now be given subthecally at any stage of the illness. 
(b) Required Desensitization; Individual Not Asthmatic, but Skin Test is 
Positive and Serum is to Be Given Subcutaneously or Intramuscularly.—Usually 
the above subcutaneous method is sufficient or the following: subcutaneous 
injection of 0.5 c.c. serum followed in one hour by 1 c.c. Two hours later 
the balance of serum by subcutaneous or intramuscular injection. 
(c) Required Desensitization when the Individual is Not Asthmatic, but the 
Skin Test is Positive and Serum is to Be Given Intravenously or Intrathecally.— 
In these cases desensitization is tedious and time consuming inasmuch as 
it should be more thorough; the following method is employed in the Rocke- 
feller hospital.1 Subcutaneous injection of 0.025 c.c. serum followed at 
one-half-hour intervals by 0.05, 0.1, 0.2, 0.4, and 1 c.c. If a mild reaction 
follows the last injection, it should be repeated once or twice; if no reaction 
follows the injections may be continued by intravenous injection at one- 
half-hour intervals beginning with 0.1 c.c. and doubling the dose at each 
injection. If a general reaction occurs, as cyanosis, dyspnea, or increased 
rapidity of the heart rate supervenes, the injections should be suspended 
for two to four hours, depending upon the severity of the reaction, and then 
be resumed, starting with the same dose as that producing the reaction. 
After 25 c.c. of serum have been given in these small doses, after a lapse of 
four hours, 50 c.c. may be given, followed by the regular dose six to eight 
hours later. Fortunately, one rarely is called upon to employ this measure; 
at the Rockefeller Hospital it has been found necessary to desensitize in 
this way only 2 or 3 patients in a series of over 150 cases. 
In addition to these attempts at specific desensitization medicinal means 
may be employed for the prevention of shock as described below. 
(d) Required Desensitization when the Individual is a Horse Asthmatic.— 
The administration of serum by any route of injection is dangerous, and, if 
possible, should be avoided. This is particularly true if skin tests with 
serum are positive; if skin tests with serum are negative the danger is less. 
Fortunately, these cases are uncommon. If, however, an individual requests 
desensitization for the relief of allergic horse asthma, bronchitis, and rhinitis 
or in preparation for the time when serum injections may be required, the 
method described on p. 734 may be employed. But if serum is subsequently 
administered for prophylactic or therapeutic purposes at a later date, the 
injections should always be subcutaneous and started with 0.1 c.c. followed 
at one-half-hour intervals by gradually increasing amounts. Atropin 
sulphate and adrenalin may be administered in addition as precautionary 
measures, the method being described in the following section. 
Medicinal Prophylaxis of Anaphylactic Serum Shock.—Two drugs have 
been found useful in the prevention of anaphylactic serum shock of guinea- 
pigs, namely, atropin sulphate and adrenalin chlorid. 
As first shown by Auer2 and confirmed by Anderson and Schultz,3 Karsner 
and Nutt,4 atropin prevents or relieves the active bronchoconstriction 
1 Monograph No. 7 on Acute Lobar Pneumonia, 1917. 
2 Amer. Jour. Physiol., 1910, 26, 439. 3 Proc. Soc. Exper. Biol, and Med., 1910, 7, 32. 
4 Jour. Amer. Med. Assoc., 1911, lvii, 1023. TREATMENT OF POLLEN ALLERGY (HA Y-FEVER) 
725 
induced by serum anaphylaxis; similar effects were observed by Bredl and 
Kraus,1 with peptone, and by Baehr and Pick,2 with both peptone and hista- 
min. Friedberger and Mita3 were unable to demonstrate any appreciable 
effects of atropin upon serum anaphylaxis. However, Hanzlik and Karsner4 
have again confirmed earlier work and found that atropin in guinea-pigs in 
the dosage of 0.01 mgm. per gram of body weight can completely prevent the 
toxic effects produced by the intravenous injection of beef-serum and peptone. 
The effects of adrenalin are more complicated and not so certain, depend- 
ing in part on the functional state of the bronchiolar musculature and in 
part on the time of administration, since the action is of short duration. 
Hanzlik and Karsner found that adrenalin in dosage of 0.0005 c.c. of 1 : 10,000 
per gram of body weight intravenously prevents death from true anaphy- 
lactic shock in guinea-pigs. Pelz and Jackson5 also found adrenalin to relax 
the constricted bronchi in anaphylactic shock if injected early. 
In the prophylaxis of serum anaphylactic shock of man atropin sulphate 
in dose of 1/50 grain may he given an adult by subcutaneous injection about 
one-half hour before the serum is administered. Adrenalin chlorid in dose 
of 1 c.c. of a : 1000 dilution should be in readiness for injection if necessary. 
Sodium chlorid has been found by I.anger,6 Brodin, Richet, and Girons7 
as a prophylactic for anaphylaxis in the lower animals, the latter adminis- 
tering 0.8 gm. per kilogram of body weight to dogs. More recently Sicard 
and Parof8 have advised the intravenous injection of human beings with 
1 gm. of sodium carbonate in 30 to 40 c.c. of water fifteeen minutes before 
the injection of serum, as a means for preventing precipitation in vivo, 
which they maintain produces anaphylactic shock. 
According to Besredka ether narcosis reduces the tendency to anaphy- 
lactic shock as discussed in a previous chapter. Under very exceptional 
circumstances it may be advisable to reduce the chances of anaphylactic 
shock in human beings by light etherization. 
The treatment of serum anaphylactic shock has been previously de- 
scribed on p. 662. 
TREATMENT OF POLLEN ALLERGY (HAY-FEVER) 
Soon after the allergic nature of hay-fever was established on a scientific 
basis Noon9 in 1911 reported the results of his investigations in the laboratory 
of Sir Almroth Wright on specific treatment with subcutaneous injections 
of pollen extracts in minute doses. He injected extracts of timothy pollen 
and controlled the doses by an ophthalmo reaction. At the same time and 
independently Koessler’0 was working along the same lines. Curtis,11 Dun- 
bar,12 and others had previously used plant and pollen extracts, but none had 
achieved results sufficiently promising or reliable to encourage a continuation 
of the work. Since then numerous reports have been made on the desensiti- 
zation treatment of hay-fever by injection of pollen extracts by Freeman,13 
1 Zentralb. f. Physiol., 1910, 24, 258. 
2 Arch. f. exp. Path. Pharm., 1913, lxxiv, 65. 
3 Ztschr. f. Immunitatsf., 1911, 11, 501. 
4 Jour. Pharm. Exper. Therap., 1920, 14, 425. 
5 Jour. Pharm. Exper. Therap., 1918, 11, 173. 
6 Miinch. med. Wchn., 1912, 59, 2545. 
7 Rev. d. med., Paris, 1920, 37, 7. 
8 Bull. d. 1. Soc. med. d. H6p., 1921, 45, 60. 
9 Lancet, 1911, 1, 1572. 
10 Forchheimer’s Therapeusis of Internal Diseases, 1914, 5, 671. 
11 New York Med. News, 1900, 37, 16. 
12 Deut. med. Wchn., 1911, 37, 578. 
13 Lancet, 1911, 2, 814; ibid., 1914, 1, 1178. 726 
TREATMENT OF HUMAN ALLERGIES 
Clowes,1 Lovell,2 Lowdermilk,3 Ulrich,4 Manning,5 Cooke,6 Goodale,7 Hitchens 
and Brown,8 Scheppegrell,9 Walker,10 and others. 
The Etiology of Hay-fever.—Hay-fever or “rose cold,” as ordinarily 
understood, is caused by the pollens of different grasses and plants. The 
effects are due primarily to three causes; (a) Sensitization to the proteins 
of these pollens; (b) direct irritation of the conjunctiva and the mucosa of 
the respiratory tract by the pollens, and especially those of the ragweed 
class, which are furnished with spicules, and (c) bacterial infection and 
especially in ragwreed pollinosis, owing to direct injury of the involved 
mucous membranes and subsequent infection. In addition to these some 
pollens apparently contain toxic substances, the hay-fever “toxin” of Dun- 
bar. Too much work has been done with these substances to deny that 
they may be present and even contribute to the lesions and symptoms of 
hay-fever, but the essential nature of hay-fever appears to be an allergic 
sensitization to the proteins of various pollens. It is well known that 
solutions of pollen proteins become toxic upon hydrolysis, and it is possible 
that a similar change may take place on the mucous membranes during the 
hay-fever season with the production of toxic substances. 
Scheppegrell11 has divided these pollens into two classes: (a) The spicu- 
lated, low in protein and causing direct hay-fever, the severity of the attack 
and its duration depending mainly on the number of pollen grains in the 
atmosphere and the length of the spicules. The ragweeds form the type 
and the principal cause of this form of hay-fever. (b) The unspiculated, 
high in protein and causing indirect hay-fever, the severity of the attack 
and its duration depending on the amount of protein and number of pollens 
in the atmosphere. The grass pollens are a type. 
A very large number of pollens have been identified as causing hay- 
fever and this greatly complicates diagnosis by skin tests and treatment by 
specific desensitization. Scheppegrell states that pollens without spicules 
and with an inappreciable amount of protein are innocuous, but there 
remains a discouragingly large number which may cause hay-fever. Walker 
states that these vary in different parts of the country and that it is advisable 
to study with particular care the hay-fever producing vegetation in different 
localities in order to secure pollens for diagnosis and treatment. In New 
England he found that the chief causes were the pollens of ragweed, timothy, 
and June grass; occasionally rose, red top grass, and various trees. 
In addition to the pollens of grasses and plants as causes of hay-fever, 
Walker has observed 12 patients who were sensitive to the pollens of trees: oak, 
maple, willow, pine, poplar, ash, and to apple blossons. These pollinate 
from April to early June. In addition to these Walker believes that sub- 
stances from the leaves of trees, like the fine hairy growths on the under 
surfaces of the willow and plantain leaves, may cause hay-fever. 
Selfridge12 has made a study of the hay-fever pollens in California, and 
Watson and Kibler13 for Arizona and the southwest. The hay-fever seasons 
1 Proc. Soc. Exper. Biol, and Med., 1912-13, 10, 69. 
2 Lancet, 1912, 2, 1716. 
3 Jour. Amer. Med. Assoc., 1914, 63, 141. 
4 Jour. Amer. Med. Assoc., 1914, 62, 1220. 
5 Jour. Amer. Med. Assoc., 1915, 64, 655. 
6 Laryngoscope, 1915, 25, 108. 
7 Boston Med. and Surg. Jour., 1915, clxxiii, 42. 
8 Jour. Lab. and Clin. Med., 1916, 1, 457. 
9 New York Med. Jour., 1918, 1016; ibid., 1919, cix, 793; Medical Record, 1918, 141. 
10 Arch. Int. Med., 1921, 28, 71. 
11 Jour. Amer. Med. Assoc., 1916, 67, 861. 
12 Californ. State Jour. Med., 1918, 16, 164. 13 Jour. Amer. Med. Assoc., 1922, 98, 719. TREATMENT OF POLLEN ALLERGY {HAY-FEVER) 
727 
vary in different sections of the country and different plants are responsible. 
These may be summarized as follows: 
' In the East: Principally the pollens of grasses, as 
timothy, June, and red-top. 
In the Middle West: June grass (blue grass) and 
sweet vernal grass. 
In all sections: The pollens of early flowering trees 
, (poplar, birch, etc.). 
(a) Spring hay-fever 
(April, May, June, and July) 
(b) Summer hay-fever 
(July, August, and September) 
Goose foots, amaranths, and docks. 
(c) Fall or autumnal hay-fever 
In the East: Principally ragweed. 
West of Rocky Mountains: Principally the artemisias. 
Southwest: Amaranths. 
East of the Mississippi river timothy pollen is the usual cause of spring 
fever and ragweed pollen the usual cause of fall or autumnal fever. 
Why the majority of human beings and apparently the lower animals 
escape hay-fever is unknown. The unfortunate sufferers do not usually 
present anatomic defects or predisposing pathologic changes to account 
for their affliction. Evidently a disposition to acquire the disease is in- 
herited, as claimed by Vander Veer and Cooke; certainly the majority of 
patients exhibit a peculiar tendency to vasomotor disturbances of the skin 
and mucous membranes to different irritants ordinarily harmless for the 
majority, and in many instances other members of the family are similarly 
afflicted. 
The Prevention of Hay-fever.—As Scheppegrell1 has stated for the 
American Hay-fever Prevention Association, it is important to educate the 
public on the relation of weeds and plants to hay-fever for the purpose of 
securing co-operation for their removal; at present neglect and careless 
cultivation are a source of disease and discomfort to a large class of sufferers. 
Undoubtedly a campaign of this kind would produce immediate results, 
even if it did nothing more than to teach recognition of the ragweed followed 
up by the removal of these as far as possible from vacant lots and gardens 
in different communities. 
Patients may help themselves by avoiding proximity to the vegetation 
to which they are sensitized, and by means of skin tests the physician can 
render great aid in determining the pollens to which sensitization exists. 
Patients who are sensitive to ragweed should be warned not to smell of 
goldenrod, golden glow, sunflower, poppy, aster, chrysanthemum, and the 
like that pollinate during the ragweed season. Those sensitive to grasses 
should avoid contact with clover, lilies, daisy, dandelion, rose, lawn grass, 
orchard grass, corn, and the like that pollinate during the grass season 
(Walker). 
Removal to a locality where the pollens do not occur, as the Adirondacks 
and White Mountains for the eastern states, is usually followed by imme- 
diate relief. A sea voyage or residence in a high locality with dry mountain 
air during the hay-fever season is particularly desirable, but only a small 
percentage of sufferers can avail themselves of these means of escape. Some- 
times a sojourn at the seaside brings relief and particularly if the neighboring 
country is free of ragweed. 
Value of Preseasonal Desensitization.—Desensitization by means of a 
series of subcutaneous injections of an extract of the pollen or pollens to 
which sensitization exists is the only means offering hope for relief for the 
majority of hay-fever sufferers. 
1 Jour. Amer. Med. Assoc., 1916, 66, 707. 728 
TREATMENT OF HUMAN ALLERGIES 
The results obtained by preseasonal desensitization have been fairly 
satisfactory. It would appear that many patients are helped in that the 
seasonal attack is less severe, others are not helped at all, and a few are 
apparently entirely relieved for the season, especially those who have had 
slight rather than severe attacks. The author believes that the principle 
of desensitization is correct and that as technic is improved upon the results 
will be more satisfactory. 
Unfortunately, many difficulties are encountered. In the first place, 
desensitization of the acutely and severely sensitized individual cannot be 
accomplished by a few injections of the pollen allergen; an error in the past 
has been dependence upon six or less injections. I am convinced that a 
long series of injections are necessary and that for the prevention of autumnal 
fever ordinarily starting about August 15th, desensitization should begin 
in January. In severe cases and especially those complicated by asthma, 
desensitization should be continuous for a year or two to bring about gradual 
relief and possibly complete desensitization. 
Second, it may not be possible to determine by skin tests the nature 
of the sensitization unless a large number of pollens are available; this is 
especially true of the early or spring types of hay-fever which may be caused 
by so many different pollens.. An extensive series of pollens are required 
for skin tests and for the preparation of extracts for injection; for this reason 
both diagnosis and treatment can be undertaken in a proper manner by 
only a few physicians. 
Naturally the question arises whether injections of a pollen extract to 
which a patient is not sensitive will induce allergic sensitization and thereby 
prove harmful. According to experiments on the lower animals this appar- 
ently does not occur, but inasmuch as the hay-fever sufferer inherits an 
increased tendency to acquire sensitization, I believe that great care should 
be exercised and that the only pollen or pollens to be injected are those which 
produce positive skin reactions. 
Some writers have claimed that preseasonal desensitization may intensify 
the symptoms of hay-fever during the seasonal attack. Doubtless this may 
occur if too large doses are given just prior to the attack, but I have never 
seen ill effects attributable to preseasonal desensitization. Of course it is 
well known that, owing to climatic and other conditions, the symptoms 
vary in severity in different years, and this is to be borne in mind in this 
connection. 
A question of considerable practical importance concerns the possibility 
of non-specific desensitization to several different pollens by injecting an 
extract of one pollen. A few authors claim that this may occur, and that if 
a patient is sensitive to several pollens, that desensitization to all may be 
effected in part, at least, by a series of injections of an extract of a single 
pollen. I have no accurate data to contribute on this subject, but probably 
this non-specific desensitization is possible to a limited degree with the 
pollens of grasses very closely related botanically. In serum allergy a 
certain amount of non-specific desensitization of animals has been observed 
by injections of a heterologous serum, but the degree of desensitization is 
slight and of short duration. Possibly a similar action may follow with 
pollen allergens, but very probably the degree of desensitization by a heter- 
ologous pollen is likewise slight and of short duration. 
Success in the specific prevention and treatment of hay-fever by desen- 
sitization appears to depend mainly upon the following factors: 
1. Accurate diagnosis of the sensitization. This refers to a complete 
diagnosis, inasmuch as some patients are allergic to more than one kind of TREATMENT OF POLLEN ALLERGY (.HAY-FEVER) 
729 
pollen. Skin tests are undoubtedly valuable in this connection, as so many 
patients do not know exactly on the basis of experience to which pollen or 
pollens they are hypersensitive. 
2. Prolonged desensitization with active allergens. Deteriorated allergens 
are to be avoided. Doses must be adjusted to bring about desensitization 
without producing ill effects. 
Inasmuch as the number of sufferers is large, means must be provided 
for affording relief by desensitization. For this purpose I believe that the 
manufacturing biologic laboratories should co-operate with the medical 
profession along these lines: 
(a) Prepare an extensive number of pollens from vegetation in different 
parts of the country and render these available for skin tests. 
(b) Be prepared to put up special antigens for desensitization for indi- 
vidual patients upon request by physicians. 
In this manner the pollens may be preserved in powder form against 
deterioration and the antigen made up fresh from time to time and according 
to the needs of individual patients instead of trying to meet the situation 
by stock allergens which have not yielded heretofore the best possible 
results. 
Preparation of Pollen Allergens.—Plants are collected during their 
season and as soon as some of the pollen pods begin to open. The pollen 
granules are collected and dried. In this condition pollens do not deterior- 
ate; according to Goodale they have been kept twenty-five to thirty years 
without any loss in antigenic activity. 
Wodehouse1 has described the following method for collection: “The 
flower heads are stripped from young plants, just coming into bloom, and 
after being allowed to become almost dry, are crushed in a mortar with 
several volumes of carbontetrachlorid. After thoroughly macerating, the 
liquid is strained off through muslin. Most of the pollen, liberated by 
crushing from the anthers, passes with the CCB through the muslin and can 
be collected on filter paper and be washed with fresh CCI4, or with the same, 
after filtering out the pollen, to remove what pollen remains, after the first 
washing. 
“The CCI4 can be used several times, or until it becomes impregnated 
with oils, etc., when it should be replaced or distilled to remove the impurities. 
“This method is very rapid and was found to yield many times more 
pollen than any other known to the writer. Experience has also shown that 
this treatment in no way interferes with the anaphylactic properties of 
extracts prepared from pollen procured in this way; but in order to get the 
best results only young plants, taken when the flowers are just opening, 
should be used, and these collected away from dusty roadsides.” 
After the pollens have been collected and dried different methods may 
be employed for the preparation of the material for skin tests and injection. 
Dunbar, Noon, and Freeman extracted the pollen with distilled water, aided 
by repeated freezing and thawing; Clowes precipitated the pollen with 
acetone and extracted with water. Koessler employed 8.5 per cent, salt 
solution followed by precipitation with 10 volumes of alcohol. Goodale 
soaked the pollens in 15 per cent, alcohol. 
Koessler has emphasized that water extracts were unstable and likely 
to deteriorate in a few weeks. Clock2 has advocated extraction with a 
solvent composed of 66§ per cent, glycerol and 33-§- per cent, saturated 
sodium chlorid solution. 
1 Boston Med. and Surg. Jour., 1916, clxxiv, 430. 
2 Jour. Infect. Dis., 1917, 21, 523. 730 
TREATMENT OF HUMAN ALLERGIES 
An excellent method is to collect the granules either mechanically or 
by the method of Wodehouse and macerate them in 10 per cent, sodium 
chlorid solution, which has been rendered faintly alkaline with sodium 
carbonate. The extract is now filtered, the filtrate dialyzed free of salts, 
precipitated with acetone, and the precipitate dried with anhydrous acetone. 
The powder so obtained keeps well and may be employed for skin tests 
and the preparation of allergen for subcutaneous injection as follows: 
For these purposes the method of Walker is satisfactory: To 0.5 gm 
the dry pollen is added 44 c.c. of sterile physiologic saline solution and the 
mixture is shaken thoroughly at frequent intervals for twenty-four hours, 
after which 6 c.c. of absolute ethyl alcohol are added. The mixture is again 
thoroughly shaken for twenty-four hours, centrifugalized at high speed, and 
the supernatant fluid removed and saved. This 1 : 100 solution of pollen 
extract is used as stock, and higher dilutions, as 1 : 500, 1 : 1000, 1 : 5000, 
and 1 : 10,000, prepared as required with a diluent of 12 per cent, alcohol in 
saline solution. These solutions are used for skin tests and treatment, 
and with the addition of a small crystal of thymol are said to keep for many 
months in a cool place. 
A second method is to grind 1 gram of the dried pollen with powdered 
glass until few or no intact granules remain. This mass is then extracted 
with 100 c.c. of N/100 sodium hydroxid in saline solution containing 0.5 per 
cent, phenol by shaking mechanically for two hours followed by three days 
in a refrigerator with occasional shakings, and finally by several hours in a 
shaking machine. The material is then filtered through paper and finally 
a Berkefeld N or Mandler filter. Coca1 employs an alkaline saline solution 
containing 0.4 per cent, phenol described on p. 681 for preparing extracts 
for skin tests and treatment and has described methods for the collection 
of pollens. 
These extracts may be standardized according to the content of Kjeldahl 
nitrogen as suggested by Cooke and Vander Veer. For routine use five 
strengths have been employed containing, respectively, 0.01, 0.1, 0.5, 1, and 
10 mg. of nitrogen per 100 cubic centimeters. 
Standardization may be effected by the complement-fixation test em- 
ploying the immune sera of rabbits, which have been injected with a gradually 
increasing amount of pollen extract. A unit of antigen is the smallest 
amount giving complete fixation with a given amount of serum. A pollen 
unit is equivalent approximately to 1/20 of a unit of antigen. 
For skin tests about 60 pollen units may be employed (0.01 c.c. of a 
solution of proper strength). For desensitization treatment begins with 
the injection of about 2 units and ends with 600-1000 units. 
Walker’s Method of Preseasonal Desensitization.—Walker advises 
skin tests to determine the degree of sensitization by testing with the 1 : 100, 
1 : 500, 1 : 1000, 1 : 5000, and 1 : 10,000 dilutions of pollen described above; 
treatment is begun with the strongest dilution which fails to give a skin 
reaction. With the majority of patients he has found a more or less positive 
reaction with the 1 : 5000 dilution and begins treatment with 0.1 c.c. or 
0.2 c.c. of the 1 : 10,000 dilution. Treatments are given subcutaneously 
once a week and each week the amount of the extract is gradually increased, 
so that as the treatment progresses, stronger and stronger dilutions are used, 
until with fourteen injections a dose of 0.25 c.c. of the 1 : 100 dilution is 
reached. 
Author’s Method of Preseasonal Desensitization.—As recently empha- 
sized by Vander Veer2 pollen extracts employed for desensitization and 
1 Jour. Immunology, 1922, 7, 163. 
2 Jour. Immunology, 1922, 7, 113. TREATMENT OF POLLEN ALLERGY (HAY-FEVER) 
731 
treatment of hay-fever are frequently too weak for the majority of patients 
and that larger therapeutic doses and more active extracts should be em- 
ployed; the writer is in thorough accord with these opinions and employs 
the following plan of treatment: 
1. Skin tests are first conducted with powders of various purified pollens 
prepared as described above for the purpose of determining the pollen or 
pollens to which sensitization exists. When this has been used tests are re- 
peated with 1 : 100, 1 : 500, 1 : 1000, 1 : 5000, and 1 : 10,000 dilutions of the 
pollen to determine the size of the first dose, which is 0.1 or 0.2 c.c. of the high- 
est dilution just failing to produce a reaction. In cases of autumnal fever 
which are so frequently due to ragweed or when the history and previous 
tests indicates this, the skin tests are done at once with the dilutions. 
2. The injections for desensitization are given subcutaneously once a 
week. When one pollen is employed the series is of 24 injections and is 
started about February 1st in order that all may be given by August 1st in 
the case of autumnal fever. If two or more pollens are employed, a series 
of 28 injections are given beginning in the early part of January. 
In spring fever the injections are started in October or November in 
order that the series may be completed by the early part of May. 
3. The proper protein or allergen is purchased or prepared for the patient 
by adding 0.1 gm. of the pollen yielding a positive skin reaction, to 8.5 c.c. 
sterile salt solution containing 0.25 per cent, tricresol. This mixture is 
extracted in an incubator for two days with frequent shakings; 1.5 c.c. 
absolute alcohol are now added and the extraction continued at room tem- 
perature for two days followed by mechanical shaking for at least two hours. 
This material is now thoroughly centrifugalized, the supernatant fluid 
removed, cultured for sterility, and stored in a refrigerator as the stock 
1 : 100 dilution. 
If skin tests are positive to two or more pollens, a mixture is made of 
these, the total amount of powder extracted being 0.2 gm. 
4. Dilutions 1 : 10,000, 1 : 5000, 1 : 1000, and 1 :500 are prepared as 
required in rubber-capped vials from the stock 1 : 100 solution, using tri- 
cresolized saline solution as the dilutent, as follows: 
1 : 10,000 dilution: 0.1 c.c. of 1 : 100+9.9 c.c. diluent. 
1 : 5000 dilution: 0.1 c.c. of 1 : 100+4.9 c.c. diluent. 
1 : 1000 dilution: 0.5 c.c. of 1 : 100+4.5 c.c. diluent. 
1 : 500 dilution: 1 c.c. of 1 : 100+4 c.c. diluent. 
5. Four injections of the 1 : 10,000 dilution are given: 0.1, 0.2, 0.4, and 
0.8 c.c., followed by 4 doses of each of the 1 : 5000, 1 : 1000, and 1 : 500 
dilutions in the same amounts. These 16 injections are followed by 8 more 
of the 1 : 100 dilution: 0.1, 0.2, 0.4, 0.6, 0.8, 1, 1.5, and 2 c.c. In case 
the solution is prepared by two or more pollens, at least two more injections 
of the 1 : 100 dilution are given, these being in doses of 1 c.c., or 1.5 and 
2 c.c. if well borne. The last dose should be given at least two weeks before 
the expected attack. 
If the treatment begins too late to give the entire series of injections 
before the expected attack, as many as possible should be given, and fre- 
quently the doses may be increased more rapidly, this depending upon 
whether or not reactions are produced. This may be accomplished by 
giving increasing amounts of 1 : 10,000 dilution every day, 1 : 5000 every 
other day, and 1 :500 and 1 : 100 dilutions every three days. 
Skin tests may be repeated after the last injection; if negative with the 
1 : 100 and 1 : 500 dilutions it is likely that no symptoms will develop 
during the season. 732 
TREATMENT OF HUMAN ALLERGIES 
The series should be repeated the following one or two years and especially 
in the asthmatic types. The duration of desensitization varies with different 
persons. Whenever the skin tests are positive to a 1 : 100 dilution of pollen, 
a series of desensitizing injections should be given. 
Seasonal Treatment.—Some patients are apparently relieved during the 
attack by subcutaneous injections of the pollen extract, but great care 
must be exercised against injecting too large doses and especially the first 
dose. Skin tests should be conducted with 1 : 1000, 1 : 5000, 1 : 10,000, 
and 1 : 20,000, dilutions and treatment begins with the subcutaneous injec- 
tion of 0.1 c.c. of the highest dilution failing to produce a skin reaction 
(usually 1 : 20,000). Subsequent injections should be increased, but the 
amounts vary with individual patients; the desired effect is relief from the 
symptoms. As a general rule the relief is of short duration and the injections 
may be cautiously given every three to five days. 
A host of drugs have been advised. Adrenalin chlorid is probably best 
for affording relief from the nasal congestion. Two drops of the following 
mixture may be instilled into each nostril two or three times daily with a 
medicine-dropper or applied with an atomizer: 
Adrenalin chlorid (1 :1000) 6 c.c. 
Rose-water 24 c.c. 
Antipyrin 1 gm. 
Adrenalin chlorid (1 : 1000) may be injected intramuscularly in dose of 
0.5 c.c. for the relief of asthma, but the effects are of short duration. The 
subcutaneous injection of atropin sulphate may also prove helpful. 
Treatment with Bacterial Vaccines.—During the hay-fever attack 
treatment with subcutaneous injections of autogenous bacterial vaccines 
are sometimes beneficial, and especially in ragweed hay-fever. Lowdermilk,1 
Strouse and Frank,2 and Medalia3 have noted beneficial effects and Walker 
states that autogenous nasal vaccine or a mixed streptococcus stock vaccine 
given during the hay-fever attack benefits somewhat an occasional patient. 
Autogenous vaccines are preferred; cultures should be made from both 
sides of the nose and preferably on blood-agar in order to facilitate the 
growth of pneumococci and streptococci if present, as well as the staphy- 
lococci, which usually are present. The author prepares a mixed vaccine 
containing approximately 2,000,000,000 per cubic centimeter; the first dose 
is 0.1 c.c. by subcutaneous injection. Subsequent injections are gradually 
increased and given at intervals of five to seven days. The injections may 
be given together with pollen extract or at alternate times. A further 
description of vaccine therapy is given in Chapter XL. 
The preseasonal use of bacterial vaccines has been used by Strouse and 
Frank with benefit in some cases. 
Treatment with Sera.—Within recent years treatment with Dunbar’s 
antiserum (pollantin) and Weichhardt’s cattle serum (graminol) have fallen 
into disuse. Dunbar’s serum is prepared by the immunization of animals 
with pollen extracts and is marketed in various ways, of which a dried serum 
diluted with lactose and a fluid serum, are mostly employed. 
The serum is used locally and usually just before and during the attack 
for the relief of symptoms. A small amount of the powder is dropped into 
each eye and insufflated into both sides of the nose. Large amounts are 
irritating and to be avoided. Best results have been reported by a daily 
1 Jour. Amer. Med. Assoc., 1914, 63,441. 
2 Jour. Amer. Med. Assoc., 1916, 66, 712. 
3 Boston Med. and Surg. Jour., 1916, 175, 201. TREATMENT OF ALLERGIC ASTHMA AND VASOMOTOR RHINITIS 
733 
dose of the powder into both sides of the nose several days before the expected 
attack. 
The early reports on the use of this serum were very enthusiastic. The 
German Hay Fever Association reported a full success in about 25 to 35 
per cent, of all cases, relief in 25 per cent., and no benefit or aggravation in 
40 to 50 per cent. 
As stated, “graminol” is nothing more than the serum of cattle collected 
during the flowering of grasses. Apparently it has yielded as good results 
as pollantin. 
Both of these sera were proposed for the neutralization of a supposed 
pollen toxin. As previously stated, pollens apparently contain toxic sub- 
stances for which normal or immune serum may be antagonistic, but the 
chief effects are probably due to allergic sensitization and sera apparently 
confer little or no protection to the sensitized cells. 
TREATMENT OF ALLERGIC ASTHMA AND VASOMOTOR RHINITIS 
As stated on p. 653 a large number of different protein substances may 
cause allergic asthma; of these, the proteins of horse, cat and dog hair and 
dandruff and goose feathers, pollens, bacteria, face and dental powders, and 
certain foods are most important. 
Vasomotor rhinitis or hay-fever-like attacks may be caused by the same 
substances or, as shown by Goodale,1 by flowers that have no pollens and to 
which the individual is not sensitive. Some cases of rhinitis with which 
asthma may be associated, are caused according to Walker, by such 
mechanical agents as room dust, hay dust, and talcum; by chemical agents 
as soap-powders, lye, and ammoniacal fumes. Also by perfumes and other 
odors, as well as by sudden changes of temperature and especially upon 
retiring and arising. 
Skin Tests as Guide to Treatment by Desensitization.—The first essen- 
tial to treatment is accurate diagnosis, and while this may be possible by the 
history alone, skin tests have proved of particular value in this connection. 
When skin tests are employed as a guide to treatment by desensitization, 
it is essential to use dilutions for the purpose of determining the degree of 
sensitization. Diagnostic tests may be conducted with the commercially 
prepared allergens or with the alcoholic extracts of different proteins as 
described on p. 681; those producing positive reactions should then be 
retested, according to Walker, with dilutions of purified proteins as 1 : 100, 
1 : 1000, 1 : 10,000, 1 : 100,000, and 1 : 1,000,000. 
Not infrequently a patient will react to more than one protein, but 
Walker is of the opinion that only that protein need be used for desensitiza- 
tion which yields a positive reaction with a dilution of 1 : 1000 or higher. 
Treatment.—This may be of two kinds: (1) The removal of the 
exciting cause if this is possible; for example, the disposal of the cat if it is 
the exciting agent, or of feather pillows, omission of the food allergen from 
the diet, etc. (2) Desensitization by weekly injections of an extract of 
the protein or proteins. 
Exclusion of the exciting agent or agents is of the utmost importance 
and may be sufficient. When this cannot be accomplished or is incomplete, 
as, for example, in horse asthma when street dust may carry sufficient 
allergen, desensitization is indicated. 
We are greatly indebted to Walker2 for his brilliant investigations and 
1 Boston Med. and Surg. Jour., 1918, 179, 293. 
2 Jour. Med. Research, 1917, 36, 423. Archiv. Int. Med., 1918, 22, 466; ibid., 1919, 23, 
220. Amer. Jour. Med. Sci., 1919, 157, 409. 734 
TREATMENT OF HUMAN ALLERGIES 
pioneer work in this field. His results have been very encouraging and 
most favorable in the treatment of all forms of allergic asthma with the 
possible exception of the food asthmas. 
Pollen Asthma (Hay-fever).—The plan of specific treatment is that 
described on p. 730. It is advisable to determine by quantitative skin 
tests the degree of sensitization and commence treatment with 0.1 c.c. of 
the highest dilution failing to give a positive reaction. These patients are 
usually extremely sensitive and desensitization requires much patience on 
the part of both physician and sufferer. I endeavor to give a course of 
28 to 30 injections as previously described, and always advise a repetition 
of the series each year for two or three years. 
Horse, Cat, Dog, and Fowl Asthma.—Quantitative skin tests are con- 
ducted with the purified proteins of the substance or substances yielding 
positive reactions in the preliminary tests and against which desensitization 
is to be tried if contact with the effluvia of the animal cannot be eliminated. 
Only such proteins are employed as yield positive reactions in dilutions 
of 1 : 1000 or higher unless there is only one protein to deal with, in which 
case it is used even if positive only in 1 : 100. As a general rule, however, 
sensitization to this degree is not accompanied by asthma or rhinitis. 
A stock 1 : 100 dilution of the protein in sterile trieresolized saline solu- 
tion is prepared from which higher dilutions are made as required. As a 
general rule treatment beigns with a 1 : 100,000 dilution, the injections being 
given subcutaneously at intervals of one week in ascending doses as follows: 
0.1, 0.2, 0.4, 0.6, and 0.8 c.c. Sometimes the same dose must be repeated 
once or twice if severe local or asthmatic reactions are produced. 
These injections are followed by five more of a 1 : 10,000 dilution and 
these in turn by 1 : 5000, 1 : 1000, and finally 1 : 100 dilutions. 
Skin tests should be conducted at intervals and the desensitizing injec- 
tions continued until negative with 1 : 100 dilutions. 
The duration of desensitization cannot be stated; Walker has reported 
cases free of symptoms for periods up to two years, but very probably the 
duration will be longer depending upon the thoroughness of desensitization. 
Horse-serum Asthma.—Sometimes a horse asthmatic will yield positive 
skin reactions to horse-serum as well as to the proteins of the hair and dan- 
druff. Desensitization may be advisable as part of the desensitization to 
the horse in general and in relation to serum therapy. 
Skin tests should be conducted with dilutions of normal horse-serum as 
1 : 100, 1 : 1000, 1 : 10,000, 1 : 100,000, and even 1 : 1,000,000. Treatment 
begins with 0.1 c.c. of the highest dilution failing to produce a skin reaction. 
These are prepared of sterile normal horse-serum diluted with sterile 
physiologic saline solution containing 0.25 per cent, tricresol. 
The doses are gradually increased and not infrequently it is necessary 
to repeat the same dose several times until it no longer produces an unusual 
local reaction or attack of asthma. All injections are given subcutaneously 
at intervals of one week. After a dose of 0.8 c.c. of a dilution is reached the 
injections are continued with the next lower dilution until the 1 : 100 dilution 
is reached. This requires prolonged treatment, and while relief from rhinitis 
and asthma are to be expected, I do not believe that desensitization can be 
so complete as to render the injection of horse-serum safe for prophylactic 
or therapeutic purposes unless the serum is given cautiously by subcutaneous 
injection and never intravenously. 
Food Asthma.—These cases are usually best treated by omission of the 
exciting food from the diet if this is possible. As a general rule sensitization 
to more than one food usually exists, and with some, as wheat, eggs and milk, TREATMENT OF ALLERGIC DERMATITIS 
735 
which enter into so many different foods, as pastries, gravies, and the like, 
omission may be a difficult matter. 
Desensitization may be essential with certain foods, as egg, because the 
sufferer cannot always escape eating this food and especially when away 
from home when it may be present in foods where its presence is not sus- 
pected. 
The food proteins for skin tests and desensitizing injections should be 
prepared of cooked foods if the allergen is usually ingested cooked, or of 
raw food if the food is generally eaten raw. Wodehouse1 has described a 
method for the preparation of these proteins. 
If desensitizing injections are to be given, quantitative skin tests should 
be made. Only those proteins yielding positive reactions with 1 :500 or 
higher dilutions need be used. Treatment begins with 0.1 c.c. of the highest 
dilution failing to give a positive reaction. Injections are given once a week 
and in gradually ascending doses and dilutions until a 1 : 100 dilution is 
borne in dose of 1 c.c. 
Bacterial Asthmas.—On p. 684 I described a method for conducting the 
skin tests with the bacteria recovered from the individual patient. In my 
opinion these autogenous vaccines are superior to stock bacterial allergens 
both for skin tests and treatment. 
Only those bacteria producing positive reactions are used for desensitiza- 
tion. A stock vaccine of equal parts of the different bacteria is prepared so 
that each cubic centimeter has a total of 2,000,000,000 bacteria. 
Treatment begins with 0.1 c.c. of a suspension of 1,000,000 bacteria 
per cubic centimeter and the doses are gradually increased. The injections 
are given subcutaneously. If no local or general reaction occurs, the suc- 
ceeding doses may be given at intervals of three days until a local reaction 
results, after which the succeeding doses are given at intervals of one week. 
When 0.8 c.c. of this suspension is well borne the series is continued with sus- 
pensions of 10,000,000, 100,000,000, and 1,000,000,000 per cubic centimeter. 
Treatment should be continued until there is complete symptomatic 
relief. Sometimes it is advisable to reculture at intervals and prepare 
fresh vaccines instead of continuing indefinitely with one vaccine. The 
duration of desensitization is variable, depending to some extent upon the 
thoroughness of treatment, but under favorable circumstances is likely 
to be periods of years. 
A special effort should be made to remove foci of infection in the accessory 
nasal sinuses, apices of the teeth, in the tonsils or elsewhere, if these exist 
and are suspected as responsible for the sensitization. I have known of 
cases relieved by the extraction of teeth and drainage of apical abscesses 
caused by streptococci. This subject is further discussed in Chapter XL. 
TREATMENT OF ALLERGIC DERMATITIS 
Desensitization in ivy poison has been reported by Schamberg2 following 
the internal administration of the tincture of rhus toxicodendron. The 
following mixture was employed: 
Tincture of rhus toxicodendron 1 c.c. 
Rectified spirit 5 “ 
Syrup of orange, sufficient to make 100 “ 
This mixture is taken about one month before the ivy season, beginning 
with 2 drops in water three times a day, and increasing by 3 drops each day 
1 Boston Med. and Surg. Jour., 1917, clxxvi, 467. 
2 Jour. Amer. Med. Assoc., 1919, 73, 1213. 736 
TREATMENT OF HUMAN ALLERGIES 
until a dose of 3 teaspoonsful per day is reached. After this 1 tea- 
spoonful per day in water should be taken throughout the season. 
Desensitization is probably not complete, but frequently sufficient to 
prevent ivy dermatitis if direct handling of the plant is avoided. 
During the attack of dermatitis Schamberg administers the tincture as 
above. 
Strickler has reported similar results following the intramuscular injec- 
tion of extracts and Alderson and Pruett1 have reported favorably upon the 
treatment of poison oak by injecting 1 c.c. of extract intramuscularly. 
Almost invariably this was followed by relief of the local symptoms. Han- 
nah2 has reported the successful treatment of ragweed dermatitis by sub- 
cutaneous injections of extracts of ragweed pollen. 
It would appear, therefore, that desensitization may be effectual in the 
prophylaxis and treatment of different forms of dermatitis venenata (poison 
ivy, poison oak, and poison sumac). It is first necessary to apply skin tests 
for diagnosis unless the patient is sure of the plant to which hypersensi- 
tiveness exists. 
Extracts for skin tests and desensitization may be prepared by macerating 
the leaves and extracting with 15 per cent, alcohol in physiologic saline 
solution containing 0.25 per cent, tricresol for a period of several days in the 
proportion of 10 grams of pulp of fresh leaves per 100 c.c. of alcohol. This 
1 : 10 extract is then filtered and is ready for use. 
Skin tests should be conducted with 1 : 10, 1 : 100, and 1 : 1000 or higher 
dilutions and treatment begun by the subcutaneous injection of 0.1 c.c. of 
the highest dilution failing to give a skin reaction. The injections are 
given once per week in ascending doses, as 0.1, 0.2,0.4, and 0.8 c.c. of 1 : 1000 
followed by 0.1, 0.2, 0.4, and 0.8 c.c. of 1 : 500 and 0.1, 0.2, 0.4, and 0.8 c.c. 
of 1 : 100. If skin reactions are positive with the 1 : 1000 dilution, but 
negative with 1 : 10,000, treatment should begin with the 1 : 10,000 dilution. 
The series should start sufficiently early to finish a course of at least 
12 to 16 injections before the season opens. Treatment should be continued 
until there are no skin reactions to the 1 : 10 extract. 
A similar series may be given during the season, but skin tests should be 
conducted in the same manner for the purpose of determining the initial 
dose. 
TREATMENT OF FOOD ALLERGY 
As previously stated on p. 721 treatment consists of: (a) Withdrawal 
of the food or foods from the diet if this is possible; (b) desensitization by 
the internal administration of the purified food protein or the food itself 
in small and ascending amounts or by subcutaneous injection of the purified 
protein when this is possible. 
Treatment by Elimination of Foods.—In some instances the number of 
foods to which a patient is susceptible is so large that treatment by with- 
drawal of the foods from the diet is impossible. It is important with such 
foods as milk and egg to avoid various foods prepared with them, as pas- 
tries, puddings, gravies, etc. 
According to Schloss3 many patients with food allergy recover spon- 
taneously and particularly cases of the congenital type due usually to milk 
or egg. The ill effects become less and less and ultimately the patient is 
able to eat the food in question without discomfort, but this does not occur 
in all cases. 
1 Californ. State Jour. Med., 1921, 19, 188. 
2 Jour. Amer. Med. Assoc., 1919, 72, 853. 
3 Amer. Jour. Dis. Child., 1920, 19, 433. TREATMENT OF FOOD ALLERGY 
This method of treatment is sufficient with such allegies as those caused 
by strawberries, buckwheat, etc., which are easily avoidable. 
In eczemas and other conditions due to food allergies, elimination should 
be prastised as far as possible without producing ill effects caused by deficient 
diet and undernutrition. 
Treatment by Desensitization.—Desensitization by subcutaneous injec- 
tion is complicated by the practical difficulty of securing the food proteins 
in solutions that are sterile and non-irritating. 
These may be prepared of some foods as described on p. 682. Solutions 
(1 : 100) in sterile physiologic saline solution containing 0.25 per cent, 
tricresol may be prepared of some of them; a faint alkalinity to aid solution 
does no particular harm. 
Skin tests should be conducted with 1 : 100, 1 : 1000, and 1 : 10,000 or 
higher dilutions and treatment begun with the subcutaneous injection of 
0.1 c.c. of the dilution failing to give a positive reaction. The injections 
are given once per week in ascending doses, as 0.1, 0.2, 0.4, and 08 c.c. 
followed by 0.1, 0.2, 0.4, and 0.8 c.c. of the next lower dilution, and so on 
until the skin tests become negative to the 1 : 100 dilution and preferably 
to the undiluted powrdered protein applied to an abrasion. 
Sometimes symptoms are produced by an injection; in this case the 
following dose should be smaller and subsequent doses cautiously increased 
Desensitization by injection is particularly serviceable with such foods 
as milk and eggs as cannot be easily avoided. In asthma due to food allergy 
and other incapacitating lesions, as angioneurotic edema and persistent 
urticaria, treatment of this kind may be imperative. 
Desensitization by Feeding.—This form of desensitization is usually 
possible, the dried food or its proteins being administered in capsules in 
small ascending doses. These dried foods should be prepared of cooked 
foods if the food is usually ingested cooked, or of raw foods, if usually eaten 
uncooked. The capsules should be prepared of uncooked milk and eggs 
and the same is true of other foods eaten cooked or uncooked; as a general 
rule uncooked foods will desensitize more efficiently. 
The doses administered should be small enough not to cause ill effects. 
Usually 0.001 or 0.0005 gm. of dried food, as dried egg-white or milk, 
diluted with starch, three times a day in capsules, are safe amounts with 
which to commence treatment. Each day the dose may be increased by 
one capsule until on the seventh day three are taken three times a day. 
In case symptoms are produced the dose is not increased until desensitiza- 
tion has occurred, when the dose is again increased. At the end of each 
week a new set of capsules are taken, each containing double the amount 
of the week previous. For example, if the treatment began with capsules 
containing 0.001 gm. of food, the second set would contain 0.002 gm. per 
capsule, the third set, 0.004 gm., and the fourth set 0.010 gm. per capsule. 
According to Schloss the patient should ultimately be able to ingest 15 to 
30 gm. of the dried proteins or food in twenty-four hours. 
Children under three of age are usually unable to swallow capsules, 
and for them a solution must be used, but owing to direct contact with 
the buccal and pharyngeal mucosa the initial dose must be very small. 
Schloss has recorded a case of marked swelling of the tongue and lips of an 
infant one year of age after taking 1 drop of milk in a teaspoonful of water. 
In this case the initial dose was 1/20 drop of milk, three times a day. 
Skin tests should be conducted from time to time as an index of the 
progress of desensitization. Thorough desensitization requires three to 
six months. After desensitization is accomplished, it is necessary for the 
737 738 
TREATMENT OF HUMAN ALLERGIES 
patient to take regularly a moderate amount of the food in order to prevent 
a return to sensitization. Sometimes urticaria or swelling of the lips develop 
in a desensitized individual, but in such cases it is only necessary to give 
treatment for a few days with larger amounts of food. 
Schloss has treated 24 cases of milk allergy in this manner, with satis- 
factory results; all were markedly sensitive, but at the conclusion of the 
treatment were able to ingest the food without discomfort. 
The duration of desensitization is variable. Schloss has followed 12 
patients who have been desensitized from three to seven years. All of these 
patients are now quite desensitized and can eat, without discomfort, the 
foods to which they were sensitive. 
TREATMENT OF DRUG ALLERGY 
As a general rule these cases require no further treatment than elimina- 
tion. It is well to bear in mind that drugs more or less in daily use, as 
powdered orris root in dental powders or creams, may be the cause of allergy. 
Diagnosis is of primary importance and is greatly aided by skin tests. As 
a rule the medicinal substance is easily eliminated. Most difficulty is 
experienced when drug allergies occur among physicians, dentists, and 
pharmacists who may be required to handle the drug in the course of their 
occupation. 
Desensitization may be effected by subcutaneous injection or internal 
administration and may be required as in the treatment of malaria with 
quinin when the patient is allergic to this drug, as was true of a case re- 
ported by Heran and Saint Girons.1 In their case the internal administra- 
tion of 0.25 gm. of quinin sulphate produced dyspnea, fever, and general 
urticaria. Treatment was given according to the following plan: 
Gram. 
First day 9.45 a. m 
0.005 
11.15 A. M 
0.1 
Second day at first 
0.005 
later 
0.2 
Third day 
0.4 
Fifth day 
0.8 
Sixth day 
0.8 
Seventh day 
1.0 
No symptoms followed any of these administrations of the drug, and 
the patient was cured of malaria. 
A similar plan may be followed for desensitizing against mercurials, 
antipyrin, aspirin, arsphenamin (dermatitis), and other drugs. As a general 
rule skin tests are elicited by 10 per cent, solutions but seldom, in dilutions 
higher than 1 : 1000. If the drug is not irritating, subcutaneous desensiti- 
zation may begin with 0.1 c.c. of the highest dilution just failing to produce 
a skin reaction. Subsequent doses should be given at intervals of a week 
and in gradually increasing amounts. 
As a general rule desensitization can be effected by internal administra- 
tion beginning with 0.001 gm. in capsules and after the same plan described 
above for desensitization in food allergy. 
1 Montpelier Med., 1917, 39, 669. CHAPTER XXXIII 
THE SCHICK TEST FOR IMMUNITY TO DIPHTHERIA 
Natural Antitoxin Immunity to Diphtheria.—It is generally well known 
that not all persons exposed to diphtheria, even by intimate contact, develop 
the disease; likewise not a few persons carry in their upper air-passages 
diphtheria bacilli possessing a high degree of virulence, as shown by the 
guinea-pig inoculation test or by the infection of other persons with whom 
they come in contact, and yet show no clinical evidences of diphtheria. 
Occasionally individuals possessing this high degree of natural immunity 
to diphtheria may show in the tissues of the throat localized areas of super- 
ficial necrosis produced by virulent diphtheria bacilli, but escape the effects 
of toxemia and the production of the typical lesions of the disease. 
Doubtless various local factors of resistance of the tissues of the nose 
and throat afford some degree of protection, but it is now known that the 
chief factor of defense is the presence of natural antitoxin in the blood. 
Infants under one year of age are generally immune and known to escape 
diphtheria even when intimately exposed; in Philadelphia the incidence of 
the disease at this age is about 3 per cent., so that the immunity is not abso- 
lute, but at least 85 to 90 per cent, are known to possess antitoxic immunity 
which, presumably, is inherited from the mother. This immunity, however, 
gradually disappears, so that at the age of two years the majority of children 
are susceptible, and the highest percentage of cases of diphtheria develop in 
children from two to five years. 
After twelve years of age immunity to diphtheria is gradually acquired 
by the majority of human beings so that at adult age from 80 to 90 per cent, 
are immune, the incidence of diphtheria in Philadelphia in individuals over 
eighteen years of age being in the neighborhood of 10 to 12 per cent. This 
immunity is probably due in part to the acquisition of a greater local resist- 
ance of the tissues in the nose and throat as compared with local resistance 
during childhood; also to the development of a natural antitoxin immunity 
probably the result of numerous minor infections with diphtheria bacilli 
capable of stimulating the production of antitoxin without producing well- 
defined clinical evidences of infection. 
Aside from the purely local factors of resistance to diphtheric infection 
of mucous membranes, antitoxin in the blood constitutes the chief means 
of defense; von Behring has determined that the presence of approximately 
1/30 unit per cubic centimeter of blood-serum will protect a human being 
or guinea-pig against diphtheria. This antitoxin in the blood apparently 
neutralizes the toxin produced by virulent diphtheria bacilli and thereby 
prevents the necrosing influence of toxin upon the tissue cells with which 
the bacilli have come in contact; it also neutralizes the toxin absorbed in 
the lymphatic and vascular channels and thereby prevents toxemia and con- 
stitutional reactions. 
Antitoxin, however, is not directly inimical to the diphtheria bacillus, 
but only to its exogenous poison; for this reason it does not directly destroy 
diphtheria bacilli and the latter may be present in the tonsils or other tissues 
with relatively large amounts of antitoxin in blood. The bacilli are probably 
largely destroyed by leukocytes, but antitoxin facilitates leukocytic de- 
739 740 
THE SCHICK TEST FOR IMMUNITY TO DIPHTHERIA 
struction by neutralization of the toxins; in this manner the administration 
of horse-serum antitoxin is accompanied by gradual disappearance of the 
necrotic exudates and bacilli. 
Tests for Antitoxin Immunity to Diphtheria.—For practical purposes 
tests for antitoxin in the blood-serum have proved satisfactory for detecting 
immunity to diphtheria. Schick1 has discovered that if human beings, are 
injected intracutaneously with one-fiftieth the minimal lethal dose of toxin 
for guinea-pigs and no reaction follows, sufficient antitoxin is present in the 
blood to confer immunity. This Schick test has been employed extensively 
during the past seven years, and when properly conducted has demontrated 
its reliability and practical value. 
In addition to this direct or clinical test for antitoxic immunity, it is 
possible to determine the presence or absence of antitoxin in the blood-serum 
by mixing serum and toxin together and after a suitable interval injecting 
a portion of the mixture into guinea-pigs subcutaneously or intracutaneously 
to determine if the toxin has been neutralized. Under proper technical 
conditions the dose of toxin is such as will produce necrosis at the site of 
injection unless neutralized and only diphtheria antitoxin in the serum being 
tested will effect neutralization. Kellogg2 has recently advocated this test 
for practical use in place of the Schick test, and a description of the technic 
is given in Chapter XIII. 
The Toxin for the Schick Test.—The proper preparation of the toxin is 
one of the most troublesome features of the Schick test. The actual amount 
to be injected intracutaneously may vary from about one-thirtieth to one- 
fiftieth the minimal lethal dose for 8- to 10-ounce guinea-pigs; my own 
preference is one-fortieth the minimal lethal dose.3 These amounts of un- 
diluted toxin are too small in bulk for injection, so that it is necessary to 
dilute with sterile saline solution. Park and Zingher dilute with sufficient 
saline to make the dose injected 0.2 c.c.; Schick uses 0.1 c.c., and the wrriter 
has always preferred this amount as being less painful, less likely to produce 
trauma, and just as accurate. 
The toxin must be potent and active and the amount injected neither 
too large (which may produce severe local reactions with necrosis) or too 
small (which may yield falsely negative reactions). The diluted toxin does 
not keep well and deteriorates within forty-eight to seventy-two hours; 
for this reason the toxin must be dispensed undiluted and diluted wTith saline 
solution just before injection. This complicates the practical application 
of the test and Zingher4 has recently discovered that the toxin dispensed for 
the test may deteriorate and become unsatisfactory. The outfit designed 
by Zingher5 and employed by the New York Department of Health is to be 
particularly recommended for conducting a small number of tests; for a 
large number of tests a vial of the undiluted bulk toxin should be used to be 
diluted by the physician according to the given directions. This insures 
against deterioration of toxin, greater accuracy in dilution is obtainable 
and the expense is only slightly greater; for these reasons the bulk toxin is 
preferable when a number of tests are to be made in hospitals and other 
institutions. 
Zingher6 has suggested that the outfit of diphtheria toxin for the Schick 
test shall contain the following maximum and minimum contents of diph- 
theria toxin in order to allow for some deterioration and provide a sufficient 
amount: 
1 Munch, med. Wchn., 1913, lx, 2608. 
2 Jour. Amer. Med. Assoc., 1922, 78, 1782. 
3 Amer. Jour. Dis. Child., 1915, 9, 189. 
4 Jour. Amer. Med. Assoc., 1920, 75, 1333. 
5 Ibid., 1915, 65, 329. 
6 Ibid., 1922, 78, 490. THE CONTROL FLUID FOR THE SCHICK TEST 
741 
“1. If the directions on the outfit call for 0.2 c.c. as the dose for the 
Schick test, the toxin content shall be no more than 1.5 M.L.D. and no less 
than 1.25 M.L.D. to each 10 c.c. of diluent; 0.2 c.c. of the diluted toxin 
shall, therefore, represent no more than 1/33 M.L.D. and no less than 1/40 
M.L.D. for a 250-gram guinea-pig. 
“2. If the directions on the outfit call for 0.1 c.c. as the dose for the 
Schick test, the toxin content shall be no more than 2.5 M.L.D. and no less 
than 2 M.L.D. to each 10 c.c. of diluent; 0.1 c.c. of the diluted toxin shall, 
therefore, represent no more than 1/40 M.L.D. and no less than 1/50 M.L.D. 
for a 250-gram guinea-pig.” 
The M.L.D. or minimal lethal dose of toxin as defined by the Hygienic 
Laboratory, is the smallest amount of toxin that will kill the average 8- to 10- 
ounce guinea-pig in less than four days. This varies with individual animals, 
some dying on the third, others on the fourth day. The technic is described 
in Chapter XIII. 
After the M.L.D. of a well ripened toxin has been accurately determined 
dilutions for the Schick test are made by first diluting the toxin so that each 
cubic centimeter contains 10 M.L.D. From this primary dilution the final 
dilution is prepared in such manner that each 0.1 c.c. (the amount injected 
intracutaneously in conducting the test) shall contain 1/40 M.L.D. The 
following table, slightly modified after that prepared by Zingher, summarizes 
the method: 
METHOD OF DILUTION OF TOXIN FOR SCHICK TEST 
M.L.D. of 
toxin (c.c.). 
Primary Schick Dilution. 
Number 
M.L.D. 
in each c.c. 
of P.S.D. 
Final Scmck Dilution. 
Amount of 
undiluted 
toxin, c.c. 
Amount of 
salt solu- 
tion, c.c. 
Total 
primary 
dilution, c.c. 
Amount of 
primary 
Schick, c.c. 
Amount of 
salt solu- 
tion, c.c. 
Total 
final dilu- 
tion, c.c. 
1/150 
1.0 
14.0 
15.0 
10 
1.0 
39.0 
40.0 
1/125 
1.0 
11.5 
12.5 
10 
1.0 
39.0 
40.0 
1/100 
1.0 
9.0 
10.0 
10 
1.0 
39.0 
40.0 
1/90 
1.0 
8.0 
9.0 
10 
1.0 
39.0 
40.0 
1/75 
1.0 
6.5 
7.5 
10 
1.0 
39.0 
40.0 
1/60 
1.0 
5.4 
6.0 
10 
1.0 
39.0 
40.0 
1/50 
1.0 
4.0 
5.0 
10 
1.0 
39.0 
40.0 
The Control Fluid for the Schick Test.—In conducting Schick tests on 
individuals over six years of age it is advisable to include a control fluid 
in order to detect and properly interpret pseudo or falsely positive reactions 
and combined pseudo and true positive reactions. These pseudoreactions 
are produced in a large degree by the autolyzed protein of the diphtheria 
bacillus present in the culture broth and not removable by filtration; they 
will be described in more detail a little later on in this chapter. 
For the purpose of checking these reactions Zingher1 recommends the 
use of the same toxin as employed for the Schick test, but heated to 75° C. 
for ten minutes which destroys all of the toxin and only a slight amount of 
the autolyzed protein. After heating the toxin the primary dilution is 
made as described for the unheated toxin. The final dilution is then pre- 
pared by diluting 1.25 c.c. of the primary dilution with 39 c.c. of saline 
solution which gives a total of 39.25 c.c. of which the dose is 0.1 c.c. carrying 
a trifle excess of the heated toxin. 
1 Amer. Jour. Dis. Child., 1916, 11, 269. 742 
THE SCHICK TEST FOR IMMUNITY TO DIPHTHERIA 
Method of Conducting the Schick Test.—As previously stated, the 
diluted toxin is injected intracutaneously and not subcutaneously. The dose 
is either 0.1 or 0.2 c.c. according to the dilution which is always stated in 
the directions accompanying the outfit. Dr. Moshage and myself have 
tried a cutaneous test by applying undiluted toxin to an abrasion of the skin 
as in the technic of the cutaneous tuberculin test; reactions occurred after 
twenty-four and forty-eight hours, but these were seldom as definite and 
clear cut as the Schick test. 
A good syringe and a fine needle are absolutely indispensable. The 
writer uses and can strongly recommend the 1 c.c. Record syringe divided 
in tenths and equipped with a tightly fitting gage No. 26 needle. The 
tuberculin syringe is also useful; it is essential that the syringe work properly 
without leakage so that an accurate and measured amount may be injected. 
The toxin is injected into the skin of the right arm either below the 
elbow or about the insertion of the deltoid muscle; the control fluid may be 
injected in the same location on the left arm, but this injection may be 
omitted in children less than six years of age. 
The skin should be cleansed with alcohol, picked up between the thumb 
and forefinger of the left hand, the needle inserted very superficially with the 
eye upward, so that the operator may be sure that the needle is introduced 
far enough and the injection given (Fig. 157). The injection is accompanied 
by a slight stinging pain and leaves a raised anemic spot (Fig. 158), showing 
the pits of the hair follicles. If this elevated area is not produced the injec- 
tion has been too deep and should be repeated. 
If one syringe is used it is better to inject the control fluid first and then 
the toxin, in order to make sure that none of the latter is carried into the 
control injection. 
If more than‘one test is to be given, the same needle may be used for all, 
being wiped off with alcohol or 1 per cent, iodin solution between each 
injection. With proper assistance it is possible to give several hundred 
tests in an hour—much less time is required than for cowpox vaccination. 
A dressing is not required; the needle puncture is so small that infection 
does not occur. The writer has never observed or heard of infection due 
to a properly conducted Schick test. 
The readings may be made twenty-four hours later, if necessary, in order 
to determine whether or not antitoxin is required. If time permits, it is 
better, however, to wait forty-eight hours, which permits very slight trau- 
matic reactions to subside. When the tests are conducted in conjunction 
with T-A immunization in which there is no necessity for haste in reading 
the reactions, it is better to make a final reading on the fourth or fifth day. 
With this interval there is no danger of missing the positive reactions; they 
are likely to be more clearly defined. Furthermore, the pseudoreactions 
will have faded to a large degree and the occasional pseudo and true com- 
bined reactions will be more clearly seen and differentiated from the negative 
pseudoreactions. 
The Negative Schick Reaction and Its Significance.—In a negative 
reaction there is no redness and no infiltration because the injected toxin 
has been neutralized by the antitoxin in the blood. If the toxin was po- 
tent and the injection properly given, a negative reaction indicates that 
the individual has at least 1/30 unit of antitoxin per cubic centimeter 
of the blood-serum which, according to von Behring and Schick and 
amply confirmed by others, is sufficient to confer immunity against diph- 
theria. In individuals over two years of age the immunity is permanent, 
that is, the antitoxin persists in the blood and subsequent Schick tests Fig. 165.—The Schick Test for Immunity in Diphtheria 
A well-marked reaction thirty-six hours after the intracutaneous injection of one-fortieth the 
minimum lethal dose of a diphtheria toxin diluted to 0.1 c.c. The patient’s blood contained no antitoxin. 
The brownish erythematous area with edematous infiltration of the subcutaneous tissues are charac- 
teristic of the reaction. (Reprinted from the Amer. Jour. Dis. Children.) THE PSEUDOPOSITIVE REACTION AND ITS SIGNIFICANCE 
743 
continue to yield negative reactions. In children under two years, however, 
the immunity (negative reaction) may be only temporary because the inherited 
antitoxin may gradually disappear; for this reason the Schick test is only of 
limited value in children under two years of age and may be omitted alto- 
gether. 
The Positive Schick Reaction and Its Significance.—When the blood 
contains no antitoxin or less than one 1/20 to 1/30 unit per cubic centi- 
meter, the injected toxin is not neutralized and acts as an irritant to the 
skin. In twenty-four hours there is a circumscribed area of hyperemia 
about the site of injection varying in size from a quarter to a silver dollar 
and accompanied by distinct edema (Fig. 165). This reaction reaches its 
height in three to four days and then gradually disappears leaving a defi- 
nitely circumscribed scaling area of brownish pigmentation (Fig. 166) which 
persists for three to six weeks. 
Severe reactions do not occur unless an error has been made and too 
much toxin injected, in which case the reaction may be quite large and 
accompanied by some vesiculation. When properly conducted there is no 
constitutional reaction and no tenderness of the axillary glands. 
A positive reaction indicates susceptibility to diphtheria, although, of 
course, by reason of other factors, not all Schick positive individuals contract 
diphtheria when exposed. A positive reaction in children does not neces- 
sarily mean that the individual will be susceptible for life; it would appear 
that many acquire natural antitoxic immunity during later years, probably 
by reason of numerous minor infections stimulating the production of 
antitoxin. 
The Pseudopositive Reaction and Its Significance.—Pseudopositive 
reactions, first described by Park, Zingher, and Serota,1 are especially apt to 
occur in individuals over five years of age and may be present in as high as 
25 per cent, of school children and adults. 
Park and his associates have explained these reactions on the basis of 
local anaphylactic reactions to the protein of autolyzed diphtheria bacilli 
in the broth. Skin tests conducted by Moshage and the writer2 with sus- 
pensions of detoxicated diphtheria bacilli idiphtherin), have yielded positive 
reactions of a similar kind, indicating that many individuals acquire ana- 
phylactic sensitization to the diphtheria proteins; these reactions, however, 
bear no relation to immunity to diphtheria.3 In addition to this principal 
cause of pseudoreactions, I am of the opinion that trauma and non-specific 
enzyme reactions to the proteins of the broth independent of those contributed 
by the diphtheria bacilli, may aid in the production of these reactions.4 
The pseudopositive reaction can be interpreted occasionally with fair 
accuracy by the trained observer from the true positive reaction; but there 
is usually some doubt as to the absolute certainty of such a reading unless 
the control injection of heated toxin has been given on the other arm for 
comparison. The pseudopositive reaction develops earlier, within six to 
eighteen hours, reaches its height in thirty-six to forty-eight hours, and is 
characterized by a central area of dusky redness about the size of a dime, 
with a characteristic secondary palish areola that shades off into the sur- 
rounding skin, and slight or no edema (Fig. 167). After fading it leaves 
none, or, but very slight pigmentation and scaling of the skin. 
Occasionally and especially among adults, a combined true and pseudo- 
1 Arch. Pediat., 1914, 31, 481. 
2 Proc. Soc. Exper. Biol, and Med., 1916, 13, 89; Amer. Jour. Dis. Child., 1916, 12, 316. 
3 Jour. Immunology, 1916, 1, 334. 
4 Jour. Amer. Med. Assoc., 1915, 65, 144. 744 
THE SCHICK TEST FOR IMMUNITY TO DIPHTHERIA 
positive reaction may be seen. The central area of brownish redness is 
larger and accompanied by distinct edema; this is surrounded by the same 
palish erythema as seen in the pseudoreaction (Fig. 168). Definite brownish 
discoloration and scaling of the skin follows as in the true reaction. If 
heated toxin has been injected as a control, a typical pseudoreaction will 
be observed and the development of these control reactions indicates that 
special caution is necessary in reading and interpreting the reactions excited 
by toxin. When in doubt, it is better to regard the toxin reaction as a true 
positive and err if necessary on this side. 
The Pseudoreaction in Relation to Toxin—Antitoxin Immunization; 
the Necessity of a Control.—Zingher1 has recently emphasized that children 
yielding pseudo-Schick reactions are most likely to develop pronounced 
local and constitutional symptoms after injections of toxin—antitoxin. 
This is to be expected when it is remembered that the pseudoreaction is 
principally due to sensitization to the proteins of the diphtheria bacillus 
contained in the toxin broth. For this reason the control injection is always 
advisable, indeed, is almost a necessity, in Shick tests on individuals over 
five years of age and especially if T-A immunization is to be given; by means 
of the control injection the pseudoreactions are readily detected and given 
special attention in active immunization. The control injection also aids in 
the interpretation of combined pseudo- and true Schick reactions; in such 
individuals the first dose of T-A should be about one-half that ordinarily 
given (0.5 c.c.) in order to reduce the degree of reaction that may follow. 
The Percentage of Positive Schick Reactions and the Influence of Age, 
Race, and Inheritance.—The percentage of positive reactions at different 
ages reported by different investigators and even by the same investigator 
at different times, have varied to a considerable degree. Doubtless these 
variations have been largely due to defective toxin (too much, too little, or 
inactive), errors in injection and interpretation of results; also according to 
locality (urban or country districts), race and social condition, and general 
health. For example, Zingher2 has observed in New York City a much 
higher percentage of positive Schick reactions (as high as 75 per cent.) 
among school children from the homes of the more well-to-do than those 
from the homes of the poorer classes living in closely crowded neighborhoods 
(as low as 16 to 20 per cent.), where the chances of contact infection and 
the development of natural immunity are greater. An unusually low pro- 
portion of positive reactions occurred among the children of Italian parents. 
An exception to this general rule was noted among colored children, the 
percentage of positive reactions being high despite the fact that the majority 
lived in the congested districts. Wright,3 however, observed about the same 
percentage of positive reactions among adult negroes as white adults. 
Park, Zingher, and Serota also found a marked tendency for all children 
of the same family to show a similar Schick reaction, whether it was positive 
or negative. Where variations occurred, the younger children as a rule 
gave a positive reaction and the older children a negative; the reverse 
condition was very rare, except in families with young infants, who often 
gave a temporary negative reaction due to inherited immunity. 
Bearing in mind these modifying factors it is almost useless to attempt 
expressing the percentage of positive Schick reactions at different ages; 
52,000 tests conducted by Zingher among the school children of New York 
varying in age from' six to fifteen years, or thereabouts, have shown positive 
1 Jour. Amer. Med. Assoc., 1922, 78, 1945. 
2 Ibid., 1921, 77, 835. 
* Jour. Infect. Dis., 1917, 21, 265. Fig. 166.—A Fading Schick Reaction Showing Brownish Discoloration and Fine Desquamation. 
Fig. 167.—A Pseudopositive Schick Reaction. 
Fig. 168.—A Combined True and Pseudopositive Schick Reaction. THE PRACTICAL VALUE OF THE SCHICK TEST 
745 
reactions varying from 13.6 to 67 per cent, and pseudoreactions varying 
from 3.9 to 26 per cent. Figures, however, will indicate in a general way 
the influence of age and may be given as follows: 
One to six months 
Per cent, positive. 
15 
Six to twelve months 
50 
One to two years 
55 
Two to four years 
65 
Four to six years 
40 
Six to eight years 
35 
Eight to twenty years 
25 
Twenty to forty years 
18 
Over forty years 
12 
The Influence of Disease Upon the Schick Reaction.—In 1915 Moshage 
and the writer1 called attention to the higher percentage of positive Schick 
reactions occurring among persons with scarlet fever. Zingher2 observed that 
the reactions were twrice as frequent in this disease, slightly more frequent 
during measles, and nearly three times as high during poliomyelitis. Lere- 
boullet,3 however, did not find that measles influenced the reactions. 
The Relation of the Schick Test to Diphtheria Carriers and to the Diag- 
nosis of Diphtheria.—There is no relation between the percentage of carriers 
of diphtheria bacilli and the Schick reaction. An individual may carry 
virulent bacilli in the nose and throat without evidences of diphtheria and 
yield a negative Schick reaction, because of the presence of sufficient anti- 
toxin in the blood to protect against that produced by the bacilli and to 
neutralize that injected into the skin. As previously mentioned, however, 
Park has observed that virulent diphtheria bacilli may produce some inflam- 
matory changes in the tonsils with superficial necrosis in Schick negative 
individuals; this indicates that the bacilli and their toxins may produce 
some tissue changes when so superficially located as to be beyond the neu- 
tralizing influence of antitoxin in the blood. 
Individuals may carry diphtheria bacilli and yield positive Schick reac- 
tions. In the majority of such cases the bacilli are found non-virulent for 
guinea-pigs. For this reason the Schick test is of some aid in the manage- 
ment of cases showing diphtheria or diphtheroid bacilli in nasal and aural 
discharges; if the Schick reaction is positive it indicates that the bacilli are 
very probably non-virulent, for the positive skin reaction indicates the 
absence of antitoxic immunity in which case the patient would probably 
show clinical evidences of diphtheria if the bacilli were virulent. I have 
observed, however, positive Schick reactions among diphtheria convales- 
cents harboring virulent bacilli; doubtless there are other local tissue and 
immunity principles that may protect an individual against virulent diph- 
theria bacilli even though lacking in antitoxic immunity. 
The Schick test, therefore, is not a substitute for the culture method in 
the diagnosis of diphtheria and the detection of carriers. It aids, however, 
in the mangement of carriers as discussed above. The test is not an index 
of the virulence of bacilli recovered in cultures; the guinea-pig inoculation 
test is the only reliable one for this purpose. 
The Practical Value of the Schick Test.—A large number of investigators 
have found the Schick test safe and a reliable and simple means for deter- 
mining the presence or absence of immunity to diphtheria. x\ccidents have 
happened, severe reactions having been produced by the injection of too 
1 Amer. Jour. Dis. Child., 1915, 9, 189. 
2 Amer. Jour. Dis. Child., 1917, 13, 247. 
3 Bull. d. 1. Soc. Med. d. H6p., 1921, 45, 1210. 746 
THE SCHICK TEST FOR IMMUNITY TO DIPHTHERIA 
much toxin; likewise a few have claimed that diphtheria developed among 
those yielding negative reactions, but the great bulk of evidence is favorable 
to the test and indicates unmistakably that when properly conducted the test 
is without danger and of much value. A large literature has accumulated 
among which may be mentioned, in addition to those previously referred 
to, the reports of Weaver and Maher,1 Bundesen,2 Moody,3 Graef and 
Ginsberg,4 Linenthall and Rubine,5 von Groer and Kassowitz,6 Lilly,7 Leet,8 
Koplik and Unger,9 Moffet,10 Conrad,11 de Elizaldi,12 Cowie,13 White,14 Mul- 
sow,15 Blum,16 and others. 
Unfortunately, the practical application of the test is hampered by the 
chances of using defective toxin and the necessary procedures for diluting 
and injecting it; also by the occurrence of pseudoreactions, but these dif- 
ficulties soon disappear with a little experience and especially when bulk 
toxin can be used as required for a large number of tests in schools, hospitals, 
and other institutions. 
The test has its greatest value as a means for detecting those lacking in 
antitoxic immunity and requiring immediate protection by an injection of 
antitoxin when exposed to diphtheria or, in the absence of acute exposure, 
slower but more lasting immunization by toxin-antitoxin mixtures (see 
Chapter XLI). By means of this test a large number of individuals may be 
spared the discomfort and useless sensitization caused by injections of 
horse-serum antitoxin and it has given great impetus to active immunization 
with T-A mixtures. 
The influence of age upon the reaction is to be remembered, that is, that 
a child under two or three years of age may react negatively and later posi- 
tively, when its inherited immunity has disappeared. For this reason young 
children should be retested at intervals to detect susceptibility should such 
develop. After four or five years of age the reaction is greatly stabilized, 
that is, continues to yield negative reactions probably for life, or positive 
reactions, unless the individual is immunized. 
It is not necessary, however, to use the test on children under five years 
in relation to toxin-antitoxin immunization; the majority give positive 
reactions anyhow, and the Schick test may be omitted in this group and T-A 
given to all children between six months and five years. After six years of 
age, however, the Schick test and its control should be conducted. 
Naturally the test has its greatest usefulness under conditions where 
the spread of diphtheria is facilitated, as in the wards for children in hos- 
pitals, in institutions for children and adults as well, among nurses, phys- 
icians, and others brought in contact with diphtheria or diphtheria carriers. 
Cooke17 and others have shown the value of the test among nurses and immuni- 
1 Jour. Infect. Dis., 1915, 16, 292. 
2 Jour. Amer. Med. Assoc., 1915, 64, 1203. 
3 Jour. Amer. Med. Assoc., 1915, 64, 1206. 
4 Jour. Amer. Med. Assoc., 1915, 64, 1205. 
5 Boston Med. and Surg. Jour., September 16, 1915. 
6 Ztschr. f. Immunitatsf., 1914, 22, 404; ibid., 1914, 23, 108. 
7 Boston Med. and Surg. Jour., 1920, clxxxii, 110. 
8 Lancet, January 24, 1920. 
9 Jour. Amer. Med. Assoc., 1916, 66, 1195. 
10 Jour. Amer. Med. Assoc., 1915, 64, 1203. 
11 Jour. Amer. Med. Assoc., 1915, 65, 1010. 
12 Prenza M6d., 1919, 5, 293. 
13 Amer. Jour. Dis. Child., 1916, 12, 266. 
14 Boston Med. and Surg. Jour., 1921, 184, 246. 
15 Jour. Amer. Med. Assoc., 1921, 77, 1254. 
16 Amer. Jour. Dis. Child., 1920, 20, 22. 
17 Amer. Jour. Dis. Child., 1922, 23, 496. THE PRACTICAL VALUE OF THE SCHICK TEST 
747 
zation of susceptibles by injection of toxin-antitoxin mixtures. In the 
writer’s opinion the Schick test should be conducted routinely under these 
circumstances and the positive reactors either actively immunized with 
toxin-antitoxin mixtures or passively with antitoxin, when intimately 
exposed. 
In private practice physicians will do well to acquaint parents with the 
advantages of the Schick test and active imminization and to encourage 
their use. By grouping individuals in private practice a number of tests 
may be done in one day, thereby minimizing the work involved. 
It is very much hoped that state and city Departments of Health will 
facilitate the application of the Schick test and active immunization by 
providing suitable outfits and even trained physicians for conducting the 
tests, especially in institutions and schools. The work being carried out 
in the schools of New York City consisting in conducting the Schick test 
and actively immunizing the positive reactors, should be closely followed, 
as it bids fair to prove a useful and practical means for really reducing the 
incidence of diphtheria. CHAPTER XXXIV 
PRINCIPLES OF ACTIVE IMMUNIZATION—VACCINES IN THE 
PROPHYLAXIS AND TREATMENT OF DISEASE 
While the importance of natural immunity must not be underrated in 
the protection it gives us after bacterial invasion has occurred, this immunity 
is, however, usually relative and seldom absolute, and may afford insufficient 
protection if the invading bacteria are numerous, or particularly virulent, 
or if the natural resistance of the organism is weakened by fatigue, disease, 
or injury. 
Active Immunization.—Usually the best and most lasting immunity is 
that actively acquired, in which our own body cells are stimulated or trained, 
as it were, to produce specific antibodies against the offensive forces of a 
particular bacterium or other pathogenic agent. A well-marked and lasting 
degree of this form of immunity usually follows recovery from many of the 
acute infections, particularly the acute exanthemata, such as smallpox, 
scarlatina, measles, typhoid fever, typhus fever, etc. In other infections, 
such as erysipelas, gonorrhea, and pneumonia, the immunity is less complete, 
of short duration, or apparently absent, and, indeed, a state of hypersus- 
ceptibility to infection may actually follow. 
It is very important, in this connection, to remember that the degree of 
immunity is not necessarily in proportion to the severity of the disease; thus 
a mild infection may be followed by the much-desired immunity, and while 
in general there is not considerable protection without infection, the latter 
does not necessarily imply that a virulent infection, or even the actual 
disease itself, is present, for discoveries have shown that an active immunity 
may be acquired by inoculation with the antigen so modified or attenuated 
that it can stimulate the production of specific antibodies without producing 
the disease or otherwise greatly disturbing the health of the individual. 
Historic.—The facts here detailed are well illustrated in the history of 
vaccination in smallpox and the development of vaccine therapy in general. 
Hundreds of years ago the people of the eastern countries were accustomed 
to expose their children to a mild case of smallpox in order that a similar 
mild infection might be acquired and a lasting immunity thus secured with 
the least danger to life. This practice, however, was not without risk, as 
the mild disease not infrequently became a virulent one. Later the dose 
of infectious agent was decreased by applying the virus to a small abrasion 
on the skin, and the resulting mild but genuine attack of smallpox usually 
conferred the much-desired immunity. But here again the severity of the 
disease was not under control, as the virus occasionally assumed increased 
virulence and induced severer infections than were desirable. Finally, 
Edward Jenner observed and showed experimentally (1796) that when 
cowpox virus is inoculated into the human being, a trivial infection, since 
called “vaccinia,” is induced, and that this is followed by an absolute or 
nearly absolute immunity of many years’ duration against smallpox. In 
other words, the virus, in its passage through the cow, becomes so modified 
that it can no longer produce smallpox, but is still able to stimulate the 
production of the specific antibodies against this disease. Jenner worked 
so hard to establish the truth of this finding that he had little time to devote 
to the mechanism involved in the process. 
748 NOMENCLATURE 
749 
In other words, the work of Jenner was largely empirical, and the ex- 
planation was not forthcoming until many years later, when Pasteur laid 
the basis of scientific immunization by discovering that light, high and low 
temperature, and exposure could so reduce the virulence of a micro-organism 
that while its injection into an animal was practically without danger or ill 
effect, it could still stimulate the protective mechanism of the host and 
induce a high degree of immunity. 
That this could be done was a fact discovered accidentally by Pasteur 
in 1879 while working with the organism of chicken cholera. After an 
absence from home he found on examining his cultures on his return that 
they had become innocuous—that hens could bear without any ill effect 
inoculation of what would formerly have been a lethal dose. The prolonged 
cultivation of the micro-organism had caused its attenuation and Pasteur 
immediately grasped the far-reaching importance of this discovery. He 
conjectured that it might be possible to produce a mild and modified form 
of chicken cholera with a vaccine of the attenuated micro-organism which 
would afford protection to the fowl against the severe form of the disease. 
This proved to be the case, and established the possibility of so modifying 
or attenuating the virulence of a virus or germ that, while its administration 
is not followed by the actual disease, it is capable of so stimulating the body 
cells that the specific antibodies are produced. This discovery formed the 
basis of prophylactic immunization or bacterin therapy in general. 
Following this discovery much work was done, for it appeared that the 
question of prevention of any bacterial disease simply depended upon 
whether the bacterium could be cultivated and so modified or attenuated 
that while its injection would not be followed by disease or other harmful 
effects, it would be capable of causing the production of specific anti- 
bodies. 
Naturally, most of the earlier work was done with the infections of the 
lower animals, and, consequently, most discoveries were directly beneficial 
to them. Pasteur soon devised a method of attenuating anthrax bacilli 
by exposing them to certain temperatures for varying lengths of time, so 
that a vaccine could be prepared that has proved of great value. Later 
the same observer discovered a method of attenuating the virus of hydro- 
phobia by a process of drying, and devised a practical method of prophylactic 
immunization against this disease. In addition to these his vaccines against 
swine erysipelas, symptomatic anthrax, and rinderpest have become well 
known. 
The knowledge gained from a study of the diseases of the lower animals 
and the aid given them has been applied to human medicine with considerable 
benefit not only in prophylactic immunization but also in therapeutics 
(bacterin therapy). The latter application is a more recent discovery, for 
which we are mainly indebted to the researches of Wright, Leishman, Doug- 
las, and their colleagues. 
Nomenclature.—The word vaccine is from the Latin vacca (a cow). 
Cowpox was called “vaccinia,” or the cow disease, and Jenner designated 
protective inoculation against smallpox with cowpox virus as vaccination. 
With true courtesy Pasteur adhered to Jenner’s nomenclature and applied 
the term vaccine to emulsions of dead or attenuated bacteria. This is 
unfortunate and tends to create confusion, as the term vaccine is inseparably 
associated with cowpox virus or lymph. The term bacterial vaccine has 
become widely known, and is used to designate bacterial suspensions pre- 
pared for purposes of immunization. There is no essential difference, 
however, between cowpox vaccine, which contains the modified germ or virus 750 
PRINCIPLES OF ACTIVE IMMUNIZATION 
of smallpox in a diluent of lymph, and a bacterial vaccine containing the 
germ, modified by some physical or chemical agency in a diluent of saline 
solution or bouillon. It is, however, well to reserve the unqualified term 
“vaccine” for cowpox virus and to retain the designation “bacterial vaccine” 
for suspensions of attenuated or dead bacteria. More recently the term 
bacterin has been suggested and applied by S. Solis Cohen to the latter, but 
this would imply an extract of bacteria, as, e. g., tuberculin, which is not 
always the case. 
Some confusion likewise exists as regards the terms serum and vaccine 
therapy. Serum therapy is a process of passive immunization induced for 
either protective or curative purposes by the injection of the blood-serum 
of another animal that has been actively immunized by inoculation with 
bacterial toxins or the bacteria themselves, as, for instance, the injection 
of diphtheria or tetanus antitoxins. Vaccine or bacterin therapy is a process 
of active immunization brought about by the injection of the bacteria or 
their products directly into a patient. Bacterial vaccines that are simple 
emulsions of dead or attenuated bacteria are not, therefore, serums, and the 
indiscriminate use of the two terms is much to be regretted. 
Methods of Preparing Vaccines.—It may be stated that, in general, the 
specific micro-organism or virus used in a vaccine should be modified as little 
as possible, or just sufficient to rob it of its disease-producing power. For 
example, typhoid bacterial vaccine is prepared by suspending the bacilli in 
salt solution and exposing them to just enough heat to modify them so that 
they can no longer multiply. The less modification, the better the vaccine. 
If the exposure is too prolonged or the temperature too high, the vaccinogenic 
power of the micro-organisms is likely to be reduced or destroyed. Therefore 
the nearer the vaccine approaches the fully viable virus or mirco-organism, 
the more potent it will be. The proper preparation of a vaccine, therefore, 
is the first step to successful vaccine therapy. 
Vaccination, using the term in its broadest sense, may be performed for 
prophylactic purposes and curative immunization in the following ways: 
1. The living micro-organism may be inoculated. This is the ideal method, 
but for obvious reasons has not been generally used and is still in the experi- 
mental stage. It is based upon experimental observations made on the 
lower animals that an organism may be so introduced as to render it incapable 
of producing disease, but may, however, stimulate the production of specific 
protective antibodies. Evidence thus far indicates quite conclusively that 
the typhoid bacillus, for example, is unable to produce typhoid fever unless 
it is introduced into the gastro-intestinal tract, and the subcutaneous injec- 
tion of living bacilli, modified only to a slight extent by artificial cultivation, 
is not followed by ill effects and produces a high grade of immunity. Rabies 
vaccine, for example, is sometimes prepared of the living virus greatly 
diluted with saline solution for the initial doses. The principle is a good one, 
i. e., in a vaccine the micro-organism should be modified as little as possible. 
For obvious reasons, however, this method must be thoroughly tried out 
on susceptible lower animals before it is applied .to human beings. 
2. By inoculation with a modified virus or with micro-organisms attenuated 
or modified according to various methods. 
{a) By passing the virus through a lower animal, as the passage of 
smallpox through the heifer when the virus is incapable of producing small- 
pox, although vaccinia confers a specific immunity against smallpox. A 
vaccine for swine erysipelas is prepared in the same manner (Pasteur) by 
passing the bacillus through the rabbit several times, which increases its 
virulence for the rabbit, but decreases it for swine. RELATION OF VACCINE TO IMMUNIZING ACTIVITY 
751 
(■b) By exposing suspensions of micro-organisms to heat. They are usually 
grown on a suitable solid medium, suspended in salt solution, and exposed 
to a temperature at or just above their thermal death-point for just sufficient 
time to kill or attenuate them in so far that they cannot multiply. The 
same result may be secured by longer exposure to a lower temperature. 
To secure a potent vaccine the principle of minimum exposure at the minimum 
temperature should he observed, the question of viability being controlled by 
culturing the vaccine. Most bacterial vaccines are prepared in this manner. 
Usually an exposure of 53° to 60° C. for from one-half to one hour is sufficient, 
and only exceptionally are these limits exceeded. 
(c) By exposing the micro-organism to air and light. The first bacterial 
vaccine (chicken cholera) was accidentally prepared by Pasteur in this 
manner. 
(id) By desiccating or drying the virus. This is one of the methods of 
vaccination in rabies, as the virus contained in the spinal cord of rabbits 
is dried for varying lengths of time, emulsified, and injected. The longer 
the period of drying, the greater the attenuation, and in this manner the 
strength of the vaccine and the progress of immunization are under control. 
(e) By exposing the micro-organism to a high temperature for varying 
lengths of time. Anthrax vaccine, for the immunization of lower animals, 
is prepared in several strengths by exposing suspensions of the bacilli to 
42° C. for varying periods of time. 
(/) By exposing the micro-organisms or their products to certain chemical 
germicides, as in the preparation of anthrax vaccine by Roux, diphtheria 
and tetanus toxins (Behring), in some preparations of tuberculin, and in 
the preparation of rabies vaccine- and ordinary bacterial vaccines by steriliza- 
tion with phenol, etc. 
3. By inoculating with bacterial constiuents, as the soluble toxins, bacterial 
extracts, and products of bacterial autolysis, as in the preparation of Koch’s 
tuberculin T. R., Koch’s old tuberculin, mallein, diththeria and tetanus 
toxins, Rosenowr’s pneumococcus vaccine, Gay’s typhoidin, etc. 
In Chapter XII a method is given for preparing bacterial vaccines, of 
which typhoid vaccine is a type. Special methods of preparing certain 
bacterial vaccines and other vaccines, such as cowpox virus and rabies 
vaccines, are given in this chapter. 
The Relation of the Method of Preparation of Vaccine to Immunizing 
Activity; Living Versus Killed Vaccines.—Barring accidents, the employment 
of a living vaccine would appear to be the most certain way of calling forth 
a maximum output of antibodies; this is indicated by the remarkable and 
unparalleled success of cowpox vaccination as ordinarily conducted with 
living virus. There is at present no satisfactory explanation for this except 
that heat-labile substances destroyed in the ordinary preparation of bacterial 
vaccines have antigenic properties (Smith). Living vaccines are also capable 
of penetrating into deeper tissues, whereas dead vaccines may remain where 
they are deposited. Similarly living viruses are capable of exerting a con- 
tinuous action and of delivering an infinite number of blows, whereas the 
injection of a dead virus produces an interrupted action and deals but a 
single blow. The actual dangers of using a living vaccine, as the possibility 
of it being too virulent and thus producing disease, or some of the symptoms 
as paralysis during the course of antirabic vaccination, or of regaining 
virulence or producing chronic “carriers” preclude their general employment 
in human practice. 
To prevent these accidents ordinary bacterial vaccines are generally 
heated at 56° to 60° C. for one hour for purposes of sterilization, but this 752 
PRINCIPLES OF ACTIVE IMMUNIZATION 
reduces vaccinogenic activity. Crandon,1 in a summary of the literature 
on the use of killed and living vaccines, concludes that unheated vaccines 
have proved superior. In order to avoid the destructive influence of heat 
Casselman2 and more recently Kisskalt3 have advocated the use of phenol- 
killed vaccines, which had previously been employed for dysentery vaccine 
by Gay; Vincent4 has employed ether, and others galactose, sodium fluorid, 
iodin, and other chemical substances. In the preparation of antibacterial 
sera, as antimeningococcus and antipneumococcus serum, killed cultures 
produce an antibody response, but more recently it has been shown that 
smaller doses of living micro-organisms produce a quicker and more pro- 
nounced reaction, and if this experience can be borne out in active immuniza- 
tion in man, it will appear as support for the contention that the superior 
vaccines are those produced with the least possible change in the bacterial 
protoplasm. It cannot be questioned, howrever, that when vaccines are 
prepared in such manner that the bacterial protein undoubtedly undergoes 
some more or less profound chemical change, as in the method of preparation 
advised by Ldffler, which consists in heating the cultures to 120° to 150° C., 
followed by drying or pulverizing, such vaccines possess some degree of 
antigenic activity capable of eliciting both specific and non-specific responses. 
Sommerville5 likewise found that heating vaccines at high temperatures, as 
boiling for fifteen minutes, did not reduce antigenic activity, but Thylor6 
observed that the activity of these was slightly less than with vaccines 
heated at 60° C. Gay and Claypool,7 in an extensive series of experiments, 
have shown that typhoid bacilli flocculated by alcohol and dried at ordinary 
temperatures followed by grinding to a fine powder retain antigenic activity; 
Brown8 has observed similar effects. 
In a study of the immunizing activity of typhoid vaccines prepared after 
various methods Perry and myself9 found that living bacilli were most active 
followed by chemically killed vaccines; heat-killed vaccines proved least 
antigenic for rabbits. 
The subject is one of considerable importance in relation to active im- 
munization both for prophylactic and therapeutic purposes, and shows that 
the method of preparing vaccines is one deserving of more attention. Un- 
questionably the highest degree of active immunization is engendered by 
living micro-organisms as indicated by the lasting immunity following one 
attack of many of the acute infectious diseases and the success of cowpox 
vaccination. On the other hand, the products of bacterial activity, as, for 
example, diphtheria toxin, and dead and thoroughly disrupted bacterial 
cells, possess antigenic activity and being free of certain toxic elements 
may actually prove more antigenic than living micro-organisms as shown 
by the success of vaccine therapy in acute bacterial infections. 
Variation in the Antigenic Activity of Different Microparasites.—While 
the method of preparation of vaccine unquestionably exerts some influence 
upon immunizing capacity, it would appear that the immunity response to 
vaccines of different microparasites likewise varies within wide limits. For 
example, a vaccine of the typhoid bacillus prepared by the usual method 
possesses a high degree of antigenic activity, whereas vaccines of Treponema 
pallidum prepared and administered in exactly the same manner are appar- 
ently followed by the production of only small amounts of demonstrable 
1 In Sanborn’s Surgical After-treatment, Saunders Co., 1910, 763. 
2 Jour. Amer. Med. Assoc., 1915, 64, 328. 6 Lancet, 1915, 2, 150. 
3 Deut. med. Wchn., 1915, 41, 393. 7 Archiv. Int. Med., 1914, 14, 671. 
4 Compt. rend. Acad, des Sci., 1912, civ, 480. 8 Indian Jour. Med. Res., 1913, 1, 589. 
5 Lancet, 1915, 2, 96. 9 Jour. Immunology, 1918, 3, 247. SENSITIZED VACCINES 
753 
antibodies in the blood and a practically negligible degree of actual immunity. 
In other words, vaccines of different microparasites prepared and adminis- 
tered in exactly the same manner stimulate widely different degrees of 
immunity, due either to a difference in the degree of antigenic activity of the 
microparasites or to a difference in the capacity for immunity response on 
the part of the body cells. 
In a series of experiments by Miss Trist and myself rabbits were injected 
with fixed and constant amounts of vaccines of glanders, typhoid, tuber- 
culosis, diphtheria, anthrax and subtilis bacilli, gonococci, meningococci, 
streptococci, pneumococci, and staphylococci; each vaccine contained the 
same number of bacteria per cubic centimeter, prepared in exactly the same 
manner and injected intravenously into rabbits in similar amounts per body 
weight. Agglutination and complement-fixation tests were conducted at 
regular intervals and marked variations in antibody response were observed. 
The vaccines of Gram-negative bacilli (glanders and typhoid) stimulated the 
production of more agglutinins and complement-fixing antibodies than the 
Gram-positive bacilli (tuberculosis and diphtheria) and the spore-forming 
bacilli (anthrax and subtilis) produced almost none of these. The Gram- 
negative cocci (gonococci and meningococci) were more antigenic than the 
Gram-positive cocci (pneumococci, streptococci, and staphylococci). 
These results are to be interpreted, of course, only in relation to the 
production of agglutinins and complement-fixing antibodies inasmuch as 
opsonic determinations were not made, but it would appear that bacterial 
proteins vary considerably in vaccinogenic activity in the same animal when 
given in the same amounts per body weight, and that this bears an important 
relation to the question of success of prophylactic and therapeutic immuniza- 
tion. 
Sensitized Vaccines.—Besredka and Metchnikoff1 have suggested a 
plan of injecting a vaccine composed of living bacteria that have been 
immersed in their specific immune serum or, in other words, have been 
sensitized. They believe that such vaccines produce practically no nega- 
tive phase, but only slight local and general reaction, and that the general 
response with antibody formation is facilitated. This principle is sup- 
ported by the observations of Theobald Smith, who found that the 
experimental injection of a toxin-antitoxin mixture aids the dissemination 
of the toxin through the body quite generally, whereas the pure toxin is 
chiefly held at or near the place of injection. This diffusion tends to cause 
maximum antibody formation over an entire portion of the body by a rela- 
tively small amount of free or easily dissociated toxin in the toxin-antitoxin 
mixture. Smith inclines to the belief that a similar phenomenon of diffusion 
may occur with sensitized dead bacteria. 
In addition, the specific immune serum may aid in the disintegration 
of the bacterial cell, either through the attachment of a bacteriolytic ambo- 
ceptor that would tend to lyse the bacterium with a complement of the 
tissues, or through a preliminary action of opsonin which prepared the 
bacterium for ultimate destruction and liberation of antigenic principles. 
Blandini,2 in a careful study of the immunizing properties of seventeen 
typhoid vaccines carried out on guinea-pigs, found that the sensitized 
vaccine of Besredka3 was the most protective. Several investigators have 
found that sensitized vaccines engender the production of agglutinins rather 
1 Ann. de l’lnst. Pasteur, 1911, xxv, 193, 867; 1913, xxvii, 597, 607. 
2 Ann. d’igiene sperimentale, 1905, 15, 295. 
3 Ann. de l’lnst. Pasteur, 1902, 16, 918; Compt. rend. Acad. d. Sci., 1902; Bull. Inst. 
Pasteur, 1910, 241. 754 
PRINCIPLES OF ACTIVE IMMUNIZATION 
poorly, blit better bacteriolysins and complement-fixing antobodies than 
plain vaccines. Garbat and Meyer1 have found that the sera of animals 
immunized with sensitized vaccines protects passively better than the sera 
of animals immunized with plain vaccines. Cecil2 employed sensitized 
vaccines in a series of 47 cases and found that the reactions are somewhat 
milder, but that the immunity is not higher than that engendered by plain 
vaccines. 
While Besredka originally advocated sensitized living vaccines, most 
vaccines of this kind are prepared of killed cultures sensitized by immune 
goat- or horse-serum. Gordon3 states that the cultures killed by phenol are 
as good as the former without the danger of infection. Wohl4 has advocated 
the use of autogenous vaccines killed by heat and sensitized by the patient’s 
own serum. 
Doubtless sensitized bacteria are more rapidly amenable to phagocytosis 
by reason of preliminary opsonification than plain vaccines, and this probably 
aids their quicker absorption from the subcutaneous tissues with less local 
reaction than follows the injection of plain vaccines. These advantages, 
however, are not marked, and the superiority of sensitized killed vaccines 
is more theoretic than practical on the basis of actual experience. 
Mixed Vaccines.—Of great practical importance is the question of the 
immunizing activity of mixed vaccines, that is, vaccines composed of two 
or more different kinds of bacteria. Castellani5 in 1903 showed that on 
injecting an animal with two different organisms at the same time, agglu- 
tinins were produced for both, and that the amount of agglutinin for each 
was about the same as in those animals immunized with but one type of 
organism. Subsequently he stated that as many as six different kinds of 
bacteria might be combined in a single vaccine with the same result, but 
that if more than six were employed, a diminished amount of agglutinin for 
each type resulted. During the recent World War Castellani advocated the 
use of these mixed vaccines for prophylactic immunization, and especially 
mixtures of typhoid, paratyphoid, cholera, and dysentery bacilli. 
Davison6 likewise found that immunization of rabbits with mixtures of 
typhoid and the paratyphoid bacilli yielded for each organism as much and 
usually a greater response of agglutinin production than the single vaccines. 
Noble and Thomas,7 however, found that the bacteriolysins engendered by 
the administration of mixtures of typhoid and paratyphoid bacilli were 
somewhat less than produced by vaccines of single strains. 
Smith8 has reported a marked example of polyvalency in the production 
of “ferments” against bacterial antigens. Huntoon and Craig9 have likewise 
found that in the immunization of horses with pneumococci, streptococci, 
and influenza bacilli there is nothing in the immunity mechanism itself 
to preclude the possibility of a wide polyvalent antibody production. 
These results indicate that antibody production may be engendered by 
mixtures of different kinds of bacteria and that under proper technical 
conditions of dosage, etc., the antibody response for each may be as great 
as when monovalent vaccines are employed. However, the immunity 
1 Ztschr. f. exper. Path., 1910, 8, 1. 
2 Amer. Jour. Med. Sci., 1918, 155, 781. 
3 Lancet, 1913, 1796. 
4 Amer. jour. Med. .Sci., 1916, 152, 262. 
5 Ztschr. f. Hyg., 1902, 40, 1; Jour. Trop. Med., 1914, 17, 326. 
5 Arch. Int. Med., 1918, 21, 437. 
7 Proc. Soc. Exper. Biol, and Med., 1920, 17, 190. 
8 Jour. Infect. Dis., 1915, 16, 313. 
9 jour. Immunology, 1921, 6, 235. SPECIFIC ANTIGENIC ACTIVITY OF VACCINES 
755 
response to mixed vaccines may not be as great as when single vaccines are 
employed when the amounts of the former injected are sufficient to produce 
more marked reactions and consequent depreciation of the general physical 
condition. For this reason the number of different bacteria incorporated 
into a vaccine should be limited as far as possible. 
Mixed vaccines for therapeutic immunization are commonly employed 
in the treatment of chronic infections of exposed mucous membranes and 
notably of the ear and respiratory tract. At the present time there is no 
practical means for determining which bacteria in cultures are pathogenic 
and thereby important from the standpoint of vaccine therapy, although 
some aid in this direction is given by skin tests and especially in allergic 
asthma. As a general rule and principle vaccines should be prepared of as 
few different bacteria as possible and only those incorporated into a vaccine 
in full dosage which are known or reasonably suspected to be pathogenic. 
Autogenous versus Stock Bacterial Vaccines.—It may be stated in 
general that autogenous vaccines, i. e., those prepared from the patient’s 
own bacteria, should be used whenever possible, especially in the vaccine 
treatment of disease. To be successful, vaccine therapy demands that the 
bacteria be as little changed as possible. Before they are killed the bacteria 
should be endowed with as many of the potencies as possible with which 
they maintain themselves in the body. As these potencies do not remain 
unchanged during artificial life, as the loss of capsules, loss of virulence, etc., 
it is advisable to secure the organism causing the infection as quickly as 
possible and prepare a vaccine without undue delay. 
Variants may occur among cultures of the same species, and the injection 
of one strain may not protect against another, as shown by Neufeld for the 
pneumococcus. In the use of an autogenous vaccine this risk of using an 
alien species or a different strain is reduced to a minimum. 
In some cases the difficulty of securing and of identifying the infective 
agent may be so great that much time is lost in preparing autogenous vac- 
cines, as, for instance, in gonorrheal and tuberculous infections, and in such 
cases it may be necessary to use a stock vaccine. 
In protective immunization stock vaccines are used, as, for instance, 
in the preparation of typhoid vaccine. In certain instances, as in gonococcus 
and tuberculous infections, stock vaccines possess but slightly inferior 
therapeutic value as compared with autogenous vaccines, not to mention 
the delay and difficulty in cultivating and preparing autogenous vaccines. 
Stock vaccines are generally required when vaccine therapy is employed 
in acute infections, as pneumonia, streptococcus cellulitis, and typhoid fever; 
at least they may be employed until the autogenous micro-organism can be 
recovered in culture and prepared in a vaccine. 
As a general rule stock vaccines should be polyvalent in order to include 
strains that may be serologically distinct. While autogenous vaccines are 
to be preferred in the treatment of disease providing they are properly 
prepared, stock vaccines possess equal powers for non-specific immunization 
and serve useful purposes. 
Specific Antigenic Activity of Vaccines.—The injection of dead bacteria 
in vaccines may cause the production of specific antibodies. The mechanism 
is similar to that involved during infection with the living micro-organism, 
and involves the first principles of immunity. The antigenic powers of a 
vaccine are probably always more or less inferior to the living antigen, as 
some principle may be lost during heating, drying, passage through animals, 
the action of germicides, etc. 
Just what portion of the bacterial cells is mainly antigenic it is difficult 756 
PRINCIPLES OF ACTIVE IMMUNIZATION 
to determine, for it probably varies with different species. With a true 
toxin, as, for example, the diphtheria bacillus, the toxin constitutes the 
main principle, and causes the production of an antitoxin as its main anti- 
body; with other bacteria, such as the typhoid bacillus, a soluble toxin and 
an endotoxin in combination with the protein of the bacterial cell are probably 
the main antigenic factors responsible for the formation of a bacteriolysin, 
opsonin, antitoxin, agglutinin, etc. 
In brief, the antigenic principles of a micro-organism are mainly thermo- 
stabile and fairly resistant substances, so that a bacillus may be so attenuated 
or altered that it cannot multiply or produce disease, and yet is capable, 
through the agency of substances that have escaped destruction, of causing 
the production of specific antibodies. 
As was previously stated, vaccines may cause the production of different 
antibodies. As curative agents, however, it would appear that they are 
most efficacious in those infections in which phagocytosis is known to be 
chiefly concerned in the defense of the host, e. g., in staphylococcus infections. 
As shown by Wright and Douglas, Neufeld and Rimpau, a bacterial vaccine 
facilitates phagocytosis, not so much qualitatively or quantitatively, as 
through the production of specific substances that act directly and primarily 
upon the bacteria and render them more vulnerable to phagocytosis (opsonin 
or bacteriotropin). Wright has advised a method of opsonic measurement, 
previously described, for measuring the immunity response, but, as will be 
understood, while the opsonin may be the chief antibody, it is seldom if 
ever the only one, so that the opsonic index is but one measure of defensive 
power. 
According to Vaughan, a micro-organism is directly responsible for the 
production of a specific proteolytic ferment capable of causing the disinte- 
gration or destruction of the bacterial cell and its products. The ferment 
is the antibody, and is produced during an infection or by a vaccine in just 
the same manner as antibodies in general are produced. In other words, 
Vaughan regards antibodies as of the nature of proteolytic ferments; thus 
the protein of the micro-organism composing a vaccine produces a specific 
proteolytic ferment capable of overcoming its substratum when it meets 
the latter in the form of the invading micro-organism of an infection. 
For example, the tissues affected may be unable to produce a sufficient 
quantity of the specific ferment necessary to overcome the infection. The 
injection of bacterial protein in another and healthier part of the body leads 
to the production, in this locality, of a specific ferment that is conveyed to 
the diseased area by way of the circulatory system, and aids in destroying 
the protein of the infecting micro-organism and its tissues. 
The Non-specific Activity of Bacterial Vaccines.—Owing to the weight 
of laboratory investigations and the fundamental laws of specificity in 
immunity reactions, the sole efficacy of a vaccine has been generally ascribed 
to the production and activity of specific antibodies, and any deviation from 
this current of thought has been received with a measure of skepticism and 
disapproval. Scattered throughout the literature are the reports of more or 
less isolated observations that in prophylactic and therapeutic immunization 
good results have been observed in the prevention and treatment of diseases 
other than that specifically concerned. 
The growing importance of this non-specific or “collateral” immunization 
by bacterial vaccines in the treatment of disease has been especially well 
summarized by Wright1 as follows: 
“I confess to having shared the conviction that immunization is always 
1 Lancet, March 29, 1919. ACTIVE IMMUNIZATION IN THE PROPHYLAXIS OF DISEASE 
757 
strictly specific. Twenty years ago, when it was alleged, before the Indian 
Plague Commission, that antiplague inoculation had cured eczema, gonor- 
rhea, and other miscellaneous infections, I thought the matter undeserving 
of examination. I took the same view when it was reported in connection 
with antityphoid inoculation that it rendered the patients much less sus- 
ceptiable to malaria. Again, seven years ago, when applying pneumococcus 
inoculations as a preventive against pneumonia in the Transvaal mines, I 
nourished exactly the same prejudices. But here the statistical results 
which were obtained in the Premier Mine demonstrated that the pneumo- 
coccus inoculations had, in addition to bringing down the mortality from 
pneumonia by 85 per cent, reduced also the mortality from ‘other diseases’ 
by 50 per cent. From that on we had to take up into our categories the fact 
that inoculation produces in addition to ‘direct’ also ‘collateral’ immuniza- 
tion. This once recognized, presumptive evidence of collateral immuniza- 
tion began gradually to filter into our minds. I suppose, many 
thousands of patients treated by vaccine therapy in private and in hospital, 
it happened every now and then that a patient was treated with a vaccine 
which did not correspond with his infection, and that that patient indubit- 
ably benefited. Again, it was not an uncommon experience for the subjects 
of a very chronic infection (such as pyorrhea) who were treated first by a 
stock vaccine, a-nd afterward with an autovaccine, to assert that they 
derived more benefit from, and to ask to be put back upon treatment by, 
the stock vaccine. 
“From such cases hints are conveyed to us that there may exist a useful 
sphere of application for collateral immunization; and that such sphere may, 
perhaps, be found in those cases where the infection is of very long standing, 
and where the patient has become very sensitive to, and has probably come 
very near the end of his tether in the matter of immunizing response to the 
particular species or strain of microbe with which he is infected. It will, 
with regard to such patients, be remembered that they constitute the third 
of those three clases of cases to which I referred at the outset of this lecture 
as very intractable to vaccine therapy. 
“We are, however, here considering primarily the question of principle; 
and in connection with this what is of fundamental importance is: that we 
should discard the confident dogmatic belief that immunization must be 
strictly specific, and that we should in every case of failure endeavor to make 
our immunization more and more strictly specific. We should instead pro- 
ceed upon the principle that the best vaccine to employ will always be the 
vaccine which gives on trial the best immunizing response against the 
microbe we propose to combat.” 
This subject is discussed in more detail in the chapter on the Use of 
Vaccines and Other Protein Substances in the Treatment of Disease. In 
prophylactic immunization specific agencies are of more importance, but in 
therapeutic immunization both specific and non-specific agencies are brought 
into operation. 
Active Immunization in the Prophylaxis of Disease.—Theoretically it 
would appear possible to actively immunize against all those infectious 
diseases in which the microparasites or viruses were procurable and capable 
of being prepared in vaccines. Practically, howrever, this has not been 
accomplished. Vaccines for smallpox, rabies, typhoid and paratyphoid 
fevers, diphtheria, and a few other infectious diseases of human beings, and 
for anthrax, black-leg, and a few other diseases of the lower animals, are 
known to engender the production of immunity principles and afford pro- 
tection varying in completeness and duration. In other diseases prophyl- 758 
PRINCIPLES OF ACTIVE IMMUNIZATION 
actic active immunization has either failed or the results are doubtful due 
either to a failure or tardiness of the body cells to produce immunity princi- 
ples, a lack of vaccinogenic activity of our vaccines or our failure to prepare 
and administer the vaccines in a proper manner. 
As was stated in the chapters on Immunity, in the presence of an infection 
the host endeavors to protect itself and overcome the invaders by various 
means, among which are phagocytosis and the production of more or less 
specific antibodies that may neutralize the poisons of the parasite (anti- 
toxins), directly kill or destroy them (bactericidans), or so lower their 
vitality or resistance that they are more easily phagocyted (opsonins or 
bacteriotropins). 
During an infection one or more of these protective forces, or all of them, 
are brought into action. After the infection has been overcome, the anti- 
bodies do not always disappear at once, but remain for some time in the body 
fluids and gradually diminish, so that if the host is invaded by the same 
parasite, the antibodies are at hand immediately to overcome it and protect 
the host absolutely, or at least so to modify the pathogenicity of the parasite 
or neutralize its products that the host will suffer but mildly while the 
parasite is being finally destroyed. The concentration and duration of the 
various defensive forces or antibodies vary in different individuals and in 
different infections, so that the degree and duration of an active acquired 
immunity are variable factors. Nevertheless—and this is the basis of 
active prophylactic immunization—an animal or a person may have specific 
antibodies for a certain parasite, produced by its own cells, without actually 
or necessarily suffering from the disease, due to the effects of inoculation 
with the germ or virus in a modified or attenuated form. The dose of vaccine 
may be so controlled that general symptoms the result of stimulation of 
the body cells are slight or not at all apparent, and, by gradually incressing 
the dose, more and more antibodies may be produced until a high degree 
of immunity is secured. 
For prophylaxis, or the prevention of a disease, active immunization 
is accomplished by the production of antibodies so that they may be at 
hand to overcome an infection if it should occur. For example, the anti- 
bodies specific against the virus of smallpox may be produced by inoculation 
with cowpox virus, so that for years the system will be protected against 
smallpox. Even if vaccination has been delayed until smallpox has actually 
been contracted, inoculation with cowpox virus early in the period of incuba- 
tion so stimulates the body cells that sufficient antibodies are produced to 
modify and lessen considerably the virulence of the smallpox virus. 
This is especially true in rabies, when the vaccine is given in such doses 
and at such intervals that sufficient antibodies are produced to neutralize 
the effects of rabic virus and actually to destroy it during the period of incu- 
bation or during the interval that elapses between the time of infection and 
the appearance of the symptoms. In this way the great majority of in- 
fected persons escape the sufferings of rabies by enduring the relatively 
slightd iscomfort consequent to a series of subcutaneous injections. 
Active Immunization in the Treatment of Disease.—This is bacterin or 
vaccine therapy, a method that owes its origin to the researches of Sir Alm- 
roth Wright and his colleagues. It was originally employed in the treatment 
of those infections that showed a tendency to chronicity in which true toxins 
played little or no part. Since recovery from an infection is in general 
dependent upon the mechanical removal of the infecting agent, aided by 
antibodies that facilitate phagocytosis or directly destroy the invading 
bacterium and neutralize its products, Wright believed that in chronic VACCINE THERAPY OF ACUTE INFECTIONS 
759 
infections autovaccination, or stimulation of the body cells to the production 
of antibodies, by reason of the fact that it is irregularly timed, is generally 
insufficient or altogether absent. For these reasons he believes that any 
stimulus that will arouse the body cells to throwing into the circulation 
substances from the invading bacterium or diseased tissues, may result in 
increased antibody formation, followed eventually by clinical improvement 
or cure. In certain cases this stimulation may be secured by judicious 
massage or manipulation of the diseased part, by passive hyperemia (Bier), 
or by similar procedures. 
However, if the micro-organism is obtained and cultivated artificially, 
it is possible, in many instances, so to modify or attenuate the organisms 
(usually by heat) that they may be reinjected into the patient in sufficient 
numbers to furnish the stimulus necessary for arousing dormant or unin- 
volved body cells to produce the opsonins and other antibodies necessary 
for overcoming the infection. In other words, with each infection the host 
endeavors to protect itself by producing antibodies. When the protection 
is insufficient, the infection will spread; when the antibodies are in excess, 
the infection is overcome; when the forces are about equal, a stage of chronic- 
ity may result in which the host becomes accustomed, as it were, to the 
invaders, and, while the infection does not spread rapidly, it does not, on the 
other hand, recede. In cases of the latter type an extra dose of bacterial 
stimulant (a bacterin) may arouse dormant or inactive cells to furnish an 
extra quantity of antibodies and thus turn the tide. 
In therapeutic inoculation, there]ore, the fundamental principle is to stimu- 
late in the interest of the infected tissues the unexercised immunizing capacities 
of the uninfected tissues. This is especially true in chronic infections, when 
the use of a bacterial vaccine may be likened to the application of the whip 
to a lazy horse that is capable of further effort and work. In acute infections, 
however, while the cells are at work they may be capable of greater effort, 
but vaccines should be given cautiously, as they may, to use the same simile, 
act as a wdiip to a willing and well-worked horse that is unable to respond 
or does respond, with resulting disastrous overexertion. 
It should he remembered, in this connection, that usual forms of treatment 
should be given while bacertin therapy is being instituted. For instance, it is 
useless to administer a vaccine to a patient with a suppurative fistula or 
sinus if an infected silk suture is directly responsible for the suppuration. 
The suture should be removed, if possible, and after this is done a vaccine 
may be of considerable aid in overcoming the coincident infection. 
In every form of bacterial infection a certain degree of success may be 
credited to vaccine therapy due not only to specific effects, but to non- 
specific effects as well. Failure to elicit beneficial therapeutic results is 
likewise common, and the whole subject of vaccine therapy is exceedingly 
difficult to evaluate in an impartial manner. 
Especially successful results have been obtained in furunculosis and 
acute inflammatory sycosis; in localized streptococcus cellulitis and lym- 
phangitis; in bronchitis, colicystitis, gonorrheal rheumatism and tuberculous 
phyctenular conjunctivitis, dactylitis, orchitis, and arthritis. 
Vaccine Therapy of Acute Infections.—The question of vaccine therapy 
in acute bacterial infections, as pneumonia, typhoid fever, puerperal sepsis, 
etc., has long been a subject of discussion with widely divergent opinions 
and practice. In acute localized infections, as streptococcus or staphylo- 
coccus cellutis, and lymphangitis, acute cystitis, etc., vaccine therapy has 
proved successful as an adjunct to treatment and clinical experience indicates 
that in acute generalized infections, as pneumonia and typhoid fever, the 760 
PRINCIPLES OF ACTIVE IMMUNIZATION 
early and careful administration of vaccines in small doses may be beneficial 
and a distinct aid in increasing resistance and recovery. 
As stated above, it would appear that vaccines are especially useful in 
chronic bacterial infections and that in acute infections vaccines may be con- 
traindicated because (a) of the danger of adding to the toxic substances to 
be combated; (b) of the danger of a “negative phase” or period of lowered 
resistance, and (c) of the delay (seven to ten days) in antibody production. 
These are factors not lightly to be ignored, but, on the other hand, should 
not be allowed to carry too much weight, inasmuch as ordinary bacterial 
vaccines in moderate dosage are not markedly toxic, beneficial specific and 
non-specific effects may result in a few days and before antibodies are 
demonstrable in the blood. 
As will be discussed shortly the fear of the “negative phase” has been 
largely responsible for hesitation in the use of vaccines in acute infections. 
I must state that I have shared this opinion, but experience is showing that 
this danger has been overemphasized. Actual decrease in antibodies by 
the injection of vaccines has not been conclusively proved. However, the 
injection of a vaccine may elicit focal and constitutional reactions due to 
toxic split protein products derived from the bacteria and the absorption 
of focus poisons; these effects in the writer’s opinion constitute the so-called 
“negative phase.” Since these substances add to the toxemia and effects 
of an acute infection they are to be avoided as far as possible; therefore if 
vaccines are given in the treatment of acute bacterial infections extra 
caution is required in dosage and intervals of injection. 
The Requisites for Success and the Causes of Failure in Vaccine Therapy. 
—The success of vaccine therapy may be said to depend upon the administra- 
tion of appropriate doses (a) to increase the defensive powers of the body 
and mainly the leukocytes and bacteriotropic substances—an increase of 
the phylactic power as stated by Wright,1 and (b) to provide means for the 
transport and access of these agencies to the foci of infection—the kata- 
phylaxis of Wright. 
The increase of defensive powers—the epiphylactic response of Wright— 
will be incomplete and interfered with in infections producing continuous 
constitutional disturbances and fever from continuous or frequently recur- 
ring auto-inoculation with bacterial products, as in pulmonary tuberculosis; 
in these diseases vaccine therapy is not generally successful. According to 
Wright the auto-inoculations must be abolished in order to obtain a free 
field for the employment of properly graduated doses of vaccine or for 
properly controlled auto-inoculations. When possible the focus of infection 
should be incised and evacuated or, when this is not possible, local or general 
(rest in bed) immobilization are required. 
Anything that prevents kataphylaxis or the carrying of defensive agents 
to the foci of infection will reduce the value of vaccine therapy as “when the 
arterial supply is interrupted or is closed down by collapse or the body 
petrified by cold, and the alkalinity of the lymph is blunted off by acid 
metabolites derived from the muscles” (Wright). 
Likewise a condition of negative chemotaxis or the ecphylaxis of Wright 
interferes with vaccine therapy; especially in septic wounds in which the 
leukocytes and protective substances of the blood are repelled by toxins or 
absorbed by bacterial products and a favorable culture-medium for bacteria 
created. For these reasons vaccine therapy may be unsuccessful in long- 
standing infections and especially in unopened abscesses and sloughing 
wounds. 
1 Lancet, March 29, 1919. THE ADMINISTRATION OF A BACTERIAL VACCINE 
761 
Under these conditions vaccine therapy should be accompanied by 
measures to counteract negative chemotaxis and promote kataphylaxis or 
the access of leukocytes and other defensive agencies to the foci of infection. 
Wright has summarized these remedial measures as follows: (a) The evacua- 
tion of pus or other fluids and replacement with fresh lymph which may be 
assisted by (b) cupping, the application of hypertonic saline solution or of 
irritant solutions, as Dakin’s fluid, hot fomentations, Bier’s hyperemia and 
massage, which excite hyperemia with increased exudation of cellular and 
serous elements and aid in the diffusion of defensive and curative agencies. 
Contraindications to Active Immunization.—It should be emphasized 
that a properly prepared vaccine is a potent substance capable, when given 
in excessive dosage or when otherwise injudiciously administered, of doing 
much harm. This is the main reason why vaccines should be used very 
cautiously, if at all, in the treatment of severe generalized infections. In 
passive immunization the conditions are different, as the body cells are not 
taxed, but rather, through the neutralization of the toxic substances which 
they are combating, they are relieved, and an antibody-laden serum may, 
therefore, be freely administered in severe infections. 
In active immunization for therapeutic purposes the conditions should 
be carefully weighed and the treatment conducted by one who is qualified 
to judge of the potencies of harm and good in a vaccine, and who has had 
sufficient experience to guide him in dosage and frequency of inoculation, 
the main objects being to tide over and aid nature during an acute infection, 
and to arouse and stimulate her during a chronic infection. 
In prophylactic immunization the physician should satisfy himself that 
the patient has no latent or active infection that may be rendered worse 
during the temporary depression that follows inoculation. It is true that 
this depression is fleeting and temporary, and that the possible harm incurred 
may be far outweighed by the ultimate good, but vaccines should be given 
with proper discernment and not carelessly and injudiciously. These 
remarks have no relation to cowpox vaccination, where the good so far 
overbalances the possible harm that in general all persons should be vacci- 
nated, especially if an epidemic is impending. 
(a) Tuberculosis.—There is at present some discussion relative to the 
harm that may be caused in tuberculosis by typhoid immunization. Prob- 
ably all will agree that a patient with an active and acute tuberculous lesion 
should be refused inoculation, but when the lesion is quiescent or healed, 
or in the early latent stage, it is indeed difficult to understand how a prophy- 
lactic dose of typhoid vaccine will do more or as much harm as an attack of 
tonsillitis, rhinitis, or some similar acute infection. 
(ib) In diabetes, carcinoma, and other debilitating conditions vaccines 
should not be administered unless the indications or requirements are 
unusually urgent. 
(c) Advanced nephritis, especially parenchymatous nephritis, may be 
regarded as contraindicating the administration of a vaccine. 
THE ADMINISTRATION OF A BACTERIAL VACCINE 
Method of Making the Inoculation.—As the administration of a vaccine 
is frequently followed by a temporary depression of the resisting powers of 
the individual and a feeling of lassitude, the injections are, as a rule, best 
given during the afternoon and evening, the night’s rest aiding in overcoming 
the depression. Since the determination of proper dosage rests mainly on 
the observation of such clinical signs and symptoms as temperature, pulse, 762 
PRINCIPLES OF ACTIVE IMMUNIZATION 
and local reaction at the site of the lesion, the patient should be watched 
during the following twenty-four to forty-eight hours. 
Vaccines are best administered with the aid of a 1 c.c. all-glass syringe, 
furnished with a sharp platinum iridium or steel needle. These may be 
sterilized in boiling water for a minute or longer. After sterilization, the 
parts should be carefully adjusted and the syringe loaded. 
The injections should be given at a point where the tissues are loose, 
where muscular action is not much in evidence, and where pressure by 
clothing or weight is not made. The most suitable localities are in front 
of the shoulder, at a site about 1| inches below the center of the clavicle; 
high up in the buttock, or in the side of the abdomen, about 2 or 3 
inches inside the anterosuperior spine of the ilium. In the majority of 
cases the injection may be given under the skin of the arm in the neighbor- 
hood of the insertion of the deltoid muscle. The skin at the point of injection 
should be touched with tincture of iodin, which is removed after the injection 
has been given by washing with alcohol. 
Both convenience and experimental work to test the comparative efficacy 
of inoculation into different tissues point to the subcutaneous tissues as the 
most suitable site for inoculation. The best method of procedure is to pick 
up a fold of skin between the finger and thumb, and then to push the needle 
well down into the middle of the fold, and slowly inject the fluid. 
Since it is known that the power of response of the tissues to the stimulus 
of a vaccine is somewhat limited, it would seem advisable to choose a new 
site for each successive inoculation. 
Vaccines are sometimes injected intramuscularly, in which case the 
muscles of the buttocks afford a suitable site. 
Intravenous injections of typhoid and other bacterial vaccines are 
sometimes made and especially for the purpose of eliciting non-specific 
reactions in the treatment of arthritis. The technic is described in Chapter 
XL. 
The Effects of Inoculation.—The local effects produced at the site of 
subcutaneous or intramuscular inoculation vary considerably, being influ- 
enced by the nature of the individual, the variety and amount of the inocu- 
lum, and the sensitiveness of the patient’s tissues to stimulation. In the 
majority of cases the local reaction is limited to a very slight reddening of 
the skin around the puncture for an area of about 1 inch. In some in- 
stances occasionally encountered where a large number of typhoid inocula- 
tions have been made, the reaction after the first dose is more severe than 
after subsequent doses, and is accompanied by considerable edema, hypere- 
mia, and pain. 
The focal effects about the lesion are exceedingly important in determining 
the reaction of the patient, and serve as a guide to the adjustment of dosage 
and intervals. Where, in a case of furuncle, and appropriate dose of staphy- 
lococcus vaccine is administered, within a few hours increased hyperemia 
is seen around the focus, and there is a slight increase in the swelling. When 
very small doses are given, these focal symptoms may practically be absent, 
but, as a rule, a slight reaction does no harm, but serves rather to show 
that the vaccine possesses some degree of potency and may aid in the curative 
process. 
The constitutional effects may also vary within wide limits. An adequate, 
but not excessive, dose may, within a few hours, produce a feeling of lassitude, 
headache, slight rise in temperature, and acceleration of the pulse-rate. 
Severe constitutional reactions are generally due to excessive dosage, but 
may occur in some persons after doses that were previously well borne. THE ADMINISTRATION OF A BACTERIAL VACCINE 
763 
Frequency and Dosage of Inoculation.—No definite rules can be laid 
down, each patient being a lawr unto himself. The opsonic index has been 
largely abandoned as a guide to the administration of vaccine, the reaction 
and condition of the patient now governing the dosage. In more acute 
infections, and in delicate persons, smaller doses are usually indicated. It 
is well to make the first dose small, and if no reaction occurs within forty- 
eight hours, a second and a larger dose may be given. If, however, the 
patient presents other symptoms of a general reaction, the dose given was 
large enough, and may be repeated, as necessary, at intervals of from five 
to seven days. It should be carefully borne in mind that an increase in 
dosage is contraindicated so long as any sign of general or focal reaction is 
produced and steady progress is maintained. One should always be on 
guard to detect any signs of fresh infection by some other organism, and if 
a given vaccine is failing to exert a beneficial effect, additional cultures should 
be made, instead of continuing to administer dose after dose of the same 
vaccine. 
The intervals at which injections are to be made are of some importance. 
It is certainly better to wait too long than to inoculate prematurely, but 
the ghost of the “negative phase” is always too prominent in the minds of 
the inexperienced. The inoculations may be given while improvement is 
still in progress or convalescence well established, in the endeavor to secure 
a summation of positive phases of clinical improvement; or one may wait 
for the first signs of retrogression before administering another dose. The 
former method is the preferable procedure, but is difficult to accomplish; 
the latter is less ideal, but is easier to perform and more devoid of risk. 
The dosage varies according to whether the infection is acute or chronic, 
the nature of the micro-organism, and the age of the patient. No fixed 
rules can be given. In acute infections the dose should be small and may 
frequently be repeated; in chronic infections larger doses may be given at 
longer intervals. If in doubt as to the size of the dose to be given, it is 
better to give a small dose, and carefully observe the effect on the patient, 
letting this serve as an index to subsequent doses. Children tolerate rela- 
tively large doses of bacterial vaccines, but the dosage should depend on 
the weight and not on the age of the child. 
The following is a list of the ordinary doses for adults of various bacterins: 
Staphylococcus aureus 100,000,000 to 1,000,000,000 
Staphylococcus albus apd citreus 200,000,000 to 1,000,000,000 
Streptococcus pyogenes 25,000,000 to 200,000,000 
Gonococcus 25,000,000 to 200,000,000 
Typhoid bacillus 250,000,000 to 1,000,000,000 
Colon bacillus 100,000,000 to 1,000,000,000 
The Mechanism of the Reaction Following the Subcutaneous and 
Intramuscular Injection of Vaccines.—The local reactions following the 
subcutaneous and intramuscular injection of vaccines may be due to a 
variety of factors as (a) irritation caused by the phenol or other preservative; 
(b) irritation caused by bacterial products; (c) irritation caused by the 
products of bacterial proteolysis, and (d) a local anaphylactic reaction to the 
bacterial protein. 
The amount of phenol or other preservative added to a vaccine should 
be the minimum required, as 0.25 to 0.3 per cent., in order to avoid the irrita- 
tion excited by larger amounts. 
The irritation excited by the bacterial products varies considerably with 
different vaccines, being especially marked with vaccines of dysentery bacilli. 
These irritant substances are probably split bacterial proteins rather than 764 
PRINCIPLES OF ACTIVE IMMUNIZATION 
preformed bacterial substances and appear in old vaccines to a larger extent 
than in freshly prepared vaccines. Vaccines prepared by heating broth 
cultures and employing the broth as a diluent are usually more irritating 
than suspensions in saline solution. A part of the local reaction may be 
due to the digestion of the bacterial proteins by tissue and leukocytic pro- 
teases with the production of toxic split products exciting large and diffuse 
areas of pale erythema analogous to the non-specific reaction excited by the 
protein constituents of the diphtheria bacillus and broth in the Schick 
reaction. 
In some instances the reaction is doubtless in part, at least, a local 
anaphylactic reaction to the bacterial proteins. This is particularly true in 
tuberculosis after the injection of tuberculin and in other chronic infections, 
as bronchitis and bacterial bronchial asthma, in which cutaneous and intra- 
cutaneous tests with bacterial proteins elicit positive reactions. It is not 
uncommon to observe that local reactions are of increased severity after 
several doses have been given, and these suggest the possibility of anaphy- 
lactic sensitization; likewise the second course of typhoid inoculations one 
or more years after a preceeding course may elicit well-marked reactions of 
anaphylactic nature. Similarly the development of local reactions during 
the course of rabies vaccination suggests the possibility of these being due to 
active sensitization to the protein constituents of the spinal cord or brain 
substance of rabbits employed for the preparation of the vaccine. There 
can be no doubt that bacterial anaphylactic reactions may occur, and 
especially after the intravenous injection of vaccines, as discussed 
in Chapter XL. The subject of bacterial anaphylaxis in general is con- 
sidered in Chapter XXIX. 
The focal reactions occurring about the foci of infection are commonly 
in the nature of hyperemia with serous and cellular exudation, and are 
especially well marked in tuberculosis following the subcutaneous injection 
of an adequate amount of tuberculin. 
These reactions are generally regarded as allergic or anaphylactic in 
character due to sensitization of the granulation and other tissues about 
inflammatory foci by bacterial proteins, the hyperemia being caused by an 
allergic reaction. For this reason these reactions have been considered 
strictly specific and diagnostic, and especially in tuberculosis, but within the 
last few years it has been shown that reactions of this kind may be elicited 
by non-specific agents and especially when these are injected intravenously 
(discussed in more detail in Chapter XXXIX). These focal reactions are very 
probably of therapeutic value by reason of the induced hyperemia and serous 
and cellular exudation resulting in the bringing in to the focus of leukocytes 
and bacteriotropic substances and the production of fibrous tissue, the latter 
being especially favorable for the walling off of tuberculous lesions. 
The constitutional reaction is ascribed to the influence of protein constitu- 
ents by Schittenhelm and Weichardt,1 and particularly of the diamino-rich 
complexes, such as histones and protamins, which, according to Ruppel, are 
present in large amounts in a number of bacteria. Petersen2 has grouped 
the toxic split products of bacterial proteins as follows: (a) Preformed 
protein split products which are toxic; (b) protein split products formed as 
the bacterial protein is fragmented in the host; (c) toxic growth products 
derived from the bacterial metabolism and excreted, and (</) toxic metabolic 
products derived from the pathologic cellular metabolism of the invaded 
organism. 
1 Munch, med. Wchn., 1910, lvii, 1769; 1911, Ivin, 840; 1912, lix, 67; 1919, lxvi, 1403. 
2 Protein Therapy and Non-specific Resistance, MacMillan Co., 1922, 99. THE ADMINISTRATION OF A BACTERIAL VACCINE 
765 
In addition to the toxic split products of the bacterial vaccine it is likely 
that the absorption of toxic substances from the foci of disease mentioned 
above are important in this connection and capable of adding to the symp- 
toms of fever, chill, tachycardia, malaise, and other symptoms. This 
absorption of toxic foci posions is facilitated by the hyperemia and the 
poisons regarded by many as the cause of the constitutional reaction following 
the injection of tuberculin. 
The Question of the Negative Phase.—In 1901 Wright1 described what 
he termed the “negative phase” (since renamed by him “apophylactic 
phase”) following the injection of a large dose of typhoid vaccine—a tem- 
porary period in which there was diminished blood resistance or lowered 
immunity. This period of lowered resistance is believed to last from several 
hours to a day or more, followed by the positive phase of increased resist- 
ance. Wright also found that this lowered resistance was most pronounced 
after the injection of large doses or giving vaccines at too short intervals. 
These views have greatly retarded the employment of vaccines in the 
treatment of acute bacterial infections. Wright and von Wassermann and 
Sommerfeld2 maintain that injections of vaccines may decrease the antibody 
content of the blood of man and the lower animals; Bull,3 however, found 
that the administration of typhoid vaccine to rabbits did not cause, as far 
as could be determined, a decrease in the antibody content of the blood; 
indeed, after intravenous injections there was an increase of the normal 
antibodies which is particularly significant in connection with the subject 
of the effects of non-specific protein therapy. 
However, as previously discussed in connection with vaccine therapy in 
acute infections, the reactions caused by vaccines due to the absorption of 
toxic split proteins from the bacteria, in addition to those from the foci of 
infection, may add to the degree of intoxication induced by a bacterial 
infection. These constitutional and focal effects may still further reduce 
resistance and constitute a real “negative phase,” not in the original meaning 
of Wright ascribed to the reduction in antibodies, but to increased protein 
intoxication. For this reason the dosage and intervals of injection of 
vaccines in the treatment of both acute and chronic infections, and for 
prophylaxis in the presence of epidemics, should be such as to avoid severe 
reactions. 
1 Lancet, September 14, 1901; Brit. Med. Jour., 1903, 1, 1069. 
2 Med. Klinik, 1915, 11, 1307. 
3 Jour. Exper. Med., 1916, 23, 419. CHAPTER XXXV 
PROPHYLACTIC ACTIVE IMMUNIZATION OR VACCINATION 
VACCINATION AGAINST SMALLPOX 
Immunity in Smallpox.—Smallpox occurs spontaneously only in man, 
although some of the lower animals may be infected experimentally. In- 
stances of natural immunity to this disease are rare; not infrequently, 
however, persons are encountered who are apparently resistant or immune 
to repeated cowpox vaccination, and in some instances these are known to 
have escaped smallpox infection. In the great majority of such instances 
of resistance to first vaccination failure is due to the use of a faulty virus 
rather than to actual natural immunity. 
Unvaccinated human beings of all ages and both sexes are extremely 
susceptible to smallpox. The mortality among the aborigines and negroes 
is especially high. The child in ulero may be infected by placental trans- 
mission of the virus providing the mother has smallpox. 
One attack of smallpox usually protects for life, but second and even 
third attacks are known to have occurred. After recovery from smallpox 
immunity principles may be found in the blood. Attempts toward inducing 
passive immunization of animals against cowpox and smallpox by means 
of the injection of blood from vaccinated animals have been occasionally 
successful; likewise in a few experiments the viruses of cowpox and smallpox 
have been successfully neutralized or destroyed in vitro through contact 
with the blood-serum of vaccinated animals and human beings recovered 
from smallpox. The presence of complement-fixing antibodies in the sera 
of human cases of smallpox and of vaccinated calves has been reported by 
Jobling,1 Sugai,2 Dalm,3 Krylofif,4 Teisser and Gastinell,5 Klein,6 Konschegg,7 
the writer,8 and others. 
History of Vaccination.—Just when and where smallpox vaccination was 
first practised is not known. The original method of inducing immuniza- 
tion against the disease by introducing the virus from a smallpox patient 
into a healthy person through an abrasion of the skin and thus greatly dimin- 
ishing the virulence of the disease was practised by the Turks during the 
eighteenth century, the chief object to preserve the beauty of the young 
Turkish and Circassian women. In 1878 Lady Mary Montagu, the wife of 
the British Ambassador at the Ottoman court in Constantinople, observing 
this practice among the Turks, had her own son and daughter inoculated 
and was largely instrumental in establishing the practice in Europe. 
As regards the prophylactic value of this method of inoculation in England 
and continental Europe, statistics are incomplete, but the literature of con- 
temporary writers shows that protection was usually complete. The induced 
disease was not, however, always mild, and not infrequently assumed an 
unexpected virulence that not only proved distressing and even fatal to 
inoculated individuals, but also constituted a source of infection to a com- 
munity. While, therefore, the underlying principles were sound, and while 
1 Jour. Exper. Med., 1906, 8, 707. 
2 Centralbl. f. Bakteriol., 1909, 49, 650. 
3 Centralbl. f. Bakteriol., 1909, 51, 136. 
4 Centralbl. f. Bakteriol., 1911, 60, 651. 
8 Ibid., 1912-13, ref., 55, 555. 
6 Munch, med. Wchn., 1914, 61, 2270. 
7 Munch, med. Wchn., 1915, 62, 4. 
8 Jour. Immunology, 1916, 1, 59. 
766 VACCINATION AGAINST SMALLPOX 
767 
these early attempts at preventive immunization mark an epoch in the 
history of medicine and of the world, it was not until Edward Jenner made 
his investigations into a theory held by farmers and by experimental evidence 
established it as true that a satisfactory method of immunizing the body 
against smallpox was introduced. 
The peasantry in various parts of Europe, and especially in England, 
had generally observed that those who had had sores on their hands con- 
tracted from similar lesions on the teats of cows, usually escaped smallpox 
infection when the disease was epidemic in a community. In fact, it is said 
that several farmers deliberately inoculated members of their family with 
cowpox lesions and that these escaped smallpox. 
Edward Jenner was a physician practising in Berkeley, Gloucestershire, 
and frequently used the method of direct inoculation from a mild case of 
smallpox among his patients. While a student he was impressed with the 
traditions of cowpox vaccination, and finding that they were largely true, 
determined to make experimental tests. On May 14, 1796 he vaccinated 
a boy, James Phipps, with virus from a cowpox lesion on the hand of a dairy 
maid,- Sarah Nehnes, and on July 1st he inoculated the same boy with pus 
from a smallpox patient without resulting infection. In 1798 he furnished 
further proof that cowpox will afford protection against smallpox by inocu- 
lating a child direct from a vesicle on the teat of a cow, and continued the 
inoculation from arm to arm through a series of 5 children, after which 
all were inoculated with smallpox virus, without a single case developing. 
In the same year he published “An Inquiry into the Causes and Effects of 
the Variolae Vaccinae,” illustrated by four plates, and within a year or two 
vaccination became general over Europe. 
Vaccination was introduced into the United States in July, 1800, by 
Dr. Benjamin Waterhouse, Professor of Physics at Harvard University, who 
vaccinated his own children. At about the same time John Redman Coxe, 
of Philadelphia, vaccinated his oldest child and then tested the experiment 
by exposing him to cases of smallpox. This bold repetition of Jenner’s 
experiment considerably strengthened public confidence in the method and 
the practice spread rapidly. Thomas Jefferson, writing in 1806 to Edward 
Jenner, said: “Future generations will know by history only that the 
loathsome smallpox existed and by you has been extirpated.” 
But Jenner and his earlier supporters met with much opposition, often 
bitter and unrelenting, and this is readily understood when it is realized 
that even at the present day, over a hundred years later, cowpox vaccination 
still has its opponents, in spite of the fact that the value of the method has 
been established, and it has been found the greatest of all boons to the 
human race, and notwithstanding that it has been definitely proved that a 
thorough and continuous practice of the operation would quickly eradicate 
smallpox from the face of the earth. This opposition is especially pernicious 
and unjust, since the practice of former years of vaccinating by direct 
transmission from arm to arm has been entirely abandoned, and that animal 
lymph, prepared and collected under strict aseptic precautions, is being 
used exclusively. 
The Relationship of Variola and Vaccinia.—The relationship of variola 
to vaccinia has been discussed since Jenner’s time, but no adequate explana- 
tion has been found. 
According to the general belief the smallpox virus, whatever it may be, 
is altered in its passage through a lower animal, and loses forever its power 
of producing smallpox, but is still so closely related that the antibodies it 
produces are sufficient to protect against smallpox. 768 
PROPHYLACTIC ACTIVE IMMUNIZATION OR VACCINATION 
The close interrelationship existing between vaccinia and variola is 
shown by several facts. First, by the presence in the virus of both, and in 
section of the skin of both, of microscopic cell inclusions, first described by 
Guarinieri in 1892. This finding has been confirmed by Pfeiffer in Germany 
and Councilman and his associates in this country. These investigators 
have made extensive studies of these bodies, and believe them to be protozoa 
intimately associated with the etiology of vaccinia and variola. More 
recently Fornet has described certain small, diplococcus-like bodies that 
were found in cowpox vaccine and in smallpox lesions. These are regarded 
as having an etiologic relationship to smallpox, and if these findings are 
confirmed, would prove the identity of variola and vaccinia. 
Second, recent investigators, particularly Copeman, of England, and 
Brinkerhoff and Tyzzer, of America, have shown, by carefully conducted 
experiments, that vaccination will protect monkeys against subsequent 
inoculation with smallpox virus, and this completely confirms the early 
experiments of Jenner and others who proved the efficacy of vaccination by 
the “variolous test.” 
Third, smallpox inoculated into calves is said by some investigators 
to produce a disease similar to cowpox after passage through several animals. 
The Preparation of Cowpox Vaccine.—During the early days of vaccina- 
tion it was customary to inoculate human beings with material obtained 
from the pustules of those previously vaccinated. The old-time physician 
carefully removed choice scabs and carried them about in a special case 
ready for inoculation. While this method served its purpose well, there 
were several drawbacks to its use, the chief of which was the danger of 
transmitting syphilis. It has now for many years been the custom to use 
virus obtained from animals, the production of which can be carefully con: 
trolled and tested, any danger of transmitting syphilis being thus obviated, 
because the heifer or cow used in the preparation of the virus is not subject 
to this disease. The opponents to vaccination, however, persist in using 
old and obsolete statistics regarding the transmission of syphilis to support 
their claims, although these have absolutely no bearing upon the modern 
methods of preparing the virus. 
Seed Virus.—This refers to the virus for vaccinating the calves or other 
animals used, and is a most troublesome factor to those engaged in this 
work. Park has found the following method the most economic, efficient, 
and reliable method found in the New York laboratories: “Crusts are 
collected from healthy children about nineteen days after successful vaccina- 
tion. These crusts are cut up and emulsified with boiled water to a muci- 
laginous paste. This humanized seed is inoculated into an area about 6 
inches square upon the abdomen of a calf, the remainder of the calf being 
vaccinated in the ordinary way. The pulp from this special area is sepa- 
rately collected and glycerinized in the usual way. It is then tested bac- 
teriologically and clinically. This bovine virus from human seed is now 
used in a dilution of 1 part to 12| parts of normal salt solution to vaccinate 
rabbits. The seed is rubbed thoroughly on the freshly shaved skin of the 
back. Five days after vaccination the pulp is removed with a curet, 
weighed, and emulsified in a mortar with the following solution: glycerin 
50 per cent., sterile water 49.5 per cent., and carbolic acid 0.5 per cent., in 
the proportion of 1 part of pulp to 8 parts of the solution. Four rabbits 
should yield from 15 to 20 c.c. of this emulsion, an amount sufficient to 
vaccinate one calf.” 
Large amounts of vaccine are then prepared by vaccinating calves with 
this seed virus. VACCINATION AGAINST SMALLPOX 
769 
The New York Vaccine Laboratory produces a virus that is never more 
than four successive transfers from a human case of vaccinia, and is guaran- 
teed to give 100 per cent, of “takes” in primary vaccination. 
Animals.—Various animals have been used, but female calves from 
two to four months of age are preferable. Older animals may be used, 
and in several European institutes cows are usually employed. With 
properly constructed operating-tables they may be handled with com- 
parative ease. Rabbits have also been used, especially in propagating 
the seed virus and to obtain pure and highly active viruses. 
The calves are kept under supervision for at least a few days. In some 
institutes they are tested with tuberculin, although with good veterinary 
inspection this test is not necessary. Soon after admission the animal is 
clipped and given a thorough cleansing, which includes the feet and the tail. 
On the day before vaccination is to be performed the belly wall is cleanly 
shaved from the cuneiform cartilage to the pubis, and well up on the inner 
sides of the thighs and the flanks. The skin is then thoroughly washed. 
Just preceding the vaccination the animal is fastened to the operating-table 
and the abdomen and inner surface of the thighs prepared as for an aseptic 
abdominal section, i. e., a thorough scrubbing with hot water, green soap, 
and soft brush, followed by alcohol and sterilized water, the parts being 
then dried with a sterile towel. All other parts are covered with sterile 
sheets, and the calf is now vaccinated under aseptic precautions. 
Vaccination.—About 100 small scarifications are now made in these 
areas, preferably by cross-scratches or in rows of lines about 1 to 2 cm. 
square and at least 1 to 2 cm. apart. The scarification is simple, but usually 
brings a small amount of blood. After they have been made, they are 
mopped with sterile gauze and rubbed with the charged slips, using one or 
two slips for each small area, depending on the amount of virus each slip 
contains. The lesions are allowed to dry, and are then covered with 
sterile gauze or a simple protective paste, or are left entirely uncovered. 
Precautions should be taken to keep the animals as clean as possible. 
Inoculated animals are to be kept in stalls or stables apart from those under 
observation. The stable should be so constructed that the floors can be 
flushed daily with a hose and hot water. Excreta should be removed 
promptly. No bedding is permissible, and means should be provided for 
fastening the legs and preventing the animal from kicking the scarifications. 
Collection.—Ordinarily, within forty-eight hours of vaccination, the 
scratches are pinkish, slightly raised, and papular, and within five or six 
days, depending upon the rate of development of the vaccine vesicles, the 
virus should be ready for collection (Fig. 169). The calf is killed and placed 
upon the operating-table. The appointments of the operating-room are 
usually equal to those in a well-equipped hospital operating-room, being 
supplied with all conveniences and means for carrying out a careful, pains- 
taking, and aseptic technic. 
The exposed parts are covered with sterile sheets. The operator and 
his assistant are clad in aseptic gowns. The vaccinated field is thoroughly 
scrubbed with soap, sterile water, and gauze, and mopped with sterile gauze. 
Crusts are carefully picked off, and the soft, pulpy mass cureted off with a 
special, spoon-like curet and collected in a sterile vessel. After the curetage, 
serum exudes from the excoriated base of the vesicle, and ivory tips may be 
charged in this. The sticky and pulpy exudate is then mixed with four times 
its weight of glycerin and water (50 per cent, glycerin, 49 per cent, water, 
1 per cent, phenol), and this is done most effectively by passing the mixture 
between the rollers of a Doring mill. The glycerinated pulp is allowed to 770 
PROPHYLACTIC ACTIVE IMMUNIZATION OR VACCINATION 
stand for three or four weeks in order to allow bacteria, which are invariably 
present, to undergo dissolution. At the end of this time the glycerinated 
pulp is thoroughly titrated in specially constructed triturating machines, 
passed through 40- and 100-mesh sieves, and put up in small capillary tubes, 
which are sealed, or “vaccine points” may be prepared. If properly pre- 
served in sealed tubes in a dark, cool place (—2 to 4° F.) the virus should 
remain active for at least three months. 
According to Park and Huddleson, 10 grams of pulp and 200 charged 
slips would be an average yield from a calf, and when made up should 
Note the lines of cowpox lesions over the abdomen and flanks of the calf. The surgeon is about 
to cleanse this area in a thorough and careful manner, after which the cowpox material is removed with 
a curet and collected in a sterile vessel. All precautions are taken to insure as thorough aseptic technic 
as possible. 
Fig. 169.—Production of Cowpox Vaccine. 
suffice to vaccinate at least 1500 persons. Calves vary greatly in their 
yield of virus. Of 2 calves vaccinated in exactly the same manner, one 
may furnish material for 500 vaccinations and the other for 10,000 inocula- 
tions. 
Testing the Vaccine.—The virus may be tested for its efficacy by a variety 
of methods. Calmette and Guerin1 inoculate rabbits upon the inner surfaces 
of the ears and estimate the potency of the virus from the speed of develop- 
ment and the size of the resulting lesions. Guerin2 estimates the potency 
1 Ann. de l’lnst. Pasteur, 1901, 15, 161. 
2 Ann. de l’lnst. Pasteur, 1905, 19, 317. VACCINATION AGAINST SMALLPOX 
771 
of virus quantitatively by inoculating rabbits with serial dilutions ranging 
from 1 : 10 to 1 : 100. Fully potent virus should cause closely approximated 
vesicles in a dilution of 1 : 500, and numerous isolated vesicles in a dilution 
as high as 1 :1000. 
Quantitative estimation of the bacteria in the glycerinated virus is made 
by the plating method, and the vaccine used only when the numbers of 
bacteria have been greatly diminished or are entirely absent. The vaccine 
is also tested for tetanus by anaerobic cultures and by injecting relatively 
large quantities subcutaneously into guinea-pigs and mice. 
Tests are also made for streptococci and Bacillus welchii (B. capsulatus 
aerogenes). 
After all of these laboratory tests for purity have been made and found 
satisfactory, and not until then, the vaccine is submitted to clinical trial. 
Fifteen inoculations are made upon previously unvaccinated children. 
These must all show a perfect take in order to pass the vaccine as up to 
standard. A clinical test is made every two weeks thereafter as long as the 
vaccine is on the market. If one of these tests fails before the end of the 
period of guarantee, the vaccine is called in (Park). 
Under the Federal Law of July 1, 1902, and the regulations framed 
thereunder, all firms manufacturing vaccine virus are required by the 
Secretary of the Treasury to obtain a license before they may sell their 
products in interstate commerce. The vaccine laboratories are carefully 
inspected by an official of the Hygienic Laboratory of the United States 
Public Health and Marine Hospital Service; the inspector carries away 
with him as many samples of virus as he wishes, and additional samples 
are purchased in the open market in different parts of the country. All 
these are subjected to a vigorous bacteriolgic examination, especially for 
tetanus bacilli, by a laboratory worker who devotes all his time to this work. 
The federal regulations require each vaccine institute to perform a careful 
autopsy on each calf after the vaccine virus has been removed, and if any 
communicable disease is found or suspected in the animal, the virus must 
not be placed on the market, but must be destroyed. In accordance with 
this law, permanent records of the bacteriologic examinations of the virus 
and of the autopsy shall be kept in each institute. 
Noguchi’s Method of Preparing Cowpox Virus.—Noguchi1 has suc- 
ceeded in freeing vaccine virus from all associated bacteria by means of 
suitable disinfecting agents, and propagated the pure virus in the testicles 
of rabbits and bulls. The virus cultivated in this manner is not only devoid 
of bacteria, but appears capable of definite transfer from one animal to 
another. The multiplication of the virus within the testicle was found by 
Noguchi to reach a maximum on the fourth or fifth day after inoculation, 
with diminution after the eighth day. Skin lesions produced in rabbits 
and calves with the original and purified or bacteria-free viruses were found 
identical, and persons have reacted to the latter in an entirely typical man- 
ner. These results are of great value and importance and tend to increase 
the safety of vaccination and incidently weaken the arguments and conten- 
tions of antivaccinationists. 
Unfortunately, however, the method has not proved successful on a 
commercial scale and the virus has been found by Force2 and others to 
rapidly deteriorate. 
Keeping of Vaccine.—Vaccine virus should always be kept at a low 
temperature (33° to 40° F.). Not infrequently failure to successfully vaccinate 
an individual is due to deterioration of the virus kept in a warm place. 
1 Jour. Exper. Med., 1915, 21, 539. 
2 Jour. Lab. and Clin. Med., 1917, 3, 220. 772 
PROPHYLACTIC ACTIVE IMMUNIZATION OR VACCINATION 
Technic of Vaccination.—The essential part of the process of vaccination 
is that the virus should be introduced through the epidermis so as to be 
absorbed by the lymphatics and blood-vessels of the corium. 
The site usually chosen is the skin of the outer side of the upper arm, 
over the insertion of the tendon of the deltoid muscle. Sometimes, in 
females, the outer side of the thigh or well above the knee on the inner 
aspect of the thigh is used. Vaccination on the foot and leg, however, is 
never advisable, as it would appear that such vaccinations are more prone 
to take on an excessive inflammatory action owing to the greater congestion 
due to the dependent position of the lower extremities; there is also more 
likelihood of secondary infection and mechanical violence occurring. 
Preparation of the Skin.—The skin should be washed with soap and 
water and in any case with alcohol on gauze or cotton, care being taken 
Fig. 170.—Technic of Vaccination. 
First Step: Proper cleansing of the skin with alcohol and cotton. 
not to rub too vigorously; if the skin is reddened, it is best to wait until 
the hyperemia subsides and always until the skin is dry (Fig. 170). 
Several methods may be employed for vaccination as follows: 
(a) The Scratch Method.—A drop of the virus is placed on the skin. If 
supplied in a capillary tube, it should be pushed through the small rubber 
bulb which accompanies it, wiped with alcohol, and the end broken off; the 
other end is broken off in the bulb and the contents expelled (Fig. 171). 
The under surface of the arm is grasped with the operator’s left hand 
so as to stretch the skin where the virus has been placed; with a sterile needle 
or other suitable instrument a scratch not deep enough to draw blood is 
made through the drop of virus, parallel with the humerus and about F 
inch long (Fig. 172). , 
The virus is then gently rubbed in with the side of the needle or other 
smooth, sterile instrument, as a toothpick (Fig. 173). Some blood-tinged VACCINATION AGAINST SMALLPOX 
773 
Fig. 171.—Technic of Vaccination. 
Second Step: Applying a drop of virus to the skin. 
Fig. 172.—Technic of Vaccination. 
Third Step: Abrading the skin with a needle through the virus. 
serum may ooze through the virus, but this should not be sufficient to wash 
the virus out of the wound. 774 
PROPHYLACTIC ACTIVE IMMUNIZATION OR VACCINATION 
The virus should be allowed to dry on the abrasion for at least fifteen 
minutes. 
Fig. 173.—Technic of Vaccination. 
Fourth Step: Rubbing in the virus followed by exposure for fifteen minutes for drying 
The vaccinal wound should now be covered with a square of sterile gauze 
fastened to the skin by adhesive strips (Fig. 174); if other shields are em- 
Fig. 174.—Technic of Vaccination. 
Fourth Step- Dressing the wound. 
ployed, care must be exercised in order that the wound shall be properly 
protected against contamination without maceration of the skin. VACCINATION AGAINST SMALLPOX 
775 
Cross-scarification, which is forbidden in Germany, favors the growth of 
anaerobic bacteria under the crust that forms on the surface of the abrasion 
where the resistance is lowered by the action of the virus. The circular 
scarification gives more control over the dosage, and there is no tendency 
to the development of excessively sore areas. 
Usually one inoculation of the virus is sufficient, but in times of threatened 
epidemic two or more inoculations are made at the same time, not only to 
insure a successful result, but rapidly to immunize the patient. It would 
appear that the degree of immunity bears some relation to the number or 
size of the vaccination lesions, and this can readily be understood if the 
infection is local and the body cells are stimulated by a diffusible toxin. 
If, however, the vaccination lesion is but the point of entry of what becomes 
a general infection, then a small lesion should suffice. This point has not 
been definitely settled, but statistics tend to show that persons vaccinated 
in two or more areas develop an immunity more quickly and that this 
immunity is more lasting. 
The subsequent care of the wound is of considerable importance. The 
operation is usually regarded as a trivial one, and justly so, but the lesion 
requires judicious after-treatment instead of being entirely neglected, as 
it so often is. The severe infections are usually attributable to gross and 
careless contamination of the wound. The best possible protection to the 
vaccinial ulceration is afforded by the formation of a hard, solid crust, due to 
desiccation of the contents of the vaccine vesicle and pustule. Unless undue 
inflammation and suppuration set in, such a crust will form. Care must be 
taken that the crust is not subjected to mechanical violence calculated to 
loosen or to detach it. 
Constricting shields are likely to be unsatisfactory. The adhesion of 
the crust to the sleeve or to a piece of protective gauze will often lead to 
forcible decrustation when the sleeve or the gauze is removed. Schamberg 
and myself1 have found that daily applications of a 4 per cent, alcoholic 
solution of picric acid upon the vaccinated area after the first forty-eight 
hours does not interfere with the success of the vaccination, and lessens the 
degree of local inflammatory reaction and constitutional disturbances by 
hardening the epithelial covering of the vaccine lesion, and thereby decreasing 
the liability of extraneous bacterial infection. 
A point to be emphasized is that severe lesions are unnecessary, and are 
usually due to scratching of the vesicle or pustule and consequent introduc- 
tion of dirt. No doubt tetanus bacilli may be introduced in this manner, 
the resulting scab affording the necessary anaerobic conditions for their 
development. 
(b) The Drill Method.—The arm is cleansed as described. A von Pirquet 
drill shaped like a very small screw driver with a moderately sharp end not 
more than 2 mm. wide, is sterilized by dipping in alcohol and flaming. The 
skin is drawn taut, the drill held between the thumb and middle finger and 
with a twisting motion and moderately firm pressure, one or two small 
circular abrasions are made without drawing blood (Fig. 175). A drop of 
virus is then added and gently rubbed into the skin by means of the drill 
or wooden applicator; or the virus may be first placed on the skin and the 
abrasions made. 
This method is especially recommended by Force,2 who makes three 
abrasions and applies virus to each. 
(c) The Multiple Acupuncture Method.-—This method has been especially 
1 The Lancet, London, November 18, 1911, 1397. 
2 Jour. Lab. and Clin. Med., 1916, 2, 597. 776 
PROPHYLACTIC ACTIVE IMMUNIZATION OR VACCINATION 
recommended by Hill.1 The skin is cleansed and three small drops of virus 
are placed at least 1 inch apart. 
The vaccinator’s hand is closed on the arm from behind, so as to draw 
the skin tight in front, and a sewing needle point, held slanting nearly 
parallel with the arm, is pressed against the skin through the drop of vaccine. 
Then it is that 1/1000 inch of the point sticks through the upper layer 
of the skin, carrying the vaccine with it. The needle is instantly withdrawn 
and similar punctures are made beside each other until a dozen are made in 
the space of 1/16 square inch or less. The two other areas are pricked in 
the same manner. The excess of vaccine is wiped away and the arm dressed 
as described. 
(d) Gauze Method.—With nervous patients the abrasion may be made by 
gauze drawn over the index-finger and held by the thumb and other fingers; 
the selected point is briskly rubbed until slightly abraded over an area about 
Fig. 175.—Technic of Vaccination Employing the von Pirquet Skin Borer for Scarifying 
the Skin. 
the size of a thumb-nail. The virus is then deposited and allowed to dry. 
The wound is then dressed in the usual manner. 
(e) Intracutaneous Method.—Wright2 has described a method consisting 
of diluting the virus as ordinarily supplied with an equal amount of sterile 
water and injecting 0.1 c.c. intracutaneously by means of a tuberculin 
syringe and a small needle (gage No. 26). The technic of injection is 
exactly similar to the Schick test. 
(f) Subcutaneous Method.—Goodale’ has recommended the subcutaneous 
injection of about one-half to three-quarters of the usual amount of virus 
supplied in a capillary tube, diluted with sifficient sterile water to make 
1 Brit. Med. Jour., 1917, 1, 189. 
2 Jour. Amer. Med. Assoc., 1918, 71, 654. 
3 Amer. Jour. Med. Sci., 1919, 158, 721. VACCINATION AGAINST SMALLPOX 
777 
1 c.c. Since there is no open wound, dressings are not required and the 
dangers of secondary infection are practically eliminated, the operation, of 
course, being conducted with the usual aseptic precautions. The main 
objection to the method is that a scar does not follow and the presence of a 
typical scar affords the only indisputable evidence of successful vaccination 
in subsequent years. 
Internal administration of vaccine by swallowing the virus generally fails 
and should never be practised. Garrison1 reported that of 25 persons 
given vaccininum by mouth, 20 were subsequently proved to be unpro- 
tected, as they were vaccinated successfully by the ordinary or scarification 
method. Force, likewise, has never observed the slightest immunity fol- 
lowing the internal administration of virus. 
Precautions.—While vaccination is a slight operation, it demands skill 
in performance and care in after-treatment in order to avoid the rare, but 
serious, complications. For the prevention of these complications vaccina- 
tion (a) should be conducted with strictly aseptic technic, (b) cross scarifica- 
tions should not be made, (c) the inoculation should cover a small rather 
than large area, although two or three areas may be inoculated at one time 
and (d) the wound should receive a proper primary dressing, and subsequent 
care to avoid secondary infection. On account of possible fouling by 
perspiration and to lessen the chance of contamination by street dirt, vaccina- 
tion should be performed preferably in winter as emphasized by Elgin.2 
The prevention of secondary infection depends essentially, according to 
Force and Stevens,3 on the use of small multiple scarifications which give 
rise to small vesicles which are not easily broken, and do not foster anaerobic 
conditions and a central slough. 
Age for Vaccination.—Ordinarily children should be vaccinated when 
about six months of age, again at six years, and preferably at puberty. At 
times of threatened epidemics everyone should be vaccinated and even 
children under four months of age. There is no special reason why healthy 
infants under four weeks of age should not be vaccinated. 
Failure of Vaccination.—When primary vaccination fails it is generally 
due to a weak or dead vaccine or defective technic. True natural immunity 
is rare, although repeated vaccinations with inert virus may result in the 
production of some immunity. A common error is to use vaccine that has 
been improperly preserved, usually one that has been allowed to remain at 
ordinary room temperatures for suflicient time to weaken or destroy the 
virus. Deep scarification with bleeding may prevent the absorption of 
sufficient virus. I have known adults to prevent successful vaccination by 
immediately washing the skin and applying a bichlorid dressing. With 
scarification methods, at least, a few minutes should be allowed for absorption 
before the excess of vaccine is removed. 
Subsequent Examinations and Reactions.—With individuals vaccinated 
for the first time and who have never had smallpox, the vaccination should 
be inspected about one week later, at which time the “take” will be well 
developed or clearly a failure. 
With individuals who have been vaccinated before, the site should be 
inspected twenty-four hours later and again on the fifth day. If there is 
redness with or without a papule which has subsided on the fifth day, it is 
considered a reaction of immunity. This reaction may be caused by inert 
or dead as well as by living vaccine, and indicates the presence of antibodies 
1 Jour. Amer. Med. Assoc., 1917, 68, 979. 
2 Amer. Jour. Public Health, September, 1915. 
3 Jour. Amer. Med. Assoc., 1917, 68, 1247. 778 
PROPHYLACTIC ACTIVE IMMUNIZATION OR VACCINATION 
in the tissues; if a vaccine has been employed which is known to be active 
on the basis of successful ‘‘takes” among others, revaccination is unnecessary. 
If redness on the first day is followed by the development of a small 
vesicle by the fifth day, which rapidly subsides, the reaction is designated as 
vaccinoid, which is mild or secondary vaccinia. 
If there is practically no change until the third day, when an areola 
develops, vaccinia will develop. 
The Mechanism of Smallpox Vaccination.—In smallpox the infection is 
general and the virus is distributed to the skin, mucous membranes, and 
internal organs by the blood; this is indicated by the successful inoculation 
of animals with the blood, bone-marrow, and other tissues of the smallpox 
victim. 
In vaccination the living virus of cowpox is deposited in the epidermis 
where it survives, multiplies, and produces the local lesion and some general 
effects designated as vaccinia. After vaccinia the individual is found 
immune to smallpox because the virus of cowpox or vaccinia has induced 
the production of antibodies which destroy and protect against the virus of 
smallpox. 
It is debatable whether the virus of cowpox finds its way into the blood 
and internal organs during vaccination (vaccinia). Some investigators 
claim that this occurs; others claim that it does not on the basis of failure to 
vaccinate animals with the blood and extracts of the internal organs from 
animals with vaccinia. 
The antibody may occur in the blood because the blood or serum of the 
individual recovering from smallpox, as likewise the blood or serum of the 
human being or lower animal after vaccinia when mixed with the virus in a 
test-tube, is known to be able to kill the virus. Furthermore, Jobling1 and 
others, including the writer,2 have found specific complement-fixing and other 
antibodies in the sera after smallpox and vaccinia. 
It is highly probable that antibodies are engendered by epithelial cells 
as well as by the bone-marrow and the tissues of other organs. Doubtless 
the virus produced in the vaccinal lesion enters the body fluids either, in a 
living state or dead. If living the amount present in the blood at any time 
must be small because vaccinal lesions rarely occur in other parts of the 
body, even in eczema, if direct inoculation is prevented. 
Apparently the epithelium is principally immunized. These cells are 
able to destroy the virus of smallpox or cowpox upon direct inoculation 
because antibodies (both destructive and allergic) occur in them. Smallpox 
virus is probably inhaled and enters by way of the lymphatic and vascular 
channels of the upper respiratory tract, unless the epithelial cells of the 
respiratory mucosa are immunized by a previous attack of smallpox or by 
vaccinia. Possibly the antibodies sometimes present in the blood are 
capable of destroying the virus upon intravenous injection, but it is more 
likely that the viruses of smallpox and vaccinia are strictly dermotropic 
and fail to infect if the epithelial cells are immune. 
The Phenomena of Vaccination; Vaccinia.—Immediately following 
vaccination a slight redness appears, which usually subsides rapidly. After 
a short period of incubation—on or about the third, day—a slight red elevation 
makes its appearance, and the lesion begins to burn and itch. On the 
sixth or seventh day the abrasion becomes a small, silvery gray, umbilicated 
vesicle with a sharply raised edge, filled with a clear serum, and surrounded 
by a narrow red areola (Fig. 176). By the tenth day the characteristic 
features are more marked, and the lesion has usually reached its height, being 
1 Jour. Exper. Med., 1906, 8, 717. 
2 Jour. Immunology, 1916, 1, 59. Fig. 176.—Vaccinia (Seven-day Lesion) 
Fig. 177.—Vaccinia (Nine-day Lesion) 
Fig. 178.—Vaccinoid. A Vaccination Scar. 
Fig. 176 shows a seven-day vaccination vesicle. Fig. 177 shows a nine-day vaccination vesicle 
just before pustulation occurred. Fig. 178 shows a recent vaccination scar with pitting and radiation, 
also a three-day “vaccinoid” or “immunity reaction” with a small vesicle. VACCINATION AGAINST SMALLPOX 
779 
accompanied by a burning sensation and an almost uncontrollable desire 
to scratch (Fig. 177). The areola is now quite angry in appearance, and 
numerous minute vesicles are seen on its surface. By the twelfth day the 
areola is smaller, the contents become turbid and commence to dry, so that 
a few days later a scab has formed that drops off in another week or two. 
About the fifth day the child becomes restless and irritable and shows 
a slight elevation of temperature. These symptoms may become more 
pronounced until the end of the second week, when they subside rapidly. 
Precautions should be taken to prevent scratching and infection of the 
vesicle. The old-time “beautiful arms,” with well-marked cellulitis and 
adenitis of neighboring glands, were largely due to secondary infection, and 
are not at all necessary in the process of vaccination. Evidence would tend 
to indicate that the vesicle is the typical lesion of both smallpox and vaccinia, 
and that the pustules are simply infected vesicles. Ordinary surgical care 
will do much to rob vaccination of its discomfort and to render the operation 
a most harmless one. 
Immunity Reaction; Vaccinoid.—The results of a vaccination, therefore, 
can be inspected and verified on or about the seventh to the ninth day. 
With persons who have been vaccinated successfully on a previous occasion 
the vaccinated area may show a slight areola at the end of twenty-four hours, 
with or without a papule, which subsides in seventy-two hours. This is 
called a “reaction of immunity,” and is due to the presence of antibodies 
against the virus. Or a small, itchy, burning papule may form, which 
develops into a small vesicle maturing on the fifth or sixth day, and then 
rapidly subsiding, constituting the reaction known as vaccinoid (Fig. 178). 
Occasionally vaccination is followed by the appearance of various eruptions. 
Vaccination Scar.—The appearance of the scar varies according to its 
age and to the degree of tissue destruction. The physician is not infrequently 
requested to examine a person and determine if the scar is satisfactory 
evidence of successful vaccination. The typical good scar is circular, and 
about the size of a ten-cent piece, with smooth, white, and depressed center 
and a raised border. The border shows numerous radiations, and the entire 
scar may show little pits of former hair-follicles when the lesion was 
sufficiently destructive to remove the upper portion of the corium 
(Fig. 178). A burn or an ordinary pyogenic infection may leave scars quite 
similar to those of vaccination, and vaccination scars may show wide 
variation, but the circumscribed character, the raised border with radiations 
and depressions, and the appearance of having been stamped on the skin 
by a sharply cut die are quite characteristic. Poor scars are those that were 
said to have been the result of vaccination, but in very many instances they 
are so indistinct as to make it difficult or impossible to recognize them as 
vaccination marks. 
Vaccination Certificates.—Physicians should exercise great caution in 
the issuing of certificates certifying to immunity. Mere failure to secure 
typical vaccinia by vaccination or repeated vaccination is insufficient. If 
there is typical vaccinia or a satisfactory scar of previous vaccination with 
the devemopment of an “immunity reaction” or vaccinoid on revaccination, 
a certificate may be issued. Otherwise revaccination should be repeated 
until one of these three reactions are observed before a certificate may be 
properly issued. Force, who has given this subject particular study, has 
stated that he has never yet seen the naturally immune individual, and that 
there will be no failures if cold, fresh, potent vaccine is used. 
Revaccination.—One successful vaccination does not necessarily confer an 
absolute immunity against smallpox, and failure to recognize certain limita- 780 
PROPHYLACTIC ACTIVE IMMUNIZATION OR VACCINATION 
tions in this respect has done harm by enabling antivaccinationists to create 
a distrust in the minds of the ignorant by pointing to individual instances 
of failure. That a person who has once been vaccinated may afterward 
suffer from smallpox is undoubted, but usually the vaccination was per- 
formed many years previously, and in any case the disease, when it does 
occur, is relatively mild (varioloid). 
There can be no doubt that the immunity gradually diminishes. Force1 
found that approximately 14 per cent, of persons successfully vaccinated 
ten to twenty years previously and showing scars, developed vaccinia and 
38 per cent, vaccinoid, upon revaccination. After twenty years approxi- 
mately 35 per cent, developed vaccinia and 27 per cent, vaccinoid. Vac- 
cinia among these individuals upon revaccination indicated considerable 
loss of immunity to cowpox virus, although such individuals exposed to 
smallpox frequently escape, whereas persons never successfully vaccinated 
almost invariably contract the disease. Vaccinoid indicated still less loss 
of immunity. 
Perhaps seven years may be taken as the average period of fairly com- 
plete protection. Children should be vaccinated within the first year after 
birth, revaccinated upon entering school, and again after leaving it. If 
smallpox is prevalent, all persons should be vaccinated, regardless of the 
fact that they have previously been vaccinated. Only those who have had 
smallpox may be excused. If, as a matter of fact, persons are still immune, 
the vaccination will not “take” and no harm is done, whereas, if it succeeds, 
such persons will have the satisfaction of knowing that their immunity has 
been increased. Hence it cannot be too strongly emphasized that not only 
vaccination, but revaccination, is indicated to protect the individual and 
society against smallpox. Dwyer claims that a person should be revaccinated 
repeatedly in succession until he fails to react; even a slight “take” would 
indicate incomplete immunity. 
Occasionally a non-immunized person refuses to “take,” but vaccination 
should be repeated three or four times, as failure is not infrequently due to 
old and inactive virus, and the actual number of persons absolutely insus- 
ceptible is very small indeed. 
Risks of Vaccination.—When vaccination is properly performed with a 
good virus, the risk of permanent injury to life or limb is almost negligible. 
1. Tetanus.—The most serious of the injuries that have been attributed 
to vaccination is tetanus. The tetanus bacillus and its spores are so wide- 
spread in nature that opportunity presents itself for contamination of the 
vaccinal wound and the virus itself. Every precaution should, therefore, 
be taken in the preparation of virus, and it is especially important that 
physicians and laymen should realize the necessity for observing ordinary 
care, and at least ordinary cleanliness, in the treatment of the vaccinal 
wound. 
It is exceedingly difficult to determine the source of infection in each 
case of vaccinal tetanus, but experimental investigations would tend to 
indicate that the virus itself is seldom, if ever, the vehicle of infection. 
Anderson2 has stated that in experiments carried out by Francis3 on monkeys 
and guinea-pigs with vaccine lymph purposely contaminated in the labora- 
tory with countless numbers of tetanus spores, it was found impossible to 
communicate tetanus in this manner, although the vaccinations were more 
severe than the ordinary vaccinations performed on man, in that several 
1 Jour. Lab. and Clin. Med., 1917, 3, 220. 
2 Public Health Report, 1915, 30, No. 29. 
3 Bull. Hyg. Lab., 95, August, 1914. VACCINATION AGAINST SMALLPOX 
781 
places were inoculated and the areas abraded were large. Anderson, further- 
more stated that his “conclusion from these experiments is that it is almost 
impossible to produce tetanus, even with vaccine virus that contains tetanus 
germs in it, by the simple act of vaccination.” 
Since 1909 there were approximately 100,000 specimens of vaccine 
virus examined in the Hygienic Laboratory, particularly with the purpose 
of determining the presence of tetanus germs or their products. The 
vaccine was purchased in the open market, and the examinations were made 
as thoroughly as it was possible to make them. To use Anderson’s words: 
“We have never succeeded in finding any evidence of the presence of the 
tetanus organism or its products in vaccine virus.” 
It would appear, therefore, that virus prepared according to modern 
methods and with all recognized precautions is safe. In view of the inci- 
dence of tetanus following other injuries, it is reasonable to conclude that 
most cases of vaccinal tetanus are secondary wound infections, and therefore 
largely preventable. 
2. Syphilis.—With the use, years ago, of humanized virus, and par- 
ticularly in the days of arm-to-arm vaccination, extremely rare instances 
of the transmission of syphilis have been known to occur. Since, however, 
the use of calf virus, which is the virus exclusively employed in this country, 
such an accident is absolutely impossible, as calves are not susceptible to 
luetic disease. 
3. Cancer, foot-and-mouth disease, tuberculosis, and various chronic 
skin eruptions have been attributed to vaccination by its opponents; none 
of these claims has, howrever, been substantiated. 
Protective Value of Vaccination.—Of the value of the protection afforded 
by vaccination against smallpox there can be no doubt in the minds of right- 
thinking and unbiased persons. The history of the world before the days 
of universal vaccination shows the wide prevalence of smallpox and its 
fearful mortality. It was regarded as a disease of childhood, owing to the 
fact that all contracted it at the earliest opportunity, and, accordingly, 
smallpox was the cause of a fearful infant mortality. 
At the present day, owing to the general employment of vaccination, 
smallpox is a rare disease, but its very rarity has fostered a certain degree 
of false security and carelessness in carrying out the process. A young 
and new generation of non-vaccinated persons in any community is a source 
of danger, and, accordingly, sporadic cases are often blessings in disguise, 
from the fact that, when they appear, compulsory vaccination is then 
instituted and large numbers seek revaccination. 
In Germany, where vaccination is compulsory, smallpox is now a com- 
paratively rare disease. While the general death-rate from all diseases is 
lower in England and Wales than in Germany, the smallpox mortality is 
seven and one-half times the mortality of Germany, and, proportionate to 
the population, over thirteen times. 
Austria, one of Germany’s neighbors, had, for the twenty years following 
1874, almost thirty times as high a smallpox mortality as Germany. During 
this period 239,800 persons perished in Austria from smallpox alone. 
Physicians should carefully impress upon those over whom they have 
any influence the necessity of being vaccinated, for only a thoroughly 
vaccinated population can solve the problem of exterminating smallpox 
as an epidemic disease. 
Unfortunately the propaganda of the antivaccinationists is capable of 
doing much harm. In 1921 the Public Health Service1 reported more than 
1 Public Health Rep., 1921, 36, 2555. 782 
PROPHYLACTIC ACTIVE IMMUNIZATION OR VACCINATION 
16,000 cases of smallpox in eight states alone in which histories were fur- 
nished in 1920; while, from information supplied by only seven states, more 
than 18,000 cases have been reported with history during the first six months 
of the present year. In Minnesota, for example, 8238 of these smallpox 
cases were actually reported. The real lesson, however, is to be found 
in the statistics of vaccination for this formidable array of patients. More 
than two-thirds of the entire 34,000 afflicted had never been successfully 
vaccinated. About one-twentieth of them had been vaccinated more than 
seven years before the attack; and of the remainder, the histories were in 
most cases uncertain with reference to their vaccination status, or were 
not obtained. Only 2 per cent, of the patients were actually reported to 
have been vaccinated within seven years prior to the attack. Again the 
lesson to the public is clear and imperative—and there is little left for the 
consolation of the antivaccination cult, which is doubtless responsible 
directly or indirectly for some of this unnecessary suffering. 
For those seeking a clear, concise, scientific, and calm statement of 
facts bearing upon vaccination, I can recommend the pamphlet prepared 
by Schamberg1 on “Vaccination and Its Relation to Animal Experimenta- 
tion,” and not a few physicians who have seen the ravages of smallpox are 
ready to shoulder the challenge of the late Osier: 
“I would like to issue a Mount Carmel-like challenge to any ten un- 
vaccinated priests of Baal. I will go into the next severe epidemic with ten 
selected vaccinated persons and ten selected unvaccinated persons. I 
should prefer to choose the latter—three members of Parliament, three 
antivaccination doctors, if they could be found, and four antivaccination 
propagandists. And I will make this promise, neither to jeer nor to gibe 
when they catch the disease, but to look after them as brothers, and for the 
four or-five who are certain to die I will try to arrange the funerals with 
all the pomp and ceremony of an antivaccination demonstration.” 
VACCINATION AGAINST RABIES 
There are but few diseases more dreaded by the laity than rabies, or 
hydrophobia. Tales of the sufferings of infected persons, especially those 
with the furious variety of this infection, characterized by maniacal symptoms 
and dread of water (hydrophobia), have been thoroughly disseminated, 
so that the cry of “mad dog!” on the public streets is sufficient to arouse a 
general state of hysteric excitement in which an otherwise harmless creature 
may be compelled to bite or snap for self-protection. Not all dogs under 
these conditions are mad or infected with rabies, and the bite of an angry 
dog, otherwise normal, is not necessarily dangerous from the standpoint 
of rabic infection. However, almost every one, upon being bitten by a dog, 
will promptly consult his physician, and this is proper and to be encouraged. 
Genuine rabies is an acute infectious disease in which the diagnosis is quite 
readily made, and whenever possible an effort should be made either to 
confirm or to disprove the diagnosis by making an examination of the animal’s 
brain, and if the dog is found to have been free from rabies, this fact should 
be carefully impressed upon the patient, as otherwise the dread of infection 
may weigh heavily upon the patient and lead to distressing nervous dis- 
turbances. 
Etiology of Rabies.—While the infectiousness of rabies has been known 
for a great many years and was proved experimentally by Galateir2 and 
1 Published by the Amer. Med. Assoc., 535 Dearborn Ave., Chicago. 
2 Compt. rend. Acad. d. sc., 1879, lxxxix, 444. VACCINATION AGAINST RABIES 
783 
Pasteur,1 it was not until Negri, in 1903, described certain bodies (Negri 
bodies), seen by him in large nerve-cells in sections of the central nervous 
system, that anything was found that seemed absolutely specific for rabies. 
Negri regarded these bodies as specific for rabies and probably of a pro- 
tozoan nature. Later investigations fully established the diagnostic 
value of these bodies, and their definite characteristic morphology, evidences 
of cyclic development, and staining qualities indicate a protozoan structure 
resembling members of the Rhizopoda, designated by Anna Williams in 
19062 as Neurorrhyctes hydrophobics. 
Rembringer,3 Poor and Steinhardt,4 Bertarelli and Volpino5 have demon- 
strated the filterability of the rabic virus, and Noguchi6 has cultivated from 
both “street” and “fixed” virus very minute granular and somewhat coarser 
pleomorphic chromatoid bodies which, on subsequent transplantation, 
reappeared in the new cultures through many generations and reproduced 
typical symptoms of rabies in dogs, rabbits, and guinea-pigs. 
The virus or parasite is contained in the saliva of the rabid animal, and 
infection is possible when the skin is abraded by bites and scratches. The 
virus travels by way of the nerve-paths to the central nervous tissue, and, 
as in tetanus, the symptoms of the disease are due to involvement of these 
tissues. 
Susceptible Animals.—A large number of animals including human 
beings are susceptible to rabies. In most countries dogs constitute the 
highest percentage because of their large numbers and freedom with conse- 
quent chances of being bitten and infected. Cats, horses, cattle, sheep, 
goats, hogs, chickens, and animals of prey, as wolves, foxes, badgers, prairie 
dogs, and martens also contract rabies when bitten by rabid animals, and 
behave quite similarly to rabid dogs, cattle, and cats. 
Immunity in Rabies.—Rabies is a disease that may attack a large variety 
of the lower animals as well as man. The disease has occurred in practically 
all of the domestic animals including chickens. The lower animals as well 
as man possess no demonstrable degree of natural immunity when the virus 
is introduced in bites. The lower animals, as rabbits and dogs, may prove 
refractory to infection when the virus is injected intravenously, but are 
readily infected by intra-cerebral and intra-ocular inoculations. Once the 
virus has been introduced in sufficient amounts in bites, the disease is in- 
variably fatal in the lower animals. Whether or not human beings inocu- 
lated with virus in bites ever escape the disease in the absence of local 
destruction of the virus in the wound or by vaccination, is not definitely 
known, but the indications are that human beings possess little or no natural 
immunity and are very liable to develop the disease when actually infected. 
After infection has occurred, the disease is invariably fatal in both 
human beings and the lower animals. Our body cells do not appear able to 
produce protective or immunity principles quickly enough to destroy the 
virus unless they are hyperstimulated by injections of vaccine, as in Pas- 
teur’s method of prophylactic vaccination. 
Skepticism Regarding Rabies.—While the exact etiology of rabies is 
unknown, the germ probably being in the Negri bodies, there can be no 
reasonable doubt of the existence of the disease in the lower animals and 
human beings. The disease, however, is not one of human beings, but of 
1 Compt. rend. Acad. d. sc., 1881, xcii, 159. 
2 Proc. New York Path. Soc., 1906, vi, 77. 
3 Ann. de l’Inst. Pasteur, 1903, xvii, 834; 1904, xviii, 150. 
4 Jour. Infect. Dis., 1913, xii, 202. 
5 Centralbl. f. Bakt., Orig., 1904, xxxvii, 51. Bertarelli ibid. 
6 Jour. Exper. Med., 1913, xviii, 314. 784 
PROPHYLACTIC ACTIVE IMMUNIZATION OR VACCINATION 
the lower animals transmissible to human beings. There are many persons, 
including physicians, who doubt the existence of rabies in human beings 
and ascribe the symptoms of so-called cases the result of hysteria and fright. 
The author has seen several cases in human beings—all fatal and Negri 
bodies found in the brains of all—presenting such a terrific and pitiable 
picture of suffering before death, that once seen the disease is never forgotten 
and leaves no room for doubt. 
Discovery of Vaccination.—To Pasteur is due the credit for having dis- 
covered (1880) the fact that the disease may be prevented by conferring 
gradual immunization with increasing doses of the attenuated virus. This 
treatment, with some modification, is now used with evident success in all 
parts of the world. 
In 1885 Pasteur1 first demonstrated the protective value of vaccination 
against rabies in dogs by means of the injection of emulsions of the dried 
spinal cord and medulla oblongata of rabic rabbits. These experiments 
were very successful and in a letter Pasteur stated: “ I have not yet dared 
to treat human beings after bites from rabid dogs; but the time is not far 
off, and I am much inclined to begin by myself—inoculating myself with 
rabies, and then arresting the consequences; for I am beginning to feel very 
sure of my results.” 
Finally on Monday, July 6, 1885, there was brought to Pasteur, a little 
Alsatian boy, Joseph Meister, and a man, Theodore Vone, who had been 
bitten by a rabic dog. The man had been bitten on the arm and Pasteur 
noting that his clothes had wiped off the dog’s saliva and that his shirt- 
sleeve was intact, reassured him and sent him back to Alsace. But the 
boy had been severely bitten and Pasteur experiencing great emotion at 
the sight of the fourteen wounds, finally consented to try vaccination after 
being encouraged to do so by his friends Vulpian and Grancher. The 
substance chosen for the first dose was a fourteen-day cord, the injection 
being made two and half days after the bites. Twelve injections were given, 
the final one being on July 16th, with some medulla only one day old and 
fatal for rabbits in seven days; the following days were indeed anxious ones 
for Pasteur, but nothing developed and the child returned home well and 
happy. On October 14th, of the same year, a shepherd lad, Jupille, had 
been severely bitten by a rabid dog while courageously defending a group 
of children; six days later Pasteur commenced the treatment which likewise 
proved successful and induced the great discoverer to establish a “service” 
in Paris for these inoculations. On November 9th he wras induced to treat a 
little girl of ten years who had been severely bitten on the head thirty-seven 
days previously. Pasteur surmized that it was too late for vaccination, 
but reluctant to stand by without an attempt, gave a few injections. True 
to his prediction, rabies developed with a fatal result which affected Pasteur 
in a profound manner. 
Incubation Period of Rabies.—The period of incubation in dogs, or the 
time elapsing between the time of injury and the first symptoms, is quite vari- 
able, ranging from fourteen to sixty days, although it may be as short as ten 
days. As in tetanus, this period depends upon—(a) the location of the injury; 
(b) the quantity or dose of virus; (c) the kind of animal responsible for the 
injury. Bites about the face and fingers, especially if they are deep and 
lacerating, are especially dangerous; bites about the back and lower limbs, 
especially if superficial, are much less dangerous, and accompanied by a 
longer period of incubation. It is to be remembered that bites may be 
infectious as early as nine days before the dog shows well-marked symptoms 
1 See the Life of Pasteur by Mrs. Devonshire, Doubleday, Page & Company, New York. VACCINATION AGAINST RABIES 
785 
of the disease. Not infrequently an animal is observed to be surly and 
snappy for several days before rabid symptoms develop, and a bite during 
this time should be regarded as dangerous. 
Reichel1 has given the period of incubation of rabies in the domestic 
animals, including man, as follows: 
Man, .fourteen to ninety days. 
Dogs, fourteen to sixty days. 
Cats, fourteen to sixty days. 
Cows, fourteen to eighty days. 
Horses, twenty-one to ninety days. 
Swine, sheep, and goats, twenty-one to sixty days. 
Birds, fourteen to sixty days. 
Rabbits, nine to ninety days. 
Guinea-pigs, eight to sixty days. 
It is an exceedingly rare occurrence for man or any of the lower animals 
to develop rabies following the one hundredth day from the time of exposure 
or infection, but there are rare instances in which exceptionally long periods 
of incubation have been observed. 
Only about 16 per cent, of human beings bitten by rabid animals and 
untreated appear to contract rabies. Since the establishment of the Pasteur 
treatment of the disease the percentage of developed cases after bites is 
much lower—about 0.46 per cent. 
Physicians may be called upon to diagnose rabies among the lower 
animals in relation to bites of human beings; the main symptoms in the 
dog and cat are as follows: 
Symptoms of Rabies in the Dog and Cat.—Two forms or stages are 
usually recognized—the furious or irritable and the paralytic or dumb. 
The furious stage is usually followed by the paralytic stage; but not infre- 
quently the paralytic stage may follow the premonitory symptoms without 
the furious stage. 
First a decided change in the disposition of the dog is observed. The 
animal becomes increasingly restless. There is a gradual and progressive 
paralysis of the muscles of the pharynx which later extends to the jaw and 
tongue. As a result there is great thirst with strenuous attempts to swallow 
and an increased flow of saliva which is whipped into froth. The appetite 
becomes depraved with attempts to swallow straw and dirt; a change occurs 
in the bark. If free, the dog will start out and travel long distances, usually 
returning greatly changed in appearance. The hind legs become paralyzed 
and the paralysis gradually extends forward, initiating the paralytic stage, 
which ends in death. The whole course of the disease is from two to eight 
days. 
Or the paralytic stage may follow the premonitory symptoms at once. 
The animal seeks a secluded spot and is rapidly overtaken with drowsiness. 
The paralysis of the muscles of the pharynx and face rapidly progresses 
and death usually follows in two to three days. 
The cat generally hides in some dark place where it may die in a day or 
two unobserved. As a rule, however, there is danger for human beings 
and especially children, the animal jumping up to the face and inflicting 
severe wounds with its teeth and claws and biting itself. The voice is lost, 
the animal mewing hoarsely; the appetite is lost, there is rapid emaciation, 
thirst, and progressive paralysis, ending in death in a few days. 
Diagnosis and Management of Rabies.—The bite of an infected animal 
may give the disease before the animal shows symptoms. Fifteen days is 
1 Amer. Vet. Rev., January, 1911 786 
PROPHYLACTIC ACTIVE IMMUNIZATION OR VACCINATION 
the longest time recorded between a bite and the appearance of symptoms 
in the dog. Therefore, if an animal is kept under observation at least three 
weeks without developing symptoms, he may be pronounced free from 
suspicion. 
In case of bites by an animal suspected as infected with rabies an effort 
should be made to quarantine the animal for a sufficient period to determine 
whether or not rabies is present. If he is killed at a time wrhen there are 
none or indefinite symptoms the diagnosis may be always in doubt because 
Negri bodies may not be found in the brain at this time. It is probably 
very rare, indeed, for an animal to recover from a slight attack of rabies; 
usually the disease progresses to a fatal termination. 
Local Treatment.—As a general rule, all animal bites should receive 
surgical attention. Wounds produced by animals clinically rabid should 
be cauterized at once with fuming nitric acid or pure phenol. This is done 
to offset the delay in securing the Pasteur treatment, and because there is 
evidence to show that thorough cauterization of the wound is in itself highly 
beneficial. 
Disposition of the Animal.—The animal should be promptly destroyed 
if it is unmistakably rabic not only to prevent further damage, but in order 
to make a microscopic diagnosis by examination of the brain for Negri 
bodies. In destroying the animal care must be exercised not to injure the 
brain. This examination is highly important and should never be omitted, 
for if it shows the absence of bodies, this fact should be carefully impressed 
upon the patient, as there is no doubt that a neurotic element, amounting 
in many instances to actual hysteria, may cause considerable harm to the 
patient even though he is definitely free from rabic infection. 
The whole dog may be packed in ice and shipped at once to a central 
laboratory, or the head alone may be cut off close to the shoulders and 
packed in ice or equal parts of water and glycerin and promptly shipped. 
The brain should not be disturbed. When it reaches the laboratory the 
diagnosis should be made at once by “smear” preparations and sections of 
the brain demonstrating the characteristic Negri bodies in the large ganglion- 
cells, and confirmed by inoculating emulsions of the brain into guinea-pigs 
or rabbits. The latter requires from ten to twenty days before the result 
may be known. A negative animal inoculation test is better evidence than 
a negative smear or section; obviously, these examinations are to be made 
only by properly trained persons. 
If the animal is not manifesting symptoms of rabies, do not kill; to do so 
may leave the diagnosis in doubt. The animal should be confined in quaran- 
tine for at least three weeks. If infected with rabies it will surely die within 
this period. If, on the contrary, he remains well throughout this period, 
he was not rabic and the wound was not infected with rabies virus. Dogs 
infected with rabies rarely recover; there are no “light” cases with recovery. 
Diagnostic Value of Negri Bodies.—According to Park, the value of the 
smear method of diagnosis may be summarized as follows: 
1. Negri bodies demonstrated, diagnosis is rabies. 
2. Negri bodies and suspicious bodies not demonstrated in fresh brains, 
not rabies. 
3. Negri bodies not demonstrated in decomposing brains, uncertain. 
4. Suspicious bodies in fresh brains, probably rabies. 
Indications for Vaccination.—If the animal was clinically rabid, the 
Pasteur treatment should be commenced as soon as possible, without waiting 
for the laboratory report, if this will be delayed for several days. Wounds 
about the face and hands, where there is no clothing to retain the infectious VACCINATION AGAINST RABIES 
787 
saliva, should receive intensive treatment; otherwise the milder course of 
immunization will suffice. 
Formerly it was a general rule to send the patient to a regular Pasteur 
institute, where there are special facilities for the proper treatment of these 
cases. At the present time, however, the physician may treat the patient 
at home. Several large manufacturing firms are prepared to ship by mail 
the fresh daily treatments properly preserved and ready for administration. 
The Pasteur treatment should be given to every patient bitten by a 
rabid animal or by one suspected of being rabid. Not all persons are neces- 
sarily infected, even by bites of rabid animals, but this should not unduly 
influence the physician, for he will not have fulfilled his duty unless he 
carefully explains the etiology of the disease and advises immediate immuni- 
zation. Aside from the actual benefits of the treatment, the mental effect 
upon the patient is deserving of consideration. Even slight wounds by 
rabid animals, wherever their location, should be regarded as dangerous, 
and the Pasteur treatment advised in addition to routine cauterization. 
With severe bites of angry but not necessarily clinically rabid dogs, 
the treatment depends upon various factors. In any case the wound should 
be thoroughly cauterized and the animal carefully guarded (not killed) for 
two or three weeks. If rabid symptoms appear, the animal should be 
destroyed, the cerebral tissues examined, and the Pasteur treatment of the 
patient begun. 
Disposition of Bitten Animals.—All animals bitten by a rabid animal 
should, of course, be promptly destroyed; even those bitten by a dog sus- 
pected of being rabid should be destroyed or closely guarded until a definite 
diagnosis can be reached. In England, where strict laws are enforced 
relative to the muzzling and control of dogs, rabies is relatively infrequent, 
and it is especially urged that similar measures be adopted and enforced in 
our own communities, particularly during the summer months. Valuable 
dogs may receive one or two doses of vaccine and quarantine for at least six 
weeks; this method of vaccination of dogs, however, is still in the experi- 
mental stage and great care is required. 
Principle of the Pasteur Treatment (Active Immunization) of Rabies. 
—This method is based upon the principle of stimulating the production 
of rabic antibodies by injecting attenuated or modified virus during the 
period of mcubation, so that the virus introduced into the wound is de- 
stroyed, neutralized, or its effects neutralized, while the virus itself is finally 
destroyed. Pasteur worked out this theory and established its truth by 
experiments upon the lower animals before applying the treatment to man. 
By passing the virus through a series of rabbits the period of incubation 
is shortened to about six to seven days, and at the same time its patho- 
genicity for man is actually diminished (virus fixe). By drying the tissues 
containing the fixed virus attenuation is secured, so that it is easily possible 
so to modify the virus that it cannot produce rabies in man, but yet is able 
to produce the specific antibodies. The Pasteur treatment is, therefore, a 
process of active immunization with emulsions of a tissue (spinal cords of 
infected rabbits) in which the virus has been attenuated by a process of 
drying and desiccation. The early doses consist of highly attenuated cords, 
and succeeding doses become gradually more potent, as is usual in the 
technic of any method of active immunization. 
After the adoption of Pasteur’s method, modifications in the preparation 
of this vaccine soon appeared. Hogyes,1 for example, claimed that Pasteur’s 
method of drying spinal cords did not modify the virus, but simply de- 
1 Nothnagel’s Spec. Path. u. Therap., 1897. 788 
PROPHYLACTIC ACTIVE IMMUNIZATION OR VACCINATION 
stroyed it and that the same result could be obtained by using dilutions of 
fresh cord. Ferran has likewise used unchanged fixed virus and in this 
country the method has been advocated by Proescher.1 Other methods of 
attenuating fixed virus have been used, such as exposure to the action of 
heat, cold, gastric juice, glycerin, and phenol. 
Mixed Serum and Vaccine Immunization of Rabies.—Fermi,2 in Italy, 
has advocated immunization with mixtures of antiserum and vaccine. 
The vaccine is a 5 per cent, emulsion of the most virulent fixed virus made 
from the brain of a rabbit or dog and rendered avirulent with 1 per cent, 
phenol. The antiserum is prepared by immunization of the horse. One 
part of serum is mixed with 3 parts of vaccine, allowed to stand on ice 
for twenty-four hours, and 3 c.c. injected subcutaneously in the morning 
and 3 c.c. in the evening. This is repeated for five or ten days, and then 
vaccine alone is injected for an additional twenty or 
fifteen days. Similar serum-vaccine mixtures have 
been employed by Marie, Remlinger, and Babes. 
Contraindications to Vaccination.—There are 
practically no contraindications. Some individ- 
uals are more apt than others, however, to ex- 
hibit reactions of hypersusceptibility to the in- 
jections. Pregnant and nursing women may be 
immunized with safety. 
Pasteur’s Method of Preparing Rabies Vaccine. 
—As a preliminary, it is necessary to prepare or 
obtain virus fixS. This may generally be procured 
from a laboratory, or may be prepared by pass- 
ing street virus from the medulla of a rabic cow 
or dog through a series of young rabbits. After 
from 30 to 50 passages the incubation period is 
gradually reduced to six or eight days (virus 
fixi). 
1. From an animal succumbing the day or night 
before, a piece of the floor of the fourth ventricle 
measuring about 2 cm. in length is emulsified in 
1 c.c. of sterile bouillon, and 3 or 4 drops of this 
emulsion are injected beneath the dura of a normal 
rabbit. In large institutes two or more rabbits 
are injected daily. The inoculation is quickly and 
easily performed by trephining a small area in the 
median line of the forehead and injecting the emul- 
sion beneath the dura mater with a syringe. The 
whole operation must be carried out in an aseptic 
and practically painless manner. 
2. After inoculation the animals are placed in clean cages; in from six 
to eight days paralytic symptoms of rabies appear, followed in three to four 
days by death. The hair is then sprayed with a solution of lysol and the 
skin removed. The cord and brain are then extracted under aseptic pre- 
cautions. The cord is severed just below the medulla, a portion is snipped 
off into sterile bouillon for culture, and then divided into two equal pieces 
which are suspended by sterilized silk threads in a sterile glass jar containing 
flakes of caustic potash (Fig. 179). The medulla is placed in a sterile dish, 
and is used to continue the inoculations, as was previously described. A 
Fig. 179.—Preparation or 
Rabies Vaccine. 
Note the cords suspended 
within the jar by means of 
sterile silk threads; sticks of 
sodium hydroxid to absorb 
moisture and hasten desic- 
cation. 
1 Arch. Int. Med., 1911, 8, 351. 
2 Supplement to Ann. d. l’lgiene, 1916-17, 26. VACCINATION AGAINST RABIES 
789 
postmortem examination is finally performed, and any cord in which the 
animal is found diseased or in which the culture of the cord shows bacterial 
contamination is rejected. 
The jars with suspended cords are kept in a special room at a temperature 
of about 20° to 25° C. 
3. After a suitable period of drying pieces of cord are prepared for 
injection. This is performed in various ways at different laboratories; no 
attempt at exact dosage is made. In the New York Board of Health labora- 
tories 1 cm. of the cord is thoroughly emulsified in 3 c.c. of sterile saline 
solution, the process being conducted in an aseptic manner. If the material 
is to be shipped, an addition of 20 per cent, of glycerin and 0.5 per cent, of 
phenol is made. 
4. After twenty-four hours’ drying the cord is known as one-day cord; 
after two days, two-day cord, etc. Pieces of cord cut off at any time and 
put into glycerin will retain the same strength for several weeks. 
Harris Method of Preparing Rabies Vaccine.—Harris and Shackell1 
have advocated a method of preparing vaccine by drying rabic brain and 
cord tissues of rabbits in a vacuum at a temperature of —18° C. Harris2 
later modified this by freezing the material with carbon dioxid snow or 
liquid air followed by drying in a vacuum and pulverizing. The material 
is then sealed free from moisture and kept in an ordinary ice-box. By this 
method one rabbit furnishes sufficient material for the immunization of 
20 to 25 patients; it is, therefore, economical and less laborious to prepare 
than the Pasteur method. 
D’Aunoy3 has recently described this method of preparing vaccine and 
has reported favorably upon its use in the prophylaxis of rabies. 
Terrell’s Method of Preparing Rabies Vaccine.—Based upon the obser- 
vations of Cummings that killed virus engendered a greater degree of immun- 
ity than the original Pasteur method, with no danger of paralysis or other 
untoward effects, Terrell4 has described the following method of preparing 
the vaccine or virus sterilized with phenol: 
Take 1 gram each of cord and brain, grinding with Wedgewood mortar 
and pestle, and add, drop by drop, distilled water until there is a thick 
homogeneous suspension; then add larger amounts until 25 c.c. of distilled 
water are added, and then add slowly 25 c.c. of distilled water containing 
1 gram of phenol crystals. This should be added, a small amount each 
time, and thoroughly mixed and the process repeated until the 25 c.c. are 
added, for if added too fast, a curdled-looking specimen results instead of a 
smooth homogeneous emulsion. Place in incubator and keep there for 
twenty-four hours at 37° C. This gives a 2 per cent, phenol dilution which 
will always kill the virus in less than twenty-four hours. At the end of 
twenty-four hours bring the volume up to 200 c.c. with sterile distilled 
water, which gives a final dilution of brain and cord, one to one hundred, 
while the phenol is 5/10 of 1 per cent. 
Terrell claims that the same amount of immunity produced with twenty- 
one doses by the original Pasteur method, can be produced with twelve 
2 c.c., doses by this method, while with twenty-one doses by this method 
there is produced three to four times the amount of immunity obtained 
with the original method. 
Administration of the Vaccine.—Injections are given with a sterile 
1 Jour. Infect. Dis., 1911, 8, 47. 
2 Jour. Infect. Dis., 1912, 10, 369; 1913, 13, 155. 
3 Jour. Infect. Dis., 1921, 29, 261. 
4 Jour. Oklahoma State Med. Assoc., August, 1920. 790 
PROPHYLACTIC ACTIVE IMMUNIZATION OR VACCINATION 
syringe. The abdominal region or arm of the patient are bared, a spot 
touched with tincture of iodin, wiped with alcohol, and the injection given 
subcutaneously. Keirle does not vary the dose according to the age, both 
the old and the young receiving the same dose. 
Cases of severe injury, such as deep bites about the face and fingers, 
should be rapidly immunized (intensive treatment); in other cases the treat- 
ment may be mild (mild treatment); the tendency at present is to give all 
cases the intensive treatment. The uniform dose of cord emulsion, prepared 
as just described, is 1 to 3 c.c. The series of inoculations may be as follows 
according to the Hygienic Laboratory method of starting treatment with 
six-day cord instead of fourteen-day cord as formerly used: 
Schedule of Inoculations in Immunization Against Rabies 
Day. 
Intensive Treatment. 
First (three injections) 
8, 7, 6 day cord 
Second (two injections) 
6, 5 day cord 
Third (two injections) 
5, 4 day cord 
Fourth 
3 day cord 
Fifth 
3 day cord 
Sixth 
2 day cord 
Seventh 
2 day cord 
Eighth 
Ninth . . 
5 day cord 
Tenth 
Eleventh 
Twelfth 
3 day cord 
Thirteenth 
3 day cord 
Fourteenth 
.. . : 2 day cord 
Fifteenth 
2 day cord 
Sixteenth 
4 day cord 
Seventeenth 
3 day cord 
Eighteenth 
2 day cord 
Nineteenth 
3 day cord 
Twentieth 
Twenty-first 
1 day cord 
With the phenol-killed vaccine of Terrell the duration of treatment and 
the number of doses administered depends upon the severity and site of the 
injury. In mild cases where the exposure to infection has been through 
contact of recent wounds with infected saliva, the treatment is one dose 
daily for twelve days. In cases presenting ordinary wounds of parts rela- 
tively free from nerve filaments or remotely located from nerve trunks or 
centers, the treatment is one dose daily for fifteen days. In case of severe 
bites on the face and hands, multiple small bites, tooth punctures or scratches 
of those parts rich in nerve filaments, and deep lacerated wounds of any part 
of the body, the treatment is one dose daily for twenty-one days. 
Reactions.—The injections are usually followed by slight soreness. A 
reaction of erythema and edema with itching, pain and tenderness, and 
sometimes with malaise and mild fever may occur; according to Geiger1 
these are especially likely to develop on the seventh and eighth days and 
again on the fifteenth and sixteenth days of the treatment. These reactions 
are probably in part traumatic and in part anaphylactic reactions to the 
injected nervous tissue of the rabbit. 
Paralysis and polyneuritis sometimes occur. Neijio2 has reported 24 
cases in a total of 19,800 persons treated in the Pasteur Institute of Buenos 
Aires, and in 4 the consequences were fatal. He is inclined to think that the 
1 Jour. Amer. Med. Assoc., 1916, 67, 1518. 
2 Semana med., 1917, 24, 10. VACCINATION AGAINST RABIES 
791 
paralysis may be prevented by great care in the preparation of the virus 
and warns against the practice of shortening the course of treatment which 
may introduce increasing amounts of virus too quickly. 
With phenol-killed virus Terrell states that paralyses do not develop 
during vaccination. 
Results.—According to reliable statistics, the mortality of rabies without 
the Pasteur vaccination is about 16 to 20 per cent.; with the vaccination 
the average mortality is about 0.46 per cent. The mortality of those bitten 
about the face or head is about 1.25 per cent.; of those bitten on the hand, 
0.75 per cent.; of those bitten on other parts of the body, a little over 0.25 
to 1 per cent. In the Pasteur Institute of Paris only such persons are 
treated as have been lacerated, so that the virus has gained entry into the 
wounds. Viala1 reports that during the year 1915 as many as 654 persons 
were treated, with a single death. Taking into consideration only those 
cases in which the diagnosis of rabies has been confirmed in the animal by a 
competent examiner, the mortality of the cases treated at the Pasteur 
Institute in Paris for the past ten years, and covering the treatment of nearly 
6000 persons, is only 0.6 per cent., which, compared to the average mortality 
of 16 per cent, without vaccine treatment, speaks most favorably for the 
value of Pasteur’s antirabic immunization. 
The bites of wolves are more fatal than those of dogs, the mortality 
being about 10 per cent, in spite of the intensive treatment, and about 
40 to 60 per cent, without treatment. 
Terrell states that of more than 550 cases treated with phenol-killed 
vaccine there have been no cases of rabies. 
When symptoms of rabies have appeared, the treatment is unavailing. 
Antirabic serums have been prepared by immunizing animals, such as 
sheep and horses, and these should be tried in human patients presenting 
symptoms, but the results in general have not been uniformly encouraging. 
Immunization of Dogs.—Active immunization may be applied to dogs 
for the direct prevention of rabies as developed by the investigations of 
Semple2 and Umeno and Doi.3 A phenolized vaccine of virus prepared of 
rabbit brain is commonly employed, a single subcutaneous injection of 5 to 
10 c.c. being given. This method has been found satisfactory in Japan 
where the incidence of rabies is large. The duration of the immunity is 
not definitely known. The method is recommended for the treatment of 
all dogs in an infected district; however, after a dog has been bitten by a 
rabic animal one dose of vaccine is insufficient and it is advisable to destroy 
the animal. Eichhorn and Lyon4 have confirmed the work of Umeno and 
Doi, finding that one large dose of phenolized fixed virus protects against 
large doses of street virus. 
Duration of Immunity After Vaccination.—There is very little data 
available on this subject, but the immunity in man following a complete 
course of vaccine injection probably does not last more than one or two 
years; the same is probably true of dogs vaccinated against rabies as de- 
scribed above. Therefore, a person bitten by a rabid animal one or more 
years after immunization should be reimmunized with a complete course of 
injections. 
1 Ann. de l’Inst. Pasteur, 1916, xxx, 422. 
2 Memoirs Med. and Sanit. Dept. Gov. India, 1911, N. S., No. 44. 
3 Kitasato, Arch. Exp. Med., 1921, 4, No. 2. 
4 Jour. Amer. Vet. Assoc., 1922, 61, 38. 792 
PROPHYLACTIC ACTIVE IMMUNIZATION OR VACCINATION 
VACCINATION AGAINST TYPHOID AND PARATYPHOID FEVERS 
Our knowledge of vaccination in typhoid fever begins with the work of 
Pfeiffer and Kolle.1 These observers, in 1896, immunized two volunteers 
with heat-killed cultures, and by complete laboratory investigations demon- 
strated the identity of the immunity following an attack of the disease with 
the artificial immunity produced by inoculation. At about the same time 
Wright,2 of London, inoculated 2 men with killed cultures, and a year later 
published the results of the successful vaccination of 17 persons. In 1896 
he continued the work in India, where 4000 soldiers were inoculated, with 
encouraging results. Later, during the Boer War, Wright and Leishman 
treated 100,000 men, and the results, while good, were not encouraging, 
due, as pointed out later by Leishman,3 to the fact that the vaccine was 
damaged during its preparation by overheating. Since 1904 an improved 
vaccine has been used among the British troops in India in ever-increasing 
quantities, with uniformly good results. 
Antityphoid vaccination was begun in the United States Army in 1908, 
the vaccine being prepared by Major Frederick F. Russel. Its value has 
been established so clearly that vaccination is now compulsory. The 
results obtained in the army, have had considerable influence in establishing 
a wide-spread general confidence in antityphoid inoculation. 
Immunity to Typhoid and Paratyphoid Fever.—Typhoid and paratyphoid 
fever are diseases only of human beings. The anthropoid apes have been 
infected experimentally by Metchnikoff and Besredka4 and Gay and Clay- 
pool6 have produced a syndrome resembling some phases of typhoid fever in 
rabbits by the intravenous injection of the bacilli, but these diseases do not 
occur spontaneously among the lower animals. 
Some human beings appear to possess a certain degree of natural immun- 
ity; at least there are several instances on record where groups were equally 
exposed and very probably swallowed about the same amount of infectious 
material, in which some contracted the disease and others did not. Further- 
more, it is well known that some individuals suffer with very light infections 
suggesting the presence of some natural resistance, while others have little 
or no evidence of resistance. City dwellers as a group appear to possess 
more resistance than country folk, who appear to be particularly prone to 
infection when coming to a city. Vincent6 has found that the Arabs possess 
more natural immunity than foreigners in Algiers, and it is also believed by 
some observers that the Japanese may possess a certain degree of racial 
immunity. 
One attack of typhoid fever does not protect against subsequent attacks 
as thoroughly as occurs in the acute exanthemata as smallpox, measles, and 
scarlet fever. The exact percentage of recurrent attacks cannot be stated 
because the evidence of the exact nature of the alleged first attack is fre- 
quently uncertain, but various observers have stated that in 1 to 3 per cent, 
of cases of typhoid fever there is a history of a previous attack. Recurrent 
attacks generally occur late in life (after thirty years of age) and during 
epidemics, when the bacilli are probably of enhanced virulence, indicating 
that the immunity conferred by one attack is relative and not absolute. 
Persons who have had typhoid fever are still susceptible to the para- 
1 Deutsch. med. Wchn., 1896, xxii, 735. 
2 The Lancet, London, September 19, 1896, 907; Brit. Med. Jour., January 30, 1897, 16. 
3 Jour. Roy. Army Corps, 1909, 12, 136. 
4 Ann. de l’lnst. Pasteur, 1911, 25, 193. 
6 Arch. Int. Med., 1913, 12, 613. 
6 Ann. de PInst. Pasteur, 1911, 25, 455. VACCINATION AGAINST TYPHOID AND PARATYPHOID FEVERS 
793 
typhoid fevers, although very probably not to the same extent as individuals 
who have never had typhoid fever. 
Various antibodies are to be found in the blood during and for a variable 
period after typhoid and paratyphoid fever. These are largely agglutinins, 
bacteriolysins, opsonins, and complement-fixing antibodies. The agglutinins 
have received most study; they usually disappear within the first year after 
recovery unless a carrier state supervenes. Posttyphoid and paratyphoid 
immunity is apparently cellular, that is, the cells remain sensitized and 
capable of producing antibodies rapidly upon stimulization by reinfection. 
Preparation of Typhoid and Paratyphoid Vaccine.—Based upon general 
principles, the vaccine should be prepared of typhoid bacilli as little changed 
by heat or chemicals as possible. Russel has prepared the army vaccine 
with a single avirulent culture which proved by animal experiments and 
laboratory methods capable of producing large quantities of immune agglu- 
tinins and bacteriolysins. As a general rule, however, the vaccine should be 
polyvalent, and particularly in view of the experiments of Hooker,1 who has 
shown by complement-fixation tests consistent antigenic differences among 
some strains of Bacillus typhosus. 
The preparation of the vaccine is comparatively simple. The bacilli 
are grown on agar for twenty-four hours, washed off with sterile normal 
salt solution, standardized by counting the bacilli, and killed by heating 
to 56° C. for one hour. As a matter of safety, 0.25 per cent, of tricresol is 
then added (Russel). The details of the technic are given in the chapter 
on Bacterial Vaccines. Lipovaccines are no longer employed. 
Vaccines should be kept in a cold place and are ordinarily useful for a 
year, but it is better not to use a vaccine more than four to six months old 
because of a tendency to loose in antigenic power. 
Sensitized Vaccines.—Metchnikoff never fully accepted the belief in the 
value of heat-killed vaccines, and was actively concerned with vaccines 
prepared of living bacilli sensitized wTith their immune serum (sensitized 
vaccines). Injection of these vaccines into chimpanzees is not followed 
by any untoward effects, and apparently the bacilli so administered are 
destroyed at once, as they have not been found in the blood, urine, and 
feces. Metchnikoff and Besredka2 have immunized persons according 
to these methods and report excellent results. Obviously, there is some 
reluctance in using a vaccine of living bacilli until extended animal experi- 
ments have proved that they are harmless and more efficient than the 
vaccines of killed bacilli. 
At the present time sensitized heat-killed typhoid vaccine has come into 
use for prophylactic immunization. These were first introduced by Bes- 
redka,3 who claimed that they were less toxic than plain vaccines and that 
the injections were followed by a quicker production of immunity. Cecil4 
has reported that the sensitized dead vaccine usually produces somewhat 
less severe local and general reactions and probably gives just as high an 
immunity. Stoner5 found that the reactions produced by sensitized and 
plain vaccines were about the same, and that the former possessed no dis- 
tinct advantage over the latter. In my experience sensitized vaccine usually 
produces somewhat less severe local reactions and probably is absorbed 
more quickly from the subcutaneous tissues; antibody production (agglu- 
1 Jour. Immunology, 1916, 11, 1. 
2 Ann. de l’Inst. Pasteur, 1911, 25, 193, 867; 1913, 27, 597. 
3 Ibid., 1902, 16, 918. 
4 Amer. Jour. Med. Sci., 1918, 155, 781. 
6 Jour. Immunology, 1916, 1, 511. 794 
PROPHYLACTIC ACTIVE IMMUNIZATION OR VACCINATION 
tinins) is sometimes slightly earlier than engendered by plain vaccines, but 
the differences are inconstant and irregular. 
Gay and Claypool,1 who have made an extended and extensive study of 
typhoid immunization, have advocated a polyvalent, sensitized typhoid 
vaccine sediment for prophylactic immunization against typhoid fever as 
being superior to other forms of typhoid vaccine. Force2 has found that 
this vaccine produces less reaction, and Sawyer3 has found it more protective 
than several other types of commercial vaccine. Mayer and Kilgore4 found 
that it produced less agglutinins than plain vaccine, although the production 
of complement-fixing antibodies was the same. Gay’s vaccine consists of 
the ground sediment of a mixed polyvalent vaccine that has been sensitized 
by an antityphoid serum and then killed and precipitated by alcohol. From 
this ground culture the endotoxins are extracted by carbolated saline solution 
and the remaining sediment of bacterial bodies alone used for prophylaxis 
and in the treatment of typhoid fever. 
For prophylactic purposes 1/10 mg. of dried bacteria, which corresponds 
to an original bacteria count of about 750,000,000, is administered by 
subcutaneous injection every other day until three or four doses have been 
given. This vaccine is claimed to produce a better and more prompt 
immunity response than other vaccines. It has also been used in the treat- 
ment of typhoid fever by intravenous injection. 
Method of Inoculation.—The vaccine is best administered at about 
4 o’clock in the afternoon, so that the reaction appears during the night 
and is least likely to be disturbing. It is well to administer a cathartic the 
day before the inoculation is made. Inoculations should not be given 
during the menstrual period, as the general reaction is likely to be somewhat 
severer at this time. 
The skin over the insertion of the deltoid muscle is touched with tincture 
of iodin and the injection given subcutaneously. Intramuscular injections 
should be avoided, as the reactions are more unpleasant and accompanied 
by unnecessary pain on movement. In making deep injections there is also 
danger of striking a nerve, a proceeding that may be followed by disagreeable 
neuritis. Subcutaneous injections in the neighborhood of the clavicle are 
less likely to be disabling if a local reaction occurs. 
After the injection has been given the iodin is wiped away with a pledget 
of cotton and alcohol; no dressing is necessary. 
The syringe and needle should be sterile. Commercial firms have 
placed the prophylactic on the market in syringes with sterilized needles, 
accompanied with full instructions as to the technic of administration. 
The vaccine should be well shaken before the injection is given. 
When a large number of injections are to be given at one time, a 
single syringe may be used with a large number of sterile needles, a 
separate needle being used for each person. The vaccine may be put up 
in individual ampules or in a stock bottle, the former being preferable. 
The oral administration of vaccine has proved worthless, although Vaillant5 
has recently reported apparent successful prophylactic immunization by 
Besredka’s method, consisting of the oral administration of bile followed by 
killed cultures of typhoid and paratyphoid bacilli by the same route. 
1 Archiv. Int. Med., 1913, xii, 613; 1913, xii, 622; 1914, xiii, 471; 1914, xiv, 662; 1914, 
xiv, 662, 669, and 671; 1916, xvii, 303. 
2 Amer. Jour. Pub. Health, 1913, iii, 750. 
3 Jour. Amer. Med. Assoc., 1915, lxv, 1413. 
4 Arch. Int. Med., 1917, 19, 293. 
6 Ann. d. l’Inst. Pasteur, 1922, 36, 149. VACCINATION AGAINST TYPHOID AND PARATYPHOID FEVERS 
795 
Dosage of Typhoid Vaccine.—Three injections are given at intervals of 
one week. For adults (150 pounds) the doses used in the army have been 
as follows: 
First dose: 500,000,000 bacilli. 
Second dose: 1,000,000,000 bacilli. 
Third dose: 1,000,000,000 bacilli. 
These amounts are contained in 1 c.c. (about 15 minims). Children 
receive doses in proportion to their weight; if the dose cannot be divided 
evenly, it is better to give a little more rather than a little less, for children 
tolerate the injections remarkably well. 
Triple Vaccine of Typhoid and Paratyphoid Bacilli.—According to recent 
experiences in the European armies prophylactic immunization with typhoid 
vaccine does not afford protection against infections with B. paratyphosus 
A and B. paratyphosus B. In this country various investigators have esti- 
mated that about 2 to 4 per cent, of the cases of so-called “clinical typhoid 
fever” are due to paratyphoid infections; apparently the majority of these 
are with B. paratyphosus B. It would appear advisable, therefore, to im- 
munize with these micro-organisms in addition to Bacillus typhosus, and 
particularly members of the Army, Navy, National Guards, and all volunteer 
organizations called into service in time of war. 
At the present time this triple vaccine (T A B) is used almost exclusively. 
It is usually prepared in such a way that each cubic centimeter contains 
1,000,000,000 typhoid bacilli and 500,000,000 of each of the two para- 
typhoids. 
The first dose should be 0.5 c.c. and the second and third doses 1 c.c.; 
I generally advise a fourth dose of 1 c.c. 
Tetravaccine of Typhoid, Paratyphoid, and Cholera Bacilli; Pentavaccine 
of Typhoid, Paratyphoid, and Cholera Bacilli and Micrococcus Melitensis 
(Malta Fever).—During the World War Castellani advocated the use of these 
vaccines as mixtures in order to afford some protection against cholera and 
Malta fever as well as against typhoid and paratyphoid fevers. In some 
cases Bacillus pestis was also included for immunization against bubonic 
plague. The vaccines were so prepared that each cubic centimeter con- 
tained 500,000,000 typhoid, 250,000,000 paratyphoid A, 250,000,000 para- 
typhoid B, and 2,000,000,000 cholera. If Micrococcus melitensis was in- 
corporated it was added to the extent of 2,000,000,000; Bacillus pestis, 
500,000,000. 
These vaccines were accordingly quite heavy suspensions; the first dose 
was 0.5 c.c. and subsequent doses 1 c.c. According to Castellani the injec- 
tions were as well borne as the usual typhoid-paratyphoid vaccine. Cas- 
tellani and Taylor1 and Lurie2 have reported favorably upon the protective 
value of these vaccines. 
Frequency and Number of Inoculations.—Ordinarily three injections 
are made at intervals of seven to ten days. If the intervals are less than 
seven days the immunity may not be so high. Dreyer, Gardner, Gibson, 
and Walker3 believe that the intervals between doses should be eighteen or 
twenty days to secure the maximum antibody response. 
Contraindications.—The administration of vaccine is frequently followed 
by some local and constitutional reaction. The later may be accompanied 
by a temporary lowering of general vitality and resistance. Accordingly, 
it is possible that individuals with an existing infection, as active tuberculosis, 
1 Brit. Med. Jour., 1917, 2, 356. 
2 Brit. Med. Jour., 1916, 1, 45. 
3 Lancet, April 6, 1918. 796 
PROPHYLACTIC ACTIVE IMMUNIZATION OR VACCINATION 
may be injured by repeated and exceptionally severe or prolonged con- 
stitutional reactions. Discretion should be exercised in this as in other 
therapeutic procedures. There is no reason to suppose that typhoid vaccine 
is any more potent for harm than vaccines in general. 
(a) In active tuberculosis with constitutional symptoms (fever and 
tachycardia) and loss of weight, typhoid vaccine should not be administered. 
In arrested and inactive pulmonary tuberculosis the vaccine may be given, 
but if the individual is below average weight, the doses should be reduced. 
Baldwin and L’Esperance1 have observed that tuberculous animals were 
apparently unharmed by typhoid vaccine. Indeed, the non-specific effects 
of vaccine upon nutrition and metabolism may prove advantageous. In 
the presence of a typhoid epidemic tuberculosis, either active or latent, is not 
a contraindication to typhoid immunization, according to Russel and Nichols.2 
(b) Under ordinary conditions typhoid vaccine should not be given 
during acute infections, as tonsillitis, rhinitis, bronchitis, or more severe 
infections. Medlar,3 however, found that there was no increased suscep- 
tibility to streptococci in animals following typhoid vaccination and the 
main contraindication under these conditions is the possible increased 
illness of the patient due to added constitutional reaction engendered by the 
vaccine. 
(c) Advanced chronic maladies, as nephritis with deficient elimination, 
and probably diabetes, should not be immunized under ordinary conditions. 
id) Pregnancy is not a contraindication according to Guerin and his 
colleagues.4 However, it is possible that the reaction may be injurious 
during the last few weeks of pregnancy, and certainly the vaccine is con- 
traindicated if albuminuria or evidences of pregnancy toxemia are present. 
(e) As a general rule it is advisable to omit injections during or just 
preceding menstruation and especially if the history is one of painful men- 
struation in underweight individuals. Lamb5 has found disturbances in 
menstruation engendered by typhoid vaccine, but none were of much import, 
and menstruation is not an absolute contraindication under ordinary con- 
ditions. 
Reactions.—Persons in poor physical condition are more likely than 
the robust to experience disagreeable after-effects. 
The local reaction consists of a red and tender area about the size of the 
palm of the hand lasting about forty-eight hours. Occasionally the edema 
and pain may be more marked, but abscess formation is practically unknown. 
The general reaction, when present, gives rise to headache, malaise, and 
sometimes to fever, chills, and occasionally to nausea, vomiting, or diarrhea. 
The severe reactions are not alarming and disappear quickly. 
The inoculated person should abstain from severe exercise for the fol- 
lowing twenty-four hours and rest; in the great majority of instances our 
soldiers have not been inconvenienced and were able to continue with their 
routine duties. 
The following tables, compiled by Russel,6 show the proportion of re- 
actions in adults and children: 
1 Jour. Immunology, 1916, 2, 283. 
2 Jour. Amer. Med. Assoc., 1921, 76, 177. 
3 Jour. Amer. Med. Assoc., 1918, 71, 2146. 
4 Gyn. et Obstet., 1920, 1, 217. 
6 Arch. Int. Med., 1913, 5, 565. 
6 Jour. Amer. Med. Assoc., 1913, lx, 344. VACCINATION AGAINST TYPHOID AND PARATYPHOID FEVERS 
797 
Dose. 
None. 
Mild. 
Moderate. 
Severe. 
First 
68.2 
28.9 
2.4 
0.3 
Second 
71.3 
25.7 
2.6 
0.2 
Third 
78.0 
20.3 
1.5 
0.1 
PERCENTAGE OF GENERAL REACTIONS IN ADULTS (128,903 DOSES) 
PERCENTAGE OF GENERAL REACTIONS IN 359 CHILDREN, TWO TO SIX- 
TEEN YEARS OF AGE 
Dose. 
None. 
Mild. 
Moderate. 
Severe. 
First 
73.54 
24.51 
1.67 
0.28 
Second 
86.26 
11.99 
1.75 
Third 
92.56 
0.38 
1.06 
A comparison of these tables shows that the general reaction is much 
more infrequent or milder in children than in adults, even after the first 
dose; after the second and third doses the difference is more marked. 
In former years considerable stress was laid upon the possibility of a 
negative phase following the inoculation, during which a person was believed 
to be more susceptible to infection. This is now believed by Leishman, 
Russel, and others of extended experience to be incorrect, the more general 
belief being that inoculations may be made and are especially indicated 
during epidemics of the disease. 
Duration and Degree of Typhoid Immunity.—It should be emphasized 
that immunity following typhoid immunization is not absolute, and an 
immunized person cannot afford to neglect ordinary precautions against 
infection. A lowered state of general body health or a large dose of infec- 
tious material may at any time result in infection. 
The prophylactic treatment should be used in conjunction with well- 
known sanitary precautions in order to obtain the best results. 
The immunity is apparently manifest soon after the first and second 
doses have been given. The duration is not known definitely. From the 
rich experience of the British Army in India Colonel Firth1 concludes that 
immunity begins to decline in about two and one-half years after inoculation. 
However, even after four and five years the typhoid rate among the inocu- 
lated is, estimated roughly, one-fourth that of unprotected troops. 
Revaccination.—Since the duration of immunity is difficult to determine 
it is advisable to revaccinate whenever typhoid fever breaks out in a com- 
munity. This is especially true of soldiers, lumbermen, campers, and the 
like, to whom vaccination may be the sole or most important means of 
defense. Physicians and nurses constantly exposed should be vaccinated 
at least every two years and preferably annually. Children may be vac- 
cinated every two or three years, and especially just before going to the 
country on their summer vacation. 
Tests for Immunity.—Agglutinins and other antibodies (bacteriolysins 
and opsonins) appear in the blood about seven to ten days after the first 
dose of vaccine and reach their maximum about three weeks after the third 
dose. After this time they gradually disappear from the serum. But actual 
immunity persists for longer periods although in diminishing degree. Ty- 
1 Jour. Roy. Army Med. Corps, 1911, xvi, 589. 798 
PROPHYLACTIC ACTIVE IMMUNIZATION OR VACCINATION 
phoid fever generally confers an immunity for life, but the agglutinins and 
other antibodies in the serum may disappear within a year of recovery 
unless the patient is a carrier. Positive typhoidin skin reactions are not 
acceptable as evidences of immunity, as discussed on p. 712. 
The subject may be summarized, therefore, with the statement that 
when typhoid agglutinins and other antibodies are present in the blood in 
amounts definitely above normal (1 : 100 or higher) after vaccination, the 
individual is probably immune to typhoid fever; if the agglutinin titer is 
1 : 30 or lower the individual may or may not be immune. 
Practical Value of Immunization with Typhoid-paratyphoid Vaccine.— 
The value of antityphoid vaccination has been repeatedly and amply demon- 
strated, and there is no question of its power to prevent infection. This 
protection is not absolute; typhoid fever has occurred among the vaccinated, 
but the incidence and mortality are much less than among the unvaccinated. 
Vaccination cannot be used as a substitute for sanitary measures. Se- 
rious contamination of water and milk supplies at their sources can be checked 
by sanitary measures, but contamination of water, ice, milk, and food on a 
smaller scale by typhoid carriers among food handlers and by flies, will 
continue as sources of infection against which vaccination will afford a large 
measure of protection. 
The value of the typhoid prophylactic therapy is best shown in the army, 
where conditions are better controlled than is possible in civilian life. In 
1911, of a division of United States troops, about 20,000 men along the 
southern boundary, only 2 cases of typhoid fever developed and both re- 
covered. During this same period of time 49 cases were reported in the city 
of San Antonio, with 19 deaths. The soldiers mixed freely in the city, ate 
of fruits and vegetables, drank of the same water, and in this manner were 
freely exposed, although the sanitary conditions in the camp were excellent. 
In 1898, during the Spanish War, there were assembled at Jacksonville, 
Florida, 10,759 troops, among whom there were certainly 1729 cases of 
typhoid, and including the suspected cases, this figure reached 2693 cases, 
with 248 deaths. This camp continued about as long as that in 1911, the 
climatic conditions and water supplies being practically the same, but the 
sanitary conditions were bad. The remarkable difference in the typhoid 
rate cannot, however, be reasonably explained by perfect camp sanitation, 
and the results in 1911 leave no doubt as to the value of antityphoid vac- 
cination. 
The results of vaccination in the Great War have been tabulated by 
Russel.1 From April 6, 1917, to November 11, 1918 approximately 4,000,000 
men served in the army; during two full years there were 1065 cases of typhoid 
fever, or one among every 3765 men. In the Spanish War there was 1 among 
every 7 men (141 per 1000). In 1917 and 1918 the total number of deaths 
was 156, or 1 death among 25,641 men. In the Spanish War there was 1 
death among every 71 men (14 per 1000). 
Compulsory vaccination in the army began in 1911; the following table, 
compiled by Russel, shows the morbidity and mortality rates for a period 
years preceding and following: 
1 Jour. Amer. Med. Assoc., 1919, 73, 1863. VACCINATION AGAINST TYPHOID AND PARATYPHOID FEVERS 
799 
RATE OF TYPHOID FEVER IN THE ARMY AND IN THE CORRESPONDING 
AGE GROUP IN CIVIL LIFE FOR THE PAST EIGHTEEN YEARS 
Year. 
Army. 
Civil Deaths from Typhoid 
Fever: Age Group, Twenty 
to Twenty-nine Years. 
Rate per Thousand 
of Population. 
Number of 
cases. 
Ratio per 
thousand. 
Deaths. 
Ratio per 
thousand. 
Total. 
Males. 
1900 
531 
5.75 
60 
0.43 
0.46 
1901 
594 
9.43 
78 
0.64 
0.42 
0.54 
1902 
565 
8.58 
69 
0.86 
0.40 
1903 
348 
5.82 
30 
0.28 
0.35 
1904 
247 
5.62 
12 
0.27 
0.33 
1905 
193 
3.57 
17 
0.30 
0.32 
1906 
347 
5.66 
15 
0.28 
0.32 
1907 
208 
3.53 
16 
0.19 
0.28 
1908 
215 
2.94 
21 
0.23 
0.28 
19091 
173 
3.03 
16 
0.28 
0.23 
1910 
142 
2.32 
10 
0.16 
0.27 
0.34 
19112 
44 
0.85 
6 
0.09 
0.23 
1912 
18 
0.31 
3 
0.04 
0.18 
1913 
4 
0.04 
0.18 
1914 
7 
0.07 
3 
0.03 
0.15 
1915 
8 
0.08 
0.18 
0.17 
1916 
25 
0.23 
3 
0.03 
0.12 
0.15 
1917 
297 
0.44 
23 
0.03 
0.11 
0.14 
1918 
768 
0.30 
133 
0.05 
0.09 
0.11 
During the Great War there were 213 deaths from typhoid fever in the 
United States Army from September 1, 1917 to May 2, 1919; if the conditions 
of the Civil War had prevailed Russel calculated that the death-rate would 
have reached 51,133, and under the conditions of the Spanish War, the rate 
would have reached 68,164. 
In France the morbidity was reduced from approximately 7 cases per 
1000 men in 1914 to less than 0.1 per 1000 in 1917, when triple vaccination 
was well under way. Vincent has calculated on the basis of 4,000,000 or 
5,000,000 soldiers sent to the French front during the thirty-eight months 
of the war, that if vaccination had not been practised the number of cases 
of typhoid would have exceeded 1,000,000 with 145,000 deaths. In all the 
armies at the front Vincent calculated that vaccination reduced the mor- 
bidity at least 7 times and the mortality 8.5 times less numerous than during 
peace times, and this does not take into account the aggravating influences 
of the war. 
As stated by Russell, it is evident “that antityphoid vaccination, carried 
out as it was by a personnel which had not been carefully trained in its 
administration, gave a high degree of protection to our forces under the 
conditions of hurried mobilization and of warfare, and reduced the rate 
not only below the rates for previous wars but also below the rate found in 
civil life in some of the older states where the entire population is protected 
by all the sanitary measures of modern life.” 
Russel records that in Hawaii, during September, 1917, there was a 
water-borne epidemic of typhoid fever of the classic type traced to a Japanese 
1 Voluntary vaccination against typhoid. 
2 Compulsory vaccination against typhoid. 800 
PROPHYLACTIC ACTIVE IMMUNIZATION OR VACCINATION 
laborer employed until taken sick in the construction of a water-supply 
system. The incidence of cases developing among 4085 vaccinated indi- 
viduals was 13.45 per 1000 and the deaths 0.97; among 812 unvaccinated 
persons the incidence was 55.41 and the deaths 8.02. 
Paratyphoid fever has never been as frequent as typhoid fever and until 
the outbreak of the World War vaccination in the army was not practised. 
In 1917 the rate among the vaccinated and unvaccinated was 0.03 per 1000 
as compared with a rate of 0.14 per 1000 of population among young men of 
corresponding age in civil life. In 1918 the rate in the army was 0.05 per 
1000, but this was still less than half the civil rate (Russel). 
Triple vaccination is now employed almost exclusively in civil practice 
as well as in the army and navy. Craig,1 Davison,2 Dryer, Gardner, Gibson 
and Walker,3 and others have shown that the results, so far as can be judged 
by agglutinin curves, are as good as follows immunization with typhoid 
vaccine alone, and neither the general nor the local reactions are rendered 
more severe. 
Excellent results have been reported by Spooner, Hachtel and Stoner, 
and others as to the prophylactic value of typhoid immunization in hospital 
training-schools for nurses, insane asylums, and other public institutions. 
Recommendations.—In view of the satisfactory results obtained in the 
army, typhoid-paratyphoid vaccination is now obligatory on all members 
of the army and navy corps. Protection of the individual by immunization 
is the only measure of protection independent of surroundings and effective 
under all conditions. 
Typhoid-paratyphoid inoculation in civilian practice cannot be as wide- 
spread or as readily performed as vaccination against smallpox, as the 
prophylactic must be administered subcutaneously and more than one dose 
is necessary. 
1. Our various state and city boards of health should endeavor to 
educate the laity, and, if necessary, offer the prophylactic free of charge 
in order to build up a vaccinated community as far as this is possible. 
2. Persons coming in intimate contact with typhoid patients, such 
as physicians, nurses, and attendants in hospitals, should be immunized. 
Hospital authorities are justified in making typhoid-paratyphoid vaccination 
obligatory on all applicants for admission to training-schools. 
3. All inmates of asylums, homes, and other public institutions under 
forty-five years of age should be immunized and the state should be ready, 
if necessary, to furnish the vaccine. 
4. The physician and nurse should urge vaccination upon all the members 
of a family when there is a typhoid patient among them, also all persons in 
a community where typhoid fever is epidemic or threatened. 
5. The physician should especially advise immunization of those about 
to leave their homes for a vacation in some neighboring seashore or mountain 
resort; also upon travelers visiting small towns and rural districts, persons 
living in towns with unsafe water-supplies, and persons dependent upon 
shallow wells for water-supply. 
6. In times of epidemics of typhoid fever the physician should urge 
vaccination and revaccination of all over whom he has influence. Thorough 
vaccination with proper sanitary conditions offers the best hope of eradicating 
this dreaded disease. 
1 Jour. Amer. Med. Assoc., 1917, 69, 1000. 
2 Arch. Int. Med., 1918, 21, 437. 
3 Lancet, 1918, 1, 498. VACCINATION AGAINST TYPHUS FEVER 
801 
VACCINATION AGAINST TYPHUS FEVER 
Immunity in Typhus Fever.—Studies in the immunity of typhus fever 
are very incomplete owing to the uncertain state of our knowledge regarding 
the etiology of the disease.1 The Plotz bacillus, typhi-exanthematici, is not 
now accepted as the primary etiologic agent, although it may be an impor- 
tant organism of secondary infection analogous to the relationship of strepto- 
cocci to scarlet fever. Wolbach, Todd, and Palfrey, believe that Rick- 
ettsia prowazeki is the cause of typhus fever transmissible by lice. For 
some peculiar and unknown reason certain strains of Bacillus proteus bear 
some relationship to the disease on the basis of the development of agglu- 
tinins for them in typhus fever (see Weil-Felix reaction). 
The disease occurs spontaneously only in man; experimentally it has 
been successfully transmitted to some of the monkeys and probably to 
guinea-pigs. There are no evidences of racial immunity—all persons, 
irrespective of race, age, and sex, appear to be extremely susceptible and 
especially if the general health and resistance have been undermined by 
privation, disease, and general misery. 
One attack of the disease generally confers an immunity for the balance 
of life and there is some evidence to the effect that the serum of a recovered 
individual may prove effective in the treatment of the disease, suggesting 
the presence of immunity principles in the blood for a short time at least 
after recovery. 
Prophylactic Immunization.—Preliminary studies upon prophylactic 
immunization of persons with Bacillus lyphi exanlhematici, a Gram-positive 
pleomorphic bacillus secured in blood-cultures from persons suffering with 
typhus fever by Plotz, Olitsky, and Baehr,2 seem to indicate that a vaccine 
of this bacillus is capable of reducing the incidence of the disease, although 
it does not produce an absolute immunity to typhus fever (Plotz, Olitsky, 
and Baehr3). The vaccine is polyvalent and is sterilized by heating to 58° 
to 60° C. from half an hour to one hour. The suspension is so diluted that 
each cubic centimeter contains about 2,000,000,000 bacteria, and is pre- 
served with 0.5 per cent, phenol or tricresol. Three injections, consisting 
of 0.5, 1, and 1 c.c. respectively, have been given in five- or six-day intervals. 
In a subsequent investigation of vaccination with this bacillus carried 
out in Europe these investigators reported encouraging results.4 
McCoy and Neill5 employed the Plotz bacillus in active immunization 
of monkeys subsequently infected with typhus virus (Mexican), but with 
negative results. Schultz6 states that there is no conclusive evidence that 
this bacillus is the cause of typhus fever. Wolbach, Todd, and Palfrey7 
were unsuccessful in cultivating it from 13 cases in Poland. It may be 
stated, therefore, that the specific etiologic relationship of the bacillus to 
typhus fever lacks confirmation, and consequently the prophylactic value 
of the vaccine is very doubtful. 
As previously stated, one attack of typhus fever generally protects for 
the balance of life. Apparently immunization requires the activity of 
living virus, as Mollers and Wolff8 were unable to vaccinate with heat-killed 
virus. The results obtained by Mitchell and Richardson9 by injections of 
1 The Etiology and Pathology of Typhus, Harvard University Press, 1922. 
2 Jour. Infect. Dis., 1915, xvii, 1. 
3 Jour. Amer. Med. Assoc., 1916, lxvii, 1597. 
4 Ibid., 1917, 68, 576. 
6 Public Health Reports, 1917, June 1, 841. 
6 Amer. Jour. Med. Sci., 1921, 161, 78. 
7 Internat. Jour. Public Health, No. 2, September, 1920. 
8 Deutsch. med. Wchn., 1920, 46, 484. 9 Lancet, 1, 742. 802 
PROPHYLACTIC ACTIVE IMMUNIZATION OR VACCINATION 
1, 2, and 2 c.c. of sterile defibrinated typhus blood are suggestive of successful 
active immunization, but inconclusive. 
Interest in the possibility of developing a successful form of vaccination 
against lobar pneumonia was greatly renewed during the recent war owing 
to the large number of cases developing in the various camps. 
Immunity in Pneumonia and Other Pneumococcus Infections.—Lobar 
pneumonia is probably the commonest pneumococcus infection in man, and 
susceptibility varies greatly in different individuals. While many of the 
lower animals are very susceptible to virulent pneumococci when the organ- 
isms are injected into the bronchi or directly into the peritoneal cavity or 
blood-stream, spontaneous pneumococcus pneumonia among them, compar- 
able to the human disease, is quite uncommon. 
Natural immunity to pneumonia is probably largely due to local factors 
in the respiratory tract. Acute infections, as rhinitis and bronchitis, pre- 
dispose to pneumonia; likewise sudden and profound chilling of the body and 
exposure to wet and cold. How these factors act in reducing natural immun- 
ity is not definitely known. It may be that phagocytosis by wandering 
leukocytes on the tracheal and pharyngeal mucosa is interferred with; that 
the normal expelling activity of the cilia is reduced or that the natural 
bactericidal activity of the blood is diminished. Rosenow, among others, 
as a result of blood-culture work, has endeavored to place pneumonia among 
the bacteremic diseases with secondary localization of the pneumococci in 
the lungs. Very probably this may occur in some cases. Blake and Cecil, 
however, have shown in recent studies that the pneumococci invade the 
tissues at some point or points near the hilus of the lung, spreading subse- 
quently throughout the lobe by way of the interstitial framework and 
lymphatics. Under these conditions pneumonia is a direct localized disease; 
patchy or bronchopneumonia is almost certainly a direct infection from the 
bronchi. 
Recovery from pneumonia and other pneumococcus infections is appar- 
ently brought about by antibodies in the blood and phagocytosis. Bull has 
shown the influence of agglutinins gathering pneumococci in the blood-stream 
in clumps and facilitating their destruction by leukocytes; Cohen and Heist 
have demonstrated the direct pneumococcidal activity of the whole blood 
as a factor of natural and acquired resistance, and many investigators have 
demonstrated the presence of opsonins and their important relation in 
aiding phagocytosis of virulent pneumococci. In addition to these immunity 
principles, enzymes derived from the leukocytes and plasma probably aid 
in the destruction of pneumococci; the products of digestion of the inflam- 
matory exudate may have a similar destructive effect. 
It is well established clinically that one attack of pneumonia does not 
confer lasting immunity against subsequent attacks. Indeed, individuals 
having had pneumonia are commonly believed to be rendered somewhat 
more susceptible. For this reason doubt has been entertained regarding 
the possibility of successful vaccination. Since it has been firmly estab- 
lished that pneumococci are divisible into groups and that the antibodies 
are highly specific for these groups, it is possible that subsequent attacks 
of lobar pneumonia in the same individual may be caused by different 
groups of pneumococci rather than by pneumococci of the same group. 
At least Cecil and Blake1 have observed that experimental pneumococcus 
VACCINATION AGAINST PNEUMONIA 
1 Jour. Exper. Med., 1920, 31, 685. VACCINATION AGAINST PNEUMONIA 
803 
Type I pneumonia in monkeys confers on them an immunity which protects 
them against subsequent infection with Type I pneumococci, but not against 
Type IV. 
Experimental Vaccination Against Pneumococci.—The injection of the 
pneumococcus vaccines into rabbits has been found by Alexander,1 Lister,2 
Howell,3 Cohen and Heist,4 and others to engender the production of specific 
antibodies. Wright5 had previously shown that antibodies may be found 
after the administration of pneumococcus vaccines to men, and Cecil and 
Austin6 have more recently observed agglutinins and protective bodies in 
some individuals after the administration of saline vaccines. Cecil and 
Blake7 found that heat-killed vaccines only sometimes engendered agglu- 
tinins and protective bodies in the blood of vaccinated monkeys, although 
active immunity was frequently present in the absence of these demon- 
strable antibodies, and particularly after vaccination with living vaccines. 
These investigations have at least shown that penumococcus vaccines 
are capable of engendering the production of protective substances which 
are probably highly specific for the type of penumococcus employed. This 
immunity appears sufficient in the lower animals to prevent pneumococcus 
bacteremia and to reduce the severity of the pneumonia following the 
intratracheal injection of virulent pneumococci. Vaccination of monkeys 
with saline pneumococcus vaccines was found by Cecil and Steffen8 to even 
prevent lobar pneumonia under these severe conditions in which large 
numbers of living virulent pneumococci are suddenly introduced into the 
lungs of highly susceptible animals—experimental conditions wdiich do not 
have a strict analogy to pneumonia in human beings—when three injections 
of large numbers of pneumococci were given. These investigators9 have 
also reported that the intratracheal inoculation of monkeys with three doses 
of pneumococcus Type I vaccine renders them completely immune against 
experimental pneumococcus Type I pneumonia. The mere spraying of 
the throat with the vaccine, however, did not produce complete immunity, 
probably because the vaccine was prevented from entering the trachea. 
The immunity appeared to be cellular in character because little or no 
protective substances could be demonstrated in the serum of monkeys 
vaccinated by this method. McCoy, Hasseltine, Wadsworth, and Kirk- 
bride10 observed that about as many cases of pneumonia developed among 
human beings inoculated with 1 c.c. of a lipovaccine containing approximately 
10,000,000,000 each of Types I, II, and III, as among a control group of 
unvaccinated individuals. There is, however, apparently sufficient experi- 
mental data at hand to encourage and justify vaccination against pneumo- 
coccus pneumonia in man. 
Kinds of Pneumococcus Vaccines.—Two kinds of vaccines are being 
employed, namely, a vaccine of pneumococci suspended in saline solution 
and autolyzea at 37° C. until about 95 per cent, of the organisms have become 
Gram-negative as prepared by Rosenow,11 and heated saline suspensions. 
Rosenow regards autolyzed vaccine as less toxic and capable of stimulating 
antibody production in a prompt manner. 
Lipovaccines are not being employed. Wright, Lister and Cecil, and 
1 Jour. Med. Research, 1917, 37, 471. 
2 Publications of the South African Institute for Medical Research, 1916, No. 8. 
3 Jour. Infect. Dis., 1920, 27, 557. 6 Ibid., 1912, 16, 665. 
4 Trans. College Phys. of Phila., 1918. 7 Ibid., 1920, 31, 519, 657. 
5 Jour. Exper. Med., 1918, 28, 19. 8 Ibid., 1921, 34, 245. 
9 Public Health Reports, November 3, 1922, 2735. 
10 Jour. Amer. Med. Assoc., 1922, 79, 1128. 
11 Jour. Amer. Med. Assoc., 1918, 70, 759. 804 
PROPHYLACTIC ACTIVE IMMUNIZATION OR VACCINATION 
Blake have used heated saline suspensions; Cecil has given the following 
method for their preparation: 
“Pneumococci are cultivated from eighteen to twenty-four hours on 
plain or glucose broth. The culture after killing may be used directly or 
it may be centrifuged, and the sediment of bacteria be suspended in physio- 
logic sodium chlorid solution. Finally it is heated at 65° C. for one-half 
hour to kill the pneumococci, and the vaccine standardized by the Wright 
method or by means of a nephelometer. Cultures are taken to test the 
sterility of the vaccine, and tricresol is added to a concentration of 0.3 per 
cent, as a preservative.” 
Dosage and Administration.—For prophylactic purposes at least three 
injections should be given at intervals of seven to ten days. Large numbers 
of pneumococci are required—much larger than of typhoid bacilli for typhoid 
vaccination. 
The vaccine should be polyvalent and prepared of several strains of 
Types I, II, and III and a large number of strains of Type IV; each cubic 
centimeter may contain 2,000,000,000 of each of the four types, making 
8,000,000,000 in all. 
The first dose may be 0.5 c.c. or 1,000,000,000 of each of the four types; 
the second and third doses may be of 1 c.c. or 2,000,000,000 of each of the 
four types. 
The injections should be subcutaneous and given in the same manner 
as typhoid vaccine (see p. 794). 
Reactions.—Cecil states that “in general it may be said that reactions 
to pneumococcus vaccine are similar to those following injections of typhoid 
vaccine. Within twenty-four hours following the injection an area of redness 
and induration appears at the site of the inoculation, which is usually 2 or 
3 cm. in diameter, but may be larger. Occasionally small sterile infiltra- 
tions, which disappear spontaneously, follow the injection of large doses. 
Such reactions appear to be an expression of cutaneous hypersusceptibility. 
“ The constitutional reaction to pneumococcus vaccine is usually insig- 
nificant. In many cases it is entirely absent. In a small percentage of 
cases, vaccination is followed by headache or backache, general malaise, 
chilly sensations, and rise in temperature. These symptoms, however, are 
of short duration.” 
The contraindications are the same as for typhoid vaccination (p. 795). 
Results of Vaccination Against Pneumonia.—Wright1 employed saline 
vaccines in South Africa where a severe type of pneumococcus pneumonia 
has claimed a high morbidity and mortality among the natives of the Rand. 
He thought that vaccination reduced the incidence of the disease during 
the first three months following inoculation. 
Lister,2 working with vaccines of different immunologic types of pneumo- 
cocci occurring in the Rand, has observed much more encouraging results. 
Three subcutaneous injections of 1 cc. of a vaccine containing three types 
and having a total content of 7,000,000,000 of cocci, have rendered a large 
native mine population absolutely resistant to pneumonic infection by any 
of these types during the observed experimental period of nine months. 
During the Great War pneumococcus vaccine was used extensively in 
some of the American Army training camps as a prophylactic against lobar 
pneumonia. At Camp Upton, New York, Cecil and Austin3 vaccinated 
12,519 men against pneumococcus Types I, II, and III. Three or four doses 
1 Drugs and Vaccines in Pneumonia, 1915, Paul B. Hoeber, New York. 
2 Publications of the South African Institute for Medical Research, 1917, No. 10. 
3 Jour. Exper. Med., 1918, 28, 19. VACCINATION AGAINST PLAGUE 
805 
of saline vaccine were given at intervals of five to seven days, with a total 
dosage of 6,000,000,000 to 9,000,000,000 of Types I and II, and 4,000,000,000 
to 6,000,000,000 of Type III. The men were under observation ten weeks 
after vaccination and during that time no cases of pneumonia of the three 
types vaccinated against occurred among the men who had received one or 
more injections of vaccine. In a control of approximately 20,000 men 
there were 26 cases of pneumonia, Types I, II, and III, during the same 
period. Altogether there were 17 pneumonias of all types among the 
vaccinated 40 per cent, of the camp, and 173 pneumonias of all types among 
the unvaccinated 60 per cent, of the camp. 
Following the Camp Upton experiment, Cecil and Vaughan1 undertook 
to vaccinate the troops at Camp Wheeler, Ga., against lobar pneumonia, 
using, in this instance, pneumococus lipovaccine in place of the saline vaccine 
which had been employed at Camp Upton. The dosage employed in all 
cases was 1 c.c. of lipovaccine containing approximately 10,000,000,000 
each of pneumococcus Types I, II, and III. Of the entire camp strength, 
13,460 men, or about 80 per cent., were vaccinated. The men were under 
observation from two to three months following vaccination. During this 
period there were 32 cases of pneumococcus Types I, II and III pneumonia 
among the vaccinated four-fifths of the camp, and 42 cases of pneumonia 
of these types among the unvaccinated one-fifth of the camp. The weekly 
incidence rate for pneumonia of all types among the vaccinated troops was 
conspicuously lower than that for the unvaccinated troops. It will be seen 
from these figures, however, that the results of pneumococcus vaccination 
at Camp Wheeler were not so striking as those obtained at Camp Upton 
the previous winter. This difference was probably due to several factors, 
among which might be mentioned the influenza epidemic, which swept 
over the camp before vaccination against pneumonia had been completed, 
and the use of a lipovaccine instead of a saline vaccine. 
Indications.—Persons susceptible to pneumonia and who suffer with 
repeated attacks may be protected by vaccination; likewise individuals 
who are exposed to cold and wet, as industrial workers, firemen, and police- 
men. Vaccination is certainly indicated in camps and other communities 
when the spread of pneumonia is especially likely to occur. 
VACCINATION AGAINST PLAGUE. 
In view of the frightful infectiousness and mortality of plague, prophy- 
lactic measures are highly desirable. Extermination of rats and ground 
squirrels, especially of the former, about the wharves of seaport cities and 
towns is highly essential, as the disease is transmitted by the fleas of these 
rodents. Aside from sanitary measures, plague vaccine has now been used 
extensively, with encouraging results. 
Immunity in Plague.—Human beings are very susceptible to infection with 
Bacillus pestis; natural immunity apparently does not exist, and especially 
during epidemics when the virulence of the bacillus is enhanced and its 
spread facilitated. Young and old, both sexes, and apparently all races are 
extremely vulnerable. Of course, not all persons contract the disease during 
epidemics and when exposed, but their escape is to be ascribed more to the 
prevention of infection, and especially bites of infected fleas, than to the 
presence of natural immunity principles. Rats, guinea-pigs, ground squirrels, 
and monkeys are likewise highly susceptible; mice, adult rabbits, cats, dogs, 
swune, cows, horses, sheep, and goats generally escape infection, being more 
1 Jour. Exper. Med., 1919, 29, 457. 806 
PROPHYLACTIC ACTIVE IMMUNIZATION OR VACCINATION 
resistant, but all of these animals may be infected artificially by large doses 
of virulent bacilli. 
Recovery from the bubonic type of plague is generally ascribed to the 
production and activity of opsonins and phagocytosis, aided by agglutinins, 
and possibly to a slight extent by bacteriolysins. The pneumonic type is 
generally fatal. One attack of the disease usually confers a lasting immu- 
nity, but second attacks are known to have occurred. 
Preparation of Plague Vaccine.—The Haffkine vaccine is prepared by 
growing pure cultures of Bacillus pestis in flasks of neutral bouillon to which 
a few drops of sterile olive oil or butter-fat have been added, to serve as 
floats from which the surface growth of the bacilli can take place. The 
flasks are cultivated at from 25° to 30° C. for five to six weeks, and are 
shaken every two or three days, by which the hanging, stalactite-like colonies 
are thrown down, so that a new crop of the bacilli can develop in contact 
with the air. 
After growing for six weeks the purity of the culture in each flask is 
tested by subcultures on agar and by direct smears. The masses of bacilli 
are broken up by shaking, and the material is sterilized by heating at 65° C. 
for from one to three hours. Phenol is added to the point of 0.5 per cent., 
and the fluid is tested for sterility by culture. If it is found sterile, it is 
finally poured into small vials of from 10 to 30 c.c. capacity. 
Kolle prepares a vaccine by cultivating the bacillus for two days in 
flasks of agar measuring 10 by 9.5 cm. Each surface of agar equals about 
15 ordinary agar slant cultures, and an agar slant holds about 15 loopsful 
of culture (4 mm. loop). A loop of this size holds about 2 mg. of organisms, 
and, accordingly, a flask of agar contains about 225 loopsful of culture, or 
225 doses of 2 mg. each. Kolle prepares the vaccine in amounts of 0.5 c.c. 
per dose (2 mg. of bacilli), and the growths in each flask are removed with 
112.5 c.c. of sterile normal salt solution. The emulsion is shaken to break 
up clumps, heated for one hour at 70° C., and tested for sterility. It is then 
preserved with phenol or tricresol and placed in ampules containing 0.5 c.c. 
each. 
The German Plague Commission strongly recommended the use of 
twenty-four- to forty-eight-hour-old agar cultures instead of the old bouillon 
cultures employed in the preparation of Haffkine’s vaccine. 
Kolle and Otto1 found attenuated living cultures of pest bacilli more 
effectual for the immunization of monkeys and other animals than killed 
vaccines. Strong2 has likewise employed attenuated cultures in the vaccina- 
tion of human beings, and found them without danger and more actively 
immunizing than killed vaccines, although for obvious reasons general 
immunization with the latter is safer and preferable. 
Lustig and Galeotti prepare a vaccine of the toxic precipitate produced 
by dissolving the bacilli in a 1 per cent, solution of caustic soda and neu- 
tralizing with 1 per cent, of acetic acid. This precipitate is dried in vacuo 
and redissolved in a weak solution of sodium bicarbonate, the dose for 
adults being 0.0133 gm. of solid substance. 
Terni and Bandi inoculate rabbits or guinea-pigs intraperitoneally with 
the bacillus, and just preceding or directly after death they collect the 
peritoneal exudate, in which the organisms are allowed to continue growing 
for twelve hours more. The bacilli are then killed at a low temperature, 
and the fluid thus obtained, after a preservative has been addedj constitutes 
the vaccine. 
1 Deutsch. med. Wchn., 1903, 29, 493; Ztschr. f. Hyg. u. Infectionsk., 1903, xlv, 507. 
2 Jour. Med. Research, 1908, 18, 325. VACCINATION AGAINST CHOLERA 
807 
Dosage.—The ordinary dose of Haffkine’s prophylactic for adult males 
is from 3 to 3.5 c.c.; for adult females, from 2 to 2.5 c.c. Haffkine himself 
has injected larger quantities without resulting harm. He recommends 
giving a second injection after from eight to ten days. The injections are 
given subcutaneously, with a sterile syringe, into the upper arm or elsewhere 
in areas where the skin is not tightly bound down. 
The local and constitutional effects are similar to those in typhoid except 
that they are slightly intensified. The inoculation is followed by redness 
and swelling at the seat of inoculation, and general symptoms in the form 
of rise of temperature and a feeling of illness. The latter pass off in twenty- 
four hours, but the patient should rest during the first day after inoculation. 
Duration of Immunity.—The immunity is apparent a few days after 
inoculation, but is of short duration. In India the protection is believed 
to last at least three months and possibly longer. In times of epidemic the 
inoculations should be repeated at least two or three times a year. The 
brief duration of the immunity is probably one reason why better results 
are not secured. Best results are observed during epidemics when protec- 
tion is afforded for a short time, or until the danger is past. In countries or 
localities where the disease is endemic, persons may refuse repeated inocula- 
tion and thus become susceptible to infection. 
Results.—In the pneumonic variety of plague the prophylactic does 
little or no good, a finding that has also been shown experimentally by 
Strong and Teague.1 
In the bubonic variety Haffkine’s vaccine has in general yielded en- 
couraging results. The protection is not absolute; the. immunity is of 
relatively short duration, and therefore good results are not so readily 
appreciated when the disease is endemic. The mortality among the inocu- 
lated is much lower, i. e., 11 to 41 per cent., as compared with 50 to 92 per 
cent, among the non-immunized. Haffkine summarized his results a few 
years ago as follows: Among 186,797 inoculated persons there were 3999 
attacks, or 1.8 per cent.; among 639,630 uninoculated persons there were 
49,433 attacks, or 7.7 per cent., with 29,733, or 4.7 per cent, of deaths. 
Teague2 has recently summarized the results of vaccination with heat- 
killed vaccines as follows: of 321,621 unvaccinated persons in India, the 
incidence of plague was 34.4 per 1000 and the case mortality 78.6 per cent.; 
among 118,148 vaccinated persons the incidence was 7.96 per 1000 and the 
case mortality 39.5 per cent. 
The Indian Plague Commission a few years ago reported as follows: 
1. Inoculation sensibly diminishes the incidence of attacks of plague. 
It is, however, not an absolute protection against the disease. 
2. The death-rate is markedly diminished by its means, not only the 
incidence of the disease but also the fatality being reduced. 
3. The protection is not conferred on those inoculated for the first few 
days after the injection. 
4. The duration of the immunity is uncertain, but it seems to last for 
a number of weeks, if not for months. 
After the disease has once developed, vaccination is of no avail. When 
there is imminent danger of infection, vaccine and antiserum should be given 
together. 
VACCINATION AGAINST CHOLERA 
Protective inoculation against cholera was first practised by Ferran, 
a Spaniard, in 1884, although little definite knowledge as to the value of 
1 Philippine Jour. Sci., 1913, vii. 
2 Jour. Amer. Med. Assoc., 1921, 76, 243. 808 
PROPHYLACTIC ACTIVE IMMUNIZATION OR VACCINATION 
the procedure resulted from his work. He is said to have used impure 
cultures of bacilli isolated from the feces of cholera patients. Broth cultures 
were prepared, and the living organisms injected subcutaneously, using 
8 drops for the first and 0.5 c.c. for the second and third doses, the injections 
being given at intervals of six or eight days. While his method and results 
have been questioned, he was, however, the first to use a method employed 
later, with some modifications, by Haffkine in India, with good results. 
Immunity in Cholera.—Asiatic cholera occurs spontaneously only in 
man. Rabbits and guinea-pigs may be infected experimentally by virulent 
bacilli and appear to suffer most from the endotoxins, but the disease does 
not occur spontaneously in the lower animals even in cholera districts. 
Not all human beings are susceptible; apparently many possess a suf- 
ficiently high natural immunity to escape infection. Undernourished, 
sickly, and elderly individuals are particularly susceptible. The acidity of 
the gastric juice apparently suffices for the destruction of the bacillus and 
may explain escape from infection in some instances, but when the bacilli 
are swallowed in water and food they may escape destruction in the stomach 
and reach the small intestine where conditions are favorable for growth. 
In some instances bacilli are found in the feces of carriers without clinical 
evidences of cholera; either the intestinal epithelium in these cases acts 
as a barrier or the individual possesses sufficient natural immunity to prevent 
the development of bacteremia and the pathologic changes of infection. 
Recovery from cholera is ascribed to the production of agglutinins and 
bacteriolysins, both of which are to be found in the serum. Doubtless the 
opsonins and phagocytosis are also important in this relation. One attack 
of the disease generally confers an immunity, but this is not absolute and 
second attacks may occur. 
Preparation of Cholera Vaccine.— Haffkine, following Pasteur’s method 
with anthrax, uses two vaccines—a weaker and a stronger—living micro- 
organisms being used in both and injected subcutaneously. Vaccine No. 1 
is weaker, and is obtained by growing the bacilli on agar at a temperature 
of 39° C. Vaccine No. 2 is composed of more virulent organisms, prepared 
by passing the vibrios through a series of guinea-pigs until a strain is obtained 
which is invariably fatal to these animals within twelve, or at least twenty- 
four, hours. Cultures are grown on agar, washed off with 8 c.c. of sterile 
bouillon or saline solution, and administered in doses of 1 c.c., which is 
equivalent to about two loopfuls (4 mm.) or 4 mg. of living bacilli. 
Kolle has shown, however, both by animal experimentation and in the 
human being, that heat-killed cultures are equally good, and that living 
cultures are unnecessary and may be undesirable. 
By this method the vaccine is prepared by cultivating a virulent strain 
on flasks of agar for twenty-four hours, as described in the preparation of 
plague vaccine, and removing the growths with sufficient salt solution so 
that 1 c.c. shall contain one loopful (4 mm.) of organisms (2 mg.). The 
emulsion is then shaken to break up clumps, heated to 53° to 60° C. for from 
one-half to one hour, cultured to determine sterility, and preserved with 
0.5 per cent, phenol. 
During the Great War heat-killed vaccines were used exclusively. 
Strong has proposed the use of a preparation of the nuceloproteins of 
the cholera vibrio. Organisms of high immunizing and peptonizing power 
and of high virulence, are selected for the preparation of the vaccine. One- 
half of a twenty-four-hour culture on agar, 1 per cent, alkaline to phenol- 
phthalein, is suspended in physiologic saline solution. This suspension 
contains approximately 30 to 35 mg. of the bacteria. It is heated for one VACCINATION AGAINST INFLUENZA 
809 
hour at 60° C., and then incubated from three to four days, after which it is 
tested for sterility and is then filtered through a Berkefeld filter. The 
remaining half of the twenty-four-hour agar culture is suspended in saline 
in the same way as the first portion. The organisms are still living in this 
portion. The suspension is shaken thoroughly from three to four days. 
After ascertaining that the growth is pure, this suspension is also filtered 
through a Berkefeld filter. The two filtrates are then mixed in equal pro- 
portion and 0.5 per cent, phenol added. The sterility of each lot of the 
vaccine is tested by animal inoculation and anaerobically. According to 
Strong, in prophylactic inoculation the degree of immunity obtained is in 
proportion to the amount of immunizing substances injected. The irritating 
substances of the cholera vibrios having been removed in the course of the 
preparation of the vaccine by this method, doses equivalent to many times 
the usual prophylactic dose of bacterial suspensions may be employed 
without severe local or general reaction. The adult dose recommended by 
Strong is 2 c.c., injected subcutaneously. Only a single dose is given. 
Dosage.—Kobe’s vaccine is given subcutaneously in two injections 
about a week apart—1 c.c. the first time and 2 c.c. the second time. 
As a general rule the heat-killed vaccines are made up to contain about 
8,000,000 per cubic centimeter. The first dose is 0.5 c.c. and the second 
dose 1 c.c., a third dose of 1 c.c. is given when possible. 
Haffkine’s vaccines are given in the same manner at an interval of five 
days. 
The local effects are usually marked by more or less pain and edema, 
which subside in forty-eight hours. The constitutional effects are not infre- 
quently severe, being marked by malaise, fever (100°-101° F.), nausea and 
vomiting, followed the next day in about 10 per cent, of persons by transient 
diarrhea. Usually all symptoms have disappeared within seventy-two 
hours. 
Results.—Haffkine’s prophylactic vaccine has yielded favorable results 
in India. Powell reports 198 cases of cholera among 6549 non-immunized 
persons, with a total mortality of 124. Of 5778 inoculated persons, there 
were 27 cases, with 14 deaths. Much better results were obtained with 
Kobe’s vaccine, and it is now generally used in preference to the Haffkine 
vaccines. 
Murata, during an epidemic in Japan in 1902, vaccinated 77,907 persons. 
Of these, 47, or 0.06 per cent., developed cholera, and 20, or 0.02 per cent., 
died. Of 825,287 uninoculated persons, 1152, or 0.13 per cent., died. Dur- 
ing a recent epidemic in Russia Franschetti inoculated 11,178 persons. Of 
these, 8 contracted cholera and 1 died. In St. Petersburg, during 1907-08, 
30,000 persons were inoculated. Of these, 12 developed cholera and 4 died. 
Of 10,000 uninoculated persons, 68 contracted the disease. 
During the Great War vaccination against cholera was extensively prac- 
tised and especially in the Balkans. The results reported by Biuwid,1 
Fornet,2 Hueppe,3 Nedrigailoff,4 Liidke,5 and others have been uniformly 
favorable to the extent that the incidence and mortality have been reduced 
among the vaccinated in cholera infected districts and especially when two 
injections were given. Savas6 reports that among the Grecian troops the 
incidence of cholera was 42 per 1000 among those inoculated once and 7 
per 1000 among those who received two injections, while among the unvacci- 
bated the incidence was 93 per 1000. The mortality among the vaccinated 
1 Wien. klin. Wchn., 1914, 28, 169. 
2 Deutsch. med. Wchn., 1914, 40, 1681. 
3 Berl. klin. Wchn., 1915, 52, 1273. 
4 Russk. Vrach., 1915, 15, 169. 
5 Med. Klinik, 1913, No. 43, 1607. 
6 Wien. klin. Wchn., 1914, 27, 1093. 810 
PROPHYLACTIC ACTIVE IMMUNIZATION OR VACCINATION 
(twice) was 10.2 per cent, and among the unvaccianted 27.5 per cent. He 
states that cholera soon disappeared from cholera infected towns after 
dual vaccinations. 
It appears justifiable, therefore, to conclude that vaccination confers 
some immunity. This protection may be apparent after the first dose, 
but is more marked about seven days after the second. The immunity 
conferred is far from being absolute, and it is noteworthy that while the 
prophylactic diminishes the liability of the inoculated person to cholera, it 
has less influence on the mortality when the disease occurs in those who 
have been vaccinated. 
Used in conjunction with modern sanitary regulations, however, Kobe’s 
vaccine certainly proves of value and should be used in combating epidemics. 
VACCINATION AGAINST DYSENTERY 
Immunity in Dysentery.—Both bacillary and amebic dysenteries are 
diseases apparently confined to human beings. Digestive disturbances 
and enteritis are predisposing and the disease is most common among young 
children, old people, and those confined to institutions. During epidemics, 
however, susceptibility is quite general, probably due to the enhanced 
virulence of the micro-organism. 
During the course of bacillary dysentery agglutinins and bacteriolvsins 
are produced and apparently are mainly concerned in the processes of 
recovery; phagocytosis, however, is also active and particularly in the intes- 
tinal wall. 
Antibodies tend to disappear from the blood rather rapidly after recovery, 
and one attack does not confer a lasting immunity; indeed, the disease tends 
to become chronic with frequent exacerbations and recurrent infections. 
The bacteriology of dysentery is discussed in Chapter XL. 
Vaccination.—Ever since the discovery of Bacillus dysenterice by Shiga at- 
tempts have been made to develop a practical means of vaccination. Owing 
to the extreme toxicity of the bacilli and the severe local reactions produced 
by the vaccines, these attempts have usually failed. Shiga1 injected himself 
with one-twelfth of a killed agar slant and the local reaction was so severe 
as to require incision, although agglutinins were produced. Similar local 
reactions were observed by Kruse2 and Rosenthal.3 
Shiga4 then injected killed cultures and immune serum simultaneously, 
followed in three to four days by a larger dose of vaccine without serum. 
He vaccinated 10,000 Japanese in this way, noting a reduction in the mor- 
tality; but the immunity lasted only three to four weeks. Vaillard and 
Dopter5 and Dopter6 have employed sensitized vaccines and obtained what 
they thought were encouraging results, but Liidke7 was unable to confirm 
these observations. 
Gay,8 Liidke,9 and others have shown that vaccines engender the pro- 
duction of specific antibodies in rabbits, but the problem of an acceptable 
method for preparing a non-toxic vaccine remains to be solved. Dean and 
1 Centralbl. f. Bakteriol., 1898, 24, 817, 870, 913. 
2 Deutsch. med. Wchn., 1900, 26, 637; 1901, 27, 386; 1903, 29, 49. 
3 Centralb. f. Bakteriol., Ref., 1905, 36, 23. 
1 Deutsch. med. Wchn., 1903, 29, 327. 
5 Ann. de 1’Inst. Pasteur, 1903, 17, 463. 
6 Ibid., 1909, 23, 677. 
7 Die Bazillenruhr, Jena, 1911. 128. 
8 Univ. Penna. Med. Bull., 1902-03, 15, 307. 
9 Centralb. f. Bakteriol., 1905, 34, 512, 649. VACCINATION AGAINST INFLUENZA 
811 
Adamson1 have proposed a method employing dilute solutions of hypochlo- 
rous acid. Vincent2 has proposed the use of ether for sterilizing young 
cultures. Olitsky3 has advocated a lipovaccine in which the bacilli are 
suspended in a bland oily medium after the method of Le Moignic and 
Pinoy. This vaccine has been found to produce local reactions, but not 
of a severe character; antibody production also occurs as the antigen is 
slowly absorbed. 
With a polyvalent vaccine of different types of dysentery bacilli Vin- 
cent4 has more recently reported encouraging results. In one series the 
morbidity was 1.6 per cent, as compared with 22.8 per cent, among the un- 
vaccinated. Another series showed among the non-vaccinated 70.57 cases 
per 1000 with 1.56 deaths; among the vaccinated 8.14 cases per 1000 with 
no deaths. 
VACCINATION AGAINST INFLUENZA 
Until the great pandemic of influenza which began in 1915 the medical 
profession generally accepted the bacillus described by Pfeiffer in 1892 as 
the causative agent of this infection; while vaccines prepared of micro- 
organisms recovered from the upper air passages had been advocated by 
a few physicians for the treatment and prevention of common “colds” and 
even sporadic cases of influenza, the profession generally had no experience 
with vaccines in the attempted prophylaxis of the disease in epidemic from. 
A large amount of investigation has been devoted to studies in etiology 
and vaccine prophylaxis with remarkable differences and shifts of opinion 
and consequent confusion concerning both subjects; the technical difficulties 
surrounding the isolation and identification of Bacillus influenza and the lack 
of scientific controls and criteria for evaluating the prophylactic value of 
vaccines, have been factors partly responsible for this confusion and incon- 
clusiveness. Furthermore, the high incidence and fatal nature of the 
respiratory complications characterizing the successive waves of the disease 
that swept the world since 1915 have very much confused the issue as to the 
primary etiology of the disease itself. However, out of the large amount and 
skilful bacteriologic work of the last few years it would appear possible and 
safe to deduce a few conclusions regarding the etiology of influenza as the 
subject stands today; while the same cannot be said of the subject of vaccine 
prophylaxis, it would appear that opinions are also beginning to crystallize 
on this subject. 
Immunity in Influenza.—Influenza occurs spontaneously only in man. 
Young rabbits and guinea-pigs are susceptible to the effects of artificial 
infection with virulent bacilli (Bacillus influenza, Pfeiffer) and injections of 
the toxin, but the disease apparently does not occur among the domestic ani- 
mals even during epidemics. Influenza of horses is regarded as caused by a 
streptococcus. 
Apparently the great majority of human beings are susceptible and 
especially during the years of later childhood and early adult age. Nursing 
infants apparently possess some immunity, but it is more probable that 
they are less exposed to infection. However, during the last great epidemic 
individual cases of natural resistance were not lacking. Doubtless factors 
predisposing to infection, as infections of the upper respiratory tract, exert 
an important role. 
Whether or not an immunity follows an attack is uncertain. Evidence 
1 Brit. Med. Jour., 1915, 1, 609. 
2 Jour. State Med., London. 1921, 29, 54. 
3 Jour. Exper. Med , 1918, 28, 69. 
Compt. rend. Soc. d. biol., 1921, 85, 965. 812 
PROPHYLACTIC ACTIVE IMMUNIZATION OR VACCINATION 
indicates in my opinion that an immunity of short duration (not over a 
year) develops in most individuals. On the other hand, some persons are 
apparently more susceptible, that is, may develop an attack of “grippe’7 
(apparently a light attack of influenza) every year or so, and peculiarly, 
about the same time each year. 
The Etiology of Influenza in Relation to Vaccination.—In the first place 
a clear distinction should be made between the primary cause of influenza 
and the secondary complications, notably the pneumonias. The etiology 
of the latter may be regarded as worked out satisfactorily and caused by 
one or several of a group of micro-organisms, including Bacillus influenza 
different types of pneumococci and streptococci, Micrococcus catarrhalis, 
staphylococci, and Bacillus Friedlander. Blake and Cecil1 have recently 
produced pneumonias in monkeys with B. influenza similar to those found 
among persons, and doubtless this bacillus is a frequent cause of pneumonia 
either alone or in conjunction with one or more of the other micro-organisms 
mentioned above. Parker2 has found that influenza bacilli may produce a 
toxin, this observation being confirmed by Huntoon and Hannum,3 Albert and 
Kelman,4 Ferry and Houghton,5 and believed capable of exerting a pathogenic 
role and especially in the production of the pulmonary lesions. As may be 
readily’ expected, the bacteriologic findings in the lesions of the lungs and 
upper respiratory passages have varied geographically, one micro-organism 
or a group predominating in one locality and another or group of micro- 
organisms in a different locality. 
These findings, however, by no means prove that any one of these bacteria 
are the primary cause of influenza. As previously stated, up to 1915 B. 
influenza was commonly accepted as the cause, although Pfeiffer had not 
found them in all cases studied in the pandemic of 1892 and the etiologic 
role of the bacillus had not then and is not now definitely established as the 
primary cause of influenza. During 1916 and 1917 the bacillus was generally 
found in such a small percentage of cases and pneumococci, streptococci 
et al in such a high percentage, that the role of this bacillus as the primary 
cause of influenza was generally discredited, although in 1919 and within 
recent months, opinion is again in its favor and bacteriologists have become 
divided into two main camps, namely, those who accept, or are inclined to 
accept, B. influenza as the primary cause, as Wolbach,6 Opie, Freeman, 
Blake, Small and Rivers,7 Medalia,8 Duval and Harris,9 Park,10 Pritchett and 
Stillman,11 Roos,12 Huntoon and Hannum,13 Small and Stangl,14 and others, 
and those who do not, the latter group regarding the germ as absolutely 
undiscovered or in the nature of a filterable virus as indicated by the re- 
searches of Nicolle, and Lebailly,15,16 Gibson and Connor,17 Bradford, Bash- 
1 Jour. Amer. Med. Assoc., 1920, 74, 170. 
2 Jour. Immunology, 1919, 4, 331. 
3 Jour. Immunology, 1919, 4. 
4 Jour. Infect. Dis., 1919, 25, 433. 
8 Jour. Immunology, 1919, 4. 
6 Bull. Johns Hopkins Hosp., 1919, 30, 104. 
7 Jour. Amer. Med. Assoc., 1919, 72, 556. 
8 Boston Med. and Surg. Jour., 1919, clxxx, 323. 
9 Jour. Infect. Dis., 1919, 25. 
10 Jour. Amer. Med. Assoc., 1919, 73, 318. 
11 Jour. Exper. Med., 1919, 29, 259. 
12 Jour. Immunology, 1919, 4, 331. 
13 Jour. Immunology, 1919, 4. 
14 Jour. Amer. Med. Assoc., 1920, 74, 1004. 
15 Compt. rend. Acad. d. sc., 1918, clxvii, 607. 
16 Ann. de l’lnst. Pasteur, 1919, 33, 395. 
17 Brit. Med Tour., 1918, 2, 645. VACCINATION AGAINST INFLUENZA 
813 
ford and Wilson,1 Yamanouchi et al,2 Olitsky and Gates.3 Many investi- 
gators, as Wahl, White and Lyall,4 Rosenau5 and others have completely 
failed to produce anything resembling influenza either in man or the lower 
animals by inoculating pure cultures of B. influenza and filtrates of influenza 
bacilli and respiratory secretions. 
The serum studies of Wilson,6 Rappaport,7 Kolmer, Trist and Yagle,8 
Gay and Harris,9 Duval and Harris10 and Fry11 have shown that agglutinins 
and complement-fixation antibodies may be found in the sera of persons 
suffering with influenza, streptococci, pneumococci, and other micro-organ- 
isms, but these studies prove nothing in regard to the primary etiologic 
agent and simply indicate that all of these micro-organisms, and notably 
B. influenza, are at least factors of secondary importance capable of exerting 
pathogenic changes and eliciting the production of specific antibodies. 
At the present time the filtrable organism described by Olitsky and 
Gates12 called Bacterium pneumosintes, is deserving of consideration as the 
primary etiologic agent of influenza. These investigators have described 
changes in the blood and lungs of rabbits and guinea-pigs which follow the 
intratracheal injection of unfiltered and filtered nasopharyngeal secretions 
obtained within thirty-six hours after onset, from patients ill with uncom- 
plicated epidemic influenza. Anaerobic cultures of these filtrates have 
developed a minute bacilloid body, not of the nature of ordinary bacteria 
and capable of indefinite propagation on artificial media. While the sera 
of normal persons do not contain agglutinis or precipitins for this organism 
these antibodies have been found by Olitsky and Gates13 in the serum of 
17 persons among 19 who were examined from 10 days to five months 
after recovery from epidemic influenza. 
Briefly, therefore, the subject of the etiology of influenza may be sum- 
marized as follows: 
1. Bacillus influenza has the strongest claim at present for recognition 
as the primary cause of influenza; if it is definitely proved that this bacillus 
is not the true and primary cause, but that the primary agent is a filtra- 
ble virus, as Bacterium pneumosintes, B. influenza will easily fall into the posi- 
tion of first importance as a micro-organism of secondary importance. 
2. The complications of influenza and notably the pneumonias may be 
accepted as caused by B. influenza, pneumococci, streptococci, possibly 
Bacterium pneumosintes, and other micro-organisms either alone or in com- 
bination, the bacteriology varying in different localities. These complica- 
tions, however, are to be separately considered from the etiologic stand- 
point from the primary cause of the disease. 
Kinds of Vaccines.—With regard to the prophylactic value of influenzal 
vaccine, it is extremely difficult to express an opinion, as is so commonly 
true of vaccine therapy in general. 
Two different kinds of vaccines have been employed, namely, vaccines 
1 Quart. Jour. Med., 1919, 12, 259. 
2 Lancet, 1919, June 7, 971. 
3 Jour. Amer. Med. Assoc., 1920, 74, 1497; ibid., 1921, 76, 640. 
4 Jour. Infect. Dis., 1919, 25, 419. 
6 Jour. Amer. Med. Assoc., 1919, 73, 861. 
6 Lancet, 1919, No. 5014, 607. 
7 Jour. Amer. Med. Assoc., 1919, 72, 633. 
8 Jour. Infect. Dis., 1919, 24, 583. 
9 Jour. Infect. Dis., 1919, 25, 414. 
10 Jour. Infect. Dis., 1919, 25, 384. 
11 Lancet, February 14, 1920, 368. 
12 Jour. Exper. Med., 1921, 33, 125, 161, 373. 
13 Ibid., 1923, 37, 303. 814 
PROPHYLACTIC ACTIVE IMMUNIZATION OR VACCINATION 
composed of influenza bacilli alone and mixed vaccines of influenza bacilli, 
pneumococci, streptococci, etc. The latter type has been used more 
extensively than the former. The mixed vaccines were prepared somewhat 
as follows: 
Bacillus influenzas 
Million per cubic 
centimeter. 
1000 
Streptococcus hemolyticus 
1000 
Streptococcus viridans 
1000 
Pneumococcus, Types I, II, III, and IV, of each 
1000 
Staphylococcus aureus 
500 
The first dose was 0.5 c.c.; the second and third doses were 1 c.c. The 
injections were given by subcutaneous injection at intervals of five to seven 
days. As a general rule the local and general reactions were of a mild 
character. 
Value of Vaccination.—The value of vaccines has been generally discussed 
under three headings, namely, their prophylactic value against influenza 
itself; second, their prophylactic value against the pulmonary conplica- 
tions. and third, their curative value in the treatment of influenza and its 
complications. 
Reports from different parts of the world by Leary,1 Rosenau,2 Paschall,3 
Wynn,4,5 Minaker and Irvine,6 Champtaloup and Drennan,7 Cadham,8 
Bezancon and Legrouse,9 Tottenham,10 Duval and Harris,11 and others indicate 
that the mixed vaccines have some value in reducing the incidence of the 
disease and especially the incidence and severity of the complicating pneu- 
monias. Two commissions,12 composed of Rosenau, Gay and McCoy and 
Whipple, Davis and Crumm, reported early in 1918, that the available 
statistical evidence indicated that influenza vaccines may have some prophy- 
lactic value, no specific value in treatment, and were probably harmless, but 
that the whole subject of vaccine therapy required careful scientific investiga- 
tion for determining its exact status. 
Recently Sir William Leishman13 has published the following statistical 
analysis of the use of two mixed vaccines in the British Army: 
43,520 non-inoculated soldiers: 
Incidence per 1000 47.3 percent. 
Pulmonary lesions 13.3 “ 
Deaths 2.25 “ 
15,624 inoculated: 
Incidence per 1000 14.1 percent. 
Pulmonary lesions ‘ 1.6 “ 
Deaths 0.12 “ 
McCoy, Murray and Teeter,14 however, have found no evidence to indicate 
the prophylactic value of influenzal vaccine and failed to demonstrate 
1 Amer. Jour. Public Health, 1918, 8, 754. 
2 Jour. Amer. Med. Assoc., 1918, 71, 1602. 
3 Jour. Amer. Med. Assoc., 1918, 71, 1602. 
4 Lancet, 1918, 26, 874. 
5 Brit. Med. Jour., 1920, February 21, 254. 
6 Jour. Amer. Med. Assoc., 1919, 72, 847. 
7 New Zealand Med. Jour., 1919, 1, 63. 
8 Lancet, 1919, 2, 885. 
9 Bull, l’acad. d. med., Paris, 1919, 81, 44. 
10 Brit. Med. Jour., 1919, 1, 41. 
11 Jour. Infect. Dis., 1919, 25, 384. 
12 Jour. Amer. Med. Assoc., 1918, 71, 1317. 
13 Lancet, February 14, 1920, 366. 
14 Jour. Amer. Med. Assoc., 1918, 71, 1907. VACCINATION AGAINST THE COMMON COLD 
815 
prophylactic value in properly conducted experiments. Hinton and Kane1 
and others have also reported negative results. Von Sholly and Park2 
observed that among 1327 inoculated persons and 3025 controls, vaccination 
with mixed vaccines had very little specific influence either in preventing 
or causing respiratory diseases other than pneumonia. The difference in 
the pneumonia incidence, however, was marked, the vaccinated showing an 
incidence of 0.075 per cent, as against 0.36 per cent, among the unvaccinated. 
Jordan and Sharp,3 in a study of 2873 persons vaccinated with a mixed vaccine, 
and 3193 unvaccinated controls, reached the conclusion that any consider- 
able degree of protection against influenza by vaccination is doubtful, but 
that there was some evidence of prophylaxis against pneumonia. 
It would appear, therefore, that the prophylactic value of influenza 
vaccine is doubtful; however, a general survey of the investigations conducted 
with mixed vaccines rather than vaccines of Bacillus influenzce alone does, I 
believe, permit the following conclusions at the present time: 
1. Large doses of mixed vaccines prepared of freshly isolated strains of 
B. influenzce, pneumococci and streptococci confer a slight degree of protec- 
tion against influenza. It is probable that this immunity is of short dura- 
tion. 
2. These vaccines have decided value in reducing the frequency and 
severity of the pulmonary complications. While the correctness of the first 
conclusion is doubtful both from a clinical point of view and especially in 
the uncertain state of our knowledge concerning the primary cause of influ- 
enza, the second may be regarded as well established and a justification 
for the use and further trial of these vaccines, and preferably under well- 
controlled conditions, for the purpose of possibly ameliorating the severity 
and the incidence of epidemic influenza and for securing data of scientific 
value. 
VACCINATION AGAINST THE COMMON COLD 
The “common cold” by reason of its tremendous morbidity is of greater 
importance than commonly considered even though the infection is not of 
itself of a fatal character. Few people escape at least one “cold” during 
the winter months and a large number suffer from several attacks each 
season. Furthermore, the “common cold” may be the starting-point for 
middle-ear infections, mastoiditis, infections of the accessory nasal sinuses, 
bronchitis, and pneumonia. It is quite possible that an acute rhinitis may 
be related to the development of carrier states for meningococci and pneu- 
mococci; considered in the light of these possibilities the “common cold” is 
worthy of very serious investigation, and especially in the direction of its 
etiology and prophylaxis. 
The “common cold,” acute coryza or acute rhinitis, is probably a definite 
entity and not to be confused with true influenza. The so-called “grippe” 
is intermediate in severity between the “cold” and influenza; the writer 
believes that the “cold” and “grippe” will be ultimately shown to have the 
same etiology, the latter being a severer infection, while influenza is caused 
by a different primary agent. 
The Etiology and Immunity.—The etiology of the “common cold” is 
still unknown. No doubt sudden changes in temperature may induce 
vascular engorgement of the nasal mucosa with increased glandular secretion 
which may resemble a “cold” of mild character; it is possible that these 
changes may favor increased activity of the bacterial flora normally present. 
1 Tenn. State Med. Assoc. Jour., 1919, 11, 442. 
2 Jour. Immunology, 1921, 6, 103. 3 Jour. Infect. Dis., 1921, 28, 357. 816 
PROPHYLACTIC ACTIVE IMMUNIZATION OR VACCINATION 
This is one view of the etiology that is widely accepted and bacteriologic 
studies have shown that streptococci, pneumococci, staphylococci and M. 
catarrhalis are commonly found. Floyd1 has also found that some patients 
exhibit cutaneous sensitiveness to these organisms which indicates their 
close relationship to the disease and opens up the possibility of the “cold” 
being an expression of clinical allergy to the proteins of these organisms. 
Krause, Foster, and more recently Bloomfield have expressed the opinion 
that the primary etiologic agent is in the nature of a filterable virus and 
unknown. Certainly no one has been able to demonstrate in cultures the 
presence of an organism that may not be found under normal conditions, 
and for this reason the streptococci, pneumococci, etc., are regarded as 
having no relation at all to the disease or but of secondary importance. The 
writer believes that if it is ultimately proved that an unknown virus is the 
cause that some of these bacteria will easily fall into a position of important 
secondary factors; certainly they have been closely identified with the 
complicating infections, as sinusitis, otitis media, etc. 
Little is known of immunity to colds. Some individuals are extremely 
susceptible; others possess a high degree of resistance. Doubtless, purely 
local factors are operative in some cases. Whether a transient immunity 
is conferred by an attack is not known—certainly such resistance is usually 
very temporary in view of the numerous attacks that may occur during one 
winter season. 
Vaccination.—The question has arisen regarding the possibility and 
advisability of attempting vaccination against the usual bacterial flora of 
streptococci, pneumococci, staphylococci, etc., as a means of prophylaxis. 
A large number of reports indicate that vaccination with vaccines of these 
bacteria confers a temporary immunity; other reports have been negative. 
I have had no experience with stock vaccines, but with autogenous vaccines 
I believe I have seen unmistakable evidences of successful prophylactic 
immunization. My practice is to make cultures on or about the third to 
fifth days of the disease during an acute attack and isolate the streptococci, 
pneumococci, staphylococci and Micrococcus catarrhalis that may be present. 
The bacteria are incorporated into a vaccine in equal proportions so that the 
final product has a total of 1,000,000,000 bacteria per cubic centimeter. Im- 
mediately after the attack has subsided, the vaccine is administered by sub- 
cutaneous injection at intervals of five to seven days in doses of 0.1, 0.2, 0.4 
0.6, 0.8, and 1 cubic centimer. In my experience this treatment has protected 
many individuals for at least one season; in some instances for two years. 
The vaccine apparently keeps well in a refrigerator for one year, and it is 
advisable to give three or four injections during the succeeding October and 
November in an attempt to confer some degree of protection for the following 
winter. 
VACCINATION AGAINST PERTUSSIS 
The hemoglobinophilic bacillus described by Bordet and Gengou in 1906 
is now commonly accepted as the cause of pertussis or whooping-cough. 
These investigators2 prepared antiendotoxic sera and vaccines and reported 
that the latter were harmless and provoked no unpleasant symptoms. 
Immunity in Pertussis.—Whooping-cough or pertussis occurs spontane- 
ously only in human beings. The lower animals escape infection even when 
intimately exposed; apes have been experimentally infected. 
Children between the first and second dentitions are especially suscep- 
1 Bost. Med. and Surg. Jour., April 15, 1920. 
2 Ann. de l’lnst. Pasteur, 1906, 20, 731. VACCINATION AGAINST CEREBROSPINAL MENINGITIS 
817 
tible; nurslings under a year of age may, however, develop the disease. 
Adults escaping the disease during childhood may become infected, and 
among these whooping-cough is frequently a serious disease. Negroes are 
especially susceptible. 
Apparently some persons are naturally immune and never contract the 
disease, even though exposed. It is highly probable, however, that examples 
of absolute immunity are rare and that some individuals apparently immune 
have had mild and unrecognized attacks. 
One attack of pertussis usually confers an immunity for life. During 
the disease agglutinins, bacteriolysins, and complement-fixing antibodies 
develop for the bacillus, but quickly disappear after recovery. Phagocytosis 
of bacilli in the mucosa of the respiratory tract also occurs and contributes 
to resistance and recovery. 
Vaccination.—In 1914 Hess1 employed pertussis vaccination and be- 
lieved that it had protective value in a certain percentage of cases; 26 per 
cent, of a group of 80 unvaccinated and exposed children did not contract 
the disease and were probably naturally immune, while 92 per cent, of 244 
vaccinated children failed to contract the disease. Luttinger2 gave prophy- 
lactic injections to 91 children constantly exposed to pertussis for two weeks 
or less; of these, 94 per cent, escaped the disease. He also vaccinated 277 
children who showed early symptoms of the disease; of these, the disease was 
apparently aborted in about 53 per cent. These results tend to indicate that 
when the disease is actually present, active immunization is much less likely 
to prevent progression of the infection than when given before any symptoms 
have appeared. Luttinger also analyzed the reports of vaccination of 239 
children by private physicians, the results indicating that 90 per cent, were 
apparently protected. Huenekens3 observed the production of complement- 
fixing antibodies after the administration of vaccine and especially when 
fresh vaccines were employed. He found prophylactic vaccination effective, 
but emphasized the necessity for using vaccines not over one week old and 
in large doses. Bogert4 and Luzzati5 have also reported favorably upon the 
prophylactic value of pertussis vaccines; likewise Davies6 on both its prophyl- 
actic and curative value, but Davison7 summarizes the literature with the 
statement that the evidence in favor of their prophylactic value is slight. 
Since the disease is one of the most dreaded of childhood, it would appear 
that immunization with freshly prepared vaccines is worth while and espe- 
cially in institutions and families. At least three subcutaneous injections 
should be given at intervals of five to seven days and as soon as possible 
after exposure. For children the doses may be 500,000,000, 1,000,000,000 
and 2,000,000,000 bacilli respectively; for adults, 1,000,000,000, 2,000,000,000, 
and 3,000,000,000. The duration of the immunity after vaccination is 
unknown, but probably is not over two or three years’ duration. 
VACCINATION AGAINST CEREBROSPINAL MENINGITIS 
Immunity in Meningococcus Meningitis.—This disease occurs only in 
man, although there is considerable evidence to indicate that horses may 
suffer from a meningitis produced by a diplococcus closely resembling the 
1 Jour. Amer. Med. Assoc., 1914, 63, 1007. 
2 Amer. Jour. Med. Assoc., 1917, 68, 1461. 
3 Amer. Jour. Dis. Child., 1918, 16, 30. 
4 Amer. Jour. Dis. Child., 1918, 15, 271. 
5 Policlinico, 1920, 27, 451. 
6 Amer. Jour. Dis. Child., 1922, 23, 423. 
7 Jour. Amer. Med. Assoc., 1921, 76, 242. 818 
PROPHYLACTIC ACTIVE IMMUNIZATION OR VACCINATION 
meningococcus. As far as I know, however, no one has shown that this 
disease of horses is transmissible to man or the reverse. 
Natural immunity of human beings to this infection probably exists 
to some degree. At least many individuals escape even when intimately 
exposed, although this may be due to other factors controlling the infection. 
Children under five years are most susceptible; adults between thirty-five 
and forty years are least susceptible, although the disease may occur still 
later in life. 
Sex has no influence; neither has race, although negroes have suffered 
more than whites in some epidemics. This is probably due to hygienic 
conditions. Catarrhal infections of the upper respiratory tract predispose 
to infection, doubtless reducing the phagocytic activity of leukocytes and 
aiding the lymphatic and vascular absorption of meningococci. 
Second attacks are rare, but do occur. During the disease agglutinins, 
bactericidans, opsonins, and complement-fixing antibodies may be found in 
the blood, but rapidly disappear after recovery. Recovery is apparently 
brought about largely by the activity of immune opsonins and phagocytosis, 
aided by agglutinins. 
Vaccination.—Sophian and Black1 have shown experimentally that a poly- 
valent meningococcic vaccine, heated to 50° C., standardized in the usual man- 
ner, and given in three injections in doses of 100,000,000, 500,000,000, and 
1,000,000,000, at intervals of a week, appears to afford a high degree of protec- 
tion. In the blood-serums of inoculated persons these observers were able to 
demonstrate opsonins, agglutinins, and complement-fixing amboceptors. All 
evidence points to the efficacy of prophylactic vaccination, as only a mod- 
erate degree of immunity may give complete protection against the disease. 
Black,2 in a subsequent investigation, found the production of antibodies 
after vaccination and expressed the opinion that the immunity probably 
lasts for two years. Quarelli3 reports favorably upon this vaccination, and 
Whitmore, Fennel and Petersen4 vaccinated 55 persons with a lipovaccine 
and observed the production of agglutinins with moderate general reactions 
after large doses. Gates5 administered polyvalent saline vaccine to 3700 
volunteers in three subcutaneous injections at weekly intervals of 2,000,000,- 
000, 4,000,000,000, and 4,000,000,000 or 8,000,000,000 cocci. These doses 
rarely caused more than the mildest local and general reactions, although 
exceptionally a more severe reaction with symptoms of meningeal irritation 
were observed. Specific meningococcus agglutinins were demonstrable 
in the blood-serum of the vaccinated individuals. 
It would appear, therefore, that immunization with polyvalent saline 
vaccines of meningococci in three doses by subcutaneous injection at weekly 
intervals may prove of value as aiding in the prophylaxis of meningococcus 
meningitis and especially wrhen the disease occurs in epidemic form. 
VACCINATION AGAINST SCARLET FEVER 
Immunity in Scarlet Fever.—Scarlet fever occurs only in human beings. 
Landsteiner, Levaditi and Prasek6 claim to have produced the disease 
experimentally in the chimpanzee, but not in monkeys; Draper and Hand- 
ford7 also failed to infect monkeys. 
Age is a most important factor in susceptibility. According to Welch 
1 Jour. Amer. Med. Assoc., 1912, lix, 527; Black, ibid., 1913, lx, 1289. 
2 Ibid., 1914, 63, 2126. 6 Jour. Exper. Med., 1918, 28, 449. 
3 Policlinico, 1917, 24, 501. 6 Ann. de l’Inst. Pasteur, 1911, 25, 754. 
4 Jour. Amer. Med. Assoc., 1918, 70, 427. 7 Jour. Exper. Med., 1913, 17, 517. VACCINATION AGAINST SCARLET FEVER 
819 
and Schamberg1 children from two to five years of age are most susceptible. 
Considerably over half of the deaths from scarlatina occur in children under 
five years, almost 90 per cent, under ten and over 95 per cent, under fifteen 
years of age. Infants under one year are seldom infected and under three 
months of age the disease is very rare, even though the child suckles a mother 
with scarlet fever. It is very doubtful if children are ever born with the 
disease. 
Negroes are less susceptible than whites and the mortality among them 
is much lower. Osier states that the natives of India enjoy a high degree 
of natural immunity. 
Undoubtedly some persons possess natural immunity, but it is not 
always absolute, as the disease may develop after repeated exposures. The 
individuals of some families are sometimes peculiarly susceptible, four and 
five members taking the disease in rapid succession. 
The unknown virus very probably gains access through the respiratory 
tract. One attack usually confers an immunity for life, although second 
and even third attacks may occur. 
Nothing is known of the mechanism of immunity, although immunity 
principles are evidently present in the blood and the serum of convalescents 
is sometimes useful in the treatment of the disease. Serum of convalescents 
has not been used for passive immunization, but it would appear that this 
may be successful as indicated by the apparent success of this method in the 
prevention of measles. 
Vaccination Against Streptococcus Infection in Scarlet Fever and Puer- 
peral Sepsis.—Several Russian physicians, particularly Gabrickevski,2 Lon- 
govi, Nitikin, Shamarin, and others, have secured good results from a method 
of prophylactic vaccination against scarlet fever with a polyvalent vaccine 
of scarlet fever streptococci. Heat-killed vaccines were given in three 
successive doses. In this country the method has been tried by Watters3 
and Kolmer,4 who found that while inoculations with a heat-killed strepto- 
coccic vaccine cannot prevent scarlet fever itself, such inoculations may, 
however, prevent a severe attack of the disease by producing some immunity 
against secondary streptococcic infections. 
Weaver and Boughton,5 Weaver and Tunnicliff,6 Meakens,7 Boughton,8 
and others have observed the production of specific opsonins and other 
antibodies after the administration of streptococcus vaccines, and it would 
appear that active immunization may be of some value in the prophylaxis 
of streptococcus infections. 
Champtaloup9 has advised the active immunization of expectant mothers 
against streptococcus puerperal sepsis, when the infection occurs in epi- 
demic form in institutions, or in private practice in which the house sur- 
roundings are not of the best. Three doses are given at intervals of two 
days of 100,000,000, 250,000,000, and 500,000.000, and sensitized vaccines 
are preferred. 
1 Acute Contagious Diseases, Lea Brothers & Company, Philadelphia and New York, 
345. (Unfortunately this valuable book is now out of print.) 
2 Russk. Vratch, St. Petersburg, 1906, x, 469; see review of literature by Smith in Boston 
Med. and Surg. Jour., 1910, clxii, 242. 
3 Jour. Amer. Med. Assoc., 1912, 58, 546. 
4 Penna. Med. Jour., February, 1912. 
5 Jour. Infect. Dis., 1908, 5, 608. 
6 Jour. Infect. Dis., 1908, 5, 589. 
7 Jour. Exper. Med., 1909, 11, 815. 
8 Jour. Infect. Dis., 1910, 7, 99. 
9 Brit. Med. Jour., 1914, 1, 1221. 820 
PROPHYLACTIC ACTIVE IMMUNIZATION OR VACCINATION 
VACCINATION AGAINST ACUTE ANTERIOR POLIOMYELITIS 
Flexner and Lewis,1 in 1910, showed that monkeys could be immunized 
against poliomyelitis by repeated subcutaneous injections of increasing 
amounts of crude unmodified virus over a period of two and one-half months. 
These animals, after a further ten days, were injected intracerebrally with 
2 c.c. of a filtrate of a highly potent virus, of which from 0.05 to 0.1 c.c. 
would prove fatal. The immunized animal therefore resisted from twenty 
to forty fatal doses. In a later report they state that artificial active 
immunity either by the injection of a single large dose or by a series of 
increasing small doses of crude virus over a period of time is not uniformly 
successful. In the former method some of the animals developed poliomye- 
litis as a result of the subcutaneous injection, and in both some of the 
animals so injected did not resist the test intracerebral inoculations of the 
rather large dose of a highly potent virus. 
Levaditi and Landsteiner2 dried cords from monkey poliomyelitis after 
the method of Pasteur for rabies, but they found that some of the animals 
developed poliomyelitis as the result of the preventive inoculations. 
Abramson3 employed subcutaneous injections of heat-modified followed 
by unheated virus for the vaccination of monkeys and found neutralizing 
principles in the blood of these animals for active virus; five of eight vacci- 
nated animals withstood the intracerebral injection of three to six lethal 
doses of virus. 
These results encourage further investigation but at the present time 
the method has not been applied to human beings. 
VACCINATION AGAINST CHICKENPOX 
Immunity in Chickenpox.—Chickenpox, or varicella, occurs only in man; 
the lower animals enjoy an absolute natural immunity even when intimately 
exposed to infection. 
The disease is most common between one and seven years, but may 
develop in nursing infants. Adults commonly escape, but those who have 
not had the disease in childhood are susceptible and may develop the disease. 
Both sexes and all races are equally susceptible. Human beings do not 
appear to have an appreciable degree of natural immunity, although the 
disease does not attack as many persons as are infected with measles. 
One attack usually confers a lasting immunity; second attacks have been 
recorded, but are exceedingly rare. 
Vaccination.—Kling4 has reported the successful inoculation of chicken- 
pox followed by protection. The virus should be the clear fluid of vesicles 
in as early stage as possible both of the disease as well .as of the vesicles 
themselves, and taken, of course, only from children otherwise healthy and 
free of syphilis and tuberculosis. Kling cleanses a vesicle, inserts the point 
of a sterile lancet, and makes three to four pricks of the cleansed skin of the 
arm of the child being vaccinated; three or four more insertions are made of 
virus from a second vesicle. The mild traumatic reaction subsides in a day 
or two, followed on the eighth day as a rule by the development of one or 
more red papules which quickly change to vesicles, followed by scabbing 
and eventually an insignificant scar. In a small percentage of cases a 
generalized eruption and slight fever may occur. Of 135 vaccinations, 
1 Jour. Amer. Med. Assoc., 1910, 54, 1780; ibid., 1910, 55, 662. 
2 Compt. rend. Soc. de biol., 1910, 68, 311. 
3 Jour. Amer. Med. Assoc.., 1918, 70, 1142. 
4 Hygiea, 1913, lxxv, 1032; Berl. klin. Wchn., 1913, 1, 2083; ibid., 1915, liii, 13. VACCINATION AGAINST MEASLES 
821 
failures occurred in 23 per cent. These experiments indicate that it is 
sometimes possible to inoculate and transmit chickenpox by this means and 
that the resulting infection is usually local. 
According to Kling successful inoculation confers immunity, while 73 
per cent, of unvaccinated children contract the disease. Handrick,1 however, 
was unable to confirm these results during two epidemics with the vaccina- 
tion of 127 children. In this country Rabinoff2 inoculated 142 susceptible 
children using Kling’s method except that scarifications were employed; 
114 or about 75 per cent, developed varicella. Of these 6 or 8 per cent, 
subsequently developed the disease, all being vaccinated within ten days 
previously and within the incubation period. Rabinoff is convinced that 
the vaccinations limited the spread of the disease. Michael3 has reported 
8 successful inoculations of 32 children by this method. 
Hess and Unger4 have vaccinated by drawing the fluid of vesicles (not 
pustules) into sterile capillary tubes to the distance of about an inch, diluting 
with sterile saline solution, and injecting intravenously. In all, 38 children, 
three or four years of age, were inoculated by this method; none developed 
any local or general symptoms or any eruption suggestive of varicella. All 
were exposed to chickenpox; one developed the disease thirty-six days after 
vaccination. 
These investigations indicate that the virus of chickenpox is contained 
in the vesicle fluid; that direct cutaneous inoculation is possible and that 
active immunization may be effected. The disease is usually so mild, how- 
ever, that vaccination may not be required as a practical measure. 
VACCINATION AGAINST MEASLES 
Immunity in Measles.—Measles occurs only in man; the lower animals 
commonly escape infection, but monkeys have been infected artificially 
by Anderson and Goldberger.5 
The disease is most common between one and ten years of age. In rare 
instances children have been born with the disease, but infants under six 
months are rarely attacked and apparently enjoy for a period of six to ten 
months a congenital maternal immunity. Adults who have escaped the 
disease during childhood may contract it in later years. 
Both sexes and all races are susceptible. Measles is probably the com- 
monest of all the acute exanthemata. In communities first attacked by the 
disease, as the Faroe Islands and Iceland, measles has proved a virulent 
infection, suggesting a gradual immunization of the population where the 
disease is endemic. 
One attack usually protects for life, but affords no protection against 
rubella (German measles). 
Vaccination.—Children under five months of age are relatively immune 
to measles and a gradually diminishing degree of resistance persists for the 
first three or four years of childhood. Owing, however, to the mortality 
of this disease and particularly when complicated by pneumonia and other 
secondary infections, attempts have been made to develop a process of 
vaccination. Hermann6 has vaccinated 40 infants five months or less of 
age by inoculating the nose with virus obtained from the nasal secretions of 
children with measles twenty-four hours before the eruption and believes 
that protection was afforded. Vaccination in measles, however, is in the 
1 Monatschr. f. Kinderh., 1914, 13, 205. 
2 Archiv. Pediat., 1915, 32, 651. 
3 Boston Med. and Surg. Jour., 1917, 702. 
4 Amer. Jour. Dis. Child., 1918, 16, 34. 
5 Public Health Reports, 1911, No. 26. 
6 Archiv. Pediat., 1915, 32, 503. 822 
PROPHYLACTIC ACTIVE IMMUNIZATION OR VACCINATION 
experimental stage. Better results have been reported following the injec- 
tion of convalescent serum for prophylactic purposes (see Chapter XL). 
Immunity in Yellow Fever.—Up to the present time this disease has 
been observed only in human beings, the lower animals escaping even when 
intimately exposed during epidemics. Noguchi has recently claimed the 
successful inoculation of guinea-pigs by the injection of blood from individuals 
with the disease. 
The negro is less susceptible than the white. Both sexes are equally 
susceptible. The disease is most common between ten and thirty years, 
young children and the aged frequently escaping. The virus is transmitted 
by the bites of infected mosquitoes {Stegomyia jasciala) and eradication of 
these mosquitoes and other hygienic measures are rapidly ridding the world 
of this pestilence. 
One attack generally confers an immunity, but second attacks are not 
unknown. 
Vaccination.—According to the French Commission, a certain degree of 
immunity could be conferred by the injection of serum heated to 55° C. 
for five minutes from yellow fever patients; some success has also been 
claimed for the injection of defibrinated blood kept under anaerobic conditions 
for eight days. They also claimed that the serum of convalescents had 
prophylactic and curative properties. 
Noguchi and Pareja1 have reported that guinea-pigs inoculated with suf- 
ficient amounts of killed culture of Leptospira icteroides are usually rendered 
resistant to a subsequent infection with living cultures. Encouraging results 
have been observed in the vaccination of human beings, but many more 
observations will be required before a final decision can be reached regard- 
ing the value of vaccination against yellow fever with leptospira vaccines. 
VACCINATION AGAINST YELLOW FEVER 
VACCINATION AGAINST SYPHILIS 
Immunity in Syphilis.—A great deal of attention has been devoted to 
studies in immunity in syphilis since the discovery of the Spirochaeta pallida 
in 1905, and the successful transmission of the disease to apes and rabbits; 
the subject is one of importance owing to the wide-spread prevalence of the 
disease and especially in relation to diagnosis and treatment. 
Natural immunity to syphilis does not appear to exist among human 
beings. All races are susceptible to infection, although it would appear that 
the severity of the disease is lessened among those peoples where syphilis 
is wide-spread and known to have been present for centuries. Of course 
not all exposed individuals contract the disease. The greatest single factor 
of resistance is an intact epithelial barrier of the skin and mucous membranes. 
It would appear that infection requires a break in this barrier before Spiro- 
chasta pallida may penetrate to the deeper tissues and produce infection, 
but the abrasion may be so small as to escape detection and, indeed, may not 
be necessary at all, in view of the recent experiments of Reasoner and of 
Brown and Pearce, who report the successful infection of the genital organs 
and eyes of normal stock rabbits by the simple procedure of applying virulent 
Spirochaeta pallida to the mucous membranes. 
Fortunately, Spirochaeta pallida cannot long exist outside of the tissues 
and syphilis does not occur spontaneously among the lower animals. Apart 
from monkeys and rabbits, the lower animals resist infection, although 
1 Jour. Amer. Med. Assoc., 1921, 76, 34. VACCINATION AGAINST SYPHILIS 
823 
syphilitic keratitis is claimed to have been produced in cats, sheep, and dogs 
by inoculation with virulent spirochetes. My own attempts to produce 
syphilis in white rats, guinea-pigs, and dogs by testicular inoculations have 
always failed, even when attempts were made to break down the resistance 
of the tissues and especially of the lymphocytes by preliminary exposure of 
the tissues to #-rays. 
The nature of natural immunity in the lower animals is not known. The 
blood may be and usually is lacking in spirocheticidal properties and demon- 
strable antibodies; the leukocytes and other cells do not appear to possess 
the power of phagocytosis or but to a very limited degree. The spirochetes 
simply die off as if lacking a suitable pabulum analogous to the death of 
transplantable mouse cancer cells in a rat or other animal and explained 
by Ehrlich with his athreptic theory of immunity. 
Human beings have a defensive mechanism against syphilis sufficient 
for offering some resistance to Spirochseta pallida after infection has occurred, 
but these factors are not nearly so effective as in the monkey and rabbit and 
infinitely less than in other of the lower animals. As long as living spiro- 
chetes are present in the tissues these defensive forces may be and usually 
are apparently sufficient to protect against superinfection, and to some ex- 
tent prevent the extension of the disease, but they are not sufficient for the 
killing of the spirochetes with the production of spontaneous cure and do not 
leave a state of immunity to reinfection when sterilization is brought about 
by the administration of mercury and arsphenamin. Brown and Pearce1 
have likewise found that a rabbit once infected continues to harbor virulent 
spirochetes for the balance of life, and the defensive forces against syphilis 
are apparently better developed in this animal than in human beings, in 
view of the usual mildness of the disease and long periods of latency. 
During the primary incubation period of syphilis, that is, the time 
elapsing between exposure and infection and the development of the chancre, 
the human being, monkey, and rabbit may be reinfected; for a brief period 
after the development of the sore auto-infection is possible. But with the 
complete development of the chancre resistance to superinfection and auto- 
infection becomes operative and persists in the great majority of syphilitics 
until cure or complete destruction of spirochetes is brought about by drug 
treatment. Instances of superinfection during the secondary stage of syphilis 
are practically unknown; during the tertiary stages it has occurred, but is 
relatively rare. As Neisser2 states, the only human beings possessing resist- 
ance to syphilis are those who are syphilitic and harbor spirochetes in their 
tissues. He quotes Rollet as follows: “Although I and my predecessors have 
a thousand times attempted to reinoculate luetic subjects, we have never 
observed a successful case. I know no single fact more thoroughly proved 
than the insusceptibility of a syphilitic to the action of a new virus, and, 
moreover, these experiments are so harmless that they may be performed 
without scruple.” Finger and Landsteiner,3 however, believe that resistance 
during syphilis is less absolute than commonly believed and that especially 
during the tertiary stages superinfection is possible; according to Landsteiner4 
the new lesions will not take the form of a chancre, but appear in the form 
of the lesions being presented by the patient at the time of reinfection. John5 
also cites a series of cases carefully collected from the literature of apparent 
1 Jour. Amer. Med. Assoc., 1921, 77, 1619. 
2 Beitrage zur Pathologie und Therapie der Syphilis, Berlin, Julius Sprinzer, 1911; Arch, 
f. Dermat. u. Syph., 1906, 78, 335; ibid., 1906, 81, No. 1. 
3 Verhandl. d. deutsch. Dermat. Gesellsch., 1907, 25. 
4 Centralbl. f. Bakteriol., Ref., 1908, xli, 785. 
5 Samml. klin. Vortr., Volkmann, 1907-09, 559. 824 
PROPHYLACTIC ACTIVE IMMUNIZATION OR VACCINATION 
reinfection during syphilis, but it is possible and, indeed, probable that some 
of these were reinfections of cured rather than uncured luetics. John 
mentions that the severity of the reinfections was similar to the first attack, 
suggesting that the first infection did not engender sufficient antibodies to 
mitigate the severity of the reinfection. 
Very probably the escape of some individuals from syphilitic infection 
after intimate exposure with the evidence of confrontation, is due to the 
fact that they are congenital syphilitics. I have known of 2 such cases 
in young men seeking advice under these conditions, who escaped infection 
only to learn of being congenitally syphilitic and both of whom presented 
syphilitic keratitis. 
. The well-known resistance to syphilis of a child born of a syphilitic 
mother (Profeta’s law) and of an apparently healthy mother exposed to an 
actively syphilitic child (Colies’ law) are not now believed due to an actual 
immunity to the disease. The Wassermann reaction has revealed that the 
child in the first instance and the mother in the second are actually syphilitic 
in most instances and that their resistance is simply the resistance to re- 
infection caused by the presence of latent syphilis. It cannot be denied, 
however, that antibodies are produced in the course of syphilis, and these 
are probably largely responsible for the periods of latency with freedom of 
symptoms; it is possible that these antibodies generated in a syphilitic mother 
may be transferred by the placenta to the child so that the resistance of the 
latter is passive, but experimental evidence indicates that the degree of this 
immunity is probably feeble and temporary and that the only sure and lasting 
resistance to infection with Spirochaeta pallida is pre-existing syphilis. 
Immunity to syphilis apparently does not persist after complete cure 
as usually occurs in smallpox, measles, typhoid fever, etc. As previously 
stated there is no conclusive evidence to indicate that syphilis is self-curative 
or self-limited. It appears that antibodies and other factors of resistance 
may limit the development of lesions and maintain variable periods of 
freedom from symptoms, but actual and complete destruction of spirochetes 
probably does not occur without the administration of antiluetic drugs. 
In some races, and notably the negro, these defensive agencies may be par- 
ticularly well developed, as, likewise, in women infected by giving birth to 
syphilitic children, but the latter is probably concerned more with phases of 
the cycle of infection than processes of immunity. Since invasion of the 
brain and cord with Spirochaeta pallida occurs in probably the majority of 
cases of early syphilis, and the incidence of late symptomatic and asympto- 
matic neurosyphilis is limited, it is probable, as recently stated by Keidel,1 
that these immune reactions tend to protect the neuraxis. 
It appears that human individuals are much less able to develop immunity 
to protozoan than to bacterial infections, and Spirochaeta pallida is now 
classed by most biologists as a protozoon. In this regard the immunity 
developed in syphilis is similar in its feebleness to immunity in other proto- 
zoan infections, as trypanosomiasis and malaria. It would appear that 
in general terms the lower in the scale of life the infectious agent, the more 
complete our processes of immunity; immunity is usually more complete 
in bacterial than in mycotic infections and more complete in mycotic than 
in protozoan infections. For this reason the writer surmises that the 
infectious agents of smallpox and measles belong to a very low order by 
reason of the completeness of the immunity following these infections. 
Furthermore, the production of antibodies for Spirochaeta pallida by the 
injection of vaccines of the spirochetes is only slight, and up to the present 
1 Jour. Amer. Med. Assoc., 1922, 79, 874. VACCINATION AGAINST SYPHILIS 
825 
time no efficient means of prophylactic immunization has been evolved; 
this subject will be reviewed in more detail later in the discussion on vaccine 
therapy in syphilis. Likewise all attempts to confer passive immunity by 
injections of blood from luetic subjects during the latent periods when 
antibodies should be expected to occur in the blood in largest amounts, have 
usually failed; this subject will also receive more discussion shortly. 
Antibodies, however, do develop during syphilis of the human being 
and lower animals, but the amounts are slight and at best only capable of 
holding the disease in check and preventing reinfection without persisting 
after cure by drugs or bringing about spontaneous cure. Agglutinin pro- 
duction has already been discussed in Chapter XV; slight amounts of 
precipitin may also be produced as discussed in Chapter XVII. With 
cultures of pallida Broadwell and myself1 were unable to discover appre- 
ciable amounts of spirocheticidal substances in the sera of syphilitics, but 
Eberson2 has recently claimed that substances of this nature are to be found 
and especially in the latent and tertiary stages of human syphilis. Phago- 
cytosis does not appear to be an important phenomenon in syphilis, although 
the spirochete has been found by Levaditi and others within body cells; 
very little work has been done on the question of immune opsonins for 
pallida, but in my own experiments with the sera of luetics these were not 
found to be increased. 
Most attention has been given the Wassermann reaction in reference to 
antibody production in syphilis, but as discussed in Chapters XXIII and 
XXIV the substance in the blood and spinal fluid of the syphilitic respon- 
sible for the complement-fixation and flocculation reactions are not anti- 
bodies in the sense of being destructive for the spirochetes; they appear to be 
peculiar cellular products endowed with the property of bringing about 
flocculation of lipoids in colloidal states with the absorption or fixation of 
complement. Specific complement-fixing antibodies in syphilis for antigens 
of Spirochseta pallida are developed only to a slight degree and not at all 
to an extent comparable to the production of the cellular product responsible 
for the Wassermann reaction. 
Resistance or immunity in syphilis presents other peculiar features. 
For example, the luetic after the primary stage is resistant to reinfection, 
but at the same time is subject to exacerbations of his own disease. In 
other words, new spirochetes are apparently unable to gain a foothold in his 
tissues, but his own spirochetes may progressively invade other organs and 
tissues. If different strains of Spirochaeta pallida exist, as indicated by the 
work of Noguchi,3 Nichols,4 Levaditi and Marie5 and Marie, Levaditi, and 
Banu,6 infection with one strain appears to protect against other strains, 
although, if Landsteiner is correct, that the uncured syphilitic can be re- 
infected more frequently than surmised, it may be that just the reverse is 
true, namely, that the uncured syphilitic is resistant to spirochetes of his 
own strain, but susceptible to other strains. 
The nature of this resistance of the syphilitic to reinfection and the 
resistance offered to the disease and responsible for the varying periods of 
latency is not well understood. Unquestionably a local tissue immunity 
develops and antibodies or other resistant substances are produced locally 
in a syphilitic lesion capable of protecting other organs. For example, 
1 Jour. Immunology, 1916, 1, 429. 
2 Arch. Dermat. and Syph., 1921, 4, 490. 
3 Jour. Exper. Med., 1912, 15, 201. 
4 Jour. Exper. Med., 1914, 19, 362; Jour. Amer. Med. Assoc., 1916, 67, 1799. 
5 Ann. de l’Inst. Pasteur, 1919, 33, 741. 
6 Compt. rend. Acad. d. Sci., 1920, 170, 1021. 826 
PROPHYLACTIC ACTIVE IMMUNIZATION OR VACCINATION 
Nichols1 found that an active lesion in one testicle of a rabbit tends to 
inhibit the development of lesions in the other testicle or other parts of the 
body. Kraus and Volk2 has previously demonstrated a local immunity of 
the skin, and Zinsser, Hopkins, and McBurney3 observed that while the second 
normal testicle of a rabbit may be infected during or after the infection of 
the other testicle, that the latter could not be reinfected and constantly 
exhibits a well-marked degree of local resistance. 
Undoubtedly antibody production occurs in the tissues involved in the 
chancre responsible for the general resistance to reinfection and auto- 
infection exhibited in the primary stage of syphilis. 
Brown and Pearce4 have recently shown in the rabbit that any procedure, 
as castration or the administration of antisyphilitic drugs, tending to reduce 
or suppress the development of the local lesion tends to intensify the general 
infection. These investigators have shown5 that measures of this kind 
modifying the local primary reaction between spirochetes and tissue cells 
may modify the clinical course of syphilis in the rabbit, and that this may 
be more important in relation to the different clinical types of syphilis in 
the human being than the question of different strains of Spirochaeta pallida. 
Levaditi6 has shown many years ago that spirochetes may be found in the 
spleen of the monkey even before the primary lesion is well developed, and 
Brown and Pearce7 have shown that the blood and lymphatic systems of 
rabbits may be invaded before the initial lesion can be detected. These 
observations indicate that syphilis becomes generalized very early in the 
primary stage and that surgical removal of a chancre alone has little chance 
of preventing general infection. While the work of Brown and Pearce8 
indicates that surgical removal of the primary lesion aids in the dissemination 
of syphilis of the rabbit probably by removing a factor for the production 
of immunity principles, and that the local struggle between parasites and 
host determines in an important manner the subsequent dissemination of 
the disease, there can be little doubt of the wisdom of destroying the spiro- 
chetes in human chancres as quickly as possible by the prompt administra- 
tion of adequate amounts of antispirochetic drugs. 
Antibody production in syphilis, therefore, appears to take place only 
in the tissues directly invaded by living spirochetes; there does not appear 
to be a general antibody production through stimulation of the antibody 
tissues in general by means of some diffusible toxin or other antigenic stimu- 
lant short of the living parasites themselves. The immune substances 
produced in foci of infection are, however, generally distributed and may 
offer an effectual resistance to reinfection, but the highest degree of resist- 
ance is found in the foci themselves where presumably the largest amounts 
of antibodies are to be found. Eberson and Engman,9 Eberson,10 Brown 
and Pearce,11 and others have expressed the opinion that the latent foci may 
elaborate antibodies and that latency itself connates a balance between 
spirochetes and antispirochetic substances. Unfortunately, however, the 
latter substances are not usually able to kill the spirochetes and the latter 
1 Jour. Amer. Med. Assoc., 1914, 63, 466. 
2 Wien. klin. Wchn., 1906, 19, 621. 
3 Jour. Exper. Med., 1916, 23, 329, 341. 
4 Arch. f. Dermat. u. Syph., 1920, 2, 675. 
5 Ibid., 1921, 3, 254. 
6 Ztschr. f. Immunitatsf., Ref., 1910, 2, 277. 
7 Arch. f. Dermat. u. Syph., 1920, 2, 470. 
8 Arch. f. Dermat. u. Syph., 1921, 3, 254. 
9 Jour. Amer. Med. Assoc., 1921, 76, 160. 
10 Arch. f. Dermat. u. Syph., 1921, 3, 111. 
11 Arch. f. Dermat. u. Syph., 1920, 2, 470. VACCINATION AGAINST SYPHILIS 
827 
may spring into activity as a result of trauma or other cause, and as the years 
go by progressively involve new organs and new tissues of the same organ. 
The question now arises regarding the nature and mechanism of the 
apparent resistance to reinfection shown by the syphilitic which develops 
during the primary stage and persists in a marked or absolute degree for 
the balance of life or until a complete cure is brought about by antisyphilitic 
remedies. When the uncured syphilitic is exposed to reinfection what 
seemingly prevents a new infection? Surely reinfection would seem possible 
when even in the old foci antispirochetic substances are not sufficiently 
powerful or numerous to effect sterilization; probably it is a matter of 
numerical infection, that is, the syphilitic carries sufficient antispirochetic 
agents to kill the relatively few spirochetes applied to the skin or mucous 
membrane upon exposure and thereby prevent reinfection, but these are 
insufficient to kill off all of the very large numbers of spirochetes of the 
previous infection now distributed in various foci throughout the body. 
If such were true it may be possible to break down this weak but usually 
sufficient resistance to reinfection by using spirochetes of enhanced viru- 
lence or in extra large numbers, and, indeed, animal experiments have indi- 
cated that this may be the case. Or it may be that there is no effectual 
resistance to reinfection at all in syphilis after the primary stage, meaning 
that new spirochetes may enter the tissues of the syphilitic upon exposure 
or inoculation, but simply mask their entrance by failure of the body cells 
to produce a local inflammatory reaction—a state of “anergie” according 
to Neisser. It may be that the local inflammatory reaction, the chancre, 
for example, is due to the irritant action of toxic substances elaborated by 
spirochetes and that in syphilis there is sufficient antibody to neutralize 
these and thereby prevent inflammation, although the spirochetes them- 
selves are not killed and successively invade the tissues with subsequent 
distribution by lymphatic and vascular channels. 
The whole subject of resistance and immunity in syphilis may be sum- 
marized as follows: 
1. Natural humoral immunity to syphilis does not exist in human beings, 
although effectual resistance to infection may be offered by intact epithelial 
barriers. 
2. After infection has occurred antibody production takes place in the 
foci of disease. 
3. These antibodies may be sufficient for holding the spirochetes in a 
state of inactivity or latency, but are not usually sufficient for killing all of 
the spirochetes and, therefore, are unable to effect spontaneous cure unaided 
by medicinal treatment. 
4. These antibodies do not persist except for a brief and unknown period 
after complete cure of syphilis; syphilis, therefore, does not leave any im- 
munity or more than a brief and temporary immunity after complete de- 
struction of all spirochetes. 
5. A second chancre cannot usually be produced in an uncured syphilitic 
after the primary stage. Resistance to second chancres is practically abso- 
lute during the secondary stage, but may be produced in rare instances in 
the tertiary stage. 
6. This resistance of the uncured syphilitic to a new chancre is due 
either to (a) the destruction upon exposure of the new spirochetes by the 
antibodies engendered by the previous infection, but which antibodies are 
unable to completely destroy the original spirochetes in the old foci of disease 
or, (b) the antibodies previously produced are able to neutralize the toxic 
products of the new spirochetes and thereby prevent a local tissue reaction 828 
PROPHYLACTIC ACTIVE IMMUNIZATION OR VACCINATION 
(the chancre or primary sore) without, however, actually destroying the 
spirochetes themselves and thereby permitting the latter to enter the tissues 
without local disturbances. 
Vaccination Against Syphilis.—Owing to the prevalence and economic 
importance of syphilis it is not surprising that early and frequent attempts 
were made to evolve a method of prophylactic immunization. Up to the 
present time the results have been largely negative and an efficient method 
has not been discovered. 
Metchnikoff and Roux1 employed filtered virus and virus heated at 
57° C. as vaccines for the immunization of monkeys, but with negative 
results in that treated animals were invariably infected as successfully as 
untreated controls. Neisser and Bruck2 employed phenol-killed vaccines 
of the virus, with similar results. Uhlenhuth and Mulzer used without 
success living spirochetes from rabbit testicular lesions by intravenous and 
subcutaneous injection; Noguchi has employed vaccines of pure culture of 
pallida for the immunization of rabbits, observing that susceptibility to 
infection was reduced in some of these rabbits, but without influence in others, 
and with generally negative results. I have repeatedly tried immunization 
of rabbits with living and heat-killed vaccines of culture pallida and sus- 
ceptibility to testicular inoculation remained practically undiminshed. 
Spitzer3 has injected individuals subcutaneously with emulsions of human 
chancre material after the primary chancre had developed, and noted in some 
individuals that generalized syphilis was prevented, but these results were 
not confirmed by Neisser with monkeys or by Brandweiner4 and Kreiblich5 
in man. 
Passage of Spirochseta pallida through monkeys and rabbits may lower 
its virulence for man and by producing a mild lesion lead to the development 
of immunity to syphilis. Metchnikoff and Roux have recorded an accidental 
laboratory infection with a monkey strain in which the chancre was mild 
and atypical and generalized symptoms did not develop. A similar result 
was observed by them in the inoculation of a volunteer seventy-nine years 
old with a strain of pallida carried for five generations in lower monkeys. 
But passage through monkeys and rabbits may not sufficiently reduce 
virulence for man, as reported by Graetz and Delbanco6 in an accidental 
laboratory infection. 
• At any rate it would appear that nothing less than living spirochetes are 
necessary for immunization; furthermore, that these must produce a local 
lesion and continue to live before immunity becomes apparent. In all 
probability these attenuated spirochetes may later acquire enhanced viru- 
lence. The immunity does not appear to be more than that developing in 
the course of syphilis, and since animal passage does not appear to safely 
modify or permanently attenuate spirochetes in a manner analogous to the 
passage of smallpox virus through cattle, and since vaccination with killed 
spirochetes has failed to immunize, it may be said that prophylactic immuni- 
zation against syphilis is not possible by present methods. 
VACCINATION AGAINST WEIL'S DISEASE (INFECTIOUS JAUNDICE) 
The cause of this disease is Spirochceta icterohcemorrhagicce discovered by 
Inada. The disease is particularly prevalent among the miners of Japan 
1 Ann. de l’Inst. Pasteur, 1903, 17, 809; ibid., 1904, 18, 1, 657; ibid., 1905, 19, 673i 
ibid., 1906, 20, 785. 
2 Beitr. z. Path. u. Therapie d. Syph., 1911, 203-216. 
3 Wien. klin. Wchn., 1905, 45; ibid., 1906, 38. 5 Wien. klin. Wchn., 1906, No. 8. 
4 Wien. klin. Wchn., 1905, No. 45. 6 Med. Klin., 1914, 375, 420. VACCINATION AGAINST ANTHRAX 
829 
the rat being a carrier and source of infection. Inada and his co-workers 
have grown the microparasite artificially and were able to actively immunize 
the lower animals. In a recent review of the subject Inada1 states that 
good results have been observed by Ito, Matsuzaki, and Wani in the prophy- 
lactic immunization of human beings. The vaccine is prepared in the usual 
manner to contain 50,000,000 to 75,000,000 of spirochetes per cubic centi- 
meter, and is administered in two doses of 1 and 2 c.c. by subcutaneous 
injection. 
PROPHYLACTIC IMMUNIZATION AMONG THE LOWER ANIMALS 
Since discoveries in bacteriology and immunity have usually been inti- 
mately associated with animal experimentation, it is not strange that the 
lower animals should have been the first to benefit from the knowledge thus 
gained. As a consequence, vaccine therapy, both prophylactic and thera- 
peutic, is being extensively used in veterinary practice with good results. 
This was one of the first vaccines studied by Pasteur, and as a prophy- 
lactic measure it has proved of great value. It finds its greatest field of 
usefulness in case of an outbreak of anthrax, when it is used to protect the 
uninfected members of a herd, as well as any animals pasturing on infected 
areas. 
In preparing the vaccine Pasteur was hampered by the fact that the 
spores of anthrax bacilli retain the virulence of the original bacilli. As the 
result of extended experiments, however, he discovered a means of attenuat- 
ing the virulence of cultures by growing the bacilli at a temperature of 
42° C.; he also found that inoculation with these attenuated bacilli would 
effectively vaccinate sheep and cattle, and so protect them against an attack 
of the disease. 
One of the most dramatic stories2 in the history of science is the account 
of the method by which Pasteur demonstrated his discovery to the public. 
Certain harsh critics, having heard of Pasteur’s ability to prevent anthrax 
in laboratory experiments, and anxious to humiliate him, sent him a public 
challenge to demonstrate the experiment on a practical scale at a farm in the 
country. A number of farmers offered to place 60 sheep at his disposal. 
The challenge was immediately accepted, and Pasteur mapped out a plan of 
action, in which he safe-guarded himself by making no half-statements, but 
boldly promised complete success. Of the total number, 25 sheep were to 
be vaccinated and 25 were to remain unvaccinated. A fortnight later all 
50 were to receive a lethal dose of the fully virulent anthrax. He declared 
that the 25 non-vaccinated sheep would die, whereas the 25 vaccinated would 
remain alive and well. The remaining 10 sheep were to serve as controls. 
The challenge and its acceptance were widely advertised in the journals, 
and Pasteur was made the subject for many witticisms. Excitement ran 
high, and a large crowd, comprised of physicians, veterinary surgeons, 
journalists, farmers, etc., accompanied Pasteur to the farm (Pouilly le Fort) 
where he was to make the final test by inoculating the deadly anthrax. 
One vaccinated animal developed a temperature overnight, a fact that 
caused Pasteur much anxiety. On going to the farm the next day, however, 
again followed by the crowd, he found all the vaccinated animals well! Of 
the unvaccinated, 22 were dead, and the others died during the following 
VACCINATION AGAINST ANTHRAX 
1 Japan Med. World, 1922, 2, 189. 
2 Narrated by Elizabeth Fraser. 830 
PROPHYLACTIC ACTIVE IMMUNIZATION OR VACCINATION 
night. Pasteur’s triumph was complete, and the possibility of preventive 
vaccination was demonstrated to the world. 
Preparation of Vaccines.—The vaccine is prepared of attenuated cultures 
of virulent anthrax bacilli according to the method of Pasteur: 
Vaccine No. 1 is weakest or lowest in virulence, and the first to be injected. 
This vaccine is prepared by growing virulent anthrax bacilli at a temperature 
of 42° C. for from six to ten weeks, or until 1/10 loopful of the culture, when 
injected into rabbits, guinea-pigs, and mice, will show virulence for mice 
only, but not for guinea-pigs and rabbits. 
Vaccine No. 2 is prepared by growing virulent anthrax bacilli at 42° C. 
for about twenty days, or until 1/10 loopful is virulent for mice, partly so 
for guinea-pigs, and not at all for rabbits. 
Vaccine No. 3 is not generally used except for immunizing sheep and 
goats. When, however, it is required, it is made as follows: Virulent 
anthrax bacilli are grown at 42° C. until 1/10 loopful, when injected into 
mice, guinea-pigs, and rabbits, will be virulent for all the mice, all the 
guinea-pigs, and some of the rabbits. 
The vaccines are prepared in ampules containing 1 c.c. of the emulsion, 
each representing one dose, to be injected subcutaneously. Vaccine No. 2 
is injected ten to twelve days after Vaccine No. 1, and No. 3 after the same 
interval following Vaccine No. 2. The resulting immunity usually lasts 
about six to twelve months. 
At the present time anthrax vaccine is generally prepared by cultivating 
the bacilli and spores on agar for about five days at 38° C. followed by sus- 
pending them in saline solution and heating at 60° C. for one hour. A 
count is then made of this stock suspension by plating methods and dilutions 
prepared preserved with tricresol to give vaccines of approximately these 
strengths: 
No. 1, No. 2, and single vaccine to contain from 1,000,000,000 to 2,000- 
000,000 per cubic centimeter. 
No. 3 and No. 4 to contain from 2,000,000 to 5,000,000 per cubic centi- 
meter. No. 4 is only occasionally used in localities where the incidence of 
anthrax is high and the maximum degree of immunization is demanded. 
The vaccines are also available in pellet form which are placed in the 
subcutaneous tissues by means of a special injector. 
Technic of Vaccination.—The animals about to be vaccinated should be 
tagged, numbered, described, and the temperature of each should be taken 
for additional assurance that they are not affected with anthrax. If the 
stable or other premises are in an unsanitary condition the animals should 
be removed to a clean, open lot for vaccination, in order to avoid risk of 
wound infection or contamination of the vaccine. This precaution is 
imperative. 
It should be carefully remembered that the vaccine is living and the 
utmost care should be taken to avoid the loss of a single drop. 
The site of injection, usually behind the shoulders (first injection upon 
the right and the second upon the left side), should be cleansed by scrubbing 
with soap and water and, after drying with swab of clean cotton or cloth, 
the area through which the needle is plunged should be disinfected with 
tincture of iodin or other reliable germicide. The dose of vaccine, con- 
sisting of 1 c.c., should be injected under the skin and the needle, on with- 
drawal, should be wiped and placed in a pledget of cotton moistened with 
tincture of iodin until the next injection is about to be made. Otherwise 
a sterile needle should be used for each animal. The used needle should be 
taken from the syringe with a pair of sterile forceps and a sterile needle VACCINATION AGAINST BLACKLEG 
831 
replaced in the same manner. Horses and mules should be injected under 
the skin of the neck, above the collar space. Draft animals usually need not 
be withdrawn from their work during vaccination and cattle in dairy herds 
may be vaccinated without affecting the milk supply. 
Animals should not be immunized within two weeks of an operation. 
During hot weather the injections should be given, as far as possible, during 
the morning or evening hours. As a general rule pregnant animals may be 
immunized as there is little danger of abortion if rough handling is avoided. 
Animals that show symptoms of the disease should not be vaccinated. 
Dosage.—Under ordinary conditions one dose of vaccine (No. 1) will 
protect animals for about one year. Preference should always be given, 
howrever, to double vaccination, the second dose (No. 2) following seven to 
ten days after the first. This immunity is more complete and durable. 
In badly infected districts a third dose (No. 3) should be given about 
ten days after the second. 
Vaccinated animals should not be turned into infected pastures for at 
least two weeks after the last injection, as it takes about this period for 
immunity to develop. 
In instances where it is desirable to immunize a herd before turning 
them out to pasture on infected areas it is well to inoculate the animals in 
the early spring, keeping them in the stable during the time required for at 
least two vaccinations, for the reason that, immediately after vaccination, 
the animals may become hypersusceptible to infection. 
Vaccination with Vaccine and Serum.—When immediate protection is 
required it is advisable to inject at least 10 c.c. of antianthrax serum sub- 
cutaneously followed at once by the subcutaneous injection of 1 c.c. of 
vaccine (No. 3) at another site. The two should not be mixed, but injected 
separately. 
Practical Value.—When conducted in a careful manner by a competent 
veterinarian anthrax vaccination has proved fairly successful. 
The immunity usually lasts for eight to twelve months at least. 
Vaccination is not without danger. Some animals may die as a result 
of the injections. These are usually hypersusceptible to anthrax and are 
beyond control. Severe reactions of this kind may be controlled by the 
prompt intravenous injection of antianthrax serum in curative doses (30 
to 100 c.c.). 
Blackleg vaccine is used entirely as a prophylactic agent, for the disease 
runs too acute a course for the vaccine to exert any therapeutic influence. 
The vaccine as employed at the present time is in the nature of an aggres- 
sin previously described. 
Preparation of Aggressin.—Young cattle over six months of age are 
injected into the gluteal muscles with approximately 40 gm. of dried muscle 
virus. As a general rule the animals succumb in thirty-six to forty-eight 
hours. Immediately after death the infected tissues are removed and the 
bloody serous fluid collected. The muscle is then ground and the juice 
expressed and collected. This is added to the serous fluid and the whole 
preserved with 0.3 per cent, tricresol, filtered through candles and tested 
for sterility by aerobic and anerobic cultures. Guinea-pigs are also injected 
with 5 c.c. amounts and should survive. The finished product has a clear 
cherry red color and after standing may show a slight sediment. 
The vaccine is prepared by the Bureau of Animal Industry in the fol- 
VACCINATION AGAINST BLACKLEG (SYMPTOMATIC ANTHRAX) 832 
PROPHYLACTIC ACTIVE IMMUNIZATION OR VACCINATION 
lowing manner: The muscle tissue from a fresh blackleg tumor is ground 
in a mortar, extracted or macerated with a little water, and the fluid squeezed 
through cheese-cloth. The expressed fluid is then evaporated at a tem- 
perature of 35° C. The dry brown scale is run through a grinding mill 
and heated for six or seven hours at a temperature of from 94° to 96° C. 
This process of heating attenuates the virulence of the bacilli present, so 
that when injected they produce but a mild attack of the disease. The 
Department of Agriculture places the ground material in packages con- 
taining a certain number of doses. These packages are, upon request, 
mailed to veterinarians, who dilute the ground muscle with as many cubic 
centimeters of sterile water as there are doses in the package. One cubic 
centimeter of the suspension is injected subcutaneously in some convenient 
area, as, e. g.} about the shoulders. 
Blackleg vaccine should be applied in the spring, before young cattle 
are turned out to pasture on infected areas; the injections are subcutaneous. 
Reactions seldom occur. 
In case of a fresh outbreak, all the healthy animals in the herd are to be 
vaccinated as soon as possible; sick animals are never to be injected. 
Blackleg vaccination has been fairly successful, and usually confers an 
immunity lasting for a period of about six to nine months. 
In young animals (up to two years) the immunity is of shorter duration 
than in older animals. After two years a natural immunity gradually 
develops. 
VACCINATION AGAINST INFECTIOUS ABORTION OF COWS 
The primary and essential cause of infectious abortion of cows is now 
regarded as the Bacillus abortus of Bang which causes an infection of the 
placental tissues and fetal membranes (chorion). Infection probably occurs 
through the intestinal tract followed by bacteremia and may take place at 
any stage of pregnancy. The bull seldom transmits the disease, but possibly 
may do so in a mechanical manner or as a result of infection of the testicles. 
Pus producing bacteria and certain fungi may also produce abortion, but 
the bacillus of Bang is the most frequent cause and the diagnosis is facili- 
tated by the complement-fixation and agglutination tests previously des- 
cribed. Calves born of infected cows may harbor the bacilli for si short 
time, but immunity is not inherited and natural immunity to the disease 
is uncommon. Acquired immunity gained by the cow passing through an 
attack usually lasts for the balance of the animal’s life. 
Dosage and Administration.—Two vaccines are being employed for the 
immunization of cattle. A vaccine of killed bacilli may be injected sub- 
cutaneously in a single dose of 20 c.c. or better, in three or four doses of 5 c.c. 
each at intervals of five to seven days. Each cubic centimeter usually contains 
about 25,000,000,000 bacilli. The results have not been as good as observed 
after vaccination with living vaccine, but killed vaccine may be employed 
in herds where actual abortions are occurring and a number of pregnant 
animals have been exposed to infection. It may be given early in preg- 
nancy, or at time of herding, in 5 c.c. doses monthly until delivery. 
Living sensitized vaccine affords better protection, but obviously is more 
dangerous. The dose is 20 c.c. by subcutaneous injection. Only non- 
pregnant animals should be injected, and they should not be bred for at least 
two months, to allow for the destruction of any living bacilli which may have 
reached the uterus. 
The immunity following vaccination with killed or living vaccines usually VACCINATION AGAINST CANINE DISTEMPER 
833 
lasts for about one year, and in infected herds it is recommended that the 
vaccine be given after each calf is born. 
Practical Value.—Vaccination with killed vaccine probably engenders 
some degree of immunity of short duration. Its use is advisable when the 
disease breaks out in a herd in order to give assistance to the pregnant 
animals in coping with the infection. 
The living vaccines given at the right time (about two months before 
pregnancy occurs) affords a greater degree of protection. Huddleson has 
shown that vaccination of non-pregnant cows and heifers with these engen- 
ders an immunity sufficient to enable the animals to withstand intense 
exposures and the intravenous injection of living bacilli. A number of 
fairly well-controlled experiments have shown that in infected herds with 
20 to 30 per cent, of animals aborting, immunization with living cultures 
reduces the incidence to about 5 per cent. Calving efficiency does not seem 
to be impaired. 
When the disease breaks out in a herd the infected animals and all having 
vaginal discharge should be isolated. All fetuses and contaminated bed- 
ding, etc., should be burned and the stable thoroughly cleaned and dis- 
infested. Living vaccine may be given all animals that have aborted or that 
freshen, as soon as they clean; also to all virgin heifers and newly purchased 
non-pregnant cows. These vaccinated animals should be bred for about 
two months. Heated or killed vaccine may be given all pregnant animals 
in the herd and all empty animals that must be bred at once. 
VACCINATION AGAINST INFECTIOUS ABORTION OF MARES 
This disease is now regarded as being caused by Bacillus abortus equi, 
first described by Smith and Kilbourne,1 and quite different from the bacillus 
of Bang, responsible for the corresponding disease of cattle previously dis- 
cussed. This bacillus is also regarded by many as the cause of infectious 
arthritis (joint-ill) of colts. 
Good and Smith2 have found that vaccination of mares before infection 
with killed vaccines engenders some degree of immunity and does no harm. 
Four subcutaneous injections of 5 c.c. each at intervals of five to seven days 
may be given. Immunization may also be effected by two injections of 
10 c.c. each at intervals of five days or one injection of 20 c.c. 
Schofield3 and Hardenbergh4 have observed good results in the prevention 
of arthritis of colts by giving several subcutaneous injections of vaccine in 
dose of 2 or 2.5 c.c. at intervals of two to five days. 
Canine distemper is now generally believed to be caused by Bacillus 
bronchisepticus discovered by Ferry.5 Other micro-organisms, as staphylo- 
cocci and streptococci, are commonly found in the discharges, but are re- 
garded as secondary invaders. 
Dogs frcm one to twelve months of age are particularly susceptible; 
animals over three years old are rarely infected. The incubation period 
varies from two to eighteen days, and young animals that have come in 
contact with the disease within three weeks should be considered as possibly 
infected. 
Vaccination Against Canine Distemper 
1 U. S. Dept. .Agriculture, Bureau of Animal Industry, 1893, Bull. No. 3, 49. 
2 Jour. Infect. Dis., 1914, 15, 347. 
3 Toronto, Canada Govt., 1915. 
4 Circular 48, Dept. Agriculture, Pennsylvania. 
5 Amer. Vet. Rev., 1910, 499. 834 
PROPHYLACTIC ACTIVE IMMUNIZATION OR VACCINATION 
Vaccination with killed vaccines is regarded by some as affording pro- 
tection for eight to twelve months and is probably worth while. For a dog 
of ordinary size three doses of 2 c.c. each (approximately 2,000,000,000 
bacilli) may be given by subcutaneous injection at intervals of five to seven 
days. Puppies of about four to six weeks should receive smaller doses 
(0.5 to 1 c.c.). 
Vaccination Against Canine Rabies 
As stated on p. 791, Semple, Umeno and Doi, Eichhorn and Lyon, have 
shown that one large dose of phenolized fixed virus by subcutaneous injec- 
tion appears to protect dogs against street virus. The duration of the 
immunity is not definitely known. When dogs are bitten by rabic animals 
it may be worth while to give them this injection not only for the protection 
of the animal against the disease, but more particularly to protect human 
beings. Until the method is put on a more conclusive basis, however, it is 
advisable to quarantine bitten dogs for a sufficient time even though they are 
apparently immunized. Of course this is not necessary with dogs not bitten, 
but immunized for protection. However, if an injected dog is subsequently 
bitten by a rabic animal, he should be destroyed or very carefully quaran- 
tined. 
Other Diseases 
Vaccination Against Hemorrhagic Septicemia or Pasteurellosis of 
Animals.—Since the discovery by Pasteur of the Bacillus avisepticus of fowl 
cholera, a similar bacillus has been isolated by many investigators in a 
similar disease of cattle, horses, buffalo, swine, sheep, goats, and rabbits. 
These bacilli are now regarded by some investigators as being the same 
micro-organism which, adapted to one species, can produce a disease peculiar 
to that species. The general symptoms and lesions of septic infections, as 
fever, difficult respiration, tremors, and edema of the region of the throat 
are much the same for all species. 
Killed vaccines are being employed for vaccination against the disease 
occurring in cattle (B. bovisepticus), swine (B. suisepticus) and sheep (B. 
ovisepticus). For cattle 5 c.c. may be injected subcutaneously every five 
days for four doses or two injections of 10 c.c. each may be given at intervals 
of ten days, or one injection of 20 c.c. For swine and sheep four injections of 
2 c.c. each may be given every five days or two injections of 4 c.c. each at 
ten-day intervals, or one injection of 8 c.c. Hardenbergh and Boerner1 
have advocated the use of living vaccines. 
The duration of the immunity is short and infected herds require immuni- 
zation every year; available reports indicate that the incidence of the disease 
is materially reduced by vaccination. 
Vaccination Against Equine Influenza (Strangles of Horses and Mules; 
Influenza; Pneumonia).—Active immunization with killed mixed vaccines 
of Streptococcus equi of Schiitz (etiologic agent of strangles). Bacillus 
equisepticus, and other micro-organisms found in influenza and pneumonia 
of horses and mules, is claimed by some investigators to be of prophylactic 
value. These vaccines are given subcutaneously in four doses of 5 c.c. each 
at intervals of five to seven days. 
1 Jour. Amer. Vet. Med. Assoc., 1917, xlix, 55. CHAPTER XXXVI 
PRINCIPLES OF PASSIVE IMMUNIZATION—SERA IN THE PRO- 
PHYLAXIS AND TREATMENT OF DISEASE 
Serum therapy may be said to have had its origin in 1890, when von 
Behring discovered diphtheria antitoxin. He found that guinea-pigs 
surviving a subcutaneous inoculation of living diphtheria bacilli may harbor 
virulent bacilli at the site of injection without showing any evidences of 
intoxication. Subsequent investigation showed that the blood-serum of 
these animals contained the protective principles, for when the serum was 
injected into other animals along with the diphtheria toxin symptoms of 
the disease did not develop, and, indeed, as was shown later, the immune 
serum was found capable of neutralizing the toxin in the test-tube. Shortly 
afterward Kitasato made similar discoveries in studying tetanus, and these 
antitoxins have since proved of great importance, not only from the new light 
that has been thrown upon the mechanism of immunity—and they were used 
as important arguments for the humeral as opposed to the phagocytic theory, 
and form the very basis and starting-point of Ehrlich’s researches—but also 
from the new and important field of therapy that was now opened, which 
gave promise and hope for the discovery of a specific serum treatment for 
each bacterial disease. 
At the time it was thought possible to immunize animals with the various 
micro-organisms known to produce disease, and that the immune serums so 
produced may be employed in the form of specific treatment. This theory 
rested on the fact that they contained the antibodies that would quickly 
overcome the infection. With a fewr of the genuinely antitoxic serums 
these hopes have been realized; but many other serums have not yielded the 
expected and wished-for results, although at the present time the reasons 
for failure are being recognized and gradually eliminated. 
Definitions.—It wTill be remembered that in active immunization our 
own body cells are stimulated to produce antibodies, either by reason of 
the presence of a disease or as the result of vaccination with the antigen of 
the disease in a modified and attenuated form. In passive immunization, 
however, our own body cells do not produce the antibodies, but we receive 
them passively in the form of an injection of an antibody-laden serum. The 
antibodies are produced by active immunization of some other animal, usually 
a horse, and we receive the antibodies or products of this immunization in a 
passive manner, i. e., our body cells receive protection against an infection and 
aid us in overcoming it through antibodies produced in some other animal. 
For this reason the process is called passive immunization; the particular 
kind of increased resistance afforded against infection is known as passive 
immunity, and since blood-serum contains the antibodies and is the usual 
vehicle by which they are transferred, the method is called serum therapy. 
Passive Immunization for the Prevention of Disease.—In prophylactic 
immunization the immune serum is introduced into our body fluids before 
infection has actually occurred, or at least in the earliest stage of infection, 
for the purpose of placing the antibodies on guard to destroy the infecting 
micro-organism or to neutralize its products before it has had an opportunity 
to produce disease. In other words, we aim to fortify our natural defenses 
by purchasing antibodies from another animal. From the fact that these 
835 836 
PRINCIPLES OF PASSIVE IMMUNIZATION 
antibodies may be introduced in a short space of time and that in this manner 
an immunity may be quickly gained, passive immunization for prophylactic 
purposes is indicated when the danger of infection is imminent, and when it 
is impossible, or when there is not sufficient time, for us to stimulate our 
own body cells to produce our own antibodies by active immunization with 
a vaccine. 
Since the antibodies are produced in another animal, the serum, when 
introduced into our body fluids, represents a foreign protein, and, accord- 
ingly, wre find that the antibodies are retained for relatively short periods of 
time and are quickly eliminated or destroyed. In active immunization, 
however, the antibodies are in native surroundings, and our body cells 
continue to produce them for some time after active stimulation has ceased, 
in this manner insuring a higher degree of immunity and one of longer 
duration. For purposes of prophylaxis, therefore, active immunization is 
always more desirable than passive immunization; not infrequently the two 
forms are used simultaneously, as the antibody-laden serum will afford 
instant protection, while the vaccine is stimulating our body cells to produce 
antibodies that will increase and maintain the protection over a longer period 
of time. This mixed form of immunization has recently received special 
study by von Behring in immunization experiments against diphtheria, and 
will be considered in detail in a later section. 
For purposes of prophylaxis only two immune serums have proved their 
efficiency, namely, the antitoxin of diphtheria and tetanus antitoxin. 
As will be pointed out further on, diphtheria antitoxin, when administered 
in sufficient amounts, affords protection for at least from four to six weeks; 
mixed immunization, by means of the simultaneous injection of a neutral 
mixture of the toxin and antitoxin, has been found to yield equally good 
and more prolonged immunity. 
Tetanus antitoxin has its greatest value as a prophylactic. When 
symptoms of tetanus have once appeared, serum treatment may be of no 
avail, whereas it has proved its efficiency beyond doubt in neutralizing the 
toxin before it reaches or unites with the nervous tissue. In all wounds 
likely to be infected with tetanus the physician should include the adminis- 
tration of tetanus antitoxin as a matter of routine treatment. 
Of the antibacterial serums, many have a prophylactic value in experi- 
mental animals, but none, with the exception of the antiplague serum, is 
in general use as a prophylactic in human practice. The reasons for this 
are apparent when it is remembered that pneumococcus, streptococcus, 
and meningococcus infections are not sufficiently epidemic in character to 
demand passive immunization. Meningococcus meningitis may, however, 
be an exception, but the method of active immunization advocated by 
Sophian is promising, easier to carry out, and should be tried during times 
of epidemic meningitis. 
In typhoid fever, cholera, and dysentery antibacterial serums have not 
been generally used in prophylaxis, although it would appear that a potent 
anticholera serum would prove of value in preventing epidemics of this 
frightfully infectious disease. 
With the exception, therefore, of the true intoxications, prophylaxis 
is more readily secured by active than by passive immunization. This 
is certainly true of typhoid fever, rabies, and smallpox. The antiserums 
of other micro-organisms, such as the pneumococcus, streptococcus, and 
gonococcus, are being used exclusively for therapeusis, rather than for 
prophylaxis, of their several infections. 
In veterinary practice hog-cholera serum has proved of value as a prophy- 837 
VARIETIES OF PASSIVE IMMUNIZATION 
lactic means of combating and limiting epidemics of hog cholera. It is not 
definitely known whether this serum is antitoxic or antibacterial, but it is 
probably a combination of both. 
Passive Immunization for the Treatment of Disease (Serum Therapy).— 
In curative immunization the conditions are somewhat different. During 
the course of an infectious disease our body cells are actively engaged in 
combating the infectious agent, so that reinforcements, in the form of 
specific antibodies, are indicated and welcomed for the aid they give in 
overcoming an infection and the relief they afford our hard-pressed pro- 
tective mechanism. 
For these reasons it may he stated that the more acute the injection, the 
greater is the indication for introducing an antibody-laden serum. In chronic 
infections and in some acute infections we may practice active immunization 
by introducing a vaccine, with the purpose in mind of stimulating dormant 
cells to produce antibodies; but, as a rule, it is reasonable to assume that 
in a severe generalized infection our body cells are doing their utmost to 
overcome the infection, and extra stimulation may be actually harmful. 
By introducing antibodies produced in some other animal, however, practi- 
cally no extra strain is thrown upon the body cells; on the contrary, they 
may be relieved when the new antibodies overcome the products of infection, 
and in this manner afford them an opportunity to recover. 
In the treatment of disease immune serums have proved of value in diph- 
theria, tetanus, cerebrospinal meningitis, and, to a lesser extent, dysentery, 
pneumonia, streptococcus infections, and plague. 
While antipneumococcus, antistreptococcus, and antigonococcus serums 
have proved of some value in the treatment of their particular infections, 
the more recent work of Neufeld.and Handel, Dochez and Cole, and their co- 
workers in pneumonia indicates that there are wide biologic differences 
among various strains of these micro-organisms, and that no curative prop- 
erties can be expected from a given serum unless this is homologous for the 
type causing the infection. Further than this, it has been found impossible to 
secure serums as rich in antibodies as are secured with diphtheria and tetanus 
antitoxins, and that the serums must be given intravenously in relatively 
large doses. A method for the quick recognition of types of pneumococci 
has been worked out in the Rockefeller Hospital, and immune serums have 
been prepared for the main types, and the results of the serum treatment of 
pneumonia along these lines have been found to be most encouraging. While 
this method is not adapted for general use, it holds out a promise for the future 
of serum therapy, and opens up a wide field of investigation with the group 
of streptococci, gonococci, and meningococci. 
Varieties of Passive Immunization.—While, strictly speaking, all anti- 
bodies are probably inimical to their antigens, from the practical standpoint 
of passive immunization three are of primary importance, namely: (1) 
The antitoxins, (2) the bactericidans and bacteriolysins, and (3) the 
bacteriotropins (immune opsonins). The antitoxins neutralize their toxins; 
bacteriolysins cause the death of their respective bacteria if suitable com- 
plements are present, and bacteriotropins accomplish the same end by 
lowering the resistance of t le bacteria, and in this manner facilitating 
phagocytosis. Other antibodies may be operative and prove of assistance, 
as, e. g., agglutinins may aid in bacteriolysis and anti-aggressins may aid in 
phagocytosis, but too little is known at the present time to allow fine dis- 
tinctions to be made, although the indications are that not one but several 
antibodies are present in each immune serum, which, acting together, lend to 
overcome an infection. 838 
PRINCIPLES OF PASSIVE IMMUNIZATION 
From the practical standpoint, therefore, immune serums may be used 
to produce two main types of passive immunization, namely: 
1. Antitoxic immunization, due to antitoxins for the true or extracellular 
toxins, as in diphtheria and tetanus (antitoxic immunity). 
2. Antibacterial immunization, due mainly to bacteriolysins and bac- 
teriotropins, as in meningococcus, pneumococus, streptococcus, gonococcus, 
and similar infections (antibacterial immunity). 
Kinds of Sera Employed in Serum Therapy.—(a) Immune horse and 
human sera are used for prophylactic immunization because of the large 
amounts of antibodies contained in them. These immune sera are com- 
monly prepared by active immunization of horses, as in the preparation of 
the antitoxins, pneumococcus, and meningococcus antisera; goats and other 
animals are sometimes employed, notably oxen, in some European labora- 
tories. 
In addition to these immune sera of the lower animals, immune human 
sera are sometimes employed and especially in the treatment of scarlet fever 
and acute anterior poliomyelitis with the sera of persons convalescent from 
scarlet fever and recovered from poliomyelitis. These sera contain curative 
principles, but cannot be prepared artificially until the etiologic agents are 
available for the immunization of the lower animals. These human immune 
sera have been employed for prophylactic as well as therapeutic immuniza- 
tion; the injection of serum from convalescent cases of measles and mumps 
have been advocated for passive immunization against these diseases. The 
success of this serum prophylaxis in measles strongly suggests that a similar 
procedure may prove efficient for conferring passive immunity against 
scarlet fever. 
(b) Normal sera are likewise employed in the treatment of some diseases. 
For example, the serum of cattle has been advocated for the treatment of 
human anthrax in the Argentine and probably owes part of its efficacy to 
the presence of natural antibodies inasmuch as the disease is very prevalent 
in that country and may result in the immunization of many of these 
animals. 
Normal human serum has also been advocated for the treatment of 
hemorrhage and various skin and other diseases, but probably owes its 
efficacy largely to non-specific factors. Under this heading is likewise to be 
included the transfusion of blood for the treatment of hemorrhage and 
sometimes for severe pyogenic infections. 
Mention is also to be made of the use of normal serum immunized in 
vitro by the method of Wright, and to which further reference will later 
be made. 
(c) Exudates and transudates are likewise sometimes employed for the 
treatment of disease. The former may contain antibodies and notably 
bacteriotropins, but both probably owe their curative activities to non- 
specific factors, a form of non-specific protein therapy. 
Specific and Non-specific Serum Therapy.—For prophylactic purposes, 
as in the prevention of diphtheria and tetanus, specific immune sera must be 
employed containing antibodies capable of neutralizing the bacterial poisons 
and aiding in the actual destruction of the microparasites. For the treat- 
ment of disease immune sera containing large amounts of various antibodies 
should likewise be employed for these purposes, but in addition to these 
specific activities, immune sera probably owe part of their efficacy to non- 
specific properties largely in the way of stimulating leukocytosis and an 
increase of natural antibodies and proteolytic ferments. Normal serum 
may owe part of its curative effects to the presence of natural antibodies, REACTIONS AFTER THE ADMINISTRATION OF SERA 
839 
but these are not sufficient for prophylactic immunization, although they 
may aid in the treatment of disease; normal sera probably owe their efficacy 
entirely to non-specific activities as previously mentioned and as will be 
discussed in more detail in Chapter XXXIX. 
Duration of Passive Immunity.—Immune horse-serum is employed in 
human medicine for prophylaxis against diphtheria, tetanus, and gas gan- 
grene; in veterinary practice against anthrax, symptomatic anthrax (black- 
leg), hog cholera, and a few other diseases. Convalescent human serum has 
been employed for prophylactic immunization against measles and mumps 
with encouraging success, and a similar prophylaxis may be developed for 
scarlet fever and other of the infectious diseases. 
The duration of the immunity conferred by horse and human sera is 
quite brief and probably not more than six weeks even when homologous 
serum is employed. Passive immunization, therefore, has its greatest 
value for tiding over a period of emergency as in epidemics; for a more lasting 
immunity active immunization, by which our own body cells are stimulated 
and probably continue to produce protective principles, after the inocula- 
tions cease, is required, and is generally more effective. 
Reactions After the Administration of Sera.—(a) The administration 
of homologous sera as normal or immune human serum to a human being by 
subcutaneous or intramuscular injection does not ordinarily produce more 
than a local reaction due to trauma and the irritation excited by the pre- 
servative; this is sometimes accompanied by mild fever that subsides within 
thirty-six hours. The local reaction is one of pain and tenderness accom- 
panied by slight erythema and edema and the severity depends largely upon 
the amount of serum injected. 
(ib) Homologous serum injected intravenously, however, may excite in 
addition to this local reaction, a mild general reaction of chilliness, slight 
fever, and possibly tachycardia, if large amounts are administered. These 
effects may be due to intravascular agglutination and hemolysis unless the 
serum employed has been found free of agglutinins and hemolysins for the 
corpuscles of the patient. They may likewise be excited in part by sodium 
citrate or other anticoagulant, if these have been employed in the prepara- 
tion of the serum. Finally this reaction may be caused by increased toxicity 
due to colloidal changes occurring during the preparation of the serum and 
the reaction closely resembles a mild protein shock; these factors are discussed 
in more detail in the chapter on Blood Transfusion. 
The administration of homologous serum, however, does not produce serum 
sickness or other evidences of an anaphylactic reaction. 
(c) The subcutaneous and intramuscular injections of normal and immune 
horse-sera do not ordinarily produce more than a local traumatic reaction, 
the severity depending largely upon the amount injected. Ordinarily this 
is accompanied by chilliness, mild fever, headache, and other general symp- 
toms subsiding in thirty-six to forty-eight hours and a form of protein 
shock reaction. 
Serum sickness may develop at a subsequent time and in rare instances 
a severe reaction of dyspnea, tachycardia, coma, and even death may be 
induced in those human beings who are “horse asthmatics” and acutely 
hypersensitive or anaphylactic to horse-serum protein (see Chapter XXX). 
(d) Intravenous and intraspinal injections of horse-serum are usually 
followed by a reaction ascribed to the effects of the various foreign proteins. 
Sometimes and, indeed, not infrequently there is practically no reaction, 
and especially if it happens to be the first dose of horse-serum ever taken 
by the patient. 840 
PRINCIPLES OF PASSIVE IMMUNIZATION 
As a general rule, however, and especially if 50 c.c. or more of serum is 
administered intravenously, a reaction sets in from fifteen minutes to one 
hour later characterized by chilliness, slight dyspnea, and cyanosis. The 
temperature rises rapidly 1° to 3° F. and then falls, often to normal. Pro- 
fuse perspiration may now take place. As shown by Matsumoto and the 
writer1 the amounts of agglutinins and hemolysins for human erythrocytes 
present in horse-sera are probably too slight to induce reactions by intra- 
vascular agglutination and hemolysis, and especially if the serum is diluted 
and slowly injected. 
Severer reactions may occur if the patient is hypersensitive to horse- 
serum proteins and serum disease may develop at a subsequent time. These 
reactions are described more fully in Chapter XXX, and their nature, 
means for prevention, and treatment should be thoroughly understood by 
every physician employing serum therapy. 
Dangers and Contraindications to Serum Therapy.—The chief con- 
traindications to the therapeutic use of a serum are those dependent upon the 
serum itself, for, as will readily be understood, the introduction of anti- 
bodies themselves does not mean an extra strain upon our body cells, 
but rather the reverse. The question before us, then, is one regarding 
the possible contraindications to the injection of a foreign serum, and 
the dangers dependent upon its use. It may be stated at once that, 
in the great majority of cases, the administration of a carefully prepared 
and properly administered serum is free from danger. Since the introduction 
of diphtheria antitoxin in the prophylaxis and treatment of that disease 
many thousands of injections have been given, in all parts of the world and 
under all sorts of conditions, and the number of fatalities is so small as to be 
regarded as almost negligible. Serum therapy should not, however, be 
abused to the extent of using the serum indiscriminately. I am opposed 
to using the serum as a prophylactic unless the indications for its employ- 
ment are distinct; for example, in diphtheria it suffices to immunize only 
those who have been brought into immediate contact with the infection. 
When, however, the indications are clear and the symptoms of infection are 
present, I believe in using the serum early and generously. 
1. Of all possible dangers consequent to the use of serum therapy, that 
of anaphylaxis is uppermost in the minds of practitioners. While it is true 
that anaphylaxis has been the cause of some fatalities, the likelihood of this 
accident taking place is so remote, in the great majority of cases, that it 
should not occupy a prominent place in the physician’s mind, nor interfere 
with the use of the serum, as, for instance, antitoxin in the treatment of 
diphtheria. It is true that serum sickness is comparatively common, and 
while the symptoms are frequently distressing, they are not dangerous and 
do not constitute the dreaded and fatal anaphylaxis. With a little dis- 
crimination and care on the part of the physician the risk of anaphylaxis 
may be rendered still more remote if attention is given to the following 
questions: 
(a) Is the patient sensitive to horse protein? This is probably the most 
important single question, as in several of the fatal cases of anaphylaxis on 
record it was learned afterward that the patient was usually rendered un- 
comfortable, and that sneezing, asthma, or even an urticarial rash would 
develop when the patient came into close proximity to horses, as in a stable, 
or when driving behind them, etc. Fortunately, these cases are very few, 
but several of the fatal cases of anaphylaxis on record occurred in just such 
persons, and at the present time a physician should generally be able to 
1 Jour. Immunology, 1920, 5, 75. DANGERS AND CONTRAINDICATIONS TO SERUM THERAPY 
841 
detect this susceptibility and avoid the dangers of anaphylaxis. In Chapter 
XXX the subject is discussed in greater detail, and skin tests are described 
in Chapter XXXI by which it may be possible to detect this condition. 
(ib) Has the patient been injected with a serum on any former occasion? 
If an injection has been given, especially a few weeks earlier, a reinjection 
of serum may cause well-marked serum sickness, but the possibilities of 
alarming anaphylaxis are so remote that serum should never be withheld 
if the clinical condition indicates that it should be given. Not infrequently 
a child receives an immunizing dose of diphtheria antitoxin, but develops 
the disease a month or two later, after the immunity has disappeared. 
Under these circumstances antitoxin should not be withheld. If time per- 
mits, the physician may inject 0.5 c.c. of the antitoxin for the purpose of 
producing anti-anaphylaxis, followed in two or three hours by the remainder 
of the serum. (See Chapter XXXII.) 7/ it were possible to obtain it, 
it would be good practice to immunize the patient with an ox-serum antitoxin, 
and then, if it was found necessary later to use an antitoxin, the usual horse- 
serum antitoxin could be employed. This would still further eliminate the 
possibility of the development of disagreeable or dangerous complications. 
2. If a patient suffers from idiopathic asthma and the condition known 
as status lymphaticus develops, serum should be given cautiously because 
of the increased respiratory difficulties that may follow. It may be well 
to give a preliminary hypodermic injection of atropin and caffein, and then, 
after a few minutes, give the serum, injected slowly and subcutaneously. 
3. Aside from these questions, the physician may be called upon to 
decide if a patient is physically able to withstand the effects of an inocula- 
tion, especially the intravenous injection of relatively large amounts of 
serum, such as are given in the treatment of pneumonia. In diphtheria in 
very young and weak children, when a large number of units or several 
injections are to be given, concentrated antitoxin is to be preferred, in order 
that injury to the subcutaneous tissues, pain, and shock may be reduced to 
a minimum. CHAPTER XXXVII 
METHODS FOR THE ADMINISTRATION OF SERUM IN THE 
PROPHYLAXIS AND TREATMENT OF DISEASE 
Upon the nature and severity of the infection wall depend the question 
whether the serums are to be given subcutaneously, intramuscularly, intra- 
venously, or intraspinously {subdurally). In diphtheria the antitoxin may 
be given subcutaneously unless the infection is quite severe; in the latter 
case it should be given intramuscularly or intravenously. In tetanus the 
serum should be given subdurally and intravenously. In epidemic cere- 
brospinal meningitis the serum is always given subdurally. In pneumo- 
coccus, streptococcus, and gonococcus infections, while the serum may be 
given subcutaneously or intramuscularly, it is best administered intrave- 
nously. It is important for the physician to know and appreciate that the route 
and method of inoculation and the amount of serum administered are important 
factors in determining the success or failure of serum therapy. 
L TECHNIC OF SUBCUTANEOUS INJECTION 
Serum given subcutaneously is slowly absorbed, and a portion of the 
antibodies may be destroyed before they reach the blood-stream. When 
large quantities of serum are to be given, as in pneumonia and strepto- 
coccus infections, this method may not be permissible on account of the 
pain and injury to the subcutaneous tissues that may result, aside from 
the more important question of slow absorption and anchorage or destruc- 
tion of the antibodies in various tissues before they reach the blood-stream 
or the focus of disease. 
1. Injections should be given where the subcutaneous tissues are loose, 
where movement is least marked, and preferably where pressure upon the 
parts is least likely to occur, for some soreness, dependent upon the bulk 
of the injection, is bound to follow. For these reasons injections may be 
given in the abdominal wall; some prefer the back, in the region of the lower 
angle of one of the scapulae, and the buttocks, but in a bedfast patient pressure 
at these points cannot readily be eliminated. 
2. The skin about the site for injection may be prepared by an applica- 
tion of tincture of iodin; this is washed off with alcohol just before the needle 
is inserted. After the injection has been given the remaining iodin should 
be removed with alcohol, to prevent the occurrence of a dermatitis, and the 
puncture wound covered with cotton and collodion or with sterile gauze 
fastened with adhesive straps. 
3. The syringe and needle should be sterile. Manufacturers of biologic 
supplies furnish antitoxin in syringes ready for injection, and these are 
usually convenient and satisfactory. The needle should be of medium size, 
and larger than that used for ordinary hypodermic medication. All glass 
or glass and metal syringes that may be boiled are to be preferred when a 
syringe is not furnished. Before boiling such a syringe the piston should be 
removed from the barrel, as otherwise it may expand so rapidly as to cause the 
latter to crack. 
4. When all is in readiness, the syringe being loaded and the air expelled, 
the skin is pinched up between the fingers and the needle quickly inserted 
842 TECHNIC OF INTRAMUSCULAR INJECTION 
843 
into the subcutaneous tissues. The injection should be given slowly, and 
during the operation, if the patient is a child, an assistant should be on 
hand to prevent struggling. The needle may be connected with the barrel 
of the syringe by means of a short piece of rubber tubing (Fig. 180). This 
permits an injection to be given without danger of the needle being broken 
off if the patient should struggle. Most pain is experienced when the first 
few drops of fluid are injected; after that the pain is not severe unless the 
tissues are suddenly distended, as by a quick injection. 
The site of injection is painted with tincture of iodin and covered with sterile gauze fastened with 
straps of adhesive plaster. Just before the injection is given the iodin is wiped off with a pledget of 
cotton and alcohol. A fold of skin is pinched up between the thumb and forefinger of the left hand, 
the needle inserted, and the serum slowly injected. The needle is then quickly withdrawn, and the 
puncture covered with the gauze and held in place by the adhesive plaster. 
Fig. 180.—Subcutaneous Injection of Serum. 
The amount of serum that may be injected in one area depends upon 
the age of the patient. Due care should be exercised against injecting too 
much serum in one area, because of slower absorption and possible necrosis 
of the skin and subcutaneous tissues. 
As shown experimentally by Meltzer and Auer, absorption occurs much 
more quickly when inoculations are given into the muscles than when they 
are given into the subcutaneous tissues. For this reason antitoxin should 
be given intramuscularly in severe cases of diphtheria, as the technic is just 
as simple as that of a subcutaneous injection. Whenever the physician 
desires more speedy absorption than that which follows a subcutaneous 
injection, and the intravenous route cannot, for some reason, be adopted, 
the inoculation should be given in the muscles, preferably those of the 
buttocks. 
1. With proper technic very large amounts of fluid even up to 100 c.c. 
may be injected without unusual pain, deep and superficial infiltrations, 
abscess formation, or other accident. The object is to introduce the inocu- 
II. TECHNIC OF INTRAMUSCULAR INJECTION 844 
SERUM IN PROPHYLAXIS AND TREATMENT OF DISEASE 
lum into the areolar tissue on the upper surface of the fascia forming the 
extension of the fascia lata covering the gluteus maximus from whence it 
appears at the lower border of the buttock in the gluteal sulcus with prompt 
absorption and little or no local irritation. 
2. When small amounts are to be injected, as 5 c.c. of milk or serum, 1 c.c. 
of vaccine, etc., the injections are best given in the same locality and by 
the technic described, but injections into other muscle groups, as the outer 
side of the thigh or shoulder, may be employed if the patient is confined to 
bed by illness. 
3. For the gluteal injection the patient should lie prone on a table or 
bed with muscles relaxed. 
4. The site of injection is the lower inner angle of the upper outer quad- 
rant shown in Fig. 181. The skin should be wiped with tincture of iodin 
and alcohol. 
The lines indicate the quadrants, the injection being given in the lower inner corner of the upper 
outer quardant. The left hand has drawn the tissues downward. The needle has been entered and an 
injection is about to be given. 
Fig. 181.—Intramuscular Injection 
5. The syringe may be a Luer or Record or some other so made that 
the needle may be easily detached. The needle should be gage No. 22 
and be 1| inches long for children and very thin adults, 2 inches for the 
average adult, and 2\ inches for obese individuals. Particular care should 
be taken to ufee sharp needles; turned and dulled points and rusty needles 
are very painful. Syringe and needle should be sterilized and loaded with 
the inoculum. 
6. With the left hand the skin and tissues are drawn downward and 
flattened; the syringe is grasped with the right hand in such manner that 
the piston is held against the index-finger and the needle introduced in the 
direction shown in Fig. 181 by a quick thrust. The injection should never 
he made until the operator is sure a vein has not been entered. For this purpose 
gently draw on the piston for five to ten seconds; if no blood appears the 
injection may be given. If the syringe is loaded wTith defibrinated blood TECHNIC OF INTRA VENOUS INJECTION 
845 
it is necessary to detach the barrel for a few seconds to determine if blood 
appears. 
7. The injection should be slowly given, the needle quickly withdrawn, 
leakage prevented by massage with a cotton pledget and the wound wiped 
with alcohol; a small collodion and cotton dressing may be applied. 
III. TECHNIC OF INTRAVENOUS INJECTION 
The necessity of administering serum intravenously in order to obtain 
the best results, or any result at all, is becoming more and more apparent. 
In severe cases of diphtheria the best results are obtained when the’ anti- 
toxin is given intravenously; in the treatment of tetanus the tetanus anti- 
toxin should be administered intravenously as well as intradurally, and both 
Fig. 182.—Intravenous Injection. 
Air is being expelled from the syringe. This is done by first attaching the needle and expelling air 
into a pledget of cotton carrying alcohol to collect any fluid that may be expelled. 
antistreptococcus and antipneumpcoccus serums should always be given by 
the intravenous route. Recent reports indicate that the proper serum 
treatment of these infections requires large doses given intravenously. 
Physicians should, therefore, be prepared to give intravenous injections. 
Since the use of salvarsan in the treatment of syphilis has become so popular 
many physicians have perfected themselves in the technic of intravenous 
administration, but there is still great hesitancy about giving intravenous 
injections, although the methods are relatively simple and easily mastered. 
With nervous patients the injections can be made practically painless by 
the preliminary infiltration of the skin about the site of puncture with a 
few drops of a sterile 1 per cent, solution of eucain or butyn. 
Syringe Method.—When small amounts of fluid are to be injected, as 846 
SERUM IN PROPHYLAXIS AND TREATMENT OF DISEASE 
from 1 to 20 c.c., a syringe is employed. The needle should be short and 
sharp; gage No. 20 is usually satisfactory. 
1. It is best to use an all-glass syringe, or at least one with a glass barrel, 
for the physician can then assure himself that all air has been expelled and 
that the fluid is free from solid particles (Fig. 182). Further than this, a 
flow of blood into the syringe will indicate that the needle has entered the 
vein. The syringes furnished by manufacturing firms are not well adapted 
for making these injections, as the rubber plunger frequently adheres to 
the glass barrel, so that the injection will be jerky and difficult, and, besides, 
it may be difficult to determine when the vein has been entered. It is 
better to empty the contents of these syringes into a large, sterilized, glass- 
barreled syringe, such as the Record, Luer, and Burroughs-Wellcome 
syringes, which have a close-fitting but easily working piston, and are 
attached to the needle by a flange and not by a screw thread. The needle 
The vein is being steadied and the needle introduced; gentle suction is then made upon the 
piston to determine if the needle has been properly introduced, as shown by a flow of blood into the 
syringe. 
Fig. 183.—Method of Making Intravenous Injection by Means of a Syringe. 
should be sufficiently large and have a sharp but short beveled edge. A 
long point may pierce the vein through and through, and permit perivascular 
bleeding or result in a subcutaneous injection. 
2. In young children with fat arms and a weak circulation it is usually 
necessary to expose a vein at the elbow by making a small incision. In 
older children and adults a vein may stand out prominently enough to 
permit the needle to be inserted directly through the skin without making 
an incision. A firm rubber tourniquet is applied above the elbow; a very 
simple one is constructed by a single turn around the arm with a piece of 
ordinary soft-rubber tubing held in place by a hemostat. After the vein 
has been entered the tourniquet should be quickly removed and this is 
quickly and deftly accomplished by releasing the hemostat. 
3. The skin about the site of injection is cleansed with soap, water, and 
alcohol, or merely painted with iodin, which is removed with alcohol just 
before the injection is to be given, in order that the vein may become visible. TECHNIC OF INTRAVENOUS INJECTION 
847 
4. An assistant steadies the patient’s arm and should be ready to release 
the tourniquet; or, a slip knot tourniquet of single garter may be employed 
which the operator may easily release himself as shown in Fig. 183. 
5. The operator then steadies the skin over a vein—usually the median 
basilic or median cephalic—with the left thumb and forefinger, and intro- 
duces the needle into the vein (Fig. 183). A flow of blood into the syringe 
indicates that the vein has been entered. The tourniquet is then released 
and the injection slowly given. Or the needle may be detached from the 
syringe and passed into the vein; when blood appears, the syringe is quickly 
attached and the injection made. The puncture wound is then sealed 
with a wisp of sterile cotton and collodion or with gauze and a bandage. 
Fig. 184.—Method of Making Intravenous Injection by Means of a Syringe. 
This syringe was devised for the intravenous administration of serum or a concentrated solution 
of salvarsan. It is provided with a three-way cock, which permits drawing fluid into the syringe and 
then injecting it into a vein. This injection may also be given by any glass syringe; the particular 
advantage of this one is that the operator may inject more than one syringeful of fluid without removing 
the needle. The same syringe may be used for the intravenous injection of any serum, as diphtheria 
and tetanus antitoxins. 
The syringe shown in Fig. 184 is well adapted for the intravenous injection 
of serum, and was devised for the administration of concentrated solutions 
of salvarsan and neosalvarsan, but any reliable and large glass-barreled 
syringe may be used. 
Gravity Method.—Larger quantities of serum or other fluid (more than 
20 c.c.) are better injected by the gravity method, the simple apparatus 
shown in Fig. 185 being quite satisfactory for the purpose. This consists 
merely of a graduated cylinder, which serves as a measuring funnel, rubber 
tubing with a glass window at the lower end and pinch cock, and is furnished 
with a metal tip that fits the needle. The needle should be of proper size, 
and have a sharp but somewhat short beveled edge. It may be curved, as 
shown in the illustration, or may be straight. 848 
SERUM IN PROPHYLAXIS AND TREATMENT OF DISEASE 
1. The cylinder, tubing, and needle should be sterilized by boiling prior 
to use. 
As shown by Stokes and Busman1 new tubber tubing may cause reactions 
when employed for the administration of arsphenamin. The same is 
true regarding the administration of serum and citrated blood. If new tub- 
ing is employed it should first be soaked in 5 per cent, sodium hydroxid 
solution over night and thoroughly rinsed with water and sterilized before use. 
The amount of tubing required may be materially reduced by using 
several windows of glass tubing. 
2. The serum or other fluid may be warmed by placing the container in 
water of a temperature not higher than 42° C. (just comfortably hot to hold 
the hand in). 
Fig. 185.—Method oe Making Intravenous Injection by Gravity. 
This method is suitable for the intravenous administration of salvarsan or antistreptococcus 
serum, etc. The needle has been entered into a prominent vein (indicated by a flow of blood); the 
tubing has been attached by means of a metal tip which fits the needle easily and snugly; the tourniquet 
has been loosened and the injection is being given. 
Since serum is viscid and may flow too slowly it is well to dilute it with 
an equal part of warm sterile saline solution. A further advantage is slower 
introduction of the serum by reason of its dilution. 
3. The injections are best given in a vein at the elbow. The arm about 
this region should be scrubbed with hot water and soap, followed by alcohol 
and 1 : 1000 bichlorid of mercury solution, or liberally painted with tincture 
of iodin. A firm tourniquet is then applied above the elbow; a single firm 
turn of rubber tubing held by a hemostat is quite satisfactory, as when the 
vein has been entered the tourniquet should be quickly released with the 
least movement and disturbance possible, and this arrangement answers all 
requirements. Sterile towels should be placed about the arm and shoulder. 
1 Jour. Amer. Med. Assoc., 1922, 78, 580. TECHNIC OF INTRA VENOUS INJECTION 
849 
4. About 20 c.c. or more of sterile distilled water or normal salt solution 
are then poured into the cylinder, and the cock opened until all air has been 
expelled from the tubing. The fluid, serum, or salvarsan is then poured into 
the cylinder. It is a good practice to filter the fluid through several layers of 
sterile gauze, especially when salvarsan is being injected, in order to remove 
any bits of glass or other foreign bodies that may be present. 
5. An assistant holds the loaded cylinder and tubing; the operator 
steadies the skin over a prominent vein and quickly inserts the needle, A 
flow of blood indicates that the vein has been penetrated. The tubing is 
then quickly and carefully attached, the tourniquet released by unfastening 
the hemostat, and the injection slowly given. As a rule, an elevation of the 
Fig. 186.—Apparatus for the Intravenous Injection of Serum (Rockefeller Hospital). 
The diluted serum is kept warm by means of two water bottles (left apart in the illustration 
to show the container). The syringe is attached with a two-way cock for the purpose of filling and 
then slowly injecting the serum into the patient. 
cylinder of 2 or 3 feet is sufficient. If swelling occurs about the site 
of puncture and the patient complains of pain, the injection is entering 
the subcutaneous tissue; when this occurs, the pinch-cock should be closed 
and the needle removed. It is then necessary to make the injection into 
another vein or into the same vein at another site. 
6. When serum is being given, as antipneumococcus serum for example, 
the injection of the first 10 to 15 c.c. should occupy ten to fifteen minutes. 
During this time the patient should be carefully watched for pallor, cyanosis, 
dyspnea, and tachycardia. If these symptoms appear, it is well to stop the 
injection for ten or fifteen minutes to see if they increase in severity. Us- 
ually they disappear, in which case the injection is resumed. As a general 850 
SERUM IN PROPHYLAXIS AND TREATMENT OF DISEASE 
rule no symptoms appear and the injection (100 c.c. serum and 100 c.c. of 
saline) can be given in ten to fifteen minutes. 
7. Figure 186 shows the apparatus employed in the Rockefeller Hospital. 
By means of the syringe the serum is aspirated from the container sur- 
rounded by hot-water bags to keep the solution warm, and injected into the 
vein. This arrangement is particularly serviceable for the injection of the 
first 15 c.c. at the rate of 1 c.c. per minute and especially if the patient is 
suspected as being hypersensitive. 
Intravenous Injection of Children.—This is usually much more difficult 
than with adults. Children over three years of age, however, frequently 
present a sufficiently prominent vein at the elbow for successful intravenous 
injections and especially when a syringe is employed with a No. 22 needle. 
Otherwise the skin may be infiltrated with sterile eucain solution and a 
vein exposed. 
The external jugular vein may be employed as shown in Fig. 18. 
In infants the injections are best given by way of the superior longitu- 
dinal sinus, the technic being described in Chapter II and illustrated in 
Figs. 21-23. 
TECHNIC OF INTRASPINAL INJECTION 
In the treatment of epidemic cerebrospinal meningitis, influenzal men- 
ingitis and tetanus, the specific serums are administered subdurally by 
means of a needle introduced in the lumbar region. Every practitioner 
should be prepared to perform lumbar puncture for the purpose of securing 
cerebrospinal fluid for making the Wassermann reaction and the bacteriologic, 
cytologic, and chemical examinations, and the administration of serum is a 
relatively simple matter when the puncture has been successfully made. 
The technic of lumbar puncture for the purpose of securing fluid for 
diagnosis is described on p. 25. But when administering serum, and espe- 
cially in the treatment of meningitis, the clinical condition of the patient 
and the danger of sudden collapse render it advisable and necessary that 
the inoculation be given with the patient lying on his side. 
Methods.—Two methods are now being employed. The older method 
consists in injecting the serum by means of a syringe, and the later one is 
a method whereby the serum is allowed to flow in by gravity. 
Not infrequently a patient will develop symptoms of collapse during a 
subdural injection, and these have been ascribed to undue pressure, the 
injurious action of trikresol or other preservative upon the respiratory 
centers, too rapid injection, and the introduction of too large a quantity 
of serum. It is now apparent that in the past too little attention has been 
paid to the patient while the injection was being made, and serum has 
usually been administered according to more or less fixed and arbitrary 
rules, instead of being guided by the clinical condition of the patient. 
If symptoms of collapse appear during a subdural injection, they may 
be relieved by allowing the fluid within the canal to flow out again, and 
this is best accomplished when the inoculation is given by the gravity method. 
The latter method has been recommended by Sophian, Flexner, and the 
Hygienic Laboratory, and is undoubtedly the method of choice. 
Blood-pressure as a Guide in Administering Serum Subdurally.— 
According to the older and customary method of injecting serum subdurally, 
fluid is permitted to flow from the needle until from 15 to 20 c.c. have been 
removed, and an equal quantity of serum is then injected. In severe cases, 
with thick plastic exudates, only a few cubic centimeters of fluid may be 
withdrawn, and, indeed, no fluid at all may be secured. To inject arbitrarily TECHNIC OF INTRASPINAL INJECTION 
851 
a fixed amount of serum under such conditions may be highly dangerous to 
the patient on account of increased pressure. On the other hand, when 
the flow is free, it may be dangerous to permit the canal to drain until 
intraspinal pressure is reduced to the normal, a fact indicated by the flow of 
a drop of fluid every three to five seconds. 
With these considerations in mind, Sophian1 has studied the value of 
cerebrospinal fluid pressure and blood-pressure as controls on the amount 
of fluid that may be safely withdrawn and on the amount of serum that 
may be injected. During the study and treatment of 500 cases of epidemic 
cerebrospinal meningitis this last-named observer found that the blood- 
pressure was a valuable guide. Cerebrospinal fluid pressure was found to be 
misleading, owing probably to a local distention of the subarachnoid space 
at the site of injection, which resulted in readings that did not represent the 
true intracranial pressure. 
1. Usually upon the withdrawal of cerebrospinal fluid a fall of blood- 
pressure occurs. With the ordinary blood-pressure in an adult patient—■ 
about 110 mm. of mercury—Sophian recommends stopping the flow when 
there has been a drop in pressure of about 10 mm. of mercury; in children, 
about 5 mm. In a few cases there is no change in blood-pressure or even a 
slight rise; in these instances fluid may be removed until the flow has dimin- 
ished to the rate of a drop every three to five seconds. 
2. With the injection of serum the blood-pressure drops still further. 
Generally the decrease in blood-pressure is proportional to the rapidity 
with which the serum is injected and the amount injected. By the gravity 
method, under ordinary conditions, at least ten minutes should be consumed 
in administering 15 c.c. of serum. A total drop of 20 mm. of mercury 
indicates that sufficient serum has been injected. If it is desired to inject 
more, as in a severe case of meningitis, close watch should be kept for other 
symptoms of collapse. 
3. Usually, under these conditions, less serum is administered than has 
been advocated heretofore. It is apparent that the more potent the serum, 
the less bulk is required—and the bulk alone is an important factor, for a 
large injection may so injure the patient as to counteract any good that the 
serum may do. Unfortunately, there is no accurate measure of the curative 
value of antimeningococci serum. It is highly desirable that a serum be 
as potent as possible, and the physician must rely upon the reputation of 
the firm producing the serum. Efforts are being made to concentrate these 
serums, much as antitoxin is concentrated, and this is an end very much to 
be desired. 
4. Blood-pressure changes are not constant in the same patient upon 
different occasions. The pressure should be taken after each puncture 
and inoculation, for the administration cannot be guided by observations 
made on a previous occasion. 
Collapse During Subdural Inoculation.—Carter has shown, by experi- 
ments on dogs, that the first mechanical effects of increased intraspinal 
pressure were respiratory depression and marked cardiac inhibition. Sophian 
has found that similar effects may be produced during subdural injections 
of serum in the treatment of meningitis. 
The symptoms of collapse, such as stupor, superficial or deep, irregular 
and slow respiration, and dilatation of the pupils, are foreshadowed by a 
marked drop in blood-pressure. The pulse may continue good or become 
slow and irregular. Incontinence of urine and feces may occur. 
The treatment consists primarily in discontinuing the injection. By 
1 Epidemic Cerebrospinal Meningitis, 1913, Mosby Co., St. Louis. 852 
SERUM IN PROPHYLAXIS AND TREATMENT OF DISEASE 
lowering the funnel fluid is allowed to flow from the spinal canal and mix 
with the serum. If a syringe is being used, it should be detached from the 
needle or gentle suction made. After a few minutes the symptoms may 
disappear and the inoculation may be cautiously resumed until the desired 
amount of serum has been injected; otherwise the needle should be with- 
drawn. 
In addition to this procedure atropin and caffein may be administered 
hypodermically in large doses, and artificial respiration resorted to if neces- 
sary. It is well to have these drugs ready for injection before the inoculation 
is begun, so that no time will be lost when they are needed. 
Anesthesia for Subdural Inoculation.—There is no doubt that lumbar 
puncture and the subdural injection of fluid are painful, the amount of 
pain depending to some extent upon the degree of meningitis, the method 
of injection, and the skill of the operator. Severe cases of meningitis that 
are stuporous or moribund may not evince any evidences of added discom- 
fort; less toxic and robust or nervous patients may, however, suffer con- 
siderably and prove difficult subjects for injection. 
Local anesthesia may be secured by injecting a sterile 1 per cent, solution 
of eucain in the region where the puncture is to be made (see Fig. 24). Gen- 
eral anesthesia for lumbar puncture in meningitis adds a considerable 
element of danger, but if it is absolutely necessary, a few whiffs of ether or 
chloroform may be given while the needle is being inserted. In giving 
subdural injections in tetanus a general anesthetic is necessary. 
Sophian has found that if water is given through a straw while per- 
forming lumbar puncture patients will frequently drink large quantities of 
it and keep very quiet. 
Gravity Method.—The apparatus required is very simple, and consists 
essentially of a proper needle and from 12 to 16 inches of soft-rubber tubing 
attached to a container or funnel for serum and furnished with a metal tip 
by which it is quickly and readily attached to the needle. 
Several manufacturers of biologic supplies are marketing antimenin- 
gococcic serum in a special container, fashioned after that devised by Sophian 
and Alexander, with the needle and tubing adapted for the administration 
of the serum by the gravity method. Such an apparatus is shown in Fig. 187. 
The physician may, however, prepare an equally efficient apparatus, 
similar to that shown in Fig. 188, which consists of the glass barrel of a 
20 c.c. syringe attached to 14 inches of soft-rubber tubing fitted with a 
metal tip that holds the needle firmly and snugly. The whole is sterilized 
by boiling, and any quantity of serum may be administered with it. The 
apparatus is adapted for the administration of antimeningococcic serum, 
tetanus antitoxin, influenza serum, salvarsanized serum, or any other fluid, 
and has given uniform satisfaction. 
The needle should be from 10 to 11 cm. in length, -with a wide, rather 
than a narrow, lumen—about 1.5 to 2 mm. This is important in adminis- 
tering serum to a case of meningitis, in which a needle with a narrow lumen 
may become plugged with exudate. The needle should be fitted with a 
trocar. The tip should have a short bevel with a sharp edge. 
Technic.—1. The serum should be warmed to body temperature by 
wrapping the sealed container of serum in towels wrung out of water com- 
fortably hot for the hands (about 42° C.). Cold serum possibly' increases 
the pain caused by the injection, although the pressure upon the sensitive 
nerve-roots is the chief source of pain and discomfort. Due care must be 
exercised that the serum is not coagulated by too high temperature. 
2. The funnel or syringe barrel, tubing, and needle should be sterilized by boiling. The outfits furnished by the manufacturers are sterilized and 
ready for use. Two sterile graduated centrifuge tubes should be on hand 
for collecting and measuring the spinal fluid. The apparatus should be 
assembled and ready for injection, so that at the appointed time the tubing 
may be attached to the needle and the inoculation given. 
TECHNIC OF INTRASPINAL INJECTION 
853 
Fig. 187.—Outfit for Intraspinal Injection of Antimeningitis Serum by Gravity (Sophian) 
3. The patient should be placed on the left side, on the edge of a bed or. 
table. An assistant places the patient in such a manner as to arch the back 
as much as possible. A second assistant takes blood-pressure readings 
during the operation (Fig. 188). 
4. Towels wet with bichlorid are arranged about the site of inoculation. 
It is well for the operator to locate the site of injection by palpating the 
spinous processes and selecting the widest interspace, which is usually on 
a level with the crests of the ilia if the back is well arched. The skin is then 854 
SERUM IN PROPHYLAXIS AND TREATMENT OF DISEASE 
cleansed with soap, water, and alcohol, and bichlorid solution or a coat of 
iodin applied. Some degree of local anesthesia may be secured by injecting 
a small amount of a sterile 1 per cent, solution of eucain. This is advisable 
in nervous adults (Fig. 24). 
The usual site of puncture is between the fourth and fifth lumbar verte- 
brae, but in repeated punctures it may be necessary to puncture one or two 
interspaces higher up in order to avoid pain and the formation of adhesions 
in the subarachnoid space. 
Lumbar puncture may be done as high as between the first and second 
lumbar vertebrae, and when no fluid is obtainable at the lower levels, high 
puncture must be conducted. Punctures may also be done between the 
twelfth dorsal and first lumbar vertebrae and even between the eleventh and 
Fig. 188.—Intraspinal Injection by Gravity. 
The line marks the crest of the ilium, and indicates the third lumbar interspace. The needle has 
been inserted and cerebrospinal fluid withdrawn. Antimeningococcus serum is being administered. 
The barrel of a 20 c.c. Record syringe is serving as a funnel, and is attached to the needle by means of 14 
inch es of soft-rubber tubing furnished with a metal tip. The needle, as shown, is reduced to about 
four times its actual size. This method may be used for making the subdural injection of tetanus anti- 
toxin’and salvarsanized serum. 
twelfth dorsals. Above the level of eleventh dorsal puncture may be useless 
because of an incomplete posterior subarachnoid space in many individuals 
at this level, but puncture may be done high in the dorsal or at the lower 
end of the cervical region and the canal washed out with sterile saline solution 
between this point and a needle in the lower lumbar region. 
However, high punctures are more dangerous than low punctures by reason 
of possible injury to the spinal cord. For this reason punctures above the 
tip of the cauda equina (third lumbar vertebra in infants and first lumbar 
vertebra in adults) must be made with extraordinary care. 
5. The operator must then choose between the median or lateral route 
of puncture. The median is the easier, and should always be adopted by 
the inexperienced operator. TECHNIC OF INTRASPINAL INJECTION 
855 
Wash off the iodin with a pledget of cotton soaked in alcohol. Locate 
the chosen interspinous space, pressing well between the spines with the 
left thumb or index-finger, and holding the finger in place pass the needle 
perpendicularly in the median line between the spines, or, better still, at 
an angle of 45 degrees upward and inward. If an obstruction is felt, with- 
draw the needle slightly and pass it in a different direction until it imparts a 
sense of “giving way,” which indicate sthat the subarachnoid space has been 
reached. Quincke has estimated the depth of lumbar puncture in adults 
to be usually from 4 to 6 cm.; in large muscular men it is from 7 to 8 cm., 
and in fat persons, about 10 cm. 
The needle should be inserted slowly and deliberately, rather than 
quickly, as puncture of a bone is likely to be followed by a dull, aching pain, 
and, indeed, the point of the needle may be bent or broken. 
The fluid may fail to flow or flow very slowly. This may be due to the 
presence of a thick exudate, impalement of a nerve filament, or adhesions 
arising from a previous puncture. The needle may be turned gently or the 
trocar inserted to remove an obstruction, after which the flow usually starts; 
if it does not do so, the needle may be withdrawn slightly or cautiously 
inserted a little further. 
6. Fluid is collected in the centrifuge tubes while blood-pressure readings 
are being made. When the pressure drops 10 mm., or if the flow is about 
a drop every three or five seconds, the tubing is connected and the serum 
injected very slowly. 
As a general rule, as much fluid should be withdrawn as can be done 
with safety, and the maximum dose of serum given. When the flow is 
scanty, a larger dose of serum may be given than counterbalances the fluid 
removed, the injection being guided by the blood-pressure. When the total 
drop reaches 20 mm. of mercury, the injection should be discontinued, or, 
if continued, the patient should be watched closely for other symptoms of 
collapse. 
7. After the injection has been completed the needle is quickly with- 
drawn and the wound covered with sterile gauze held in place by adhesive 
straps. All iodin should be washed off with alcohol to avoid irritation or 
an actual dermatitis. 
Syringe Method.—1. The technic is practically the same as that just 
described, except that the injection is given with a syringe. 
Manufacturing concerns market their products in syringes all ready 
for injection, but at the present time the gravity method is used almost 
exclusively. When injecting tetanus antitoxin it is necessary to empty 
the syringe into another sterile syringe, as shown in the accompanying 
illustration (Fig. 189), which will fit an appropriate needle. Since the 
plunger of the purchased syringe oftentimes adheres to the barrel and 
renders the injection jerky and difficult, I frequently transfer the serum to a 
sterile, all-glass syringe which I know will work smoothly and satisfactorily. 
2. Lumbar puncture is performed as for the gravity method while blood- 
pressure observations are being made. When sufficient fluid has been 
removed, the loaded syringe is attached to the needle and the injection 
slowly given. The physician is frequently tempted to inject the serum and 
complete the operation quickly, but it is better to inject it in amounts of 
0.5 to 1 c.c. every half to one minute, being guided by the blood-pressure 
readings and general condition of the patient. If the pressure falls below 
20 mm. of mercury or other symptoms of collapse appear, the fluid may be 
drained from the canal by gentle suction with the piston. This is usually 
impossible when the manufacturers’ syringe is used. Otherwise the syringe 856 
SERUM IN PROPHYLAXIS AND TREATMENT OF DISEASE 
is detached and the fluid collected in tubes until the patient’s condition 
improves and the injection is resumed or the needle removed. 
At times the patient is so restless that this slow method is not feasible. 
In such instances the physician should make the injection as slowly as 
possible, endeavoring to put into the canal as much serum as fluid was 
removed or at least a reasonable amount. 
Difficulties in Lumbar Puncture.—In failure to enter the subarachnoid 
space it is well to withdraw the needle entirely and try again, rather than 
withdrawing part way and probing. 
Obstruction of the needle may be due to impaction with a nerve root, in 
which case rotation of the needle suffices. The needle may not be intro- 
duced far enough, a part of the eye being outside of the arachnoid space, or 
The line indicates the crest of the ilium, and usually passes between the third and fourth lumbar 
vertebrae, which is the proper point for inserting the needle. The site of injection has been painted 
with tincture of iodin after cleansing with soap, hot water, and alcohol. An assistant holds the patient 
to prevent sudden jerking and possible accident. 
Fig. 189.—Intraspinal Injection by Means of a Syringe. 
it may be introduced too far. The needle, therefore, may be gently intro- 
duced or withdrawn to a slight extent. A flake of pus may obstruct; careful 
introduction of the stylet and rotation suffices for correction. 
Dry tap may be due not only to these causes, but likewise to hydro- 
cephalus, either internal by obstruction of the foramina of Magendie and 
Luschka, or external, the upper part of the subarachnoid space being cut 
off by adhesions. 
Hemorrhage may occur due to puncture of a vein somewhere between 
the skin and subarachnoid space. In this case only the first few cubic 
centimeters of fluid are stained. 
A vein belonging to the plexus lying on the ventral aspect of the spinal 
cavity may be punctured wrhen the needle has been introduced too far; this 
is indicated by more copious and continuous bleeding. METHODS OF PREPARING HUMAN SERUM 
857 
Blood may appear at the end of the flow due to the rupture of a venule 
as a result of injury or lowering of intraspinal pressure. 
The fluid may be uniformly hemorrhagic due to the extreme intensity 
of the inflammation; in this case the fluid does not clot. Cases of this kind 
have a very unfavorable prognosis. 
Sequelae of Lumbar Puncture.—There is usually a drop of temperature 
within four to eight hours due to the withdrawal of the purulent fluid and 
regardless of whether or not serum has been given. 
Sometimes a nerve root is struck during puncture, producing pain in the 
legs or pelvis, but paralyses of the bladder, rectum, or of muscle group or 
paraplegia are very rare. Usually the patient complains of pain during the 
injection of serum owing to pressure, but after the injection is completed 
this distress rapidly disappears. Pain and weakness in the back, so often 
complained of for some considerable time after recovery from meningitis, 
is more probably a sequel of the disease itself than of repeated lumbar punc- 
ture (Worster-Drought and Kennedy). 
Serum sickness develops in many cases and the physician should be 
perfectly familiar with the symptoms and warn the family of its possible 
occurrence in order to allay fears of recrudescence of the meningitis. 
METHODS OF PREPARING HUMAN SERUM FOR AUTOSERUM AND CON- 
VALESCENT SERUM INJECTIONS 
As will be discussed shortly in Chapter XL human convalescent serum 
and blood have proved of value in the treatment of scarlet fever, acute 
anterior poliomyelitis, and some other diseases; injections of the individual’s 
own serum (autoserum) have been found useful in the treatment of some skin 
diseases and for intraspinal injection in the treatment of neurosyphilis 
according to the Swift-Ellis method. Normal human serum and blood 
are also employed for the treatment of hemorrhage. 
Method for Securing Small Amounts of Human Serum.—1. When only 
5 to 10 c.c. of serum are to be given, 25 c.c. of blood may be aspirated from 
a vein at the elbow with a Luer or Record syringe and expelled into a sterile 
centrifuge tube or a test-tube that may be centrifuged. The skin over the 
vein should be cleansed with tincture of iodin and alcohol and every step 
carefully guarded against contamination. 
The serum may be allowed to separate spontaneously (usually over night) 
or after coagulation has occurred the clot may be gently broken up with a 
sterile glass rod and the serum secured within a few hours by centrifuging. 
The supernatant serum is then carefully pipeted into a sterile test-tube. 
2. For larger amounts of serum, as 15 to 20 c.c., blood may be secured 
from a vein at the elbow into a special large centrifuge tube shown in Fig. 
190. This tube easily carries 50 c.c. of blood and fits the International 
centrifuge. The neck is narrowed so that a rubber stopper may be em- 
ployed when centrifuging. Osborne1 has recently described a very con- 
venient type of stopper, but the ordinary rubber stopper suffices. A cotton 
or gauze stopper may be used during the collection of the blood, but is not 
adapted for centrifuging because it may be driven into the tube unless 
fastened with rubber bands, and even then fibers may be driven into the 
blood, increasing the-risks of contamination. 
It is essential to use a sharp, short, and rather wide bored needle. That 
shown in Fig. 190 is gage 18, 1 inch in length, and fits the Record syringe. 
To the end is attached 3 inches of soft-rubber tubing, the free end being 
1 Jour. Amer. Med. Assoc., 1922, 78, 580. 858 
SERUM IN PROPHYLAXIS AND TREATMENT OF DISEASE 
in the centrifuge tube, and the needle enclosed in a small test-tube strapped 
to the centrifuge tube by means of rubber bands. 
3. Blood is usually secured from a vein at the elbow a little below the 
crease in order to minimize the amount of discomfort during the following 
day or two. A firm tourniquet is applied and the patient requested to open 
and close the hand rapidly and vigorously which aids in distending the 
superficial veins. The skin is cleansed and tincture of iodin applied; which 
is later wiped away with alcohol in order that the vein may be seen. 
The needle is removed from the test-tube and grasped with a hemostat 
in the right hand, avoiding contamination of the tip; the tube is held in the 
palm of the hand and the needle inserted into the 
vein (Fig. 191). The patient is requested to 
quietly open and close the hand, which increases 
the flow of blood and prevents coagulation in the 
needle and tubing. After securing the desired 
amount of blood the tourniquet is released before 
the needle is removed in order to prevent bleeding 
Fig. 190.—Apparatus Em- 
ployed by Author for 
Aseptic Collection of 
Small Amounts of 
Blood. 
Consists of a 50-c.c. 
centrifuge tube fitted with 
a short piece of tubing and 
No. 18 needle (length 1 
inch). Needle enclosed in 
a test-tube and the whole 
sterilized by autoclaving. 
Fig. 191.—Collection of Blood. 
This method is especially useful for the collection of blood for auto- 
serum treatment. 
into the tissues. The needle and attached tubing are replaced in the test- 
tube. 
4. The blood is allowed to coagulate at room temperature and then 
placed in a refrigerator for the separation of serum; or after coagulation has 
taken place the clot may be broken up very thoroughly with a sterile glass 
rod, the rubber stopper sterilized by boiling for a few minutes, and the 
serum secured at once by centrifuging. 
5. The supernatant serum is now carefully removed with a sterile pipet 
into a sterile test-tube ready for immediate injection. If serum is to be 
kept for some time it is well to add 0.5 c.c. of 5 per cent, solution of phenol 
or tricresol to each 10 c.c. of serum (0.25 per cent.) with storage in a re- 
frigerator. METHODS OF PREPARING HUMAN SERUM 
859 
Method of Securing Large Amounts of Human Serum.—For securing 
blood from scarlet fever or poliomyelitis convalescents the above method 
may be employed when only small amounts of blood are to be taken. 
As a general rule, however, larger amounts up to 200 to 500 c.c. are 
usually withdrawn from adult subjects in good condition. 
Since the serum is to be used for the treatment of other individuals 
reasonable care should be exercised against the selection of syphilitic and 
tuberculous subjects. After the serum has been secured a small amount 
should be reserved for the Wassermann test and 5 c.c. planted in a flask of 
glucose broth for culture. 
The technic for collection is the same as described above except that a 
larger container is used and means provided for facilitating the flow of 
blood and preventing coagulation in the needle and tubing by creating a 
partial vacuum in the container. The apparatus shown in Fig. 192 is 
Fig. 192.—Apparatus por Aseptic Collection of Large Amounts of Blood 
Tube A is for suction; tube B is for collection of blood. 
convenient and readily assembled. It consists of a 500-c.c. Erlenmeyer 
flask with a two-holed rubber stopper fitted with two pieces of glass tubing 
for attaching the needle and suction tube, the whole being sterilized by 
autoclaving. 
After blood has been collected the stopper may be replaced by a sterile 
solid rubber or gauze and cotton stopper. If separation of the serum is 
incomplete or a portion becomes bloody, the material may be removed to 
centrifuge tubes with sterile pipets and clear serum secured by centrifuging. 
Each serum should be cultured for bacteria and tested for syphilis as 
mentioned above. If these tests are satisfactory the sera of several indi- 
viduals may be mixed, preserved with 0.25 per cent, phenol or tricresol, and 
stored in a refrigerator in 20-c.c. bottles. CHAPTER XXXVIII 
PROPHYLACTIC SERUM OR PASSIVE IMMUNIZATION 
SERUM PROPHYLAXIS OF DIPHTHERIA 
Immunity in Diphtheria.—Diphtheria is a disease almost entirely con- 
fined to the human race. Cats have been known to suffer with it and act 
as a source of infection for children. Several other of the lower animals, as 
rabbits, dogs, birds, and notably the guinea-pig, are susceptible to injections 
of the bacilli or toxins, but do not contract the disease spontaneously. 
Not all human beings are susceptible. Since the advent of the Schick 
test it is now known that some may possess in tbeir blood sufficient natural 
antitoxin to protect against diphtheria, the minimum required amount 
being at least l/30th unit per cubic centimeter. Children under six months 
of age are generally immune by reason of congenital antitoxic immunity 
conferred by the mother. After six to eight months this immunity dis- 
appears from the majority of children. The disease is most common be- 
tween one and seven years of age. After ten to twelve years a natural 
immunity develops in many individuals and particularly in the crowded 
districts of cities; this is probably due to the production of antitoxin as the 
result of numerous slight infections. The majority (60 to 70 per cent.) of 
adults are naturally immune. All races are susceptible; the disease is 
somewhat more common among boys than girls, probably owing to the 
greater danger of contact infection. 
One attack of diphtheria does not always confer a lasting immunity; 
second and even third attacks are not uncommon. This is probably because 
our body cells respond slowly to stimulation by diphtheria toxin with the 
production of antitoxin, and the ordinary attack of diphtheria is too brief in 
duration for adequate stimulation of antitoxin production. This is indicated 
by the fact that several months are required for the production of sufficient 
amounts of antitoxin after the injection of toxin—antitoxin mixtures. 
Toxins are the most important pathogenic agents in diphtheria and 
antitoxin is the most important antibody. Probably other immunity 
principles are also concerned both in affording protection and in recovery— 
especially phagocytosis. Bacteriolysins and opsonins are sometimes demon- 
strable during diphtheria and probably contribute to the mechanism of 
recovery. 
Serum Prophylaxis.—The subcutaneous administration of relatively 
small doses of antitoxin will usually confer a passive immunity against 
diphtheria lasting from two to four weeks. 
The object is to introduce antibodies (antitoxin and opsonin) into the 
body fluids in order that they may neutralize the toxin as rapidly as it is 
produced, aid in the destruction of the bacilli, and thus protect the individual 
in case virulent bacilli should be inspired or otherwise gain access to the 
tissues. 
Dosage and Administration.—The doses advised are relatively small, 
and the injection does not usually produce any discomfort other than sore- 
ness about the site of inoculation. For infants under one year of age 500 
units suffice; for older children and adults from 1000 to 1500 units should 
be given. Antitoxin has been administered orally, but the results are 
860 SERUM PROPHYLAXIS OF DIPHTHERIA 
861 
irregular and the degree of immunization inconstant; for these reasons the 
serum should be given by subcutaneous injection. 
Duration of the Immunity.—The duration of this passive immunity is 
relatively short, owing to the fact that the antitoxin is eliminated rapidly, 
as it is part and parcel of a foreign serum that tends to be excreted 
or destroyed soon after its introduction into the body. It will endure, 
however, for at least two weeks, and frequently two to four weeks longer. 
Since the incubation period of diphtheria is only a matter of a few days, 
this suffices, in the majority of instances, to protect the individual. But if 
the danger of infection lasts for more than six weeks a second injection may 
be required. 
Indications.—The indications are to immunize all persons who have come 
in intimate contact with a case of diphtheria. If time permits the Schick 
test should he conducted, as only those persons yielding positive reactions require 
the antitoxin (see p. 739). It is especially valuable in families and small 
communities, such as go to make up hospital wards and asylums. The 
physician who is attending a case of diphtheria in a private home should 
urge immunization upon all members of the household who have been ex- 
posed to infection. 
The Schick Test in Relation to Passive Immunization Against Diph- 
theria.—As stated above only those persons showing a positive Schick 
reaction require immunization when exposed to diphtheria because of sus- 
ceptibility to the disease. Children over two years and adults who have 
yielded negative Schick tests properly conducted do not require immunization 
because they are immune to the disease probably for life. For this reason 
the Schick test is particularly valuable to physicians, nurses, and others 
especially likely to be exposed to diphtheria; if known to be Schick negative 
the administration of antitoxin is not required. 
When time permits the Schick test may be conducted before antitoxin 
is injected on a large scale to adults, as inmates of a ward or institution; this 
delays the injection of antitoxin, but in my opinion is justifiable as the 
majority of adults are immune and will yield negative reactions. With 
children, however, from two to twelve years of age this delay may not be 
advisable if there is reason to believe that they have been intimately exposed, 
because the majority are susceptible to the disease and yield positive Schick 
reactions. 
Results.—The immediate results are usually good. The main disadvan- 
tage is the short duration of the immunity, so that no matter how faithfully 
it is carried out, persons do not remain immune for long periods of time, 
and accordingly the total morbidity of the disease is not influenced to any 
extent. In homes from which the case of diphtheria is promptly removed 
to a special contagious hospital and in which the remaining members are 
promptly immunized the percentage of secondary cases is practically nil. 
Of 6772 patients who were removed to the Philadelphia Hospital for Con- 
tagious Diseases, the remaining members of the family not being immunized, 
secondary cases developed in 164 persons, or in 2.4 per cent. Of 4063 cases 
of diphtheria treated at home with antitoxin, the other members of the 
family not being immunized, secondary cases developed in 219 persons, or 
in 5.3 per cent. Of 639 diphtheric patients treated at home who did not 
receive antitoxin and where immunization was not practised, secondary 
cases developed in 151 persons, or 23.6 per cent. These figures, compiled 
by Dr. A. A. Cairns, chief medical inspector of Philadelphia, and taken 
from the annual reports of the Philadelphia Bureau of Health for the years 
of 1909, 1910, and 1911, show that the best results are obtained when the 862 
PROPHYLACTIC SERUM OR PASSIVE IMMUNIZATION 
diphtheric patient is promptly removed to a special hospital and the remain- 
ing members of the household are immunized. Similar results have been 
reported by Braun,1 who found that only 1.62 percent, of a group of 2218 
children exposed to diphtheria and immunized with serum developed the 
disease, whereas nearly 35 per cent, developed diphtheria in families not 
immunized. Even when the patient is removed promptly there is some 
danger of other persons having been infected, and immunization should, 
therefore, always be promptly practised. When the patient is treated at 
home, other members of the household, even if immunized, are liable to 
develop the infection, probably owing to the fact that the patient harbors 
virulent bacilli for varying periods of time after the passive immunity in 
other persons has passed away and the quarantine is broken. Certainly 
in those homes where antitoxin is not used either for therapeutic or for 
prophylactic purposes, the percentage of secondary infections is so high as 
to leave no doubt as to the value of antitoxin. 
In this connection I may mention the desirability of using an antitoxin 
prepared by immunization of cattle for the general purpose of prophylaxis, 
and especially for the treatment of those persons who are hypersensitive to 
horse serum. In these cases horse antitoxin could be used later if a person 
contracted diphtheria without danger of anaphylaxis. 
VON BEHRING METHOD OF TOXIN-ANTITOXIN VACCINATION AGAINST 
DIPHTHERIA (T-A IMMUNIZATION) 
Historical.—Owing to the fact that the antibodies produced through the 
activities of our own body cells (active immunization) persist for longer 
periods of time than those that are introduced passively (passive immuniza- 
tion), von Behring and his assistants have developed a method of active 
immunization in diphtheria whereby our own body cells are stimulated to pro- 
duce our own antitoxin in sufficient amounts to protect us against the disease. 
It has long been known that more or less balanced mixtures of this kind 
produce immunity in animals. Babes in 1895 was the first to inject experi- 
mentally diphtheria toxin-antitoxin mixtures and to appreciate that not 
only slightly toxic but also that slightly overneutralized mixtures would 
engender the production of antitoxin in animals. At about the same time 
and independently, Park of New York made the same observation and 
began to use this knowledge in the immunization of horses for the production 
of antitoxin. Park noted that the use of large amounts of overneutralized 
toxin was a safe and rapid method of beginning immunization of horses; 
that is, the antitoxin protected against fatal reactions, but yet active immu- 
nization occurred because small amounts of toxin became gradually disso- 
ciated and served to stimulate the body cells to produce antitoxin. Exactly 
the same mechanism is operative in the immunization of human beings with 
these T-A mixtures. 
In 1907 Theobald Smith2 suggested that it might be possible to employ 
this method for the purpose of producing immunity in man. Subsequently 
Smith3 studied the effects of injections of neutral mixtures in guinea-pigs 
and horses, and again pointed out the applicability of the method to human 
beings. Finally this was accomplished by von Behring4 in 1913, by demon- 
strating in a few human beings the safety of the injections. Shortly after- 
1 Deutsch. med. Wchn., 1913, 40, 1145. 
2 Jour. Med. Research, 1907, xvi, 359. 
3 Jour. Exper. Med., 1909, xi, 241; Jour. Med. Research, 1910, xxiii, 433. 
4 Deutsch. med. Wchn., 1913, 39, 873 TOXIN-ANTITOXIN VACCINATION AGAINST DIPHTHERIA 
863 
ward Hahn and Summer1 injected 1097 children in the district of Magde- 
burg, where diphtheria was endemic. Of these, 633 received the full series 
of three injections, 255 received two injections, and 209 received one. The 
Schick test was not employed. For the first two weeks there was no differ- 
ence in the number of cases of diphtheria in the vaccinated and unvaccinated, 
but after that time there was a lessening of the number in the vaccinated 
group. In 1920 Bieber was able to study the histories of the majority of 
these individuals and reported that during the six years, 15 per cent, of the 
unvaccinated group contracted diphtheria, whereas only 4.6 per cent, of 
the vaccinated contracted the disease. 
Mechanism of Toxin-antitoxin Immunization.—Active immunization in 
diphtheria could probably be accomplished by the administration of minute 
and increasing doses of toxin, but there would be some danger of producing 
local necroses, an acute toxemia or paralysis, and the process may require 
so much time as to be useless in the presence of epidemics. 
von Behring’s method, according to his report read before the German 
Convention on International Medicine in 1913, is based upon the principle 
that the Union of toxin and antitoxin is not stable, and when a neutral 
mixture of the two is injected into animals, sufficient toxin becomes disso- 
ciated to unite with body cells and stimulate the production of antitoxin. 
The amount of antitoxin injected does not confer an appreciable immu- 
nity; therefore this method of immunization cannot be regarded as a com- 
bined active and passive immunization, but is purely a process of vaccina- 
tion or active immunization by means of the toxin stimulating our body 
cells to produce antitoxin. For convenience only the subject is here con- 
sidered instead of in the chapter on Active Immunization. 
Preparation of the Toxin-antitoxin.—Formerly immunization was con- 
ducted with mixtures of toxin-antitoxin which were slightly overneutralized 
or just neutral; at the present time mixtures slightly toxic for the guinea- 
pig are being employed. According to Zingher2 these may be prepared as 
follows: 
“Only a well-ripened toxin with an L+ dose of 0.4 to 0.2 c.c. should be 
employed. Concentrated antitoxin of carefully determined strength is 
added in such proportions that 85 per cent, of each L+ dose of toxin is 
neutralized by one unit of antitoxin. The actual amount of antitoxin for 
each L+ dose of toxin is 1.17 units. For example, a toxin with an L+ 
dose of 0.4 c.c. contains 2.5 L+ doses per cubic centimeter and will require 
2.5 X 1.17 or 2.9 units of antitoxin to each cubic centimeter or 2900 units 
for the liter of toxin.” 
The mixture is thoroughly shaken, allowed to stand for two hours, and 
sterility and toxicity tests conducted. For the latter purpose a guinea-pig 
is injected subcutaneously with 1 c.c. of the mixture and should develop 
only a slight induration; a second pig is injected with 5 c.c. and should show 
a moderate or marked local edema and late paralysis, but should not die 
of acute poisoning. If the animal dies before the fifth day, it indicates an 
excess of free toxin in the mixture which has to be carefully neutralized. 
The mixtures should be kept cold in a refrigerator and are effective for at 
least three months. As the mixture stands the toxin tends to deteriorate 
faster than the antitoxin and for this reason perfectly fresh mixtures are 
better. This, however, is not practical and not necessary if well ripened 
toxin and antitoxin are employed. 
The mixtures as described contain from 2 to 3 L+ doses of toxin in each 
1Deutsch. med. Wchn., 1914, 40, 13. 
2 Jour. Infect. Dis., 1917, 21, 493. 864 
PROPHYLACTIC SERUM OR PASSIVE IMMUNIZATION 
dose, properly neutralized. In older children and adults they sometimes 
produce severe reactions, and recently Park has been using in each dose 
only 1/10 L+ dose of toxin, properly neutralized. Preliminary results 
indicate that the antitoxin production is just about as good, and that these 
severe reactions are avoided. 
Dose and Method of Administration.—The dose is 1 c.c. injected sub- 
cutaneously once a week until three doses have been given. For children 
under one year of age three injections of 0.5 c.c. each are given. The in- 
jections may be given in the arm at the insertion of the deltoid muscle or 
under the skin of the abdomen. Zingher has found that two injections of 
larger amounts (1.5 c.c.) do not give as good results as three injections of a 
smaller amount. As previously stated, the mixture should be underneu- 
tralized, but is perfectly safe for the human being. In about 5 to 8 per cent, 
of Schick positive individuals a fourth and even a fifth dose may be required 
to render them immune (Schick negative). 
Age in Relation to Immunization.—Apparently the most favorable age 
is from six months to two years. According to Park1 infants under six 
months do not usually respond with the production of antitoxin because 
of the combined effect of immature cells and the overneutralization of the 
toxin-antitoxin mixture (due to the presence of natural antitoxin in the 
blood); the injections, however, produce no ill effects. 
As previously stated, it is scarcely worth while conducting a preliminary 
Schick test with children six months to two years of age because if a negative 
reaction occurs this result may be temporary and misleading and the child 
lose its immunity later on as the natural antitoxin disappears. Toxin- 
antitoxin may be given to all, even though it means that a few children may 
not actually be in need of immunization. With older children and adults 
the Schick test should be done first and only the positive reactors selected 
for immunization. 
Children bear the injections better than adults, but immunization may 
be safely practised at any age; the best time, however, is during early child- 
hood from six months to six years of age. 
Reactions.—Adults and older children may develop local reactions of 
slight edema, erythema, and tenderness regarded as anaphylactic reactions 
to the dissolved proteins of the diphtheria bacilli in the toxin. Children 
under four years commonly do not show any reaction. Persons who have 
shown a combined true and pseudo-Schick reaction are especially likely to 
develop a local reaction, the latter indicating a hypersensitiveness to the 
proteins in the toxin. Occasionally a local reaction is accompanied by slight 
fever and malaise. I have never observed serum sickness, probably because 
the amount of concentrated antitoxin actually injected is too small to elicit 
this general reaction. According to Park in children of school age about 
10 per cent, develop fairly sore arms and temperatures of from 99° to 103° F. 
About 5 per cent, feel miserable enough to stay at home from school for one 
day, and a very few for two days. In adults the reactions are apt to be 
more frequent and severe, but not quite as marked as follows injections of 
typhoid-paratyphoid vaccine. In infants the reactions are practically 
negligible. 
Probably some degree of anaphylactic sensitization to horse-serum pro- 
teins is engendered by the injection of T-A, but the degree of sensitization 
is very slight and does not constitute a contraindication to the administra- 
tion of horse antiserum (like tetanus antitoxin, antimeningococcus serum, 
etc.) at a later time in life. 
1 Jour. Amer. Med. Assoc., 1922, 79, 1584. TOXIN-ANTITOXIN VACCINATION AGAINST DIPHTHERIA 
865 
The Development of Immunity.—After each injection a portion of the 
toxin becomes dissociated from the antitoxin and serves to stimulate the 
body cells with the production of antitoxin, but the processs is slow and 
generally three or six months are required after the last dose for the produc- 
tion of sufficient antitoxin to render the patient Schick negative. 
According to Zingher one injection of T-A will immunize about 60 per 
cent, of Schick positive reactors; two injections immunizes about 80 per 
cent., and three injections about 90 to 95 per cent. By reinjecting those who 
still give a well-marked positive Schick reaction at the end of three months, 
with two or three more doses of T-A, an active immunity may be developed 
in almost all susceptible persons. If the Schick test is only faintly positive 
one additional injection of T-A will probably suffice. 
Test for Immunity.—It is possible to actually measure the amount of 
antitoxin produced in the blood as described in Chapter XIII; recently 
Kellog1 has described a new test for this purpose which is described in the 
same chapter. For practical purposes, however, the Schick test is com- 
monly employed and when properly conducted and found to yield a negative 
reaction, is a safe index of a sufficient degree of immunity to diphtheria. 
As stated above, the test should be conducted three to six months after the 
last injection of T-A. 
Duration of Immunity.—According to Park and Zingher, who have had 
an extensive experience in New York, the immunity after three doses of T-A, 
prepared as described above, apparently lasts for six years and the indications 
are that it may persist for a lifetime. A child successfully immunized at 
four years of age may expect protection at least until twelve to fifteen years 
of age, at which time natural immunity will probably have developed to such 
degree as to afford protection for the balance of life. Schroeder2 found in 
one series of children that the immunity persisted for five years, and in a 
second group of 570 children immunity had persisted for at least two years 
in 90 to 95 per cent. Meyer3 has observed that immunity has persisted in 
94.4 per cent, of a group of children for at least forty-four months. It is 
entirely likely that retesting (Schick) will show that the majority of these 
children are still immune and that they will continue to yield negative 
reactions. 
Dangers and Contraindications.—When the T-A mixtures are properly 
prepared extensive experience has shown that the method is without danger. 
Zingher reports that many thousands of injections have been given with the 
toxin-antitoxin preparation of the New York Research Laboratory of the 
Bureau of Health without a single untoward result. Deplorable accidents 
have occurred due to the injection of mixtures too highly toxic for the 
guinea-pig and these emphasize the necessity for extreme care in preparing 
and testing the mixtures. According to Zingher mixtures properly pre- 
pared always remain safe, that is, do not acquire increased toxicity and 
remain effective for immunization for at least six months. 
Von Behring advises against the immunization of atrophic infants and 
individuals suffering with acute active tuberculosis. The fear expressed by 
some that immunization is contraindicated in those persons harboring 
diphtheria bacilli or in the presence of an epidemic, by reason of producing 
a negative phase of temporary depression of resistance, has been finally 
dissipated by practical experience. Not one of numerous carriers injected 
with T-A has developed diphtheria. Whether or not active immunization 
1 Jour. Amer. Med. Assoc., 1922, 78, 1782. 
2 Arch. Pediat., 1921, 38, 368. 
3 Jour. Amer. Med. Assoc., 1922, 78, 716. 866 
PROPHYLACTIC SERUM OR PASSIVE IMMUNIZATION 
with T-A aids in getting rid of the bacilli is still an open question, but very 
probably does not have this effect. 
Practical Value.—Abroad von Behring and his assistants1 and in this 
country Park and Zingher2 have employed this method of immunization 
against diphtheria on a broad scale and have amply demonstrated the safety 
and practical value of the procedure. Blum,3 Meyer,4 Lilly,5 White,6 Mul- 
sow,7 and others have reported favorably. My own experience has likewise 
been most favorable in that I have observed no accidents or untoward 
results aside from a large local reaction in an occasional adult. In about 
88 per cent, of Schick positive children three injections have engendered 
sufficient antitoxin in my experience to yield negative Schick reactions in 
tests conducted about four months after the last dose. 
Park and Zingher have observed that the immunity produced by three 
injections of T-A prepared by them has persisted for at least seven years, 
and the indications are that it may persist in the majority of individuals 
for life. This opinion is based upon the results of Schick tests, but so far 
there is a very large amount of evidence to show that a negative Schick 
reaction properly conducted is a sure sign of sufficient antitoxin immunity to 
protect against diphtheria. 
The results, therefore, are better than obtained in typhoid immunization 
and approach in perfection the immunity engendered against smallpox by 
vaccination. Park has recently reported the following results in New York 
City: 
(a) Among 57,000 originally Schick-negative children, 5 cases of diph- 
theria were reported. It is possible that some of these were not true clinical 
diphtheria, but cases of tonsillitis showing carrier bacilli in the cultures. 
ib) Among 33,000 Schick-positive children who received one to three 
injections of toxin-antitoxin, 9 cases of diphtheria were reported. 
(c) Among a total of 90,000 Schick-negative or immunized children, 
14 cases of diphtheria were reported, whereas among a total of 90,000 un- 
treated control children, 56 cases of diphtheria were reported. That is, 
about jour times as many cases of diphtheria developed among the unvaccinated 
children. 
Diphtheria still remains an active and widely prevalent disease through- 
out the world even though the mortality has been reduced from 70 to 75 
per cent, to 10 per cent, since .the advent of antitoxin in 1894. For the 
United States it is calculated that the yearly incidence of the disease is from 
150,000 to 200,000 cases per year and the deaths 20,000 to 22,000 per year. 
Despite vigorous public health measures and the prophylactic use of anti- 
toxin, the disease demands greater efforts toward its eradication, and for 
this purpose the wide-spread use of active immunization with T-A mixtures, 
is proving most valuable. 
Recommendations.—1. In the presence of exposure to diphtheria when 
rapid protection is required antitoxin should be injected; the immunity en- 
gendered by T-A mixtures develops too slowly for this purpose. It is to 
be remembered, however, that antitoxin protects for only a few weeks. 
1 See Semaine medicale, 1913, xxxiii, No. 18; Berl. klin. Wochenschr., 1914, lix, 917; 
ibid., 1914, lix, 917; Therap. Monatsch., 1913, xxvii, No. 11; Deutsch. med. Wchnschr., 1913, 
xxx, 460; ibid., 1913, xxix, 977; ibid., 1913, xxxix, 1977; ibid., 1913, xxxix, 2500; ibid., 1914, 
xl, 13; ibid., 1914, xl, 582. 
2 Jour. Amer. Med. Assoc., 1914, 63, 859; ibid., 1915, 65, 2216. 
3 Amer. Jour. Dis. Child., 1920, 20, 22. 
4 Jour. Amer. Med. Assoc., 1922, 78, 716. 
s Boston Med. and Surg. Jour., 1920, clxxxii, 110. 
8 Ibid., 1921, 184, 246. 7 Jour. Amer. Med. Assoc., 1921, 77, 1254. SERUM PROPHYLAXIS OF TETANUS 
867 
2. Physicians, nurses, and others especially likely to be exposed to diph- 
theria should have the Schick test conducted and if positive should be 
immunized with T-A mixtures. This is especially advisable for nurses who 
may thereby avoid the disagreeable effects of one or more injections of 
antitoxin during their training and in subsequent practice. 
3. Since diphtheria is pre-eminently a disease of children, it may be said 
that the Schick test and T-A immunization have proved so reliable when 
properly administered that all children known to be susceptible to diphtheria 
should be immunized. This is particularly true of children in institutions, 
as hospitals and asylums, where close contact increases the incidence, but 
even in private practice physicians should acquaint parents with the advan- 
tages of these procedures and advise their use. 
Zingher1 has demonstrated in the schools of New York City that it is 
possible to conduct the Schick test and apply T-A immunization on a large 
scale, and it is very much hoped that other cities and states adopt similar 
procedures for the solution of the diphtheria problem. 
4. The Schick test is a great aid, but may be dispensed with under cer- 
tain conditions as stated by Zingher2: 
(a) Infants under six months do not require the Schick test or active 
immunization, inasmuch as 80 to 90 per cent, are immune to diphtheria 
(presumably inherited from the mother). Furthermore, Zingher has shown 
that they do not produce antitoxin very well even when immunized. 
(.b) All children six months to two years of age should be immunized, as 
this is the most important period. The Schick test may be omitted because 
it is not reliable at this age, inasmuch as a child reacting negatively may 
lose its natural immunity and later react positively. 
(c) Children two to five years of age may first be tested by the Schick test 
because after two years a negative reaction is reliable, inasmuch as it remains 
constant. But since the majority of children react positively the Schick 
test may be omitted as a routine procedure and active immunization applied 
to all. 
(d) Children six to fifteen years and adults should first have a Schick test; 
T-A immunization is only required when the reactions are positive. 
(g) Immunity to diphtheria after T-A immunization can never be 
assumed to have been produced; the only reliable evidence of this is actual 
demonstration of antitoxin in the blood and for this purpose*the Schick test 
should be conducted three to six months after the last injection. 
Immunity in Tetanus.—Tetanus occurs spontaneously in man, horses, 
cattle, sheep, and rarely in dogs and goats. Birds and chickens are highly 
resistant. 
It is impossible to state whether or not human beings possess natural 
immunity analogous to the natural antitoxic immunity possessed by many 
individuals to diphtheria. Since infants may suffer from tetanus infections 
of the umbilical stump, children of all ages from vaccinal and wound tetanus, 
and adults of both sexes and all races from wound tetanus, it is highly 
probable that natural antitoxic immunity to tetanus is very uncommon or 
does not exist to sufficient degree to confer immunity. Studies of the blood 
of human beings for tetanus antitoxin have generally failed to detect its 
presence; Lowry3 examined the sera of 5 normal persons, with negative 
SERUM PROPHYLAXIS OF TETANUS 
1 Jour. Amer. Med. Assoc., 1921, 77, 835. 
2 Jour. Lab. and Clin. Med., 1920, 6, 117; Amer. Jour. Dis. Child., 1918, 16, 83 
3 Wien. klin. Wchn., 1915, 28, 1273. 868 
PROPHYLACTIC SERUM OR PASSIVE IMMUNIZATION 
results. Escape from tetanus infection is to be ascribed more to fortuitous 
circumstances, preventing the growth of the bacilli and production of toxins, 
than to the presence of natural immunity principles aside from the degree 
of protection afforded by phagocytosis of bacilli in the tissues. Necrotic 
tissue products tend to decrease the degree of phagocytosis as well as favoring 
the production of tetanus toxins. 
The mortality of the disease is very high. Our body cells react slowly to 
stimulation by tetanus toxins with the production of antitoxin. For this 
reason the bacilli and spores may live for several weeks in wounds, the 
individual being protected in the meantime by horse-serum antitoxin, only 
to produce tetanus when the heterologous antitoxin has been eliminated. 
I have not been able to find data bearing upon the question of duration of 
immunity after recovery. It is highly probable that the immunity is of 
short duration and the individual subject to reinfection, but records of 
two attacks at different times in the same person have not been found. 
Both Lowry and Wintz1 have found, however, that human beings produce 
antitoxin during tetanus infection. 
Natural resistance to tetanus is probably almost entirely cellularphago- 
cytic. Recovery from tetanus is probably largely by means of the neu- 
tralization of the free and to some extent of the combined toxin, by antitoxin 
aided by the phagocytic destruction of the bacilli in the infected tissues. 
Dangerous Wounds from the Standpoint of Tetanus.—One of the greatest 
dangers from this terrible infection lies in the fact that while the local lesion 
may show no signs of disturbance, the central nervous system may suddenly 
manifest symptoms of poisoning. The wounds that are likely to contain the 
tetanus bacillus are the following: All wounds that may contain dirt con- 
taminated by manure, such as that from the streets, stables, barns, and 
even fields; wounds made by firecrackers or toy pistols; gunshot wounds, 
especially those made by blank cartridges; crushing injuries, made by ma- 
chinery or in other ways. The feet and hands are especially prone to be 
infected with tetanus germs. Crushing injuries of bones are especially 
dangerous and, as shown during the late war, frost bites with phylctenules, 
ulcerations, and gangrene produce many cases of tetanus. Stone2 has recently 
reported that of 49 cases of tetanus, 46.6 per cent, followed injuries of the 
feet and legs and 24.3 per cent, followed injury of the upper extremity. 
Street injuries that are not deep or perforating, but grinding and lacerating, 
are very likely to develop tetanus infection It has also been stated that 
tetanus bacilli may be harbored in an old injury, and yet cause no symptoms 
until some additional injury or general disturbance of the body causes the 
normal protection against infection to be broken down, when toxins from 
the bacilli may be absorbed and tetanic symptoms develop. This theory 
would seem to be responsible for an otherwise apparently unaccountable 
development of tetanus. 
From what has been said it will be seen that any injuries received on 
the street, or those inflicted on workers about horses or cattle and in stables, 
are more likely to develop tetanus than are injuries received in other ways. 
Newborn babies may be infected through the stump of the umbilical cord. 
Likewise a suppurating wound, or even a fresh wound, which may be innocent 
at first, may become infected with the tetanus bacillus if the wound or 
suppurating focus is improperly cared for. Many cases of vaccinal tetanus 
can thus be accounted for, i. e., due to negligence in the care and treatment 
of the wound. It is now generally agreed that proper treatment of the original 
1 Miinch. med. Wchn., 1915, 62, 1561. 
2 Jour. Amer. Med. Assoc., 1922, 78, 1939. wound, combined with the administration of tetanus antitoxin, will surely 
prevent the development of lockjaw. 
Value of Antitoxin Prophylaxis.—In former years Fourth of July wounds 
claimed a heavy toll of fatalities due to tetanus. Owing to the efforts of 
the American Medical Association municipalities have been urged to adopt 
legislative measures for enforcing a saner form of celebration, and efforts 
have been made to educate physicians in the proper care of these wounds 
and to impress upon them the great prophylactic value of tetanus antitoxin. 
These efforts have been crowned with success, as statistics collected from 
all parts of the country will show. In 1903 there were in the United States 
406 deaths from tetanus; in 1904, 91; in 1905, 87; in 1906, 75; in 1907, 73; 
in 1908, 76; in 1911, 18 cases and 10 deaths; in 1912, 7 cases with 6 deaths, 
and in 1913, 4 cases with 3 deaths. 
Experience during the recent war has furnished ample new evidence of 
the great value of serum prophylaxis. Since tetanus bacilli and spores are 
to be found in dirt and earth, the earth-digging methods of warfare practised 
during the early period of this war were responsible for a high incidence of 
tetanus, and especially in wounds grossly contaminated with dirt and foreign 
bodies and in wounds involving bones. During the first years the sudden 
demand for antitoxin was met by the preparation of sera below the average 
in potency, so that disappointments in its protective value were encoum 
tered, but with this correction the evidence of army surgeons, as Tizzoni,1 
MacConkey,2 Bruce,3 Bazy,4 Leischman,5 Vaillard,6 and others,7 are over- 
whelmingly in favor of the high prophylactic value of the serum given 
subcutaneously or intramuscularly. 
In the American Army the prophylactic dose of serum was 1500 units 
by subcutaneous injection as soon as possible after injury. This was re- 
peated ten days later and again when subsequent operation was done. Of 
224,089 injured men only 36 developed tetanus or one to each 6224 wounds 
(Stone). This is in marked contrast to the incidence of tetanus in the Civil 
War when, according to Sanford,8 the incidence was one case of tetanus to 
each 487 wounds. Bazy9 observed 200 wounded men in the same sector 
in France; 100 received antitoxin and none developed tetanus, while of 
100 who were not immunized, 18 cases developed. 
In the British Army, Bruce10 has stated that the incidence of tetanus 
among the immunized never rose above 2 or 3 per thousand; before the 
routine use of antitoxin the incidence was from 15 to 32 per thousand. 
The British statistics also show that of the wounded immunized with 
500 units of antitoxin, the average period of incubation among those de- 
veloping tetanus was 45.5 days with a mortality of 22.5 per cent.; among 
those not immunized the average incubation period was 10.9 days and the 
mortality was 53.3 per cent. These figures indicate therefore, that prophy- 
lactic immunization not only lengthened the period of incubation, but greatly 
reduced the mortality from tetanus even when the disease developed despite the 
injection of antitoxin. 
SERUM PROPHYLAXIS OF TETANUS 
869 
1 Gazz. d. osp. e d. clin., 1914, 35, No. 56. 
2 Brit. Med. Jour., 1915, 2, 849. 
3 Brit. Med. Jour., 1917, 1, 118. 
4 Lancet, 1918, 2, 523. 
5 Lancet, 1917, 1, 131. 
6 Bull. d. l’Acad. d. Med., 1916, 76, 167. 
7 Amer. Jour. Med. Sci., 1919, clvii, 764. 
8 Bull. Internat. A. M. Museum, 1918, 7, 365. 
9 Bull, et mem. Soc. de Chir., 1916, 42, 2919. 
10 Brit. Med. Jour., 1917, 1, 118. 870 
PROPHYLACTIC SERUM OR PASSIVE IMMUNIZATION 
While it is not within the province of this chapter to deal with surgical 
technic, the proper cleansing and care of a wound constitute so important 
a part of the prophylaxis of tetanus that I shall refer to this subject, quoting 
largely from the technic described by Ashhurst and John.1 
Prophylactic Surgical Treatment of Wounds.—1. The surrounding skin 
should be painted with a 3 per cent, alcoholic solution of iodin; for simple 
wounds this will ordinarily suffice. 
2. All foreign material should be removed from the wound, and to do 
this properly all parts of the wound should be made accessible by wide 
incision, under ether, if necessary. This is especially true of a puncture 
wound. It should then be freed from all tags and loose shreds of tissue by 
means of the scissors, and the whole wound swabbed with the 3 per cent, 
iodin solution. The wound should next be dressed with gauze soaked in 
the same solution. The use of strong caustics is inadvisable, as they cause 
sloughing and tend to produce a good focus for the growth of tetanus bacilli. 
Bruce2 has stated that in lacerated wounds primary excision of the tissues 
should be done before the wound is sutured. The cautery should not be 
used. 
3. The wound should be dressed daily at first, being exposed and thor- 
oughly irrigated with hydrogen dioxid solution, and then dressed with the 
gauze saturated with the iodin solution. As soon as healthy granulations 
have formed, balsam of Peru applications should be made. 
4. Antitetanic powders have been prepared, made up with antiseptics, 
and although Robertson3 and others have found experimentally that their 
use has seemed to be successful in preventing the development or absorption 
of tetanus toxin, still it has not as yet shown that these results were not 
merely due to the strong antiseptic that was combined with the antitoxin 
powder. It might, however, be well to apply tetanus antitoxin and anti- 
tetanic powder to the open wound, but these remedies are not to be relied upon 
nor accepted as substitutes for the injection of antitoxin. 
Serum Prophylaxis.—The most successful preventive treatment, and 
practically the only successful curative one after the disease has developed, 
is by means of tetanus antitoxin. As a prophylactic remedy this antitoxin 
exceeds in value even diphtheria antitoxin; therapeutically, however, it is 
far inferior to the latter, for the reason that part of the toxin produced by 
the tetanus bacilli soon unites with the nerve-cells of the spinal cord where 
it may not be neutralized. The prophylactic use of tetanus antitoxin, how- 
ever, has not infrequently been unsuccessful, due probably to the fact that 
it was used incorrectly. 
1. Antitoxin should be given as soon as possible after the wound has been 
inflicted, and best at the time the primary treatment is given. The antitoxin 
should be injected “as near the wound as possible, so as to flood the tissues 
in the immediate vicinity,” and, if possible, it should be given intramuscu- 
larly, so that the motor nerves may absorb it rapidly. If any nerves are 
exposed, antitoxin should be injected into them. The injection of 1500 units■ 
of antitoxin is generally advised as the first prophylactic dose. 
2. From the fact that tetanus antitoxin is one of the albuminous con- 
stituents of horse-serum that are foreign to the human system, in the human 
being the antitoxin is rapidly eliminated in from eight to ten days after the 
injection is administered. Knorr has found, as the result of animal experi- 
ments, that by the sixth day only one-third, and by the twelfth day only 
1 Amer. Jour. Med. Sci., 1913, cxlvi, No. 1. 
2 Jour. Hyg., 1920, 19, 1. 
3 Amer. Jour. Med. Sci., 1916, 151, 668. SERUM PROPHYLAXIS OF GAS GANGRENE 
871 
one-fiftieth, of the optimum quantity remained in the blood. Hence it is 
important, if antitoxin is to prove useful, that it should be present in the system 
for two or three weeks after receipt of the injury, especially as it cannot be deter- 
mined when the tetanus bacillus first gained access to the wound. There should 
be a second intramuscular injection of 1000 units of antitoxin about the end of 
the first week, and perhaps a similar dose at the end of the third week in severe 
cases. While certain individuals may develop serum sickness, no dangerous 
symptoms have been observed to result from the use of tetanus antitoxin. 
3. If the surgeon is first consulted several days after the injury has been 
inflicted, and there are reasons for suspecting that tetanus infection has 
occurred, the wound should be opened and dressed as previously described, 
and, in addition to the intramuscular injection of 1500 units of antitoxin 
in the neighborhood of the wound, it will be good practice to inject an 
additional 5000 or 10,000 units intravenously. It requires at least twenty- 
four hours for the antitoxin to be absorbed form the tissues, and immediate 
neutralization of any toxin present in the blood may mean a great deal 
from the standpoint of prognosis if tetanus should develop. 
Antidysentery sera are prepared by the immunization of horses with the 
Shiga dysentery bacillus and its toxins; the serum is largely in the nature 
of an antitoxin. Horses may also be immunized against bacilli of the 
Flexner group; the serum is largely antibacterial. A polyvalent serum may 
be prepared by mixing these, or horses may be immunized simultaneously 
with Shiga and Flexner bacilli and toxins. Antiserum for the Shiga toxins 
will not protect against infections with the Flexner bacill, and the reverse. 
For prophylactic purposes 10 c.c. of polyvalent serum may be injected 
subcutaneously. The immunity only lasts for twelve to fourteen days. 
Prophylactic immunization has its greatest value in institutional out- 
breaks of bacillary dysentery. The serum is of no value in amebic dysentery 
or in non-dysenteric ileocolitis of children. 
SERUM PROPHYLAXIS OF BACILLARY DYSENTERY 
Bacteriology of Gas Gangrene.—The bacteriology of gas gangrene was 
the subject of a great deal of investigation during the World War because 
0.5 to 3.5 per cent, of wounded developed this dreadful complication with a 
mortality of 10 to 50 per cent., depending upon how early treatment was 
given. Until these studies were made Bacillus aerogenes capsulatus (B. 
welchii, B. perfringens) was regarded as the sole cause of this infection, but 
experience has shown that pure infections of war wounds with this bacillus 
were so rare as to be almost negligible. Other spore-bearing anaerobic bacilli 
were commonly present in gas gangrene, as B. oedematis maligni (vibrion 
septique) and B. oedematiens (B. bellonensis), although frequently overgrown 
by other micro-organisms. 
Immunity in Gas Gangrene.—Immunity in gas gangrene is probably 
similar to that in tetanus. Nothing is definitely known of natural immunity 
to the various micro-organisms and their toxins identified with these infec- 
tions. In all probability the serum of normal human beings possesses little 
or no neutralizing effects upon these toxins. Phagocytosis is probably the 
chief means of defense and resistance to infection; phagocytosis is likewise, 
probably, the chief mechanism concerned in recovery and especially after 
the negative chemotactic substances (toxins and products of necrotic tissue) 
are removed by drainage. As in diphtheria and tetanus, antitoxin produc- 
SERUM PROPHYLAXIS OF GAS GANGRENE 872 
PROPHYLACTIC SERUM OR PASSIVE IMMUNIZATION 
tion against the toxins of gas gangrene by our body cells is probably tardy 
and never reaches a high degree. 
Preparation of Immune Sera.—Bull and Pritchet1 have found that 
B. aerogenes capsulatus (B. welchii, B. perfringens) produces an exotoxin in 
fluid culture-media, and that horses immunized with sterile filtrates con- 
taining the toxin produce an antitoxin affording protection to pigeons and 
guinea-pigs infected with the bacillus or its soluble toxin. The part played 
by this bacillus in gas gangrene, however, is of minor importance; some 
investigators regard it as a saprophyte, and Taylor2 and others doubt that 
the hemolytic poison produced by its growth in muscles is a true exotoxin 
for wrhich an antitoxic serum can be produced. While Bull’s antitoxin is 
undoubtedly protective and curative in experimental infections of laboratory 
animals, the serum has proved generally disappointing in the prevention and 
treatment of gas gangrene of war wounds; Gibbon3 found that it sometimes 
arrested infections, had no effect on others, and occasionally produced severe 
reactions. 
Caulfield4 has recently described in detail a method for the immunization 
of horses with toxins of B. perfringens and other gas-producing organisms; 
also methods for maintaining the virulence of the cultures, preparation of 
the toxins, etc. 
The standard adopted by the Hygienic Laboratory for B. perfringens 
antitoxin has been recently described by Ida Bengtson.5 The unit shall be 
1 c.c. of the standard serum which is kept in cold storage. To estimate the 
potency of a fresh serum, the test toxin shall first be standardized by inocu- 
lating pigeons intramuscularly with 1/100 unit of standard serum mixed 
with varying amounts of toxin to determine the smallest amount of toxin 
which will overcome this amount of serum and kill the pigeon within twenty- 
four hours. This dose of toxin, called the “test dose,” is usually somewhat 
greater than 10 minimal lethal doses. The test dose of toxin is then to be 
mixed with varying amounts of the serum to be standardized and injected 
into a second series of pigeons; that amount of serum which gives protection 
against the test dose shall be considered to contain 1/100 unit. 
Antisera have also been produced in the Pasteur Institute of Paris by 
Weinberg and Sequin,6 Leclainche and Vallee,7 and Sacquepee8 by immunizing 
horses with dead and living cultures of B. aerogenes capsulatus (B. welchii, 
B. perfringens); this serum is designated as antitoxin serum; for B. oedematis 
maligni (vibrion septique), designated as antivibrion septique or anti- 
edematous serum, and for B. oedematiens (B. bellonensis), designated as 
anticedematiens or antibellonensis serum. In an experimental study of 
these sera Nevin9 found that Bull’s antitoxin alone was ineffective in mixed 
infections and that best results followed the use of a mixture of these various 
antisera. These results appear to be in accord with clinical opinions, 
namely, that the antitoxic serum for B. welchii alone is unsatisfactory, and 
that best results in the serum prophylaxis and treatment of gas gangrene 
may be expected with a mixture of the antisera for B. welchii, B. oedematis 
maligni, and B. oedematiens or a polyvalent serum secured by immunizing 
1 Jour. Exper. Med., 1917, 26, 119. 
2 Bull. Johns Hopkins Hosp., 1916, 26, 297. 
3 Amer. Jour. Med. Sci., 1919, 157, 764. 
4 Jour. Infect. Dis., 1920, 27, 151. 
5 Hygienic Lab. Bull., No. 122. . 
6 Proc. Roy. Soc. Med., 1916, 9, 119-144. 
7 Bull, et mem. Soc. de chir. de Paris, 1916, 42, 1804-46. 
8 Presse med., 1918, 26, 197-199. 
9 Jour. Infect. Dis., 1919, 25, 178-188. SERUM PROPHYLAXIS OF MENINGITIS 
873 
horses with the three micro-organisms and their products. For example, 
Duval and Vaucher1 have reported favorably upon the prophylactic value 
of such sera, the mortality rate being reduced from 16 to 3.5 per cent.; of 
77 cases of gangrene treated with serum, 16, or 20.7 per cent., died. Mar- 
quis,2 in the treatment of 10 cases with serum, lost 2, a mortality of 20 per 
cent.; Rouvillois, Guillame, Louis, Pedeprade, and Thibierge3 treated 25 
cases, 3 of whom were moribund, with a mortality of 24 per cent. Mairesse 
and Regnier4 had 30 cases, or 10 per cent., of 297 wounded men showing 
gas bacillus infection, develop gangrene after the prophylactic adminis- 
tration of antiperfringens, antivibrion septique, or antioedematiens sera, 
depending upon the bacteriology; of 30 cases receiving curative injections 
the mortality was 16 per cent. 
Van Beuren5 states that the prophylactic use of antigas serum in the 
British and American armies yielded favorable and encouraging results, 
and both the Third and Fifth Interallied Surgical Congresses for the Study 
of War Wounds reported that antiperfringens serum seemed to have given 
favorable results as a prophylactic, and that antivibrion septique and anti- 
oedematiens sera yielded appreciable results, both from a prophylactic and 
curative standpoint. All investigators are agreed, however, that serotherapy 
is but an adjuvant to surgery, and that surgical treatment “is still indicated 
and in no way modified.” In general the lowest mortality from gas gan- 
grene treated by operation alone is about 25 per cent., and by operation 
plus serotherapy about 19 per cent. 
Serum Prophylaxis.—In the treatment of gas gangrene best results appear 
to have followed: (1) A prophylactic dose of polyvalent antigas gangrene 
serum and tetanus antitoxin; (2) earliest possible surgical treatment and 
bacteriologic examination of the wound; (3) the administration of sera, 
either single or polyvalent, according as there are one or more gas bacilli 
in the wound. 
Intravenous injections of polyvalent serum in combination with deep 
muscular injections proximal to but in the vicinity of the wound are to be 
preferred; the total amount injected for prophylaxis should be from 30 to 
100 c.c., half intravenously and half intramuscularly. The serum treatment 
of gas gangrene is described in Chapter XL. 
SERUM PROPHYLAXIS OF MENINGITIS 
Prophylactic Immunization in Meningococcus Meningitis.—In Chapter 
XXXV mention has been made of the probable value of active immunization 
against epidemic meningitis by the subcutaneous injections of three doses 
of a polyvalent meningococcus vaccine at intervals of a week. Sophian 
and others have shown experimentally that opsonins, bacteriolysins, agglu- 
tinins, and other antibodies are produced, and while meningococcus men- 
ingitis is ordinarily but mildly infectious (about 5 per cent, of secondary 
cases in homes), the method is practically devoid of danger and worthy of 
trial, especially for physicians, nurses, and members of a household who are 
exposed to the infection over a period of many weeks. 
Passive immunization by means of the subcutaneous injection of anti- 
meningitic serum was advised by Jochmann in 1906, but has not come into 
general use. The immunity resulting from the injection of from 10 to 20 c.c. 
1 Bull, et mem. Soc. de chir. de Paris, 1918, 44, 1535. 
2 Bull, et mem. Soc. de chir. de Paris, 1918, 44, 1522. 
3 Bull, et mem. Soc. de chir. de Paris, 1918, 44, 1226. 
4 Presse med., 1918, 26, 461, 462. 
6 Jour. Amer. Med. Assoc., 1919, 63, 239-242. 874 
PROPHYLACTIC SERUM OR PASSIVE IMMUNIZATION 
of serum is only temporary, and probably lasts about a month. Another 
drawback is the resulting serum sensitization, which renders subsequent 
injections of serum more likely to be followed by serum sickness. In the 
presence of an active epidemic, such as that which occurred in Texas during 
1912, immediate passive immunization of physicians, nurses, and attendants 
by the subcutaneous injection of 15 c.c. of serum, followed by active immuni- 
zation with three doses of vaccine (500, 1000, and 1000 millions) at intervals 
of a week, may be advisable. While there are no available statistics to prove 
the value of this procedure, it is, at least, a rational one, and since there is 
danger of contracting the disease, especially after prolonged contact, phys- 
icians should practice immunization during epidemics of this dreaded disease. 
In mixed passive and active immunization the serum probably affords 
immediate protection, and tides over any temporary negative phase or 
period of lowered resistance following the injections of vaccine. 
SERUM PROPHYLAXIS OF MEASLES 
Owing to the high mortality in measles of children between the ages of 
six months and five years attempts have been made to immunize against 
the disease and to reduce its severity, at least over this period of greatest 
susceptibility. In Chapter XXXV mention was made of the work of 
Hermann, who employed active immunization by means of injections of 
nasal secretions taken from children with measles. This method, however, 
may excite a severe attack, and more attention has been given the possi- 
bility of developing a successful form of passive immunization by the injec- 
tion of blood or serum taken from measles convalescent patients. 
Degkwitz1 has injected 3 c.c. or more of serum taken from cases of 
measles between the seventh and fifteenth days after the disappearance of 
fever, into 172 children exposed to the disease, and states that all of these 
escaped the disease. Richardson and Connor2 injected 7 to 25 c.c. of serum 
and reported that 6 were definitely exposed to measles and escaped infec- 
tion; one control case developed measles. Eight others were partially 
exposed and escaped infection. In 3 others serum and nasal virus were 
injected simultaneously; in 2 there was no reaction, while the third developed 
a mild atypical rash. Maggiore3 injected susceptible children with 2 c.c. of 
serum upon entrance into the hospital and double this amount the following 
day. Kutter4 reports that of 145 susceptible children injected after the 
method of Degkwitz success was complete in 107 instances. Rietschel5 
also reports favorably upon this method of immunization, but draws attention 
to the difficulty of securing supplies of serum from young children; the 
serum of adults who had had measles in childhood was found less effective, 
v. Torday6 inoculated 261 children with amounts of serum varying from 3.5 
to 22 c.c.; all had been exposed to measles and only 15 developed the disease. 
The short duration of the immunity is indicated by the fact that 3 of the 
children developed measles from seventy-two to seventy-five days later. 
Nicolle and Conseil7 also report successful immunization by this method. 
McNeal8 inoculated 16 children exposed to measles with 5 c.c. of serum 
1 Ztschr. f. Kinderh., 1920, 25, 134; ibid., 1920, 27, 171. 
2 Jour. Amer. Med. Assoc., 1919, 72, 1045. 
3 Pediatrica, Naples, 1921, 29, 873. 
4 Ztschr. f. Kinderh., 1921, 30, 90. 
5 Ztschr. f. Kinderh., 1921, 29, 127. 
6 Ztschr. f. Kinderh., 1921, 29, 148. 
7 Bull. d. 1. Soc. m6d. d. hop., 1918, 42, 336. 
8 Jour. Amer. Med. Amer., 1922, 78, 340. SERUM PROPHYLAXIS OF MUMPS 
875 
intramuscularly; of these 4 developed the disease in a mild form. Hirarshii 
and Okamoto1 inoculated 44 children with blood taken from cases between 
the time of appearance of Koplik’s spots and the height of the eruption. 
The initial dose was small, being about 0.0001 c.c. of blood. A second 
injection was given about three weeks later. These investigators were 
inclined to regard this a form of active immunization with virus in the blood 
and regard the method as tending to increase resistance without conferring 
an absolute immunity. 
Collection and Administration of Serum.—Blood should be removed 
aseptically from children with measles from the seventh to fifteenth day 
after the disappearance of fever and preferably from the seventh to ninth 
days. The children should be free of syphilis and tuberculosis. The serum 
is collected, cultured for sterility, and preserved with 0.5 per cent, phenol. 
The dose for immunization of children should be 5 to 10 c.c. by subcutaneous 
or intramuscular injection. The injections should be made as soon as 
possible after exposure; injections made late in the period of incubation may 
not prove prophylactic, although apparently mitigate the severity of the 
attack. Reactions have not been reported, and since the serum is homolo- 
gous, there is no danger of anaphylaxis. 
The duration of the immunity is apparently not over sixty to seventy 
days. 
As stated by McNeal, the method recommends itself for the prevention 
of measles between the ages of five months and six years, and especially in 
institutions in which large numbers of frail children are intimately asso- 
ciated. 
Naturally these results indicate that similar attempts in passive immuni- 
zation against scarlet fever may prove worthy of investigation. 
Immunity in Mumps.—Mumps or epidemic parotitis occurs only in 
man. The etiology is unknown. 
Infants possess some degree of natural immunity, inasmuch as the disease 
is uncommon under one year of age. After this time children are highly 
susceptible and in institutions as many as 90 per cent, may become infected. 
Adults enjoy a certain degree of immunity, but this is not absolute; in men 
the testicles may become involved and in women the ovaries and mammary 
glands. 
Both sexes and all races are susceptible, but boys are probably more 
susceptible than girls. 
One attack generally confers a lasting immunity; second attacks, how- 
ever, may occur. 
This type of parotitis is probably different from “postoperative parotitis” 
which sometimes follows injuries and operations upon the pelvic and abdomi- 
nal organs. 
Serum Prophylaxis.—Hess2 has employed intramuscular injections of 
whole blood for prophylactic immunization. The dose was 6 to 8 c.c. by 
intramuscular injection. The donors were selected from children just 
recovering from mumps and those who had had the disease from ten days 
to one or two years previously. Of 17 children inoculated none contracted 
the disease although intimately exposed, and Hess believes that these inocu- 
lations conferred some degree of immunity and aided in checking an epidemic. 
SERUM PROPHYLAXIS OF MUMPS 
1 Japan Med. World, 1921, 1, 10. 
2 Amer. Jour. Dis. Children, 1915, 10, 99. 876 
PROPHYLACTIC SERUM OR PASSVIE IMMUNIZATION 
SERUM PROPHYLAXIS OF SYPHILIS 
The immunity of syphilis and active immunization have been discussed 
in Chapter XXXV. 
Neisser and Bruck1 and their associates have injected horses, sheep, and 
monkeys with living and dead syphilitic virus, organ extracts, and the blood 
of syphilitic human beings and monkeys, without being able to produce 
immune sera possessing the slightest effect either in vitro or in vivo upon 
active virus. They cite a large number of similar attempts for some of 
which success was claimed, but all of their experiments in addition to those 
of Finger and Landsteiner2 and others yielded negative results. 
As stated in the discussion of immunity in syphilis in Chapter XXXV, 
antibodies may be found in the blood of syphilitic human beings, and Eberson3 
believes that spirocheticidal substances may be present in sufficient amounts 
to prevent syphilis in rabbits when human syphilitic serum is injected, but 
the amounts are too small in both human beings and the experimentally 
infected rabbit and monkey, to be of value in the serum prophylaxis of 
human syphilis. As mentioned in Chapter XXXV, animals injected with 
dead pallida vaccines do not produce more than very small amounts of 
antibodies which renders impossible the immunization of large animals, as 
the horse, for the preparation of immune sera; furthermore, horses are not 
susceptible to infection with living pallida, and at the present time it appears 
that nothing less than an actual syphilitic lesion with living spirochetes is 
required for antibody production. 
The Serum Prophylaxis of Hog Cholera.—While the cause of hog cholera 
has not as yet been discovered, it is a well-known fact that the virus is 
present in the blood of infected animals, and it is possible, by immunizing 
healthy hogs with gradually increasing doses of infected blood, to prepare 
a potent immune serum that will prove of value in the prophylaxis and cure 
of hog cholera. The nature of this serum is unknown. It possesses many 
of the features of an antitoxin, and for the present may be classed with these. 
Methods for the preparation of antihog cholera serum have been de- 
scribed in detail by Graham and Hummelberger,4 Haslam,5 and others. 
Production and Standardization of Hog Cholera Serum.—Healthy hogs weighing about 100 
pounds are selected, and injected subcutaneously with 40 c.c. of hog cholera serum per hun- 
dred pounds of weight. Two or three days following this protecting dose they are injected 
intravenously with 3 or 4 c.c. of sterile, defibrinated blood, obtained from an animal suffering 
from the disease; or the animals may be exposed in pens known to be infected. If the 
animals live for one month without showing evidences of toxemia, they receive another 
injection of 5 c.c. of infected blood (virus). Two or three weeks later they receive another 
injection of from 15 to 20 c.c. of infected blood; in from fifteen to twenty-one days after this 
inoculation they receive a final injection of from 4 to 5 c.c. of virus per pound of body weight. 
Animals tolerating the last injection are said to be hyperimmune, and are bled in ten days. 
In hyperimmunizing the animal some prefer to inject the virus intraperitoneally instead of 
intravenously. If this method is adopted, about double the dose of infected blood (virus) 
is required. About 5 per cent, of animals succumb during the period of immunization. 
The immunized hogs are bled aseptically from the tail by snipping off the tip, and 5 c.c. 
of blood per pound of weight is collected in sterile vessels. Each animal is bled once a week 
until three or four bleedings have been made. 
SERUM PROPHYLAXIS IN DISEASES OF THE LOWER ANIMALS 
1 Beitr. z. Path. u. Therapie d. Syph., 1911. 
2 Centralb. f. Bakteriol., Ref., 1906, 38, 107. 
3 Arch. Dermat. and Syph., 1921, 4, 490. 
4 Jour. Infect. Dis., 1916, 18, 118. 
5 Jour. Immunology, 1921, 6, 263. In about one week after the last bleeding they are hyperimmunized again by giving 
them 4 or 5 c.c. of infected blood per pound of body weight, and additional bleedings are 
made so long as the tail lasts. 
The serums secured in the several bleedings are mixed together. The third or fourth 
bleeding is said to be most potent or to contain the greatest number of antibodies. 
In testing and standardizing the immune serum, six hogs, weighing about 100 pounds each, 
are placed in an infected pen. No. 1 receives no serum and is a control; No. 2 receives 15 c.c. 
of serum; No. 3 receives 20 c.c.; No. 4 receives 30 c.c.; No. 5 receives 35 c.c.; and No. 6 receives 
40 c.c. of immune serum. The animals are allowed to remain in the pens until the control 
succumbs and the protecting dose of serum has been determined. In this manner we can 
determine just about the amount of serum required to protect 100 pounds of hog. The 
Pennsylvania Live Stock Sanitary Board has found that it does not require quite 40 c.c. of 
serum, but this amount is recommended for safety, and because the weight may not be 
judged accurately. 
SERUM PROPHYLAXIS IN DISEASES OF LOWER ANIMALS 
877 
More recently the following method has been adopted: Seven hogs 
susceptible to hog cholera and weighing not less than 45 or more than 90 
pounds each, are selected. Each animal is injected with 2 c.c. of the virus 
blood. Five of the animals are now injected with one-half of the field dose 
of serum which is usually about 20 c.c. 
The serum test is satisfactory when both of the control pigs become 
sick within four to seven days after the test has started and succumb; all 
of the serum treated pigs must remain well throughout the test, or not more 
than one of these become sick subsequent to the fourth day, but fully recover 
before the expiration of three or four weeks, when the test is finished. 
Eberson1 and Duval and Couret2 have described methods for concen- 
trating the immune serum; whole serum, however, is generally administered. 
Hog-cholera serum is used in both the prophylaxis and the therapeutic 
management of this disease. For prophylactic purposes for each 100 pounds 
of hog 40 c.c. of serum are injected. These injections are usually given 
subcutaneously and occasionally intramuscularly. In some instances active 
and passive immunization is practised by the simultaneous injection of 
2 c.c of virus, together with 40 c.c. of immune serum per 100 pounds of 
weight. Owing to the danger of spreading the disease, this method is not 
generally employed. 
The results of prophylactic immunization are excellent, and the method 
is valuable in checking epidemics. Immunity is said to last for two to four 
months after vaccination. 
Prophylaxis of Calf Cholera.—Serum has been prepared by immunizing 
horses with strains of colon, paracolon, and other bacilli belonging to these 
groups, isolated from calves dying of calf cholera. The immune serum 
should agglutinate the micro-organisms used in its production in dilutions of 
1 : 2000 to 1 : 500. 
This serum has proved of value in the prevention of calf cholera, and 
may be of benefit in the treatment, providing that it is prepared with the 
same strain or strains of micro-organisms responsible for the infection. 
The amount of serum ordinarily administered is 15 to 30 c.c. by sub- 
cutaneous injection. In infected herds it is recommended that this amount 
of serum be injected into calves within forty-eight hours of birth. The 
immunity lasts for only a few weeks and subsequent injections may be 
required. 
Serum Prophylaxis of Infectious Abortion.—A serum for the prophylaxis 
of infectious abortion of mares is prepared by the immunization of horses 
with the bacillus of equine abortion and streptococci; Good and Smith3 have 
1 Jour. Infect. Dis., 1915, 17, 339. 
2 Jour. Med. Research, 1921, 42, 503. 
3 Jour. Infect. Dis., 1916, 18, 397. 878 
PROPHYLACTIC SERUM OR PASSIVE IMMUNIZATION 
found such sera to contain antibodies, but the practical use of the serum is 
still in the experimental stage. For the immunization of mares 40 to 50 c.c. 
serum may be injected intravenously or subcutaneously; the immunity is 
probably only of a few weeks’ duration. 
A similar serum is available for the prophylaxis of infectious abortion of 
cows; it is administered in the same manner and appears to confer some 
degree of immunity of a few weeks’ duration. 
Serum Prophylaxis of Anthrax.—Serum prophylaxis has not proved very 
effective and active immunization with vaccines or the simultaneous injec- 
tion of immune serum and vaccine are generally employed. The dose of 
serum alone is 20 to 50 c.c. by subcutaneous injection; the resulting immunity 
is untrustworthy and of a week or two in duration. 
Serum Prophylaxis of Botulinus.—Graham and Schwarze1 have prepared 
an antitoxin for the prophylaxis of botulism in cattle which yielded encourag- 
ing results. Since Bacillus botulinus produces a very potent soluble toxin 
it would appear possible to prepare antitoxin, and that such serum may 
prove of value in the treatment of botulism of human beings. 
Serum Prophylaxis of Rinderpest.—The cause of rinderpest, or cattle 
plague, so fatal among European and African cattle, is still unknown. The 
virus is present in the blood and bile and is filtrable through Berkefeld and 
Chamberland filters. The disease is characterized by inflammation of the 
intestinal mucous membrane. One attack usually confers a lasting 
immunity. 
Kolle and Turner have prepared an immune serum by injecting healthy 
oxen subcutaneously with gradually increasing amounts of blood and bile 
from diseased cattle. The subcutaneous injection of 100 to 200 c.c. of 
immune serum is said to confer a passive immunity of several weeks’ dura- 
tion. 
Simultaneous injections of small amounts of virus, blood, and immune 
serum leads to a combined active and passive immunization of longer dura- 
tion and more effectiveness. 
1 Jour. Bacteriology, 1921, 6, 83. CHAPTER XXXIX 
THE PRINCIPLES OF NON-SPECIFIC PROTEIN THERAPY 
Non-specific Biologic Therapy.—Within recent years considerable 
attention has been given the subject of “protein therapy” or the use of 
various non-specific agents in the treatment of disease for their therapeutic 
power in altering the reactivity of the whole body with general tissue stimu- 
lation and activation, rather than influencing directly the cause of the 
disease as in specific therapy. 
In Chapter XXXIV brief mention was made of the non-specific activities 
of bacterial vaccines, that is, of clinical observations showing that they may 
exert beneficial and curative effects in diseases to which the bacteria incor- 
porated in the vaccine have no etiologic relationship. Renaud1 in 1911 
had noted these effects with typhoid vaccine employed in the treatment of 
various non-typhoidal infections. In 1914 Kraus and Mazza2 reported 
that the intravenous injection of colon vaccine produced practically the 
same therapeutic results in typhoid fever as the injection of typhoid vaccine, 
and Ichikawa3 observed equally good results in paratyphoid fever following 
the intravenous injection of typhoid vaccine. 
From this non-specific or heterovaccine therapy it was logical to attempt 
the intravenous injection of other substances to elicit the shock reaction; 
for example, protein split products, of bacterial and non-bacterial origin, 
milk, sera, colloidal metals, hypertonic saline solution, water, etc. Jobling, 
Petersen and Eggstein4 were among the first to study the mechanism of the 
reaction following the injection of these split proteins, and found that a 
part of the therapeutic results must be due to the mobilization of enzymes. 
At the same time Liidke5 reported upon the treatment of typhoid fever with 
the intravenous injection of albumoses, and Miller and Lusk6 upon the 
treatment of arthritis with the intravenous injection of typhoid vaccine and 
other foreign proteins. 
Immune sera have also been administered for their non-specific effects 
in the treatment of disease as diphtheria antitoxin for streptococcus infec- 
tions, tuberculosis, pneumonia and pleurisy, typhoid fever, dysentery, 
iritis, arthritis, etc. Equally good results have been observed with normal 
horse-serum and the effects are probably non-specific and due to serum 
constituents entirely apart from the antitoxin or other antibodies. 
Since these pioneer investigations were made the subject of “protein 
therapy” has attracted considerable attention with the employment of many 
different substances and the accumulation of a large literature.7 These 
non-specific agents have been used solely for the treatment of disease; to the 
best of my knowledge none have been employed for prophylactic therapy, 
1 Presse med., 1911, 19, 665. 
2 Deutsch. med. Wchn., 1914, xl, 1556. 
3 Ztschr. f. Immunitatsf., 1914-15, 23, 32. 
4 Ztschr. f. Immunitatsf., 1915-15, 24, 219; Jour. Exper. Med., 1915, 21, 239; ibid., 1915, 
22, 141 and 401. 
6 Munch, med. Wchn., 1915, lxii, 321. 
8 Jour. Amer. Med. Assoc., 1916, 66, 1756. 
7 See Protein Therapy and Non-specific Resistance by Petersen, The MacMillan Com- 
pany, New York, 1922. 
879 880 
THE PRINCIPLES OF NON-SPECIFIC PROTEIN THERAPY 
although the immediate non-specific reaction following the administration 
of some probably aids in raising resistance to infection for a brief time. 
Non-specific Agents.—The non-specific agents mostly employed for the 
treatment of disease have been typhoid and colon vaccines; normal and 
immune horse sera, cow’s milk, and peptone. These, however, by no means 
include all. The agents of autoserum therapy, as human blood, serum 
and plasma, pleural exudates, leukocytic and tissue extracts, the products 
of sterile abscesses, altered serum proteins, etc., are included, briefly sum- 
marized as follows: 
1. Vaccines: Stock vaccines of typhoid, colon, pyocyaneus bacilli, etc 
Autogenous vaccines. 
2. Bacterial extracts and yeasts. 
3. Heterologous sera: Normal and immune horse-serum; chicken-serum, 
etc. 
4. Homologous sera and blood: Normal and convalescent human blood 
and serum; plasma. 
5. Homologous exudates, transudates, and secretions: Human blister serum 
and pleural fluid; cerebrospinal fluid; human milk; pus; products of tissue 
excitation and destruction as produced by the cautery, rebefacients, tur- 
pentine, hypertonic saline solution, Roentgen rays, radium, etc. 
6. Foreign proteins and protein-split products: Milk; casein; egg albumen; 
gelatin; nucleoproteins; proteoses; deutero-albumose; peptone; histamin, 
etc. 
7. Tissue extracts and enzymes: Extracts of leukocytes; tumor autolysates; 
trypsin. 
8. Altered homologous blood proteins: Products of hemolysis caused by 
the intravenous injection of water and hypotonic saline solution; altered 
proteins due to the injection of formalin, colloidal metals, etc. 
That human blood and serum, blister fluid, pleural and other exudates 
and pus, the products of tissue necrosis as produced in abscesses, by the 
cautery, etc., should be included as “foreign proteins” for the human being 
appears unlikely and strange, yet their injection is capable of eliciting a 
response very similar to the reaction excited by the administration of truly 
foreign proteins, as bacterial vaccines, horse-serum, and cow’s milk. Petersen 
believes that a part at least of the therapeutic effects following the injection 
of turpentine, hypertonic saline, and other substances producing “fixation 
abscesses,” the application of rubefacients with vesiculation, the seton, 
cautery, etc., are due to the production and absorption of altered homologous 
proteins capable of eliciting in mild degree a non-specific reaction analogous 
to that produced by the administration of foreign proteins. 
The mere withdrawal of human blood into citrate solution or the separa- 
tion of serum by spontaneous coagulation or defibrination apparently 
results in profound changes tending to increase their toxicity and especially 
if administered at once. Undoubtedly withdrawal of the blood of an 
individual and reinjection of his serum (autoserum therapy) is sometimes 
beneficial in the treatment of some diseases, and the mild reaction that may 
be elicited appears to be quite similar to that produced by the injection of a 
foreign protein. In other words, our own body proteins, as our plasma, 
whole blood, transudates, and exudates, by very slight manipulation may be 
changed sufficiently to act as non-specific stimulants upon reinjection. 
Likewise the administration of dilute formalin solution and some of the 
colloidal metals appear to owe their effects to the production of sufficiently 
new or “foreign” products in the circulating blood to elicit non-specific 
reactions of the kind designated as “shock” or “foreign protein” reactions. THE NON-SPECIFIC GENERA L REACTION 
881 
Probably the effects of the intravenous administration of water and hypo- 
tonic saline solution are due in part to the products of hemolysis acting as 
“foreign” agents. 
Probably the most commonly employed non-specific agents are typhoid 
and colon vaccines, cow’s milk, horse-serum and various split proteins, as 
proteoses, deutero-albumose, and Witte’s peptone. In Chapter XL will 
be given the doses and administration of these and other non-specific agents 
in the treatment of disease. Here it may be stated, however, that the 
reactions elicited by these agents vary considerably in degree and that the 
method of their preparation apparently influences their effects. For example, 
a vaccine of the colon bacillus generally elicits a severer reaction than 
typhoid vaccine and the effects of both vary according to the culture em- 
ployed, method of sterilization, etc. 
Leukocytic Extracts in the Treatment of Disease.—In Chapter XIX the 
nature of these was discussed. It has been shown experimentally by Hiss 
and Zinsser,1 Petterson,2 Opie3 and Zinsser, McCoy and Chapin4 that these 
extracts possess some degree of prophylactic and curative values in experi- 
mental bacterial injections of the lower animals; Youland,5 however, was 
not able to demonstrate any marked or constant degree of antibacterial 
activity. Hiss and Zinsser, Lambert, Lloyd and Lucas, and others have 
employed these extracts in the treatment of staphylococcus, streptococcus, 
pneumococcus, and other infections of man, to which reference will be made 
in the succeeding chapter. 
The nature of the antibacterial substances in leukocytic extracts is not 
definitely known. It would appear, however, that bactericidans (leukins) 
may be present; likewise various enzymes. These substances are not 
specific for any one particular micro-organism, and their influence, therefore, 
is non-specific and properly included in the category of non-specific agents. 
Furthermore, since leukocytic extracts are generally prepared of the leuko- 
cytes of the horse or rabbit, they constitute foreign protein substances, and 
it is now believed that their therapeutic effects are largely dependent upon 
the production of the non-specific protein reaction. 
The Non-specific General Reaction.—The general reaction following the 
intravenous injection of vaccines, sera, protein split products, etc., and the 
intramuscular injection of these substances in larger doses as well as milk and 
other agents, usually elicits a general reaction which varies greatly in severity 
according to the substance administered, its dosage and route of adminis- 
tration, the physical condition of the patient, the disease, temperature, 
and other individual factors. This great variation in effects, the danger of 
eliciting too much reaction and the lack of standardized products and dosage, 
greatly complicates non-specific therapy and, indeed, may render it hazardous 
until experience has been gained. Probably most experience has been gained 
with the intravenous injection of 25,000,000 to 50,000,000 of typhoid and 
colon vaccines and the intramuscular injection of cow’s milk. 
The intravenous injection of typhoid vaccine is generally followed in 
about half an hour (six to eight hours with colon vaccine) by chilly sensa- 
tions and trembling or a frank chill. This lasts about twenty minutes to an 
hour and is treated by hot drinks, the application of hot-water bottles, etc. 
1 Jour. Med. Research, 1908, 19, 323, 399, 411, 429, 455; ibid., 1909, 20, 245; ibid., 1910, 
22, 397; ibid., 1913, 28, 385. 
2 Centralbl. f. Bakteriol., 1904, 36, 71; ibid., 1905, 39, 423, 613; ibid., 1906, 40, 537; ibid., 
1906, 42, 56. 
3 Jour. Exper. Med., 1905, 7, 316, 759; ibid., 1906, 8, 410, 536. 
4 Jour. Med. Research, 1911, 24, 483. 
5 Jour. Med. Research, 1914, 31, 367. 882 
THE PRINCIPLES OF NON-SPECIFIC PROTEIN THERAPY 
In cases of arthritis with painful and tender joints a severe chill is to be 
avoided by reason of the resulting increased pain and distress. The intra- 
muscular injection of milk is not usually followed by a chill. 
As the chill subsides a fever develops varying from 100° to 104° F. and 
reaching its maximum in about three to four hours; with intramuscular 
injections (especially of milk) fever may not appear for six to eight hours 
and persist for twenty-four to forty-eight hours. 
Apparently the production of fever depends more upon the individual 
and type of disease than upon the dosage; when the foreign agent is adminis- 
tered during a febrile period the temperature may not be increased, but 
actually show a decline. 
The pulse-rate is generally increased by 15 to 30 beats per minute, coinci- 
dent with the development of fever. In typhoid and other acute infections 
a pulse-rate of over 100 is regarded by Petersen as a contraindication to 
non-specific therapy. 
Sweating generally occurs shortly after the subsidence of the chill and 
particularly in arthritic patients who appear to derive special temporary 
benefit from profuse perspiration. Sometimes there appears to be an 
increased production of milk in lactating women. 
Frontal headache, nervous irritability, which may amount to delirium, 
nausea and vomiting, and herpes may occur and especially after the intra- 
venous injection of vaccines in the treatment of acute infections, as 
typhoid fever, pneumonia, and erysipelas accompanied by considerable 
toxemia. 
The Influence of Non-specific Agents Upon the Blood.—The most 
notable blood changes affect the leukocytes. As a general rule a leukopenia 
first appears, which is temporary, and followed in five to twelve hours by 
leukocytosis. The leukopenia has been ascribed to depression of the bone- 
marrow and accumulation of the leukocytes in the internal organs. The 
reactive leukocytosis is of a myeloid type—that is, due to stimulation of 
the bone-marrow and is largely made up of neutrophil polymorphonuclears, 
large mononuclears, and transitionals; the eosinophils are sometimes in- 
creased. 
The degree of leukopenia and reactive leukocytosis varies with different 
foreign agents. Immunized animals appear to develop a higher leukocytosis 
than normal animals, and Gay and Claypool1 thought that this leukocytosis 
was specific after the injection of typhoidin, but the phenomenon has since 
been shown to be non-specific. In human beings the leukocytic changes 
are most marked after the first injection and tend to be reduced after subse- 
quent injections. 
An increase of erythrocytes is commonly noted and especially in anemia; 
in pernicious anemia this increase is transient and uncertain. Cowie and 
Calhoun2 have reported that the platelets increase in number and size after 
the injection of typhoid vaccine. Sometimes an increased coagulability of 
the blood is found due to increase of fibrinogen and thrombokinase. 
The Influence of Non-specific Agents Upon Ferments and Antiferments.— 
Jobling, Petersen and Eggstein3 have shown that in the lower animals the 
injection of various non-specific agents is followed by a mobilization of 
proteolytic enzymes as well as of lipases and especially after severe reactions 
induced by typhoid and colon vaccines. Davis and Petersen4 have shown 
1 Arch. Int. Med., 1914, 14, 662. 
2 Arch. Int. Med., 1919, 23, 69. 
3 Jour. Exper. Med., 1915, 21, 239; ibid., 1915, 22, 141, 401, 568. 597. 
4 Jour. Exper. Med., 1917, 26, 699. NON-SPECIFIC AGENTS AND ANTIBODY PRODUCTION 
883 
similar changes in the lymph. In human beings the enzyme changes are 
not so marked; Petersen has summarized these as follows: 
(a) Concentration of the serum. 
(b) Practically no changes in the non-protein nitrogen of the blood. 
(c) A primary decrease of serum protease followed by a progressive 
increase for a period of three or four days. 
(d) An increase of serum pepidase in cases showing clinical improvement. 
(e) No changes in the serum diastase and irregular changes in the lipases. 
Jobling and Petersen have also shown that there may be an increase of 
the antiferment (antitryptic) activity of the serum after the injection of non- 
specific agents and in human beings, especially in cachexia, during acute 
febrile diseases, in pregnancy, serum disease, etc. These investigators have 
identified serum antiferment (anti-enzyme or antitrypsin) as consisting of 
finely dispersed lipoids which contain unsaturated carbon bonds in their 
chemical structure; they have been discussed in more detail in Chapter XIV. 
The Influence of Non-specific Agents Upon Antibody Production.—In 
Chapter VIII mention was made of the possibility of drugs stimulating 
body cells to the greater production of antibody for some antigenic substance, 
notably pilocarpin in relation to the production of diphtheria antitoxin. 
Of greater importance and interest in relation to the treatment of infectious 
diseases is the question of the possibility of non-specific agents stimulating 
the production of antibodies for micro-organisms and their products. 
Hektoen1 observed that rabbits sensitized to horse-serum yielded specific 
antihorse precipitin when injected with some other serum; the cells ensi-s 
tized to horse-serum with the production of precipitin were thereby stimu- 
lated by a non-specific agent to produce, or at least, cast off, additional 
amounts of specific antibody. 
Similar results have been reported by Bieling2 with rabbits immunized 
with typhoid bacilli. The subsequent injection of other non-specific sub- 
stances, as colon, dysentery, and diphtheria bacilli, resulted in the production 
of specific typhoid agglutinin. In Chapter XX, we have noted similar 
observations on the non-specific production of hemolysins. 
In typhoid fever, however, Baluit,3 Liidke,4 v. Groer,5 Svestka and 
Marek6 claim that the administration of various non-specific agents did not 
increase typhoid antibodies, although positive results were claimed by 
Flachseder7 and Parlavecchio8 and by Dolken9 in dysentery (increase of 
agglutinins following injections of milk). Culver10 also found that the 
injection of primary and secondary proteose preparations stimulate antibody 
production or mobilization for specific organisms in gonococcal arthritis 
in a manner not to be distinguished from that produced by the injection 
of the specific organisms themselves. Culver thought that when the injec- 
tion of proteose was large enough to excite a chill, that antibody production 
was engendered by the consequent massage of the infected joints, but it is 
entirely likely that the injections of non-specific proteose stimulated the sensi- 
tized antibody producing tissues with the production or casting off of specific 
1 Jour. Infect. Dis., 1917, 21, 279. 
2 Ztschr. f. Immunitatsf., 1919, 28, 246. 
3 Therap. Monatschr., 1915, 29, 307. 
4 Berl. klin. Wchn., 1920, lvii, 344. 
5 Munch. med. Wchn., 1915, lxii, 1312; Therap. Monatsch., 1916, 30, 521. 
6 Wien. klin. Wchn., 1915, 28, 1054. 
7 Wien. klin. Wchn., 1916, 29, 641. 
8 Arch. f. klin. Clin., 1909, xc, 202. 
9 Munch, med. Wchn., 1919, lxvi, 480. 
10 Jour. Amer. Med. Assoc., 1917, 68, 362. 884 
THE PRINCIPLES OF NON-SPECIFIC PROTEIN THERAPY 
gonococcal opsonins and lysins. Uddgren1 has observed that the intra- 
muscular injection of milk into Wassermann negative syphilitic patients 
frequently acted as a provocative followed by a positive Wassermann 
reaction. 
It is not improbable, therefore, that non-specific agents may stimulate 
the production or elaboration of specific antibodies provided the antibody- 
producing tissues have been previously sensitized or keyed up by the specific 
antigen and that a part of the good effects observed clinically after the 
injection of non-specific agents may be due to this antibody production. 
The evidence of antibody production by non-specific agents in normal rabbits 
or human beings not previously sensitized by vaccines or disease is very 
slight and probably does not occur in any constant or appreciable degree. 
The Influence of Non-specific Agents Upon Metabolism.—As stated by 
Petersen “the nitrogen balance shows considerable variation both experi- 
mentally and clinically following the parenteral introduction of the proteins 
and their split products.” In patients the injection of the non-specific 
agents is frequently associated with an increase in the nitrogen excretion, 
the urine showing an increase of total nitrogen of 20 to 30 per cent, above 
the excretion before the injection. Petersen states that after about two 
days the nitrogen excretion again reaches the normal, and for a variable 
period after this time there exists in many patients a diminution in excretion. 
Diuresis may follow sometimes the injection of foreign proteins, but Uddgren2 
and others have not observed albuminuria or other evidences of kidney 
irritation. 
Not infrequently patients experience a general feeling of improved well 
being after one or more injections of milk or some other non-specific agent 
and an increase of body weight; Uddgren has observed an increase of weight 
of some patients during a course of intramuscular injections of milk, and 
especially among those receiving market milk, with subsequent temperature 
reactions. 
The Focal Reaction Following the Injection of Non-specific Agents.— 
In Chapter XXXIX mention was made that the injection of foreign proteins 
(particularly typhoid and colon vaccines intravenously and milk intra- 
muscularly) may elicit a reaction of hyperemia, edema and pain about 
inflammatory foci including tuberculous lesions, analogous to the reaction 
about tuberculous foci excited by the subcutaneous injection of a sufficient 
amount of tuberculin. 
This focal reaction has been commonly accepted as a specific allergic 
phenomenon not only in tuberculosis following the injection of tuberculin, 
but in other infections as well, as, for example, reactions around furuncles 
or other lesions after the injection of an adequate amount of specific vaccine. 
As previously stated, this focal reaction is to be interpreted as both 
specific and non-specific. In my opinion the specific reaction is one of 
allergy due to allergic sensitization of the inflammatory tissues with the 
production of allergic cellular shock upon the introduction of the specific 
antigen and its union with the allergic antibody in or upon the sensitized 
(allergic) cells. The non-specific reaction generally requires the adminis- 
tration of larger amounts of agent to elicit a focal response than is required 
of the specific agent or antigen. The non-specific agent, however, is capable 
of eliciting focal reactions by means of its stimulating effect upon cellular 
activity (plasma-activation of Weichardt) which finds expression in increased 
secretory activity of gland cells, increased activity of leukocytes, etc., as 
1 Berl. klin. Wchn., 1918, lv, 354. 
2 Milchinjection in der Opthalmologie, Stockholm, 1918. THE MECHANISM OF THE NON-SPECIFIC REACTION 
885 
well as by increasing the permeability of the lymphatics and capillaries with 
hyperemia and an outpouring of lymph and plasma, enzymes, etc. Doubt- 
less the cells of inflammatory tissue about foci of infection are more sus- 
ceptible of stimulation by non-specific agents than normal cells; the con- 
stitutional reaction of fever, tachycardia, etc., is probably due to the absorp- 
tion of poisons from the foci, not only performed poisons derived from 
bacteria and disintegrated tissues, but poisonous digestion products pro- 
duced by the increased enzymic activity. Schmidt1 and Wolf-Eisner2 
have explained focal reactions on this dual basis; Petersen3 likewise accepts 
the specific as well as non-specific mechanism of the focal reaction, explaining 
the former on the basis of the liberation of disintegrating bacterial substance 
and possibly even living bacteria, and resulting in a secondary specific 
response on the part of the body. 
Whatever may be the true explanation of the focal reaction and the 
relative importance of the specific and non-specific mechanisms, the fact 
remains that the administration of non-specific agents including not only 
vaccines, milk, sera, proteoses, etc., but other agents, as Roentgen rays, 
counterirritants, trauma, etc., may elicit focal reactions. The flare-up 
of tuberculous lesions after the injection of milk, proteoses, and nucleins 
are notable examples; likewise the exacerbations of non-tuberculous in- 
flammatory foci in the eye, skin, joints, appendix, gall-bladder, respiratory 
and genito-urinary tracts, of the lancinating pains of tabes, epileptic attacks 
and the psychic symptoms of paresis, after the injection of these and other 
non-specific agents, are to be regarded as examples of non-specific focal 
stimulation as indicated on the reports of Schmidt and Kraus,4 Holler,5 
Dollken,6 Josephson,7 v. Jauregg,8 and others. 
The Mechanism of the Non-specific Reaction.—The effects and results 
of non-specific protein medication are better known and understood than 
the mechanism of their production. Weichardt,9 who has conducted a great 
deal of investigation in this field, believes that the essential mechanism 
concerned is one of cellular stimulation and that following the intravenous 
injection of the non-specific agent there is a universal stimulation of body 
cells (the omnicellular plasma-activation theory) to greater activity in the 
production of either specific antibacterial substances or merely an increase 
of general resistance to intoxication—either synthetic (the formation of 
conjugate proteins from the toxic forms) or lytic (the degradation of the 
toxic fragments to the amino-acids) or in some other way accelerating the 
elimination of the toxic substances. The theory of Dollken is similar except 
that he considers certain proteins and other agents as exciting a selective 
stimulating effect. 
This stimulation does not involve any alteration in function or call into 
play any new method of defense; it represents rather an accumulation of the 
defensive agencies normally present, and for this reason non-specific therapy 
may be contraindicated in the terminal stages of disease when the defensive 
agencies are exhausted. 
In addition to this cellular stimulation theory of Weichardt and Dollken, 
1 Deutsch. Arch. f. klin. Med., 1919, cxxi, 1. 
2 Miinch. med. Wchn., 1920, lxvii, 93. 
3 Arch. Int Med., 1918, 21, 14. 
4 Med. Klinik, 1919, 15, 503. 
5 Med. Klinik, 1917, 13, 1038. 
6 Miinch. med. Wchn., 1919, lxvi, 480; Neurol. Centralbl., 1919, 38, 354. 
7 Med. Record, 1920, xcviii, 607. 
8 Psych.-neurol. Wchn., 1918-19, 20, 132, 251. 
9 Miinch. med. Wchn., 1919, lxvi, 289; ibid., 1920, lxvii, 91. 886 
THE PRINCIPLES OF NON-SPECIFIC PROTEIN THERAPY 
which can be measured by the increased activity of the glandular 
parenchyma, increased motility of smooth muscle, etc., Starkenstein1 has 
demonstrated an increased permeability of the endothelioma of the capil- 
laries due probably to actual changes in the membrane of the cell followed 
by a decreased permeability. Due to this effect and the general cellular 
stimulation, enzymes, antibodies, fibrinogen, thrombokinase, and glycogen 
are thrown into the circulation. According to Pemberton2 there seems to 
be some relation between the glycogen metabolism and the pathologic 
changes of arthritis; that as stored glycogen is exhausted improvement 
follows and one effect of non-specific protein therapy is to hasten the cata- 
bolism of glycogen. The alteration in the cell membrane is also indicated 
by augmentation of the lymph flow and by a change in the irritability of 
the central nervous and sympathetic nervous systems. 
A combination of these two theories affords at present the most satis- 
factory explanation of the mechanism of non-specific protein therapy. The 
omnicellular plasma-activation of Weichardt explains the fever due to 
direct or indirect stimulation of the thermoregulating mechanism, the 
leukocytosis due to stimulation of the bone-marrow, the increase of serum 
and lymph ferments, increase of antibodies providing the patient has been 
previously sensitized, the increase of platelets, fibrinogen, and other coagu- 
lating substances and an increase of antiferments, all of which are to be 
regarded as defensive agencies. The increased capillary permeability theory 
of Starkenstein affords an explanation of the access of these agencies to 
foci of disease. 
The Relation of the Non-specific Reaction to Therapeutic Activity.— 
The treatment of disease by the administration of non-specific agents has 
attracted a great deal of attention during the past few years. Indeed, it 
would appear that this form of therapy is in danger of being overused and 
its importance overemphasized. One commercial firm has gone so far as 
to prepare different proteins, presumably of vegetable origin, for use in 
different diseases, apparently a specific protein for specific medication as a 
form of non-specific protein therapy. 
Undoubtedly the effects produced by the injection of some non-specific 
agents are frequently beneficial, notably the intravenous injection of typhoid 
and colon vaccines, sera and proteoses, and the intramuscular injection of 
milk. The effects produced by adequate doses including general cellular 
stimulation, improvement of elimination and metabolism, leukocytosis, 
sweating, mobilization of enzymes and antibodies, are doubtless beneficial. 
As a general rule some reaction is required before beneficial effects are to be 
expected, as chilliness, moderate fever, sweating, and leukocytosis; but the 
degree of clinical improvement is by no means in ratio to the severity of 
the reactions, at least not in all diseases. 
The focal reaction engendered about foci of infection may likewise prove 
beneficial; in my opinion, always so when of a mild character. The hypere- 
mia, leukocytosis, and enhanced phagocytosis and mobilization of ferments 
resulting from the focal reactions are capable of exerting beneficial results 
and are desirable in the treatment of localized infections. Doubtless no 
small part of the success attending the treatment of tuberculosis by tuber- 
culin and non-specific agents is due to the enhanced production of fibrous 
tissue following repeated mild focal reactions. 
However, there is distinct danger from too severe focal reactions from 
the standpoint of breaking down local defensive measures and thereby 
1 Miinch. med. Wchn., 1919, lxvi, 205. 
2 Arch. Int. Med., 1920, 25, 351. RELATION OF NON-SPECIFIC REACTION TO THERAPEUTIC ACTIVITY 
887 
facilitating local and general infection. This is particularly true in the 
treatment of tuberculosis calling for extraordinary care in the administra- 
tion of tuberculin or any other agent capable of eliciting focal reactions. 
The whole subject of non-specific protein therapy may be said to be still 
in its infancy. It is potentially a dangerous therapy by reason of the 
damage that may be caused by injudicious selection of the agent and error 
in dosage. Great caution must be exercised until more information is at 
hand regarding the relative merits and demerits of the different protein 
substances and other agents that may be used in the treatment of different 
diseases; likewise more information on the very important subject of dosage 
according to the disease and the patient as well as the route of administra- 
tion to be chosen. 
In the succeeding chapter will be given the treatment of different diseases 
by non-specific substances, as well as by vaccines and sera for specific pur- 
poses, in order to group the different forms of biologic therapy under each 
disease instead of handling the different subjects of vaccine, serum, and 
non-specific therapy in different chapters. CHAPTER XL 
VACCINES, SERA, BLOOD. AND NON-SPECIFIC PROTEINS IN 
THE TREATMENT OF DISEASE 
In this chapter it is proposed to present together, instead of separately, 
the subjects of specific vaccine and serum and non-specific protein treatment 
of disease. As previously discussed it is not always possible to differentiate 
specific from non-specific effects; indeed, it would appear that a part of the 
beneficial effects observed in vaccine and serum therapy are always of a 
general or non-specific character. The author hopes that the arrangement 
of this chapter will prove more serviceable than the plan pursued in previous 
editions of considering the subjects of serum and vaccine therapy in separate 
chapters. 
The Administration of Vaccines, Sera, and Blood.—The technic of 
subcutaneous injection of bacterial vaccines is described on p. 761; of intra- 
venous injection on p. 845. Methods for the subcutaneous, intramuscular, 
intraspinal, and intravenous injection of serum are described in Chapter 
XXXVII; also methods for obtaining blood and injecting intramuscularly. 
The subject of Blood Transfusion is considered in Chapter XLII. 
Reactions Following the Administration of Vaccines, Sera, and Non- 
specific Agents.—These have been discussed in previous chapters, and the 
physician employing these substances in the treatment of disease should 
be familiar with the reactions that may be elicited, the dangers, and con- 
traindications. The reactions that may follow the injection of serum are 
described on p. 839; skin tests for anaphylactic sensitization are described 
on p. 680 and methods for desensitizing on p. 722. 
Reactions following the injection of vaccines are described on p. 762 
and of non-specific substances on p. 881. 
Ordinary abscesses or furuncles are usually caused by Staphylococcus 
aureus; stitch abscesses or similar superficial lesions may be caused by 
Staphylococcus albus and occasionally by Staphylococcus citreus. Car- 
buncles are usually caused by Staphylococcus aureus and occasionally by 
Streptococcus pyogenes. 
When numerous abscesses occur the urine should be examined for sugar; 
not infrequently the blood-sugar is found higher than normal, even though 
sugar is not found in the urine. 
Immunity in Staphylococcus Infections.—Man and practically all of 
the lower animals are subject to staphylococcus infections. Most of the 
lower animals and notably the rat, appear, however, to possess a greater 
degree of natural immunity than man and can withstand very large injec- 
tions of virulent organisms. 
Natural immunity is largely dependent upon intact epithelial barriers 
and the secretions. Any agent producing lowered resistance of the skin, 
mucous membranes, and the cutaneous glands greatly increases suscepti- 
bility to staphylococous infections. Practically every organ and tissue is 
vulnerable to infection. 
Phagocytosis aided by opsonins constitute the chief means of defense 
and recovery when the cocci have gained access to the deeper tissues. The 
TREATMENT OF ABSCESSES (FURUNCULOSIS) AND CARBUNCLES 
888 TREATMENT OF ABSCESSES AND CARBUNCLES 
889 
blood and serum possess staphylococcidal activity, but phagocytosis is 
undoubtedly of more importance. 
Immunity to staphylococcus infections may be raised by appropriate 
immunization by means of vaccines; this immunity is principally due to the 
production of specific immune opsonins. But the duration of staphylococcus 
immunity after active immunization or after recovery from the disease 
tends to be of short duration. For this reason the writer advocates a second 
course of vaccine injections within four to six months after the first in the 
treatment of furunculosis, even though the patient is free of abscesses at 
this time. 
Vaccine Treatment of Staphylococcus Infections.—Stock staphylococcus 
vaccine frequently yields as good results as autogenous vaccine and may 
be employed while the latter is being prepared. Cultures of pus may be 
made on any suitable medium. There is considerable evidence indicating 
the existence of serologic strains among the staphylococci, which renders 
advisable the use of autogenous vaccine whenever it can be properly pre- 
pared. 
The vaccine may contain 2,000,000,000 cocci per cubic centimeter. The 
first dose may be 0.1 c.c. or 200,000,000. The injections should be sub- 
cutaneous at intervals of seven days and never less than five days. Each 
succeeding dose may be increased by 0.1 or 0.2 c.c. until 1 c.c. is being given 
at one time. 
In infants and children under six years of age the vaccine may contain 
500,000,000 per cubic centimeter, the first dose being 0.1 c.c. and gradually 
increased in the same manner. In children six to twelve years of age the 
vaccine may contain 1,000,000,000 per cubic centimeter. Children over 
twelve years may receive the adult doses. 
As a general rule four to ten injections are required; sometimes one or 
two suffice. In chronic furunculosis I believe it is a good plan to give two 
or three more injections four to six months later in order to reinforce the 
immunity, even though the patient is free of furuncles at that time. 
As a general rule slight soreness follows at the site of injection and occa- 
sionally a mild constitutional reaction. With the first few doses a slight 
focal reaction may develop characterized by tenderness of the lesions with 
increased discharge of pus; in chronic furunculosis these focal reactions are 
desirable, providing care is taken not to render them severe by excessive 
doses. 
The results of vaccine treatment of recurring furuncles and carbuncles 
have been generally very good. This is especially true of deep-seated 
lesions. Superficial lesions and particularly those produced by Staphy- 
lococcus albus are frequently not improved by this therapy. In chronic 
furunculosis of infants particularly good results may be obtained. In both 
adults and children recurrent furunculosis and the development of car- 
buncles are usually accompanied by underweight, anemia, and other con- 
stitutional conditions which lower staphylococcus immunity and require 
appropriate attention. 
Serum Treatment of Staphylococcus Infections.—While several attempts 
have been made to treat staphylococcus infections with an immune serum, 
the investigations have been too few and too brief to warrant a statement 
in regard to the value of serum therapy in these infections. Thomas1 pre- 
pared a serum by immunizing a ram with 18 different strains of Staphylo- 
coccus pyogenes aureus, and reported good results in the treatment of 28 
cases of furunculosis and carbuncles. 
1 Jour. Amer. Med. Assoc., 1913, lx, 1070. 890 
VACCINES, SERA, BLOOD, PROTEINS IN TREATMENT OF DISEASE 
It is probable that, with more extensive use of antistaphylococcus 
serum, its value will be proved, especially in the treatment of severe furun- 
culosis of infants as well as of adults, when the low general vitality of the 
patient contraindicates the use of a bacterial vaccine. With extended use 
it may also be found to be of benefit in staphylococcus bacteremia, osteo- 
myelitis, arthritis, carbuncle, and other severe infections. 
The activity of the serum is probably largely dependent upon the presence 
of bacteriotropins and antitoxins, the former promoting phagocytosis and 
the latter neutralizing the staphylolysins or hemotoxic poisons produced 
by staphylococci. 
Non-specific Treatment of Staphylococcus Infections.—Turpentine in- 
jections for the purpose of producing “fixation abscesses” and the conse- 
quent absorption of these products, were introduced by Fochier,1 who in- 
jected 1 c.c. subcutaneously or intramuscularly. Cerioli2 has modified this 
plan by injecting 1 c.c. of a mixture composed of 16 parts eucalyptol and 
4 parts turpentine. The abscesses are quite large and more recently French 
and German clinicians have modified these methods and are now injecting 
intramuscularly at frequent intervals a few drops of a mixture of 4 parts of 
turpentine in 16 parts of sterile olive oil. 
Milk injections have sometimes yielded particularly good results. Or- 
dinary pasteurized or raw milk is boiled for five to ten minutes and when 
cooled 3 to 5 c.c. are injected intramuscularly. As a general rule a mild 
fever and leukocytosis develop in six to eight hours which subside in twrenty- 
four to forty-eight hours. The injections may be given at intervals of 
seven to ten days. 
TREATMENT OF SEPTIC WOUNDS AND SINUSES; LYMPHADENITIS 
Bacteriology.—A variety of aerobic and anaerobic bacteria have been 
found in the septic wounds of the war, including the chronic sinuses, with and 
without sequestra. Among the aerobes, Streptococcus pyogenes has been 
most frequent and most important; staphyloccoci, Bacillus coli, B. proteus, 
B. pyocyaneus, and numerous strains of coliform bacilli have also been found 
and usually in mixed culture. Of the anaerobic spore-bearing bacilli, 
B. tetani, B. perfringens (B. aerogenes capsulatus, B. welchii), B. malignant 
edema (vibrion septique), and B. cedematiens were the most common. One 
of the most remarkable lessons of the bacteriology of war wounds was the 
long periods of time in which bacteria and particularly the anaerobes re- 
mained in the tissues, and especially in sinus cases with sequestra; tetanus 
and gas gangrene developed many weeks after trivial wounds and operations, 
and healed wounds frequently broke down with severe sepsis as the result 
of too early exercise. 
Vaccine Treatment of Septic Wounds.—A great many cases of septic 
wounds have been treated with vaccines during the recent war, and espe- 
cially by the English surgeons; it is exceedingly difficult, however, to evaluate 
the results because of widely divergent opinions. Certainly the results have 
not been impressive—temperatures did not fall by crisis nor discharges 
cease as if by magic, and a decision whether vaccines have done any good 
relies mainly upon clinical impressions. Sir Berkeley Moynihan3 believes 
“that a case has been made out for the trial of vaccines in appropriate cases 
and in suitable conditions,” and is thoroughly convinced of the great value 
of streptococcus vaccine in serious wounds. 
1 Progres med., 1892, 15, 355. 
2 Gazz. d. osp. e d. clin.. 1915, 36, 721. 
* Lancet, 1916, 1, 333. It is the general consensus of opinion that vaccines had none or but 
slight direct influence upon the bacterial flora of septic wounds and sinuses; 
the basis for vaccine treatment rests upon the principle of producing anti- 
bodies and thereby raising the antibacterial power of the serum, providing 
for its access to the wound as with the use of Wright’s1 hypertonic saline 
solution which promotes the flow of lymph into the wound, and providing 
proper drainage and dressings. Wright2 claims that this lymph not only 
brings into the wound bacteriotropic and bactericidal substances, but 
promotes tryptic activity resulting in digestion of sloughs and aiding in the 
cleansing of the wound. Wright has also observed that the pus of these 
sloughing wounds affords a good medium for numerous varieties of bacteria 
and the gas gangrene group in particular (the serosaprophytes), whereas in 
fresh lymph—more or less unaltered blood fluids—only a few of these, of 
which the streptococci and staphylococci are most important, are able to 
grow (the serophytes). 
Vaccines are prepared of the aerobic bacteria only, and especially of 
streptococci; mixed vaccines are advisable if staphylococci, Bacillus proteus, 
B. coli, and B. pyocyaneus are also present. Such vaccines are given as 
soon as possible in the attempts to increase the resistance to general con- 
stitutional disturbances subsequent to surgical manipulation of the infected 
tissues, and to prevent the spread of bacterial infection of the tissues in the 
neighborhood of the wound; Goadby3 has advised at least four injections 
■over a period of two weeks preceding operation, and in a large series of 
cases found that this practice eliminated “flares” and greatly reduced the 
percentage of secondary hemorrhages due to septic infection. Tidy4 and 
Shera5 and Swan6 have had similar experiences; the presence of anaerobic 
bacteria was ignored, provided the patient received these immunizing 
inoculations and adequate drainage was provided. 
Prophylactic immunization among soldiers with antisepsis vaccine 
prior to exposure to injury has not been tried on a large scale, although 
strongly endorsed by Sir Berkeley Moynihan, Sir Almroth Wright, and Sir 
Watson Cheyne; stock vaccines prepared of strains of bacteria from war 
wounds have been commonly employed in individual cases while autogenous 
vaccines were being prepared. 
The doses should not be too large; the administration of a streptococcus 
vaccine may begin with 50,000,000 and gradually increased; mixed 
vaccines may begin with 25,000,000 of each of the important aerobic bacteria 
present, followed by increasing doses every three to five days over a period 
of two to three weeks. Operations should not be attempted for at least 
two days after the last dose. 
TREATMENT OF STREPTOCOCCUS INFECTIONS AND SEPTICEMIA 
891 
TREATMENT OF STREPTOCOCCUS INFECTIONS AND SEPTICEMIA WITH 
SPECIAL REFERENCE TO CELLULITIS, PUERPERAL SEPSIS, AND EN- 
DOCARDITIS. 
Kinds of Streptococci in Relation to Biologic Therapy.—There appear 
to be many saprophytic and non-pathogenic as well as pathogenic varieties 
of streptococci. Owing to variations in morphology, biologic characters, 
and virulence it has so far been impossible to make a satisfactory classifica- 
tion. The relation of the pneumococcus to the group has not been fully 
1 Lancet, June 23, 1917. 
2 Lancet, April 29, 1919, Brit. Med. Jour., 1915, 1, 625; ibid., 1915, 2, 629, 717. 
3 Lancet, 1916, 2, 89. 
4 Lancet, 1915, 2, 329. 
5 Vaccines and Sera in Military and Civil Practice, Hodder and Stoughton, 1918. 
,s Lancet, November 18, 1916. 892 
VACCINES, SERA, BLOOD, PROTEINS IN TREATMENT OF DISEASE 
decided. It was first thought that the streptococci of erysipelas, puerperal 
sepsis, scarlet fever, etc., belonged each to a different species, but it is now 
generally believed that the slight differences among the majority of strepto- 
cocci from these diseases were but acquired variations of organisms derived 
from the same species. Differences exist among streptococci, but these 
are not constant according to the source (disease). 
Some streptococci are able to produce definite hemolysis, and this type 
is probably the original Streptococcus pyogenes of Rosenbach, since desig- 
nated by Roily as Streptococcus hemolyticus and by Blake1 as Streptococcus 
hemolysans. Dochez, Avery, and Lancefield2 have recently shown that at 
least four biologic types exist among these as shown by agglutination and 
protection tests. A second group does not produce hemolysis; some of these 
produce methemoglobin including Streptococcus viridans, and others do not, 
designated as Streptococcus nonhemolyticus. 
The group of streptococci is, therefore, very large and indicates that in 
vaccine therapy autogenous vaccines should be prepared in order to make 
sure that the same strain or strains are being employed. In serum therapy 
the antistreptococcus serum should be polyvalent, but probably better 
results will be obtained by the immunization of horses with streptococci 
from the one disease instead of preparing one serum for all streptococcus 
infections. 
Immunity in Streptococcus Infections.—Human beings and most of the 
lower animals possess a fairly well-marked degree of natural immunity to 
streptococci. Of the lower animals, young rabbits are particularly sus- 
ceptible. Highly virulent streptococci, however, readily overcome the 
barriers to infection, and when gaining access to the lymph and blood are 
able to proliferate rapidly, produce virulent poisons, and prove very dan- 
gerous infections. In other words, natural defenses are readily overcome 
by highly virulent streptococci. 
Natural immunity is largely dependent upon intact ephithelial barriers, 
the secretions and phagocytosis. Streptococcus poisons are in the nature 
of aggressins, that is, repel leukocytes and retard phagocytosis. Recovery 
from infection is largely brought about by the production of anti-aggressins 
and opsonins aiding phagocytosis, and probably, to a lesser extent, by the 
production of agglutinins and streptococcidal lysins in the blood. 
Immunity to streptococci is built up very slowly. One attack does not 
confer more than a transient resistance and, indeed, as in recurring erysipelas, 
there may be no evidences of local or general acquired immunity at all even 
to the streptococci of the same disease. 
Vaccine Therapy of Streptococcus Infections.—There is a great differ- 
ence of opinion on the subject of using vaccines in acute streptococcus 
infections. Probably the majority of physicians believe that the patient 
should not receive this extra stimulation. In streptococcus cellulitis and 
lymphadenitis the results of judicious vaccine therapy is, however, usually 
very satisfactory. In puerperal streptococcus endometritis the writer has 
observed good results when the vaccine injections were started early instead 
of late in the infection. In acute streptococcus endocarditis with recovery of 
the micro-organisms in blood-culture, as likewise in streptococcus bac- 
teremia (septicemia), Barr and Bell,3 Thompson,4 Rosenow,5 and others have 
recorded encouraging results. Horder,6 Billings,7 and others, however, 
1 Jour. Med. Res., 1917, 36, 99. 
2 Jour. Exper. Med., 1919, 30, 179. 
3 Lancet, February 23, 1907, 499. 
4 Amer. Jour. Med. Sci., 1909, 138, 169. 
5 Jour. Amer. Med. Assoc., 1910, 55, 1719. 
6 Quart. Jour. Med., 1909, 2, 289. 
7 Arch. Int. Med., 1909, 4, 409. TREATMENT OF STREPTOCOCCUS INFECTIONS AND SEPTICEMIA 
893 
have reported negative results. It would appear that the acute types of 
endocarditis are more favorable for vaccine therapy than the slow, insidious 
types associated with progressive anemia and due to streptococci of low- 
grade virulence. 
The vaccine in small doses cannot be said to be primarily toxic; an 
injection does not add to the amount of toxic substances to be combated. 
It is largely a question of whether or not benefit is to be derived from extra 
stimulation of antibody producing tissues and whether these tissues are 
sensible to this extra stimulation and capable of a beneficial response. My 
decision in individual cases is based upon the general condition of the patient 
and the duration of the disease. If the patient is fairly robust I believe 
that small doses of vaccine are advisable in the early stages; if the patient 
is anemic and emaciated, as in the later stages, vaccine therapy is contraindi- 
cated. 
Stock vaccine may be employed while attempts are being made to 
recover the infecting streptococcus for the preparation of an autogenous 
vaccine. The vaccine should contain approximately 500,000,000 per cubic 
centimeter; the first dose is 0.1 c.c. or 50,000,000. Injections should be 
subcutaneous, repeated every five days and in gradually increasing amounts. 
Serum Treatment of Streptococcus Infections.—The acute character of 
streptococcus infections and their relative frequency and severity have 
made them the subject of numerous efforts on the part of various investi- 
gators toward developing an efficient serum therapy. To Marmorek belongs 
the credit of first attempting, in 1895, to prepare a curative serum on a large 
scale. Since then Aronson, Tavel, Krumbein, Moser, Meyer-Ruppel, 
Menzer, and others have prepared immune serums with various cultures 
and according to various methods. While many antistreptococcus serums 
will show undoubted protective value, especially against their homologous 
cultures as tested in experimental animals, the general opinion regarding 
their curvative value in streptococcus infections of man have been con- 
flicting and, as a rule, unfavorable. Occasionally the rapid improvement of 
a patient following an injection of the serum would indicate that it has 
proved beneficial, and the same is occasionally true of a particular group of 
infections treated with a specially prepared serum. The tendency of acute 
streptococcus infections to end spontaneously by crisis must, however, be 
borne in mind, and the good results observed in individual cases may be 
coincident with, rather than the result of, the administration of the serum. 
Several causes for the failure of antistreptococcus serum therapy are 
now understood, and if these can be eliminated the value of this form of 
therapy will be greatly augmented. 
1. The serum should he given in large doses, and by intramuscular and 
intravenous injection. In a true streptococcic infection the cocci are likely 
at some time to be found in the blood-stream, and an attempt to destroy 
these organisms or to limit a local infection by injecting 10 c.c. of serum in 
the subcutaneous tissues is almost sure to result in failure. As in the serum 
treatment of pneumonia, at least 100 c.c. of serum should be given intra- 
venously and the dose repeated if necessary. 
2. The serum should be used as early in the disease as possible instead 
of waiting until the patient has become moribund. In puerperal sepsis and 
scarlet fever, for instance, the question of serum therapy should be considered 
early, for when properly administered the serum will at least do no harm 
and may prove efficacious. In order to determine the value of the serum 
a bacteriologic diagnosis should always be attempted, especially by means 
of blood-cultures obtained by placing from 2 to 5 c.c. of blood in a flask 894 
VACCINES, SERA, BLOOD, PROTEINS IN TREATMENT OF DISEASE 
containing at least 100 c.c. of dextrose broth just prior to injecting the 
serum. 
3. The serum should be polyvalent. Marmorek maintains that all strepto- 
cocci are alike, and he has, accordingly, prepared his serum from a single 
highly virulent strain. Thus far no adequate methods for classifying these 
organisms have been discovered, but it is likely that future researches will 
show that streptococci from different infections possess different immuno- 
logic characters similar to the variations observed among the pneumococci 
causing lobar pneumonia. If this is found to be true, under these conditions, 
a similar serum treatment, while complicated, is likely to prove valuable in 
the treatment of streptococcus infection. Tunnicliff1 has recently reported 
that the serum of a sheep immunized with hemolytic streptococci from the 
throat in the acute stage of scarlet fever showed agglutinins, opsonins, and 
protective power highly specific for these micro-organisms. For the treat- 
ment of streptococcus infection in scarlet fever the serum should be prepared 
of numerous strains isolated from patients having this disease. What has 
just been said is also true of the other three infections so frequently strepto- 
coccal, namely, puerperal sepsis, phlegmonous cellulitis, and ulcerative 
endocarditis. If not these four, at least two antistreptococcus serums 
should be available; one for scarlet fever and the second for other infections; 
both, and especially the latter, should be prepared by immunizing horses 
with a large number of various strains. 
Mode of Action of Antistreptococcus Serum.—Virulent streptococci 
exert a powerful negative chemotactic influence upon leukocytes, repelling 
them and effectively resisting phagocytosis for varying periods of time. 
The early researches of Bordet showed that antistreptococcus serum neu- 
tralizes this influence and promotes phagocytosis. Since then numerous 
investigators have supported Bordet’s findings, so that it may be accepted 
as true that one of the chief antibodies in antistreptococcic serum is of the 
nature of a bacteriotropin or immune opsonin. A potent serum also contains 
an antitoxin, as may be shown experimentally by neutralization of the 
hemotoxic poison of streptococci, and also clinically, when the rapid sub- 
sidence of fever and general improvement of the patient are probably due, 
in part, to neutralization of streptococcal toxins. Thus far the presence of 
bacteriolysins has not been definitely proved, although they may be present 
and operative in vivo. I have found that antistreptococcus serum contains 
an antibody capable of fixing complement with streptococcus antigens.2 It 
may be stated, therefore, that the action of antistreptococcus serum is 
dependent primarily upon bacteriotropins, and secondarily upon antitoxins 
and, possibly, bacteriolysins. 
Preparation of Antistreptococcus Serum.—Some differences of opinion have been 
expressed regarding the advisability of passing cultures that are being used for purposes of 
immunization through a lower animal in order to increase their virulence. For example, 
the unsatisfactory results that have followed the use of Marmorek’s serum have been ascribed 
not only to the fact that it is monovalent, but also to possible alteration of the strain in its 
biologic characteristics by animal passage, so that its virulence for the human being was 
diminished or lost, and, accordingly, while the antiserum is protective for the animals through 
which the passage has been conducted, it is inactive for the human being. Tavel, Krum- 
bein, and Paltauf have prepared polyvalent serums with different strains from human infec- 
tions without animal passage. Menzer has prepared a serum with strains of cocci derived 
from acute rheumatic fever, and Moser with strains obtained from scarlet fever, which have 
also not been passed through animals. Aronson has attempted a combined procedure, 
making use of passed and unpassed cultures conjointly, and this appears to be the method 
of choice. In other words, those who prepare antistreptococcus serums should use as many 
1 Jour.' Amer. Med. Assoc., 1920, 74, 1386; ibid., 1920, 75, 1339. 
2 Arch. Int. Med., 1912, ix, 220. TREATMENT OF STREPTOCOCCUS INFECTIONS AND SEPTICEMIA 
895 
fresh strains as possible, and in several of the older cultures the virulence should be increased 
from time to time by passage through animals. 
In preparing the serum young and healthy horses should be used. The injections should 
first commence of dead cultures given subcutaneously, then of autolysates, and finally of living 
cultures administered intravenously. Occasionally severe local and general reactions are 
observed, and the whole procedure should be conducted under careful supervision. 
First Method.—Cultures are grown on a solid medium, and an emulsion and autolysate 
prepared as described for immunizing with meningococci. Begin by injecting subcutaneously 
5 c.c. of emulsion heated to 60° C. for one hour, and increasing the dose each week by S c.c. 
until 100 c.c. are given at one time. If the reactions are mild (general and local), the doses 
may be increased more rapidly. Then begin with 2 c.c. of living culture and increase the 
dose each week. When a dose of 10 c.c. is reached, inject with autolysate and living cultures 
alternately, gradually increasing the dose until a dose of 50 c.c. is reached. Living cultures 
are then given intravenously and the dose rapidly increased until 100 c.c. and more are given 
at one time. Intravenous injections are not infrequently tolerated better than subcuta- 
neous injections. The horses may be bled several times during the course of immunization 
and their serums tested. When the serum is to be used therapeutically, the animals should 
not be bled in less than from ten to fourteen days after the last injection was given. In 
view of the large doses required, a concentrated serum is advisable (Heinemann and Gate- 
wood1). Since trikresol has an inhibiting influence on phagocytosis (Weaver and Tunnicliff2) 
the minimal quantity (0.2 per cent, or less) should be used. 
Second Method.—Kolle prepares a polyvalent serum with cultures derived from cases of 
erysipelas, puerperal sepsis, scarlet fever, etc. Their virulence is increased from time to 
time by passage through rabbits. 
Horses are used for immunization purposes and all injections are given intravenously 
at intervals of a week. Cultures are grown on test-tubes containing a solid medium, and 
immunization is started with half a culture, heated. The dose is increased each week with 
an additional culture until it equals 16 cultures. Then living and killed cultures are mixed, 
giving in one week 2 living and 14 killed cultures and so on until 10 living and 6 killed cultures 
are given at a single dose. The horses are bled two weeks after the last dose is administered. 
Caulfield3 has recently described new methods of promising value. 
Standardization of Antistreptococcus Serum.—There is at present no single satis- 
factory method for standardizing these serums, although a satisfactory method is a desidera- 
tum for testing serums placed on the market. In order to obtain an approximate idea as to 
the value of a serum, the following tests may be employed: 
1. Protective Value.—At the Serum Institute in Vienna a passed culture (one used in the 
process of immunization) is selected, and that dose which will kill a mouse at the expiration 
of or just preceding the end of four days is regarded as a single lethal dose. In testing an 
antiserum ten times this quantity of culture is used with decreasing doses of immune serum 
injected twenty-four hours previously or simultaneously. A normal serum is one of which 
0.1 c.c. will afford protection, and 1 c.c. of such a serum is said to be one immunity unit, 
i. e., it affords protection against 1000 lethal doses of culture. For conducting these pro- 
tection tests the same technic and standard described for the standardization of antipneu- 
mococcus serum may be employed. 
2. Bacteriotropic Value— The technic of Wright or Neufeld may be employed with a 
virulent culture and human leukocytes. Weaver and Tunnicliff have observed better results 
when using 1 part of immune serum reactivated with 9 parts of fresh guinea-pig serum 
The technic and standard for conducting these tests may be similar to that 
Evans in the Hygienic Laboratory for the testing of antimeningococcus serum. 
3. Agglutination Tests.—These may be conducted in the same manner as tests employing 
antipneumococcus or antimeningococcus sera; a macroscopic technic is employed with an 
incubation of sixteen to eighteen hours at 55° C. A satisfactory serum should have a titer 
of 1 : 400 or higher. 
4. Complement-fixation Tests.—These tests may be employed with the bacterial emulsion 
or autolysate used in immunization as the antigen. 
Preservation.—The serum should be kept in a cool, dark place. After a few months it 
loses some of its protective value, and much of it on the market is worthless. 
Administration of Antistreptococcus Serum.—It must be emphasized 
here that, in order to obtain the best results, antistreptococcus serum must be 
given intravenously. In an adult patient with a severe general infection 
from 30 to 100 c.c. of serum, diluted with an equal amount of sterile normal 
1 Jour. Infect. Diseases, 1912, 10, 416. 
2 Jour. Infect. Diseases, 1911, 9, 130. 
3 Jour. Path, and Bacteriol., 1916-17, 21, 28. 896 
VACCINES, SERA, BLOOD, PROTEINS IN TREATMENT OF DISEASE 
salt solution, may be given in one dose. If improvement follows, subsequent 
doses should be given subcutaneously or intramuscularly in order to prolong 
the action of the serum. If no improvement follows in from twelve to 
twenty-four hours, or if an acute exacerbation sets in, a second dose should 
be given intravenously. Since the various manufacturers use different 
cultures in the preparation of these serums, it would be well to use a different 
brand of serum if the first does not exert a beneficial effect. In patients with 
severe infections the activity of the serum may be enhanced by adding, 
just before injection, 5 c.c. of fresh sterile guinea-pig serum to each 50 c.c. 
of the immune serum; for example, Tunnicliff has shown that sheep anti- 
streptococcus serum for scarlet fever rapidly lost its opsonic and protective 
power, which was restored by the addition of fresh normal sheep serum. 
Value of Antistreptococcus Serum.—Although the serum has been in 
use for almost twenty-five years, the true exact value of the remedy has not as 
yet been estimated. It may be stated that a carefully prepared and properly 
administered serum will do no harm and may do good, and that its use should 
form a part of the treatment of severe streptococcal infections. 
In some cases of wound infections with severe cellulitis and septicemia 
the serum may at times exert a most pronounced and happy effect. In 
other cases, and especially in those in whom the cocci are found in the blood, 
even repeated injections may be of no value. 
In puerperal sepsis and endocarditis of streptococcal origin the results 
of serum treatment have not been uniform, but are generally unfavorable. 
If serum is administered at all, it should be given early, in large doses, and 
intravenously. Not all cases of puerperal sepsis are streptococcic, and 
while the physician may not be justified in withholding serum until a bac- 
teriologic diagnosis has been made, this factor must be considered when 
estimating the value of a serum. 
Blood Transfusion in the Treatment of Streptococcus Septicemia.— 
Immunized Blood.—In Chapter XLII this subject is considered in more 
detail. Occasionally in severe, prolonged streptococcus infections with 
anemia, as in puerperal sepsis, ulcerative endocarditis and cellulitis, the 
transfusion of 500 c.c. of compatible human blood turns the tide in favor of 
recovery. One drawback is the possible harmfulness of the reaction that 
may follow transfusion. 
If time permits it may be advisable to immunize the donor beforehand, 
as recently reported by Dick.1 The donor may receive three injections of 
heat killed vaccine on successive days of 1,000,000,000, 2,000,000,000, and 
3,000,000,000. Transfusion may be done a week later, injecting 50 to 100 
c.c. of blood and repeating again next day or giving 500 c.c. at one time. 
Wright2 has recently stated that the transfusion of plain blood may be of 
no avail because more and better pabulum is supplied the micro-organisms. 
His experiments have suggested the practical utilization of adding vaccine 
to the blood and after a suitable interval of about three hours injecting the 
whole into the patient. Wright has observed that this “immunization in 
vitro” is non-specific, enhances the bactericidal activity of the serum, and 
calls the method “immuno-transfusion.” In a case of streptococcus wound 
sepsis Wright drew off 1 liter of blood from a compatible donor into a 
paraffin coated receptacle, added 1,000,000 killed staphylococci, and later 
injected this mixture intravenously with a successful result. Whether or 
not citrated blood may be used is not stated and the method requires further 
trial before an opinion of its value may be given. 
1 Jour. Amer. Med. Assoc., 1922, 78, 1192. 
2 Lancet, March 29, 1919. TREATMENT OF ERYSIPELAS 
897 
Leukocytic Extracts, Peptone, and Other Non-specific Agents in the 
Treatment of Streptococcus Infections.—Leukocytic extract may be injected 
subcutaneously or intramuscularly in doses of 10 to 15 c.c. daily for several 
days. 
Gow1 has employed intravenous injections of 10 per cent, solutions of 
Witte’s peptone in the treatment of septicemia; 8 to 10 c.c. are injected very 
slowly by means of a very fine needle (No. 28) and repeated every other day 
in doses increased by 2 c.c. until 16, 18, or 20 c.c. are given at one time. 
Gow makes no extravagant claims for this method, although he thinks it is 
a valuable adjunct to treatment. Nolf2 has likewise employed intravenous 
injections of peptone in the treatment of streptococcus, staphylococcus, and 
pneumococcus septicemias, and claims good results. 
Liidke3 has employed injections of deutero-albumose. Lindig and 
Arweiler4 have used injections of casein. Intramuscular injections of milk 
have been employed by Bianchi,5 but the results have not been particularly 
encouraging when streptococci were present in the blood. 
TREATMENT OF ERYSIPELAS 
Erysipelas is also a streptococcus infection, and much that has been 
stated in the preceding paragraphs applies to its treatment with biologic 
agents. 
Immunity in Erysipelas.—Disposition to this disease is wide-spread, but 
susceptibility is specially marked in the cases of recently delivered women, 
after injuries, burns, and surgical operations. Sometimes infection appar- 
ently begins within the nose and spreads to the skin. 
One attack does not prevent subsequent attacks; indeed, the patient 
may be left even more susceptible. The disease has been experimentally 
produced in the same area of skin many times, indicating that even a local 
tissue immunity is not engendered. As previously stated in discussing 
immunity to streptococcus infections in general, natural resistance is largely 
due to phagocytosis. In the mechanism of recovery, phagocytosis aided by 
specific opsonins are apparently of most importance. 
Vaccine Treatment of Erysipelas.—Vaccine treatment has generally 
failed to influence the disease and probably most physicians are opposed to 
the use of vaccines in acute erysipelas; it cannot be denied, however, that 
when vaccines have been employed early in the infection, that not infre- 
quently the lesions and symptoms have rapidly subsided. Owing to the 
self-limited character of erysipelas it is not always easy, however, to evaluate 
the worth of a therapeutic measure. Because of the difficulty and delays 
in obtaining the streptococcus in pure culture, it is necessary to employ a 
stock streptococcus vaccine for a time at least. During the acute stage 
the doses should be small, beginning with 0.1 c.c. of a vaccine containing 
500,000,000 killed streptococci per cubic centimeter; the injections should 
be given subcutaneously every five days and in gradually increasing amounts. 
In chronic and recurring erysipelas vaccine therapy is more strongly 
indicated; the dosage and intervals of injections should be as stated above 
until 1 c.c. of vaccine is being given at one time. 
Serum Treatment of Erysipelas.—Antistreptococcus serum has been 
frequently employed in the treatment of erysipelas, with indifferent results. 
1 Brit. Med. Jour., 1920, 2, 268. 
2 Jour. Amer. Med. Assoc., 1919, 73, 1579. 
3 Berl. klin. Wchn., 1920, 47, 344. 
4Therap. Halbmonatsch., 1920, 34, 470. 
6 Riforma med., 1920, 36, 1026. 898 
VACCINES, SERA, BLOOD, PROTEINS IN TREATMENT OF DISEASE 
The usual procedure has been to inject 10 to 20 c.c. of the serum subcuta- 
neously and these amounts were too small. In severe spreading infections 
accompanied by high fever, considerable edema, and delirium, serum should 
be given, but as early as possible, intravenously and in dose of 50 to 100 c.c., 
repeated the following day if necessary. Administered in this manner 
beneficial effects are to be expected. Jez1 has reported favorably upon the 
treatment of erysipelas by subcutaneous and intravenous injections of 
10 to 15 c.c. of the patient’s own serum. Kaiser2 has reported a case success- 
fully treated by injections of whole blood from a convalescent donor- 
Leukocytic Extract and Other Non-specific Agents in the Treatment of 
Erysipelas.—Lambert,3 Hess and Uwver,4 and others have reported favorably 
upon the subcutaneous injection of 10 to 15 c.c. of leukocytic extract once 
a day for several days in the treatment of acute erysipelas. 
Petersen5 has reported that the intravenous injection of 0.5 to 2 c.c. of 
a 2 per cent, solution of proteoses has yielded very satisfactory results; also 
the administration of milk and typhoid vaccine. Holler6 has employed 
intravenous injections of deutero-albumose, but the doses must be small 
in order to avoid severe general reactions which are readily induced (not 
over 0.1 to 0.5 c.c. of 10 per cent, solution). Blumenau7 has employed 
nuclein injections; sodium nucleinate may be injected subcutaneously in 
dose of 0.5 gm. 
Schmidt8 and others have employed intramuscular injections of 5 to 
10 c.c. of market milk boiled for ten minutes and cooled. 
Normal horse-serum and diphtheria antitoxin have been employed, espec- 
ially by the French clinicians, in dose of about 10 to 20 c.c. by cubcutaneous 
injection. 
My impression is that these non-specific agents are especially indicated 
in chronic and recurrent erysipelas; great care must be exercised in dosage 
because of the chances of eliciting severe general reactions. 
TREATMENT OF SCARLET FEVER 
The etiology of scarlet fever being unknown, the only true form of specific 
treatment is by means of injections of serum from convalescent cases. One 
attack of the disease generally confers an immunity for life, and immunity 
principles are doubtless in the serum, at least for a short space of time after 
recovery. 
Streptococci are known to have a very close relationship to the disease;, 
probably they are the most important bacteria of secondary infection. 
These streptococci are usually of the hemolytic variety, are responsible for 
the severe throat lesions and adenitis, and may be more or less specific for 
scarlet fever, as indicated by recent investigations by Tunnicliff, to which 
reference has been previously made. 
Vaccine Treatment of Complications of Scarlet Fever.—This refers al- 
most solely to the use of vaccines in the treatment of complications, as 
suppurative adenitis, otitis media, etc. Streptococci and Staphylococcus 
aureus are the usual findings in the acute cases; in chronic infections other 
1 Wien. klin. Wchn., August 31, 1901. 
2 Arch. Pediat., 1915, 32, No. 7. 
3 Amer. Jour. Med. Sci., 1909, 137, 506. 
4 New Jersey Med. Rec., 1913, lxxxi-, No. 11. 
6 Protein Therapy and Non-specific Resistance, MacMillan Co., 1922, 191. 
6 Med. Klinik, 1917, 13, 1038. 
7 Ztschr. f. Immunitatsf., Ref., 1911, 3, 1075. 
3 Med. Klinik, 1916, 12, 171; ibid. 1920, 16, 691. TREATMENT OF SCARLET FEVER 
899 
bacteria may be found. I believe it is always a good plan to culture the pus 
from these lesions and hold an autogenous vaccine in readiness for later 
use in case of necessity. 
While streptococcus vaccines have been advocated as a means of prophy- 
lactic immunization against scarlet fever, they have not proved of value in 
the treatment of the disease. 
Treatment of Scarlet Fever with Antistreptococcus Serum.—In severe 
anginose or malignant scarlet fever large doses of serum from horses especially 
immunized with strains of streptococci from scarlet fever patients have, on 
the whole, yielded favorable results. Not all cases of severe scarlet fever, 
however, are due to secondary streptococcal infections: those patients who 
are overwhelmed and prostrated at the very outset are probably intoxicated 
with the true scarlatinal virus, whatever that may be, and such cases are not 
likely to be benefited by serum treatment. The patients most likely to 
improve under serum therapy are those who become severely ill after the 
onset of the disease and the appearance of the eruption. 
The serum should be prepared by the immunization of horses with 
strains of streptococci from scarlet fever cases. Very probably it has no 
specific effect upon the scarlatinal virus, but solely upon the secondary 
streptococcus infection and should be employed in severe infections for this 
purpose. In Russia, where scarlet fever has a high morbidity and mortality,. 
Moltchankoff1 has found that serum treatment reduced the mortality from 
47.4 to 16.1 per cent. 
For adults the dose of antistreptococcus serum should be 50 to 100 c.c. 
by intravenous injection and repeated twelve hours later and again the 
next day. Children may receive 25 to 50 c.c. by intravenous injection if 
possible; otherwise 25 to 50 c.c. by intramuscular injection, repeated daily 
for three or four days if necessary. 
Treatment of Scarlet Fever with Convalescent Serum and Blood.— 
Weisbecker2 in 1897 first employed the sera of persons convalescent from 
scarlet fever in the treatment of acute cases; Huber and Blumenthal,3 von 
Leyden,4 Rumpel,5 Scholtz,6 Reiss and Jungman,7 Koch,8 Rowe,9 Zingher,10, 
and others have reported favorably upon the treatment of scarlet fever and 
particularly severe infections, with injections of serum from persons who 
have recovered from this disease. 
Kling and Widfelt11 in 1918 reported on the use of convalescent serum in 
the treatment of 237 cases. Of these, 17.7 per cent, died, while in a corres- 
ponding group of 91 severe cases in which no serum was given, the mortality 
was 70 per cent. Of those receiving serum on the second day, 93.7 per cent, 
recovered; with each day’s delay the mortality increased, until by the sixth 
day only 50 per cent, recovered. They believed that severe toxic cases were 
benefited by the convalescent serum. They found the serum from persons who 
had mild attacks as effective as that from persons who had severe cases, but 
they also observed a variation in the effectiveness of sera from different 
individuals. 
1 Jahrb. f. Kinderh., 1907, lxvi, 503. 
2 Ztschr. f. klin. Med., 1897, 32, 188. 
3 Berl. klin. Wchn., 1897, xxxiv, 671. 
4 Deutsch. Archiv. f. klin. Med., 1902, lxxiii, 616. 
5 Munch, med. Wchn., 1903, 50, 38. 
6 Fortschr. d. Med., 1903, 21, 353. 
7 Deutsch. Archiv. f. klin. Med., 1912, cxii, 70. 
8 Munch, med. Wchn., 1913, lx, 2611; Deutsch. med. Wchn., 1915, xli, 372. 
8 Med. Klinik., 1913, ix, 1978. 
10 Jour. Amer. Med. Assoc., 1915, 65, 875. 
11 Hygiea, 1918, 80, 2. 900 
VACCINES, SERA, BLOOD, PROTEINS IN TREATMENT OF DISEASE 
Barker,1 Weaver,2 Bode,3 Synnott,4 Glaser,5 Schultz,6 and others have 
likewise reported most favorably upon this treatment of severe toxic cases 
of scarlet fever. An initial fall of temperature usually occurs with improve- 
ment of the restlessness, delirium, and great improvement of the general 
condition of the patient. 
Reiss and Jungman obtained the serum from persons about the third 
week of the disease and injected 50 to 100 c.c. intravenously. Koch empha- 
sizes the necessity of starting the treatment as early in the disease as possible 
and injecting large doses of the serum intravenously. Rowe was unable 
to convince himself that there were any different effects in cases treated 
with normal and with convalescent serum; Koch suggests that the con- 
valescent serum be reserved for the gravest cases, and that otherwise normal 
human serum be used. Zingher has used intramuscular injections of whole 
citrated blood, administering from 2 to 8 ounces in a series of injections at 
close intervals. Normal human blood was also found of value in the treat- 
ment of a group of later septic cases, seen from the fifth to the eighth day of 
the disease. It is apparent that these methods of treatment are worthy of 
further trial, and particularly in the treatment of severe or anginose cases 
of the disease. 
The technic is very simple and the method of treatment particularly 
applicable in hospitals for the treatment of scarlet fever. Donors are 
selected after the second week of the disease, and preferably from among 
adolescents or adults who show no evidences of tuberculosis. From each 
person 30 to 100 c.c. of blood may be obtained under aseptic precautions 
and the serum carefully separated, submitted to the Wassermann test, 
Cultured for sterility, and preserved with 0.2 per cent, tricresol in sterile 
Containers in a refrigerator. In young children an intravenous injection 
may not be possible unless one of the larger veins, as the external jugular 
or longitudinal sinus, is selected. It is advisable to inject the serum as 
early in the disease as possible in large doses (20 to 50 c.c. for a child seven 
years of age) and, preferably, by intravenous injection. The injections 
appear to be harmless and no special precautions as to the presence of hem- 
agglutinins or hemolysins appear necessary. Otherwise the serum may be 
injected intramuscularly in the gluteal region, and in the absence of specific 
convalescent serum, normal human serum collected in the same manner may 
be injected. As a general rule three or four injections are required at 
intervals of six to twelve hours to influence the disease, and particularly the 
severe infections. 
In using whole blood the method employed by Zingher is very satis- 
factory. Two c.c. of a sterile 10 per cent, solution of sodium citrate in 
normal salt solution is placed in a sterile 100-c.c. bottle; 2 ounces of blood 
are collected, added, and briefly shaken; the blood is now ready for intra- 
muscular injection. Wassermann reactions should be made beforehand on 
all possible donors so that the proper persons may be selected. In private 
practice the physician may take the blood from either one of the parents 
or close relatives. 
Weaver has injected the serum intramuscularly in amounts of from 60 
to 90 c.c. A second dose was given after twenty-four hours when the first 
was not followed by satisfactory improvement or when improvement was 
followed by a tendency to relapse. 
1 Arch. Pediat., 1914, 31, No. 8. 
2 Jour. Amer. Med. Assoc., 1921, 77, 1420; Jour. Infect. Dis., 1918, 22, 211. 
3 Arch. f. Kinderh., 1921, 69, 256. 5Ztschr. f. klin. Med., 1916, lxxxiii, 41. 
4 Med. Rec., 1917, xci, 106. 6 Therap. Monatsh., 1918, 32, 12. TREATMENT OF TYPHOID FEVER 
901 
Blood was drawn aseptically into sterile bottles, 200 to 300 c.c. being 
taken from adults and proportionately less from the larger children. No 
ill effects from bleedings were observed. The sera are separated from the 
clots, mixed, and preserved with 0.3 per cent, tricresol in bottles of 30 c.c. 
in a refrigerator. The serum tends to lose in efficiency after keeping more 
than eight weeks. 
The value of the treatment of scarlet fever by means of intramuscular 
injections of convalescent serum or blood or by intravenous injections of 
serum is well supported by numerous investigators mentioned above. It is 
well known that the course of scarlet fever is uncertain, some patients 
suddenly becoming rapidly better under any plan of treatment. But in 
severe cases treated early with convalescent serum the almost constant fall 
of temperature and rapid general improvement are striking. It would 
appear that the treatment is worthy of adoption not only in hospitals for 
the treatment of scarlet fever, but in private practice as well. In the 
absence of available convalescent serum normal blood may be used, but 
the results are not as satisfactory. 
Treatment of Scarlet Fever with Normal Serum and Other Non-specific 
Agents.—Moog1 and Rehder2 have treated scarlet fever with injections of 
normal serum and the former thought the results were as good as observed 
with convalescent serum. Holler3 has employed intravenous injections twice 
daily of 1 c.c. of a 10 per cent, solution of deutero-albumose, and Ludke4 
has employed albumose injections in the same manner. 
According to these investigations the beneficial effects of serum treat- 
ment with normal or immune sera are non-specific and the choice of agent 
merely a matter of individual preference. The consensus of opinion, how- 
ever, is to the effect that the results produced by convalescent serum are in 
a large part specific and that this serum is to be preferred in the treatment of 
scarlet fever. 
TREATMENT OF TYPHOID FEVER 
By reason of being the first acute infectious disease treated with vaccines, 
typhoid fever has since commanded a great deal of attention from the 
standpoint of treatment with both specific and non-specific vaccines and 
other biologic agents. Indeed the subject of non-specific protein therapy 
has been largely developed, clinically at least, by studies of this nature in 
typhoid fever and arthritis. 
Vaccine Treatment of Typhoid Fever.— Treatment with Plain Vaccine 
Subcutaneously.—While killed preparations of the typhoid bacillus have 
been injected subcutaneously as a method of treatment in typhoid fever 
since the work of Frankel5 in 1913, the subject is still in the experimental 
stage. An analysis of the literature upon this subject by Callison,6 Watters,7 
and Krumbhaar and Richardson8 show that at best the ordinary type of 
heat-killed vaccine administered subcutaneously does no harm and may 
shorten the course of the disease, prevent fever relapses and complications, 
and yield a slightly lower mortality. The treatment should be given early 
if at all, while the resistance and general condition of the patient are good. 
The injections are given subcutaneously and the doses should be modified 
in individual cases in order to avoid severe reactions. The initial dose may 
be 100,000,000 bacilli; three days later 250,000,000 or 500,000,000 may be 
1 Therap. Monatsh., 1914, 28, 37. 
2 Deut. Arch. f. klin. Med., 1916, cxx, 237. 
3 Med. Klinik, 1917, 13, 1038. 
4 Berl. klin. Wchn., 1920, lvii, 344. 
6 Deutsch. Med. Wchn., 1893, xix, 985. 
6 Amer. Jour. Med. Sci., 1912, cxliv, 350. 
7 Med. Rec., 1913, Ixxxiv, 518. 
8 Amer. Jour. Med. Sci., 1915, cxlix, 406. 902 
VACCINES, SERA, BLOOD, PROTEINS IN TREATMENT OF DISEASE 
injected, followed by three or four similar doses at intervals of five days. 
During convalescence two or more additional doses may be given in an 
effort to prevent relapses. 
Treatment with Sensitized Vaccine Subcutaneously.—Garbat1 has found 
the subcutaneous administration of sensitized typhoid bacilli or serobacterin 
superior to the non-sensitized vaccine. His vaccine was sensitized with 
the sera of typhoid convalescent cases and usually given subcutaneously in 
dose of 500,000,000 (250,000,000 to very sick individuals) every five to seven 
days for three or four doses; at least one injection was given during con- 
valescence to prevent relapse. Seventeen patients were treated, with one 
fatality. Harmful effects were not observed and the general course of the 
disease seemed milder and the complications less frequent. Ardin-Delteil2 had 
previously used Besredka’s living vaccine, and since then several observers 
have employed living and dead sensitized vaccines by subcutaneous injec- 
tion, but the results are quite similar to those observed with plain vaccines 
administered by subcutaneous injection. 
Treatment with Vaccine Intravenously.—In 1913 Thirloix and Bardon* 
reported the successful treatment of typhoid fever by the intravenous 
injection of 2,000,000 to 10,000,000 bacilli. More striking results have been 
reported by Ichikawa,4 Kraus,5 and others from intravenous injections of 
sensitized vaccine. Ichikawa prepared his vaccine by suspending ten loops 
of fresh typhoid culture in 10 c.c. of human typhoid convalescent serum and 
incubating for five or six hours. The organisms were then secured by 
centrifuging, washed three times and suspended in 100 c.c. of saline solution 
containing 0.3 per cent, tricresol. This vaccine was used unheated in doses 
of 0.5 c.c. diluted with a syringeful of saline and slowly injected intravenously. 
Gay and Chickering6 have reported upon the treatment of 53 cases of 
typhoid fever with intravenous injections of 1/50 to 1/25 milligram of a 
sensitized, polyvalent, killed typhoid vaccine sediment prepared after the 
method of Gay and Claypoole. 
A reaction usually follows these injections. A rigor or chill develops 
in ten to fifteen minutes accompanied by a rise in temperature of 1 to 3 
degrees, reaching its maximum in three hours and then falling. There is 
also a slight increase of the pulse-rate, slight cyanosis and respiratory distress, 
a sense of discomfort, and a leukopenia. 
The temperature reaches normal or subnormal in about twelve hours 
accompanied by sweating and amelioration of general symptoms and accom- 
panied by a leukocytosis. 
Mild reactions of this kind are desirable and the amounts of sediment 
required vary with different patients. As a general rule the injections are 
given every two or three days in slightly increasing amounts until three or 
four have been given. Gay has not observed any harmful effects and as 
many as fifteen or sixteen injections have been given a single patient. 
The mortality in a series of 98 cases was 6.6 per cent.; in 66 per cent, of 
the cases a distinct benefit was obtained, as shown by lowered temperature, 
disappearance or amelioration of subjective symptoms, and an apparently 
accelerated recovery. 
1 Jour. Amer. Med. Assoc., 1915, 64, 489. 
2 Compt. rend. Acad. d. sc., 1912, cxl, 1174. 
3 Soc. med. d. Hop., 1913, 36, 108 
4Ztschr. f. Immunitatsf., orig., 1914, xxiii, 32. 
5 Wien. klin. Wchn., 1914, xxvii, 1443. 
6Archiv. Int. Med., 1916, xvii, 303. (This paper contains a very good review of the 
literature upon the treatment of typhoid fever with the intravenous administration of sensi- 
tized vaccines.) TREATMENT OF TYPHOID FEVER 
903 
Comparative Value of Different Vaccines by Subcutaneous and Intra- 
venous Infections.—Gay1 has recently made a careful survey of the results 
of treatment with different vaccines administered subcutaneously and 
intravenously by different observers: 
SUMMARY OF RESULTS OBTAINED BY RECENT OBSERVERS (1913-17) IN THE 
TREATMENT OF TYPHOID FEVER BY VACCINES ADMINISTERED IN 
VARIOUS WAYS 
Observ- 
ers. 
Total 
Cases. 
Estimates 
Based on. 
Bene- 
fited, 
Per Cent. 
Mortal- 
ity, 
Per Cent. 
Untreated vaccine subcutaneously 
30 
1001 
512 
46 
14.5 
Sensitized vaccine subcutaneously 
14 
593 
239 
69 
8.0 
Untreated vaccine intravenously 
22 
501 
233 
62 
13.0 
Sensitized vaccine intravenously 
12 
487 
316 
85 
11.0 
These mortality records are within the normal limits of untreated cases, 
but Gay believes that the results in general were more favorable to the use 
of sensitized vaccine injected intravenously. 
Gay has attributed most of the beneficial results to non-specific effects 
and principally to the hyperleukocytosis. Apparently the typhoid protein 
acts much as other non-specific proteins described below; at least the re- 
actions are the same. In addition, however, it is possible that typhoid 
vaccines in general, including the sensitized sediment, have the property of 
engendering specific effects in some degree. At least there is generally an 
increase of agglutinins, and Gay believes that specific antibodies aid in the 
destruction of typhoid bacilli by the increased numbers of leukocytes. 
The Effect of Vaccine Therapy Upon Relapses and Complications in 
Typhoid Fever.—Vaccine therapy has had little or no influence upon the 
occurrence of relapses. In Gay’s series they occurred in 10.2 per cent, which 
was about the expected number. Gay believes, however, that the adminis- 
tration of a few doses of vaccine by subcutaneous injection on alternate days 
after the temperature has remained normal for twenty-four hours, may aid 
in preventing relapses; the dose of his vaccine for this purpose is 0.1 miligram 
corresponding to about 800,000,000 bacilli. This plan has also been recom- 
mended by Meyer,2 Meyer and Alstaedt,3 and others. 
In the treatment of complications, as periostitis, glandular suppuration, 
cholecystitis, and similar complications due to localized infections, a typhoid 
vaccine may be of considerable aid. The vaccine should be autogenous if 
possible, and mixed if the lesions are open and other bacteria of probable 
pathogenic activity are found. Stone4 has reported favorably upon the 
treatment of cases of typhoid carriers with the ordinary vaccine, and it would 
appear that this form of treatment is worthy of trial. 
Serum Treatment of Typhoid Fever.—The serum of Chantemesse is the 
only serum that has been used on a large scale in the treatment of typhoid 
fever in man. 
The serum is derived from horses that have been immunized for several 
years with bouillon filtrates containing typhoid toxin, chiefly endotoxin, 
and with typhoid bacilli. Kraus and von Stenitzer, Meyer, Bergell, and 
Aronson use bouillon filtrates and aqueous bacterial extracts; Besredka 
1 Typhoid Fever, MacMillan Co., 1918, 222. 
2 Berl. klin. Wchn., 1915, liii, 870. 
3 Berl. klin. Wchn., 1915, No. 52, 677. 
4 Jour. Amer. Med. Assoc., 1910, lv, 1708; Amer. Jour. Med. Sci., 1912, cxliii, 544. 904 
VACCINES, SERA, BLOOD, PROTEINS IN TREATMENT OF DISEASE 
injects dead and then living cultures; MacFadyen uses an endotoxin secured 
by breaking up cultures frozen at very low temperature. It would appear 
that a serum should be bacteriolytic and endotoxic, and this probably is best 
secured by prolonged intravenous immunization of horses with a large 
number of dead cultures and then with autolysates and living cultures 
conjointly. 
According to Chantemesse, the subcutaneous injection of a few drops 
of his serum produces leukocytosis and raises the opsonic index of the pa- 
tient’s serum. He emphasizes the fact that the serum should be given 
early—before the seventh day—and reports that, by its use, the mortality 
has been reduced from 17 to 4.3 per cent. These results have not been 
generally confirmed. 
Gay and Chickering have treated a few cases with goat immune serum 
and vaccine, and thought that the beneficial effects were more pronounced 
than with vaccine alone; further experiences, however, have not confirmed 
these hopes, although Gay believes that the subject of combined serum and 
vaccine therapy is one that should be more extensively tested. 
Kraus and V. Steritzer1 and Rodet2 had previously used a typhoid im- 
mune serum by subcutaneous or intravenous injection, with apparently 
good results; it is highly probable, however, that the beneficial results were 
due to non-specific protein shock reactions rather than to specific effects. 
Nicolae and Conseil3 have treated 5 patients with the serum of con- 
valescent typhoid patients. The dose was about 130 c.c.; in their experience 
the duration of the disease was not shortened, but certain nervous symptoms 
appeared to be improved. Garbat4 treated 3 severe cases with subcutaneous, 
and intramuscular injections of convalescent serum; a third case received 
intravenous injections because the bloods of donors and patient were com- 
patible. Two to three doses of 50 to 100 c.c. of serum were given, with good 
results, and Garbat believes that this treatment should be considered for 
the treatment of severe cases when donors are available. 
In cases of hemorrhages it would appear advisable to transfuse the whole 
blood of a convalescent donor or some one who has had typhoid fever, in 
order not only to furnish immune substances, but to secure a styptic effect 
as well. 
Non-specific Treatment of Typhoid Fever.—From what has been stated 
above it is apparent that most observers are of the opinion that the beneficial 
effects to be secured in the treatment of typhoid fever with vaccines and 
sera and especially by intravenous injection are due to their non-specific 
effects. 
Heterovaccines Intravenously.—Kraus and Mazza5 found that the injec- 
tion of colon vaccine did just as well as typhoid vaccine. Reibmays6 and 
Decastello7 employed cholera and colon vaccines, and Stein,8 in presenting 
the results in about 1500 cases of typhoid treated by him both subcutaneously 
and intravenously, found that by either method of injection colon vaccine 
gave results similar to typhoid vaccine. 
Albumoses Intravenously.—Liidke,9 in 1915, treated 23 cases of typhoid 
with intravenous injections of deutero-albumose and later 78 cases of typhoid 
and paratyphoid B infections; no deaths occurred in the latter group and 
the results were good, especially when the treatment was instituted early in 
1 Deutsch. med. Wchn., 1911, 37. 
2 Bull. Acad. d. Sci., 1916, 76, 85, 114. 
3 Ann. de l’Inst. Pasteur, 1912, 26, 332. 
4 Jour. Immunology, 1916, 1, 387. 
5 Deutsch. med. Wchn., 1914, xl, 1556. 
6 Munch, med. Wchn., 1915, lxii, 610. 
7 Mitt. d. Gesellsch. f. inn. Med., 1915, 14, 105. 
8 Wien. klin. Wchn., 1919, 32, 895. 
9 Miinch. med. Wchn., 1915, lxii, 321. TREATMENT OF TYPHOID FEVER 
905 
the disease. He injected 0.5 to 1.5 c.c. of a 10 per cent, solution every two 
or three days as indicated. 
Holler1 treated 350 cases of typhoid fever with injections of deutero- 
albumoses, giving two to four daily injections of 10 per cent, solutions be- 
ginning with 0.5 c.c. and increasing as indicated. The average duration 
of the disease was ten days and the mortality | per cent. 
Galambos2 treated 25 cases with injections of 1 c.c. of a 4 per cent, 
solution of deutero-albumose and obtained a critical drop in the temperature 
in 50 per cent, of the cases. These treatments appeared to be more effective 
than the injection of heterovaccines of staphylococci, gonococci, and colon 
bacilli. Galambos has also employed injections of physiologic saline 
solution (100 c.c. intravenously) and methylene-blue by mouth in dose of 
1.2 gm. per day divided into four-hour doses. The intravenous injection 
of saline was found too bothersome for routine use; methylene-blue was found 
to be a stimulant and the euphonia so commonly observed in non-specific 
therapy was apparent in most instances, but protein-split products appeared 
to yield the best results. Medication was continued some days after the 
temperature reached normal. 
In regard to dosage of proteoses, one of two plans may be followed: 
either to give sufficient to provoke a general reaction and perhaps repeat 
the dose after several days if the fever has not been altered, or to give small 
daily doses as employed by Holler, Jobling, and Petersen. 
Milk Intramuscularly.—Owing to the greater ease of administration, 
intramuscular injections of 5 to 10 c.c. of milk boiled for ten minutes and 
cooled, have become quite popular among non-specific agents in the treat- 
ment of typhoid and paratyphoid fevers. Decided advantages are the 
styptic effects of milk and the greater ease in controlling the reactions by 
attention to dosage, in addition to the greater ease of administration. 
Schmidt,3 Saxe,4 Muller,5 Cormaldesi,6 Grote,7 and others have rendered 
favorable reports, and especially when the injections were started early. 
In most cases a rise of temperature persisting for about two days was followed 
by lysis to a normal level. There was only rarely chills and the reactions 
were generally mild. 
Karell and Luksch8 have also employed milk injections for the treatment 
of bacillus carriers with apparently good results. 
Mechanism of Action of Non-specific Agents in the Treatment of Typhoid 
Fever.—A great deal of attention has been given the subject of antibody 
production and particularly of agglutinins, under stimulation by non-specific 
agents. Some investigators have reported an increased production of 
agglutinins, and a few an increase of bacteriolysins and opsonins; others 
have reported negative results. Very probably the injection of non-specific 
agents does tend to stimulate the antibody producing tissues and particularly 
after these have become sensitized. 
Doubtless the leukocytosis is an important factor in overcoming the 
infection, and especially when aided by the increased production of anti- 
bodies, as suggested by Gay. Ltidke and Petersen also suggest that after 
the reactions with alterations in cellular permeability and stimulation, the 
cells acquire an increased resistance to the intoxication. 
Precautions in the Treatment of Typhoid Fever with Vaccines and Non- 
specific Agents.—There are two chief dangers according to Petersen to keep 
iMed. Klin., 1917, 13, 1038. 
2 Wien. klin. Wchn., 1916, 29, 1041. 
3 Med. Klin., 1916, 12, 171. 
4 Wien. klin. Wchn., 1916, 29, 1043. 
6 Deutsch. med. Wchn., 1918, xliv, 545. 
6 Riforma med., 1920, 36, 296. 
7 Munch, med. Wchn., 1919, lxvi, 307. 
8 Wien. klin. Wchn., 1916, 29, 187. 906 
VACCINES, SERA, BLOOD, PROTEINS IN TREATMENT OF DISEASE 
in mind, and especially when typhoid or non-specific vaccines and protein- 
split products are injected intravenously: 
(a) Over intoxication.—With a profoundly toxic patient we must be 
reasonably certain that he is able to bear the increase in intoxication which 
is probably due to a rapid destruction of typhoid bacilli and poisons from 
the disease foci. 
(b) The possibility of hemorrhages and perforation due to increased peris- 
talsis and increased congestion of the intestinal tract. 
Petersen states (italics are mine): 11 If the pulse is over 100 and there 
are evidences of vasomotor instability, if the patient is profoundly toxic or 
cachectic, if there has been any bleeding—epislaxis, gastro-intestinal, etc.— 
if cyanosis is present, if the disease has continued for several weeks before 
treatment is commenced, or if there is any evidence of pneumonic complications, 
it is not advisable to try non-specific therapy. If given under such conditions 
the clinician must consider the dangers involved and seriously weigh the chances 
for collapse or hemorrhage or perforation, and determine whether or not they are 
overbalanced by possible advantages.” 
TREATMENT OF INFLUENZA 
The bacteriology, immunity, and vaccine prophylaxis of influenza have 
been discussed in Chapter XXXV. From the standpoint of mortality 
and biologic therapy most interest in the last pandemic was commanded 
by the complicating bronchopneumonia. The biologic therapy of influenza 
itself, however, is to be briefly considered, the treatment of bronchopneu- 
monia being discussed later. 
Serum Treatment of Influenza.—Ordinarily influenza is of such short 
duration and mild character that serum therapy is not to be considered. 
Anti-influenzal serum is available, but its use is confined to the treatment 
of influenzal meningitis. During the last pandemic the intravenous injec- 
tion of 20 to 50 c.c. of convalescent human serum was being employed, but 
especially in the treatment of the bronchopneumonia; I shall refer to this 
treatment in more detail under that subject. 
Huff1 has mentioned the beneficial effects following the subcutaneous 
injection of 8 c.c. of human convalescent influenza serum in the treatment of 
influenza of a child. Meille2 reports favorable results from the subcutaneous 
injection of 1 or 2 c.c. of the patient’s own serum (autoserum). 
A number of clinicians, as Kautsky,3 Bettinger,4 Lustig,5 and Crohn6 
have employed subcutaneous and intramuscular injections of diphtheria 
antitoxin in doses of 1 to 10 c.c. and claim good results in uncomplicated 
influenza and cases with pneumonia. The effects were evidently non- 
specific and doubtless would have been observed to the same degree if normal 
horse-serum had been employed. 
Vaccine Treatment of Influenza.—The usual stock mixed vaccine of in- 
fluenza bacilli, streptococci, pneumococci, and staphylococci (see under 
Vaccine Prophylaxis of Influenza) has been employed by some physicians 
for the routine treatment of influenza. When administered early in the 
disease it is believed to reduce the incidence of bronchopneumonia and 
bronchitis and to shorten the course of the general infection. 
Borden and Leopold7 observed that vaccine treatment of 60 cases of 
1 Brit. Med. Jour., May 10, 1919, 575. 
2 Policlinico, 1918, 25, 1055. 
3 Med. Klin., 1919, 15, 69. 
4 Munch, med. Wchn., 1919, 66, 125. 
5 Med. Klin., 1919, 15, 42. 
6 Munch, med. Wchn., 1920, 67, 1521. 
7 U. S. Naval Med. Bull., 1919, 13, No. 4. TREATMENT OF PNEUMOCOCCUS LOBAR PNEUMONIA 
907 
influenza had little immediate effect on the course of the disease, except a 
fall of temperature probably twenty-four hours earlier than control cases; 
5 per cent, of these cases developed pneumonia, while of 221 control case's 
35.3 per cent, developed pneumonia. Wynn has treated 107 cases with 
subcutaneous injections of mixed vaccine, the dose being 80,000,000 to 
100,000,000 of each organism. He states that in simple uncomplicated 
influenza a single injection may abort the disease. The following table by 
Wynn, emphasizes the importance of early treatment, not only upon the 
duration of the disease, but upon the mortality as well: 
Temperature Normal in 
Day of Injection. 
Number. 
Recovered. 
Died. 
Twenty.four 
hours. 
Forty-eight 
hours, 
per cent. 
per cent. 
First 
28 
28 
71.4 
85.7 
Second 
23 
22 
1 
47.8 
56.5 
Third 
22 
20 
2 
50 
72.7 
Fourth 
20 
15 
5 
30 
40 
Fifth 
14 
12 
2 
35.7 
63.5 
Total 
107 
97 
10 
50 
65 
TREATMENT OF PNEUMOCOCCUS LOBAR PNEUMONIA 
Acute lobar pneumonia, with its clear-cut clinical course, unsatisfactory 
and difficult treatment, uncertain prognosis, and high mortality, was one 
of the diseases in which the earliest efforts were directed towrard discovering 
a specific serum therapy. Since the pioneer work of the Klemperers in 1891, 
numerous investigators have prepared serums that have yielded either 
indifferent results or proved beneficial in but a limited number of cases, so 
that there has been no well-established form of specific therapy. 
Recent investigations by Neufeld and Handel1 in Germany, and by 
Dochez,2 Cole,3 and Gillespie,4 in the Rockefeller Institute, have disclosed 
several reasons for the failure of serum therapy in pneumonia, and have 
emphasized the importance of the following factors: 
1. The serum should correspond to the type of pneumococcus causing 
the infection. 
2. The serum must be administered in large doses, and preferably 
intravenously. 
3. To be most effective the treatment should be given as early as possible. 
The investigators in the Rockefeller Institute have divided the pneu- 
mococci causing lobar pneumonia into four main groups, and have worked 
out a method for the rapid identification and classification of the particular 
pneumococcus that is the etiologic factor in a given infection, so that in 
Type I infections the proper administration of the corresponding immune 
serum has yielded encouraging results in the serum treatment of Type I 
pneumonia. These researches are of importance not only in this connection 
but also from the fact that they may have disclosed the reasons for failure 
1Ztschr. f. Immunitatsf., orig., 1909, 3, 159; Arb. a. d. k. Gsndhtsamte; 1910, xxiv; 169; 
ibid., 1910, xxxiv, 293; Berl. klin. Wchn., 1912, xlix, 680. 
2 Jour. Exper. Med., 1912, xvi, 665, 680, 693; Jour. Amer. Med. Assoc., 1913, Lxi, 727. 
3 Jour. Exper. Med., 1912, xvi, 644; Arch. Int. Med., 1914, xiv, 56. 
4 Jour. Amer. Med. Assoc., 1913, lxi, 727; Jour. Exper. Med., 1914, xix, 28. 908 
VACCINES, SERA, BLOOD, PROTEINS IN TREATMENT OF DISEASE 
in the treatment of streptococcus and other infections, and that similar 
studies in these conditions may insure for serum therapy a definite and 
valuable role in the treatment of disease. 
The Nature of Lobar Pneumonia.—The frequency with which the 
Diplococcus pneumonia is found in the local lesion and in severe cases in 
the blood-stream of pneumonia patients, and the more recent experimental 
studies of Wadsworth,1 Meltzer,2 Wollstein and Meltzer,3 Winternitz, Kline 
and Hirchfelder,4 leave little doubt regarding the etiologic relationship of 
this micro-organism to lobar pneumonia. Much still remains to be learned, 
however, regarding the method of infection and the nature of the resulting 
disease. While pneumococci are to be found living in the upper air-passages 
as harmless parasites, it is probable that those causing infection differ 
inherently as regards adaptation or virulence for man. In addition it is 
likely that general resistance is lowered in some more or less peculiar manner, 
and experimental studies in animals, as well as the course of the disease in 
man, suggest most strongly that local changes in the respiratory tract may 
precede the infection, so that a combination of factors, such as the virulence 
of the organisms and the diminished general and local resistance, play a 
part in the production of lobar pneumonia. 
The investigations of Blake and Cecil,5 previously referred to in Chapter 
XXXV, have been very instructive and indicate that pneumococci probably 
penetrate the tissues at or near the root of the lung and reach the pulmonary 
tissues along the framework and lymphatics with the production of pneu- 
monitis. According to this view, lobar pnuemonia is a bronchogenic infec- 
tion, a direct infection, rather than a primary general pneumococcus bac- 
teremia, with subsequent selective localization in the lungs. 
While pneumococci may be found in the blood of the most severe cases, 
the general symptoms are apparently due to intoxication with a poison or 
toxin derived primarily from the pneumococci, and secondarily from the 
exudate in the local lesions. The studies of Rowntree,6 Medigreceanu,7 
and Peabody,8 showing chlorin retention; of Peabody,9 showing progressive 
loss in the oxygen-combining power of the hemoglobin, due to the formation 
of methemoglobin; of Medigreceanu,10 showing a deficiency of oxydase or 
lessened power of the tissues to carry on proper oxidation, and of Neufeld 
and Dold,11 Rosenow,12 Cole,13 Wadsworth,14 Jobling and Strouse,15 Weiss and 
the writer16 indicating the presence of exotoxins and endotoxins within the 
pneumococci—all these support the view that in pneumonia there is well- 
marked intoxication, and this, in addition to the effects of the local pul- 
monary consolidation on the heart, respiration, and nervous system, con- 
stitute the main features of the infection. 
Regarding the mechanism of recovery from pneumonia, there is little 
1 Amer. Jour. Med. Sci., 1904, cxxvii, 851. 
2 Jour. Exper. Med., 1912, xv, 133. 
3 Jour. Exper. Med., 1913, xvii, 353; ibid., 1913, xviii, 548. 
4 Jour. Exper. Med., 1912, xvii, 657; ibid., 1913, xviii, 50. 
6 Jour. Exper. Med., 1920, 31, 403. 
6 Bull. Johns Hopkins Hosp., 1908, xix, 367. 
7 Jour. Exper. Med., 1911, xiv, 289. 
8 Jour. Exper. Med., 1913, xvii, 71. 
9 Jour. Exper. Med., 1913, xviii, 7. 
10 Jour. Exper. Med., 1914, xix, 309. 
11 Berl. klin. Wchn., 1911, xlviii, 1069. 
12 Jour. Infect. Dis., 1911, ix, 190. 
13 Jour. Exper. Med., 1912, xvi, 644; ibid., 1914, 20, 346. 
14 Jour. Exper. Med., 1912, 16, 78. 
15 Jour. Exper. Med., 1913, xviii, 597. 
16 Jour. Immunology, 1918, 3, 395. SERUM THERAPY OF PNEUMOCOCCUS PNEUMONIA 
909 
definite information. The recent studies of Neufeld, Dochez, and Clough 
indicate that antibodies are produced at or about the time of the crisis, and 
that these are probably responsible for the destruction of the bacteria in 
the circulating blood, and, to a greater extent, in the local lesion. In the 
resolution of the local lesion it is probable that ferments play an important 
part. That resolution does not occur earlier may be due to the overbalancing 
of the leukocytic ferments by the antiferments of the serum, and the lytic 
ferments become active only when they reach a point of excess over the 
antiferments, causing a solution of the fibrin, relieving tension, and affording 
an outlet for the exudate. According to Vaughan, the pneumococci may be 
considered as furnishing a ferment that brings about the production of a 
specific antiferment, capable of reacting upon its substratum, the ferment 
and the new bacterial tissue, and causing its destruction by a process of 
solution. Pneumococci in the resolving lesion are probably destroyed by 
leukocidins released through disintegration of leukocytes, by fatty acids, 
and probably by antibacterial substances in the blood. 
While it is true that immunity does not usually follow an attack of 
pnuemonia, and, indeed, the patient is apparently hypersusceptible, it has 
been found experimentally that the antibodies are highly specific for the 
particular organism causing an infection. Reinfection is, therefore, possible 
with an organism belonging to another group, and liability to reinfection 
may be increased because of lowered local and general resistance due to the 
previous attack. 
Thomas1 has recently reported a recurrent attack of Type I pneumonia 
in the same individual within a month after a previous attack with treatment 
by Type I serum; this indicates that an active immunity, if produced at all, 
was of brief duration, and that passive immunity conferred by the immune 
serum was less than a month in duration. 
Serum Therapy of Pneumococcus Pneumonia 
Indications.—The indications for specific serum therapy are mainly 
twofold: first, to destroy any pneumococci present in the blood and in 
the local lesion; or, if the latter is impossible because of mechanical obstacles 
that interfere with the circulation and prevent access of the antibodies to 
the cocci, to at least prevent extension of the lesion by preventing the 
multiplication of organisms at its margin; second, to neutralize the toxins 
produced during the course of the disease. 
It would, of course, be highly desirable to have at our command a serum 
that would cause solution of the local exudate and bring about a crisis and a 
cure. It is hardly reasonable to expect, however, that a serum can be 
produced that will contain digestants for fibrin and leukocytes. The local 
lesion is most likely to be harmful because of the toxic substances that 
emanate from it, and not because so large an area of lung is temporarily 
incapacitated and the heart embarrassed. A serum that will prevent general 
bacteremia, limit the extension of the local lesion, and neutralize the toxins 
while nature is preparing to react upon the exudate with a ferment, is proba- 
bly fulfilling all that may be expected of a specific serum therapy. 
Types of Pneumococci.—Neufeld and Handel have shown that an immune serum 
produced by the injection of a given variety of pneumococci into an animal was not effective 
against all forms of pneumococci. In the Rockefeller Hospital a serum, known as Serum 1, 
prepared by immunizing a horse with a culture obtained from Neufeld, was found to protect 
against only about one-half the types of pneumococci (Group I) studied. By immunizing 
1 Amer. Jour. Med. Sci., 1920, 161, 103. 910 
VACCINES, SERA, BLOOD, PROTEINS IN TREATMENT OF DISEASE 
rabbits to each of the types that were not acted upon by Serum 1, and testing the immune 
serum against all strains by cross-agglutination and by cross-protection experiments, it was 
found that a number of the serums possessed the same properties, thus indicating that their 
respective cultures belonged to the same general group (Group II). By immunizing a horse 
with one of these, Serum 2 was produced. In Group III are placed all the organisms of the 
so-called Pneumococcus mucosus type. In Group IV are included all the varieties of pneu- 
mococci against which Serums 1 and 2 are not effective, and which, from their other proper- 
ties, do not belong in Group III. Animals may readily be immunized to any member of this 
Group IV, and the serum of the immunized animal is protective against the strain used for 
immunization, but in no instance has this serum been found effective against any other 
member of this group or against the organisms of the other groups. While no cultural or 
morphologic differences between the members of Group I, II, and IV exist, it has been found 
possible to group them by the agglutination reaction in exactly the same manner as by pro- 
tection experiments. 
Determining the Type of Pneumococcus.—The method is described on page 286. 
Preparation of Antipneumococcus Serum.—The method commonly employed is 
that developed by Cole and his associates1 in the Rockefeller Institute. Horses are immunized 
with mixed cultures of pneumococci belonging to Types I, II, III, and IV or of Type I alone; 
the Hygienic Laboratory requires a minimum standard of protective value against only 
virulent Type I pneumococci. 
Sound and fairly heavy horses are chosen proved free of glanders. 
Broth cultures of the pneumococcus should be of such virulence that 0.000001 c.c. kills 
mice regularly. Cultures are grown in beef peptone broth +0.3 to 0.5 for twelve to fifteen 
hours and should show 200,000,000 to 300,000,000 bacteria per cubic centimeter. About 
350 c.c. of this broth is thoroughly centrifuged and the cocci suspended in 125 c.c. of sterile 
saline solution, heated in a water-bath at 56° C. for thirty minutes and kept on ice. 
Twenty c.c. of this suspension are injected intravenously every day for six days, followed 
by a resting interval of seven days. 
A fresh culture is now prepared in the same manner and a second series of six daily injec- 
tions given. 
Seven days later three daily intravenous injections of living pneumococci are given, 
the first dose being a suspension of cocci removed from 20 c.c. of the original broth culture, 
the second dose being the cocci from 40 c.c., and the third, the cocci from 80 c.c. of broth. 
If the febrile reactions are severe, smaller doses are given. 
Seven days later the series is repeated giving the cocci from 100, 150, and 200 c.c. of 
broth on the three days respectively. Seven days later a third series of injections are given 
comprised of the cocci from 100, 200, and 300 or 400 c.c. of broth on the three days 
respectively. 
About seven days later a small amount of blood is drawn and the serum tested. If satis- 
factory a large bleeding may be made followed by a rest of three or four days, when a fourth 
series of intravenous injections of cocci from 50, 80, and 100 c.c. of broth are given on the 
three days respectively. After a week the serum is again tested, and if of standard strength, 
bleeding may be done on the tenth day. If the serum is not sufficiently strong, a second series 
of live cultures is given, again keeping the dosage fairly small, never giving more than the 
bacteria from 300 to 400 c.c. of broth and always bleeding about ten days after the last 
injection. 
After large bleedings which are conducted in an aseptic manner, the serum is tested for 
sterility by cultural and animal tests and preserved with 5 c.c. of chloroform per liter. 
In view of the large doses of serum required in the treatment of lobar pneumonia and 
the possible injurious effects of the large amount of horse-serum protein injected, Gay and 
Chickering2 and Chickering3 have endeavored to concentrate the immune serum by adding 
relatively small amounts of the extract of pneumococcus and recovering the resulting pre- 
cipitate. These precipitates have been found to contain practically all the protective anti- 
bodies of the original serum and relatively small amounts of protein as compared with the 
original serum. At the present time, however, this method is not being employed and whole 
serum is generally administered. 
Standarization of Antipneumococcus Serum.—At the present time the protection test 
alone is being employed. The following method, modified after that devised by Neufeld,4 is 
employed in the Rockefeller Institute and the majority of laboratories concerned in the 
preparation of this serum: 
White mice weighing 18 to 22 gm. are employed. The culture of Type I pneumococcus 
1 Monograph No. 7 Rockefeller Institute. 
2 Jour. Exper. Med., 1915, xxi, 389. 
3 Jour. Exper. Med., 1915, xxii, 248. 
4 Ztschr. f. Immunitatsf., orig., 1909, 3, 159; Arb. a. d. k. Gsndhtsamte., 1910, 34, 166; 
ibid., 1910, 34, 293; Berl. klin. Wchn., 1912, 39, 680. SERUM THERAPY OF PNEUMOCOCCUS PNEUMONIA 
911 
should be of such virulence that 0.000001 c.c. of an eighteen-hour broth culture injected 
intraperitoneally kills a mouse in forty-eight hours. 
Various dilutions of the culture are made with peptone broth, using a fresh pipet for each, 
in such manner that 0.5 c.c. carries 0.2, 0.1, 0.01, 0.001, 0.0001, 0.00001, 0.000001 c.c. of the 
original broth culture. 
For the test 2 c.c. of serum are diluted with 3 c.c. of broth so that 0.5 c.c. carries 0.2 c.c. 
undiluted serum. 
When injecting the animals, 1 c.c. of diluted serum and 1 c.c. of the diluted culture are 
taken up into a 2 c.c. Luer or Record syringe, mixed, and 1 c.c. injected intraperitoneally into 
each of two mice. It is well to start with the weakest culture first, working up to the largest 
amount, separate syringes being used for each, or the one syringe washed out with hot sterile 
water and cooled between the injections. 
Three control mice receive injections of 0.00001, 0.000001, and 0.0000001 c.c. of the cul- 
ture alone. 
The mice are properly marked and observed for at least forty-eight hours. 
To be satisfactory a serum in dose of 0.2 c.c. should protect against at least 0.1 c.c. of the 
culture of the above-mentioned virulence, that is, against about 100,000 fatal doses of culture. 
Action of Antipneumococcus Serum.—The curative and protective 
value of this serum depends mainly upon bacteriolysins, bacteriotropins, 
and antitoxins. The first are readily demonstrated in protection tests 
and also in pneumonic patients when pneumococci in the blood-stream 
are destroyed. Bacteriotropins may likewise be demonstrated experi- 
mentally and in the blood of patients if the corresponding organism 
is used in the tests (Neufeld, Strouse). The antitoxic properties are shown 
clinically and also in viiro by neutralization of the hemotoxic poison obtained 
by dissolving pneumococci in bile. As shown recently by Bull,1 the agglu- 
tinins in the serum may play a very active role by producing an agglutination 
of the cocci in the blood-stream followed by phagocytosis. 
According to Avery2 the antibodies are associated or combined with 
that fraction of the globulins precipitated by 38 to 42 per cent, saturation 
with ammonium sulphate. 
Types of Pneumococci in Relation to Serum Treatment and Mortality 
of Pneumonia.—The value of serum treatment has been established for 
Type I pneumonias, is doubtful in Type II, and is without beneficial effects 
in Type III and Type IV infections. 
The incidence of these types of pneumococci in pneumonia is indicated 
by the investigations of Cole and Longcope in New York, and Richardson 
in Philadelphia; since the influenza epidemic the Type IV strains have been 
more abundant than the combined fixed Types I, II, and III. 
INCIDENCE OF TYPES OF PNEUMOCOCCI IN LOBAR PNEUMONIA 
Cole. 
Longcope. 
Richardson. 
Per cent. 
Per cent. 
Per cent. 
Type I 
33 
23 
31 
Type II 
32 
21 
20 
Type III 
9 
14 
6 
Type IV 
Other bacteria, 6 per cent. 
20 
40 
43 
MORTALITY IN LOBAR PNEUMONIA (NO SERUM) 
Cole. 
Longcope. 
Richardson. 
Per cent. 
Per cent. 
Per cent. 
Type I 
25 
12.5 
30 
Type II 
29 
72.7 
25 
Type III 
45 
85.7 
50 
Type IV 
12.5 
23.8 
12 
1 Jour. Exper. Med., 1915, 22, 466. 
2 Jour. Exper. Med., 1915, 21, 133. 912 
VACCINES, SERA, BLOOD, PROTEINS IN TREATMENT OF DISEASE 
Type I pneumonias, therefore, constitute about 30 per cent. The 
mortality, however, has varied greatly at different times in different parts 
of the country, but in general terms ranges from 15 to 30 per cent. 
Cole states that serum should be given only to Type I cases; this requires, 
of course, a delay of twelve to twenty-four hours or longer for the necessary 
typing. Park has always stated that in severe cases when a delay in typing 
is unavoidable, that a dose of Type I or polyvalent serum should be given as 
soon as possible and the later injections only in case the infection proves to 
be Type I infection. In routine practice this means the administration of 
serum to some cases other than Type I infections, but appears justifiable in 
the treatment of severe infections when facilities are poor for the necessary 
preliminary typing. 
Polyvalent sera prepared by biologic laboratories doubtless contain 
small amounts of antibodies for Types II and III in addition for Type I, 
but it is doubtful that they are of benefit in the treatment of any but Type I 
pneumonias, except for the possible beneficial results from non-specific 
effect's. 
Administration of Antipneumococcus Serum.—To obtain the best results 
the serum should be given as early as possible and intravenously. In young 
children, when the giving of an intravenous injection is quite difficult or 
impossible, the muscles of the buttocks should be substituted. Certainly 
small doses given subcutaneously are almost devoid of effect. The procedure 
in use in the Rockefeller Institute1 consists in injecting 0.5 c.c. of serum 
subcutaneously to discover if hypersensitiveness exists and to produce 
anti-anaphylaxis. About one hour later from 90 to 100 c.c. of the serum, 
diluted one-half with salt solution, are injected intravenously. The condi- 
tion of the patient serves as a guide in the later treatment. 
Cole2 has recently stated that “this dose should be repeated every eight 
hours until the fall of temperature and amelioration of symptoms indicate 
that the infection has been overcome. The serum injected should be at 
body temperature, and it should be injected very slowly. The total amount 
required in the average case is from 200 to 300 c.c., though in severe cases, 
treated late in the disease, it may be necessary to employ much larger 
amounts, even as much as 1000 c.c.” 
It has been shown experimentally that, in the presence of a maximum 
degree of infection, no amount of serum, however large, is effective. This 
suggests that the body must furnish a second substance to act with the 
antibodies in the serum, and indicates the early administration of serum 
before the infection has reached too extreme a grade. I would also suggest 
that the body may be deficient in bacteriolytic complements, and that the 
effect of a serum may be enhanced by adding fresh sterile human serum— 
say 5 c.c. to each 100 c.c. of immune serum—just prior to administration. 
Meyer3 has recently shown that the opsonins in antipneumococcus 
serum may be reactivated by the addition of small amounts of fresh human 
serum. 
Results in the Serum Treatment of Pneumonia.—It is hardly necessary 
to review the numerous reports that have been made in past years, because 
in most instances the serum was administered subcutaneously and in too 
small doses to be of value, even granting that it contained antibodies for 
the particular infection. Prior to the investigations of recent years showing 
the serologic grouping of pneumococci, horses were immunized with a 
1 Avery, Chickering, Cole and Dochez, Monograph No. 7, Rockefeller Inst., 1917. 
2 Jour. Amer. Med. Assoc., 1921, 76, 111. 
3 Jour. Infect. Dis., 1920, 27, 82. SERUM THERAPY OF PNEUMOCOCCUS PNEUMONIA 
913 
variety of strains and the sera were polyvalent. Furthermore, the sera 
were not very potent and in view of the administration of small amounts, 
the results were not encouraging. 
At the present time the serum treatment of pneumonia is practically 
confined to the Type I infections. Richardson,1 in an excellent investigation 
and review of the literature, has shown that roughly about one-fourth of all 
cases of lobar pneumonia are due to Type I pneumococci. The mortality 
of this group treated without serum is from 25 to 30 per cent, or higher. 
Cole reports that “among 195 cases treated with serum in the Hospital of 
the Rockefeller Institute, only 18 deaths have occurred, a case mortality 
rate of but 9.2 per cent. Reports of 300 additional cases treated with serum 
have been collected from the literature, making a total of 495 cases. The 
case mortality rate in these 495 cases has been 10.5 per cent.” 
Cecil and Blake2 have shown the specific curative effects of Type I serum 
in the treatment of severe experimental Type I pneumonias of monkeys. 
All treated animals recovered; all untreated controls died. Frequent injec- 
tions were found important, and when serum treatments were instituted late 
in the disease, the injections were required over a longer period of time. 
Normal horse-serum was observed to be without beneficial effects. 
Thomas3 has recently reported upon the serum treatment of 50 Type I 
cases in New York City. The course of the disease appeared to have been 
shortened in 4 cases; in 8 cases the improvement appeared to have been 
but transitory, while in the remaining 38 the duration and outcome of the 
disease do not appear to have been demonstrably affected by the serum. 
Aside from an apparent influence upon mortality the general effects of the 
serum are sometimes good. The blood of the patient becomes sterile follow- 
ing the injection of serum. The progression of the local lesion in the lung is 
usually arrested. The subjective and objective symptoms of the disease are 
lessened. Following the injection of serum most patients seem to feel better, 
and in a number of them there is an apparent lessening in the degree of intoxi- 
cation. While in no case is one injection sufficient to bring about a crisis, 
in all except the fatal cases the serum has apparently an ultimate favorable 
effect, lowering the temperature and shortening the course of the disease. 
The sum total of experience indicates, however, that the serum treatment 
of lobar pneumonia is not as promising as the earlier reports indicated; that 
it should be confined to Type I pneumonias and that the serum must be 
given early, in large amounts and by intravenous injection. 
The procedure is not wholly without danger. Thomas states that 2 
deaths from anaphylaxis are reported in the literature. It is necessary to 
give the serum with considerable caution. The skin tests for anaphylaxis 
and the possible reactions, anaphylactic, so-called thermal and serum disease, 
have been considered in preceding chapters. 
Kyes’ Antipneumococcus Serum.—Kyes4 has described a method of 
preparing antipneumococcus serum by the immunization of chickens with 
massive intraperitoneal doses of virulent penumococci. As is well known 
chickens possess a high degree of natural immunity to pneumococci, and 
both Kyes and more recently Bull and McKee,5 have found that normal 
chicken serum is capable of protecting mice and guinea-pigs against infec- 
tion. Bull and McKee found that protective substances were present for 
1 Jour. Lab. and Clin. Med., 1919, 4, 484. 
2 Jour. Exper. Med., 1920, 32, 1. 
3 Jour. Amer. Med. Assoc., 1921, 77, 2101. 
4 Jour. Amer. Med. Assoc., 1911, 56, 1878. 
5 Amer. Jour. Hyg., 1921, 1, 284. 914 
VACCINES, SERA, BLOOD, PROTEINS IN TREATMENT OF DISEASE 
each serologic type of pneumococcus and that they may be selectively 
removed from the serum by adsorption. 
Kyes1 administers this serum intravenously in doses of 2.5 c.c. irrespec- 
tive of the type of pneumonia. Usually one injection is given daily until 
the temperature remains below 100° F. Sometimes two injections are given 
per day during the first few days of especially sick individuals. The average 
number of injections is three. 
Of 538 cases not treated with serum the mortality was 45.3 per cent. 
Of 115 cases under identical conditions treated with his serum, the mortality 
was 20.8 per cent, or a reduction of more than one-half. 
Gray2 has reported upon the use of this serum in the treatment of lobar 
pneumonia in Camp Grant. The doses were increased from 2.5 c.c. once or 
twice daily to 5 to 10 c.c. twice daily. In a few cases as much as 30 c.c. was 
given daily. Commonly a total of 60 to 90 c.c. was given in cases that 
recovered. 
In one group of 322 cases of typical lobar pneumonia the mortality was 
7.7 per cent. A second group of 234 cases of epidemic pneumococcus-bron- 
chopneumonia the mortality was 16.7 per cent, as compared with 53.6 per 
cent, of 1684 control cases. In a third group of 118 cases of a typical pneumo- 
coccus pneumonia of lesser virulence, the mortality was 4.3 per cent. 
Treatment of Pneumococcus Pneumonia with Huntoon’s Antibody 
Solution.—Huntoon and his co-workers3 have recently described a solution 
of antibodies for pneumococci Types I, II and III combined, secured from 
polyvalent antipneumococcus serum by dissociating antigen-antibody com- 
plexes by various means (for a more complete description, see page 150). 
The resulting solution of antibodies is remarkable because it yields negative 
or poorly defined reactions for proteins, does not regularly sensitize guinea- 
pigs, and proves protective for mice inoculated with virulent pneumococci. 
From the experimental standpoint it would appear worthy of trial in the 
treatment of lobar pneumonia, but in the experience of Cecil the intravenous 
injection sometimes elicits very severe reactions. Recently Cecil and 
Larsen4 have reported the following results in the treatment of pneumonia 
with this polyvalent antibody solution: 
(a) In 424 cases of pneumococcus pneumonia treated with pneumo- 
coccus antibody solution, the death-rate was 21.4 per cent. A control 
group of 410 cases in the same institution showed a death-rate of 28.3 per 
cent. 
(b) Pneumococcus antibody solution produces its most striking effect 
in pneumococcus Type I pneumonia. In a series of 156 treated cases 
the death-rate was 13.3 per cent.; while a control series of 162 cases showed 
a death-rate of 22.2 per cent. A definite but less marked effect was observed 
in cases of pneumococcus Types II and IV pneumonia which wTere treated 
with antibody solution. The antibody solution had no effect whatever 
on the death-rate in pneumococcus Type III pneumonia. 
(c) The death-rate of streptococcus pneumonia was not favorably 
influenced by antibody solution treatment. 
(d) In the series of patients with pneumococcus pneumonia treated with 
antibody solution, 28.8 per cent, recovered on or before the fifth day. In 
the control series, only 7.9 per cent, recovered on or before the fifth day. 
(e) There were forty-four severe complications in the series of 424 
1 Jour. Med. Research, 1918, 38, 495. 
2 Amer. Jour. Med. Sci., 1920, 159, 885. 
3 Jour. Immunology, 1921, 6, 117, 123, 185. 
4 Jour. Amer. Med. Assoc., 1922, 79, 343. VACCINE TREATMENT OF PNEUMOCOCCUS PNEUMONIA 
915 
pneumococcus pneumonias treated with antibody solution, while the control 
series of 410 pneumococcus pneumonias showed fifty-four severe complica- 
tions. 
The solution may be given subcutaneously or intravenously—the latter 
has yielded the best results, but is likewise more likely to produce severe 
reactions. 
The initial dose by subcutaneous injection should be at least twice the 
intravenous dose or about 50 c.c. and by intravenous injection from 25 to 
50 c.c. for adults. The usual procedure is to administer doses at twenty- 
four-hour intervals as long as the temperature remains above 100° F. In 
cases where the temperature has not dropped promptly, a second dose should 
be given within the first twenty-four hours. 
Subcutaneous injection produces a local reaction; intravenous injection 
almost always produces a chill and sharp rise of temperature in from thirty 
to forty minutes, the reaction being similar to the “foreign protein reaction.” 
Severe reactions of this sort have generally occurred after the first injection 
given on the fifth or later days of the disease. 
Best results have followed the early administration of the solution, that 
is, as soon as the diagnosis has been made. Typing of the pneumococcus 
is not required except as a matter of interest and record. Cases first treated 
on or after the fifth day should be injected subcutaneously rather than 
intravenously. In children twelve years or less of age this treatment may be 
omitted unless the disease is particularly acute. Further experience is 
required before a final opinion may be expressed of the practical therapeutic 
value of this solution. 
Vaccine Treatment of Pneumococcus Pneumonia 
Vaccines by Subcutaneous Injection.—While pneumococcus vaccine has 
been employed by a large number of physicians in the treatment of lobar 
pneumonia, the curative value of this procedure is still doubtful. Stoner,1 
Allen,2 Morgan,3 Harris,4 Lyons,5 Wynne,6 and others have reported favorably 
upon the vaccine treatment of pneumonia as based upon mortality statistics 
and clinical impressions; Charteris,7 on the other hand, failed to observe 
beneficial results. Rosenow and Hektoen8 have prepared a vaccine of 
partially autolyzed pneumococci which they believe proved of value in the 
treatment of pneumonia. 
Recently Rosenow9 has described the preparation and administration 
of this vaccine and believes that it favorably influences the mortality and 
complications of pneumonia. The vaccine is prepared by growing virulent 
strains of pneumococci of the different types in glucose broth from eighteen 
to twenty-four hours, centrifugalizing, and suspending the sediment in salt 
solution so that 1 c.c. contains about 15,000,000,000 pneumococci. The 
suspension is shaken with ether and then incubated at 37° C. until autolysis 
has occurred, that is, until 95 per cent, of the organisms have become Gram- 
negative, and 5 c.c. of the suspension produce few or no toxic symptoms in 
guinea-pigs. The dose of this vaccine for adults is 1 c.c. administered 
1 Amer. Jour. Med. Sci., 1911, cxli, 186. 
2 Vaccine Therapy and Opsonic Treatment, Blakiston, Philadelphia, 1913. 
3 Proc. Roy. Soc. Med., 1910, iii, Suppl. 65. 
4 Brit. Med. Jour., 1909, 1, 1530. 
5 Brit. Med. Jour., 1913, 1, 992. 
6 Brit. Med. Jour., 1915, 1, 458. 
7 Glasgow Med. Jour., 1912, lxxvii, 19. 
8 Jour. Amer. Med. Assoc., 1913, lxi, 2203. 
8Jour. Amer. Med. Assoc., 1918, 70, 759. 916 
VACCINES, SERA, BLOOD, PROTEINS IN TREATMENT OF DISEASE 
subcutaneously every day until the temperature reaches normal. Putnam1 
has reported favorably upon the use of this vaccine. Cecil,2 however, be- 
lieves that great caution is to be observed in recommending the treatment of 
pnuemonia by vaccines of whatever modification; and that it is difficult to 
evaluate the results of vaccine therapy, reported in this disease, because so 
many factors may possibly determine the question of death or recovery. 
If vaccine therapy is employed it would appear necessary to administer 
it as early as possible. For this reason stock polyvalent vaccines are re- 
quired. The dose and administration of the Rosenow-Hektoen vaccine is 
given above. Ordinary saline vaccines may be prepared in such manner 
that each cubic centimeter contains 250,000,000 of each of the four types; 
the first dose may be 0.1 c.c. by subcutaneous injection followed by daily 
or every other day injections in gradually increasing amounts until two to 
four doses have been given. Cohen,3 who is an ardent believer in the use of 
Yaccines as part of the treatment of lobar pneumonia, employs a mixed stock 
vaccine of pneumococci Types I, II, and III, streptococci, staphylococci, 
and M. catarrhalis, containing a total of 800,000,000 per cubic centimeter. 
The first dose is about 0.2 c.c. (125,000,000) before the case is typed. Three 
days later 0.4 c.c. (250,000,000) are given and repeated at three-day intervals 
ip gradually increasing amounts. The usual case receives two or, at most, 
three doses, the injections being stopped when resolution of the pneumonia 
takes place. 
• In delayed resolution vaccine therapy may prove of distinct aid. It is 
my custom to prepare autogenous vaccines of pneumococci, streptococci, 
and staphylococci from the sputum, using equal proportions of each organism 
in such manner that each cubic centimeter contains a total of 2,000,000,000. 
The first dose is 0.1 c.c. and subsequent injections are given at three- to five- 
day intervals in gradually increasing amounts. 
Vaccines by Intravenous Injection.—Rosenow and Falls4 have injected 
vaccines of autolyzed pneumococci intravenously in amounts sufficient to 
provoke reactions, usually 10 to 20 c.c. The temperature generally dropped 
2. to 6° F., followed by a rise to its former level or slightly lower. In a series 
of 35 cases, chills were noted in 12. Since it was impossible to control the 
occurrence and degree of the chills and since the beneficial effects were 
transient and the course of severe cases uninfluenced, this method was 
abandoned, as the possible beneficial effects were more safely secured by 
subcutaneous injections. 
Non-specific Vaccine Therapy of Pneumococcus Pneumonia.—Miller5 
has summarized his experience with 15 cases as follows: 
“Fifteen consecutive patients with lobar pneumonia entering Cook 
County Hospital were treated by a single intravenous injection of typhoid 
vaccine. The dosage used was 30,000,000, the minimum amount required 
to give a chill. All reacted by a rise in temperature and a leukocytosis. 
In 9 patients the vaccine did not modify the course of the disease. In 6 
the patient was detoxicated following the injections. The pulse, temperature, 
and respiration returned to normal, the cough and pleural pain subsided, 
and the patient stated that he felt much better. In 3 of the 6 cases the 
improvement was temporary, as after the lapse of from twelve to twenty- 
four hours the symptoms returned with unmodified severity. In 3 cases 
1 Jour. Amer. Med. Assoc., 1915, 64, 2160. 
2 Jour. Amer. Med. Assoc., 1921, 76, 178. 
3 New York Med. Jour., June 8, 1918; Penna. Med. Jour., 1919, 12, 506; Jour. Amer. 
Med. Assoc., 1919, 73, 1741. 
4 Jour. Amer. Med. Assoc., 1916, 67, 1929. 
8 Jour. Amer. Med. Assoc., 1920, 74, 1598. INFLUENZAL AND STREPTOCOCCUS BRONCHOPNEUMONIA 
917 
the detoxication was permanent; however, the patients had a moderate 
temperature for from three to four days, to the time at which the crisis would 
normally appear. They were, however, entirely free from evidence of 
intoxication. There was no relation between the severity of the chill, the 
temperature reaction and degree of increased leukocytosis, and the beneficial 
results of the vaccine.” 
As stated by Petersen these results suggest a temporary detoxication 
that leaves the general course of the disease process unaltered. Very proba- 
bly the effects are purely non-specific in which antibodies play no part, 
inasmuch as the disease has too short a period of incubation to permit the 
production of antibodies in time to materially influence the course of the 
disease during the first week. 
Cowie1 has reported favorably upon the results of intravenous injection 
of typhoid vaccine in the treatment of influenzal pneumonia, providing the 
injection is given before the fourth day, finding that the acute symptoms 
may subside in from one to three days. Cases showing undoubted evidence 
of advanced myocardial insufficiency or acute endocarditis should not 
receive these injections. ’ 
Leukocytotherapy of Pneumococcus Pneumonia.—Hess2 has reported 
the treatment of 7 cases of lobar pneumonia by the daily subcutaneous 
injection of 10 c.c. of leukocytic extract. In every instance the injection 
was followed by a drop in temperature and a decrease of the toxic symptoms, 
although the disease was not materially shortened. Floyd and Lucas3 
have reported favorably upon the results of this treatment in 41 cases of 
lobar pneumonia, and particularly the beneficial effect upon the pulse and 
respiration. Lambert4 has likewise observed good results in a small series 
of cases. Williams and Youland,5 however, were unable to see any bene- 
ficial effects in a series of 7 cases receiving amounts of extract varying from 
20 to 320 c.c. by subcutaneous injection per day. The temperature, leu- 
kocyte count, general condition of the patient, course, and termination of 
each case did not appear to be influenced. 
It would appear that the direct bactericidal activity of leukocytic extracts 
is feeble and that beneficial effects are to be ascribed to non-specific agencies. 
For this purpose 10 to 15 c.c. may be administered per day by subcutaneous 
injection. 
TREATMENT OF INFLUENZAL AND STREPTOCOCCUS BRONCHOPNEUMONIA 
Etiology of Bronchopneumonia.—The bacteriology of bronchopneumonia 
is quite complex. The disease is caused principally by a group of organisms 
that are not as invasive as pneumococci, but gradually creep down the 
bronchial tree and involve patches of lung by direct involvement of the 
bronchioles. For this reason bronchopneumonia is usually a complication 
developing in diseases of the upper respiratory tract as whooping-cough, 
influenza, measles, grip and the “common cold,” and especially acute 
bronchitis. From the etiologic standpoint the influenza bacillus, strepto- 
cocci, Type IV pneumococci, Staphyloccus aureus and Micrococcus 
catarrhalis are of most importance. 
Influenzal Pneumonia; Bacterium Pneumosintes.—The pneumonia of 
influenza may be caused by a variety of micro-organisms and mainly Ba- 
cillus influenzae, streptococci, and staphylococci. The pneumonic lesions 
1 Jour. Amer. Med. Assoc., 1919, 72, 1117. 
2 Jour. Med. Research, 1908, 14, 325. 
3 Jour. Med. Research, 1909, 21, 223. 
4 Amer. Jour. Med. Sci., 1909, 137, 506. 
6 Jour. Med. Research, 1914, 31, 391. 918 
VACCINES, SERA, BLOOD, PROTEINS IN TREATMENT OF DISEASE 
have seldom revealed B. influenzae in pure culture. It is highly probable 
that in influenza resistance is greatly lowered to secondary infections with 
these other micro-organisms. 
Direct evidence of this has been apparently furnished by the studies of 
Olitsky and Gates.1 These investigators have observed that the intra- 
tracheal injection of rabbits with the nasopharyngeal secretions of influenza 
cases collected within thirty-six hours after the onset of the disease, pro- 
duces severe injury of the lung tissue with edema and hemorrhages and 
favoring the production of severe pneumonia by ordinary bacteria. Fil- 
trates of these secretions have produced similar results and anaerobic cul- 
tures of the filtrates in the Smith-Noguchi ascites broth kidney medium 
have developed a minute bacillary body to which has been given the name 
Bacterium pneumosinles— a bacterium that injures the lung. 
Rabbits recovering from intratracheal injections of this filter-passing 
■organism or of nasopharyngeal secretions of influenza cases are apparently 
immune to reinoculation. Injection of rabbits with cultures results in the 
production of specific antibodies, and recently antibodies for Bacterium 
pneumosinles were found in the blood of 17 of 19 human cases of influenza. 
These studies indicate that Bacterium pneumosinles may be the primary 
etiologic agent of influenza, but further experience is required before a 
final statement may be made. 
Characteristically this is a broncho- or nodular hemorrhagic pneumonia. 
As shown by Blake and Cecil2 experimentally in monkeys the influenza 
bacillus is capable of setting up a bronchitis and broncholitis with peri- 
bronchial consolidation—a true bronchopneumonia with hemorrhage and 
edema, sometimes succeeded in the later stages by emphysema and bron- 
chiectasis. In human infections, however, other bacteria are usually present 
and it is often difficult to determine just what part of the process is due to 
the influenza bacillus. 
According to Blake and Cecil, the influenza bacillus travels down the 
bronchial tree into the bronchioles and involves the neighboring tissues by 
direct contiguity, producing patches of inflammation probably by means of 
exogenous toxins. The bacillus does not appear to enter the lymphatic 
channels as does the pneumococcus with subsequent wide-spread distribu- 
tion and more general involvement of the lung tissue. 
Streptococcus Pneumonia.—This pneumonia may develop in the course 
of influenza, measles, and other infections; apparently it may also occur 
spontaneously and without predisposing causes. According to Cecil3 it is 
the organism most frequently found in bronchopneumonia and may be 
classified into the “hemolytic” and “viridans” group. 
Streptococcus hemolyticus produces a severe form of lobular pneumonia, 
frequently complicated by multiple abscesses and empyema with a mor- 
tality of about 50 per cent. 
Streptococcus viridans produces a mild pneumonia with a mortality of 
less than 10 per cent. 
The experiments of Blake and Cecil have indicated that Streptococcus 
hemolyticus pneumonia is produced in much the same manner as pneumo- 
coccus pneumonia, that is, by entering the lymphatics at the root of the 
lung and gradually involving the lung tissue, as well as by extension along 
the connective-tissue framework. 
‘Jour. Exper. Med., 1921, 33, 125, 361, 373, 713; ibid., 1921, 34, 1; ibid., 1922, 35, 1, 
553, 813; ibid., 1922, 36, 685. 
2 Jour. Exper. Med., 1920, 32, 401. 
3 Amer. Jour. Med. Sci., 1922, 164, 58. INFLUENZAL AND STREPTOCOCCUS BRONCHOPNEUMONIA 
919 
Staphylococcus Aureus Pneumonia.—This pneumonia has been de- 
scribed by Chickering and Park1 as a complication of influenza during the 
recent pandemic. Of 1409 cases of pneumonia in Camp Jackson during 
1918, 153 were found to be associated with this organism. The striking 
clinical features of the disease were the “peculiar cherry-red indigo-blue 
cyanosis, the fulminating course, the lack of definite signs of consolidation 
of the lungs, the dirty salmon-pink purulent sputum, and usually a leuko- 
penia. Pathologically, if the disease is of long enough duration, it is charac- 
terized by innumerable abscesses in the lungs. The prognosis is always 
grave, though a few patients occasionally survive.” Only 2 patients in 
their series recovered. 
Treatment of the Bronchopneumonia of Influenza with Antipneumo- 
coccus Serum.—Ordinarily serum has not been employed in the treatment 
of bronchopneumonia. During the pandemic of influenza, however, the 
incidence and mortality of pneumonia was so high as to lead to numerous 
trials of serum treatment. 
At Camp MacArthur pneumococci, and especially Type IV, was found 
so frequently in 440 cases (85.8 per cent.) that polyvalent pneumococcus 
serum was employed in treatment. Medalia and Schiff2 report that of the 
serum treated cases, omitting those that received serum in a moribund 
condition, the mortality among 286 cases was 7.6 per cent., while among 
277 cases not treated with serum, the mortality was 24 per cent. It was 
believed that the serum treatment had been of value. 
Treatment of the Bronchopneunomia of Influenza with Human Con- 
valescent Serum and Blood.—Toward the close of the pandemic of influenza 
considerable attention was being given the treatment of pneumonia by 
injections of serum and citrated blood from convalescent human beings. 
McGuire and Redden3 in 1918 reported upon the successful treatment 
of a group of 37 cases by this method. Donors were selected from among 
adult convalescent cases about seven to ten days after subsidence of fever 
and about 800 c.c. of blood removed from each under aseptic conditions. 
Wassermann and blood grouping tests were then conducted. Treatment 
was conducted by the intravenous injection of 75 to 125 c.c. of homologous 
serum followed by reinjections at intervals of eight to sixteen hours until 
the patient improved; the average amount of serum for each case was 300 c.c. 
Sanborn4 treated 101 cases of bronchopneumonia in influenza with 
convalescent human serum. The dosage for adults was 100 c.c. and for 
children 50 c.c. From one to six doses were given each patient according 
to indications, at eight- to twenty-four- hour intervals. Those patients only 
were used as donors who had recovered from an undoubted bronchopneu- 
monia which had followed epidemic influenza. Of the 67 patients who re- 
covered, 59 required one or two doses; 7, three, and 1, six. The maximum 
dosage of serum in a given case in the recovered group was 700 c.c. Of 
34 who died, 17 who received one dose did not live a sufiicient length of time 
to receive a second. The remaining 17 received two to four doses and 2 
then received 500 c.c. and 700 c.c. respectively as the total dosage. Re- 
actions following serum were conspicuous by their absence. There were 
practically no complications in recovered cases. The mortality rate in this 
series of 101‘serum treated cases was 33.6 per cent. In 184 cases of influenzal 
pneumonia not treated by serum, the mortality was 21 per cent. Sanborn 
1 Jour. Amer. Med. Assoc., 1919, 72, 617. 
2 Jour. Amer. Med. Assoc., 1918, 71,1821; Boston Med. and Surg. Jour., March 20, 1919. 
3 Jour. Amer. Med. Assoc., 1918, 71, 1311; Amer. Jour. Public Health, 1918, 8, 741. 
4 Boston Med. and Surg. Jour., 1920, 18, 171. 920 
VACCINES, SERA, BLOOD, PROTEINS IN TREATMENT OF DISEASE 
maintains that the mortality rate for this series was not a measure of the 
value of convalescent serum, but was an index of the severity of the disease 
in the group as a whole. Convalescent serum appeared to have value 
when it was administered during the first three days of the pneumonia. 
Its value rapidly decreased when administered after the third day of the 
pneumonia as indicated by rapidly increasing mortality rates, according 
as administration was delayed from day to day. The time factor, that 
is, the period between onset and day of treatment, had a close rela- 
tion to the mortality rate and seemed largely to determine success or 
failure. 
Francis, Hall and Gaines1 have also expressed the opinion that intra- 
muscular injections of 50 to 100 c.c. of convalescent serum proved helpful 
in the treatment of the bronchopneumonia of influenza. 
Maclachlan and Fetter2 have treated 54 cases of bronchopneumonia of 
influenza with intravenous injections of 75 to 100 c.c. of blood from non- 
syphilitic convalescent cases; the mortality was 27 per cent. This amount 
of blood was removed into 40 c.c. of 2 per cent, sodium citrate in saline 
solution and injected without attention to compatability of the bloods of 
donor and recipient. Reactions of chills and fever occurred in some instances, 
but no other ill effects were observed. Ross and Hund3 have also treated 
a group of cases with intravenous injections of 250 to 500 c.c. of citrated 
blood taken from compatible donors among convalescent cases of influenza; 
the mortality rate was 21.4 per cent, as against 42.8 per cent, among cases 
symptomatically. 
Treatment of the Bronchopneumonia of Influenza with Antistrepto- 
coccus Serum.—Rosenow4 has employed intravenous injections of 25 to 
100 c.c. of monovalent antistreptococcus serum prepared by the immuniza- 
tion of horses with streptococci recovered from cases of influenza during 
the epidemic. Of 12 patients treated by him, all but one were critically 
ill at the time of serum treatment. Five recovered and 7 died. Three of 
the patients who died were practically moribund at the time of the serum 
treatment and good effects could scarcely be expected. Two others that 
died showed green-producing streptococci immunologically different from 
the strain with which the serum was prepared, and in 2 hemolytic strepto- 
cocci caused death. In these cases also improvement could not be expected. 
In all cases in which specific agglutination was obtained marked improve- 
ment followed the injection of the serum, and in no case were good effects 
noted at a time when agglutination tests were negative. 
Edgewood5 and Hughes6 have also reported favorable results from the 
use of antistreptococcus serum. 
Treatment of the Bronchopneumonia of Influenza with Non-specific 
Agents.—Intramuscular injections of 5 to 20 c.c. of sterile milk for one to 
three injections have been employed by Von den Velden,7 Munzes and 
Ptitz,8 Patschkowski,9 and others; satisfactory results are claimed. Bor- 
chardt and Ladwig10 have employed intravenous injections of small amounts 
of 5 to 10 per cent, solutions of sodium chlorid in the treatment of 98 cases, 
and remark particularly on the detoxication of severe cases. Ludke has 
treated 10 cases with intravenous injections of 1 to 3 c.c. of 10 per cent, 
solutions of albumose. In 5 of these an immediate improvement was noted; 
1 Military Surgeon, 1920, 67, 177. 
2 Jour. Amer. Med. Assoc., 1918, 71, 2053 
3 Jour. Amer. Med. Assoc., 1919, 72, 640. 
4 Jour. Infect. Dis., 1920, 26, 614. 
5 Brit. Med. Jour., 1918, 2, 515. 
6 Lancet, 1919, 2, 782. 
7 Deutsch. med. Wchn., 1918, xliv, 1446. 
8 Munch, med. Wchn., 1919, 66, 227. 
9 Munch, med. Wchn., 1916, 66, 531. 
10 Berl. klin. Wchn., 1920, 57, 1123. TREATMENT OF DIPHTHERIA 
921 
in 2 cases pneumonia developed with empyema, and in 3 the temperature 
came down only after a number of injections. 
Subcutaneous injections of turpentine (0.1 to 1 c.c. of 20 per cent, solutions 
in sterile olive oil) for the production of fixation abscesses have been claimed 
by Taillens,1 Netter,2 Pehu and Pillon,3 Hodel,4 Probst,5 and others to yield 
good results probably by the production of leukocytosis in influenza and 
particularly its bronchopneumonia in children. 
Serum (Antitoxin) Treatment of Diphtheria 
TREATMENT OF DIPHTHERIA 
The discovery of diphtheria antitoxin and its use in the treatment of 
this infection constitute one of the triumphs of modern medicine. 
Twenty-five years ago diphtheria was one of the most dreaded of diseases, 
accompanied ordinarily by a mortality of at least 30 per cent., while the 
loss of life from the laryngeal form of the disease, particularly after tra- 
cheotomy, was simply appalling. 
Shortly after Roux and Yersin (1888) had demonstrated that the symp- 
toms of diphtheria were due largely to a soluble poison or toxin secreted 
by the bacilli, Ferran, and later Frankei and Brieger (1890), undertook 
experiments in active immunization against diphtheria. About the same 
time von Behring discovered the antitoxin, and in a series of extensive 
researches with Wernicke he established experimentally its prophylactic 
and therapeutic value in diphtheria. The first attempt to apply this dis- 
covery to the cure of this infection of the human being was made in von 
Bergmann’s clinic (1891). The results, while encouraging, were not alto- 
gether satisfactory, owing largely to the fact that the serums were weak and 
the doses given too small. The discovery, however, resulted in creating an 
extraordinary stimulus to researches in immunity, and during the following 
two years more powerful serums were prepared, so that in 1896 a marked 
drop in the mortality of diphtheria was apparent in those places where the 
antitoxin was being used. 
Since then diphtheria antitoxin has been the means of saving countless 
thousands of lives, and the treatment of diphtheria, instead of being a 
reproach to medicine, has become the model of what the scientific treatment 
of an infectious disease ought to be. Statistics and the individual expe- 
riences of those especially engaged in the treatment of diphtheria show that 
when the antitoxin is used on the first day of the disease, practically no 
mortality occurs. Parents and guardians should be taught this fact, and 
cautioned to seek medical advice promptly whenever a child complains of 
sore throat. While the use of diphtheria antitoxin is still decried and opposed 
by a few members of the medical profession—and this is not be to wondered 
at when it is remembered that cowpox vaccination still has its opponents— 
it is at least to be hoped that no physician will deprive a patient suffering 
from diphtheria of the benefits to be derived from the antitoxin treatment. 
In the absence of special contraindications the refusal or neglect to use antitoxin 
in the treatment of diphtheria would, in the opinion of most physicians, con- 
stitute an act of criminal negligence and malpractice. 
Preparation of Diphtheria Antitoxin.—The methods of preparing and 
standardizing diphtheria antitoxin are given in Chapter XIII. Since anti- 
toxin gradually deteriorates with time, physicians should carefully observe 
1 Arch. d. med. d. enf., 1919, 22, 527. 
2 Bull. Acad. d. Med.. 1918, 80, 427. 
3 Lyon med.. 1913, 121, 941. 
4 Cor.-Bl. f. Schweiz. Aerzte, 1919, xlix, 310. 
6 Rev. med. d. 1. Suisse rom., 1920, xl, 159. 922 
VACCINES, SERA, BLOOD, PROTEINS IN TREATMENT OF DISEASE 
the date printed upon each package of antitoxin, which is the time limit 
calculated by the manufacturers beyond which they do not guarantee that 
the full antitoxic strength is maintained. 
Nature of Diphtheria.—In the great majority of cases diphtheria is a 
local infection of some portion of the upper respiratory tract. The bacilli 
are usually inhaled, find lodgment upon a mucous membrane, and secrete 
a toxin that produces necrosis of the cells of the mucosa and effectually 
resists phagocytosis of the bacilli. From this area of infection, which now 
becomes a prolific source of toxin production, the toxin or poison is absorbed 
by the body fluids, and the resulting toxemia is chiefly responsible for most 
of the symptoms of the disease. 
Other micro-organisms, such as staphylococci, pneumococci, and strepto- 
cocci, which may be unable to infect a healthy mucous membrane, readily 
multiply in the necrotic tissue and add to the severity of the local lesion, 
the lymphadenitis, and the toxemia. 
Rarely the diphtheria bacilli gain access to the blood-stream. The 
severity and danger of diphtheria are dependent primarily upon the strength 
and amount of toxin produced by the bacilli, and secondarily upon the size 
and location of the primary lesion and the amount of antitoxin present in 
the patient’s blood. A lesion in the larynx is far more dangerous than one 
of equal size on a tonsil, because in the former the edema and necrotic 
exudate obstruct the trachea and may produce death by suffocation. On 
the other hand, the size of the local lesion alone is not an indication of the 
severity of the infection, because virulent bacilli in a small patch may produce 
more toxin than less virulent ones in a larger area, and the degree of local 
tissue necrosis is not an absolute indication of the toxicity of the soluble 
poison. Other things being equal, the patient who has most antitoxin, 
either naturally or acquired as the result of a previous injection with anti- 
toxin or of an attack of diphtheria, will present least evidences of toxemia, 
although the bacilli causing the infection may be most virulent. For 
example, the highly virulent strain of diphtheria bacillus used extensively 
in the past twenty years in the production of antitoxin was isolated by Park 
and Williams from the throat of a patient presenting no clinical symptoms 
other than redness of the fauces and slight toxemia. 
The primary lesion of diphtheria is usually located in the throat (tonsils, 
uvula, larynx), and frequently in the nose; more rarely the ears, conjunctiva, 
vulva, prepuce, and wounds are the seats of primary infection. 
Purposes of Serum Treatment of Diphtheria.—If we were always certain 
of seeing our patients on the first day of their illness, and if the disease could 
always be diagnosed in this stage, the treatment of diphtheria would resolve 
itself into an immediate dose of antitoxin and rest in bed twro or three weeks. 
But patients frequently come under treatment comparatively late in the 
disease, or the true nature of the condition may not be fully diagnosed at 
first and treatment thus be delayed. Diphtheria is, therefore, still to be 
regarded as a dangerous infection, and while the proper use of antitoxin 
constitutes the most important part of the treatment, local applications, 
general constitutional measures, and the management of the various con- 
ditions that may complicate the disease are all to be considered. Here we 
will discuss only the serum treatment of the disease. 
Bearing in mind the pathology of diphtheria, serum treatment aims to 
fulfil the following primary indications: 
1. To neutralize all free toxin circulating in the body fluids, and also to 
neutralize, as much as possible, the toxin that has already united with the 
tissue cells. TREATMENT OF DIPHTHERIA 
923 
2. To cause the destruction or removal of the bacilli producing the toxin 
n,s quickly as possible. 
3. To furnish the patient with sufficient excess of antitoxin to neutralize 
the toxin as rapidly as it is produced until the virulent bacilli disappear. 
Action of Diphtheria Antitoxin.—Diphtheria antitoxin best fulfils these 
requirements. In fact, there are no substitutes. In former days the 
powerful and irritant caustics and germicides that were freely applied to 
the throat, instead of limiting the disease and destroying the bacilli, probably 
actually encouraged its extension by excoriating and depressing the resist- 
ance of the surrounding mucous membranes. 
1. The chief action of antitoxin is just what its name implies, namely, 
a substance that neutralizes the toxin. This is regarded as a chemical 
reaction analogous to the neutralization of an acid by an alkali. When 
the antitoxin molecule has united with the toxin molecule, it is believed that 
the toxin is neutralized and that both are rendered inert. As the result of 
experimental studies, however, we know that this union is not always a 
firm one, and it is possible for the toxin to become dissociated and attack 
body cells or other molecules of antitoxin, thus explaining in part the neces- 
sity for giving quite large doses of antitoxin—doses that are out of all pro- 
portion to what we would expect to be necessary, when considered weight 
for weight between guinea-pig and man, to effect complete neutralization 
■of the toxin. 
It is reasonable to presume, and may be accepted as true, that a stronger 
affinity exists between diphtheria toxin and antitoxin than between body 
-cells and toxin. Just as the union between this toxin and its antitoxin is 
somewhat unstable, so, in like manner, it is probable that the union of toxin 
and body cells is not so firm but that it may, during an early stage, be dis- 
sociated to some extent by the more attractive antitoxin. This factor is 
to be borne in mind in the treatment of diphtheria, and is an additional 
argument for the administration of large doses of the serum. 
2. Antitoxin pure and simple does not, however, constitute the only 
factor of value in antidiphtheric serum. While the pure antitoxin neutralizes 
the toxin, it has no injurious action on the bacilli themselves, and, indeed, 
it is said that the bacilli may live and multiply in the presence of large 
.amounts of antitoxin. How, then, are we to explain the gradual disappear- 
ance of the membranous exudate and the bacilli at the local site of infection? 
It has been shown experimentally that virulent diphtheria bacilli resist 
phagocytosis. This condition is probably due to an actual leukotoxic action 
of the diphtheria toxin, aided by an aggressin-like action of the toxin which 
neutralizes opsonin, and in this manner prevents phagocytic activity. The 
writer in common with other observers, has shown that the antiserum as 
■ordinarily produced neutralizes the antiphagocytic action of diphtheria 
bacilli, and enables the polynuclear leukocytes to ingest them readily.1 
Whether this is brought about through neutralization of the toxin by anti- 
toxin, or is due to the presence of an immune opsonin (bacteriotropin) that 
lowers the resistance of the bacilli, or to the presence of an anti-aggressin 
that neutralizes the intrinsic defensive mechanism of the bacilli and thus 
favors phagocytosis, I am unable to state, but probably all three factors 
are operative. 
3. As ordinarily prepared, diphtheria antitoxin possesses little or no 
bacteriolytic activity. I have found, however, that antitoxin will fix com- 
plement with an antigen of diphtheria bacilli,2 indicating, therefore, the 
presence of bacteriolytic amboceptors. Serums produced by immunizing 
1 Jour. Med. Research, 1912, xxvi, 373. 
2 Jour. Infect. Dis., 1912, xi, 44. 924 
VACCINES, SERA, BLOOD, PROTEINS IN TREATMENT OF DISEASE 
horses with unfiltered cultures of diphtheria bacilli instead of with the 
filtered toxin alone have been advocated for the general and the local treat- 
ment, in the hope of securing lysis of the infecting bacilli. Certainly it 
would appear wise to raise the bacteriotropic and bacteriolytic content of 
an immune serum by injecting the horses occasionally with unfiltered cul- 
tures, for it is highly probable that the action of the antiserum depends not 
only upon an antitoxin but also, to some extent at least, upon a bacterio- 
tropin and possibly a bacteriolysin. 
Subcutaneous Administration of Diphtheria Antitoxin.—Antitoxin is 
usually administered by subcutaneous injection into the tissues of the back, 
abdomen, or buttocks. Experimental studies have tended to show that 
complete absorption does not occur until forty-eight hours after, although 
it is common clinical experience to observe improvement take place during 
the first twenty-four hours after injection. 
Park1 has proved that it takes twenty-four hours for the major part of 
the antitoxin to be absorbed by the blood from the subcutaneous tissues 
and some twelve hours from the muscles; two or three days are required for 
total absorption. 
As will be emphasized later, it is highly desirable and necessary to get 
antitoxin into the circulation as soon as possible after infection has occurred. 
Usually this is best accomplished by giving antitoxin to every patient even 
suspected of being dipntheric, and making the diagnosis afterward; the next 
best method is to administer the antitoxin in such manner as to favor quick 
absorption. For this reason intramuscular and intravenous injection should 
be resorted to in all severe cases. As previously pointed out the Schick 
reaction in diphtheria has indicated that diphtheria toxin may be dissociated 
from tissue cells by large doses of antitoxin. Park and his associates have 
shown experimentally by this reaction that 20,000 units of antitoxin given 
subcutaneously were necessary to yield an effect equal to 1000 units given 
intravenously. 
The subcutaneous route, therefore, should be used only in the treatment of 
mild infections. Intramuscular injections are just as easy, not more painful, 
and, as will be discussed in the succeeding paragraph, serum is absorbed 
at least three times more quickly. 
Intramuscular Administration of Diphtheria Antitoxin.—Antitoxin does 
not become effective until it is absorbed in the blood and then passing 
through the capillary walls to the tissue fluids and cells. The greater the 
quantity in the blood, the more thorough and rapid will be the neutralization 
of toxin. 
As previously stated, intramuscular injections are easily given in the mus- 
cles of the thighs and bullocks, and since absorption is three to four limes more 
rapid than by subcutaneous injection, there is no valid reason why intramuscular 
injections should not be used routinely instead of subcutaneous injections; 
particularly in the treatment of late cases and cases of more severe infections. 
Rolleston and MacLeod,2 Veeder,3 and numerous others have emphasized 
the value of intramuscular injections. 
Intravenous Administration of Antitoxin.—Intravenous injections are far 
more difficult, especially in children with fat arms and feeble circulations. 
An anesthetic or ten minutes’ struggling may do the patient harm, and 
unless the injection can be given with little disturbance and danger, the 
serum should be given by intramuscular injection. Not infrequently, 
1 Jour. Amer. Med. Assoc., 1912, 58, 1976; Jour. Infect. Dis., 1914, 14, 338. 
2 Brit. Jour. Dis. Child., 1913, 11, 289. 
3 Missouri State Med. Assoc. Jour., 1915, 12, 145. TREATMENT OF DIPHTHERIA 
925 
however, an intravenous injection yields splendid results in severe and 
apparently hopeless cases, and in older children and adults this route of 
administration should be considered. 
There can be no doubt, however, of the value of intraveneous injections 
in severe cases and especially in late advanced cases. By these are meant 
cases already sick for twenty-four to seventy-two hours with rapid extension 
of the disease in the throat or nose, swelling of the tissues and glands, and 
severe constitutional symptoms. According to Park1 antitoxin given 
intravenously passes out to the tissues about ten times as rapidly as when 
given subcutaneously and four times as rapidly as when given intramuscu- 
larly. 
Intravenous injections are being given in increasing frequency; numerous 
reports advocating this route in the treatment of severe infections, whenever 
possible, have been made by Tachan,2 Nickolson,3 Dupaquier,4 Schone,5 
Mixsell,6 Schorer,7 Alber,8 and others. If it were not for the fact that the 
technic of administration is more difficult and the reactions sharper and 
more dangerous, the intravenous route would be the one of choice in the 
treatment of all but mild early cases. 
In laryngeal diphtheria, however, the intravenous injection of antitoxin 
while giving a quicker response, does not greatly lower the death-rate. In 
a series of 79 cases treated by Park and Mixsell with intravenous injections 
the mortality was 53.1 per cent. Bronchopneumonia developed in 45.5 
per cent, of the former series and in 51.9 per cent, of the latter, and out of a 
total of 77 cases developing this dreaded complication only 5 survived. 
Intravenous injections are apt to be followed in from fifteen to sixty 
minutes by a rise in temperature of 2° to 5° F. and a chill. The chill is not 
usually severe, lasts for five to forty-five minutes, and occurs in about 40 per 
cent, of cases. The temperature falls in a few hours to its former level 
or below and no bad effects are noticeable. These reactions are doubtless 
caused by the serum proteins producing a “protein shock reaction,” they 
are not dangerous, but are disturbing at the time and may impress the 
family unfavorably. 
The syringe method is well adapted for the intravenous injection of 
antitoxin, as the bulk method of serum is usually small, especially if a con- 
centrated antitoxin is being used. 
The technic of these injections has previously been described in Chapter 
XXXVII. 
Oral Administration.—Antitoxin has also been given by the mouth and 
even by rectal injection. The presence of a preservative, usually phenol, 
renders the oral administration objectionable. 
McClintoch and King9 have shown experimentally that antitoxin is ab- 
sorbed after oral administration, but in an irregular and unreliable manner. 
If digestion is inhibited absorption is facilitated, and they have recommended 
that 1 minim of fluidextract of opium and 4 to 10 minims of saturated 
solution of salol in chloroform be added to each dose and that no food be 
taken for at least four hours before administering the mixture. 
1 Trans. Cong. Amer. Phys. and Surg., 1916, 10, 118. 
2Therapie der Gegenwart, 1910, 51. 
3 Bull. Med. and Chir. Faculty of Maryland, 1914, 8, 45. 
4 New Orleans Med. and Surg. Jour., 1915, 68, 145. 
5 Deutsch. Arch. f. klin. Med., 1913, 110. 
6 Californ. State Jour. Med., 1913, 11, 297. 
7 Amer. Jour. Dis. Child., 1915, 9, 59. 
8 Jahrb. f. Kinderh., 1913, 30, No. 2. 
9 Jour. Infect. Dis., 1906, 3, 701; ibid., 1909, 6, 46. 926 
VACCINES, SERA, BLOOD, PROTEINS IN TREATMENT OF DISEASE 
While a therapeutic effect may be secured after large doses have been 
swallowed, there are very few occasions when this should be the method of 
choice. 
Serum sickness does not usually follow oral administration and the 
method may be of value in the treatment of diphtheria occurring in an 
individual extremely susceptible to horse-serum. Furthermore, the method 
may be of value for continuing the effects of antitoxin when by reason of fear 
on the part of the patient or for other reasons injections cannot be given. 
Local Application of Antitoxin.—There does not appear to be any advan- 
tage in applying either fluid or dried serum antitoxin to the involved tissues- 
in the nose or throat. The antitoxin itself is not directly bactericidal and 
the only good to be expected is that of cleansing, which may be done just 
as well with simple saline solution. 
Importance of Early Treatment.—The most important point to he ob- 
served in the treatment of diphtheria is to give antitoxin at once. It may he 
fatal to wait for the result of a culture, except perhaps in the case of the mildest 
of infections. In fad, the necessity for early administration of antitoxin cannot 
he overestimated. When once suspicion is aroused, antitoxin should be given 
at once and the diagnosis may follow. A few hours may make an enormous 
difference in the prognosis of any case, and while a dose of 2000 units of 
antitoxin may prove of the utmost value in checking diphtheria, it will 
do no harm whatever in case the disease is not present. 
It is true that a physician will naturally hesitate to administer antitoxin 
unnecessarily; nevertheless, diphtheria is frequently a difficult disease to 
diagnose clinically, and is quite likely to be mistaken for tonsillitis. For 
this reason many physicians prefer to wait for the result of a culture, and 
this is proper, provided that the patient, especially in the case of a child, 
is given the benefit of the doubt by receiving 2000 units of antitoxin. It is 
to he emphasized that a single negative culture does not exclude diphtheria. As 
ordinarily made, about 20 per cent, of primary cultures from genuine cases 
of diphtheria fail to show the presence of bacilli, whereas subsequent cultures 
will showT them to be present in large numbers. A primary negative culture 
is most likely to be obtained from a patient having a heavy exudate, as the 
physician may rub lightly over the membrane, culturing the micro-organisms 
of secondary infection and overlooking the diphtheria bacilli in the depths 
of the membrane adjacent to the diseased mucous membrane. To wait 
another twenty-four hours for a second culture still further reduces the 
patient’s chances for recovery. In the vast majority of instances, therefore, 
antitoxin should he given at once and repealed as often as is necessary until the 
correct diagnosis is established, and not one but at least two successive negative 
cultures should be obtained before diphtheria is to be excluded with any reasonable 
degree of safely. 
The following table, compiled from the annual reports of the Philadelphia 
Hospital for Contagious Diseases, shows the decided influence of early 
treatment upon the mortality of diphtheria. This table comprises cases 
of diphtheria alone, and does not include cases complicated by other diseases, 
such as scarlet fever and measles. It is worthy of special notice that of 741 
cases treated with antitoxin on the first day of the disease, not one died. Ker,1 
in an exceptionally rich experience, has never seen a fatal result occur in a 
case that developed in a hospital and in w’hich antitoxin wras administered 
on the first day of the disease. 
Recently he has published the following table, in wdiich the effects of 
early treatment upon the mortality is brought out very clearly: 
1 Infectious Diseases, 1920, Oxford Med. Press. TREATMENT OF DIPHTHERIA 
927 
MORTALITY OF DIPHTHERIA ACCORDING TO DAY ON WHICH ANTITOXIN 
WAS FIRST INJECTED IN 8591 CONSECUTIVE CASES 
Day of 
Number of 
Percentage of 
Illness. 
Patients. 
Deaths. 
Mortality. 
First 
. 329 
5 
1.52 
Second 
2269 
77 
3.39 
Third 
2407 
165 
6.85 
Fourth 
1612 
176 
10.91 
Fifth 
911 
136 
14.92 
Sixth 
416 
54 
12.98 
Seventh 
320 
53 
16.56 
327 
50 
15.29 
Totals 
8591 
716 
8.33 
MORTALITY OF DIPHTHERIA ACCORDING TO THE DAY OF ADMISSION 
(WHICH ALSO INCLUDES THE TIME OF GIVING THE ANTITOXIN) IN 
THE PHILADELPHIA HOSPITAL FOR CONTAGIOUS DISEASES 
Year. 
Total 
Number 
of 
Cases. 
Mortality According to the Dax on which Antitoxin was 
First Injected. 
Average 
Mor- 
tality 
in Diph- 
theria 
Alone. 
First 
Day. 
Second 
Day. 
Third 
Day. 
Fourth 
Day. 
Fifth 
Day. 
Sixth 
Day. 
After 
Sixth 
Day. 
1904 
712 
4.09 
13.72 
17.54 
14.75 
19.44 
12.94 
10.81 
1905 
862 
4.43 
9.22 
16.66 
13.04 
9.52 
22.89 
10.09 
1906-07 
1,239 
3.45 
12.90 
10.85 
13.08 
29.41 
6.09 
8.70 
1908 
1,426 
6.62 
4.7 
10.60 
11.71 
21.43 
23.96 
8.55 
1909 
2,153 
4.61 
7.13 
10.72 
9.35 
7.25 
13.33 
6.60 
1910 
1,870 
5.50 
5.13 
8.41 
13.74 
6.04 
8.33 
6.42 
1911 
1,895 
6.91 
9.41 
7.12 
11.04 
7.22 
2.66 
6.86 
1912 
1,676 
2.92 
6.33 
8.0 
9.53 
14.97 
9.63 
6.02 
1913 
1,273 
3.92 
6.51 
6.52 
6.87 
4.76 
12.5 
18.8 
7.63 
Average 
13,106 
0.4 
5.0 
8.3 
10.7 
11.2 
14.2 
13.1 
7.96 
It is generally believed that the paralyses of diphtheria are due to a 
toxone, and not to the true toxin. Ehrich believes toxone to represent a 
late secretory product of the diphtheria bacillus, whereas others regard it 
as a modified toxin. It is certainly apparent, however, that the bacilli 
should be gotten rid of as soon as possible, so as to eliminate the possibility 
of toxone production. This is best accomplished by the early use of large 
doses of antitoxin, aided possibly by judicious local treatment. 
Dosage of Diphtheria Antitoxin.—While it is now generally agreed that 
the doses of from 100 to 200 units, such as were commonly given during the 
early years of antitoxin therapy, were far too small, there is still some differ- 
ence of opinion regarding the proper doses to employ. Since the severity 
of the disease varies so markedly, no hard-and-fast rules can be given. The 
physician who has a clear idea of the nature of diphtheria and of the action 
of antitoxin, and knows what to expect of the latter in the treatment of the 
disease, should have no difficulty in properly treating a case of diphtheria. 
General Treatment.—It is to be emphasized, however, that while anti- 
toxin constitutes the most important part of the treatment of diphtheria, 
it is not usually the whole treatment. Absolute rest in bed, a generous diet, 
combined with the use of tonics and local applications, are all part of the 
treatment. Special treatment of the laryngeal form of the disease and the 
treatment of complications are matters of considerable importance that 
influence the prognosis in a given case. I may mention, in passing, the 928 
VACCINES, SERA, BLOOD, PROTEINS IN TREATMENT OF DISEASE 
value of continuous enteroclysis of normal salt solution during the early 
periods of severe infections; this appears to dilute the toxins and aid in their 
destruction and excretion. 
In administering antitoxin the physician must be guided by the clinical 
condition of the patient, as we have as yet no practical laboratory method 
for estimating the degree of the toxinemia. Treatment may be regarded as 
satisfactory when— 
1. The local patch of exudate has ceased to spread and shows indica- 
tions of disappearing. 
2. The general condition of the patient is improved, i. e., the toxinemia 
is decreased, the pulse grows stronger and more regular, and the patient 
feels more comfortable. 
The temperature is not a reliable guide, for not infrequently in severe 
infections it may be normal or subnormal throughout. 
So long as the exudate shows no signs of loosening and disappearing, but 
tends to spread, and so long as the general condition remains unimproved 
or grows worse, large amounts of antitoxin should be given. No case should 
be regarded as hopeless until death supervenes. 
Every case of diphtheria is to be treated individually, rather than by 
any set rule. The amount of antitoxin given in the initial dose, and in 
subsequent doses as well, is dependent on the following factors: 
1. The situation and extent of the lesion. 
2. The general condition of the patient. 
3. The day of the disease. 
4. The age of the patient. 
1. The Silualion and Extent of the Lesion.—In ordinary tonsillar diph- 
theria in which there is a small patch on one tonsil and which is first seen on 
the second day of the disease, the initial dose should be at least 5000 units. 
When the exudate involves the pillars of the fauces, the uvula, the posterior 
pharyngeal wall, or is well forward on the palate, this dose should be doubled, 
and 10,000 units be given. 
Cases that present laryngeal symptoms in addition to faucial lesions 
should never receive less than 12,000 units. In well-marked laryngeal 
diphtheria with dyspnea and partial suffocation at least 20,000 units should 
be given, preferably by intramuscular or intravenous injection. 
Bronchopneumonia is the most dreaded complication of laryngeal 
diphtheria and is responsible for the majority of deaths. Out of a series 
of 77 cases studied by Park and Mixsell only 5 recovered, making a death-rate 
of 93.5 per cent. Of 81 cases of laryngeal diphtheria not developing broncho- 
pneumonia the mortality was 7.4 per cent. 
Hogan1 has recently stated that 82.11 per cent, of the deaths from diph- 
theria in Baltimore have been due to laryngeal infections, and he makes a 
special plea for earlier diagnosis, earlier administration of antitoxin, and 
intubation. 
In nasal diphtheria a distinction must be made between those cases 
that exhibit merely a dirty, chronic discharge containing bacilli, in which 
2000 units may suffice, and those that present an actual membrane accom- 
panied by well-marked toxic symptoms, when a large amount of serum— 
at least 10,000 units—should be given. Owing to the ready absorption of 
toxin by the nasal mucosa a small patch in the nose may be accompanied 
by severe general toxinemia and frequently requires energetic treatment. 
In diphtheria of the eye, vulva, or of wounds relatively large doses should 
be given—at least from 10,000 to 20,000 units. 
1 Jour. Amer. Med. Assoc., 1921, 77, 662. TREATMENT OF DIPHTHERIA 
929 
All these doses must be regarded merely as suggestions for the initial 
dose in cases seen on about the second day of the disease. Physicians with 
extensive hospital experience, for example, Woody and McCullom, generally 
favor large doses, and while in private practice the question of expense and 
economy may be a factor, the physician will be wise to err on the side of 
safety and give a little too much serum rather than too little, especially in 
the first dose, when so much depends upon how soon antitoxin is introduced 
into the body fluids. 
2. The general condition of the patient, or the effect which the toxinemia 
will have on the patient, is highly important in estimating the dosage of 
antitoxin. A patient who is pale, drowsy, prostrated, and has a weak and 
irregular pulse; who has large masses of glands around the neck, or who 
has marked albuminuria, will require a much larger dose than one who 
presents none of these signs. Two persons of about the same age and 
suffering from the same lesions may show very different degrees of toxinemia. 
In the severe cases we must administer the maximum dose and repeat it at 
suitable intervals until an effect is produced. 
3. The day of illness on which the patient comes under observation is 
important in deciding the initial dose of antitoxin. For corresponding 
tonsillar lesions a dose of 2000 units on the first day may do more good than 
5000 units given on the fourth day. Ker gives second-day cases of purely 
tonsillar diphtheria 3000 units, and adds an additional 1000 on the following 
day. 
4. If the age of the patient exerts any influence at all on dosage, it indicates 
that more antitoxin should be given to children than to adults with cor- 
responding lesions, as the disease is more fatal in children. So far as infants 
are concerned, Ker seldom gives more than 4000 units at a single dose, which 
should be an adequate amount when we consider the small size of the patient. 
Children over one year of age may be given from 5000 to 10,000 or more 
units, depending upon the location of the lesion and the degree of toxinemia. 
Repeating the Dose.—Whether or not subsequent doses of antitoxin will 
be required is dependent upon the circumstances of the individual case. 
In ordinary cases if on the day after treatment is commenced there is no 
diminution in the amount of membrane visible and the general symptoms 
have not improved, the dose should be repeated. If the membrane has 
spread and the toxinemia is worse, the second dose should be larger than the 
first. In septic cases the second dose may be given in from six to ten hours 
after the first. If the symptoms are less urgent, the interval may be ex- 
tended to twelve, but should never exceed twenty-four hours. 
One thing is certain: a part of what should be the first dose should not 
be held back to be given later. When the first dose has been of sufficient 
size, the second and third injections are not required. As previously stated, 
subsequent injections may result in the neutralization of toxin already 
united with body cells as indicated by the investigations of Kleinschmidt,1 
but most depends upon a large single dose as early in the disease as possible. 
The Importance of Large Doses of Antitoxin.—As to the total amount of 
serum to be administered, continued injections at relatively short intervals 
are required until improvement has taken place. As long as a membrane is 
present and the patient is toxic it is probably worth while to push the treat- 
ment unless these show a tendency to clear away. Time must be allowed 
for absorption to take place, and the serum should not be pushed so far as 
to be wasted, and, quite possibly, excreted unchanged. The remedy is 
expensive, especially in private practice, and it is obviously desirable to have 
1 Jahrb. f. Kinderh., 1912, 76, 1. 930 
VACCINES, SERA, BLOOD, PROTEINS IN TREATMENT OF DISEASE 
due regard for economy. While, as previously mentioned, physicians of 
such wide experience as McCullom, of Boston, and Woody, of Philadelphia, 
frequently give 200,000 or more units in severe cases of diphtheria, others, 
e. g., Ker, of Edinburgh, have never given more than 64,000 units to a single 
patient, and, indeed, several of my colleagues of wide experience claim that 
they have secured excellent results in severe infections with doses that 
seldom exceeded 10,000 units. 
The following table by Park summarizes well these questions of dosage 
and route of administration: 
DOSAGE OF UNITS OF ANTITOXIN IN DIPHTHERIA—SINGLE DOSE ONLY 
For infants, 10 to 30 pounds (under two years): 
Mild Cases. 
Moderate Cases. 
Severe Cases. 
Malignant Cases. 
(Subcutaneous or 
intramuscular in- 
jection) 2000 to 
3000. 
(Intramuscular or 
subcutaneous in- 
jection) 3000 to 
5000. 
(Intramuscular or 
one-half intrave- 
nous and one-half 
intramuscular or 
subcutaneous) 
5000 to 10,000. 
(One-half intravenous 
and one-half intra- 
muscular) 10,000. 
For children, 30 to 
90 pounds (under fifteen years): 
3000 to 4000. 
4000 to 10,000. 
10,000 to 15,000. 
10,000 to 20,000. 
For adults, 90 pounds and over: 
3000 to 5000. 
5000 to 15,000. 
10,000 to 20,000. 
15,000 to 40,000. 
Antitoxin Treatment of Carriers.—Diphtheria antitoxin has no effect 
upon ridding the tissues of the nose and throat of carrier diphtheria bacilli. 
Fluid and dried sera have been applied, but have proved ineffectual, although 
under any plan of treatment a few apparent successes may be encountered. 
Recent experiments by Gelien, Moss and Guthrie1 have shown that the 
subcutaneous injection of antitoxin may not prevent the successful lodgment 
and growth of diphtheria bacilli in the nasopharynx of individuals so treated. 
Treatment of Relapses.—Occasionally after a patient has recovered 
from an attack of diphtheria the infection recurs after several weeks. It is 
in such cases that the physician hesitates to administer antitoxin on account 
of the discomforts occasioned by serum sickness. It is true that serum 
sickness is more likely to follow in these than in primary cases, and the very 
profuse and itchy eruption, joint pains, and fever do indeed render the 
patient quite uncomfortable. Since a relapse is usually, although not 
always, comparatively mild, the serum may be dispensed with if there is no 
involvement of the larynx and if there is not much evidence of toxinemia; 
otherwise full doses of antitoxin should be given without hesitation. The 
subcutaneous injection of 0.5 c.c. of the serum two or three hours before the 
main dose is given may possibly produce a condition of anti-anaphylaxis 
and ward off the more dangerous symptoms. If respitatory difficulties should 
follow a reinjection of serum, adrenalin chorid, atropin, and caffein should 
be administered hypodermically (see Chapter XXXII). 
Antitoxin Sequelae.—A certain percentage of cases will present a group 
of symptoms that constitute the condition known as serum sickness, occurring 
at varying times following the administration of antitoxin. This condition 
has been shown to be due to certain constituents of horse-serum other than 
the antitoxic antibodies. It is noteworthy that the serum of one horse may 
cause more serum sickness than that of another; in general, concentrated 
antitoxins produce fewer cases than whole serum. 
1 Bull, Johns Hopkins Hosp., 1920, 31, 381. TREATMENT OF DIPHTHERIA 
This condition is characterized by the development of a rash (urticarial,, 
multiform, or scarlatiniform), mild fever, joint pains, and possibly adenitis. 
The scarlatiniform rash may be extremely difficult to differentiate from that 
of true scarlatina, especially in the wards of a diphtheria hospital, where 
outbreaks of scarlet fever are not uncommon. 
This subject has been considered in greater detail in Chapter XXX. 
It may be stated here that while the patient is decidedly uncomfortable, 
and even quite sick, for several days, serum sickness is not a dangerous 
condition; the treatment is largely symptomatic and palliative. 
Value of Diphtheria Antitoxin.—At the present day it seems hardly 
necessary to introduce elaborate statistics to prove the value of antitoxin 
in the prophylaxis and treatment of diphtheria. 
1. It is generally admitted that most of the reduction in the mortality 
of diphtheria cannot be attributed solely to the use of antitoxin, for un- 
questionably bacteriologic diagnosis has permitted the inclusion, in our 
statistics, of a certain number of cases that, twenty years ago, would not 
have been classed as diphtheria. Generally speaking, however, the mor- 
tality of diphtheria, considering all types of infection coming under observa- 
tion at varying intervals after the disease has developed, is at least five 
times less under antitoxin treatment than when it is treated without antitoxin. 
This proportion is true, whether we compare the general mortality before 
1896 with the present rate, or whether we take a large city and compare 
the mortality under both forms of treatment for a single or for several years. 
For example, in Philadelphia the mortality rates per 100,000 of population 
for the five years preceding the use of antitoxin were as follows: 
931 
1891 127.4 
1892 156.3 
1893 103.9 
1894 122.5 
1895 115.9 
In five years following the general use of antitoxin the mortality rates 
per 100,000 population were as follows: 
1906 37.78 
1907 34.60 
1908 33.35 
1909 33.6 
1910 31.7 
In Philadelphia, during the years 1909-10 and 1911, the average mor- 
tality of diphtheria treated with antitoxin in the Philadelphia Hospital for 
Contagious Diseases was 9.9 per cent., and in the private practice of phys- 
icians, 13.07 per cent. In contrast to these figures is the mortality of 40.34 
per cent, in the private practice of those physicians (fortunately few) who 
refused to give antitoxin or in those families opposed to its use. 
Kossel1 has recently stated that in Germany prior to the use of antitoxin 
(1886 to 1894) the mortality from diphtheria was from 85 to 130 per 100,000 
inhabitants; since 1899 it has never reached 30. 
The average mortality in 13,106 cases of pure diphtheria treated with 
antitoxin in the Philadelphia Hospital for Contagious Diseases during the 
years 1903-1914 was only 7.96 per cent. As stated elsewhere, when this is 
compared with the average mortality of about 41 per cent, when no anti- 
toxin was used, it is not difficult to appreciate the therapeutic value of the 
remedy. 
1 Deutsch. med. Wchn., 1915, 41, 1445. 932 
VACCINES, SERA, BLOOD, PROTEINS IN TREATMENT OF DISEASE 
2. As was previously stated, it is also worthy of note that there is practi- 
cally no mortality in diphtheria cases receiving antitoxin on the first day of 
illness. During nine consecutive years (1904-1913), covering the treatment 
of 741 such cases in the Philadelphia Hospital for Contagious Diseases, not 
a single death occurred. During 1913, 2 of the 51 first-day cases died. 
It will also be noted that the patient’s chance for recovery grows steadily 
less with each day that the administration of serum is delayed, and this 
should be evidence enough to convince any right-minded person that we 
possess in antitoxin a remarkable remedy for the treatment of diphtheria. 
3. The influence of antitoxin is also noted in the mortality of laryngeal 
diphtheria. While the mortality in this condition is still high, owing to the 
frequency and dangers of bronchopneumonia and the necessity for operative 
measures, it has been reduced at least one-half since antitoxin came into 
general use. Prior to 1896 the mortality was at least 70 per cent.; since 
then it has been reduced to 35 per cent, or less. As shown in the following 
table, of 1207 cases treated in the Philadelphia Hospital for Contagious 
Diseases, the average mortality was 35.6 per cent. 
MORTALITY OF LARYNGEAL DIPHTHERIA (INTUBATION CASES) IN THE 
PHILADELPHIA HOSPITAL FOR CONTAGIOUS DISEASES 
(This table excludes those patients dying in the ambulance and moribund cases dying 
within twenty-four hours.) 
Year. 
Total 
Number 
of 
Cases. 
Mortality According to the Day Upon which Intubation 
was Performed. 
Average 
Mor- 
tality. 
First 
Day. 
Second 
Day. 
Third 
Day. 
Fourth 
Day. 
Fifth 
Day. 
Sixth 
Day. 
After 
Sixth 
Day. 
1903 
67 
66.6 
25.3 
21.8 
30.9 
22.5 
45.0 
26.60 
1904 
125 
39.20 
29.03 
73.68 
33.33 
70.0 
36.84 
39.20 
1905 
136 
30.76 
39.39 
50.0 
21.42 
16.66 
42.30 
36.76 
1906-07 
72 
50.0 
55.0 
40.74 
45.45 
71.92 
28.57 
49.29 
1908 
162 
29.26 
50.0 
33.33 
30.76 
9.09 
38.88 
33.95 
1909 
183 
50.0 
44.18 
38.88 
23.81 
53.84 
40.0 
42.62 
1910 
89 
100.0 
55.0 
33.33 
28.58 
50.0 
36.46 
39.34 
1911 
104 
50.0 
76.66 
55.55 
52.63 
36.36 
27.27 
54.80 
1912 
133 
30.0 
58.33 
41.93 
48.27 
50.0 
63.67 
50.0 
48.87 
1913 
136 
100.0 
65.91 
63.64 
35.29 
25.0 
47.82 
68.0 
58.82 
Average 
1207 
35.6 
- Formerly when a child contracted diphtheria the parents were warned 
of the likelihood and danger of the infection involving the larynx and trachea; 
nowadays this possibility is quite remote. 
4. Finally the claim of the opponents of the serum therapy of diphtheria 
that antitoxin increases the percentage of paralyses is without foundation. 
While it is true that this percentage is somewhat higher than was noted in 
former years, this increase is to be explained by the fact that antitoxin 
saves a larger number of severe cases long enough for them to manifest 
paralyses, and, second, by the greater attention that has recently been 
directed to its milder forms. Since diphtheric paralysis is regarded as 
caused by toxone or a later secondary toxic product of the bacilli, the indica- 
tions are to rid the patient of the bacilli as quickly as possible, and this is 
best and most surely accomplished by the proper administration of anti- 
toxin. TREATMENT OF TETANUS 
933 
Normal Horse-serum and Other Non-specific Agents in the Treatment 
of Diphtheria.—Bingel1 has treated 466 cases of diphtheria with subcutaneous 
injections of normal horse-serum and the results based upon mortality, 
duration of disease, and number of complications showed no marked differ- 
ences from 471 cases treated with immune horse-serum (antitoxin). Others 
have employed normal horse-serum with similar results, the subject having 
been recently reviewed by Klotz,2 and doubtless a part of the beneficial 
results of the antitoxin treatment of diphtheria are to be ascribed to non- 
specific effects produced by the serum proteins. This may be one reason 
why not a few of the older physicians believe that the old-fashioned raw 
serum antitoxin gave better results than the solutions of concentrated 
globulins now in common use, although the incidence of serum disease was 
higher. Calhoun,3 Kraus and Sordelli4 found that normal horse-serum had 
little or no effect upon neutralizing diphtheria toxin injected into guinea-pigs, 
but this does not exclude the possible beneficial non-specific effects in human 
beings from injections of normal or antitoxic horse-serum. 
Liidke has treated 15 cases with two to three subcutaneous injections of 
3 to 5 c.c. of 10 per cent, solution of albumoses and claims that the results 
compared favorably with those observed with antitoxin therapy. 
Treatment of Diphtheria Carriers with Vaccines and Non-specific 
Agents.—A large number of substances have been advocated for the treat- 
ment of diphtheria carriers to rid the tissues of the upper respiratory tract 
of both virulent and non-virulent bacilli. Hewlett and Nankivell5 have 
advocated the administration of a vaccine of diphtheria bacilli, and recently 
Fraser and Duncan6 have renewed interest in this subject, advocating the 
subcutaneous injection of a detoxicated diphtheria bacillus vaccine. The 
vaccine is so prepared that each cubic centimeter contains 100,000,000,000 
and injections are given every four days for twelve injections beginning 
with 0.05 c.c. in the first dose and gradually increasing the amounts until 
as much as 3.5 c.c. are given with the twelfth or last injection. 
Petersen states that both Paschen and Muller have reported the effective 
treatment of these cases by intramuscular injections of sterile milk. 
TREATMENT OF TETANUS 
Tetanus antitoxin was discovered by Behring and Kitasato in 1892. 
Since then it has proved of great value in the prevention of lock-jaw. While, 
in general, authoritative opinions regarding its curative value vary, the 
statistics and the individual experience of many investigators of more recent 
years show quite conclusively that tetanus antitoxin does possess curative 
value, and is of distinct aid in the treatment of tetanus, especially when the 
nature of the disease is understood and the serum is administered accordingly. 
Preparation and Standardization of Tetanus Antitoxin.—This technic 
has been described in Chapter XIII. Since antitoxin gradually deteriorates 
with time, physicians should carefully observe the date printed on each 
package of antitoxin, which is the time limit calculated by the manufac- 
turers beyond which they do not guarantee that the full antitoxic strength 
is maintained. 
Action of Tetanus Toxin.—It may be well to recall briefly the main 
features concerning the pathogenesis of tetanus, as successful treatment 
depends upon a thorough understanding of these principles. 
1 Deutsch. Arch. f. klin. Med., 1918, cxxv, 284. 
2 Berl. klin. Wchn., 1919, lvi, 987. 
3 Amer. Jour. Dis. Child., 1921, 21, 107. 
4Ztschr. f. Immunitatsf., 1921, 31, 107. 
6 Lancet, 1912, 1, 143. 
6 Lancet, 1920, 2, 994. 934 
VACCINES, SERA, BLOOD, PROTEINS IN TREATMENT OF DISEASE 
1. Tetanus is a local infection accompanied and characterized by a 
general toxinemia. The bacilli and spores never gain access to the blood, 
and are never distributed through the tissues and internal organs, but reside 
at the local site of infection, where they produce a powerful toxin, which, 
when absorbed, is responsible for the main lesions and symptoms of the 
infection. Therefore while the blood of the tetanus patient is sterile, it 
usually contains the toxin. Neisser has produced tetanus in mice by giving 
them subcutaneous injections of the blood of a tetanus patient. 
2. Tetanus toxin has a strong affinity for nerve tissue, and this con- 
stitutes the most important feature in the pathogenesis of the disease. 
The toxin is rapidly absorbed from the local site of infection into the blood 
and lymph-streams, where it is distributed to other muscles and reaches 
the central nervous system indirectly through absorption by the end-plates 
of motor nerves. As expected, absorption is most likely to occur along the 
motor nerves supplying the parts injured, and for this reason the muscles 
and nerves should be infiltrated with antitoxin as soon as possible after an 
injury has been received. 
According to Meyer and Ranson,1 Marie and Morax,2 absorption occurs 
along the axis-cylinders of motor nerves, the intramuscular endings of 
which the toxin penetrates. The experiments of Field, Cernovodeanu, 
Henni and Teale3 indicate, however, that the toxins are absorbed by way of 
the lymphatics of the nerves, and not by way of the axis-cylinders; the latter 
view is now most general accepted. 
Robertson4 believes that the toxin passes throughout the body exclu- 
sively by the blood and lymph streams and, theoretically, may be neu- 
tralized by antitoxin and at any stage in this passage before the final and 
comparatively* indissoluble union with the ganglion cells occurs. 
Even though the toxin gains entrance to the blood, it cannot injure the 
motor nerve tissue directly, as, for example, by means of the blood-vessels 
supplying the central nervous system. As was previously stated, the toxin 
in the blood and lymph channels may reach the central nervous system 
only in an indirect manner, through the end-plates or lymphatics of other 
motor nerves. 
Ascending centripetally along the motor plates and lymphatics, the 
poison reaches the motor spinal ganglia on the side inoculated; it then affects 
the ganglia on the opposite side, making them hypersensitive. The visible 
result of this hypersensitiveness is the highly increased muscle tonus—i. e., 
rigidity. If the supply continues, the toxin next affects the nearest sensory 
apparatus: there is an increase in the reflexes, but only when the affected 
portion is irritated. In the further course of the poisoning the toxin as it 
ascends continues to affect more and more motor centers and also the neigh- 
boring sensory apparatus. This leads to spasm of all the striated muscles 
and general reflex tetanus (Park). 
3. Regardless of the severity of the infection, there is always an incuba- 
tion period in tetanus during which the bacilli multiply and produce toxin 
which is traveling toward the tissues of the central nervous system. Anti- 
toxin in sufficient amount will neutralize the toxin as quickly as it is produced, 
and thus protect the infected individual until the leukocytes and other body 
cells have destroyed the bacilli and spores. 
In general terms, the severer the wound and the heavier the infection, 
1 Arch. f. Path. u. Phar., 1903, xlix, 369. 
2 Ann. de l’lnst. Pasteur, 1902, 16, 818; ibid., 1903, 17, 335. 
3 Jour. Path, and Bacteriol., 1919, 23, 50. 
4 Amer. Jour. Med. Sci., 1916, 152, 31. TREATMENT OF TETANUS 
935 
the shorter will be the incubation period and the higher the mortality. 
In acute tetanus the incubation period is less than ten days; in chronic tetanus 
this period is much longer and more variable. 
The toxin is produced, and may be absorbed during or at least soon after 
the first twenty-four hours following infection; this explains the necessity for 
administering antitoxin as soon as possible after the injury has been received. 
Action of Tetanus Antitoxin.—1. Tetanus antitoxin will neutralize free 
toxin in a chemical manner similar in some respects to that by which neu- 
tralization of an acid by an alkali is effected. It is generally believed that 
as soon as the molecule of antitoxin has become united with a molecule of 
toxin, the latter is rendered inert. It may be possible, however, for the 
toxin molecule to become dissociated and attack nerve-cells or other mole- 
cules of antitoxin; this is one reason for the necessity of giving large doses 
of antitoxin in order than an excess may be at hand. 
2. When tetanus toxin has once united with nerve-cells, it is difficult 
or impossible for the antitoxin to effect its neutralization. Hence the 
greatest value of the antitoxin lies in prophylaxis; when properly adminis- 
tered, however, it is possible for the antitoxin to aid in the cure of tetanus, 
and its use should never be omitted in the treatment of any case. 
The value of antitoxin in the treatment of tetanus is probably dependent 
upon the following two factors: (1) Neutralization of all free toxin as 
quickly as it is secreted and before it is absorbed by the nervous tissue; 
(2) actual dissociation or neutralization of some of the toxin “loosely united” 
with nerve-cells or suspended in the lymph after it has left the capillaries 
and before it is taken up by the nerve-cells. 
Of interest in this connection is the question whether different serologic 
strains of tetanus bacilli exist. This is particularly important because 
antitoxin is commonly prepared by the immunization of horses with the 
toxin of a single strain. Since antitoxin has proved so efficacious in the 
prophylaxis of tetanus during the recent war, it would appear that the 
antitoxin engendered by one strain will neutralize the toxins of all tetanus 
bacilli. The question, however, is not definitely settled and the investiga- 
tions of Tulloch1 suggest that there may be different strains. 
3. Aside from its chief antitoxic action, antitetanus serum probably 
contains anti-aggressins or bacteriotropins that aid phagocytosis by over- 
coming their repelling or negatively chemotactic influence. This may, 
however, be accomplished by simple neutralization of the toxins, which 
impairs their leukotoxic action sufficiently to permit living leukocytes to 
engulf and destroy the bacilli. 
Methods of Administering Tetanus Antitoxin.—Recent investigations 
and case reports show quite conclusively that in the treatment of tetanus 
as much depends upon the method of administering antitoxin as upon the 
quantity administered. 
1. Absorption by the subcutaneous route is so slow that it should not be 
depended upon in the treatment of tetanus. While it is true that the mor- 
tality of tetanus has been reduced about 20 per cent, by the administration 
of large amounts of serum by this route, it should be emphasized that a 
smaller amount, given subdurally or intravenously, will yield even better 
results. Knorr has shown experimentally that after subcutaneous injection 
the maximum quantity of antitoxin is not found in the blood until twenty- 
four hours have elasped. Since every hour counts heavily in the chances 
for recovery when symptoms of tetanus have appeared, it may be laid down 
as a general rule that the first doses of antitoxin should be given subdurally 
1 Quoted by Bruce, Lancet, 1918, 1, 577. 936 
VACCINES, SERA, BLOOD, PROTEINS IN TREATMENT OF DISEASE 
or intravenously. The subcutaneous route may be chosen when serum is 
given for prophylactic purposes at the lime of injury, but should not be relied 
upon in the treatment of tetanus. 
2. Intramuscular injections may be given to keep up the good effect 
of antitoxin after the first doses have been given subdurally and intra- 
venously, and are to be preferred to the subcutaneous route whenever 
the physician is unable to inject the serum subdurally and intravenously. 
3. For the rapid neutralization of toxin free in the body fluids serum 
should be given intravenously. In the treatment of tetanus from 10,000 
to 20,000 units may be given by this route as early as possible. While 
it is conceded that when the toxin has become firmly bound to the tissues 
of the central nervous system dissociation by the use of antitoxin is not 
possible, yet one can never know, in a given case, how firm this union has 
become, and clinical results, supported by animal experimentation, show 
that life may be preserved by large doses of antitoxin injected into the 
blood. 
4. The experimental work of Permin,1 Park and Nicoll,2 supported by 
the clinical results reported by various observers, shows quite conclusively 
that the subdural route is a very efficacious and valuable avenue by which 
to administer antitoxin in the treatment of tetanus. The serum should be given 
by the gravity method, in exactly the same manner as in giving antimeningitic 
serum. To insure its thorough dissemination throughout the spinal meninges 
the antitoxin should be diluted, if necessary, with normal salt solution. As 
a rule, the amount injected should be slightly less than the amount of fluid 
withdrawn. In the case of a “dry tap,” if the operator is reasonably sure 
of having entered the canal, from 3 to 5 c.c. of serum may be injected. It 
is generally necessary to repeat this injection within twenty-four hours. 
The reason for administering antitoxin subdurally is apparent when it 
is remembered that neither the central nervous system nor the peripheral 
nerves take up antitoxin direct from the blood (Park). Only after very 
large intravenous doses are traces of antitoxin found in the cerebrospinal 
fluid, and animals passively and actively immunized may be rendered tetanic 
if the toxin is injected directly into the central nervous system or into the 
nerve. While antitoxin injected subdurally finally passes over into the 
blood, it will neutralize any free or dissociated toxin before the latter has 
developed any harmful tendency. 
5. To neutralize any toxin that may have been absorbed by a nerve 
it may be advisable to inject antitoxin directly into the nerve, and these 
inlraneural injections under anesthesia are advised by Ashhurst, John, and 
others as part of the rational treatment of tetanus. 
Indications for Treatment.—Every patient, and especially with a history 
of wound, who presents the symptoms of a stiff neck and rigidity of the 
masseter muscles on attempting to open the mouth, general muscular 
rigidity, and especially of the recti muscles of the abdomen causing a board- 
like resistance to the touch, coming on suddenly and not otherwise definitely 
accounted for, should receive immediate intraspinal and intravenous injec- 
tions of tetanus antitoxin, as as appropriate surgical and general treat- 
ment. 
General Treatment.—A large number of substances have been advocated 
in the treatment of tetanus; of these, the most common are injections of 
phenol and intraspinal injections of magnesium sulphate. While phenol 
may be well tolerated by tetanus patients, Ashhurst and John believe that 
1 Mitt. a. d. Grenzgeb. d. Med. u. Chir., 1913, xxvii, 1. 
2 Jour. Amer. Med. Assoc., 1914, lxiii, 235. TREATMENT OF TETANUS 
937 
all these treatments are of little value, and that spinal injections of magne- 
sium sulphate are dangerous. 
1. Chloral hydrate and potassium bromid should be given by mouth 
or by rectum, in sufficient quantities to produce sleep and quiet. Drugs, 
such as those of antagonistic physiologic activity, are more or less successful 
and frequently of aid when given in conjunction with the antitoxin. 
Stone1 has recommended for these purposes the oral or rectal adminis- 
tration every three to six hours of 3 to 5 grains of phenol-barbital (luminal); 
or chlorbutanol in dose of 15 to 30 grains in diluted alcohol or whisky every 
three to six hours. Morphin, \ grain, with atropin, 1/100 grain, may be 
necessary, but Nicoll2 states that conservatism in the administration of 
depressing sedatives is extremely advisable. The patient should be spared 
from every possible source of irritation, from noise, bright light, and fussy 
attention. 
2. While combating the disease, the general care of the patient should 
not be forgotten. A purgative should be administered early. Simple, 
nourishing, non-stimulating food should be given by the mouth, if possible, 
or by the nasal tube, if necessary. Absolute quiet should be maintained. 
Distention of the bladder from retention of urine should be guarded against. 
If water is not well absorbed, and especially if there is peritoneal or pelvic 
inflammation, saline injections into the colon should be given. 
Surgical Treatment.—1. The site of the wound should be located, and 
if possible incised under ether or chloroform anesthesia and thoroughly 
cleansed of foreign material and necrotic tissue. It should then be swabbed 
with the 3 per cent, alcoholic solution of iodin, washed with hydrogen 
dioxid solution, and packed loosely with gauze soaked in the iodin solution. 
2. Cauterization with pure phenol, followed by alcohol, may be em- 
ployed, but, as a rule, the weaker germicide is preferable in order not to 
produce necrosis of the tissues, which furnishes pabulum for bacterial 
growth. The wound should be dressed daily. 
The technic of these injections has been described in Chapter XXXVII. 
Serum Treatment; Administration of Antitoxin.—1. Administer 1 c.c. of 
antitoxin subcutaneously, and if no anaphylactic symptoms occur, give 
intravenously from 10,000 to 20,000 units of antitoxin one hour later, and 
repeat the dose if no effect is apparent or if the good effect wears off in about 
six to eight hours (the technic is described in p. 845). Give similar intra- 
venous injections on the second and third days, and the good effects may 
be maintained by direct intramuscular injections of from 5000 to 10,000 
units for one or two additional doses on subsequent days. 
2. From 10,000 to 12,000 units should be given inlraspinally by means 
of lumbar puncture. This dose should be repeated every eight hours for 
three or four doses unless the symptoms have markedly ameliorated. The 
technic of this injection is described on p. 850. A quantity of cerebrospinal 
fluid should be removed before the serum is injected. After the first injec- 
tion the fluid may be found to have become cloudy, with a large increase 
of cells, especially of the polynuclear variety, although bacteriologically it 
may be sterile. This outpouring of leukocytes is probably a reaction to 
the irritant effect of the serum, and especially of the preservative it contains. 
The intraspinal injections are given under light chloroform anesthesia 
if necessary. 
Children require almost as much antitoxin as adults. Intraspinal 
injections should never be omitted, although in young children intravenous 
1 Jour. Amer. Med. Assoc., 1922, 78, 1939. 
2 Jour. Amer. Med. Assoc., 1921, 76, 112. 938 
VACCINES, SERA, BLOOD, PROTEINS IN TREATMENT OF DISEASE 
injections may not be possible, in which case intraspinal and intramuscular 
injections are given. In young infants serum may be injected intravenously 
through the longitudinal sinus. 
Treatment of the average cases of tetanus in persons twelve or more 
years old will require from 120,000 to 200,000 units of antitoxin as follows: 
First day: 10,000 units intraspinally and 20,000 units intravenously. 
Repeat both injections eight hours later. 
Second day: 10,000 units intraspinally and 20,000 units intravenously. 
Repeat both injections eight hours later. 
Third day: 10,000 units intravenously. If necessary 10,000 units intra- 
spinally. 
Fourth day: 10,000 units intramuscularly. 
In some cases the first day treatment with antitoxin is sufficient; with 
others, in which treatment is delayed, more antitoxin may be required. 
Results of the Antitoxin Treatment of Tetanus.—As was previously 
stated, the prophylactic value of tetanus antitoxin has been proved beyond 
any reasonable doubt. This does not imply, however, that the simple intro- 
duction of 1000 units of antitoxin beneath the skin will surely protect the 
patient, as the percentage of cases developing tetanus even after the serum 
has been given is altogether too high. As was pointed out under Prophy- 
lactic Treatment, the wound must receive thorough surgical attention, and 
the antitoxin must be injected in such places where it will have the greatest 
opportunity to neutralize the toxin. Even if tetanus should develop under 
these conditions, it is likely to be mild and the prognosis would be much 
more favorable. 
Tetanus antitoxin has likewise been very successful in veterinary prac- 
tice, especially after castration and other operations, in injuries, and among 
horses used for the purpose of producing diphtheria antitoxin and other 
immune serums. 
While the curative value of tetanus antitoxin has not come up to expecta- 
tions, more recent carefully prepared statistics indicate that, with serum leeal- 
menl, the mortality is reduced at least 20 per cent, as shown by an investigation 
made by Irons.1 This includes cases treated by the subcutaneous injection 
of antitoxin, and it must be emphasized that, in order to secure the best 
results, tetanus must be treated in a rational manner according to its pathol- 
ogy. Under these circumstances we can confidently expect a greater reduc- 
tion in mortality. But at any rate no physician should withhold antitoxin 
in the treatment if there are any possible means of obtaining it. If only 
3000 units may be had, it is far better to inject this amount intraspinally 
than subcutaneously. 
Permin,2 in a thorough review, reaches the general conclusion that 
antitetanus serum has reduced the mortality of tetanus approximately 
20 per cent. He gives figures from Denmark that are especially valuable, 
because they were gathered from a small area, and hence represent fairly 
uniform conditions: Of 199 cases not receiving serum, only 21 per cent, 
recovered; whereas of 189 cases treated with serum, 42.3 per cent, recovered. 
Of 92 acute cases with an incubation period of less than ten days 24.2 per 
cent, recovered when serum was used, whereas of 94 cases treated without 
serum only 5.3 per cent, recovered. It is significant that these Danish 
figures correspond closely to the American statistics and those of other 
countries. Irons3 has recently tabulated the results of 225 cases of tetanus 
1 Jour. Infect. Dis., 1914, 15, 367. 
2 Mitt. a. d. Grenzgeb. d. Med. u. Chir., 1913, xxvii. 
3 Jour. Amer. Med. Assoc., 1914, lxii, 20, 25. SERUM TREATMENT OF GAS GANGRENE 
939 
treated with antitoxin collected from hospitals and private records for the 
years 1907 to 1913. The mortality in all cases receiving serum was 61.77 
per cent.; in 21 cases without serum the mortality was 85.7 per cent. The 
latter figures correspond quite closely with the general mortality of about 
85 per cent, of tetanus treated without serum. Irons’ figures also show the 
influence of large doses of antitoxin; of 57 cases receiving a small dose of 
antitoxin (3000 units or less subcutaneously), the mortality was 73.7 per 
cent.; of 143 cases receiving large doses (over 3000 units subcutaneously, 
or 3000 or less intraspinally or intravenously), the mortality was 57.3 per 
cent. Magnesium sulphate was given intraspinally in 18 cases which also 
received serum; in 4 cases—2 acute and 2 chronic—the patients recovered, 
giving a mortality for the group of 77 per cent. In 2 cases death occurred 
shortly after injection with symptoms of respiratory failure. 
The greater the interval elapsing between the onset of symptoms and 
the time treatment is begun, the greater the mortality. 
Likewise whether or not a prophylactic injection of antitoxin was given 
at the time of injury, has considerable influence on the outcome. For 
example, in the British Army during the recent war, the mortality was 34.8 
per cent, among those developing tetanus after a prophylactic injection of 
antitoxin and 71.2 per cent, among those not immunized. 
The virulence of tetanus infection apparently varies from year to year 
and modifies the mortality rates. Stone,1 for example, reports that in the 
Los Angeles County Hospital the mortality during four years from 1916 to 
1921 varied from 14.3 to 71 per cent, with the same general plan of treatment. 
All recent writers on the treatment of tetanus have emphasized the value 
of large doses of antitoxin by intravenous and intraspinal injection, as 
Fitzgerald,2 Ross and Stirrett,3 Bruce,4 Golla,5 Andrewes and Horder,6 
Gibson,7 Bacti,8 Robertson,9 Woolf,10 and others. 
In view of this evidence in favor of antitoxin in the treatment of tetanus 
it is apparent that the physician is compelled to give every patient with 
tetanus the opportunity to obtain this 20 per cent, or more benefit by ad- 
ministering the serum promptly and correctly. 
Non-specific Treatment of Tetanus.—Liidke11 reports that the treatment 
of 7 cases of acute tetanus with intravenous injections of 3 to 5 c.c. of a 10 
per cent, solution of deutero-albumose at intervals of twenty-four to forty- 
eight hours, resulted in greatly diminishing the intensity of the spasms and 
recovery of all. Kaznelson12 treated 2 cases in a similar way with albumose 
injections, with 1 recovery. 
The bacteriology of gas gangrene and prophylactic value of antigas 
gangrene serum, have been discussed in Chapter XXXVIII. During the 
Great War a large number of these infections occurred resulting in consider- 
able investigation in reference to etiology and treatment. At the close of 
the war it was the consensus of opinion that an efficient means of serum pro- 
phylaxis and treatment had been finally worked out. 
Surgical treatment consisting of thorough cleansing and drainage of 
the wound to remove the favorable soil for bacteria and particularly the 
SERUM TREATMENT OF GAS GANGRENE 
1 Jour. Amer. Med. Assoc., 1922, 78, 1939. 
2 Brit. Jour. Surg., 1916, 4, 14. 
3 Canad. Med. Assoc. Jour., 1915, 5, No. 4. 
4 Lancet, 1917, 1, 680. 
5 Lancet, 1917, 1, 686. 
6 Lancet, 1917, 1, 682. 
7 Amer. Jour. Med. Sci., 1916, 152, 781. 
8 Bull. d. l’Acad. d. m£d., 1916, 86, 316. 
9 Amer. Jour. Med. Sci., 1916, 151, 781. 
10 Jour. Amer. Med. Assoc., 1922, 72, 1266. 
11 Berl. klin. Wchn., 1920, lvii, 344. 
12Therap. Monatsh., 1917, 31, 437. 940 
VACCINES, SERA, BLOOD, PROTEINS IN TREATMENT OF DISEASE 
anaerobes, and to encourage the inflow of fresh lymph and facilitate phagocy- 
tosis, is of maximum importance. 
The serum should be polyvalent, that is, prepared by the immunization 
of horses not only with the toxins of Bacillus welchii (B. perfringens) but 
with B. edemaciens and Vibrion septique as well. 
The serum should be given as early as possible and before profound 
toxemia has developed. The administration should be by intravenous 
injection of 50 to 100 c.c. diluted with saline solution and repeated every 
six to eight hours until three or four doses have been given. Serum may 
also be injected in the tissues about the wound in order to effect neutraliza- 
tion of the toxins at the site of production. 
Bull,1 Vincent,2 McGlannan,3 Bouchet,4 Gibbon,5 Van Beuron,6 and others 
have rendered favorable reports, although available statistics are yet too 
meagre to- warrant further statements at this time. The polyvalent serum 
would appear to be worthy of thorough trial in the treatment of this highly 
fatal disease. 
Serum Treatment of Bacillary Dysentery.—Soon after the discovery of 
a bacillus of dysentery by Shiga in 1892 the treatment of bacillary dysentery 
by the use of immune serums was undertaken. Following the discovery 
of the etiologic importance of Shiga’s bacillus in the dysenteries of Asiatic 
countries, similar investigations were made in various parts of the world 
and various bacilli were isolated. At first these micro-organisms were all 
regarded as being identical, but further investigation has shown that marked 
differences are apparent, and two main types are now recognized: one type 
(Shiga) does not ferment mannite and produces a soluble or extracellular 
toxin, and a second type (Flexner-Harris, Hiss, etc.) ferments mannite and 
does not produce an extracellular toxin. A further discussion of these bacilli 
will be found in Chapter VII. 
Antidysentery Sera.—The Shiga dysentery bacillus produces a soluble 
or exogenous toxin and the antiserum is largely antitoxic. This serum is 
ineffectual for dysentery caused by the Flexner group of bacilli. 
The Flexner bacillus produces largely an endotoxin; its antiserum is 
largely bactericidal and is ineffectual against Shiga infections. 
A polyvalent serum may be prepared by mixing these or by immunizing 
horses with both strains and toxins, as is the usual custom. 
Dysentery caused by the bacilli of the Kruse-Shiga type may be regarded 
as a form of intoxication analogous to the intoxication of diphtheria. The 
intestine, where the bacilli lodge, corresponds to the throat, which is the 
site of infection in diphtheria; here the bacilli develop and produce their 
toxins, and these toxins, when absorbed into the circulation, in turn produce 
the symptoms of the disease. 
Antitoxin has been prepared for the bacilli of the Kruse-Shiga type, and 
these have yielded fairly satisfactory results in the prophylaxis and cure of 
this variety of bacillary dysentery. Antiserums for the mannite fermenting 
group of bacilli (Flexner, Harris, Hiss, Duval, etc.) have not proved of as 
much value in the treatment of these infections. Bacilli of the latter group 
are largely responsible for the dysenteries in this country, and also for a 
percentage of cases of ileocolitis of infancy. Since the antiserum of the 
Shiga bacillus is of practically no value in the treatment of infections caused 
TREATMENT OF BACILLARY DYSENTERY 
1 Brit. Med. Jour., 1918, 1, 62. 3 Surg., Gyn., and Obst., 1918, 26, 336. 
2 Lancet, 1918, 2, 178. 4 Bull. d. l’Acad. d. med., 1919, 81, 556. 
6 Amer. Jour. Med. Sci., 1919, 157, 764. 
6 Jour. Amer. Med. Assoc., 1919, 73, 239 (includes an excellent review of the subject). by bacilli of other groups, the serum treatment of dysentery is employed 
mainly in European and Asiatic countries, where infections with this group 
are common. After fairly extensive trials in this country the serum treat- 
ment of infantile diarrheas and true dysenteries has proved disappointing. 
The preparation and standardization of dysentery antitoxin is described 
in Chapter XIII. 
Administration and Uses.—Dysentery antitoxin has been used both 
in the prophylaxis and in the cure of this infection. The doses of serum 
advised by various observers vary considerably, owing to the marked 
differences that exist in the potency of these serums. Since the various 
manufacturers do not employ the same standards, the physician should 
use the serum in accordance with the printed directions that accompany 
each package. 
For curative purposes, Shiga has advised 10 c.c. for mild cases and from 
20 to 60 c.c. for severer cases. It may be necessary to repeat the injections 
several times. Vaillard and Doyle have given as much as from 80 to 100 
c.c., and have repeated this dose on the following days. When the serum is 
being used during an epidemic, it is advisable to ascertain beforehand the 
nature of the infection, as the antiserum for the Shiga bacillus is highly specific 
and is not likely to prove of value in the treatment, of infections caused by the 
Flexner type of bacillus. Otherwise a polyvalent serum should be used. 
The injections have usually been given subcutaneously. Better results 
would, no doubt, be obtained in the treatment of severe infections by ad- 
ministering large doses of serum intravenously. 
Flexner1 has stated “that in cases of ordinary severity, a single sub- 
cutaneous dose may be followed by such marked alleviation of the symptoms 
as not to call for repetition. In other and severer cases the dose may need 
to be repeated in from twelve to twenty-four hours, and again in forty-eight 
hours. 
“The effects of the injection of the serum tend to appear promptly. They 
consist first in the amelioration, often within a few hours, of the nervous 
symptoms and the general prostration. Usually within twenty-four hours 
the tenesmus and colic disappear and the stools become markedly reduced 
in number. Along with this improvement there goes diminution in the 
blood and mucus content of the discharges, coincident with which there is a 
return of their feculent character. According to the severity of the attack, 
the stools return to normal in from two to five days. 
“Acute cases of bacillary dysentery are especially subject to the serum 
treatment, but cases in their second to third week may still be favorably 
influenced, as may also relapses in the course of convalescence from acute 
attacks. On the other hand, chronic cases and the recrudescences in the 
course of chronic dysentery, even if originally induced by the dysentery 
bacilli, do not as a rule respond to the serum. 
“It should be added that the employment of the serum does not con- 
traindicate the usual eliminative and dietetic forms of treatment.” 
Results.—Adequate statistics regarding the value of the serum in the 
prophylaxis and treatment of dysentery are favorable in so far as Shiga 
infections are concerned. From the prophylactic standpoint, encouraging 
results have been reported by Kruse, Shiga, Vaillard and Dopter, Rosculet, 
and others, and it would appear that passive immunization is of value in 
combating localized outbreaks, such as occur in institutions and armies. 
From the curative standpoint, most observers agree that the use of a 
potent serum will reduce the mortality of acute cases at least from 30 to 
TREATMENT OF BACILLARY DYSENTERY 
941 
1 Jour. Amer. Med. Assoc., 1921, 76, 108. 942 
VACCINES, SERA, BLOOD, PROTEINS IN TREATMENT OF DISEASE 
50 per cent. The mortality has varied widely from 5 to 50 per cent, in 
different outbreaks and epidemics. As a general rule the Shiga infections 
are more severe than the Flexner infections, but this relation may be re- 
versed. Flexner states that in mild outbreaks with a mortality of 10 to 
15 per cent., serum treatment may reduce this rate to 1 to 2 per cent. In 
severer epidemics serum treatment may reduce the mortality from 50 per 
cent, to about 10 per cent. Shiga reports a drop in the mortality in Japan 
of from 22 to 26 per cent, to 9 to 12 per cent.; Kruse obtained a reduction in 
mortality of about 10 per cent. Vaillard and Dopter1 treated 96 cases, 
with but 1 death; Rosenthal2 treated 157 cases with 7 deaths—a mortality 
of 4.5 per cent, as compared with that of 10 to 11 per cent, occurring in other 
German hospitals. Coyne and Auche3 treated 11 cases due to the Flexner 
type of bacillus and report good results. Ruffer and Willmore,4 Violle,6 
Rogers,6 Brau,7 and Baht8 have also reported favorably upon the use of 
the serum. 
During the recent Great War bacillary dysentery proved a frequent and 
troublesome infection and particularly in 1917-1919. The Flexner infec- 
tions appeared to predominate. Nolf2 reports that serum treatment failed 
whereas vaccine treatment and particularly by intravenous injection, 
yielded very good results (see later). Klein,10 however, reported good results 
when 60 to 100 c.c. of serum was injected intravenously early in the disease. 
Finlayson,11 Waller,12 Pribaum,13 Klesk,14 and others have published generally 
favorable reports. 
A highly potent and polyvalent serum should be in readiness during war, 
and particularly in camps and hospitals. In a thorough investigation made 
in the United States in 1903 by the Rockefeller Institute, under the direction 
of Flexner, it was found that the results of the serum treatment of ileocolitis, 
among children at least, were quite disappointing. Josephs and Davison15 
for example, found that the intramuscular and subcutaneous injections of 
from 20 to 50 c.c. of serum did not reduce the mortality or influence the 
course of the disease in a series of acute dysenteries among 20 children in 
this country (17 being Flexner and 3 Shiga infections). This is largely due to 
the fact that several different strains of bacilli may be the cause of an infec- 
tion, and unless a corresponding antiserum is employed for the particular 
type causing the infection in a given case, good results cannot be expected. 
Probably if some means were discovered for making a prompt bacteriologic 
diagnosis, and if several immune serums were on hand for the treatment of 
infections caused by the main types, after the methods worked out by 
Neufeld, Dochez, and Cole in the treatment of pneumonia, better results 
may be obtained. 
1 Ann. de PInst. Past., 1906, xx, 321; 1907, xxi, 241. 
2 Deutsch. med. Wchn., 1904, xxx, No. 19. 
3 Compt. rend. Soc. de biol., 1906, lx, No. 26. 
4 Brit. Med. Jour., 1908, 11, 1176; ibid., 1909, ii, 462. 
6 Arch, de mid. et pharm. nav., 1912, xcviii, 61. 
6 Dysenteries, Their Differentiation and Treatment, London, 1913, 290. 
7 Ann. d. hyg. et m£d., 1913, xvi, 710. 
* Brit. Med. Jour., 1914. 
9 Jour. Amer. Med. Assoc., 1919, 73, 1177. 
10 Lancet, 1919, 2, 775. 
“Brit. Med. Jour., 1917, 1,46. 
“Lancet 1919 2 778. 
13 Centralb. f. Bakteriol., 1918, 80, 33; ibid., 1918, 81, 37. 
14 Med. Klin., 1915, 11, 1147. 
15 Jour. Amer. Med. Assoc., 1921, 77, 1863. (This paper gives a good bibliography of the 
serum treatment of dysentery.) TREATMENT OF BACILLARY DYSENTERY 
943 
According to Davison1 the oral and rectal administration of 5 to as 
much as 1381 c.c. of D’Herelle’s dysentery bacleriolysani had no influence 
upon the course or mortality of bacillary dysentery in children. This 
material is prepared by cultivating a small portion of feces (dysenteric 
convalescent or normal) in bouillon and filtering through a Berkefeld filter; 
when the filtrate is added to cultures of dysentery bacilli they are killed and 
dissolved. 
The curative effect of dysentery antitoxin is shown by a reduction in the 
number of stools, by the fact that blood and pus disappear from the dis- 
charge, pain and tenesmus are relieved, the temperature becomes normal, 
and the patient gains in weight. Individual observers are frequently 
enthusiastic over the results obtained in individual cases, and no doubt 
these are striking in those instances where the antiserum appears to be 
specific for the particular form of infection. 
Vaccine Treatment of Bacillary Dysentery.—Vaccines by Subcutaneous 
Administration.—In the very acute choleric types characterized by violent 
diarrhea and vomiting with rapid dehydration of the tissues, vaccines have 
not generally been employed; in these the subcutaneous injection of 1 to 2 
liters of saline solution and fluid diet has proved a successful plan of treat- 
ment. 
In subacute ulcerative dysentery due to Flexner bacilli, Nolf and his 
associates2 have reported good results from the subcutaneous injection of 
heat-killed autogenous vaccines beginning with a dose of 10,000 and gradually 
increasing to 5,000,000,000 to 10,000,000,000, which were well tolerated by 
most individuals. 
In old chronic cases Nolf3 has likewise reported favorably upon the 
results of treatment with heated vaccines beginning with a dose of 1,000,000 
and increasing until 5,000,000,000 to 10,000,000,000 are being injected at 
one time. “In every case the general condition was improved, and the 
intestinal symptoms steadily decreased. In the majority the cure was 
complete and definite. At times there seemed to be complete cure at the 
end of treatment, but at a later period the symptoms returned. In some 
cases the stools, though regular and only one or two a day and without 
blood and mucus, yet remained soft, and there persisted a little intestinal 
instability and discomfort.” 
Vaccines by Intravenous Administration.—Nolf has reported even more 
favorably upon the results of the intravenous administration of heat-killed 
vaccines: 
“The doses were given at four-day intervals, the initial dose being 
regularly 10,000 germs, then 30,000, then 50,000, then 100,000, etc. In 
general, the betterment of the patient did not long delay. The fever dropped 
by lysis, with some recrudescences more or less marked on the days of the 
vaccine therapy and the next day; and the intestinal symptoms improved 
coincidently. In many cases of moderate intensity a complete cure was 
effected when the dose of 500,000 was reached. In the more refractory 
cases it was necessary to push the vaccine up to about 10,000,000. 
“In 52 cases treated in this way we had only 2 deaths. All the other 
patients left the hospital cured except 2, whom military necessity forced us 
to send away too soon. We have no doubt that in these 2 cases also the 
continuation of the treatment would have resulted in a cure in a relatively 
short time. By vaccinotherapy we were thus able to avoid the dangerous 
1 Amer. Jour. Dis. Child., 1922, 23, 531. 
2 Arch. m6d. beiges, May, 1918. 
3 Jour. Amer. Med. Assoc., 1919, 73, 1177.  VACCINES, SERA, BLOOD, PROTEINS IN TREATMENT OF DISEASE 
944 
tendency toward chronicity which in 1917 was produced in a considerable 
number of our patients. This last result we considered particularly gratify- 
ing. 
“The complete record of the epidemic of bacillary dysentery of 1918 
shows a complete cure, at the latest in a few weeks’ time, in 500 cases, 
except only 2 patients who died, and 2 who left before the cure was com- 
pleted.” 
Non-specific Treatment of Bacillary Dysentery.—Doubtless part of the 
beneficial results observed by Nolf in the treatment of bacillary (Flexner) 
dysentery with intravenous injections of vaccine are to be attributed to 
non-specific effects. 
Adler1 and Dollken2 have reported favorable results from the intra- 
muscular injection of 3 to 5 c.c. of sterile milk and especially as a means for 
treating intestinal hemorrhage; this form of treatment appears very worthy 
of trial in conjunction with other therapeutic measures. 
Dollken, Liidke, and Holler have also employed intravenous injections 
of 10 per cent, solution of deutero-albumose in dose of 1 to 2 c.c. for three to 
six injections in the treatment of dysenteries mostly caused by the Shiga 
bacillus; of 12 cases treated in this manner by Ludke, good results were 
secured in 10. 
TREATMENT OF MENINGOCOCCUS MENINGITIS (CEREBROSPINAL FEVER) 
Types of Meningococci.—Prior to 1909 meningococcus meningitis was 
regarded as caused by a single strain of the meningococcus. In 1909 
Dopter3 discovered the para-meningococcus, differentiable from the menin- 
gococcus by immunologic reactions and especially by the agglutination 
reaction. In 1915 Gordon and Murray4 classified all meningococci into 
four groups by means of agglutination tests. According to Flexner, Gordon’s 
Type I appears to correspond with the “parameningococcus” of Dopter, 
and Type II with the normal or regular meningococcus. Types III and IV 
appear to conform to the more common intermediates. According to 
Nicolle’s classification, his Type A corresponds to Gordon’s Types I and III 
and his Type B to Gordon’s Types II and IV. The differentiation is made 
by agglutination tests, the technic being described in the chapter on Agglu- 
tinins. It is very much hoped that an international classification may be 
adopted. 
Worster-Drought and Kennedy5 have classified the meningococci from 
42 cases according to Gordon’s types with these results: 
(a) 10, or 23 per cent., due to Type I; all were severe infections, 6 prov- 
ing fatal. 
(b) 20, or 49 per cent., due to Type II; this coccus was most frequently 
met with both in carriers and in actual cases; 14 recovered and 6 died. 
(c) 10, or 23 per cent., due to Type III; 6 recovered and 4 died. 
(d) 2, or 5 per cent., due to Type IV; 1 case died. 
In a general way it would appear that Types I and III are more virulent 
as regards the meninges than Types II and IV, but both Rolleston6 and 
Adshead7 believe that observations on a larger scale are required before any 
1 Wien. med. Wchn., 1917, 67, 509. 
2 Munch, med. Wchn., 1919, 66, 480. 
3 Compt. rend. Soc. debiol., 1909, 66, 772,1055; ibid., 1909, 67, 74; ibid., 1909, 69, 600. 
4 Jour. Roy. Army Med. Corp., 1915, 25, 410, 456. 
5 Cerebrospinal Fever, A. & C. Black, London, 1919, 270. 
6 Lancet. 1919, 1, 541. 
7 Med. Research Committee, Special Report Series, No. 17. TREATMENT OF MENINGOCOCCUS MENINGITIS 
945 
statement can be made on the possible relationship of these strains to the 
severity of the disease. 
The Mechanism of Infection of the Meninges.—The path by which 
meningococci reach the meninges has been thought to be (a) direct invasion 
of the cerebral meninges by way of the lymphatics from the nasopharynx 
and accessory nasal sinuses; (b) infection along the lymphatics of the spinal 
nerve roots; and (c) primary invasion of the blood-stream from the naso- 
pharynx with secondary localization of the meningococci in the meninges. 
Westenhoeffer1 advanced the theory that infection took place through 
the sphenoidal sinuses and Netter and Debre through the ethmoid sinuses. 
Elser and Huntoon,2 Worster-Drought and Kennedy,3 and others have been 
unable, however, to find any evidence in support of the sphenoid pathway, 
and the latter point out that when sphenoid suppuration occurs it may be 
secondary. 
Flexner4 in 1917 considered that meningococci probably passed along 
the lymphatics around the olfactory nerves to the meninges, but Austrian’s5 
experiments yielded negative results. 
The theory that infection may occur along the lymphatics of spinal 
nerves with the primary foci of meningococcus infection in the intestines, 
lungs, or cervical glands, was put forward particularly by McDonald,6 but 
has since been discredited by Symmers7 and others, and now commands 
but little attention. 
Meningococcus Bacteremia.—Herrick8 has recently called particular atten- 
tion to the occurrence of meningococci in the blood-stream and that by 
special cultural methods they may be found in 50 to 80 per cent, of cases 
examined at an early stage and frequently before meningitic symptoms are 
well developed. In some cases the bacteremia may apparently exist without 
the development of meningitis. Baeslack9 obtained positive results in 
36 per cent, of early cases, but the general average of positive blood-cultures 
is about 25 per cent. Bloedorn10 gives a good review of the literature and 
reports a case. 
There cannot be any doubt that meningococcus bacteremia occurs in 
many cases, although it tends to be intermittent and temporary and may 
require a series of blood-cultures for its detection. This blood infection, 
however, is important in relation to serum therapy as well as in relation to 
the mechanism of infection of the meninges. 
Elser and Huntoon, Austrian and Herrick believe that meningitis is 
produced first by blood invasion with secondary involvement of the meniges 
by way of the choroid plexes which either filters out the cocci and is first 
infected or by selective localization of the cocci in the pia mater. 
It is probable that the pathway of infection varies in different cases. 
As Huntoon and Elser have pointed out infection may occur by extension 
from the ear, mastoid, or sinus infection when the onset is gradual, just as 
pneumococcus meningitis is so frequently produced. In the majority of 
instances, however, and especially in the epidemic form of the disease with 
sudden onset, infection appears to take place by passage of meningococci 
1 Berl. klin. Wchn., 1905, 62, 737. 
2 Jour. Med. Research, 1909, 20, 517. 
3 Lancet, 1917, 2, 711. 
4 Jour. Amer. Med. Assoc., 1917, 69, 639. 
5 Bull. Johns Hopkins Hosp., 1918, 29, 183. 
6 Rev. Neurol, and Psych., Edinb., 1907, 5, 702. 
7 Lancet. 1908, 2, 472. 
8 Arch. Int. Med., 1918, 21, 541; Jour. Amer. Med. Assoc., 1918, 71, 612. 
9 Jour. Amer. Med. Assoc.. 1918, 70, 684. 10 Amer. Jour. Med. Sci., 1921, 172, 881. 946 
VACCINES, SERA, BLOOD, PROTEINS IN TREATMENT OF DISEASE 
from the nasopharynx into the lymphatic and vascular channels with 
invasion of the blood-stream and localization in the meninges. As stated 
by Bloedorn, “it would appear that cases of meningococcus septicemia are 
becoming more frequent, or at least more frequently recognized, and that 
it is well to be on the alert for such cases in order that the diagnosis may be 
made and treatment instituted before the development of meningitis or 
endocarditis. The appearance in an acute febrile case of petechiae or a 
maculopapular eruption resembling at times the rose spots of typhoid 
fever, together with a moderate leukocytosis and a temperature chart, which 
may either be of the septic type or may resemble markedly the chart of 
malarial fever, should riase the suspicion of meningococcus sepsis and 
should lead to repeated blood-cultures being made in an effort to identify 
the meningococcus. The skin lesions of meningococcus sepsis are par- 
ticularly striking. The eruption may be purpuric, hemorrhagic, or maculo- 
papular. In the acute fulminating cases, which usually die within the 
first few days, a purpuric generalized eruption is the rule. These cases 
are quickly overwhelmed and die frequently before the development of 
meningitic symptoms. In the protracted cases of meningococcus sepsis 
the character of eruption shows more of a tendency to assume the maculo- 
papular or hyperemic type.” 
The main symptoms and lesions of the disease, and several of the com- 
plications, for example, the paralyses, eye complications, deafness, hydro- 
cephalus, and mental disturbances, are probably directly due to suppuration 
of the meninges, with involvement of accessory and motor nerve-roots, 
meningeal irritation, and pressure from the accumulation of exudate in the 
ventricles and subarachnoid space. Complications, such as arthritis, 
pyelitis, endocarditis, adenitis, etc., are due to the bacteremia, which may 
become chronic and be accompanied by deposits of meningococci in the 
various tissues and organs. In addition to these complications there is 
probably a varying degree of general toxemia, due to a soluble toxin or endo- 
toxin liberated through lysis of the cocci. 
Although cerebrospinal meningitis may be considered primarily as a 
general infection, in the majority of instances local suppuration of the 
meninges constitutes the main lesion. For anatomic and physiologic reasons, 
however, it is impossible to treat the disease according to the ordinary 
principles governing the treatment of localized suppuration, as, for example, 
by continuous drainage and by cleansing the affected parts with germicidal 
solutions. Unaided, the leukocytes and body fluids are generally unable to 
destroy the cocci and terminate the infection before serious harm to impor- 
tant nerves and nerve-centers has resulted, so that epidemic meningitis, 
with a mortality of from 75 to 90 per cent., and followed by more or less 
serious sequelae, which few survive, is one of the most dreaded of infections. 
In administering antimeningitic serum we aim to assist the patient’s 
leukocytes and body fluids to overcome the infection. Repeated spinal 
punctures remove portions of the infective material mechanically, but 
the greatest dependence in bringing about quick destruction of the cocci 
and effecting recovery of the patient is to be placed upon the serum. 
The Serum Treatment of Meningococcus Meningitis.—Historical.— 
During the pandemic of meningococcal cerebrospinal meningitis in 1904-05 
several laboratories sought to produce an immune serum for the purpose of 
treating human cases of this infection. 
After pursuing experimental studies on the subject on the lower animals, 
Jochmann,1 in 1905, immunized a horse and used the serum in the treatment 
1 Deutsch. med. Wchn., 1906. xxii, No. 20, 788. TREATMENT OF MENINGOCOCCUS MENINGITIS 
947 
of 38 cases of epidemic meningitis. At first he employed the subcutaneous 
method of injection, and later he used the intraspinal method. The results 
were quite encouraging, and during the following year 30 more cases were 
treated, with a resulting mortality of 27 per cent., as against a mortality of 
53 per cent, in untreated cases. 
At about the same time Kolle and Wassermann1 reported that they had 
prepared an antimeningococcus serum, which had not, however, up to that 
time been used in the treatment of human infections. A year later serum 
was administered subcutaneously and then intraspinally, with encouraging 
results. 
In this country Park had, in 1905, prepared an antimeningococcus 
serum, which was used in the treatment of 20 cases in Hartford, Conn., by 
subcutaneous injection, but without beneficial results. Jochmann had, in 
the mean time, shown the superiority of intraspinal injections, and this 
method soon supplanted the subcutaneous method. 
In 1905 Flexner began a series of studies regarding experimental menin- 
gococcus infections in the lower animals, and the therapeutic value of anti- 
meningococcus serum. These valuable experiments attracted the attention 
of the world, and placed this method of treatment upon a firm basis. In 
1906 Flexner2 proved that a specific immune antimeningococcus serum could 
be produced that, if injected intraspinally, would save the lives of monkeys. 
Later horses were immunized and the serum used in the treatment of human 
infections during an epidemic in Akron, Ohio, in May, 1907. In a short 
time the serum was used extensively in other epidemics, and a report of 
these early cases was made by Flexner3 in 1907. A later report by Flexner 
and Jobling,4 covering the treatment of 400 cases, showed that the mortality 
had been reduced from 75 per cent, to below 30 per cent. In 1909 they 
reported5 upon 712 cases treated with the serum, with a mortality of 31.4 
per cent. In a more recent report Flexner6 reviewed all the cases—1300 
in number—gathered from all parts of the world, treated with serum pre- 
pared in the Rockefeller Institute. The general mortality rate is given as 
30.9 per cent., as against 75 to 80 per cent, among cases not receiving serum 
treatment. Of 1394 cases treated with serum during the Texas epidemic7 
(1912), the mortality was 37 per cent., as compared with a mortality of 77 
per cent, among 562 cases receiving no serum. 
Good results have been reported by many observers with Jochmann’s, 
Kolle and Wassermann’s, Ruppel’s, Paltauf’s, and Dopter’s serums, and the 
serums have been prepared by several commercial biologic laboratories, so 
that the curative value of antimeningococcus serum is definitely established. 
Preparation of the Antimeningococcus Serum.—Method of Flexner and Jobling.— 
1. Many strains of meningococcus are used in order that a polyvalent serum may be pre- 
pared. Fresh strains from new epidemics and sporadic cases are constantly added. “Fast” 
strains, or those isolated from cases in which the serum has produced no beneficial effect, are 
especially desirable. For polyvalent sera the more strains employed, the better the serum; 
at the Rockefeller Institute as many as 50 different strains have been employed. As shown 
by Amoss, Gates and Olitsky,8 horses immunized with five strains including the different 
types of meningococci, produce antibodies for the various strains, but these sera do not keep 
1 Deutsch. med. Wchn., 1906, xxxii, No. 16. 
2 Jour. Amer. Med. Assoc., 1906, xlvii, 560. 
3 Jour. Exper. Med., 1907, ix, 168. 
4 Jour. Exper. Med., 1908, x, 141. 
5 Jour. Amer. Med. Assoc., 1909, liii, 1443. 
6 Jour. Exper. Med., 1913, xvii, 553. 
2 Sophian: Epidemic Cerebrospinal Meningitis, St. Louis, 1913. Report of Dr. Steiner, 
President of the Texas State Board of Health. 
8 Jour. Exper. Med., 1920, 32, 767. 948 
VACCINES, SERA, BLOOD, PROTEINS IN TREATMENT OF DISEASE 
as well or prove ultimately as therapeutically active as sera prepared by immunizing with 
a large number of strains. Wadsworth,1 however, has recently reported that by immunization 
with a limited number of representative strains, four or six, carefully selected on account 
of their antigenic and agglutination properties, the potency was increased three- to tenfold 
without sacrificing the polyvalency; that is, as tested with at least 70 heterologous strains 
of the meningococcus. 
2. Immunization is performed first with an autolysate of the meningococci, and later 
with living cultures. 
3. Stock cultures are kept alive by transplanting them every four days in slants of 
ascitic glucose agar, neutral to phenolphthalein. 
4. In preparing the autolysate, the cultures are subcultured first on glucose-agar slants 
without serum. After twenty-four hours’ growth about 3 c.c. of salt solution are added to 
each slant, and the culture emulsified. One c.c. is then poured over the surface of glucose- 
agar slants in large 500 c.c. Blake bottles. After twenty-four hours’ incubation heavy 
uniform, and diffuse growths are secured. 
Add 10 c.c. of normal salt solution to each bottle and wash off the culture. If necessary 
a long heavy platinum loop may be used. Each bottle is tested for contamination by stain- 
ing a smear according to the method of Gram. Each bottle is emptied into a common vessel; 
2 per cent, toluol is added, mixed well, and incubated for from eighteen to twenty-four hours. 
The toluol is then allowed to evaporate, or it may be immediately filtered off through sterile 
gauze saturated with salt solution. The preparation is kept in a refrigerator and should be 
prepared fresh every month. 
5. The injections are given subcutaneously about the neck and abdomen. Young and 
healthy horses are selected for the purpose. The first dose consists of 2 c.c. of the auto- 
lysate, and this is gradually increased, depending on the manner in which the animal reacts, 
until 10 c.c. are given in a single dose. Then inject 2 c.c. of living culture diluted with 2 
parts of salt solution, and increase the doses, the same as with the autolysate, until 10 c.c. of 
culture are given at one injection. Next inject living cultures and autolysate alternately, 
until a maximum of from 30 to 35 c.c. are given in one dose; this last is then used as the 
constant dose until the immunization has been completed. 
Injections are given every five to seven days until the large doses are reached, when they 
are given every ten days or two weeks. 
Horses usually show quite marked reactions, such as fever, depression, and induration 
about the site of injection, but if due care is exercised, few animals are lost during the im- 
munization. 
6. Bleedings are usually begun about the fourth month after immunization has been 
instituted. The horses are bled aseptically from a jugular vein about every two weeks, 
from 8 to 8 liters of blood being removed at each sitting. The serum is separated and pre- 
served with trikresol. Recent investigations indicate that trikresol may be partly responsible for 
paralysis of the respiratory centers, and at present every effort should be made to collect and market 
the serum in a strictly aseptic manner, so that none or but very little preservative is required,2 
Efforts are being made at present to discover an efficient volatile antiseptic that may be 
driven off by warming the serum at body temperature. 
Rapid Method of Amoss and Wollstein.3—These investigators have recently advocated 
a rapid method of immunization consisting in inoculating alternately several strains of 
living meningococci and parameningococci and the autolyzed products of each, by which a 
potent polyvalent serum may be produced in eight to twelve weeks instead of in the ten 
months required by the subcutaneous method. In the preparation of a polyvalent serum a 
twenty-four-hour agar slant culture of meningococci is removed with 2 c.c. of salt solution 
and 0.1 c.c. suspended in 15 c.c. of salt solution and injected very slowly into the circulation 
of a horse. The temperature is taken hourly and should not rise over 3° C.; twenty-four 
hours later 0.2 c.c. of suspension, and on the third day 0.3 c.c. are given. After the lapse of 
seven days a series of three injections of living parameningococci are given in doses of about 
0.3 c.c. of the emulsion or sufficient to give a temperature reaction of about 2.5° to 3° C. 
After a lapse of seven days three intravenous injections of an autolysate composed of equal 
parts of autolysate of meningococci and parameningococci are given. A series of three 
injections of living meningococci and parameningococci follow in doses which may be run up 
to 0.6 c.c. of emulsion, with every third series consisting of injections of the mixed autol- 
ysates; after ten to twelve weeks a potent serum is generally produced. In order to make 
the serum as polyvalent as possible a large number of strains of meningococci and para- 
meningococci should be used. 
Severe reactions due to hypersensitiveness of the horse to the meningococci or its prod- 
ucts are especially lilely in the first dose of each series after three or four series of injections 
have been given. In order to desensitize, a portion of the first injections of each series is 
1 Jour. Exper. Med., 1921, 33, 107. 
2 Hall, W.: Bull. No. 91, Hyg. Lab., U. S. P. H. S., 1914. 
3 Jour. Exper. Med., 1916, 23, 403. TREATMENT OF MENINGOCOCCUS MENINGITIS 
949 
injected intravenously, and two hours later the remainder of the dose is given. Danger due 
to agglutinated cocci is lessened by diluting the dose in 15 to 20 c.c. of salt solution and 
injecting very slowly. 
The Preservation of Antimeningococcus Serum.—The subject of proper preserva- 
tion of antimeningococcus serum is one of considerable importance by reason of the untoward 
effects that may be produced when the serum is injected intraspinally. Ordinarily 0.2 to 
0.3 per cent, tricresol is employed. Kramer,1 however, has declared that tricresolized serum 
is dangerous. Hale2 has shown that tricresol injected intraspinally in dogs may produce 
paralysis of the respiratory center, while Auer3 showed that the respiratory changes were 
much less marked in monkeys and due to a stimulation of the inhibitory function. 
Flexner4 and Fitzpatrick, Atkinson and Zingher6 have attributed untoward effects in 
human beings to the pressure exerted by intraspinal injections of serum rather than to the 
presence of tricresol. 
Chloroform has also been employed as a preservative, but as reported by Neal and 
Abramson6 the serum produces so much pain when injected that this method is objectionable; 
furthermore, chloroform is much less bactericidal than tricresol. 
Tricresol, therefore, is being usually employed at the present time in 0.2 to 0.3 per cent. 
Both Hale7 and Leake and Corbitt8 have found tricresol and phenol of about equal toxicity, 
the M. L. D. for mice being approximately 0.00037 gram per gram weight of mouse. Tric- 
resol, however, is more bactericidal. 
Krumwiede and Banzhaf9 have advised using equal parts of cresol and ether as a means 
for preventing the immediate precipitation of the serum proteins and clouding of the serum, 
although subsequent precipitation may occur. Masucci10 found that the mixture of ether and 
cresol had no advantage as a preservative of serum over straight ether in the total precipitate 
formed on standing or in germicidal value. 
Standardization of Antimeningococcus Serum.—An accurate method of standard- 
izing antimeningococcus serum has not as yet been devised. In the selection of a serum 
physicians must, therefore, be guided by the reputation of the manufacturers. 
The main point at issue concerns the nature of antimeningococcus serum, that is, the 
mechanism of its curative action and the chief antibodies concerned in the process. Ac- 
cording to Jochmann,11 Flexner,12 Flexner and Amoss,13 Evans,14 and others the chief activity 
of the serum is by increasing phagocytosis of the meningococci; Flexner also thought that 
the serum may be antiendotoxic, and Flexner and Jobling16 have found that potent sera may 
directly injure the organisms and impair their power of propagation. 
Five different methods have been advocated for testing the potency of these sera: (a) 
By determining the opsonin or tropin content as originally advocated by Jobling18 and regarded 
at present by the Hygienic Laboratory as the method of choice. Certainly in the treatment 
of meningococcus meningitis one is impressed by the evidences of increased phagocytosis 
of the cocci in cases favorably influenced by the serum, and this test appears to the writer as 
being a direct measure of at least one of the important curative activities. (b) By deter- 
mining the agglutinin content as described by Amoss and Wollstein,17 Wadsworth, Kirkbride, 
and Gilbert.18 (c) By determining the complement-fixation titer as advocated by Krumbein 
and Schatiloff19 and widely employed in Germany, (d) By determining the protective power 
of the serum in mice infected with virulent cultures as advocated by Hitchens and Robinson.20 
1. Bacteriotropic Titration.—While the antimeningitic serum was being prepared at the 
Rockefeller Institute Jobling used the opsonic test in standardization as the method of choice 
on account of the part taken by specific opsonins in promoting recovery from meningococcus 
infections. As a definite and suitable standard of strength Jobling has suggested that a 
serum be accepted as satisfactory when it shows unmistakable phagocytic activity in dilu- 
tions up to 1 : 5000. The method of Neufeld is used, as described on page 184. 
The method described by Evans as used in the Hygienic Laboratory (Bulletin No. 124) 
is as follows: 
(1) The serum is diluted with Locke’s solution 1 : 25, 1 : 50, and 1 :150; 0.2 c.c. of each 
1 Jour. Amer. Med. Assoc., 1913, 60, 1348. 
2 Hygienic Laboratory Bulletin, 1913, No. 91. 
3 Jour. Amer. Med. Assoc., 1914, 62, 1799; Jour. Exper. Med., 1915, 21, 43. 
4 Jour. Amer. Med. Assoc., 1913, 60, 1937. 
6 Collected Studies from Bureau of Laboratories of New York City, 1914-15, 37. 
6 Jour. Amer. Med. Assoc., 1917, 68, 1035. 
7 Hygienic Laboratory Bulletin, 1913, No. 88. 
8 Hygienic Laboratory Bulletin, 1917, No. 110, 35. 
9 Jour. Infect. Dis., 1921, 28, 367. 15 Jour. Exper. Med., 1908, 10, 141. 
10 Jour. Infect. Dis., 1922, 30, 379. 16 Jour. Exper. Med., 1909, 11, 614. 
11 Deutsch. med. Wchn., 1911, 37, 1733. 17 Jour. Exper. Med., 1916, 23, 403. 
12 Jour. Exper. Med., 1907, 9, 168. 18 Arch. Int. Med., 1919, 23, 269. 
13 Jour. Exper. Med., 1916, 23, 683. 19 Deutsch. med. Wchn., 1908, 24, 1002. 
14 Hygienic Lab. Bull., 1920, No. 124, 43. 20 Jour. Immunology, 1916, 1, 345. 950 
VACCINES, SERA, BLOOD, PROTEINS IN TREATMENT OF DISEASE 
dilution are placed in test-tubes measuring 10 x 75 mm. A fourth tube carries 0.2 c.c. of 
1 :25 normal horse-serum, and a fifth, 0.2 c.c. of Locke’s solution (control). 
(2) Young and rapidly growing cultures of meningococci are employed (thirteen- to 
twenty-one-hour cultures on serum glucose agar). The suspensions are made by washing 
off the growths in each tube with 2 c.c. of equal parts of ordinary broth and Locke’s solution 
and diluting until the turbidity equals the standard equivalent to 300 parts per million of 
silica adopted by the American Public Health Association; 0.2 c.c. of this suspension is 
added to the tubes carrying the serum dilutions which now gives final dilutions of 1 : 50, 
1 :100, and 1 : 300. These mixtures are placed in a water-bath at 37° C. for forty-five min- 
utes and in the meantime the leukocyte suspension is prepared. 
(3) On the preceding day a rabbit is injected intrapleurally on each side with 5 c.c. of a 
sterile aleuronaut suspension (starch 3 gm.; aleuronaut 5 gm., and ordinary broth 100 c.c.). 
On the following day the animal is chloroformed and the exudates in both cavities washed 
out into a 50 c.c. centrifuge tube with 1 per cent, solution of sodium citrate warmed to body 
temperature. Care is exercised against using bloody exudates. If aleuronaut is removed, 
the heavy particles soon settle and the supernatant suspension of leukocytes is transferred 
to a second 50 c.c. tube and centrifuged for four minutes at such speed as throws the leuko- 
cytes to the bottom without too much impaction. 
The supernatant fluid is discarded, the tube filled up with warm saline solution, the 
leukocytes gently distributed, and the tube centrifuged. The resulting mass of leukocytes 
are now resuspended in about 10 c.c. of warm Locke’s solution. 
Leukocytic suspension (0.2 c.c.) is added to each tube at the end of the forty-five minutes 
of incubation and the contents gently mixed and reincubated for forty-five minutes. Twice 
during this period each tube is rolled vigorously between the palms of the hands in order to 
keep the leukocytes in suspension. 
(4) Smears are now rapidly made, dried in the air, fixed with methyl alcohol, dried with- 
out washing, stained with a weak solution of carbol toluidin blue for twelve seconds (5 gm. 
toluidin blue dissolved in 100 c.c. alcohol, 500 c.c. water and 500 c.c. of 5 per cent, phenol, 
filtering after one or two hours. One part of the stain is diluted with 2 parts of water for 
staining the smears), and then with a solution of safranin (4 drops of a 0.5 per cent, watery 
or alcoholic solution of safranin in 50 c.c. of water) for ten minutes or longer. 
(5) The controls usually show that 4 to 20 per cent, of the polymorphonuclear leukocytes 
have engulfed cocci. The highest dilution of the immune serum showing 32 per cent, or at 
least twice as many phagocytes as the controls expresses the titer. No attempt is made to 
average the cocci contained in the leukocytes, but simply the highest dilution, producing at 
least twice as much phagocytosis as observed in the two controls. 
2. Complement-fixation Tests.—The advantages of these tests are that the same poly- 
valent antigen may be used as is employed for purposes of immunization; the technic is 
simple and the reactions are usually sharp and definite. According to Sophian, in a series 
of comparisons with opsonic and complement-fixation tests the results corresponded in 
every instance, a high opsonic content being accompanied by a high complement-fixation 
reading. The latter indicates, at least, that the horse has responded to immunization and 
that curative antibodies probably are present. Laboratories usually adopt their own stand- 
ards in preparing antimeningococcic serum. In the complement-fixation test Kolle requires 
complete inhibition of hemolysis with 0.1 c.c. of serum. 
Hitchens and Hansen1 have advocated a dried meningococcus antigen prepared as 
follows: 
(1) The cultures are grown on salt-free agar at 37° C. for sixteen to eighteen hours. 
(2) The growth is collected and suspended in distilled water; about 10 c.c. distilled water 
being used for each 20 square inches of agar surface. 
(3) To this suspension is added an equal volume of 95 per cent, ethyl alcohol. 
(4) This mixture is immediately centrifugalized. 
(5) After the supernatant fluid has been removed the precipitated bacteria are again 
suspended in 95 per cent, ethyl alcohol. 
(6) This process of centrifugalization and resuspension is repeated a second and a third 
time, but with ethyl ether instead of alcohol, the original volume of the suspension being 
reduced one-half each time. 
(7) The ether adhering to the bacteria after the precipitation and after removal of the 
supernatant ether is removed by vacuum. 
(8) The bacterial mass now constituting the antigen is further dried over phosphorus 
pentoxid in vacuo for three days. 
(9) The antigen is stored in tubes—about 0.05 gram to each—and kept in vacuo over 
phosphorus pentoxid. 
For use, a small amount of the powder, 0.02 gram, is carefully ground in a mortar, 20 
c.c. of normal salt solution being gradually added. 
The technic of these tests is given in Chapter XXV. 
1 Jour. Immunology, 1916, 1, 355. TREATMENT OF MENINGOCOCCUS MENINGITIS 
951 
3. Agglutination Tests.—These tests are readily conducted with the polyvalent antigen 
used in immunization, a macroscopic technic, as that described in Chapter XVI, being 
employed. 
Wadsworth, Kirkbride, and Gilbert have described the following method as employed 
by the Laboratory of the New York State Department of Health: 
(1) Standard cultures of meningococcus representative of the principal types may be 
obtained from the Laboratory of the State Department of Health. They are cultures which 
have been identified by serologic tests and which possess distinctive agglutinative and anti- 
genic properties. 
(2) Standard suspensions for performing the agglutinating tests are so prepared as to 
correspond in opacity with a standard suspension of barium sulphate containing 3 c.c. of a 
1 per cent, barium chlorid solution in 97 c.c. of a 1 per cent, sulphuric acid solution. Such a 
standard meningococcus suspension contains approximately 2,000,000,000 cocci per cubic 
centimeter. 
(3) In testing the potency of serum normal saline solution (0.85 per cent.) should be used 
as the diluent. After dilution with the serum the final mixture should contain 1,000,000,000 
cocci. The standard culture suspensions should be added in agglutination tubes (8 to 10 
mm. in diameter), to the diluted serums to be tested, and also similar dilutions of the standard 
serum. These test mixtures should be incubated at 55° C. for from sixteen to twenty-four 
hours. The titer thus determined is the maximum dilution of the serum in which definite 
clumping of the meningococcus in the suspension can be easily detected by the unaided eye. 
The agglutinative titer which is to be recorded is the potency of the serum. 
(4) The potency of antimeningococcus serum intended for therapeutic use, when tested 
and compared with the standard serum supplied on request by the Laboratory of the New 
York State Department of Health, shall have an agglutinative action in a dilution of 1 : 800 
with at least the four representatives of the principal types of meningococcus, under condi- 
tions in which a standard suspension is employed and the tubes maintained for sixteen to 
twenty-four hours at 55° C. 
4. Animal Inoculation Tests.—These tests have been found quite irregular and imprac- 
ticable for general use. As stated by Jobling, not only does the pathogenicity of the menin- 
gococcus vary considerably from day to day, but the resistance of animals to this micro- 
organism is also quite variable. By preparing a large quantity of bacterial emulsions and 
using sufficient controls to determine the fatal dose, and by employing a standard lethal 
dose of emulsions mixed with varying quantities of serum injected intraperitoneally into 250- 
gram guinea-pigs, some conception of the protective value of a serum may be obtained. 
Hitchens and Robinson1 have recently described an animal inoculation test which they 
hoped would prove of value for the standardization of antimeningococcus serum, but these 
hopes have not been realized. Amoss and Marsh,2 and Neill and Taft3 found the method 
too irregular in its results and insufficient for a guide to the therapeutic potency and stand- 
ardization of antimeningococcus serum. 
Action of Antimeningococcus Serum.—As previously stated, experi- 
ments in vitro show that a potent antimeningitic serum possesses three chief 
antibodies upon which its curative powers probably depend, namely: (1) 
Bacierioiropins (immune opsonins), which lower the resistance of the menin- 
gococci and facilitate their phagocytosis; (2) bactericidins, which kill the 
cocci extracellularly, either with or without final lysis; and (3) antitoxins, 
which neutralize the true extracellular toxin, which some strains of menin- 
gococci apparently produce in varying degree, and anti-endotoxins. Other 
than these are the agglutinins, which probably aid in bacteriolysis, and 
anti-aggressins, which may assist in the process of phagocytosis. 
Microscopic examination of a direct stained smear of the sediment of 
cerebrospinal fluid obtained from fresh cases will show large numbers of 
polynuclear leukocytes and cocci, the majority of the latter being extra- 
cellular. As the case improves, whether under serum treatment or spon- 
taneously, the micro-organisms diminish in number and become intracellular, 
frequently appearing clumped and failing to grow in culture. It would 
appear, therefore, that a cure is brought about partly by means of phago- 
cytosis aided by bacteriotropins; by bacteriolysis through the agency of 
1 Jour. Immunology, 1916, 1, 345. 
2 Jour. Exper. Med., 1918, 28, 779. 
3 Hygienic Laboratory Bulletin, 1920, No. 124, 93. 952 
VACCINES, SERA, BLOOD, PROTEINS IN TREATMENT OF DISEASE 
specific bacteriolytic amboceptors in the immune serum and complements 
in the spinal fluid and blood-serum, and to some extent by neutralization 
of a toxin with antitoxin. 
A potent antimeningococcus serum furnishes these main antibodies, 
and since the first two must act locally upon the cocci infecting the meninges, 
the serum must be applied locally and directly by intraspinous and subdural 
injection, since only traces of immune serum could eventually find their 
way into the cerebrospinal fluid if the serum were injected subcutaneously 
or intravenously. On the other hand, in the treatment of meningococcus 
bacteremia and toxemia the serum should be injected intravenously and 
subcutaneously. 
Polyvalent and Monovalent Sera.—Unfortunately, an immune serum 
may not contain the antibodies for the cocci producing a given infection, 
and hence the serum, even though it is skilfully administered in large doses, 
will have no influence upon the disease. Apparently the cocci of these 
resistant or “fast” strains are uninjured by the antibodies in the serum. 
To overcome this difficulty, a large number of different strains of menin- 
gococci are used in immunizing horses. In the Rockefeller Institute horses 
have been immunized with as many as 40 strains and during the recent 
war this serum yielded excellent results. If, however, the serum of one 
laboratory is found to exert no beneficial effect, the physician should use 
the serum of another, for different laboratories probably immunize their 
horses with cultures not in common use. It is highly desirable to secure 
cultures of these “fast” strains. These should be sent at once to laboratories 
engaged in the production of anlimeningilic serum, for the larger the number 
of these strains used in immunization, the more potent and valuable will be 
the serum. Very probably the use of monovalent sera will still further 
improve the results of the serum treatment of this disease. Under these 
conditions the type of infection would first require determination by the 
agglutination test applied to cocci removed from the spinal fluid, the technic 
being similar to the agglutination test for the typing of pneumococci. Gor- 
don and Hine1 have recently published the results of the treatment of 90 
cases by monovalent serum. The plan was to first inject a mixture of equal 
parts of Type I and Type II (Gordon’s classification) sera in order to lose 
no time while the typing was being done. Treatment was then continued 
with monovalent serum. Seven cases were fatal from other causes and 
may be excluded. The results with the remaining 83 were as follows: 
(a) 34, or 41 per cent., were due to Type I, with 1 death, or 3 per cent. 
(■b) 32, or 38 per cent., were due to Type II, with 7 deaths, or 21.9 per 
cent. 
(c) 10, or 12 per cent., were due to Type III, with no deaths. 
id) 7, or 9 per cent., were due to unknown types and 2 deaths, or 28.6 
per cent. 
The general mortality, therefore, was 12 per cent., which is an improve- 
ment on the general results of serum treatment of this disease. Type II 
serum was the least successful of these monovalent sera. This Type II 
group is more complex than the other groups and probably contains sub- 
groups. Favorable reports on the use of monovalent sera have also been 
made by Banks2 and Munro.3 
The Importance of Early Diagnosis and Treatment.—Statistics have 
shown that the best results in the serum treatment of meningitis are secured ■ 
1 Report of Medical Research Committee, January 28, 1919. 
2 T anppf 1 1 
» Brit. Med. Jour., March 27, 1920, 430. TREATMENT OF MENINGOCOCCUS MENINGITIS 
953 
in the early cases. Lumbar puncture and examination of the spinal fluid 
should always be done at once and without delay when symptoms are 
present suggestive of meningitis; this is especially true in times of epidemics. 
When meningitis develops acutely and spontaneously without previous 
infection of the ear, mastoid or accessory sinuses, the infection is usually 
meningococcic, and I believe it is good practice to give the first dose of serum 
at the time of the first spinal puncture if the fluid is opalescent or cloudy. 
Bacteriologic diagnosis can usually be made within an hour by examina- 
tion of smears of sediment of spinal fluid stained according to the method 
of Gram. Not infrequently, however, prolonged search may reveal only 
a few or none of the Gram-negative diplococci, although many pus-cells are 
present. These are usually meningococcus infections because in pneu- 
mococcus and streptococcus meningitis very large numbers of Gram-positive 
cocci are to be found. Cultures in the early cases are apt to prove sterile 
and if the smears are likewise negative or doubtful, twenty-four or more 
hours may be lost in treatment unless serum has been introduced with the 
first puncture. Even though the case does prove to be a pneumococcus 
or other infection, no particular harm has been done by the injection of the 
antimeningococcic serum. 
It is true that experiments on the lower animals have indicated that 
when meningococci are in the blood their localization in the meninges may 
be favored by spinal puncture. This would tend to make the physician 
hesitate to do lumbar puncture. The writer believes, however, on the 
basis of experiments, that this danger is overemphasized unless the meninges 
are irritated by an injection of normal serum. When symptoms of menin- 
gitis are present I am sure spinal puncture should be done at once and a 
good antiserum injected. 
Intraspinal Administration and Dosage of Antimeningitic Serum.—In 
Acute Meningitis.—As a rule, serum should he injected into the spinal canal as 
early in the disease as possible, and in such maximum amount as is compatible 
with safety. Intraspinal injection is absolutely necessary, for the serum 
must be brought into contact with the infected membranes, and only a trace 
would reach the spinal fluid if the serum were injected subcutaneously or 
intravenously. The advantages of early administration are obvious, and 
if the symptoms are indefinite, the physician should not hesitate to perform 
lumbar puncture and to secure fluid for microscopic examination, just as 
he would take a throat or nose culture to aid in the diagnosis of diphtheria. 
The maximum, or at least an adequate, amount of serum should be injected, 
care being observed to avoid undue pressure as a result of injecting too 
quickly or too large an amount. This administration of antimeningitic 
serum is, therefore, an important and delicate, though relatively simple, 
procedure. 
1. The technic of intraspinal injection has been described on p. 850. 
Whenever possible, the serum should be injected by the gravity method, 
and the amounts of fluid withdrawn and serum injected controlled by 
blood-pressure readings. 
2. Lumbar puncture is performed, and the fluid collected in graduated 
tubes. In the ordinary case fluid may be allowed to drain until the blood- 
pressure drops about 10 mm. of mercury, or if the pressure is unchanged 
or rises, until the fluid flows about 1 drop every three seconds, provided 
there are no other evidences of collapse, such as faintness, headache, and 
great restlessness. 
3. As a rule, the amount of serum injected should be slightly less than 
the amount of fluid withdrawn. When the injection is controlled by the 954 
VACCINES, SERA, BLOOD, PROTEINS IN TREATMENT OF DISEASE 
blood-pressure readings—the amount varies considerably—usually the 
injection should stop when the pressure falls another 10 or 15 mm. For 
adults, the dose of serum should be about 30 c.c.; for an infant, about 15 c.c. 
The serum should be allowed to flow in slowly, an ordinary injection 
consuming at least ten or fifteen minutes. If symptoms of collapse should 
appear before an adequate amount of serum has been injected, the funnel 
may be lowered and the spinal fluids allowed to flow out. When the symp- 
toms have disappeared, the injections may be continued and satisfactorily 
completed. 
3. When the physician cannot administer the serum by the gravity 
method or under blood-pressure control, the injection may be given by 
means of a syringe (see p. 855). It should be given slowly, and the patient 
observed closely in order to detect the general symptoms of collapse. The 
amount of serum injected should not be larger than the amount of cerebro- 
spinal fluid withdrawn. According to Sophian, the average doses are as 
follows: 
One to five years 
Dose of Antimeningitic 
Serum. 
Amount of Fluid 
Withdrawn. 
12 c.c. 
Five to ten years 
15 c.c. 
Ten to fifteen years 
20 c.c. 
Fifteen to twenty years 
30 c.c. 
Twenty years and over 
40 c.c. 
The injection of too large a dose of serum may be followed by headache, 
pain in the back and legs, and restlessness. When the amount of serum 
injected exceeds the amount of spinal fluid withdrawn the symptoms just 
named must be regarded as the signal to stop; otherwise they may be dis- 
regarded. 
Reactivation of Serum.—According to the experiments of Matsunami, 
Toyama, and the writer,1 the addition of fresh normal human or guinea-pig 
serum to commercial antimeningitis serum increases its opsonic and bac- 
tericidal activities; Hektoen and Tunnicliff2 have confirmed these observa- 
tions in so far as opsonic activity is concerned. Since the spinal fluid is 
generally free of opsonins and complements, the writer has advocated3 the 
addition of fresh normal serum to antimeningitis serum before intraspinal 
injection. Independently Fairley and Stewart4 have made the same recom- 
mendations. 
Unfortunately the trouble and labor involved retard the adoption of 
this form of treatment, but I believe it is of distinct value in the treatment 
of some cases that fail to improve and especially show in smears that phagocy- 
tosis of meningococci is not proceeding as rapidly as necessary. My pro- 
cedure now is to remove 10 to 20 c.c. of blood from healthy older children 
or adults into a sterile centrifuge tube, secure the serum, and keep it on ice. 
When necessary 2 to 5 c.c. of this serum are added to the immune serum 
just before intraspinal injection. 
Intravenous Administration of Antimeningoccus Serum.—During epi- 
demics of meningitis it may be possible to detect cases in the bacteremic 
stage when meningococci are present in the blood and clear fluid is collecting 
in the ventricles. In these and in all severe fulminant infections it is good 
practice to inject from 30 to 100 c.c. of serum intramuscularly or intra- 
venously. It is advisable to secure a culture of blood in ascites dextrose 
1 Jour. Immunology, 1918, 3, 157, 177. 
2 Jour. Infect. Dis., 1921, 29, 553. 
3 Jour. Amer. Med. Assoc., 1918, 71, 1856. 
4 Commonwealth of Australia, Service Publication No. 9. TREATMENT OF MENINGOCOCCUS MENINGITIS 
955 
broth in all cases in adults and older children. If sufficient serum may be 
obtained and the expense is a secondary consideration, an intravenous or 
intramuscular injection, given at the outset and once or twice during the 
acute stage, may benefit the patient by neutralizing toxins and possibly 
prevent complications due to the entrance of meningococci into the blood- 
stream. Furthermore, as shown by Flexner and Amos,1 in poliomyelitis 
the intraspinal injection of serum increases the permeability of the choroid 
plexus and meninges, permitting the passage of immunity principles from 
the blood into the cerebrospinal fluid. For this reason the intravenous or 
intramuscular injection of antimeningitis serum in conjunction with intra- 
spinal injections may be particularly advantageous in the treatment of 
severe infections. 
With intravenous injections of serum care must be exercised against 
severe reactions. It is good practice to first inject subcutaneously 1 c.c. 
•of the serum and give the balance about an hour later, injecting slowly. 
Blackfan2 has directed particular attention to these reactions and advises 
against intravenous injections of serum for children. 
Herrick3 has reported particularly good results from treatment by 
intravenous injections of serum in severe adult cases, the patients often 
•coming out of coma with rapid recession of the rash and symptoms. The 
mortality was reduced from 64 to 19 per cent. Out of Herrick’s 265 cases, 
64, or 25 per cent., proved fatal, but out of 137 cases given ordinary intra- 
spinal injections and less than 45 c.c. of serum intravenously 47, or 34 per 
cent., proved fatal, whereas out of 128 cases treated intraspinally with small 
doses and intravenously with large doses of serum 19, or 15 per cent., proved 
fatal. 
Of the two forms of serum therapy—intraspinal and intravenous—the 
former is the more important. Acute primary meningococcus bactermia 
without involvement of the meninges is rare, but when meningococci are 
found in the blood, serum should be given intravenously. In severe adult 
•cases seen within the first three days serum may be given intravenously as 
a routine; at least one dose of 100 c.c. may be given. 
Repeating Doses of Serum.—It is the general rule to give an intra- 
spinal injection of serum every day for four days, and then on alternate 
days until the acute symptoms have subsided, and to resume the treatment 
if an exacerbation or a relapse occurs. In severe fulminant infections, and 
especially if the exudate is thick and only small amounts of serum can be 
introduced under pressure, it is well to repeat the injection every twelve 
hours until several doses have been given. Some cases require daily con- 
secutive injections for six or more days; the average case will require from 
four to six injections if the treatment is begun during the acute stage; in 
the subacute and chronic cases many more treatments are required. There 
are two main indications and guides: 
1. The condition of the cerebrospinal fluid. 
2. The clinical condition of the patient. 
1. In most instances the cerebrospinal fluid tends to clear up macro- 
scopically as the disease improves. This is, however, occasionally mis- 
leading, as the fluid may become more turbid as the result of an aseptic 
meningitis or excitation of a polynuclear leukocytosis due to the serum, 
while, in reality, the numbers of meningococci are diminishing and the patient 
is improving. 
1 Jour. Exper. Med., 1917, xxv, 499, 525. 
2 Jour. Amer. Med. Assoc., 1921, 76, 36. 
3 Jour. Amer. Med. Assoc., 1918, 71, 612. 956 
VACCINES, SERA, BLOOD, PROTEINS IN TREATMENT OF DISEASE 
More accurate information is obtained by the microscopic examination 
of a stained smear of the sediment of the cerebrospinal fluid withdrawn. 
In fresh acute cases the cocci are numerous and mostly extracellular; improve- 
ment is indicated by a diminution in their number, and by the fact that 
they are mostly intracellular. I generally determine the phagocytic index 
or the relative proportion of leukocytes that have engulfed cocci and the 
opsonic index or the relative number of cocci per leukocyte as determined 
by counting a large number. When many cocci are present, or if they are 
few in number but mostly extracellular, the indications are to puncture 
next day, even if the clinical condition of the patient is good and the tem- 
perature is lower. The number and position of cocci are, therefore, of more 
importance as. a guide to subsequent injections than is the total number of 
pus-cells. 
2. As an indication for repeating the doses of serum the clinical con- 
dition is of most value when combined with the examination of the cerebro- 
spinal fluid. Occasionally the patient’s condition may seem to improve, 
although the fluid may show numerous cocci, which will subsequently 
aggravate the clinical condition unless the serum is administered. In 
favorable cases there is usually a lower temperature the day following an 
injection, and frequently delirium becomes less marked and there is some 
return to consciousness. The complexion, which is often cyanosed at first, 
regains a healthy color; the pain in the head, neck, and limbs becomes less 
severe, although the neck and spine may remain stiff for several days. 
Finally the mind becomes clear and the patient is cheerful, and no longer 
irritable, apathetic, and hypersensitive. He feels better and his appetite 
returns. . When this favorable outcome supervenes, the serum injections 
may be discontinued, to be resumed, however, upon the first evidence of a 
relapse. The physician should be on his guard for the appearance of acute 
hydrocephalus, which condition is relieved by repeated lumbar puncture. 
The Serum Treatment of Cases with Thick Plastic Exudate.—In very 
severe cases the exudate may be so thick that it will not flow from the needle. 
In these cases the serum should be injected in small doses under pressure, 
and the injections repeated every eight to twelve hours. As they are likely 
in any case to terminate fatally, the physician is justified in taking the risk 
of increasing intracranial pressure. It may be well carefully to inject a 
small amount of warm sterile salt solution, which will dilute the exudate 
and possibly start a flow; or a second needle may be inserted higher up, 
when a thinner exudate is found, or washing may be possible by injecting 
salt solution in the upper needle and draining through the lower. 
The Serum Treatment of Cases with a Dry Canal and Cases of Posterior 
Basal Meningitis.—Occasionally a patient improves clinically and the 
amount of cerebrospinal fluid becomes very scanty, the spinal tap being 
dry, although it is certain that the needle has entered the subarchnoid space. 
In such instances a small amount of serum may be injected, or the injection 
may be dispensed with if the clinical condition continues to improve. If, 
however, cases with dry canals present evidences of toxemia and general 
aggravation of symptoms, small doses of serum should be injected under 
pressure and the injections repeated as often as necessary. The physician 
must be very cautious, however, for if there are clinical evidences of severe 
intracranial pressure, it is probable that there is an encapsulation of fluid 
within the ventricles, and shutting off of the communication between the 
ventricles and the subarchnoid space. In this posterior basic meningitis 
intraspinal injections are dangerous and aggravate the process. In infants 
it is necessary to puncture the ventricles through the anterior fontanel as TREATMENT OF MENINGOCOCCUS MENINGITIS 
957 
first practised by Cushing and Sladen,1 and in older children and adults by 
trephining at Kocher’s point, removing the fluid, and if it is found to be 
cloudy or purulent, injecting serum. It may be necessary to tap both 
ventricles alternately at intervals of several days, depending upon the 
reaccumulation of fluid and pressure symptoms. Permanent drainage may 
be instituted in severe cases by means of small catheters. While the opera- 
tion is not usually dangerous, the ultimate prognosis is very unfavorable. 
In adults trephining and injecting serum into the ventricles has yielded, 
according to Marcland,2 4 cures among 18 cases. Landry and Hamley3 
had 1 cure out of 5 cases, but Fairley and Stewart lost all of 9 cases. In 
place of trephining and injecting the serum Herrick recommends Cobb’s 
method of breaking down adhesions around the foramen magnum and fourth 
ventricle by manipulations of the head under general anesthesia to relax 
the neck muscles. This method may produce an increased flow of spinal 
fluid. 
Cistern puncture and injection of serum may also be tried. Ayer4 has 
conducted the puncture in 20 cases without mishap and his method is de- 
scribed in Chapter XXXVII. Recently Mitchell and Reilly5 have recorded 
the recovery of an infant of four months following repeated cistern punctures 
and drainage and the injection of 5 to 8 c.c. of serum each time. 
The Serum Treatment of Subacute and Chronic Meningitis.—If there 
is no evidence of sepsis; if the mind is clear and the neck limber; if the 
general conditions are good and the cerebrospinal fluid is practically cleared 
up, the affection is most likely hydrocephalitic, and may be relieved by 
repeated spinal punctures, with removal of as much fluid as is safe, using 
blood-pressure as an index. If meningococci are present in cultures of the 
fluid, small amounts of serum may be injected. The prognosis in these 
cases, however, is generally bad, as the process is prolonged and the patient 
finally succumbs. 
In the second form of chronic meningitis, when the meningeal symptoms 
are active, intensified, and persistent, serum should be administered every 
few days in the same manner and in the same dosage as in the acute cases. 
Improvement is, however, usually temporary, and the ultimate prognosis 
is very grave. 
Serum Sickness.—Intraspinal injections of serum result in the sen- 
sitization of the patient in just the same manner as if serum were injected 
by other routes. The percentage of cases developing serum sickness is 
likely to be high, since antimeningitic serum is not refined (Sophian reports 
60 per cent, of his cases as developing the condition), and while the symptoms 
are distressing, they are seldom alarming, and fatal anaphylaxis is extremely 
rare. Occasionally the onset of serum sickness may be mistaken for a re- 
currence of meningitis, but if the meningeal condition has been responding 
as well as could be expected, it is wise to let the patient alone, rather than 
to make additional punctures and cause further depression. Local sedatives, 
laxatives, atropin, sedatives, and, at times, morphin, are indicated (see 
Chapter XXXII). 
Summary and General Plan for the Serum Treatment of Meningococcus 
Meningitis.—For cases of average severity first diagnosed on or about the 
third day of the disease the following plan of serum treatment is advised: 
1 Jour. Exper. Med., 1908, 10, 548. 
2 Bull, et mem. Soc. d. hop., 1918, xlii, 1218. 
3 Amer. Jour. Med. Sci., 1919, 157, 210. 
4 Arch. Neutrol. and Psych.. 1920, 4, 529. 
5 Amer. Jour. Med. Sci., 1922, 164, 66. 958 
VACCINES, SERA, BLOOD, PROTEINS IN TREATMENT OF DISEASE 
spinal fluid should always be drained off, in amounts slightly more than the 
amount of serum injected. If the spinal fluid pressure is high, it is well to 
drain spinal fluid until it flows at the rate of about 1 drop in five seconds. 
Each spinal fluid should be examined for meningococci. 
First day: For children under twelve years: 15 c.c. of serum intra- 
spinally at the time of first spinal puncture or as soon as possible there- 
after. Repeat the drainage and injection of serum eight to twelve hours 
later. 
For children over twelve years and adults: 20 to 30 c.c. of serum intra- 
spinally as soon as possible. Blood culture. Severe cases in adults to 
receive 100 to 150 c.c. serum intravenously one hour after the subcutaneous 
injection of 1 c.c. of the serum for desensitization. Intraspinal injection 
to be repeated eight to twelve hours later. 
Second day: Repeat the intraspinal injection. If very severe give a 
second intraspinal injection eight to twelve hours later. If blood culture 
is positive, give a second intravenous injection of serum. 
Third and fourth days: One intraspinal injection of serum. If patient 
is not improving or growing worse, give a second injection on these days 
about eight to twelve hours later. 
Fifth day: If spinal fluid removed in fourth day shows no meningococci, 
spinal puncture may be done and fluid removed if it is under pressure, but 
serum may be omitted. As a general rule, however, it is well to inject serum 
intraspinally. 
The appearance of the fluid if cloudy may be misleading because leukocytes 
may be present by reason of irritation of the meninges by the serum and menin- 
gococci absent. 
Serum should be injected intraspinally at least once a day until two 
successive spinal fluids have proved sterile. At that time serum may be 
stopped, but a sharp watch kept up for relapses. Serum sickness will 
probably develop and should not be mistaken for a relapse. 
Results of the Serum Treatment on Meningococcus Meningitis.— 
(a) Upon the Course of the Disease.—In the majority of cases the subdural 
injection of a potent antimeningitic serum is followed by some immediate 
improvement in the local suppurative meningitis and general sepsis, for the 
temperature usually drops, the mental condition improves, and delirium is 
diminished, although the Kernig sign may persist, partly as the result of 
meningeal irritation and partly on account of fear. Hydrocephalus is gen- 
erally relieved, as indicated by lessening of the pressure symptoms, as, for 
example, severe headache, vertigo, and vomiting; breathing becomes more 
regular, and the pulse also becomes slower and more regular. The duration 
of the illness is usually shortened. According to Holt, in the New York 
epidemic of 1904-05, antedating the use of serum, among 350 cases that 
recovered the duration in 3 per cent, was one week or less, and in 50 per 
cent, five weeks or longer. Of 288 cases reported by Flexner and Jobling, 
the average duration of active symptoms in those receiving serum was 
eleven days. Sophian, in an experience of several hundred cases, reports 
that many acute cases were relieved in five or six days and discharged as 
cured in two weeks. 
(b) Upon Complications.—Next to its influence upon mortality, the 
good effects of antimeningitic serum are apparent in that it lessens the inci- 
dence and severity of the terrible complications of this disease. The most 
severe and permanent sequels are those resulting from affections of the 
internal ear and the essential structures of vision. A conservative estimate 
of the incidence of the former among cases not receiving serum treatment TREATMENT OF MENINGOCOCCUS MENINGITIS 
is 12 per cent. (Goffert), whereas Flexner’s1 analysis of 1300 cases treated 
with serum shows that deafness occurred in but 3.5 per cent, of cases. At 
least from 12 to 24 per cent, of cases not receiving serum treatment will 
develop more or less serious eye complications, whereas Flexner’s report 
shows that impairment of vision among serum-treated cases occurred in 
about 0.9 per cent, of cases. The latter report shows the occurrence of 
arthritis in but 0.9 per cent, of cases, whereas among cases untreated with 
serum this is a frequent complication. Whereas chronic meningitis is 
relatively common among cases treated without serum, it is uncommon 
among those treated with serum. 
(c) Upon Mortality.—The gross mortality among cases treated without 
serum varies from 70 to 90 per cent.; among serum-treated cases the mor- 
tality is about 30 per cent. For example, of 1294 cases treated with serum 
prepared in the Rockefeller Institute, the general mortality was 30.9 per 
cent. 
The importance of early diagnosis and prompt institution of serum treat- 
ment is shown in the following table: 
959 
MORTALITY ACCORDING TO THE PERIOD OF INJECTION OF SERUM 
Time of Injection. 
Mortality, Per Cent. 
Flexner, 
1294 Cases. 
Dopter, 
402 Cases. 
Netter and 
Debre, 
99 Cases. 
Sophian, 
161 Cases. 
First to third day 
18.1 
8.2 
20.9 
9.0 
Fourth to seventh day 
27.2 
14.4 
33.3 
14.9 
Later than seventh day 
36.5 
24.1 
26.0 
22.6 
Average mortality 
30.8 
16.44 
28.0 
15.5 
The influence of age upon morlaliiy was early pointed out by Flexner. 
The very high mortality in infants and in old persons is due to their lowered 
vitality and enfeebled resistance. An additional factor in young children 
is their greater tendency to develop extreme hydrocephalus and convulsions. 
MORTALITY ACCORDING TO AGE 
Age. 
Reported by 
Flexner. 
Dopter. 
Netter. 
Sophian. 
Under one year 
49.6 
48.6 
50.0 
50.0 
One to two years 
31.0 
20.1 
0.0 
21.2 
Two to five years 
28.4 
9.3 
16.6 
17.5 
Five to ten years 
15.1 
8.5 
12.5 
9.0 
Ten to twenty years 
29.4 
10.2 
0.0 
18.0 
Over twenty years or age not given. 
38.2 
14.1 
0.0 
32.0 
The statistics show indubitably that the mortality of epidemic meningitis 
can be greatly reduced by the administration of serum. While the ordi- 
nary type of epidemic meningitis responds best to the specific treatment, 
the fulminant cases may also receive some of the beneficial influence 
of the serum. To quote from what Flexner wrote in 1909, and repeated in 
1 Jour. Exper, Med., 1913, xvii, 553. 960 
VACCINES, SERA, BLOOD, PROTEINS IN TREATMENT OF DISEASE 
1913: “In view of the various considerations presented, the conclusions 
may be drawn that the antimeningitis serum, when used by the subdural 
method of injection, in suitable doses and at proper intervals, is capable of 
reducing the period of illness; of preventing in large measure the chronic 
lesions and types of the infection, of bringing about complete restoration of 
health, thus lessening the serious, deforming, and permanent consequences 
of meningitis; and of greatly diminishing the fatalities due to the disease.” 
During the great war meningococcus meningitis was of wide-spread 
occurrence. During the first year or two a sudden demand for serum was 
met by supplies of poor quality sera, but this situation was soon corrected 
with an improvement in results and a lowering of the mortality to the 
neighborhood of 20 per cent. 
Treatment of Meningitis with Human Serum.—MacKenzie and Martin1 
in 1908 treated 16 acute cases with intraspinal injections of 15 to 20 c.c. of 
fresh sterile human serum, with a mortality of 38 per cent. 
Cases of meningitis have also been treated with intraspinal injections 
of serum from convalescent cases of the disease as well as by injections of 
the patient’s own serum. 
Convalescent serum may doubtless prove worth administering in excep- 
tional cases, if derived from an individual recovered from the same type of 
infection, but the curative value of normal human serum is questionable, 
although its addition to antimeningococcus serum may prove distinctly 
worth while in the treatment of some cases, as previously discussed, by 
increasing the opsonic and bactericidal activity of the latter. 
Vaccine Treatment of Meningococcus Meningitis.—Vaccines have been 
advocated in the treatment of subacute and chronic cases when serum appears 
to be losing its effect. Rolleston2 reports the treatment of 21 cases with 
vaccines, always in addition to other treatment, with a mortality of 23.5 
per cent. 
Crowe3 considered autogenous vaccines of aid in treatment and began 
early in the acute stage with small doses of 1,000,000 cocci. Hall,4 Collins,5 
Sophian,6 Colebrook,7 Fairley and Stewart,8 Worster-Drought and Kennedy,9 
Chalmers and O’Farrel,10 MacLagan,11 and others have also reported favorably 
upon the use of vaccines in subacute and chronic cases of the disease. 
The vaccines should be autogenous; if a stock vaccine is used it should 
be polyvalent in order to make sure that it includes the type producing the 
disease. In subacute and chronic cases the vaccine may contain 1,000,- 
000,000 cocci per cubic centimeter. Subcutaneous injections may be given 
every five days beginning with 0.1 c.c. for persons fifteen years or older and 
gradually increasing until 1 c.c. is being given at one time. 
Since lumbar puncture as an aid to the diagnosis of meningitis is coming 
into more general use, the important fact has been revealed that the influenza 
THE SERUM TREATMENT OF INFLUENZAL MENINGITIS 
1 Jour. Path, and Bacteriol., 1908, 12, 539. 
2 Lancet, 1919, 1, 645. 
3 Lancet, 1915, 2, 1127. 
4 Medical Research Committee Report, 1916. 
5 Brit. Med. Jour., 1915, 1, 287. 
6 Epidemic Cerebrospinal Meningitis, C. V. Mosby Co., 1913, 192. 
7 Lancet, 1915, 1, 1026. 
8 Commonwealth of Australia Service Publication, No. 9, 1916. 
9 Cerebrospinal Fever, A. and C. Black, London, 1919, 428. 
10 Jour. Trop. Med. and Hyg., 1915-16, xl, 101. 
11 Edinb. Med. Jour., 1918, 20, 375. THE SERUM TREATMENT OF INFLUENZAL MENINGITIS 
961 
bacillus is not an infrequent cause of severe, and usually fatal, seropurulent 
cerebrospinal meningitis. In 1911 Wollstein1 collected 58 cases of this 
infection, all but 6 ending fatally, and as the bacterial diagnosis of meningitis 
is becoming more widely known and more commonly employed, the number 
of reported cases is increasing rapidly. The mortality of 90 per cent., which 
is exceeded only by the tuberculous and pneumococcus infections of the 
meninges, and the encouraging results following the use of a specific anti- 
influenzal serum in experimental infections in monkeys, render this subject 
one of great importance from the standpoint of serum therapy. 
Influenzal Meningitis.—Like the acute meningeal infections in general, 
influenzal meningitis is more prevalent among children than among adults. 
Infection of the meninges is probably always secondary to infection of 
the respiratory tract with virulent influenza bacilli, the route of infection 
being chiefly through the blood-current. Direct infection from the nose 
cannot be excluded, and should be considered a possibility. However, all 
or nearly all cases of spontaneous influenzal meningitis in human beings are 
the result of influenzal bacteremia, since the bacilli have been cultivated in 
large numbers from the heart’s blood before and after death. The same is 
true of experimental influenzal meningitis in the monkey. 
According to Flexner,2 the cerebrospinal fluid removed by lumbar punc- 
ture from human patients is always turbid, and deposits a yellowish or whitish 
sediment on standing. “As the disease advances, the fluid becomes more 
heavily charged with pus-cells, until toward the end, and as late as the 
seventh day of illness, the puncture may yield merely a viscid mass of 
purulent matter. The number of influenza bacilli present in the fluid is 
usually large, and the bacilli lie chiefly extracellular among the pus-cells, 
although a variable but small number is usually found ingested by the 
leukocytes. In morphology the bacilli vary somewhat, and in this respect 
the observer may readily be deceived as to the nature of the bacteria present. 
While some of the fluids contain the typical, minute rods, others show quite 
irregular and knobbed or even filamentous bacteria that have little resem- 
blance to the influenza bacillus as seen in recent cultivations. These bizarre 
or involution forms, however, are met in old and exhausted cultures; and 
when they are recultivated on a suitable hemoglobin medium, they yield 
the typical minute rods.” 
The cerebrospinal fluid removed from monkeys inoculated by subdural 
injection with virulent cultures of the influenzal bacillus resembles in all 
essential particulars the fluid removed from patients with spontaneous 
infections. 
The bacieriologic diagnosis can usually be made by microscopic examina- 
tion of stained smears of the fluid, but whenever possible, the diagnosis 
should be confirmed by cultural methods. 
Anti-influenza Serum.—After having satisfactorily demonstrated experi- 
mental influenza meningitis in the monkey, Flexner and Wollstein prepared 
an immune serum and showed that the experimental infection could be 
controlled and cured by injecting the serum directly into the seat of disease 
by intraspinal inoculation. The immune serum was prepared by the 
ordinary methods, first a goat and then a horse being injected with non- 
virulent and finally with virulent bacilli, covering a period of many months, 
until their serums showed the presence of agglutinins and bacteriotropins. 
The serum lacked bacteriolytic properties, and did not give rise to comple- 
ment fixation in dilutions greater than 1 : 100. 
1 Jour. Exper. Med., 1911, xiv, 73; Amer. Jour. Dis. Child., 1911, i, 42. 
2 Jour. Amer. Med. Assoc., 1913, lxi, 1872. 962 
VACCINES, SERA, BLOOD, PROTEINS IN TREATMENT OF DISEASE 
Following the administration of serum, the cerebrospinal fluid tends to 
clear up; the bacilli become fewer in number and are mostly ingested by the 
phagocytes; the pus-cells become less numerous, and while the bacilli may 
persist in the fluid for a longer period, they ultimately disappear. 
Administration of Anti-influenza Serum.—The Rockefeller Institute 
has distributed serum throughout different parts of the country, and is 
prepared to furnish it to physicians upon request. Physicians, and especially 
pediatrists, should resort to lumbar puncture early in all suspected cases of 
meningitis, for only in this manner may influenzal meningitis be detected 
early to derive any possible benefit from serum treatment. The serum shoidd 
be injected directly into the spinal canal by the gravity or syringe method in 
exactly the same manner and with the same precautions as are observed in 
administering serum in the treatment of epidemic meningitis. Since the 
disease is usually accompanied by a bacteremia, it is well to inject serum 
intravenously, although serum injected intraspinally soon finds its way into 
the blood-stream. 
TREATMENT OF PNEUMOCOCCUS MENINGITIS 
Meningitis is caused more frequently by the pneumococcus than by the 
influenza bacillus. Its mortality is certainly no less than in influenzal 
meningitis. During the past four years I have seen 24 cases and all were 
fatal. 
The few instances in which antipneumococci serum has been employed 
have not yielded results that inspire confidence in its employment alone. 
Even when the homologous serum is used in treating experimental pneumo- 
coccus meningitis in monkeys, the fatal termination may be delayed, but is 
not prevented. For this reason the outlook for its successful employment 
alone in human infections is not encouraging. Investigations by Lamar1 
have shown, however, that mixtures of homologous antipneumococcus 
serum, sodium oleate, and boric acid exert a marked and decided curative 
influence upon a virulent experimental meningitis, and while this method 
has not thus far been generally applied in the treatment of the disease in 
humans, it is deserving of trial and offers some encouragement for an other- 
wise highly fatal infection. 
More recently Litchfield2 and Gray3 have reported the successful treat- 
ment of the disease with Ryes’ fowl antipneumococcus serum, to which 
further reference will be made. 
The Nature of Pneumococcus Meningitis.—This infection is usually 
secondary, and follows on pneumonia or on inflammations of serous mem- 
branes by indirect transmission by the blood or by direct infection from the 
nasopharynx, mastoid cells, frontal, sphenoid and ethmoid sinuses, and 
internal ear. 
The diagnosis is usually made as the result of microscopic examination 
of stained smears of cerebrospinal fluid removed by lumbar puncture. 
Large numbers of polynuclear leukocytes with intracellular and extra- 
cellular Gram-positive diplococci, occurring in pairs or in short chains, 
usually indicate a pneumococcus infection. Whenever possible, the diagno- 
sis should be confirmed by making cultures of the fluid on dextrose blood- 
agar, and by injecting portions intraperitoneally and subcutaneously in 
mice. 
Pneumococcus infections of the cerebral meninges have been found 
1 Jour. Exper. Med., 1911, i; ibid., 380; xiv, 256; 1912, xvi, 581. 
2 Jour. Amer. Med. Assoc., 1919, 72, 1345. 
3 Amer. Jour. Med. Sci., 1920, 159, 885. TREATMENT OF PNEUMOCOCCUS MENINGITIS 
963 
experimentally to be more refractory to treatment than infections of the 
spinal meninges, hence human infections following injuries to the head, or 
occurring as the result of direct extension from neighboring sinuses, are 
likely to be more refractory than indirect infections by way of the blood. 
Importance of Early Diagnosis and Drainage.—Early diagnosis by 
spinal puncture and examination of spinal fluid is of great importance; ordi- 
narily much valuable time is lost. The exudate in pneumococcus meningitis 
rapidly thickens and greatly hinders drainage by spinal puncture. The 
spinal fluid contains untold millions of pneumococci which proliferate rapidly 
and soon overwhelm the patient. The writer regards pneumococus menin- 
gitis among the most fatal of infections, and in view of its frequency and 
particularly after infections and operations upon the mastoid cells and 
accessory nasal sinuses the disease is one urgently demanding investigation 
for the purpose of evolving an efficient treatment. 
Some means should be found for affording drainage. Something in this 
direction can be accomplished by two or more spinal punctures daily, re- 
moving as much spinal fluid as possible. Likewise washing out the spinal 
canal between a needle in the upper dorsal and one in the lower lumbar 
regions, may aid. But the blunt of the attack is in the cerebral subarachnoid 
space and sera introduced in the lumbar region are soon blocked from 
diffusion to these parts by the plastic exudate. Cistern puncture (described 
in Chapter XXXII) and drainage is to be thought of in this connection; 
also washing out of the meningeal cavities by puncture of the lateral ven- 
tricles, as described by Beilin and his associates.1 Some means should be 
devised for more or less continuous drainage of the cerebral and spinal 
meninges with facilities for flushing out the subarachnoid space at intervals 
with saline solution, serum, or some chemical agent. The successful institu- 
tion of continuous drainage will open up hope for success in the treatment 
of this dreadful disease and the writer and his associates are now investigating 
the problem. 
Treatment with Antipneumococcus Serum, Sodium Oleate, and Boric 
Acid.—Numerous investigations by Conradi,2 Korschun and Morgenroth,3 
Levaditi,4 and Noguchi5 have shown that substances may be obtained 
directly from tissue-cells and leukocytes or after autolysis which are bac- 
tericidal and hemolytic, and, as shown by Noguchi, are largely in the nature 
of higher saturated fatty acids or their alkaline soaps. As shown by Klotz,6 
soaps occur in inflammatory foci, and the origin of the fatty acids and soaps 
is readily accounted for since Achaline7 has shown the presence of lipase in 
such foci. With the death of leukocytes in an inflammatory focus, brought 
about by a bacterial poison, leukocidins, or lack of nutriment, disintegration 
occurs, and by autolysis and lipolysis fatty acids and soaps are produced 
that in themselves seem to exert a destructive action upon the infecting 
bacteria. 
With these considerations in mind, Lamar, investigated the influence 
of soaps upon pneumococci. Solutions of 0.5 to 1 per cent, of sodium 
oleate were found rapidly to kill pneumococci; much weaker solutions, as, 
e. g., 0.1 per cent., or even 1 part of soap in 10,000 parts of water, were found 
to lessen their virulence, and, what is more important and significant, to 
1 Lyon Chirurgical, 1918, 15, 455. 
2 Beitr. z. chem. Phys. u. Path., 1902, i, 193. 
3 Berl. klin. Wchn., 1902, xxxix, 870. 
4 Ann. de l’Inst. Pasteur, 1903, xvii, 187. 
5 Biochem. Ztschr., 1907, vi, 327. 
6 Jour. Exper. Med., 1905, vii, 633. 
7 Compt. rend. Soc. de biol., 1899, li, 568. 964 
VACCINES, SERA, BLOOD, PROTEINS IN TREATMENT OF DISEASE 
render the organisms peculiarly susceptible to lysis by a homologous anti- 
pnuemococcus serum. 
A serious drawback to the application of these discoveries was that 
protein constituents of serum and exudates were found to inhibit the bac- 
teriolytic and hemolytic action of unsaturated fatty acid soaps, as was shown 
by Noguchi1 and then by von Liebermann.2 The latter and von Fenyvessy3 
later found that this inhibition can be prevented in the test-tube and also 
in the animal body by adding a minute quantity of boric acid, which prevents 
the union of soap and protein matter when the latter is not too greatly in 
excess. 
The experiments of Lamar with mixtures of homologous antipneu- 
mococcus serum, sodium oleate, and boric acid in the treatment of pneu- 
mococcus meningitis in the monkey have yielded excellent results, especially 
when used early in the infection. These experiments have proved that 
sodium oleate lowers the virulence of pneumococci and renders them pecu- 
liarly and highly susceptible to solution by bacteriolysins present in the 
serum, and that boric acid largely prevents the inhibitory action of protein 
constituents upon this sensitizing action of sodium oleate. 
One drawback to the use of this method in the treatment of human 
infections is the necessity of using an antipneumococcus serum corresponding 
to the organism causing the infection. But a polyvalent serum may be 
employed or the type of infection quickly determined by agglutination reac- 
tions. 
It is highly desirable that the treatment be administered as early as possible, 
when the exudate is largely serous or at most seropuruleni. The mixture should 
be injected into the spinal canal after the withdrawal of the fluid by the 
gravity or syringe method and under blood-pressure control, as previously 
described. The amount injected and the number of injections depend upon 
the clinical condition of the patient, and in general may be administered in 
the same way as is antimeningitic serum. 
The initial dose may be 20 c.c., and is prepared as follows: 
Antipneumococcus serum (sterile) 4 c.c. 
5 per cent, aqueous solution of boric acid (sterile) 15 c.c. 
2 per cent, aqueous solution of sodium oleate (Kahlbaum’s or Merck’s) 
(sterile) 1 c.c. 
Ordinarily two spinal punctures should be made daily and as much spinal 
fluid removed as possible; the mixture or antipneumococcus serum alone 
should be injected after each drainage. 
Intravenous injections of serum are also advisable not only in the treat- 
ment of meningitis developing during the course of pneumonia, but also 
in that more fatal form developing by direct extension from the mastoid 
cells and accessory sinuses. For this purpose 50 to 100 c.c. may be adminis- 
tered as described in the serum treatment of pneumonia. 
Treatment with Kyes Chicken Antipneumococcus Serum.—During 
the epidemic of influenza in 1918 Litchfield4 treated 10 cases of pneumo- 
coccus meningitis developing in the course of lobar pneumonia. Kyes’ 
antipneumococcus serum was injected intraspinally and intravenously, with 
5 recoveries. Gray,5 subsequently reported an additional case treated with 
this serum, with a fatal outcome. 
1 Biochem. Ztschr., 1907, vi, 327. 
2 Biochem. Ztschr., 1907, iv, 25. 
3 Ztschr. f. Immunitatsf., orig., 1909, ii, 436. 
4 Jour. Amer. Med. Assoc., 1919, 72, 1345. 
5 Amer. Jour. Med. Sci., 1920, 159, 885. TREATMENT OF TUBERCULOUS MENINGITIS 
965 
Treatment with Mixtures of Antipneumococcus Serum and Ethyl- 
hydrocuprein Hydrochlorid.—Morgenroth and Levy1 have shown that 
optochin (ethylhydrocuprein) hydrochlorid, a synthetic compound of 
quinin, is highly pneumococcidal. These observations have been amply 
confirmed by Moore,2 Cohen, Heist, and the writer,3 and others. Idzumi 
and the writer4 have found that the intraspinal injection of this drug has a 
favorable influence upon experimentally produced pneumococcus meningitis 
in dogs and rabbits when given early (within first thirty-six hours). 
The drug and further observations on its use in the treatment of pneu- 
mococcus infections will be discussed in my monograph on Chemotherapy, 
but it is mentioned here for the possible benefit it may have in the treat- 
ment of pneumococcus meningitis of human beings. 
A 1 : 200 solution is prepared in sterile saline solution and heated at 
60° C. for one hour. For children under fifteen years of age a mixture of 
5 c.c. of this solution and 10 c.c. of polyvalent antipneumococcus serum may 
be injected intraspinally twice daily after thorough drainage of spinal 
fluid. For adults 10 c.c. of the solution and 20 c.c. of serum should be 
injected twice daily for three or four days. The internal administration of 
ethylhydrocuprein also raises the pneumococcidal power of the blood and 
for an adult 0.5 gm. may be given three times daily in capsules. To the 
above mixtures for spinal injection I sometimes add 2 to 5 c.c. of fresh chicken 
serum for the purpose of reactivating the immune serum and for the natural 
immune substances in fowl serum. 
SERUM TREATMENT OF STREPTOCOCCUS MENINGITIS 
Streptococcus meningitis is much less common than pneumococcus 
meningitis. The disease almost invariably develops by extension to the 
cerebral meninges from suppuration foci in the mastoid cells, middle ear, or 
accessory nasal sinuses. Personally I have never seen streptococcus menin- 
gitis develop spontaneously, that is, without direct extension from these 
neighboring parts. 
Bacteriologically the infection is readily mistaken for pneumococcus 
meningitis. 
Treatment, such as it is, may be along the same lines as described for 
pneumococcus meningitis. The prognosis is always extremely grave. 
Spinal puncture should be done once or twice daily to release pressure and 
remove pus and organisms. Antistreptococcus serum may be injected 
intraspinally twice a day in dose of 10 to 30 c.c. I believe something is to 
be gained by adding to each dose of antistreptococcus serum 2 to 5 c.c. 
of fresh sterile human serum, for the purpose of reactivating the immune 
serum. 
A great many drugs and methods have been advocated for the treatment 
of this highly fatal disease. Hollis and Pardee5 have recently renewed the 
literature and report that 38 authentic and 15 doubtful cases of tubercu- 
lous meningitis have been recorded as cured, including 2 cases of proved 
and 2 doubtful cases of their own treated with intraspinal injections of anti- 
meningococcus serum. They also report on 3 additional cases treated with 
this serum, but with a fatal outcome in all. These cases, in addition to 
TREATMENT OF TUBERCULOUS MENINGITIS 
1 Berl. klin. Wchn., 1911, 48, 1561, 1650, 1779, 1983. 
2 Jour. Exper. Med., 1915, 22, 551. 
3 Jour. Infect. Dis., 1917, 20, 313. 
4 Jour. Infect. Dis., 1920, 26, 355. 
6 Arch. Int. Med., 1920, 26, 49. 966 
VACCINES, SERA, BLOOD, PROTEINS IN TREATMENT OF DISEASE 
the one reported by Schaeffer,1 makes a total of 8 treated with intraspinal 
injections of antimeningococcus serum, 5 of whom recovered. 
Treatment consists of spinal puncture, drainage of the fluid and the 
injection of 15 to 30 c.c. of antimeningococcus serum once a day for three 
or four days, and thereafter at intervals of two or three days for two or 
more injections. The beneficial effects are probably entirely non-specific 
and mainly due to the local irritation of the meninges with the outpouring 
of serous and cellular elements and principally polymorphonuclear leuko- 
cytes. Probably the same effects could be secured by the injection of 
sterile normal horse-serum. 
Tilli2 has reported the recovery of 1 case of tuberculous meningitis in a 
child ascribed to the subcutaneous injection every three days of 1 to 3 c.c. 
of its own spinal fluid (autoserum). 
Manwaring3 has observed that the intraspinal injection of extracts of 
dog leukocytes had some slight beneficial influence upon the course of experi- 
mental tuberculous meningitis of dogs. The injection of extracts of horse 
leukocytes had the effect of slightly delaying the development of the disease. 
The injection of rabbit leukocytes into experimentally infected monkeys 
also had a tendency to prolong the infection. These results suggest that 
it may be worth while to inject leukocytic extracts (commercially prepared 
of horse leukocytes) in the treatment of this highly fatal disease; for a child 
the dose may be 5 to 10 c.c. by intraspinal injection. 
TREATMENT OF ANTHRAX 
Infection and Immunity in Anthrax.—Anthrax is a disease of the lower 
animals and especially of sheep and cattle, transmissible to man. Natural 
immunity among the lower animals varies greatly; thus mice, guinea-pigs, 
and rabbits are more susceptible than sheep and cattle, while the dog and 
rat are relatively immune and fowls and cold-blooded animals highly immune. 
Infection of man and the lower animals usually takes place by the en- 
trance of virulent spores in the hair follicles or injuries of the skin; the 
resulting lesion is known as the “malignant pustule.” Spores may be 
inhaled in dust and produce lesions in the respiratory tract, designated as 
“wool-sorters’ disease.” Bacilli or spores may be swallowed, escape de- 
struction by the gastric juice and produce intestinal lesions. The latter 
infections are highly fatal forms of anthrax. Man is usually infected by 
handling hides, hair, and wool contaminated with the spores, which may 
survive drying or immersion in pickling fluids and brines for long periods 
of time. 
For susceptible animals the anthrax bacillus is highly invasive, that is, 
capable of surviving in the body fluids and rapidly multiplying in the lym- 
phatic and vascular channels. Curiously the presence of exogenous and 
endogenous toxins have not been conclusively demonstrated in cultures, 
although the general or bacteremic infection is accompanied by symptoms 
and degenerative changes in the parenchymatous organs characteristic of 
acute toxemia and especially in the terminal stages of the disease, when the 
very rapidly developing toxemia is overwhelming. 
Natural immunity to anthrax and recovery from infection are apparently 
largely due to phagocytosis, as claimed by Metchnikoff and supported by the 
experiments of his associates. At least the presence of bactericidal sub- 
stances in the serum does not appear to bear a relation to immunity, inas- 
1 New York Med. Jour., January 11, 1913. 
2 Policlinico, 1916, 33, 1357. 
3 Jour. Exper. Med., 1912, 15, 1; ibid., 1913, 17, 1. TREATMENT OF ANTHRAX 
967 
much as the serum of the susceptible rabbit may be bactericidal, and that, 
of the resistant dog and rat almost lacking in bactericidal activity. Op- 
sonins in the serum have been shown by Wright to possess an important 
bearing upon immunity by reason of their relation to phagocytosis. An- 
thrax bacilli are also able to produce an aggressin or toxic substance exerting 
negative chemotaxis; a part of the action of antianthrax serum is the 
neutralization of these substances which thereby facilitates phagocytosis. 
One attack of anthrax does not confer a lasting immunity against the 
infection. 
Antianthrax Serum.—The production of antianthrax serum by the 
immunization of horses or other animals with the bacilli and spores was 
among the earliest attempts in the development of serum therapy; owing 
to the virulence of the spores, the high invasiveness of the bacilli for the 
blood-stream, and the susceptibility of the horse to infection the production 
of immune sera is a matter of some difficulty and especially since the horse 
produces antibodies for anthrax rather slowly and seldom to a high degree. 
Marchoux1 produced a sheep antiserum that protected rabbits against 
fatal doses of anthrax, but Sclavo2 was first to produce the serum on a large 
scale for the treatment of human beings. Sclavo immunizes animals over a 
long period of time, sometimes as long as two years. This serum has been 
extensively used in Europe and South America and to some extent in this 
country. 
In the United States antianthrax serum has been produced by Eichhorn, 
Berg and Kelser3 in the Bureau of Animal Industry. Horses are immunized 
with injections of living spore vaccine while protected against infection with 
injections of anthrax serum. Great care is exercised to prevent contamina- 
tion of the serum and elaborate tests made to insure the absence of anthrax 
spores. Eichhorn and his associates4 have succeeded in concentrating the 
serum in the same manner as diphtheria antitoxin, finding that the immune 
principles are carried down in the globulin fractions. 
Antianthrax serum is now being prepared by the different manufacturing 
firms and is readily available for the treatment of the disease. 
The literature on the results of the serum treatment of anthrax is quite 
large and generally favorable to the curative value of the serum for anthrax 
in human beings and the lower animals. Regan5 has recently published a 
good summary. 
According to Sclavo the general mortality of anthrax in Italy without 
serum treatment is about 24 per cent.; with serum treatment the mortality 
is lowered to about 6 per cent. Mendez reports a mortality of 4.19 per cent, 
among cases treated with his serum.6 The reports of Cicognani,7 Legge,8 
Herley,9 Royer and Holmes,10 Schwartz,11 Regan,12 Regan and Regan,13 Sym- 
mers,14 Hubbard and Jacobson,15 and numerous others indicate that the 
1 Ann. de l’lnst. Pasteur, 1895, 9, 785. 
2 Berl. klin. Wchn., 1901, 18, 19, 481, 520. 
3 Jour. Amer. Vet. Assoc., 1915-16, xlvii, 669. 
4 Jour. Agricult. Research, 1917, 8, 37. 
8 Amer. Jour. Med. Sci., 1921, 162, 406 
6 Centralbl. f. Bakteriol., 1899, 26, Nos. 21 and 22. 
7 Gaz. d. ospeded. e. d. chir., 1901, No. 114. 
8 Brit. Med. Jour., March 18, 1905, 589. 9 Lancet, 1909, 2, 1662. 
10Penna. Med. Jour., 1907, 2, 937. 
11 New York Med. Jour., 1918, cvii, 1171. 
12 Jour. Amer. Med. Assoc., 1919, 72, 1724. 
13 Amer. Jour. Med. Sci., 1919, 157, 782. 
14 Weekly Bulletin of New York Health Depart., August 7, 1920. 
18 Monthly Bulletin of New York Health Depart., November, 1920. 968 
VACCINES, SERA, BLOOD, PROTEINS IN TREATMENT OF DISEASE 
administration of potent antiserum is effective in reducing the general 
mortality of anthrax. 
The serum may be injected locally around the base of the malignant 
pustule in anthrax of the skin as advocated by Regan and doubtless aids 
in the processes of phagocytosis, but local anthrax is not particularly dan- 
gerous and various other forms of treatment may be employed with success. 
From the standpoint of serum treatment most interest attaches to the 
treatment of anthrax bacteremia and internal types of the disease (pul- 
monary and intestinal) which are accompanied by a mortality of 90 to 100 
per cent. Pied1 in 1913 states that up to that date he found reports of 
7 cases of anthrax bacteremia treated with serum, with recoveries in 5 cases. 
Bandi,2 Bissel,3 Graham and Detweiler,4 and others have reported recoveries 
of blood infections under serum treatment. 
Treatment of Anthrax of the Skin (Malignant Pustule).—This is the 
usual form of anthrax in human beings and generally occurs among workers 
in hair, wool, and hides. The disease has also been caused by infected 
shaving brushes, hair brushes, and other articles made of hair. 
Doubtless many cases of skin anthrax escape diagnosis and are treated 
by simple incision, drainage, and the application of wet dressings as in the 
surgical treatment of any local pyogenic infection. 
However, the marked edema and involvement of the neighboring glands 
are apt to lead to bacteriologic examinations by smear and culture and 
correct diagnosis. The problem is then primarily one of prevention of 
anthrax bacteremia. Histologic examination of a pustule shows enormous 
numbers of bacilli in the lymphatic spaces with some walling off of leukocytes, 
but the zone of leukocytes is poorly developed, inasmuch as negative chemo- 
taxis is exerted by toxic aggressins of the organisms. 
Incision is dangerous and tends to break down the local defenses with 
consequent spread of the infection. When the lesion is large, in which case 
it may be mistaken for a carbuncle, simple incision is frequently followed 
by the development of bacteremia and death. 
Destruction of the lesion with the thermocautery is one of the oldest 
treatments still widely practised in many parts of the world. When large 
lesions are removed it may lead to slow convalescence and the production 
of large scars. One other objection is that coagulation of the tissues inter- 
feres with subsequent drainage and free drainage is important for the pre- 
vention of the spread of the infection and more particularly of the dreaded 
generalized bacteremia. 
The local injection of germicides, as phenol, iodin, bichlorid of mercury, 
zinc chlorid, and other substances, has long been a favorite treatment. The 
plan is to inject a germicide, 3 per cent, carbolic acid being mostly employed, 
at different points in the base of the pustule in order to destroy the bacilli, 
excite leukocytosis and liquefaction necrosis of the pustule. This treatment, 
however, is apt to prove very painful and due care must be exercised against 
the injection of toxic amounts of the germicide. 
Excision of the lesion is widely practised. When thoroughly done 
including not only the pustule, but a portion of the edematous base, it 
removes at once the necrotic tissue and an enormous number of bacilli 
leaving a clean base for drainage. A drawback, however, is the resulting 
scar and deformity, although in the Philadelphia Hospital for Contagious 
1 Bull, med., 1913, 1137. 
2 Lancet, 1904, 2, 372. 
2 New York Med. Jour., July 21, 1917. 110. 
4 Jour. Amer. Med. Assoc., 1918 70, 671. TREATMENT OF ANTHRAX 
969 
Diseases, where excision is commonly practised, I have seen many cases 
with extensive excision heal promptly without secondary pyogenic infection 
and with surprisingly slight scar formation and deformity. 
The local injection of antianthrax serum has been advocated particularly 
by the Regans. “The method consists in the injection of 2, 2.5, or even 
3 c.c. of serum at each of three or four points equidistant from one another 
at the various sides of the pustule. The needle is best inserted into the 
red indurated area of the pustule just beyond the blanched zone, the serum 
being directed toward the base of the eschar and injected so as to circum- 
scribe the lesion. The injections are given once, twice, or three times in the 
twenty-four hours, depending on the severity of the case, and not more 
than 7 to 10 c.c. are injected at one time. Commonly four to six injections 
suffice.” 
The size of the lesion is not a reliable index of the chances for blood 
infection; in 1 case I observed a rapidly fatal bacteremia from a very small 
pustule developing in a hair follicle on the chest of a male worker in hides. 
The following plan is advised for the systematic treatment of the malig- 
nant pustule, the various steps being given in consecutive order: 
1. Give 1 c.c. of antianthrax or normal horse-serum subcutaneously for 
the purpose of desensitizing the patient in preparation for the intravenous 
injection of serum. This is not necessary if the large dose of serum is given 
subcutaneously or intramuscularly. 
2. Under ether anesthesia excise the lesion, taking in a wide base and 
handling the lesion as little as possible, being particularly careful not to 
squeeze or handle it roughly. After excision and attention to hemostasis, 
the wound may be dusted with calomel and powdered ipecac and dressed 
with sterile gauze. Ipecac was advocated in 1888 by Muskett1 and has 
been used in the Philadelphia Hospital for Contagious Diseases with con- 
siderable success. Both ipecac and its alkaloid emetin are anthracidal, the 
latter being studied by Smith and the writer.2 Or the wound may be dressed 
with 1 : 1000 solution of bichlorid of mercury or mercurophen in 10 per cent, 
sodium chlorid solution, the bichlorid or mercurophen acting as a preventive 
for secondary pyogenic infection and an anthracidal agent, while the hyper- 
tonic saline excites a reversal of lymph flow into the wound and encourages 
drainage. Mercurophen has been produced by Schamberg, Raiziss, and the 
author, and possesses an extremely high anthracidal activity even in the 
presence of blood and serum. 
3. If, however, excision is refused or cannot be practised, the local 
injection of antianthrax serum after the method of Regan and described 
above, is advised. Furthermore, even if the lesion is excised, but the edema 
persists with signs of extension of the disease around the margins of the 
wound, serum should be injected locally as described. 
4. The next step is to conduct a blood-culture, removing 5 to 10 c.c. of 
blood from a vein at the elbow into a flask containing 100 to 200 c.c. of 
nutrient broth. Special care should be exercised against contamination, 
because Bacillus subtilis may be mistaken for B. anthracis. 
5. It is now advised to give an injection of antianthrax serum. If the 
lesion is quite small it will suffice to give 40 c.c. intramuscularly. If the 
lesion is large, with considerable edema and lymphadenitis, it is better to 
give 50 to 100 c.c. intravenously. 
6. The blood-culture is closely watched. If anthrax bacteremia is 
present the bacilli will grow up within eighteen to twenty-four hours and 
indicate vigorous intravenous treatment. If the blood-culture is sterile 
1 Lancet, 1888, 1 269. 
Jour. Infect. Dis., 1916, 18, 247. 970 
VACCINES, SERA, BLOOD, PROTEINS IN TREATMENT OF DISEASE 
at the end of twenty-four hours, serum need not be given, but it is advisable 
to repeat the blood-culture daily for the following two or three days. 
Treatment of Anthrax Bacteremia and Internal Anthrax.—Antianthrax 
serum should be injected intravenously in dose of 100 to 200 c.c. as soon as 
possible. It is also a good plan to include an injection of neoarsphenamin, 
by reason of the reports indicating that arsphenamim and neoarsphenamin 
exert a curative influence. 
My plan is to dissolve 0.9 gm. neoarsphenamin in 100 c.c. of sterile saline 
solution, add 100 c.c. of antianthrax serum, and inject the mixture intra- 
venously. 
Twelve hours later a second injection of serum alone should be given. 
Twelve hours later the injection of serum and neoarsphenamim may be 
repeated. Blood-cultures should be made daily and the injections of serum 
kept up until the blood is found to be sterile. If and when this fortunate 
result is obtained, at least one intramuscular injection of 30 to 50 c.c. of 
serum is advised as a precautionary measure. 
If antianthrax serum is not available it is well to inject fresh sterile beef 
serum, which may be collected by bleeding from a jugular vein or an abattoir, 
blood being collected in a sterile vessel and the serum separated. Kraus, 
Penna and Cuenca1 have treated anthrax in the Argentine Republic by 
injecting normal beef serum, perviously heated twice at 56° C. for thirty 
minutes, in doses of 30 to 50 c.c. intramuscularly and repeating in twelve, 
twenty-four, or thirty-six hours, as the case may require. According to 
their latest report on 200 cases the mortality has been reduced to 0.5 per 
cent, as compared with 10 per cent, in 250 cases treated in the usual 
manner during the preceding ten years. Solari2 and Langon3 have also 
reported favorably, but Lignieres4 states that beef serum is inferior to 
antianthrax serum. 
Koehler, Wanner, and the writer5 found normal beef serum in this country 
only feebly anthracidal and without protective value for experimental 
infections in mice. However, since anthrax is widely prevalent among 
cattle in the Argentina, it is probable that beef serum there possesses higher 
curative values by reason of natural immunization of the cattle. A more 
probable explanation for the curative activity of normal beef serum is, 
however, that it excites a non-specific protein shock reaction of which the 
resulting leukocytosis is probably curative. At any rate the work of Kraus, 
Penna, and their associates indicates that normal beef serum is worthy of 
trial, although I believe that preference should be given antianthrax serum 
when it is obtainable. 
THE SERUM TREATMENT OF PLAGUE 
In so far as specific therapy is concerned the prompt administration of 
aniiplague serum in large doses and by intravenous injection has proved of 
decided value especially in the treatment of the bubonic type of the disease; 
in the highly fatal pneumonic type it has also apparently saved some lives. 
Antipest Serum.—The preparation of this serum is a difficult matter and 
different methods have been described. 
The serum at present mostly employed is obtained from horses after 
repeated intravenous injections of killed cultures sometimes followed by 
1 Prenza Med. Argentina, 1917, 3, 297; ibid., 1917, 4, 91, 147; ibid., 1918, 4, 455. 
2 Semana med., 1917, 24, 98. 
3 Prenza Med. Argentina, 1917, 4, 49, 370. 
4 An. de Facul. med., 1918, 3, 258. 
5 Jour. Infect. Dis., 1920, 26, 148. THE SERUM TREATMENT OF PLAGUE 
971 
living cultures after the method described for the preparation of anti- 
dysentery serum. It is bactericidal for the bacillus of pest and also contains 
agglutinins that aid in bacteriolysis and phagocytosis. Potent sera also 
contain opsonins, and a part of its protective and curative properties are 
ascribed to their presence. Some sera are feebly antitoxic for the small 
amounts of exogenous toxins produced by the organisms. 
(a) In addition to Yersin’s serum, which is prepared at the Pasteur Institute of Paris 
by immunizing horses with dead and then with living cultures of pest bacilli, other serums 
have been prepared. For example: 
Cb) Kolle immunizes horses with intravenous injections of heat-killed cultures, begin- 
ning with I agar slant culture and doubling the dose each week until 15 cultures are given 
at one time. The horses are bled fourteen days after the last dose is given. 
(c) Lustig immunizes horses with pest-nucleoproteins, obtained by breaking up the 
bacilli with 1 per cent, of potassium hydroxid and precipitating the proteins with acetic 
acid. These are then suspended in sterile normal salt solution, as in the preparation of 
Lustig’s vaccine. 
(d) Terni-Bandi immunizes donkeys and sheep with aggressins obtained by intraper- 
itoneal injection of guinea-pigs with pest bacilli. 
(e) Markl immunizes horses with filtrates of old pest bouillon cultures. He believes 
that the value of pest serum is largely dependent upon antitoxins. 
The serums are usually tested by injecting mice with lethal doses of pest culture and 
decreasing doses of antiserum. The agglutinin content may also be measured. According 
to Strong pest immune sera do not contain appreciable amounts of bacteriolysin, but are 
largely bacteriotropic in action. Whenever cultures are used in immunization, the serum 
should always be cultured carefully and tested by animal inoculation to guard against the 
possibility of living bacilli being present. 
Dosage and Administration of Serum.—A common mistake made in the 
serum treatment of this disease has been the administration of too small 
doses (10 to 20 c.c.) by subcutaneous injection. Larger amounts of serum 
are required at the earliest possible lime. If intravenous injection is not 
possible, as may be the case in children, 40 to 80 c.c. of serum should be 
injected intramuscularly. Fontes has given the serum intraperitoneally to 
children and has published the following table showing the results: 
Method. 
Cases Treated. 
Died. 
Death-rate. 
Subcutaneous 
21 
8 
38 
Intraperitoneal 
11 
2 
18 
Intravenous 
69 
5 
7.2 
Chocky,1 Moreno,2 Guiteras and Ricio,3 and others have also advocated 
the intravenous injection of large doses of serum immediately upon the 
confirmation of the diagnosis. 
The importance of early treatment is shown in the table published by 
•Choksy in Bombay: 
Day or 
Number of 
Re- 
Mortality, 
the Disease. 
Cases. 
Died. 
COVERED. 
Per Cent. 
First 
323 
98 
225 
30.3 
Second 
311 
164 
147 
52.7 
Third 
248 
155 
93 
62.5 
Fourth 
106 
60 
46 
56.6 
Fifth 
52 
32 
20 
61.5 
Sixth 
14 
8 
6 
57.1 
Seventh 
4 
4 
0 
100 
1 Brit. Med. Jour., 1908, 1, 1282. 
2 Series of seven papers in Semana med., 1915, 22, beginning with No. 30 and ending with 
No. 45. 
3 Library and Press: La Moderna Poesia, Havana, 1915. 972 
VACCINES, SERA, BLOOD, PROTEINS IN TREATMENT OF DISEASE 
The following plan of serum treatment, slightly modified after that 
employed by Chosky, is recommended: 
1. The intravenous injection of 80 to 100 c.c. of serum as soon as the 
diagnosis is made. For children under twelve years the dose may be 50 c.c. 
If an intravenous injection cannot be given a child, the serum may be in- 
jected intramuscularly or intraperitoneally. 
2. Repeat the injection six hours later if the temperature has not fallen. 
If this has occurred the second injection may be given twelve to eighteen 
hours later. 
3. The amount of serum to be injected subsequently will depend upon 
the rise of the temperature the previous evening, and the general condition 
of the patient. If the temperature be the same as on the first evening, the 
same amount may be injected; if it be lower, 30 c.c. or less should be injected. 
4. The quantity of serum injected should be lessened gradually until the 
temperature falls to the normal in the mornings. 
5. A sudden fall of the temperature between the second and the seventh 
day should not indicate a suspension of the treatment. 
6. The injections are given every twelve hours if secondary buboes 
present themselves, or if the temperature rises rapidly one or two degrees. 
7. If the evening temperature be lower than that of the morning, the 
dose is reduced on the following day. 
8. From six to eight injections are generally sufficient to complete the 
treatment. 
9. The total quantity of serum required varies between 150 and 300 c.c. 
according to the gravity of the symptoms and the condition of the serum. 
Value of Serum Treatment of Plague.—When serum was first introduced 
by Yersin in 1896 the doses were small (10 to 20 c.c.) and usually adminis- 
tered by subcutaneous injection. Since then Yersin and others have em- 
ployed larger doses, but the mortality statistics vary greatly owing not only 
to the use of sera of varying potencies, but to different doses and routes of 
administration and to the fact that plague epidemics vary greatly in viru- 
lence. The table shown on page 973, prepared by Dujardin-Beaumetz and 
quoted by Guiteras and Ricio, shows the wide fluctuations in mortality rates 
of plague treated with serum by different observers. 
Chosky’s mortality varied from 30 per cent, among cases treated on the 
first day to 100 per cent, treated on the seventh day, as shown in the table 
given above. His general mortality was 63.5 per cent, for 400 cases treated 
with serum as compared with a mortality of 74 per cent, among 200 cases 
treated without serum. Burnett1 reported a mortality of 29.7 per cent, 
among serum treated cases as against 73.9 per cent, treated without serum. 
Penna in Argentina, reports that among 664 cases treated with Yersin’s 
serum during the period 1905 to 1912, the mortality ranged from 23 per 
cent, in 1906 to 7.3 per cent, in 1912, with an average mortality of 12.5. 
From 1914 to the middle of 1919 Kraus’ serum was used with an average 
mortality of 7.8. per cent. While the Plague Commission of India2 did not 
report very favorably upon the use of Yersin’s or Lustig’s sera in 1913, 
later statistics indicate an inprovement in the mortality rates ascribed to 
the use of more potent sera in larger doses and by intraveneous injection. 
1 Report on Plague in Queensland, 1900-07, Brisbane, 1907. 
2 British Commission, Seventh Report on Plague Investigation in India, Jour. Hyg., 
Supplement II, 1913, 326. ' TREATMENT OF ASIATIC CHOLERA 
973 
Author. 
Epidemic. 
Treated. 
Died. 
Mortal- 
ity, 
Per Cent. 
Canton and Amoy, 1896 
26 
2 
7.6 
Bombay, 1897 
50 
17 
34.0 
Nha-Trang, 1898 
33 
14 
42.4 
Mandvi, 1898 
136 
89 
65.4 
Simond 
Bombay, Karad, Moundra, 1898... 
171 
99 
57.8 
Kuratchi, 1898 
75 
37 
49.3 
Mongolia, 1898 
16 
12 
75.0 
Delay 
Mongtze, 1898 
10 
4 
40.0 
Tamatave, 1898-99 
20 
11 
55.0 
Calmette and Salimbeni. . 
Oporto, 1899 
142 
21 
14.7 
Oporto, 1899 
6 
1 
15.6 
Noumea, 1899-1900 
7 
2 
28.5 
Noc 
Noumea, 1901 
17 
8 
46.9 
Anher 
Reunion, 1899 
8 
1 
12.5 
Reunion, 1900-01 
13 
9 
15.3 
Majunga, 1902 
71 
32 
45.0 
Foa-Tcheou, 1902. 
67 
34 
50.7 
Tonkin, 1903 
101 
51 
50.5 
Choksy 
Bombay 
51 
37 
72.5 
Rosario 
26 
11 
42.3 
Buenos Aires, 1905 
204 
29 
19.3 
Del Rio and Zegers 
Inique, 1903 
85 
38 
44.7 
Autofogasta 
50 
3 
6.0 
Chanarai, 1904 
18 
1 
5.5 
Santos, 1900 
19 
7 
36.8 
Rio Grande, 1902 
45 
7 
15.5 
Rio Janeiro, 1900 
410 
138 
33.6 
Rio Janeiro, 1901 
278 
99 
35.6 
Rio Janeiro, 1902 
268 
68 
25.3 
Rio Janeiro, 1903 
541 
124 
22.9 
Rio Janeiro, 1904 
504 
103 
20.4 
Rio Janeiro, 1905 
149 
25 
16.5 
Rio Janeiro, 1906 
187 
54 
28.8 
Campos, 1902 
136 
27 
19.1 
Tavares de Marcedo 
Campos, 1906 
14 
2 
14.2 
A. Ferrari 
Rio Janeiro, intravenous, 1907 
69 
5 
7.2 
TREATMENT OF ASIATIC CHOLERA 
Numerous attempts have been made to develop an efficient serum 
therapy for this disease. 
Anticholera Serum.—In some respects cholera would seem to be due 
mainly to a toxin elaborated by the bacilli in the intestinal tract of infected 
persons, similar to the action of the toxin of the Kruse-Shiga type of dysen- 
tery bacillus. Various attempts have been made to prepare an efficient 
anticholera serum, but the only one that has yielded encouraging results 
in experimental infections as well as in cholera of human beings is that 
prepared by Kraus. This serum is prepared by immunizing horses with a 
true toxin derived from a cholera-like vibrio isolated by Gottschlich from 
the intestinal contents of pilgrims dying at El Tor from a dysentery or 
cholera-like infection. According to Kraus, this antiserum is largely anti- 
toxic, and serves to neutralize the toxin of true cholera more effectively than 
does the antiserum resulting from immunization with cholera cultures. 
Antisera have also been prepared by immunization of horses with ground 
organisms in order to render it anti-en do toxic; also with killed and living 
cultures of the vibrio and extracts of these, by Carriere and Tomarkin.1 
1 Ztschr. f. Immunitatsf., orig., 1909, 4, 30. 974 
VACCINES, SERA, BLOOD, PROTEINS IN TREATMENT OF DISEASE 
These antisera in addition to being antitoxic for exogenous and endog- 
enous toxins of virulent cultures, are likewise bactericidal, agglutinative, 
and opsonic; their value in the treatment of the disease probably depends 
upon a combination of these activities. 
Value of Anticholera Serum.—The results have not been very good. 
Doubtless the mortality rates have been influenced not only by the varying 
severity of different epidemics, but likewise by the use of sera of varying 
potencies and more especially by the doses employed and route of adminis- 
tration. 
In most instances the serum has been given by subcutaneous injection. 
In the Russian epidemic of 1908-09 Stiihlern1 employed large doses by 
intravenous and subcutaneous injection combined with saline infusion; the 
best mortality rate was about 30 per cent. Slightly better results were 
reported by Salimbeni2 in the treatment of 42 cases. Best results were 
obtained with the sera prepared by Schurupoff, and especially with that 
prepared by Carriere and Tomarkin.3 Jegunoff4 treated 12 patients with 
intravenous injections of Kraus’ serum diluted with 500 to 700 c.c. of saline 
solution, with a mortality of 25 per cent, as compared with a general 
mortality of 75 per cent, in cases receiving no serum. Hundogger5 has 
reported unfavorably upon the use of the serum; Strong6 has thoroughly 
reviewed the literature up to 1907 and reports that the results have not been 
very encouraging. 
Ketscher and Kering used Kraus’ antitoxic serum in 119 cases, with a 
mortality of 58 per cent., among those treated with subcutaneous injec- 
tions and 50 per cent, among those treated intravenously; the mortality 
among those not treated with serum was 63.4 per cent. Von Stiihlern and 
Tuschinski observed a mortality of 29.9 per cent, among serum-treated 
cases. 
More recently Livierato7 has reported the successful use of anticholera 
serum injected intravenously along with injections of hypertonic saline 
solution according to Rogers, and believes that the serum has an unmis- 
takable antitoxic action and effectually supplements treatment with saline 
solution. It may be stated that in general terms the mortality of severe 
cholera has been reduced from 70 per cent, to about 25 per cent, when serum 
has been intravenously injected in large doses; doubtless a part of these 
beneficial results are to be ascribed to the replenishment of the blood volume, 
and best results are to be secured by a combination of serum and saline 
injections given intravenously. 
Dosage and Administration.—Anticholera serum should be administered 
as early in the disease as possible and preferably by intravenous injection. 
A good plan is that employed by Jegunoff, who administered 100 to 150 c.c. 
of serum diluted with 500 to 700 c.c. of saline solution. The intravenous 
injection of saline solution alone has proved of great benefit, and especially 
during the first few days of the disease, to counteract the depletion of the 
tissues by diarrhea and vomiting. 
Rogers employs a hypertonic saline solution prepared by dissolving 
16 gm. sodium chlorid, 0.5 gm. calcium chlorid, and 0.74 gm. potassium 
1 Med. Klinik, 1909, 5, 1452. 
2 Ann. de l’Pasteur Inst., 1908, 22, 172; ibid., 1910, 24, 34. 
3 Ztschr. f. Immunitatsf., 1909, 4, 30. 
4 Wien. klin. Wchn., 1909, No. 24. 
6 Wien. klin. Wchn., 1909, No. 52. 
6 Bull. No. 16, Bureau of Govt. Lab., Manila, 1904; ibid., No. 21; Philippine Jour. Sci., 
1906, 1, 501; ibid., 1907, 2, 413. 
7 Riforma med., 1915, 31, 673. TREATMENT OF PERTUSSIS 
975 
chlorid in 1000 c.c. of sterile distilled water. From 1000 to 2000 c.c. are 
injected intravenously, aiming to restore the blood-pressure to 110 mm. Hg. 
The dose of serum should be repeated at intervals of twelve to eighteen 
hours until improvement is apparent. In children the injections may be 
given intramuscularly if intravenous injections cannot be given, but better 
results are likely to follow the intraperitoneal injection of serum (50 to 
100 c.c.) and saline solution (500 c c.) under these conditions. 
Vaccine Treatment of Cholera.—The disease has not until recently been 
treated with vaccine owing to its very rapid course. During 1915, Petro- 
vitch1 employed a vaccine for the treatment of 1153 mild cases, with a mor- 
tality of 2 cases. In 90 cases of moderate severity none died, while the 
mortality was 9.4 per cent in a similar group not treated with vaccine. In 
157 severe cases the mortality was 14.4 per cent., while of 120 similar control 
cases the mortality was 58 per cent. Saline infusions were also employed. 
Serum Treatment.—Antipertussis sera have not been prepared on a 
large scale by the immunization of horses with the bacillus of Bordet and 
Gengou. Serum treatment has been confined to the use of human serum 
or blood from normal persons and convalescent cases of the disease. 
Bleyer2 treated 45 cases during the early weeks of the disease, with 
intramuscular injections of compatible blood taken from donors within 
three months of recovery, from donors who had had the disease at more 
remote periods, and from donors who had never had pertussis. The dose 
varied from 40 to 125 c.c., divided into two, three, or four doses injected 
into the gluteal muscles. Blood not used at once was citrated to 1 per cent. 
Bleyer found that a few cases were apparently benefitted by the injections, 
but was not able to ascribe any of these effects to a specific action of the 
injected blood. 
Pentz3 reports that the intramuscular injection of 3 children with con- 
valescent serum had an apparent good effect in the way of mitigating the 
severity of the infection. 
Vaccine Treatment.—Graham4 was first to use pertussis vaccine in the 
treatment of the disease, injecting 20,000,000 bacilli every four days and 
later increasing the dosage to 40,000,000. Twenty-four cases were treated 
and 17, or 71 per cent., were apparently benefited. Since then a large litera- 
ture has accumulated on the use of vaccines for the prophylaxis and treat- 
ment of whooping-cough. The vaccines employed have not always been 
prepared of Bacillus pertussis alone; in some instances mixed vaccines, 
including B. influenza, streptococci, and other organisms, have been included. 
Sill5 reports 33 cases treated with vaccine. In all the number and 
severity of the attacks were markedly diminished. He advises 50,000,000 
every second day in moderately severe cases and 100,000,000 in very severe 
cases. He reports 13 additional cases and 3 prophylactic cases which were 
exposed and proved to be immune. All 16 cases were treated with mixed 
pertussis vaccine: Bacillus pertussis, 50,000,000; Staphylococcus aureus, 
20,000,000; micrococcus catarrhalis, 20,000,000. With best drug treatment 
the cough persisted from seven to twelve weeks; with pertussis vaccine, 
from four to four and half weeks. Of the last 10 treated with combined 
vaccine, all were cured in three and a half weeks. 
TREATMENT OF PERTUSSIS 
1 Bull, de l’Acad. de med., 1915, 74, No. 33. 
2 Amer. Jour. Med. Sci., 1917, 154, 39. 3 Nederl. Tijdschr. v. Geneesk., 1917, 1, 714. 
4 Tr. Amer. Pediat. Soc., 1911, 23, 157; Amer. Jour. Dis. Child., 1912, 3, 41. 
5 Amer. Jour. Dis. Child., 1913, 5, 379. 976 
VACCINES, SERA, BLOOD, PROTEINS IN TREATMENT OF DISEASE 
Sill cites Dr. Freeman, of London, who treated 1140 cases with vaccine, 
with the following results: Much better, vaccine, 31 per cent.; control 
saline, 21 per cent. Better, 37.1 per cent.; control saline, 34.5 per cent. 
Unchanged, 15 per cent.; control saline, 23.9 per cent. Worse, 15.2 per 
cent.; control saline, 18.15 per cent. Much worse, 1 per cent.; control saline, 
1.85 per cent. 
Sill believes that no bad effects follow the vaccine treatment; that the 
average length of the cough under this treatment is about four and one-half 
weeks. He suggests giving the vaccine treatment every day for a few days 
in very severe cases and he is guided more by the severity of the disease 
than the age of the child. The cases seen earlier in the disease before the 
height of the paroxysms is reached respond most quickly to treatment. 
The younger child receives relatively larger doses, and responds more quickly 
to treatment. He considers this important because of the grave danger 
of pneumonia developing in the young child. He also believes that vaccine 
may confer immunity. 
Nicolle and Conor1 used living cultures of the Bordet bacillus, injected 
under the skin, every two or three days. The treatment was given to 122 
children, but 18 were not seen after the first injection. Of the 104 that 
were treated, 37 were cured, 40 improved, and 27 remained stationary. 
In the cases cured the improvement was manifest early. The improvement 
was noticed after the first or second injection, the number and severity of 
the paroxysms being diminished. 
Saunders, Johnson, White and Zahorsky,2 Ladd,3 Scott,4 Wilson,5 Luttin- 
ger,6 Meyers,7 Bamberger,8 Hartshorn and Moeller,9 Huenekens,10 Polozker,11 
Shaw,12 Reynolds,13 Bloom,14 Rewalt,15 Davies,16 and others, have reported 
favorable results from the treatment of pertussis especially when the injec- 
tions were given early in the disease. 
Bacher and Menschikoff,17 Hess,18 von Sholly, Blum and Smith,19 Baren- 
berg,20 and others did not find the vaccine of any particular value in treat- 
ment, although the consensus of opinion is to the effect that the vaccine 
when freshly prepared and given early in the disease in appropriate doses 
is of distinct value. Bloom states that vaccine treatment tends to minimize 
loss in weight, reduces the duration of the disease, decreases the severity 
of the disease, and reduces the percentage of complications and mortality. 
Dosage and Administration of Pertussis Vaccine.—Stock vaccines are 
generally employed. These may be prepared of different strains of B. 
pertussis alone or in combination with pneumococci, streptococci, and other 
organisms. 
The author employs a stock vaccine of several strains of B. pertussis 
with different strains of pneumococci in order to obtain some degree of 
1 Acad. d. Sci., 1913. 
2 Pediatrics, 1912, 24, 161. 
3 Arch. Pediat., 1912, 29. 
4 New York Med. Jour., 1913, xcvii, 176. 
5 New York Med. Jour., 1913, xcvii, 823. 
6Med. Record, 1913, lxxxiv, 1125; Jour. Amer. Med. Assoc., 1917, 68, 1461. 
7 Pediatrics, 1914, 26, 450. 
8 Amer. Jour. Dis. Child., 1913, 5, 33. 
9 Arch. Pediat., 1914, 31, No. 8 (gives a review of the literature covering 1445 cases). 
10 Amer. Jour. Dis. Child., 1918, 17, 29. 
11 Amer. Jour. Obstet., 1916, 73, 551. 16 Amer. Jour. Dis. Child., 1922, 23, 423. 
12 Amer. Jour. Obstet., 1917, 76, 161. 17 Centralb. f. Bakteriol., 1912, 61, 218. 
13 Arch. Pediat., 1919, 36, 290. 18 Amer. Jour. Obstet., 1914, lxx, 510. 
14 Arch. Pediat., 1919, 36, 1. 19 Jour. Amer. Med. Assoc., 1917, 68, 1451. 
1BPenna. Med. Jour., 1921, 24, 404. 20 Amer. Jour. Dis. Child., 1918, 16, 23. TREATMENT OF GONOCOCCUS INFECTIONS 
977 
prophylactic immunization against pneumococcus pneumonia at the same 
time. This vaccine is prepared of 400,000,000 of B. pertussis and an equal 
number of each of the four types of pneumococci per cubic centimeter, 
yielding a vaccine containing a total of 2,000,000,000 per cubic centimeter. 
For children under five years of age the dose is 0.1 to 0.2 c.c. (200,000,000 
to 400,000,000) every other day by subcutaneous injection; for children 
five to twelve years the dose is 0.5 to 1 c.c. every three to four days by sub- 
cutaneous injection. 
After the paroxysmal stage has been passed, the writer employs an 
autogenous vaccine for the treatment of the bronchitis and residual infection. 
This is usually prepared of staphylococci, streptococci, and pneumococci 
depending upon the bacteriologic findings after plating sputum on blood- 
agar plates. The vaccine is made up of equal proportions of the different 
bacteria, each cubic centimeter containing 2,000,000,000. The doses begin 
with 0.1 c.c. and are gradually increased at intervals of four to six days. 
Immunity in Gonococcus Infections.—Human beings only are naturally 
susceptible to infection by the gonococcus, although the urethral and con- 
junctival mucous membranes of some of the apes have been experimentally 
infected with virulent cultures with the production of slight and rapidly 
disappearing inflammatory changes resembling gonococcus urethritis and 
conjunctivitis of human beings. Other lower animals, as rabbits and guinea- 
pigs, may be intoxicated by the intraperitoneal and subcutaneous injection 
of virulent gonococci and autolysates, but the mucous membranes resist 
infection when the gonococcus is applied in culture or in pus. 
Little is known of the nature of natural immunity to the gonococcus in 
human beings. The organism appears to possess a selective affinity for 
certain tissues, notably the mucous membranes of the urethra and adnexa 
of both sexes, the eye (conjunctival mucosa and iris), endocardium, and 
synovial membranes. Infections of the nasal and oval cavities are very 
rare; the rectum is sometimes infected and the meninges and other interval 
organs are very rarely attacked. The mucous membranes of the young are 
more susceptible than adult tissues; the conjunctival mucosa of babies 
and the mucosa and skin of the urogenital system of female children are 
especially susceptible, whereas the conjunctival mucosa of adults apparently 
acquires natural resistance in view of the relative infrequency of involve- 
ment considering the opportunities for infection. 
Serologic strains of gonococci probably exist analagous to the different 
types of meningococci. The investigations of Louisa Pearce1 indicate that 
the gonococci producing infantile vaginitis may be different serologically 
from those producing adult infections. Torrey and Buckell2 have recently 
studied the problem in a most exhaustive manner and found on the basis 
of agglutinin absorption tests, that it was not possible to group gonococci 
in distinct immunologic types. However, they were able to classify 
their strains under three general headings of (a) regular, (b) intermediate, 
and (c) irregular strains. No definite serologic distinction could be 
drawn between strains isolated from vulvovaginitis cases in children 
and those from gonorrheal infections in adults. There appears to be 
abundant clinical evidence to the effect that the gonococci from urogenital 
infections of adults of both sexes may produce gonococcus infection of the 
urogenital and other mucous membranes of children of both sexes; the 
TREATMENT OF GONOCOCCUS INFECTIONS 
1 Jour. Exper. Med., 1915, 21, 289. 
62 
2 Jour. Immunology, 1922, 7, 305. 978 
VACCINES, SERA, BLOOD, PROTEINS IN TREATMENT OF DISEASE 
evidence is much less conclusive that gonococci from infantile infections 
are able to excite the disease in adults, although this is probably the case. 
The great majority of human beings are apparently susceptible to 
gonococcus infection; the escape of individuals in a group exposed to infection 
is to be ascribed more to fortuitous circumstances than to individual resist- 
ance. However, a natural local or systemic immunity has been observed 
by Hess1 in a small number of female children who have escaped infection 
when exposed to cases of gonococcus vaginitis one or more times. 
Acquired immunity to the gonococcus never reaches a high degree. An 
individual recovering from gonorrhea is susceptible to reinfection. Probably 
the latency of chronic infections is due in part to the production of immunity 
principles retarding the activity of the organisms; it is known, however, 
that these gonococci are able to infect others and that, in turn, these may 
superinfect the original host. It would appear, therefore, that the tendency 
to latency of gonococcus disease in both sexes and the spontaneous cure of 
vaginitis of children as they approach puberty are to be ascribed more to 
tissue changes, alteration of secretions, and reduction of virulence of the 
organisms than to the acquisition of curative and protective immunity 
principles. 
Resistance to gonococcus infection and recovery from infection is largely 
a function of phagocytosis, although there is abundant evidence to indicate 
that the gonococcus enjoys a peculiar resistance to the destructive activities 
of the endolysins (cytases) of leukocytes, and that the leukocytes in turn 
are relatively slightly harmed by the gonococci. Gonococci produce toxins, 
largely endogenous, which are regarded as the principle pyogenic substances. 
In chronic and wide-spread gonococcus disease various specific anti- 
bodies, as opsonins, bacteriolysins, agglutinins, precipitins, and complement- 
fixing antibodies, may be found in the blood, but of these, the opsonins are 
of most importance. The gonococcus, however, is not actively antigenic, 
that is, is not very active in stimulating the antibody-producing tissues. 
For this reason antibody production in ordinary gonococcus infection does 
not reach a high degree, active prophylactic immunization against the 
disease has not been accomplished, vaccine treatment has not proved gen- 
erally successful, and the immunization of horses and other animals for the 
production of immune sera is prolonged and seldom results in the produc- 
tion of highly potent immune sera. 
Antigonococcus Serum in the Treatment of Arthritis; Acute Epididymitis; 
Prostatitis and Orchitis; Gonococcus Septicemia.—Antigonococcus serum 
is prepared by the immunization of goats and horses with different strains 
of gonococci and was first described by Torrey and Rogers2 in 1906. 
Torrey’s serum is prepared by immunizing rams with gradually increasing 
intraperitoneal doses of dead, and later of living, cultures of gonococci. 
Larger amounts of serum may be secured by immunizing horses according 
to the methods described for the preparation of meningococcus and strep- 
tococcus immune serums, and in view of the larger doses now advocated, this 
is advisable. This serum has been successfully concentrated in the same 
manner as is diphtheria antitoxin by Heinemann and Gatewood.3 
According to Torrey, antigonococcus serum is largely bactericidal in 
nature. The presence of antitoxins has not been demonstrated, but potent 
sera are also capable in some degree of neutralizing the endotoxins liberated 
or formed when the cocci are broken up. The serum also contains agglu- 
1 Amer. Jour. Dis. Child., 1916, 12, 466. 
2 Jour. Amer. Med. Assoc., 1906, xlvi, 261, 273. 
3 Jour. Infect. Dis., 1912, 10, 259. TREATMENT OF GONOCOCCUS INFECTIONS 
979 
tinins, complement-fixing antibodies, and opsonins, and its curative proper- 
ties are probably due to the activities of these antibodies in addition to the 
production of fever, leukocytosis, and other non-specific agencies. 
Antigonococcus serum has been employed to some extent in the treatment 
of ordinary acute gonorrhea of the male and female, but only in doses of 2 to 
3 c.c. by subcutaneous injection which were insufficient for bringing into 
play specific and non-specific agencies. There is evidence indicating that 
fever of 101° to 103° F. is curative in gonococcus infections, inasmuch as the 
gonococcus is highly susceptible to heat, and rapid improvement has been 
noted by some observers when the serum was given intravenously followed 
by a febrile reaction and leukocytosis. These effects are purely non-specific, 
and since the reaction is unpleasant, serum therapy is not employed in the 
treatment of ordinary acute gonorrhea. 
In the treatment of acute gonorrheal complications, however, anti- 
gonococcus serum is sometimes useful and especially in severe fulminating 
epididymitis, prostatitis, and orchitis, salpingitis, gonococcus bacteremia 
(sepsis), and metastatic lesions, especially arthritis. This is particularly 
true when the serum is given early, intravenously or intramuscularly, and 
in amounts of at least 30 to 50 c.c. instead of 2 to 3 c.c. by subcutaneous 
injection. Gonococcus bacteremia is not commonly detected, but when 
the condition is found it may be stated without reserve that the intravenous 
administration of the serum is indicated in order to rid the blood of the 
cocci and probably prevent the involvement of other mucous membranes, 
as the joints and endocardium. Given early in arthritis, upon the first 
evidences of inflammation, it may abort the infection. After the arthritis 
is well developed, however, the serum may bring about only temporary 
relief, and equally good results have been observed following the intravenous 
injection of normal horse-serum and other protein agents, eliciting non- 
specific reactions. That antigonococcus serum is useful in the treatment 
of arthritis is indicated by the reports of Uhle and MacKinney,1 Herbst 
and Belfield,2 Swinburne,3 Schmidt,4 Corbus,5 and others. 
In acute gonococcus infections antigonococcus serum is, therefore, indi- 
cated and probably useful, and particularly when the infection is especially 
acute and wide-spread, with metastases to neighboring or distant tissues. 
If serum is given at all it should be given in adequate doses, as stated above. 
A second injection should follow in about twelve hours and probably a third 
on the second day, until there are evidences of improvement. Debre and 
Paraf6 have recently reported the successful treatment of 15 cases of acute 
gonococcus arthritis by the injection of antigonococcus serum directly into 
the joints at intervals of one to three days, followed by the application of 
compressing dressings. At the same time serum is given intramuscularly 
or intravenously. 
Autoserum therapy has been occasionally practised. Wagon7 reports 
that in 4 cases of orchitis with effusion the removal of the fluid and sub- 
cutaneous injection of 1 c.c. was followed by good results. It is highly 
probable, however, that the tapping alone was the more beneficial of the 
two procedures. 
Dufour and Debray8 have also treated arthritis by the same method, 
aspirating fluid from the larger joints and reinjecting it without heating. 
In 3 cases the general symptoms and local pain rapidly subsided. 
1 Jour. Amer. Med. Assoc., 1908, 41, 105. 
2 Illinois Med. Jour., 1908, 13, 689. 
3 Trans. Amer. Urol. Assoc., 1909, 3, 170. 
4 Therap. Gaz., 1909, 26, 609. 
5 Jour. Amer. Med. Assoc., 1914, 62, 1462. 
6 Bull. d. 1. Soc. m£d. d. hop., 1919, 43, 908. 
7 Bull, de l’Acad. de med., 1916, 76, 81. 
8 Bull. d. 1. Soc. med. d. hop., 1920,44,1399. 980 
VACCINES, SERA, BLOOD, PROTEINS IN TREATMENT OF DISEASE 
Subcutaneous Injections of Gonococcus and Mixed Vaccines in the 
Treatment of Acute Urethritis.—Gonococcus vaccines injected subcutane- 
ously in ordinary doses have been used extensively in the treatment of 
acute gonorrheal urethritis of the male and female with a wide divergence 
of opinion in regard to their value. Some physicians believe that their 
administration is useful for shortening the disease and preventing extensions 
and complications, while others, and probably the majority, believe that 
they are useless, as recently stated by Geraghty.1 
Boyd2 has treated 250 cases with vaccines prepared of gonococci and 
staphylococci, the latter isolated from cases of mixed infection, and con- 
cludes that there is no evidence indicating that this mixed vaccine in ordinary 
or in large doses had any effect upon the course of early acute cases of gonor- 
rhea in men. 
Lumb,3 however, believes that vaccine treatment results in a material 
reduction in the duration of the disease and that relapses occurred in less 
than 1 per cent, of a series of cases. Baril and Creuze4 and Gibson5 have 
also advocated vaccines in acute and chronic gonorrhea for the purpose of 
building up general resistance and preventing chronic complications. 
Stock vaccines are generally employed and in larger doses than usual; 
injections are made subcutaneously every two or three days, beginning 
with at least 100,000,000 cocci and gradually increasing. Slight reactions 
are apparently necessary for therapeutic results 
The cultivation of the gonococcus is not particulatly difficult during 
the acute stages of gonorrhea and the autogenous strain may be secured, 
prepared in a vaccine, and held in reserve for later use if the infection 
becomes chronic with involvement of neighboring organs. 
Subcutaneous Injections of Gonococcus and Mixed Vaccines in the 
Treatment of Chronic Urethroprostatitis, Epididymo-orchitis, and Ar- 
thritis.—These chronic infections so difficult of cure have received special 
attention from the standpoint of vaccine therapy. In the treatment of 
chronic urethritis and the extensions, mixed stock vaccines of gonococci, 
staphylococci, colon bacilli, and diphtheroid bacilli have been generally 
employed. 
Given by subcutaneous injection and in ordinary doses at intervals of 
five to seven days it may be stated that in the opinion of many physicians 
the results have been frequently negative, that is, beneficial effects have not 
been observed in the majority of cases. 
It is apparent, as recently stated by Sezary,6 that the doses must be larger 
than usually given in order to secure a febrile reaction and a focal reaction 
(increased discharge, etc.) before beneficial effects are to be obtained. For 
this purpose the writer prepares an autogenous vaccine of staphylococci and 
other bacteria secured in cultures of secretion after prostatic massage, and 
adds a stock polyvalent vaccine of gonococci. Each cubic centimeter 
contains 2,000,000,000 (1,000,000,000 gonococci and 500,000,000 of each of 
the other organisms); the injections are given intramuscularly beginning 
with 0.2 c.c. (200,000,000) and rapidly increased at intervals of five to seven 
days. Used in conjunction with local measures many cases are apparent- 
ly benefited; at least the focal reactions combined with prostatic massage 
appear to throw off large numbers of organisms with gradual disappearance 
of urethral discharge. 
1 Jour. Amer. Med. Assoc., 1921, 76, 35. 
2 Jour. Royal Army Med. Corp., London, June, 1921. 
3 Brit. Med. Jour., October 6, 1917, 450. 5 Lancet, 1919, 2, 739. 
4 Paris Med., 1919, 9, 202. 6 Progres med., Paris, May 14, 1921. TREATMENT OF GONOCOCCUS INFECTIONS 
981 
In arthritis, vaccines of gonococci administered by subcutaneous injec- 
tion have sometimes proved useful adjuvants in treatment. Heaworth1 has 
reported favorably on the use of sensitized vaccine. It would appear that 
in the past the doses have been too small. Better effects are to be secured 
with doses large enough to produce a general reaction of fever associated 
with some degree of focal reaction in the joints. For these purposes intra- 
muscular injections are better than subcutaneous injections. 
Best results have been secured by intravenous injections, which elicit 
well-marked non-specific general reactions; this treatment will be described 
later, but should be employed very cautiously by reason of the possibility 
of eliciting too severe reactions. 
Pearson2 has recently advocated the subcutaneous injection of gonococcus 
vaccine as a provocative measure to determine cure of urethritis. In none of his 
cases did the vaccine produce a general reaction nor was any discomfort 
caused to the patient and no complications followed its use. On the morning 
of the first day a dose of 3,000,000 of polyvalent gonococcal vaccine is admin- 
istered subcutaneously, and at the same time the seminal vesicles, prostate, 
and Cowper’s glands are massaged thoroughly to liberate any toxins confined 
in them. The patient is then instructed to hold his urine from midnight 
of that night until a smear is taken the next morning, and to do the same for 
each successive night until four smears have been obtained. On the morning 
of the second day a dose of 5,000,000 of the vcacine is administered. Should 
the case still be infected with the gonococus, a positive smear may be ob- 
tained on one of the four mornings. Should there be no infection with the 
gonococcus, the smear will remain negative for that organism. Of 100 
consecutive cases the provocative vaccine gave a reliable result in 96 per 
cent, of cases, but Pearson advises that it should not be relied on solely; 
it should be used in conjunction with other routine tests. Its us& renders 
unnecessary the irritation of the urethra produced by strong provocative in- 
jections of chemicals. 
Subcutaneous Injections of Gonococcus Vaccine in the Treatment of 
Vaginitis and Cervicitis.—A large literature has accumulated on the results 
of the vaccine treatment of vulvovaginitis and cervicitis and especially 
of children. The reports are contradictory, but the results have been 
generally negative. 
Hess has reported that subcutaneous injections of 100,000,000, 200,000,- 
000, or 400,000,000 of killed cocci sometimes elicits a focal reaction of in- 
creased vaginal discharge of pus cells and gonococci in cases of chronic and 
latent infections of children, and that one or two doses have proved useful 
for diagnosis by reason of these provocative reactions. Hess also believes 
that the injections have some prophylactic value when used routinely among 
female children prior to admission in the wards of a hospital where gono- 
coccus vaginitis may make its appearance. 
The disease in both children and adults soon involves the cervix, almost 
invariably becomes chronic, and proves extremely difficult of cure. It would 
appear that the administration of vaccine in amounts large enpugh to elicit 
a focal reaction may be useful as an adjuvant to local treatment. Doses 
less than required to produce hyperemia and increased discharge (focal 
reaction) have been proved to be useless. 
Intravenous Vaccine and Non-specific Agents in Arthritis, Prostatitis, 
and Epididymitis.—Bruck and Sommer3 have reported exceptionally good 
results in the treatment of acute and subacute gonococcus arthritis, prosta- 
1 Brit. Med. Jour., 1918, 1, 4. 
3 Munch, med. Wchn., 1913, 60, 1185. 
2 Jour. Urology, 1918, 2, 455. 982 
VACCINES, SERA, BLOOD, PROTEINS IN TREATMENT OF DISEASE 
titis and epididymitis by the intravenous injection of polyvalent gonococcus 
vaccine. Culver1 has injected intravenously 100,000,000 killed gonococci 
in the treatment of these infections, giving the same or slightly different 
amounts every four or five days, until about five injections had been given. 
“The injection is followed by a chill of variable severity coming on in 
from twenty minutes to one hour and lasting from fifteen to thirty minutes. 
This chill may or may not be accompanied by headache, which is usually of 
short duration and passes away shortly after the chill is over. Exceptionally 
there are nausea and vomiting during the first few hours, but it is never 
severe and is always transient. This occurs always in patients who have 
disobeyed instructions by eating heartily within a few hours of the injection. 
“At the onset and during the chill the patient often complains of severe 
pain in the affected parts. Whether this is due to some internal cause or 
to the unavoidable motion produced by the chill it is impossible to say. 
The pain, however, is variable in different patients and in the same patient 
at different times. 
“It is usual for the severity of the reaction to decrease following repeated 
injections, and this decrease is directly proportional to the number of injec- 
tions previously given in order to maintain a constant reaction with each 
succeeding injection of gonococcal gradually and cautiously increased. 
“The leukocyte counts during a reaction and following it are somewhat 
variable. Usually there is a mild leukocytosis just before and during the 
first part of the chill, soon followed by a marked leukopenia, which appears 
toward the end of the chill. A count as low as 2000 has been observed 
repeatedly during this stage of the reaction. This condition is soon followed 
by a gradually developing leukocytosis which reaches its maximum in from 
five to seven hours, and remains moderately high for from twenty-four to 
thirty hours. A return to normal occurs in about forty-eight hours.” 
Similar reactions and clinical results were observed after the intravenous 
injection of a vaccine of 100,000,000 meningococci as well as after the injec- 
tion of 25,000,000 colon bacilli and 2 c.c. of a 4 per cent, solution of secondary 
proteose made from casein. For these reasons Culver ascribed the effects 
and beneficial results to non-specific reactions embracing the febrile reaction, 
leukocytosis, mobilization of ferments, and to some extent an increase of 
gonococcus antibodies. 
“Twenty-four patients suffering from arthritis associated with gonorrheal 
urethritis were treated. Naturally, most of these cases were acute or sub- 
acute, but some were of five months’ duration and many were of over ten 
weeks’ duration when the treatment was begun. 
“As might be expected, the most striking results were obtained in the 
acute and subacute cases; the most refractory cases, however, were also in 
the acute class. Those suffering for long periods appear to respond more 
slowly to the treatment, but fortunately seem to suffer from no recurrences 
or new joint involvements during the course of the treatment, as some of 
the more acute cases do. All but two of the arthritic patients were com- 
pletely cured or manifested a decided improvement. The length of treat- 
ment varied from two days to one month. 
“Unusual effects were seen in three patients with acute arthritis so severe 
that sedatives were necessary to give them rest for the first two days in the 
hospital. After a single reacting dose in each instance they felt so well 
that they insisted on getting out of bed, and in three days they walked 
out of the hospital. Two of the patients had effusions in the knee-joint, 
which completely disappeared before their discharge from the hospital. 
1 Jour. Amer. Med. Assoc., 1917, 68, 362; Jour. Lab. and Clin. Med., 1917, 3, 11. TREATMENT OF GONOCOCCUS INFECTIONS 
983 
“Twelve patients with acute epididymitis were treated, and invariably 
the pain would subside after the first injection. Usually not more than two 
injections were necessary and, indeed, in most instances one, to effect a 
cure. The swelling begins to subside within twenty-four hours after the 
first injection.” 
In a subsequent review of this work Culver1 states that in his experience 
at one of the military camps during the war where intravenous injections 
of gonococcus protein were made in every case of epididymitis, the results 
were satisfactory, the average stay in the hospital being from five to six 
days. Even the primary localization of the cocci in the urethra or extensions 
of the infection into the prostate and seminal vesicles responded more readily 
to local treatment after these injections than cases not injected. 
Miller and Lusk2 have likewise observed encouraging results in the 
treatment of acute and chronic gonorrheal arthritis from intravenous injec- 
tions of typhoid vaccine (100,000,000); Cecil,3 however, noted but little 
improvement after the intravenous injection of typhoid and gonococcus 
vaccines in arthritis. Luithlen4 has had considerable and favorable expe- 
rience in arthritis treated with intravenous injections of 100,000,000 of 
gonococci; he recommends, though, that local treatment should not be 
neglected. 
Block5 has also employed intravenous injections of gonococcus and 
typhoid vaccine and obtained the best clinical results when well marked 
reactions were produced, particularly a sharp rise in temperature. 
Muller and Weiss6 have reported very favorably upon the therapeutic 
effects of intramuscular injections at intervals of four to five days, of 5 to 
10 c.c. of market milk, boiled for ten minutes and cooled, in the treatment 
of acute and subacute gonococcus arthritis, prostatitis, and orchitis. Kon- 
teschweller7 also reports very satisfactory results in the treatment of gono- 
coccus arthritis from the intramuscular injection of milk and kefir and 
the intravenous injection of peptone. Trossarello,8 however, did not ob- 
serve any beneficial results in arthritis from milk injections. 
Weiss9 has treated a number of cases of epididymitis by injecting 5 to 
10 c.c. of sterile milk in the scrotal skin and claims excellent results. 
Smith10 has observed that patients suffering from gonorrheal complica- 
tions are frequently benefited when decided general reactions follow the 
intravenous injection of normal horse-serum as well as horse antigonococcus 
serum; Brown11 has also reported the beneficial effects following the injection 
of normal horse-serum and diphtheria antitoxin in the treatment of epididy- 
mitis. These results are purely non-specific effects. Muller12 and Saudek13 
have also treated with success gonorrheal complication by the injection 
of 1 to 2 c.c. of sterile normal horse-serum about the site of the lesion. 
Subcutaneous injections of turpentine have been employed by several 
observers and especially Krebs14 in the treatment of both acute and chronic 
gonococcus infections. Krebs has treated several hundred cases with 
injections every three to five days combined with appropriate local treat- 
ment. He claims good results and particularly in the severer infections; 
the acute disease was shortened and complications lessened. 
These results indicate that non-specific therapy has proved useful in the 
1 Jour. Amer. Med. Assoc., 1920, 76, 311. 
2 Jour. Amer. Med. Assoc., 1916, 68, 2010. 
3 Arch. Int. Med., 1917, 20, 951. 
4 Wien. klin. Wchn., 1919, 32, 448. 
6 Cor.-Bl. f. schweiz. Aerzte, 1914, xliv, 1377. 
6 Wien. klin. Wchn., 1916, 29, 249. 
7 Presse med., 1919, 27, Annex, 629. 
8 Riforma Med., 1920, 36, 350. 
9 Wien. klin. Wchn., 1919, 32, 840. 
10 Jour. Amer. Med. Assoc., 1916, 66, 1758. 
11 Glasgow Med. Jour., 1918, xc, 280. 
12 Wien. klin. Wchn., 1917, 30, 805. 
13 Dermat. Wchn., 1918, lxvi, 384. 
14 Miinch. med. Wchn., 1919, lxvi, 1441. 984 
VACCINES, SERA, BLOOD, PROTEINS IN TREATMENT OF DISEASE 
treatment of subacute and chronic gonorrheal complications; much less so, 
however, in acute gonorrhea. The beneficial effects are apparently due to 
a combination of agencies, but it is significant that those who have employed 
this therapy have found the febrile reaction necessary for beneficial effects. 
It is possible that fever alone is destructive for gonococci; it is well known 
that the organisms are very susceptible to temperatures above 98° to 99° F. 
Due care and caution, however, must be exercised in employing these 
remedial measures and especially when sera and vaccines are given intra- 
venously. 
Vaccines and Non-specific Agents in the Treatment of Buboes.—Vac- 
cine treatment has been generally confined to the treatment of chronic 
discharging, suppurative adenitis; autogenous vaccines prepared of organ- 
isms secured in cultures of the pus are to be preferred. If administered 
subcutaneously the injections should be given at intervals of five to seven 
days in doses large enough to secure mild focal reactions. For this purpose 
the vaccine may contain 2,000,000,000 organisms per cubic centimeter and 
treatment begun with 0.1 c.c. and gradually increased. 
Luithlen has reported favorably upon the treatment of old torpid venereal 
ulcers associated with buboes by intramuscular injections of vaccine in doses 
of 300,000,000 to 800,000,000; better results were observed from two or three 
intravenous injections of 50,000,000 at intervals of several days. 
Trossarello has employed intramuscular injections of milk in the treat- 
ment of these cases. 
Vaccines and Non-specific Agents in the Treatment of Salpinigitis.— 
Much less has been reported on the treatment of gonococcus complications 
in women by vaccines and non-specific agents than in the treatment of men. 
Gonorrheal salpinigitis and ovaritis or more general pelvic inflammation 
is usually less acute than gonorrheal metastases in men and medical assist- 
ance is less frequently sought. 
Ordinary gonococcus vaccine administered by subcutaneous injection 
in usual amounts has proved valueless in the treatment of subacute and 
chronic pelvic infections of women. 
Intravenous injections of 50,000,000 to 100,000,000 gonococci, producing 
a general reaction similar to that previously described and usually associated 
with increased pelvic pain and tenderness, has yielded much better results. 
Five intramuscular injections at intervals of three to four days of 5 to 10 c.c. 
of milk boiled for ten minutes and cooled have yielded particularly good 
results in the hands of Trossarello.1 All of a series of cases of tubal and 
ovarian disease were improved, some after a single injection. Best results 
were observed in those patients developing febrile reactions. 
More recently Fuchs,2 Schubert,3 Zoeppritz,4 and Sonnenfeld5 have re- 
ported particularly good results in the treatment of adnexal disease by 
injections of turpentine. The usual dose has been 0.5 c.c. of a mixture of 
1 part turpentine and 4 parts sterile olive oil by intramuscular injection 
(gluteal). A small amount of novocain may be added to prevent local pain. 
Fuchs states that cases with a temperature reaction of 1° to 2° C. gave the 
most striking clinical results and that in over two hundred injections there 
was no abscess formation or untoward effects; even in acute early adnexal 
inflammation accompanied by much pain and hemorrhage the injection of 
one to two doses is usually followed by a reduction of the adnexal swelling 
and a decrease of pain and hemorrhage. 
1 Riforma med., 1920, 36, 350. 
2 Zentralbl. f. Gynak., 1920, xliv, 52. 
3 Zentralbl. f. Gynak., 1919, xliii, 468. 
4 Zentralbl. f. Gynak., 1919, xliii, 297. 
6 Berl. klin. Wchn., 1920, lvii, 707. TREATMENT OF ACUTE AND CHRONIC ARTHRITIS 
985 
TREATMENT OF ACUTE AND CHRONIC ARTHRITIS 
Infection and Immunity in Arthritis.—Acute and chronic arthritis, in- 
cluding rheumatoid arthritis or arthritis deformans, are believed to be 
microbic infections; metabolic changes are considered secondary rather than 
primary etiologic factors. Gout is an exception; the irritants deposited 
in the joints are metabolic products capable of exciting sterile inflammatory 
changes, but even in this disease the probability of bacterial infection cannot 
be excluded in all cases and particularly since the chemical injury may favor 
the localization and activity of bacteria. Closed traumatic arthritis is also 
an exception, but aside from these the majority of arthritides are to be 
regarded as microbe infections, and biologic therapy is of importance in 
relation to both prophylaxis and treatment. 
The anatomic structure of the joints and synovial membranes favors 
bacterial embolism; for this reason it is to be expected that in blood infec- 
tions (bacteremia, septicemia) the circulating microbes may lodge in these 
tissues and produce disease. It is highly probable that gonococcus arthritis 
is of this nature rather than caused by toxic substances in the blood or 
gonotoxins produced in some focus in the urogenital tract. Likewise tuber- 
culosis, typhoid, streptococcic, staphylococcic, pneumococcic, and meningo- 
coccic infections of the joints are to be regarded as embolic infections. To 
the best of my knowledge no one has produced arthritis experimentally 
with injections of bacterial toxins, and a more reasonable assumption is that 
the arthritis is excited by the presence of the actual germs. However, we 
do know that some bacterial toxins show selective affinities for certain tissues, 
as tetanus toxin for the spinal cord, diphtheria toxin for the ganglia of the 
heart, etc., and it is possible, but not probable, that poisons produced by 
known and unknown microbes may have a similar selective affinity for the 
synovial membranes of joints; simply because blood-cultures fail to detect 
a bacteremia is not of much significance because bacteremia is usually 
intermittent. Sterile cultures of joint fluid are of more significance, but do 
not preclude the presence of bacteria in the tissues of the joint. 
Whether or not races of bacteria acquire selective affinities for the tissues 
of the joints is still an open question; undoubtedly the work of Rosenow and 
others with the streptococci are of a convincing nature and especially in 
relation to the subject of focal infection. 
That bacteria and particularly pathogenic cocci may gain access to the 
blood from foci of infection in the tonsils, accessory nasal sinuses, gums and 
apices of the teeth, chronic bronchitis, intestines and genito-urinary tracts 
of both sexes with secondary localization in the joints and elsewhere, may be 
regarded as definitely proved. The sudden exacerbation of an arthritis 
following the extraction of teeth or the administration of an autogenous 
vaccine prepared of the bacteria from apical abscesses, indicate a very direct 
relationship aside from the abundance of clinical and experimental data in 
support of this contention. 
Tuberculosis, typhoid, gonococcic, staphylococcic, pneumococcic, and 
meningococcic infections of the joints are distinct clinical entities and 
almost invariably secondary to some well-known primary infection. Chronic 
rheumatism, rheumatic arthritis, or arthritis deformans is a more or less 
well-defined clinical entity, but the exact nature of the infecting organism is 
unknown except that it is almost certainly a coccus. Whether it is an 
ordinary streptococcus or a special streptococcus rheumaticus or the diplo- 
coccus rheumaticus of Poynton and Paine cannot be stated. Likewise the 
exact etiology of acute rheumatic arthritis is unknown, although the majority 986 
VACCINES, SERA, BLOOD, PROTEINS IN TREATMENT OF DISEASE 
of investigators have found cocci and the diplococcus rheumalicus of Poynton 
and Paine has received most confirmation. It is highly probable that both 
the acute and chronic forms of arthritis are due to the same coccus or cocci, 
belonging to the group of streptococci and pneumococci. 
Acquired immunity in arthritis infections never reaches a high degree. 
One attack of rheumatic arthritis never confers a lasting immunity against 
subsequent attacks; indeed, a hypersusceptibility to recurrent attacks is 
usually apparent probably due more to the results of injury to the joints than 
a decrease of natural defenses. Age, sex, seasons, exposure, and trauma are 
modifying factors, but with the exception of trauma, exert their effects more 
by exercising a modifying influence upon the primary foci producing the 
organisms than upon the joints. Immunity to the pathogenic cocci is never 
developed to a high degree; this is particularly true of gonococci and strepto- 
cocci. One attack of gonorrhea and even of gonococcus arthritis does not 
confer any immunity against subsequent attacks; indeed, the involved joints 
are more susceptible than ever even though apparently healed. The same 
is true of the streptococci which probably include the exciting cause of acute 
rheumatic fever and chronic rheumatic arthritis. Doubtless some degree of 
local and general immunity are acquired, and particularly the development 
of opsonins, but these defenses are apparently feeble against fresh accessions 
of the organisms from the feeding places of primary focal infection. For 
this reason detection and removal of the primary focus is the first principle 
of treatment. In acute rheumatic fever, however, a primary focus of 
infection may not exist; indeed, it would appear that the germ may enter 
through the upper respiratory tract and particularly the tonsils, and induce 
the disease as a specific infection without producing a way station of infection. 
Resistance and recovery from arthritis is probably largely brought about 
by the destruction of the microbes by leukocytes aided by high temperature 
and other non-specific agencies. 
Serum Treatment of Arthritis.—The serum treatment of gonococcus 
arthritis has been previously described. In the acute embolic arthritis due 
to streptococci, meningococci, and pneumococci developing in the course 
of general infections by these organisms, serum treatment if demanded 
consists of the intravenous injection of large doses of the corresponding 
immune serum. Moore1 has found that the intravenous injection of poly- 
valent antistreptococcus serum in rabbits prevents the development of 
arthritis and other lesions following the injection of streptococci capable of 
producing arthritis in untreated controls. Nicoll2 found that antistrepto- 
coccus serum was of little or no value in chronic arthritis; however, some 
benefit from non-specific reactions would be expected. 
Some cases of gonococcus arthritis have been treated with injections of 
serum directly into the involved joints; I do not know of cases of other 
infections treated in this manner, although the local treatment of arthritis 
by the injection of chemotherapeutic agents and sera directly into the joints 
will probably receive an increasing amount of attention in the future. 
Further investigations in the etiology of acute rheumatic fever and the 
preparation of an immune serum for treatment are urgently required; the 
disease would appear to be one demanding a therapy of this kind in view of 
its sudden onset, severe symptoms, and progressive involvement of different 
joints. 
Treatment of acute and chronic arthritis by normal serum will be con- 
sidered under Non-specific Therapy. 
1 Jour. Infect. Dis., 1914, 15, 215. 
2 Jour. Amer. Med. Assoc., 1914, 63, 2225. TREATMENT OF ACUTE AND CHRONIC ARTHRITIS 
987 
Cases of acute and chronic arthritis with effusion have also been treated 
—and apparently with some degree of success—by aspirating the joint and 
immediately injecting 1 to 5 c.c. of the fluid subcutaneously or intramuscu- 
larly (autoserum therapy of arthritis). When the fluid is clear, heating before 
injection is not required; if, however, there is reason to believe that bacteria 
are present, the fluid should first be heated at 60° C. for one hour. The 
fluid should be aspirated into sterile citrate solution (1 c.c. of 2 per cent, 
for 10 c.c. of fluid) to prevent coagulation. The effects are probably non- 
specific and probably could be improved by injecting the fluid intravenously. 
Vaccine Treatment of Arthritis by Subcutaneous Injection.—The vac- 
cine treatment of gonococcus and typhoid arthritis has been previously 
described. 
Lacking the specific organism of acute rheumatic fever, vaccines adminis- 
tered by subcutaneous injection have not been commonly employed; this 
disease, however, has been successfully treated by intravenous injections 
of typhoid vaccine and other non-specific agents described below. 
Chronic rheumatic arthritis and arthritis deformans have been extensively 
treated with vaccines prepared by streptococci, pneumococci, staphylococci, 
and other bacteria secured in cultures of the tonsils, apices of teeth, accessory 
sinuses, and other places of suspected or known foci of infection. Vaccines 
have also been prepared of streptococci secured in cultures by Rosenow’s 
methods from excised lymph glands draining the involved joints, from con- 
tracted and diseased adjacent muscles, and from excised portions of the 
diseased capsule of the joint itself. These vaccines have been injected 
subcutaneously at intervals of five to seven days in amounts varying from 
10,000,000 to 2,000,000,000 of cocci. 
The results have not been encouraging. The most to be expected in the 
treatment of chronic arthritis would be the gradual relief of subjective 
symptoms and particularly pain and a decrease in the number and severity 
of acute exacerbations of the disease. Restoration of function and dis- 
appearance of deformities cannot be expected except to a slight degree and 
is entirely dependent upon the amount of absorption of new fibrous and 
calcareous tissue that may take place. In some instances these objects are 
apparently attained, but in well controlled studies it is evident that the same 
beneficial effects may be secured by the usual hygienic treatment. 
To secure any benefit at all it is necessary to inject amounts of vaccine 
capable of eliciting a mild febrile reaction and some focal reactions in the 
involved joints. For this purpose 500,000,000 to 2,000,000,000 cocci per 
cubic centimeter may be required, depending upon the activity of the disease, 
and preferably by intramuscular injection. Not infrequently these injec- 
tions are followed by considerable symptomatic relief and general improve- 
ment of the patient for temporary periods. 
From a theoretic standpoint continuous vaccine treatment is indicated 
and especially if focal reactions occur after small doses, indicating that the 
vaccine very probably contains the causative organism. Antibody pro- 
duction is slow, as previously discussed, but apt to be continuous, with the 
probability of reaching a point when some therapeutic usefulness may become 
apparent even if nothing more than to delay or prevent the progression of 
the disease with the involvement of new joints. 
Likewise when arthritis is early and involving a few of the smaller joints, 
but develops acute exacerbation following the extraction of teeth for apical 
abscesses, I believe it is good practice to secure the streptococci from such 
foci and institute a course of vaccine treatment for its possible prophylactic 
value against progression of the arthritic infection. For this purpose I 988 
VACCINES, SERA, BLOOD, PROTEINS IN TREATMENT OF DISEASE 
prepare an autogenous vaccine containing 3,000,000,000 cocci per cubic 
centimeter and start treatment with the subcutaneous injection of 0.1 c.c. 
Subsequent doses are given at intervals of five to seven days in gradually 
increasing amounts until as much as 1 c.c. is being given at one time; a 
series of twelve to fifteen doses are given and the balance of the vaccine 
carefully preserved in a cold place for future use if necessary. 
Non-specific Treatment of Arthritis with Special Reference to the 
Intravenous Injection of Typhoid Vaccine.—Miller and Lusk1 were the first 
to report on 24 cases of arthritis treated with intravenous injections of 
proteoses and typhoid vaccine. A second series of 85 cases were given 
intravenous injections of 40,000,000 to 75,000,000 typhoid vaccine. Forty- 
five of these were cases of acute arthritis, and 33 of the patients had already 
been treated with salicylates, without benefit, except in 4 cases. Twenty- 
nine out of the 45, after receiving one to four doses of vaccine, recovered 
in one to five days; 8 showed marked improvement. The 4 gonococcus 
cases in this series did not respond well to the vaccine. There were 9 re- 
lapses in the acute series. Of 12 cases of subacute arthritis, 10 cleared up 
in three to five days, and the other 2 patients improved. Nineteen chronic, 
but still active cases were treated, in 10 of which the patients showed im- 
provement. Culver2 reports a series of 31 gonococcus arthritis cases, in 
28 of which the patients either recovered or greatly improved after intra- 
venous injection of gonococcus, meningococcus, or colon vaccine. Manier, 
Petersen, and Jobling3 have treated 13 patients with arthritis by intravenous 
injections of secondary proteoses. Three of the cases were acute, 3 sub- 
acute, and 7 chronic. Of the acute cases, 2 cleared up promptly after the 
injections, while in the third, a gonococcus arthritis, the patient was not 
relieved. In the 3 subacute cases the patients all recovered rapidly after 
treatment. In the chronic series there was complete relief in 3, marked 
improvement in 3, and no change in 1. Cecil4 has treated 40 cases of 
arthritis, of which 26 were of the ordinary rheumatic type, 7 acute toxic 
arthritis, and 7 of gonorrheal origin. Of the rheumatic and toxic arthritides, 
40 per cent, recovered in from two to ten days without the use of salicylates. 
Of 20 patients receiving salicylates in addition to the vaccine, 17 were cured 
or greatly benefited. The 7 patients with gonococcus arthritis made very 
slow improvement.. Cecil noted that while the arthritic pains were generally 
relieved, muscular pains persisted in many cases. 
Snyder,5 Thomas,6 Scully,7 Pemberton,8 Cross,9 Cadbury,10 Cowie and 
Calhoun,11 Boyd,12 Gow,13 and others have employed intravenous injections 
of typhoid and other vaccines with favorable results especially in the acute 
and subacute cases; Hardinger14 reports that his results were not satisfactory. 
In Europe intramuscular injections of milk and casein and intravenous 
injections of casein have been more commonly employed, and a few investi- 
1 Jour. Amer. Med. Assoc., 1916, 66, 1756; ibid., 1916, 67, 2010. 
2 Arch. Int. Med., 1917, 19, 1042. 
3 Jour. Amer. Med. Assoc., 1916, 66, 1753; Arch. Int. Med., 1917, 19, 1042. 
4 Jour. Amer. Med. Assoc., 1917, 69, 20. 
6 Arch. Int. Med., 1918, 22, 224. 
6 Jour. Amer. Med. Assoc., 1917, 69, 770. 
7 Jour. Amer. Med. Assoc., 1917, 69, 20. 
8 Arch. Int. Med., 1920, 25, 351. 
9 Jour.-Lancet, 1917, 37, 764. 
19 China Med. Jour., 1919, 33, 213. 
11 Arch. Int. Med., 1919, 23, 69. 
12 Jour. Lab. and Clin. Med., 1919, 5, 88. 
13 Brit. Med. Jour., 1920, 1, 284; Quart. Jour. Med., 1919, 13, 82. 
14 Colorado State Jour. Med., 1921, 19, 26. TREATMENT OF ACUTE AND CHRONIC ARTHRITIS 
989 
gators have used intravenous injections of proteoses and peptone; extracts 
of cartilage, the “sanarthrit” of Heilner,1 have been employed abroad and 
especially in the treatment of gout by Umber,2 Sterna,3 and others. 
Dosage and Administration.—Ordinary typhoid vaccine has been most 
commonly employed for eliciting the non-specific reaction in arthritis. The 
amount injected intravenously has varied from 10,000,000 to 100,000,000; 
larger amounts, up to 500,000,000, have been injected in a few instances 
without greatly increasing the reaction and with no ill effects, but the 
smaller amounts are sufficient and safer. 
For beneficial effects a sharp reaction must be produced, and the degree 
of reaction varies in different individuals receiving the same substance in 
the same amount. 
Intravenous injections have yielded the best results. As previously 
stated, typhoid vaccine has been mostly employed, but colon vaccine and 
albumose may be employed. For the average adult case the amounts 
injected for the first dose may be as follows; the subsequent injections should 
be larger or smaller depending upon the degree of reaction and results: 
(a) Ordinary typhoid vaccine in dose of 50,000,000. The typhoid-para- 
typhoid vaccine used for prophylactic immunization is satisfactory and the 
stock vaccine containing a total of 500,000,000 bacilli per cubic centimeter 
may be employed by taking 1 part and diluting with 9 parts of sterile 
saline solution and injecting 1 c.c. intravenously. If the vaccine contains 
1,000,000,000 bacilli per cubic centimeter, 1 part should be diluted with 19 
parts of saline and 1 c.c. injected to give a dose of 50,000,000. 
(.b) Colon bacillus vaccine in dose of 25,000,000 intravenously. 
(c) A 5 per cent, solution of Merck’s albumose in sterile saline solution 
sterilized by boiling for ten minutes; dose 1 c.c. diluted with 4 c.c. of sterile 
saline intravenously. 
(d) Ordinary market milk boiled for ten minutes, cooled, and injected 
intramuscularly in dose of 10 c.c. Usually one or two intramuscular injec- 
tions are sufficient; if beneficial effects are not observed after the first, pro- 
viding a reaction has been produced, it is entirely likely that further injec- 
tions will not bring about an improvement. 
Injections should not be given until the effects of the previous one has 
subsided. The interval is generally five or more days. The average case 
requires one to three injections, but as many as ten have been given. 
The• Reactions— These have been described in detail on p. 881. As 
previously stated, a general and focal reaction must be produced in order 
to elicit a beneficial effect upon the joints. Delirium may occur among 
alcoholics and marked dyspnea has developed in a few cases. There can 
be no doubt that the reaction is unpleasant and may be dangerous when 
the injections are given to improperly selected patients. The chill some- 
times is unusually severe and results in rendering the joints more painful 
than ever by reason of the involuntary movements of muscles and joints. 
As a general rule, however, the after-effects and particularly relief from pain 
are so marked that many patients request subsequent injections, and if 
alcoholics and patients with acute endocarditis and advanced myocardial 
degeneration and hypertension are excluded, there is very little risk. 
Contraindications.—Since this treatment may excite delirium tremens, 
alcoholics should be selected and treated with extra caution or excluded 
altogether; cases of arthritis with acute endocarditis and chronic cases with 
1 Munch, med. Wchn., 1916, 63, 997; ibid., 1918, 65, 983. 
2 Miinch. med. Wchn., 1918, 65, 988. 
3 Miinch. med. Wchn., 1920, 67, 632. 990 
VACCINES, SERA, BLOOD, PROTEINS IN TREATMENT OF DISEASE 
well-marked myocardial degeneration are also to be considered as risks for 
intravenous medication. Intramuscular injections of milk are less dan- 
gerous and may be employed in these cases. Young vigorous individuals 
are best adapted for intravenous therapy. 
Results and Practical Value.—Non-specific therapy probably has no 
influence upon the primary foci of infection, and a diligent search should be 
made for these and adequate treatment applied. It is probable, however, 
that the leukocytosis, fever, ferment, and antibody increase elicited by the 
sharp protein reaction may have a direct destructive effect upon the microbes 
in the joints, so that further progression of the arthritis may be prevented, 
but direct proof of these facts is lacking. “Whether as a result of the 
reaction the local tissues become immune to the toxic effect of bacteria 
still alive in the focus, whether it means merely an increased tolerance to 
toxic split products set free at a distance and to which the local tissues had 
heretofore been sensitive, or whether we deal with the actual destruction 
of bacteria which had become localized in the joint is not determined” 
(Petersen). 
Cowie1 has recently summarized the results very well as follows: 
“It cannot be questioned that very remarkable beneficial effects have 
followed the intravenous injection of foreign protein in the arthritides. 
There is seldom a case of acute arthritis or periarthritis which does not 
respond to a certain degree, but this improvement is often of only a few 
hours’ duration. Taking cases at random, including all varieties, there 
will be found a large percentage in which no permanent beneficial effect 
can be secured by this method of treatment. One should not expect perma- 
nent beneficial effects in cases of periarthritis of a less chronic nature, unless 
great care has been taken to rid the body of all foci of infection that can 
possibly be found. Accordingly, therapeutic intravenous injections of 
foreign protein should follow failure to secure successful results by the 
removal of such foci of infection, or they should be used in conjunction with 
an attempt to remove the focus. 
“Unquestioned relief from pain will often follow protein therapy, even 
though the focus of infection is not removed, and, in some cases, in addition 
to improvement in the joint condition, the focus itself may cease to be 
active.” 
“Thus far, experience has taught us that acute or subacute processes 
that have not progressed beyond the first year are the ones that give the 
best results, particularly those that have not gone on to marked structural 
change of the articular or periarticular tissue; next to these, cases which 
have progressed longer and which may show structural change, but which 
have not produced definite ankylosis and its consequential results.” 
I believe that the work that has been done justifies the statements that 
acute and subacute arthritis and periarthritis are the forms which respond 
more promptly and more surely to this method of treatment. A few cases 
of chronic arthritis of as long as three years’ standing have been recorded 
in which apparently complete cure has followed this method of treatment; 
and in still more chronic forms, unquestioned benefit has occasionally 
resulted. However, at present there are not enough properly classified 
cases recorded to enable us to say what percentage of each class is benefited 
by this method of treatment. 
Petersen states that “in perhaps 40 per cent, of the cases one or two 
injections completely terminate the disease, in another 30 per cent, the 
improvement is marked and recovery made complete on further injections, 
1 Jour. Amer. Med. Assoc., 1921, 76, 310. TREATMENT OF DISEASES OF THE SKIN 
991 
while in the balance there may be either a transient improvement with a 
relapse later, or no marked clinical improvement.” 
Acute and subacute cases of non-gonorrheal origin have yielded the best 
results. The most to be hoped for in chronic arthritis is relief of the joint 
pains—muscular pains may not be relieved—and a cessation or retardation 
of further progress of the disease. Restoration of function is not to be 
expected unless gradually brought about by passive movements. The 
absorption of inflammatory deposits and release of muscular fibrosis and 
contraction does not take place, although miracles of this sort are commonly 
expected to occur when this and other forms of biologic therapy are em- 
ployed. 
Non-specific therapy has no curative effect upon cardiac disease; both 
Torrey and Petersen believe, however, that under this treatment serious 
cardiac involvement is less likely to occur. 
TREATMENT OF DISEASES OF THE SKIN 
The Skin in Relation to Infection, Immunity, and Biologic Therapy.— 
In a broad and general manner diseases of the skin may be divisible into 
those produced by local microparasites and those developing during the 
course of general infections and pathologic alterations of internal organs. 
As discussed in Chapter IX, the skin is not only an important barrier to 
infection but is probably concerned in an intimate manner in the mechanism 
of recovery from infectious and non-infectious diseases, and by reason of 
these relationships is subjected to many different kinds of injury. 
The treatment of skin diseases, therefore, is a broad and general field. 
As Ravaut is quoted: “To treat skin disease wholly from without is as 
irrational as treating the skin lesions of syphilis by local applications alone.” 
Furthermore, it would appear that the skin is an organ capable of pro- 
ducing ferments,' antibodies, and possibly internal secretions, profoundly 
influencing the metabolism and diseases of the internal organs. By reason 
of these activities it is very probably concerned in the processes of recovery 
from many diseases not directly involving the skin; furthermore, it would 
appear that these remedial agencies may be brought into play by various 
therapeutic agencies. For example, the curative effects of light rays and 
sunlight upon different diseases of the skin and of the internal organs as 
well, notably tuberculosis, may be due to mild non-specific shock instead 
of the production of some hypothetic internal secretion stimulating the 
organism. As stated by Petersen1: “The epithelial tissues become hy- 
peremic and absorption from them is accelerated. Skin enzymes-—protease 
and lipase, perhaps some ereptase in younger individuals—are swept into 
the circulation, together with some protein-split products due to digestive 
stimulation in the skin. The agents that ordinarily provoke the non- 
specific reaction are, therefore, available; the enzymes present in the serum 
can now attack seminecrotic or necrotic foci and there set up a focal reaction 
with its resulting train of increased temperature, malaise, etc.” Hoffmann2 
has also called attention to the fact that other therapeutic measures, as 
counterirritation, blistering, etc., may involve precisely the same mechanism. 
The work of Muller3 also affords a striking example of the importance of 
skin stimulation on remote disease processes. He has shown that the 
amount of antigen required to produce a focal reaction about a disease proc- 
ess remotely situated, i. e., gonorrheal urethritis, may be one-thirtieth less 
1 Protein Therapy and Non-specific Resistance, MacMillan Co., 1922, 138. 
2 Strahlentherapie, 1919, 7, 1. 
3 Munch, med. Wchn., 1920, lxvii, 9; Med. Klinik, 1920, 26, 579. 992 
VACCINES, SERA, BLOOD, PROTEINS IN TREATMENT OF DISEASE 
when injected intracutaneously than when injected intramuscularly or intra- 
venously. These observations indicate that slight skin reactions may exert 
an important effect on remote pathologic lesions irrespective of the sub- 
stance injected and to be ascribed to changes and processes produced in the 
skin. On the basis of these observations Ponndorf1 and Kroschinski2 have 
employed intracutaneous injections of tuberculin not only for the treatment 
of tuberculosis but in neuralgia, neuritis, acne, furunculosis, and other 
diseases. 
Treatment of Acne.—Acne vulgaris was among the first diseases in which 
vaccine therapy has been applied. The cause of the disease is not definitely 
known and may have some relationship to overstimulation of the sebaceous 
glands by disturbances of glands of internal secretion and especially those 
concerned with sexual changes at puberty and early adult life. Schamberg 
also regards intestinal auto-intoxication (predisposed to by constipation) 
as a factor of importance; in many cases the foodstuffs favoring the de- 
velopment of toxic substances are the starches and sugars. Of interest 
in this connection are the results of complement-fixation tests with the 
sera of individuals suffering with acne conducted by Strickler, Schamberg, 
and the writer3; a large percentage were found to yield positive reactions 
with antigens of colon bacilli, indicating that this group of organisms may 
bear a relationship to the disease in at least some cases. 
It would appear that while metabolic changes of this kind are of etio- 
logic significance, doubtless bacterial activity in the skin is also of impor- 
tance. The acne bacillus described by Unna, Sabouraud, and Gilchrist is 
especially found in myriads, in comedones, and in sebaceous glands. Staph- 
ylococci are invariably found in the pustular lesions. It is held by some 
that in acne there is a follicular hyperkeratosis which obstructs the hair 
follicles and sebaceous glands and results in retention of secretion, inflam- 
mation, and suppuration. 
It is apparent, therefore, that vaccine therapy is only a part of the treat- 
ment of acne. The results of vaccine treatment are sometimes brilliant, 
at other times very disappointing, but probably in the majority of cases of 
distinct aid. 
Engman4 has observed particularly good results in the treatment of acne 
indurata and certain forms of cystic acne, with stock vaccines of the acne 
bacillus. Less favorable results were observed in the more superficial types, 
known as acne simplex and acne pustulosa. Engman has given the following 
directions as to dosage: “The initial dose should be from 3,000,000 to 
5,000,000, to be repeated in from five to seven days. The interval should 
be gaged according to the reaction to the dose, which is usually exhibited 
by the appearance of a few new lesions within forty-eight hours after the 
injection. On the third day after the injection comedones may be expressed 
and the lesions opened if necessary. The manipulation of the lesions at 
this time helps bring the immunizing blood to the part and it is at the height 
of the “tidal wave of immunity.” Local hyperemia in the form of hot 
towels or that resultant from manipulation should be used on the third day 
after each injection and not before then. 
If, after a few such doses, new lesions continue to appear after the third 
day, a larger dose of from 7,000,000 to 10,000,000 should be given and 
continued until a proper therapeutic result is obtained, when the interval of 
1 Munch, med. Wchn., 1914, lxi, Nos. 14 and 15. 
2 Miinch. med. Wchn., 1921, lxviii, 205. 
3 Jour. Cutan. Dis., March, 1916. 
4 Jour. Amer. Med. Assoc., 1921, 76, 176. TREATMENT OF DISEASES OF THE SKIN 
993 
administration should be lengthened to two weeks, then three weeks, and 
finally four weeks, thus continuing the remedy in a prophylactic manner. 
If there is an outcrop of many new lesions within forty-eight hours after 
injection the dose should be lessened and the interval of dosage extended. 
However, if many new lesions appear after the third day it is an indication 
for a much larger dose. In no instance should more than 15,000,000 be 
administered in any injection.” 
Engman states that staphylococcus vaccine or a mixed vaccine of staphyl- 
ococci and the acne bacillus are of little value, except in some cases of acne 
varioliformis, which occurs later in life. 
In my experience best results have been observed when a mixed vaccine 
of the acne bacillus, staphylococci, and the colon bacillus has been employed, 
and especially in the treatment of the pustular infections. Since the acne 
bacillus is a difficult organism to cultivate a stock suspension is employed; 
the staphylococci, however, are secured in cultures of the lesions and the 
colon bacillus secured by plate cultures of the feces. The mixed vaccine 
is so prepared that each cubic centimeter contains 200,000,000 acne bacillus 
and 400,000,000 each of the colon bacillus and staphylococci. The first 
dose is 0.1 c.c.; subsequent injections are given at intervals of five to seven 
days in gradually increasing amounts according to the plan of treatment 
employed by Engman for the administration of vaccines of the acne bacillus. 
Treatment with non-specific agents has been employed by numerous 
investigators. Intravenous injections of 50,000,000 typhoid bacillus vaccine 
has brought about improvement and may be considered in the treatment of 
severe cases. Turpentine injections have been employed by a number of 
physicians and especially Klingmueller,1 who injects about 4 drops of a 
20 per cent, solution in sterile olive oil subcutaneously every three or four 
days over a long period of time. Intramuscular injections of 5 to 10 c.c. of 
milk, boiled and cooled, have been employed and apparently with success 
in some cases. 
Treatment of other Pyogenic Dermatoses; Sycosis Vulgaria; Impetigo 
Contagiosa; Ecthyma (Pyodermia).—Autogenous vaccines are sometimes of 
value in the treatment of these pustular infections. As a general rule 
cultures reveal the presence of staphylococci. Vaccines of staphylococcus 
aureus, frequently found in cultures of the deeper lesions, are particularly 
apt to prove helpful in some cases, as reported by Dennie and Bufford,2 
McLeod,3 and others. 
The vaccine may be so prepared that each cubic centimeter contains 
1,000,000,000 cocci. The first dose should be 0.1 c.c. and subsequent 
injections given at intervals of five to seven days in gradually increasing 
amounts. 
Subcutaneous injections of 20 per cent, dilutions of turpentine and 
intramuscular injections of milk, have likewise been employed with some 
success; the doses are the same as those given above for the treatment of 
acne. 
Treatment of Eczema.—True eczema is very probably the cutaneous 
manifestations of general, functional, and metabolic disturbances in which 
bacterial infection may be responsible for secondary changes. 
The relation of food allergy to eczema, its diagnosis by skin tests, and 
treatment, have been described in previous chapters. 
The term “eczema” is a very broad and general one, including a variety 
1 Berl. klin. Wchn., 1918, 45, 896; Miinch. med. Wchn., 1918, 55, 896. 
2 Boston Med. and Surg. Jour., 1916, 173, 905. 
3 Practitioner, London, 1920, 104, 338. 994 
VACCINES, SERA, BLOOD, PROTEINS IN TREATMENT OF DISEASE 
of dermatoses, produced by a variety of widely different causes. No other 
skin affection requires more thorough-study and search for etiologic factors 
than this group. 
In eczema papulosum, vesiculosum and pustulosum, and in infectious 
eczematoid dermatitis of Fordyce originating around suppurating wounds 
autogenous vaccines may be of some aid in treatment as reported by Medalia1 
and others. As a general rule staphylococci and various diplococci are to be 
found; the vaccine may contain 1,000,000,000 per cubic centimeter. Treat- 
ment is begun with 0.1 c.c. and the injections given every five to seven days 
in gradually increasing amounts. 
Subcutaneous injections of 20 per cent, turpentine in sterile olive oil in 
dose of 2 to 4 drops every three or four days have been employed by Kling- 
mueller,2 Singermann,3 Becker,4 and others. Intramuscular injections of 
milk boiled for ten minutes and cooled have also been employed with success; 
the dose for adults is 5 to 10 c.c. and for children correspondingly smaller 
amounts. Typhoid vaccine in dose of 50,000,000 has been injected intra- 
venously by a few observers. 
Spiethoff,5 who has been one of the earliest advocates of the autoserum 
treatment of certain skin diseases, has reported favorable results from the 
autoserum treatment of eczema. He draws from 50 to 100 c.c. of blood from 
adults, separates the serum and reinjects the serum intravenously on alter- 
nate days at first and then twice a week. Heuck,6 Ullman,7 and others, 
however, have not noted any beneficial effects of autoserum injections in 
eczema. 
Danysz's8 enlerovaccine has been employed in the treatment of various 
non-infectious diseases of the skin, including eczema, with apparent success. 
He has worked on the theory that these diseases may be anaphylactic 
phenomena and that albuminoid material or microbian constituents in the 
intestinal canal may pass into the blood and act as antigens with the pro- 
duction of anaphylaxis. He now employs a stock antigen prepared by 
cultivating a scrap of feces in ordinary bouillon and then securing pure 
cultures, which are mixed in the same proportions as found originally. This 
vaccine is diluted with saline, sterilized with heat and the dose determined 
by weight. For oral administration, the dose is 1/10 to 5/10 mg.; for 
subcutaneous injection 1/1000 to 1/1200 mg. 
Treatment of Psoriasis.—The etiology of psoriasis is still unknown. 
Vaccines prepared of various staphylococci, diplococci, diphtheroid bacilli, 
etc., secured in cultures of psoriatic lesions, have been employed in treatment 
by a large number of physicians. In not a few instances cures have been 
claimed, but the disease is so subject to periods of spontaneous improvement 
that the evidence in favor of the efficacy of vaccines is not convincing. In 
the writer’s experience autogenous vaccines of the cutaneous organisms have 
proved of no value. 
Various non-specific biologic agents, however, have been employed 
with some evidence of success, and particularly injections of the patient’s 
own serum (autoserum therapy), while chrysarobin ointment (2 to 10 per 
cent.) is being applied locally. 
1 Boston Med. and Surg. Jour., 1915, 173, 187. 
2 Boston Med. and Surg. Jour., 1915, 173, 187. 
* Derm. Zentralb., 1920, 23, 130. 
4 Derm. Wchn., 1920, lxxi, 472, 481. 
B Med. Klinik, 1914, 11, 29; ibid., 1916, 12, 1223. 
6 Miinch. med. Wchn., 1912, lix, 2608. 
7 Arch. f. Dermat. n. Syph., 1913-14, cxviii, 125. 
8 Presse M6d., 1918, 26, 367; Bull. Med., 1920, 34, 155, 657. TREATMENT OF DISEASES OF THE SKIN 
995 
Autoserum treatment was first employed by Spiethoff1 for the treatment 
of psoriasis and apparently with considerable success. Gottheil and Saten- 
stein,2 in this country, have likewise reported favorably. Fox3 found auto- 
serum injections in general very satisfatory and in some cases the results 
were decidedly brilliant, but only when used in conjunction with chrysarobin. 
Similar results were observed by Hilario,4 Ravitch,6 and Trimble and Roth- 
well6; Willock7 has not observed any beneficial effects from autoserum 
injections in psoriasis. Schamberg and myself have observed apparently 
beneficial results in some cases and especially those with acute exacerbations 
unable to bear with chrysarobin ointment. Not infrequently the injections 
resulted in rendering the lesions quiescent and when employed with applica- 
tions of neorobin ointment yielded beneficial results. 
The technic of these injections has been described in Chapter XXXVII. 
As a general rule 50 c.c. of blood is drawn from a vein at the elbow under 
aseptic precautions, the serum separated (usually 15 to 20 c.c. obtained) 
and injected intravenously. Several injections are given at intervals of 
about seven to ten days. 
Linser8 and Perry9 have employed six to nine intravenous injections of 
3 to 5 c.c. of horse-serum. Engman and McGarry10 and Scully11 have employed 
intravenous injections of typhoid vaccine in dose of 75,000,000 to 100,000,000. 
The latter reports that these injections do not clear up the lesions of psoriasis, 
but lessen the induration and inflammatory reaction. When employed with 
the local application of weak chrysarobin ointment the lesions frequently 
cleared away with rapidity. Cadbury12 observed temporary improvement 
in 5 cases, all relapsing sooner or later. Turpentine and milk injections 
have also been employed in doses similar to those given above in the treat- 
ment of acne. 
Sabouraud13 considers that non-specific therapy has introduced a new era 
in the treatment of psoriasis and finds that improvement occurs in most 
cases and lasts longer under treatment with Danysz’s enterovaccine than any 
other measures yet known. This vaccine has been described under Eczema. 
Treatment of Ringworm (Trichophytosis).—Vaccines of the small- 
spored fungus Microsporon audouini and of the large-spored fungus tri- 
chophyton, have been reported by Strickler,14 Engman and McGarry,16 and 
others as yielding good results in the treatment of ringworm of the body 
(tinea circinata) and of the scalp (tinea tonsurans). The deep-seated 
lesions have improved under this treatment more rapidly than the super- 
ficial ones. Autolytic products of the fungi designated as trichophyton 
have been employed as vaccines by various European observers. 
Strickler has prepared the vaccines by cultivating the fungi on Sabou- 
raud’s solid “French proof agar,” removing the growth, grinding with 
1 Munch, med. Wchn., 1913, lx, 521. 
2 Med. Record, 1914, lxxxv, 620; Jour. Amer. Med. Assoc., 1914, 63, 1190. 
3 Jour. Amer. Med. Assoc., 1914, 63, 2190; Jour. Cutan. Dis., 1915, 33, 616. 
4 Jour. Cutan. Dis., 1914, 32, 780. 
5 Jour. Amer. Med. Assoc., 1915, 64, 1228. 
6 Jour. Cutan. Dis., 1915, 33, 621. 
7 Jour. Amer. Med. Assoc., 1916, 65, 14. 
8 Verhandl. d. deutsch. Kong. f. inn. Med., 1911, 28, 125. 
9 Boston Med. and Surg. Jour., 1916, 174, 274. 
10 Jour. Amer. Med. Assoc., 1916, 67, 1741. 
11 Jour. Amer. Med. Assoc., 1917, 69, 1684. 
12 China Med. Jour., 1919, 33, 213. 
13 Presse med., 1920, 28, 53. 
14 Jour. Amer. Med. Asscc., 1915, 65, 224. 
15 Jour. Amer. Med. Assoc., 1917, 68, 543. 996 
VACCINES, SERA, BLOOD, PROTEINS IN TREATMENT OF DISEASE 
weighed amounts of crystals of sodium chlorid, and diluting with sufficient 
sterile water to give a fairly turbid emulsion in isotonic saline solution. 
This suspension was heated to 60° C. for one hour, cultured for sterility, and 
preserved with 0.25 per cent, phenol. The dose was 1 to 4 c.c. by sub- 
cutaneous injection every three to six days. 
Lavinder1 prepares the vaccine by cultivating the fungi in flasks of about 
100 to 150 c.c. of Sabouraud’s liquid medium. The growth is removed by 
filtration through sterile paper, washed once or twice with saline, transferred 
to a dish or foil, and dried in the incubator. The dried material is 
now weighed and each gram ground with one gram of salt and gradually 
rubbed with 100 c.c. of sterile water. Each cubic centimeter, therefore, 
contains 10 mg. in isotonic saline. The emulsion is heated to 60° C. for 
one hour, preserved with 0.25 per cent, phenol or tricresol, and cultured for 
sterility. The first dose is 0.3 or 0.4 c.c. (3 or 4 mg.) and subsequent doses 
are gradually increased. 
Fischl,2 Muller,3 Grabisch,4 and others have reported good results in the 
treatment of trichophyton infections with subcutaneous injections of 20 
per cent, solutions of turpentine in sterile olive oil; the initial dose is about 
0.2 c.c. and subsequent doses at intervals of three to five days are gradually 
increased. Especially good results were observed in deep lesions. Lowen- 
feld and Paulay5 observed as good results in the treatment of these infections 
with such non-specific agents as tuberculin and turpentine as with specific 
vaccines (“trichon”)- Ruete,6 however, did not observe beneficial effects 
from injections of turpentine. 
Reese,7 Scholz and Kraus,8 and others have treated large series of cases 
with intramuscular injections of 5 to 10 c.c. of boiled milk and regard this 
type of non-specific treatment as yielding better results than turpentine 
injections. 
It is apparent, therefore, that in severe cases of ringworm of the body, 
scalp, and bearded region treatment by subcutaneous injections of vaccine, 
turpentine, or milk (intramuscularly) may prove valuable adjuvants to local 
therapy; the beneficial effects ascribed to the vaccine are probably in large 
part non-specific. 
Treatment of Pemphigus; Urticaria; Dermatitis Herpetiformis; Prurigo; 
Pruritus; Lichen Planus; Strophulus and other Dermatoses; Pellagra.— 
Many observers have reported favorably upon the treatment of various 
dermatoses, notably the various forms of urticaria and pruritus with intra- 
venous injections of the patient’s own serum. Urticaria is frequently a 
manifestation of allergy to foods or other agents and the subject has been 
previously discussed from this standpoint. 
In the herpes and pruritus of pregnancy and vomiting of pregnancy, Mayer 
and Linser,9 Mayer,10 Freund,11 Linser,12 Fetzer,13 Veiel,14 Rubsamen,15 and 
others have reported marked relief following the intravenous injection of 
the patient’s own serum in doses varying from 10 to 50 c.c. In senile 
pruritus marked relief is sometimes afforded by the intramuscular injection 
of 5 to 10 c.c. of milk, boiled for ten minutes and cooled. A few hours 
later a mild fever develops with local soreness, but generally with total 
1 Jour. Amer. Med. Assoc., 1816, 66, 945. 
2 Wien. klin. Wchn., 1919, 2, 2. 
3 Miinch. med. Wchn., 1918, Ixv, 697. 
4 Dermat. Wchn., 1918, lxvii, 624. 
6 Wien. klin. Wchn., 1919, 32, 498. 
6 Dermat. Ztschr., 1919, 28, 28. 
7 Miinch. med. Wchn., 1919, lxvi, 747. 
8 Dermat. Wchn., 1918, lxvii, 857. 
9 Miinch. med. Wchn., 1910, Ivii, 2757. 
10 Miinch. med. Wchn., 1911, 35, 1299. 
11 Med. Klinik, 1911, 7, 371. 
12 Dermat. Ztschr., 1911, 17,'217. 
13 Deut. Gesellsch. f. Gyn., 1911, 14, 712. 
14 Miinch. med. Wchn., 1912, lix, 1911. 
15Deutsch. med. Wchn., 1913, 39, 931. TREATMENT OF DISEASES OF THE SKIN 
997 
relief of the intolerable itching of body and scalp which lasts for several 
days and even weeks, when a second injection is required. 
In urticaria, dermatitis herpetiformis, prurigo, pruritus, lichen planus, 
strophulus, and hemorrhagic diseases of the skin and mucous membranes 
good results from this therapy have been reported by Linser,1 Heuck,2 
von Zumbusch,3 Ullman,4 Spiethoff,5 Praetorius,6 Holobut and Lenartowicz,7 
Gottheil and Sattenstein,8 Willock,9 Swann,10 and others. In many cases 
relief was afforded by one to three injections of the patient’s own serum in 
dose of 10 to 25 c.c. or more, but not infrequently the beneficial effects are 
only temporary. The treatment would appear to be especially worthy of 
trial in the toxic dermatoses of pregnancy, severe urticaria, dermatitis 
herpetiformis, and pemphigus; cures are not commonly effected and this is 
especially true of pemphigus. 
Palmer and Secor11 have reported favorable results in the treatment of 
pellagra by producing blisters with cantharides, drawing off the serum and 
injecting 1 c.c. at once subcutaneously; these injections were made about 
once a week. 
The technic of securing the blood and serum and for giving the injec- 
tions, has been described in Chapter XXXVII. 
Subcutaneous injections of 20 per cent, turpentine in sterile olive oil in 
ascending doses beginning with 0.1 or 0.2 c.c. and intramuscular injections 
of 50,000,000 to 75,000,000 of typhoid bacilli have also been employed in the 
treatment of these skin affections as well as in the treatment of lupus erylh- 
ematosis and lupus vulgaris, expoliative dermatitis, purpuras (especially 
with injections of milk owing to its styptic qualities), leg ulcers, erythema 
multiforme, erythema nodosum, and actinomycosis. 
Vaccine Treatment of Pruritus Ani, Vulvce and Scroti.—Murray12 has 
reported a series of cases of pruritus ani relieved or apparently cured by the 
administration of autogenous vaccines of Streptococcus fecalis secured in 
cultures of the effected skin as follows: After preliminary cleansing with 
soap and water, flushing with sterile water, and drying with gauze, the 
deeper layers of the skin are exposed with a sterile skin curet, avoiding 
bleeding as much as possible, and cultures made on blood or ascites agar. 
The vaccine is sterilized with 0.5 per cent, phenol or 0.3 per cent, tricresol, 
and not with heat. 
Injections are given subcutaneously once a week in ascending doses. 
With a vaccine containing 1 ,'000,000,000 per cubic centimeter the first dose 
may be 0.1 c.c. 
Murray states that of 181 cases distinct relief was obtained in 113, 
including some with severe pruritus involving the neighboring parts. Frick13 
has also reported favorably upon the treatment of 40 cases of pruritus ani, 
and it would appear that autogenous vaccine therapy is worthy of trial in 
conjunction with appropriate local treatment. 
The intramuscular injection of 5 or 10 c.c. of milk, boiled for ten minutes 
and cooled, has, in my experience, afforded complete relief to several patients 
for variable periods of time. 
1 Arch. f. Dermat. u. Syph., 1912, cxiii, 701. 
2 Miinch. med. Wchn., 1912, lix, 2608. 
3 Wien. klin. Wchn., 1913, 38, 2348. 
4 Arch. f. Dermat. u. Syph., 1913-14, cxviii, 125. 
6 Miinch. med. Wchn., 1913, lx, 521; Med. Klinik, 1914, 11, 29; ibid., 1916, 12, 1223. 
6 Miinch. med. Wchn., 1913, lx, 867. 10 Jour. Amer. Med. Assoc., 1915, 64, 737. 
7 Dermat. Wchn., 1914, lviii, 41. 11 Jour. Amer. Med. Assoc., 1915, 64, 1566. 
8 Jour. Amer. Med. Assoc., 1914, 63, 1190. 12 Jour. Amer. Med. Assoc., 1918, 71, 1449. 
9 Jour. Amer. Med. Assoc., 1916, 65, 14. 13 Ohio State Med. Jour., October, 1919. 998 
VACCINES, SERA, BLOOD, PROTEINS IN TREATMENT OF DISEASE 
Vaccine Treatment of Actinomycosis.—Wynn,1 Whittier,2 Kinnicutt and 
Mixter,3 Collie,4 Malcolm,5 and Dean6 have reported the successful treat- 
ment of actinomycosis with vaccines when iodides have failed. These 
vaccines have been prepared of the ray fungus cultivated on Sabouraud’s 
or other suitable media, broken up by grinding, suspended in saline solution, 
and sterilized by heating. The methods described above for the preparation 
of ringworm vaccine may be employed. 
Standardization of the vaccine is rather difficult. Dean states that no 
more than 10,000,000 actino fragments should be given in one dose. My 
practice is to prepare a suspension of a density equalling tubes No. 5 to 8 of 
the McFarland nephelometer and begin treatment with 0.1 c.c. Subsequent 
injections are given at intervals of one week in ascending doses. 
Treatment of Other Skin Diseases.—The treatment of dermatitis venenata 
has been discussed in Chapter XXXII; furunculosis has been considered in 
this chapter and tuberculosis of the skin will be considered in the following 
chapter on Tuberculosis Therapy. 
TREATMENT OF GLANDERS 
Immunity in Glanders.—This disease is caused by Bacillus mallei and 
occurs most frequently among horses and mules. Man is occasionally 
infected and especially stablemen. Cattle and rats possess a high degree 
of natural resistance; sheep, goats, dogs, and rabbits likewise possess some 
natural immunity but may be infected experimentally, while guinea-pigs 
and animals of the cat family are highly susceptible. In man the disease 
is usually chronic, has a mortality of about 50 per cent., and, according to 
Fitch,7 is frequently mistaken for pyemia, rheumatism, syphilis, or other 
disease. 
The nature of the natural immunity exhibited against B. mallei is not 
known; very probably phagocytosis is an important means of resistance to 
infection. 
During the disease in man and the lower animals various antibodies 
are to be found in the blood, notably agglutinins and complement-fixing 
antibodies. These are very useful for serologic diagnosis, as discussed in 
previous chapters. Allergic sensitization to B. mallei proteins also occurs 
and a product of the bacillus, called mallein, prepared in the same manner as 
old tuberculin, is widely employed by veterinarians for diagnostic purposes. 
These antibodies can be produced artificially by the immunization of 
horses and other animals with dead and living bacilli, but vaccines are only 
of doubtful value for prophylactic immunization and the immune serum is 
of little benefit in treatment. 
Immunity to glanders in the horse and other lower animals is poorly 
developed. In this respect the disease resembles tuberculosis. Chronic 
lesions may become acute and one attack does not confer immunity to 
subsequent attacks. This is particularly perplexing and paradoxic in view 
of the readiness with which antibodies are produced during the disease or 
by artificial immunization. 
Vaccine Prophylaxis and Treatment.—The use of vaccines for the pro- 
phylaxis of glanders among horses is sometimes employed; according to 
some investigations a temporary immunity may be secured, but corrobora- 
tion is lacking. 
1 Brit. Med. Jour., 1908, 1, 554. 
2 Jour. Amer. Med. Assoc., 1909, 53, 1453. 
3 Boston Med. and Surg. Jour., 1912, 67, No. 3. 
4 Brit. Med. Jour., 1913, 1, 991. 
6 Brit. Med. Jour., 1916, 2, 488. 
6 Brit. Med. Jour., 1917, 82. 
7 Cornell Veterinarian, July, 1914. TREATMENT OF LEPROSY 
999 
Vaccines and mallein have also been employed in the treatment of glan- 
ders of horses and of human beings. The dose for human beings may be 
0.1 c.c. of a vaccine containing 1,000,000,000 per cubic centimeter; injections 
are subcutaneous and repeated at intervals of five to seven days in increasing 
amounts. Bristow and White,1 Cramp,2 Zieler,3 and others have reported 
success with autogenous vaccines; mallein has been employed by Robins4 
in a manner analogous to the subcutaneous injection of old tuberculin. 
While the disease is fortunately rare in human beings it is nevertheless 
important, by reason of the high mortality, and vaccine treatment should 
be employed along with the usual medicinal and surgical measures. 
TREATMENT OF LEPROSY 
Immunity in Leprosy.—Leprosy is caused by Bacillus lepra, but it is 
still doubtful whether this organism has been successfully cultivated outside 
of the tissues. The disease is one of great antiquity and all races of human 
beings are susceptible to infection. Children are probably more susceptible 
than adults; whether or not the disease is transmitted in the uterus is still 
undecided, but probably is not. 
The ordinary domestic animals enjoy an absolute immunity to the 
disease; rat leprosy occurs, but whether it is caused by the human bacillus 
is unknown, but probably is a separate infection. Monkeys have been 
successfully inoculated according to some investigations, but these observa- 
tions require corroboration. 
Man is probably infected through the upper respiratory passages and the 
bacilli are constantly found on the nasal mucosa of lepers. Prolonged and 
intimate contact is usually required for infection, and many individuals 
apparently escape altogether in spite of frequent and intimate exposure. 
Either the bacilli are but feebly pathogenic and largely destroyed by phago- 
cytic cells (mononuclear leukocytes) or else a prolonged period of incubation 
is required before lesions develop—a matter of years. There is evidence 
indicating that bacteremia occurs and that the bacilli may be transmitted 
to the skin and other organs and tissues by way of the blood-stream. 
Antibodies are apparently produced during the course of disease, at least 
to a degree sufficient for bringing about long periods of latency, spontaneous 
improvement, and even apparent cure in some cases. The degree of 
immunity, however, is usually insufficient for bringing about a cure and in 
the majority of cases the disease progresses to a fatal end. Furthermore, 
this immunity is not transmitted; indeed, the children of leprous mothers 
are quite susceptible to infection. 
The nature of this acquired immunity is unknown. Probably immune 
opsonins for the bacilli are of most importance; complement-fixing anti- 
bodies have been found in the blood by some observers and allergic sensitiza- 
tion is said to occur as in tuberculosis. 
Vaccine and Serum Treatment.—Various sera have been employed, 
but with indifferent or negative results. Probably the oldest method is that 
of Carrasquilla,5 who immunized horses with the sera of leprous individuals. 
A large literature has accumulated on the results of treatment with this 
supposedly immune serum, but the results are probably not better than those 
to be obtained with normal horse-serum and other non-specific agents. 
1 New York State Jour. Med., 1910, 236. 
2 Jour. Amer. Med. Assoc., 1911, 56, 1379. 
3 Deutsch. med. Wchn., 1920, 46, 209. 
4 Studies from the Roy. Viet. Hosp., 1906, 2, 1. 
5 Wien. klin. Wchn., 1897, Nos. 41 and 42. 1000 
VACCINES, SERA, BLOOD, PROTEINS IN TREATMENT OF DISEASE 
Immune sera have also been prepared by Sugai1 and others by immunizing 
horses with the expressed juices of leprous nodules; Currie, Clegg, and Hol- 
man2 have immunized with an acid-fast bacillus regarded as Bacillus leprae. 
None of these sera can be said to have proved of value. Janin3 has used in- 
jections of blister serum from the patient, giving 8 to 10 c.c. by subcuta- 
neous injection every ten days. Improvement was claimed for some cases. 
A large number of vaccines have also been employed including the juices 
of ground up and expressed leprous nodules containing large numbers of 
bacilli; vaccines of various acid-fast and chromogenic bacilli regarded as 
the leprosy bacillus and autolytic extracts of these prepared after the manner 
of Koch’s old tuberculin. Favorable results were claimed in some instances, 
but the vaccines were ordinarily prone to produce abscesses, and the con- 
sensus of opinion is to the effect that they have not proved beneficial. 
A large literature, however, has accumulated on the use of nastin in 
treatment. Schumacher,4 Peiper,5 Scott,6 and others believe that the results 
are beneficial and encouraging, the latter observing improvement in 85 per 
cent, of cases. Wise and Minett7 observed temporary improvement in some 
cases only, while Minett8 and others have reported negative results. 
Tuberculin has also been employed in treatment, but with generally 
negative results. 
TREATMENT OF DISEASES OF THE EAR AND MASTOID 
Furunculosis of the external auditory canal is practically always a 
staphylococci infection, Staphylococcus pyogenes albus being found in the 
small superficial lesions and S. aureus in the deeper lesions. In chronic and 
recurrent infections vaccine therapy with stock or autogenous staphylo- 
coccus vaccines is frequently of great benefit. Dosage and administration 
have been previously discussed under Furunculosis. 
Vaccine Treatment of Acute and Chronic Otitis Media.—Otitis media 
is usually an ascending infection from the nasopharynx by wray of the 
eustachian tube. Cultures generally show the presence of a single organism, 
but two or more may be the etiologic agents. Streptococci, pneumococci, 
and Staphylococcus aureus are most frequently found; the Friedlander 
bacillus, Micrococcus catarrhalis, and the influenza bacillus are found occa- 
sionally. 
After rupture of the tympanum and in chronic otitis media other organ- 
isms may gain access through the external auditory canal and in cultures 
overgrow the primary organisms. Diphtheroid bacilli. Bacillus pyocyaneus, 
B. proteus, B. coli, and staphylococci are commonly found, usually in com- 
binations of two or more different organisms. 
In acute otitis media it would appear advisable to make cultures of the 
pus immediately after incision or rupture of the typanum; blood-agar is 
the medium of choice. These cultures generally disclose the primary 
organism in pure culture. 
In chronic otitis media cultures should be made from the middle ear 
after cleansing of the external auditory canal in order to avoid as far as 
possible picking up secondary organisms. The pus should be streaked over 
1 Arch. f. Dermat. and Syph., 1912, cxii, 88. 
2 Lepra, 1912, 13, 25. 
3 Rev. d. Med. et d’Hyg. Trop., 1913, 10, 81. 
4 Arch. f. Schiffs u. Trop.-Hyg., 1913, 17, 15. 
8 Arch. f. Schiffs u. Trop.-Hyg., 1913, 17, 183. 
6 Indiana Jour. Med. Res., 1913, 1, 352. 
7 Jour. Trop. Med. and Hyg., 1912, 15, 259. 
8 Brit. Guiana Med. Ann., 1913, 24. TREATMENT OF DISEASES OF THE EAR AND MASTOID 
1001 
plates of blood-agar in order to permit the less hardy and slower growing 
but more important organisms a chance to multiply, especially pneumococci 
and streptococci. This is of considerable importance in relation to treatment 
with autogenous vaccines, inasmuch as vaccines of the more rapidly growing 
secondary and frequently saprophytic bacteria are of little use. 
In acute otitis media more observers have reported prompt relief of pain 
and resolution of the inflammatory process following the early subcutaneous 
injection of a mixed stock vaccine of streptococci, pneumococci, and staphyl- 
ococci. 
In chronic otitis media autogenous vaccines are frequently of decided 
help in treatment while the usual local measures are being employed. It is 
essential to have very carefully prepared cultures; much of the success 
depends primarily upon the bacteriologic work. I believe that it is a 
good routine practice to institute vaccine therapy in all subacute cases 
before chronicity and involvement of the bony tissues is established. Even 
in the long-standing cases good results are sometimes observed from autog- 
enous vaccine therapy, although a long series of injections are usually re- 
quired, as is true of superficial infections in general. 
In the otitis media of scarlet fever Weston and myself1 observed that 
in general terms the suppuration was of shorter duration among the vaccine- 
treated cases; in some instances particularly good results were observed, 
although it is to be borne in mind that the disease is frequently cured in an 
equally short time by the usual forms of local treatment. Coates,2 Haughey,3 
and others have reported favorably upon the treatment of otitis media, 
"especially with autogenous vaccines. They have proved especially useful 
in children, among whom local treatment is frequently difficult. 
My practice is to prepare a vaccine for children up to twelve years of 
age containing a total of 500,000,000 organisms per cubic centimeter. 
Diphtheroid bacilli and Bacillus proteus are not included because they are 
generally saprophytes. B. pyocyaneus, however, is always included, and 
sometimes when found in pure culture in chronic infections has yielded 
satisfactory results. The organisms chosen for the vaccine are employed 
in equal proportions. For individuals over twelve years of age the vaccine 
is made to contain 1,000,000,000 per cubic centimeter. 
The first dose is 0.1 c.c. by subcutaneous injection; subsequent injections 
are given at intervals of five to seven days in gradually increasing amounts 
until 1 c.c. is being given at one time. As a general rule six to eighteen 
injections are required, but if there is not distinct improvement after eight 
to ten injections I believe it is good practice to discard the vaccine and 
prepare a second. 
If autogenous vaccines are not to be had, good results may be obtained 
with a stock vaccine of streptococci, pneumococci, and staphylococci; indeed, 
they sometimes yield better results than improperly prepared autogenous 
vaccines. 
Rauch4 and Lawner5 have reported good results in the treatment of acute 
and chronic otitis media with intramuscular injections of 5 c.c. of market 
milk boiled for ten minutes and cooled. Gomperz,6 however, did not 
observe beneficial effects with this therapy. Hirsch7 employed subcutane- 
ous injections of turpentine (20 per cent, dilutions), with negative results. 
1 Amer. Jour. Med. Sci., 1911, 142, 403. 
2 Jour. Amer. Med., Assoc., 1915, 65, 356; ibid., 1917, 68, 162. 
3 Ann. Otol., Rhinol., and Laryngol., 1915, 24, No. 1. 
4 Wien. klin. Wchn., 1917, No. 43. 
5 Wien. klin. Wchn., 1917, 30, 17. 
6 Wien. med. Wchn., 1917, 67, No. 37. 7 Arch. f. Ohrenh., 1919, cv, 62. 1002 
VACCINES, SERA, BLOOD, PROTEINS IN TREATMENT OF DISEASE 
In tuberculosis of the ear the subcutaneous injection of tuberculin 
frequently yields excellent results, and in the opinion of Randall and other 
otologists of extensive experience should be included in the plan of treat- 
ment. This subject will be discussed in more detail in the succeeding chapter 
on Tuberculosis Therapy. 
Vaccine Treatment of Acute and Chronic Mastoiditis.—In the opinion 
of some physicians the prompt administration of a stock vaccine of staphylo- 
cocci, streptococci, and pneumococci may shorten the course of acute mas- 
toiditis and relieve the pain; the writer has had no personal experience with 
this therapy. Davis1 has recently advocated the administration of stock 
vaccines of staphylococci and streptococci in acute mastoiditis complicating 
otitis media along, of course, with local treatment. He believes that the 
prompt institution of vaccine therapy reduces the percentage of cases 
requiring operation. 
In chronic suppurative mastoiditis following operation autogenous, vac- 
cines are worthy of trial, although in my experience they have not yielded 
encouraging results. Superficial infections of bony tissue are generally 
refractory to immunotherapy. 
If vaccines are prepared great care should be exercised in the bac- 
teriologic work in order to secure the streptococci or pneumococci, which 
are usually the organisms of primary etiologic importance. 
TREATMENT OF DISEASES OF THE RESPIRATORY ORGANS AND ACCES- 
SORY SINUSES 
Skin tests for the diagnosis and treatment of allergic rhinitis, hay-fever, 
and asthma, and the treatment of diphtheria and pneumonia have been 
previously described. 
Vaccine Treatment of Acute Rhinitis.—The uncertain state of our 
knowledge of the etiology of the “common cold” has been discussed on 
p. 815. According to some observers the organisms to be found in the nasal 
secretions by ordinary cultural methods are to be regarded as the etiologic 
factors, and especially pneumococci, streptococci, Micrococcus catarrhalis, 
staphylococci, and B. influenza; Tunnicliff has described a small anaerobic 
bacillus, and Foster, Kruse, and others a filterable virus. 
It is obvious that until the etiology is conclusively determined that 
treatment of acute rhinitis with vaccines of the cultivatable bacteria has 
little or no standing in so far as specific therapy is concerned. On the other 
hand, it has been the experience of many physicians that the prompt sub- 
cutaneous injection of these mixed vaccines has served to abort or shorten 
the attacks of acute rhinitis. Stock vaccines have been commonly em- 
ployed. 
The writer has observed that the administration of autogenous vaccines 
of this kind prepared of cultures made during a previous attack has fre- 
quently been followed by a prompt relief of the symptoms of acute rhinitis 
in some cases; in the treatment of the rhinitis of scarlet fever, particularly 
dangerous because favoring the presence and dissemination of the scarlet 
fever virus, Weston and the writer2 found autogenous vaccine of aid in 
shortening the course of the infection in some cases. Good results in the 
treatment of “common colds” with autogenous and stock vaccines have been 
reported by Fisher,3 Rowlette,4 and others. 
1 Penna. Med. Jour., 1922, 25, 306. 
2 Amer. Jour. Med. Sci., 1911, 142, 403. 
3 Boston Med. and Surg. Jour., 1913, 168, 834. 
4 Brit. Med. Jour., 1915, 1, 1046. DISEASES OF RESPIRATORY ORGANS AND ACCESSORY SINUSES 
1003 
Whether or not the effects of vaccine therapy in the treatment of rhinitis 
are purely non-specific cannot be stated; the writer believes, however, that 
bacteria of the pyogenic group to be found in the thick nasal discharges are 
at least important from the standpoint of secondary infection, even though 
their significance as primary etiologic factors is uncertain. Furthermore, 
there is evidence indicating that these vaccines apparently possess some 
prophylactic value for some individuals (p. 816), and the whole subject is 
worthy of further investigation in view of the high incidence and importance 
of the disease. 
Vaccine Treatment of Hay-fever.—Hay-fever is generally regarded as 
primarily due to allergic sensitization to various pollens; skin tests for 
diagnosis and desensitization by subcutaneous injections of pollen extracts 
have been previously described. 
During the actual attack of hay-fever, however, the injection of extracts 
of the pollen or pollens responsible for the sensitization and the attack may 
afford slight or no relief. 
In these cases Scheffegrell1 and others have employed a mixed treatment of 
vaccine and pollen extract. Stock vaccines of staphylococci, pneumococci, 
Micrococcus catarrhalis, etc., have been generally employed. Injections are 
given at intervals of one or two days until the severity of the attack subsides. 
The pollen extract is then used, the vaccine injections being resumed if a 
severe paroxysm develops. In 3 per cent, of a series of 1000 cases Scheffe- 
grell states that the treatment was limited to vaccine therapy alone and 
that the results in general were satisfactory when given in the acute attacks. 
The only logical basis for vaccine treatment in hay-fever aside from the 
possible beneficial non-specific effects is to combat secondary bacterial 
infection facilitated by the changes in the mucous membranes instituted 
by the allergic reaction. It would appear that in some cases this secondary 
infection may be worthy of treatment with vaccines as indicated. I have 
noted beneficial effects in cases treated with autogenous vaccines, and especi- 
ally those in whom the attacks were very severe and accompanied by asthma. 
Chronic Rhinitis; Ozena.—In chronic or frequently recurring rhinitis 
the administration of autogenous vaccines are sometimes of distinct value 
in treatment before hypertrophic or atrophic changes have occurred. Cul- 
tures on blood-agar usually show the presence of one or more different 
organisms, and especially staphylococci, streptococci, pneumococci, Micro- 
coccus catarrhalis, and Friedlander’s bacillus. In my experience the ad- 
ministration of autogenous vaccines in cases of frequently recurring rhinitis 
in whom a state of chronicity may be said to have been established, has fre- 
quently yielded very good results. The vaccine is prepared of equal num- 
bers of the bacteria secured in pure culture from plate cultures, employing 
blood-agar in order to favor the growth of the less easily grown organisms 
so that each cubic centimeter contains 1,000,000,000. Treatment is begun 
with 0.1 or 0.2 c.c. by subcutaneous injection and the injections given at 
intervals of five to seven days in increasing amounts. If improvement 
results after three to six injections, the intervals of injection is raised to two 
weeks, then once per month, and finally once in two to four months. 
Vaccine therapy has been extensively employed in the treatment of 
ozena. The vaccines of the ordinary bacteria recovered in cultures of the 
fetid secretions or of the mucosa beneath the crusts have failed to establish 
their value in the treatment of this exceedingly disagreeable affection. 
Vaccines of the cocco-bacillus of Perez,2 however, are claimed to have 
1 Hay-fever and Asthma, Lea & Febiger, 1922, 225. 
2 Ann. de l’Inst. Pasteur, 1899, 13, 937. 1004 
VACCINES, SERA, BLOOD, PROTEINS IN TREATMENT OF DISEASE 
proved of value by Hofer,1 Skillern and Holmes,2 Horn,3 Klenk,4 Guggenheim,5' 
and others. The bacillus is rather difficult to cultivate and isolate in pure 
culture, so that stock polyvalent vaccines have been generally employed. 
The available reports indicate that it may be worth while using a vaccine 
of this bacillus, although its etiologic relationship to the disease cannot be 
said to have been definitely proved. Hofer regards the organism (Bacillus 
ozcena-foelidce) as related to the Friedlander bacillus, while Horn claims 
that it is related to B. bronchisepticus, regarded by Ferry and others as the 
course of canine distemper. Perez found the bacillus in 8 of 22 cases of 
ozena; in a recent study of 50 cases Ward6 found it in 22 cases. 
An effort should be made to prepare an autogenous vaccine to contain 
1,000,000,000 per cubic centimeter; the initial dose may be 0.1 c.c. by 
subcutaneous injection, and subsequent doses increased at intervals of five 
to seven days until some focal reaction occurs in the nose after each injec- 
tion. A long series of injections are ordinarily required in conjunction 
with the usual local measures. 
Vaccine Treatment of Acute and Chronic Sinusitis.—Infections of the 
accessory nasal sinuses are usually secondary to nasal and postnasal disease; 
in my experience pneumococci are the most frequently found organisms in 
smears and cultures, but staphylococci and streptococci also occur in many 
cases. 
In acute sinusitis the prompt administration of a stock vaccine of staphy- 
lococci, streptococci, and pneumococci in total dose of 100,000,000 is reported 
by Davis7 and others to afford prompt relief in a large number of cases. 
Second and third injections are commonly given at intervals of forty-eight 
hours. 
In chronic sinusitis (frontal, ethmoidal, sphenoidal, antral) autogenous 
vaccines are sometimes of value as part of the treatment. A special effort 
should be made, however, to obtain the cultures of pus with as little con- 
tamination as possible and preferably upon plates of blood-agar in order 
that the growth of pneumococci and streptococci so readily overgrown 
by staphylococci and diphtheroid bacilli may be obtained. The vaccine 
may contain a total of 1,000,000,000 per cubic centimeter and treatment 
begun with 0.1 c.c. Subsequent injections may be gradually increased at 
intervals of five to seven days until 1 c.c. is being given at one time. Slight 
focal reactions (increased hyperemia and discharge) are desirable; for this 
purpose intramuscular injections of large doses are sometimes more effica- 
cious than subcutaneous injections. 
Vaccine Treatment of Acute Tonsillitis, Pharyngitis, and Laryngitis.— 
Some physicians have observed good results from the subcutaneous injec- 
tion of stock vaccines of staphylococci, pneumococci, streptococci, etc., 
early in the course of acute tonsillitis, pharyngitis, and laryngitis. Cultures 
usually show the presence of staphylococci or streptococci or a combination 
of these. The initial doses are small, varying from 25,000,000 to 100,000,000 
of each organism per cubic centimeter; second and third injections are 
given at intervals of forty-eight to seventy-two hours. 
Vaccine Treatment of Chronic Bronchitis.—In chronic bronchitis the 
administration of autogenous vaccines frequently proves of great value as 
an adjuvant to treatment, and especially in cases without advanced bron- 
chiectasis and emphysema. In my experience the results have been among 
1 Wien. klin. Wchn., 1913, 25, 1011. 
2 New York Med. Jour., August 15, 1908. 
3 Jour. Amer. Med. Assoc., 1915, 65, 788. 
4 Missouri State Med. Jour., 1916, 13, 197. 
5 Interstate Med. Jour., 1915, 22, 2. 
6 Jour. Infect. Dis., 1916, 19, 153. 
7 Penna. Med. Jour., 1922, 25, 306. DISEASES OF RESPIRATORY ORGANS AND ACCESSORY SINUSES 
1005 
the most satisfactory in vaccine therapy. Cases of bronchitis due primarily 
to passive congestion from cardiac decompensation are not included in this 
group. Those due primarily to bacterial infection from a preceding attack 
of influenza, pneumonia, or repeated attacks of acilte bronchitis and dis- 
playing what Johnson has called the “pneumocatarrhal diathesis,” are to 
be selected for vaccine therapy. In Chapter XXXII the work of Walker 
with bacterial vaccines in chronic bronchitis and asthma has been discussed, 
especially in relation to allergic asthma; Gillett,1 Pirie,2 Babcock,3 and others 
have reported favorably upon the use of autogenous vaccines in chronic 
bronchitis and asthma. 
Cultures on blood-agar or other suitable medium should be made of the 
bronchial secretions, and preferably of that coughed up upon arising in the 
morning. As a general rule streptococci predominate, although this varies 
in different cases, and staphylococci, pneumococci, Micrococcus catarrhalis, 
the Friedlander bacillus, and diphtheroid bacilli may be found. Smears of 
the secretions should be examined at the same time, as in this manner an 
idea may be gained regarding which organism predominates. 
Skin tests may be conducted to determine if allergy exists for any of the 
organisms secured in pure culture; the method has been described on p. 684. 
If asthma is absent these tests, however, are apt to be negative or indefinite, 
in which case mixed vaccines are to be employed. If positive skin reactions 
are observed the vaccine may be prepared of the organism or organisms 
producing the positive reactions. 
My practice is to prepare a vaccine containing 1,000,000,000 per cubic 
centimeter if more than one organism is used in equal proportions. If 
plates and smears show a great preponderance of streptococci these alone 
are employed; diphtheroid bacilli, Bacillus proteus, fungi, and the like, 
which may be present by reason of contamination of the bronchial secre- 
tions with saliva, are not included. 
The initial dose is 0.1 c.c. by subcutaneous injection; subsequent doses 
are gradually increased and given at intervals of five to seven days. Doses 
large enough to produce slight focal reactions characterized by increased 
expectoration during the twenty-four hours succeeding an injection are 
advisable, and especially with the first three or four doses. 
It is not well to give more than six to ten doses of the same vaccine unless 
decided improvement is noted; a fresh vaccine should be prepared rather 
than one vaccine continued over a long period of time. 
Treatment of Asthma.—The diagnosis of allergic asthma by skin tests 
has been described on p. 666; likewise treatment of allergic asthma including 
that type due to allergic sensitization to the proteins of bacteria present 
in the secretions. Walker,4 Sicard,6 Babcock,6 Hutcheson and Budd,7 and 
others have reported very favorable results in the treatment of some cases of 
bacterial asthma with vaccines. The latter authors have simply planted 
1 c.c. of washed sputum in 10 c.c. of broth to which is added a few drops of 
sterile guinea-pig serum. After forty-eight hours’ incubation the culture is 
heated to 60° C. for two hours, preserved with phenol, and injected subcuta- 
neously as a vaccine, beginning with 1 minim and gradually increasing at 
weekly intervals until 15 minims are being given at one time. Of 20 cases 
treated in this manner, 12 secured complete relief after one to five injections, 
5 showed improvement, 2 no improvement, and 1 became worse. 
1 Brit. Med. Jour., 1913, 1, 387. 
2 Brit. Med. Jour., 1913, 2, 1268. 
3 Jour. Amer. Med. Assoc., 1915, 65, 1942. 
4 Arch. Int. Med., 1919, 23, 220. 
6 Amer. Jour. Med. Sci., 1917, 153, 856. 
6 Jour. Amer. Med. Assoc., 1915, 65, 1942. 
7 Amer. Jour. Med. Sci., 1918, 155, 826. 1006 
VACCINES, SERA, BLOOD, PROTEINS IN TREATMENT OF DISEASE 
Various non-specific agents have likewise been employed. Auld1 has 
used Witte’s peptone by intravenous and subcutaneous injection, the dose 
being 5 to 10 c.c. of a sterilized 5 per cent, solution in saline solution. Sharp 
reactions usually follow and great care must be exercised. Some cases are 
promptly benefited by one or two injections; others are relieved by injections 
at intervals, and in others no beneficial effects are observed. Pagniez2 and 
Widal, Abrami and Brissand3 have administered peptone by mouth. Kahn 
and Emsheimer have reported definite improvement in 6 cases treated with 
subcutaneous injections of the patient’s defibrinaied blood in dose of 20 to 
30 c.c. by subcutaneous injection at intervals of one week for ten in- 
jections. 
Danysz4 has employed injections of his “enterovaccine” described on 
p. 994. Miller and Petersen have employed intravenous injections of typhoid 
bacilli and report that in some cases the results were quite satisfactory; in 
others there were no apparent effects, and especially those in which food 
sensitization was demonstrable. 
Treatment of Pleuritis (Empyema).—Acute pleurisy with empyema is 
generally a pneumococcus infection developing during lobar pneumonia or 
a streptococcus infection following bronchopneumonia. After drainage 
has been established, secondary infection with staphylococci, diphtheroid 
bacilli, and B. pyocyaneus may occur. Chronic primary pleurisy is usually 
a tuberculous infection; other organisms may gain access from cavities and 
bronchi and produce secondary infections. 
Treatment of acute pneumococcus and streptococcus pleurisies with 
intrapleural injections of aniipneumococcus and aniislrepiococcus sera has 
not been generally employed, although encouraging results have been ob- 
served in the treatment of experimentally produced pleurisies of the lower 
animals by injections of these sera. Floyd5 has treated 20 cases of pneumo- 
coccus pleuritis and empyema produced during pneumonia by Types I and II 
pneumococci, with intrapleural injections of Type I or Type II serum. In 
each case thoracotomy was performed for drainage, the pus examined, and 
the pneumococci typed. Subsequently, every other day, when it was 
possible, the chest cavity was irrigated with saline solution, and 4 ounces 
of equal parts of appropriate immune serum and saline solution were put 
into the chest. It was found that healing occurred much more rapidly if 
secondary invaders, as staphylococci and diphtheroid bacilli, could be ex- 
cluded; the serum therapy appeared to aid in phagocytosis and to shorten 
the course of the infection. 
Vaccines have been employed, with indifferent results. The writer is of 
the opinion that cultures should always be made and treatment with autog- 
enous vaccines instituted as part of the treatment of pneumococcus, strepto- 
coccus, and staphylococcus pleurisies along with adequate drainage. These 
vaccines are so prepared that each cubic centimeter contains 1,000,000,000 
organisms; the initial dose is 0.1 c.c. by subcutaneous injection. Subsequent 
injections are given at intervals of five to seven days and gradually increased 
until 1 c.c. is being given at one time. 
Autoserum therapy, consisting of the withdrawal of 5 to 10 c.c. of fluid 
and injecting at once subcutaneously, has been advocated by a number of 
investigators for the treatment of tuberculous pleurisy. 
Numerous investigators, as, for example, Gilbert,6 Macon,7 Schniitgen,8 
1 Brit. Med. Jour., 1918, 2, 49. 
2 Presse med., 1920, 28, 65. 
3 Presse med., 1921, 29, 181. 
4 Arch. Int. Med., 1916, 18, 445. 
5 Jour. Immunology, 1920, 5, 321. 
6 Gaz. des Hop., 1894, 560. 
7 La Presse Medicale, 1909, No. 71, 627. 
8 Berl. klin. Wchn., 1909, No. 3, 97. DISEASES OF RESPIRATORY ORGANS AND ACCESSORY SINUSES 
Fishberg,1 Pfender,2 Dodal,3 and others, have reported favorable results in 
the treatment of tuberculous pleurisy with effusion, cases that arise either 
insidiously or abruptly with pain in one already tuberculous, or in one in 
whom tuberculosis is suspected, following withdrawal of a portion of the 
fluid and immediate injection of from 2 to 5 c.c. into the subcutaneous tissues. 
In these cases there is usually a sharp reaction, consisting of a rise in tem- 
perature, occasionally accompanied by chill, lassitude, and, in the majority 
of cases, diuresis or, more rarely, diarrhea, followed by gradual absorption 
of the fluid within the following few days up to two or three weeks. Fish- 
berg mentions the disappearance of pain, dyspnea, and prostration within 
two or three days in favorable cases. 
It is difficult to state whether the improvement is due to autotherapy 
or simply to the puncture and removal of so much fluid. Eisner4 has seen 
a leukocytosis follow injection of serum in experimental tuberculous infec- 
tions of rabbits and guinea-pigs, and believes that this explains the good 
results in this particular form of therapy. Zimmermann5 has expressed a 
similar opinion. Other investigators assert their belief in the presence of 
aggressins, bacteriolytic amboceptors, complements, and endolysins from 
disintegreated leukocytes as explaining the results. It is more likely that 
these fluids contain the bacilli or their products, and constitute a form of 
vaccine or autotuberculin, stimulating body cells to produce antibodies 
largely in the nature of bacteriotropins and bacteriolysins. Levy, Valenzi 
and Ponzin,6 Szurek,7 and Arnsperger8 are inclined to believe that the bene- 
ficial results are obtained independently of the injections, and while the 
procedure is quite generally regarded as perfectly safe, Jousset9 has recorded 
a case of cold abscess following an injection. This mode of treatment seems 
to have failed in about 10 to 15 per cent, of cases. Lyter10 has recently 
reviewed the subject and has come to the conclusion that the injections 
have little or no influence in hastening absorption of the exudate. In 23 
cases 8 were completely absorved in two weeks’ time—34 per cent., while 
in the balance the effusion did not lessen as a result of the treatment. Lyter 
observed practically no leukocytosis, and in only 2 of the 8 rapidly absorbing 
cases did he observe any increase of temperature. 
The technic is very simple, and the injections may be given by any 
physician who can make an ordinary exploratory puncture. In all cases 
where puncture shows the presence of a serous fluid the needle should not 
be withdrawn completely, but when its point has reached the subcutaneous 
tissues, from 2 to 5 c.c. should be injected then and there. In some patients 
it will be necessary to repeat the treatment several times every two or three 
days before any effect becomes evident. 
In view of the fact that the fluid may contain a sufficient number of living 
tubercle bacilli to produce secondary infection, it would appear advisable 
to withdraw fluid into an equal volume of 2 per cent, sodium citrate in 
normal salt solution, heat at 60° C. for one-half to one hour, and preserve 
in a sterile container with the addition of a few drops of 5 per cent, phenol 
until ready for subcutaneous injection in doses of from 2 to 5 c.c. 
1007 
1 Jour. Amer. Med. Assoc., 1913, lx, 962. 
2 Wash. Med. Ann., 1914, xiii, 83. 
3 Wien. med. Wchn., 1910, 455. 
4 Zeitschr. f. klin. Med., 1912, lxxvi, 34. 
8 St. Petersb. med. Wchn., 1909, No. 34, 461. 
6 Bull. et. mem. de la Soc. d. hop. d. Paris, 1910, xxvii, 265. 
7 Med. Klinik, 1909, No. 44, 1665. 
8 Therap. d. Gegenwart, 1911, lii, 495. 
9 Arch. gen. de med., 1912, xci, 141. 10 Amer. Jour. Med. Sci., 1918, 156, 665. 1008 
VACCINES, SERA, BLOOD, PROTEINS IN TREATMENT OF DISEASE 
The biologic therapy of typhoid fever, bacillary dysentery, and Asiatic 
cholera has been previously considered in this chapter. 
Vaccine Treatment of Chronic Gingivitis and Alveolitis; Apical Abscesses 
and Focal Infection.—Chronic gingivitis and pyorrhea alveolaris (Rigg’s 
Disease) are frequently bacterial infections in which different kinds of 
streptococci, pneumococci, staphylococci, etc., are to be found in the pus. 
Endamoeba buccalis are frequently found in smears, and are regarded by 
Smith and Barret as bearing a secondary relationship to the disease, that 
is, these protozoa are not the primary cause of gingivitis, but by reason of 
their burrowing movements may mechanically carry bacterial infection 
into the deeper tissues of the gums and even to the periosteum of the alveolar 
processes or tooth sockets just as Endamoeba histolytica are known to do 
in the wall of the colon in amebic dysentery. In a large number of cases of 
this disease smears reveal a great increase of various spirochetes sometimes 
associated with the banana-shaped or fusiform bacillus of Vincent’s angina. 
This type of spirochetic gingivitis usually responds to the local application 
of 1 per cent, solutions of arsphenamin, and the subject will be discussed 
again under Chemotherapy. 
In those cases of gingivitis in which smears of the secretions removed 
from the pockets about the teeth show only a few spirochetes and enormous 
numbers of bacteria with or without amebas, autogenous vaccines may 
prove of value as an adjuvant to the usual treatment applied to the gums. 
Cultures should be made on blood-agar in order to facilitate the growth 
of streptococci and staphylococci, which may be overgrowm by staphylo- 
cocci, diphtheroid bacilli, and other organisms. After cleansing the surface 
of the gums the pus expressed from the pockets is especially to be cultured. 
Head employs the following method: The pocket from which the material 
is to be obtained is first protected from mouth contamination by a sterile 
napkin. The gingival margin of the pocket is then washed with sterile 
cotton dipped in sterile salt solution. The margin of the pockets are slightly 
seered with an electrocautery so that the gum is distinctly whitened. The 
root of the tooth adjacent to the margin is thoroughly gone over with the 
cautery to destroy all extraneous flora of the mouth which may cause con- 
tamination. A thin spear of platinum about 3/1000 inch in thickness is 
is then heated to a cherry-red color and plunged into the bottom of the 
pocket. It is particularly essential that this blood-serum should be obtained 
from the wall of the pocket, so that any bacteria lurking within the tissues 
of the pocket may be secured, as it is presumable that the germs within the 
walls are most responsible for the disease. The smear should then be drawm 
directly out without touching any portion of the mouth to prevent con- 
tamination and the material inoculated in tubes of blood-agar or other 
suitable medium in the usual manner. 
The vaccine should be a mixed one if more than one organism is recovered 
in the pus, but streptococci and staphylococci are of most importance. 
The writer prepares these vaccines to contain 1,000,000,000 per cubic 
centimeter; the first dose is 0.1 c.c. by subcutaneous injection. Subsequent 
doses are given at intervals of five to seven days and in gradually increasing 
amounts. 
Head,1 who has had an exceptional experience in the vaccine therapy 
of gingivitis, commences treatment with much smaller doses, as low as 
30 000 organisms. This dose is raised weekly as long as the patient bears 
TREATMENT OF DISEASES OF THE DIGESTIVE TRACT 
1 Medical Record, July 6, 1918. TREATMENT OF DISEASES OF THE DIGESTIVE TRACT 
1009 
the injections well, even if the dose goes up to 1,000,000,000. As long as 
improvement is noted under a dose, that dose is maintained. 
Goodby1 was probably first to apply vaccine therapy in gingivitis. 
Beebe,2 Eyre and Payne,3 Williams,4 Beebe,5 Medalia,6 and others have made 
favorable reports, and in the bacterial forms of the disease vaccine therapy 
appears worthy of trial as part of the treatment. 
Aside from the possible value of vaccine therapy as an aid in the treat- 
ment of the purely local infection of the gums and alveolar tissues, these 
vaccines may possess an important prophylactic and curative activity for 
systemic infections due to the distribution of bacteria from these foci. This 
is especially true when abscesses are found at the roots of teeth. It is not 
an uncommon experience to observe that individuals with these infections 
may develop acute exacerbations of arthritis, iritis, neuritis, endocarditis, 
etc., as a result of the extraction of teeth or surgical treatment of the infected 
gums, probably caused by the introduction of extra amounts of bacteria or 
their products into the blood-stream from these focal infections. Reactions 
of this kind appear to show indubitably a direct relation between the local 
infection of the gums and tooth sockets and the distant lesions, and in these 
cases at least the employment of an autogenous vaccine appears a logical 
procedure not only for the aid it may render in the treatment of the local 
infection, but, even more importantly, for its possible value in the prophy- 
laxis and treatment of the secondary infections. Daland7 has emphasized 
the necessity of careful search for primary foci of infection in ulcerative 
endocarditis and especially the relation of oral sepsis to this infection. He 
recommends the removal of the foci if at all possible and treatment with 
autogenous vaccines, usually of Streptococcus hemolyticus. 
Focal infections of the upper respiratory tract embracing the nose and 
tonsils as well as the gums are regarded by Rehfuss and others as bearing 
an important relation to some bacterial infections of the gastro-intestinal 
tract. The investigations of Rosenow, previously referred to in a discussion 
of the subject of Focal Infection, indicate that streptococci from these foci 
may be productive of gastric ulcer, cholecystitis, and appendicitis. For 
these reasons more and more attention is being paid the important relation 
of “oral sepsis” to diseases of the gastro-intestinal tract, as well as of more 
distant organs, and autogenous vaccines of the bacteria of the primary foci 
are being more widely employed as part of the general plan of treatment. 
Treatment of Cholecystitis and Hepatitis.—Until recently vaccine therapy 
has had no wide application in the treatment of bacterial infections of the 
gall-bladder and its ducts and the biliary ducts of the liver. Stock vaccines 
have been administered in some cases prepared of streptococci, colon bacilli, 
etc., but without preliminary bacteriologic examinations. Autogenous 
and stock vaccines have also been employed in the treatment of these 
infections after drainage or removal of the gall-bladder. 
By means of Meltzer-Lyon method of gall-bladder drainage it is now 
possible to secure specimens of bile for microscopic and cultural study. 
The writer has had the opportunity of examining a large number of specimens 
for Dr. Lyon and his former associate Dr. Bartle, the technic and findings 
1 Lancet, 1907, 1, 663; ibid., 1909, 2, 1875. 
2 Boston Med. and Surg. Jour., 1909, clxi, 613. 
3 Proc. Roy. Soc. Med., Odontological Sec., 1909, 3. 
4 Amer. Jour. Med. Sci., 1911, 141, 666. 
5 Boston Med. and Surg. Jour., 1909, cxli, 613. 
6 Dental Cosmos, January and February, 1913; Boston Med. and Surg. Jour., 1913, 
clxix, 786. 
7 Medical Clinics of North America, Saunders Co., September, 1917. 1010 
VACCINES, SERA, BLOOD, PROTEINS IN TREATMENT OF DISEASE 
being reported elsewhere.1 Staphylococci, colon bacilli, and streptococci 
are commonly found, and the administration of autogenous vaccines has 
rendered important aid in the treatment of some cases of infection of these 
organs, especially vaccines of the staphylococci and streptococci. It is our 
practice to prepare these autogenous vaccines to contain 2,000,000,000 
organisms per cubic centimeter. The first dose is 0.1 c.c. by subcutaneous 
injection. Subsequent doses are given at intervals of five to seven days 
and gradually increased. 
Mention may also be made here of autoserum treatment of hepatitis 
with ascites. This therapy, first introduced by Gilbert for the treatment of 
pleurisy with effusion, has also been employed by Audibert and Monges,2 
Vitry and Sezary,3 Lahiri,4 and others for cirrhosis of the liver with ascites. 
The technic consists of the subcutaneous injection of 3 to 5 c.c. of the pa- 
tient’s own sterile ascites fluid at intervals of five to seven days. The above- 
mentioned observers report an increased output of urine and gradual absorp- 
tion of the transudate. The beneficial results ascribed to this treatment 
are probably due to the effects of a non-specific protein therapy. 
Vaccines have also been employed in the treatment of colitis, chronic 
intestinal toxemia, appendicitis, etc. Drew5 reports that in 14 out of 15 
cases of membranous colitis smears and cultures showed a great preponder- 
ance of Streptococcus fecalis. Autogenous vaccines beginning with small 
doses were found of aid in treatment. Langeron6 has reported a case of 
sepsis due to Streptococcus fecalis (enterococcus) successfully treated with 
an autogenous vaccine. Satterlee7 states that the predominating factor 
in the symptomatology of chronic intestinal toxemia is the colon bacillus 
and that autogenous vaccines of this bacillus should be administered as part 
of the treatment to immunize against the toxic substances being produced. 
Of interest in this connection is the work of Torrey and Rahe,8 who found 
that in a number of dogs it was possible to effect a temporary suppression 
of a certain variety of Bacillus coli normal to the intestinal tract by injec- 
tions of vaccine. Autogenous vaccines in large doses are apparently nec- 
essary for bringing about a decrease in numbers of the colon bacilli with a 
coincident use of specific antibodies in the blood. 
Vaccine Treatment of Sprue.—Michel9 has recently reported the success 
ful treatment of sprue with vaccines prepared of a monilia (Monilia psilosis), 
which Ashford has isolated from the tongue and feces of individuals with 
this disease and regards as the etiologic agent. 
Michel has described a method for preparing the vaccine which he 
administers in dose of 0.1 c.c. by subcutaneous injection; subsequent injec- 
tions are given at intervals of ten days in gradually increasing amounts. 
Of 81 cases diagnosed by Ashford and yielding Monilia psiiosis in the 
feces and positive complement-fixation tests, 62 were treated with vaccine. 
Of these, 49 were discharged cured, 12 were improved, and 1 died. These 
results were regarded as very encouraging. 
1 Gall Tract Disease, its Diagnosis and Treatment, Lea & Febiger. 
2 Presse m£d., 1910, 18. 
3 Rev. d. med., 1913, 33. 
4 Practitioner, London, 1912, 88, No. 3. 
5 Jour. State Med., London, 1918, 26, 217. 
6 Bull. d. 1. Soc. M6d. d. H6p., 1918, 42, 43. 
7 Jour. Amer. Med. Assoc., 1916, 67, 1729. 
8 Jour. Immunology, 1920, 5, 133. 
9 Jour. Infect. Dis., 1918, 22, 53. TREATMENT OF DISEASES OF KIDNEY AND BLADDER 
1011 
TREATMENT OF DISEASES OF THE KIDNEY AND BLADDER 
Vaccine Treatment of Bacteriuria.—Not infrequently bacteria are 
found in the urine collected under rigid aseptic precautions, without any 
evidences of local or systemic disturbances. In fully 50 per cent, of these 
cases the colon bacillus is found. Bacteriuria is more common in women 
than in men, as shown by the investigations of Willians and Murray1 and 
others. Colon bacilli are likewise found in the urine of children, especially 
female children, and frequently without demonstrable evidences of pyelitis 
or other infection. How colon bacilli gain access to the urinary tract in 
apparently normal individuals is not definitely known; of course if cath- 
erization has been practised this avenue of contamination is first suspected. 
It is probable that in some cases, and especially in women, contamination 
occurs by way of the urethra. Otherwise the entrance of colon bacilli into 
the blood from the intestinal tract and their elimination through the kidneys 
without the production of infection is to be suspected. Staphylococcic 
bacteriuria may also occur, but the majority of cases are due to colon bacilli. 
Of special interest in this connection is the possibility of elimination of 
tubercle bacilli in the urine without infection of the urogenital organs. It 
would appear that this is only possible in cases of tubercle bacilli in the 
blood and their ability to pass into the uriniferous tubules. Personally I 
doubt that this occurs, and in my experience the presence of tubercle bacilli 
in urine has always been found due to a tuberculous infection of the kidney or 
other organs of the urogenital tract. 
Vaccines have been employed in the treatment of bacteriuria, but without 
much success. Whenever employed it is essential that autogenous vaccines 
be prepared of cultures of the urine secured by catherization under rigid 
aseptic conditions. Many different strains of Bacillus coli are known to 
exist, and nothing is to be hoped for in vaccine therapy of colon infections 
unless the strain or strains present in the urine are employed. 
Vaccine Treatment of Cystitis—Bacterial infections of the bladder may 
occur as a result of ascending infections from the urethra and prostate, 
catheterization, and direct injury; doubtless, infection may descend from 
the kidneys and especially in tuberculosis. 
Colon bacilli are frequently the primary etiologic factors, but strepto- 
cocci, staphylococci, tubercle bacilli, and other organisms may be present 
either as primary infection or secondary invaders. 
The vaccine treatment of tuberculous cystitis is considered in the suc- 
ceeding chapter under Tuberculin Therapy. In other bacterial infections 
autogenous vaccines may prove of aid as part of the general and local treat- 
ment. 
It is essential, however, that the urine for culture is collected under rigid 
aseptic precautions in order that the infecting bacteria are obtained. Bil- 
lings,2 Williams and Murray, and others have found colon vaccines of value; 
even better results have been observed in staphylococcic and streptococcic 
infections. 
During the Great War, Walker and Shera found autogenous vaccines very 
useful in the treatment of paraplegic soldiers infected by rough-and-ready 
catheterization on the field and 90 per cent, of whom later developed cystitis 
due to B. coli or a staphylococcus, or a mixture of these. 
Vaccines may be prepared containing 1,000,000,000 per cubic centimeter; 
the initial dose may be 0.1 c.c. by subcutaneous injections given at intervals 
1 Jour. Obst. and Gyn., Brit. Emp., 1912, 22, 65 
2 Amer. Jour. Med. Sci., 1910, 139, 625. 1012 
VACCINES, SERA, BLOOD, PROTEINS IN TREATMENT OF DISEASE 
of five to seven days in increasing amounts. It is well to secure fresh cul- 
tures and prepare a second vaccine after six to ten doses have been given, 
instead of administering the same vaccine over a prolonged period. 
Treatment of Pyelitis of Children.—This infection is almost invariably 
due to a strain of the colon bacillus and the route of infection usually obscure 
or unknown. The infection may be present without producing well-marked 
symptoms and is a troublesome infection from the standpoint of treatment. 
Vaccines are apparently helpful in the treatment of some cases as re- 
ported by Freeman1 and others; it is essential that autogenous vaccines be 
employed. Securing urine under proper conditions for cultures is frequently 
very difficult and requires special methods. 
For children varying in age from one to twelve years the vaccine may 
contain 500,000,000 bacilli per cubic centimeter. The initial dose may be 
0.1 c.c. by subcutaneous injection; subsequent injections are given at inter- 
vals of five to seven days in gradually increasing amounts. 
Karo2 has employed subcutaneous injections of lerpichin; 20 per cent, 
solutions of turpentine in sterile olive oil ‘may be injected subcutaneously 
in dose of a few drops to produce sterile abscesses and non-specific effects. 
The intramuscular injections of 1 c.c. of milk, boiled for ten minutes and 
cooled, is also worthy of trial. 
Treatment of Pyelonephritis.—Pyelonephritis may be caused by Staphyl- 
ococcus aureus, streptococci, tubercle bacilli, and other organisms during 
the course of bacteremia from some primary focus of infection, or develop 
from a pyelitis and ascending infection from the urinary bladder. In the 
latter cases the colon bacillus is usually the cause. 
Vaccines have been employed in the treatment of some cases, but it is 
not possible to express an opinion of their value. It would appear that the 
consensus of opinion among surgeons is in favor of nephrectomy and espe- 
cially in tuberculous nephritis. Bumpus3 has found that the administration 
of mixed colon bacillus vaccine in prostatic cases does not engender an 
immunity to renal infection preliminary to operations on the prostate and 
bladder. 
If operation is refused or contraindicated it would appear worth while 
to employ an autogenous vaccine prepared of cultures of the urine secured 
by catheterization, along with general measures, and particularly a search 
for and removal of primary foci in suppurative nephritis of embolic origin. 
Such vaccines may contain 1,000,000,000 organisms per cubic centi- 
meter and treatment begun with the subcutaneous injection of 0.1 c.c. 
Gow4 has reported 1 case of colon bacillus pyelonephritis successfully 
treated with intravenous injections of 50,000,000, 75,000,000, and 125,000,00*0 
bacilli. 
TREATMENT OF DISEASES OF THE EYE 
Infection and Immunity in Diseases of the Eye.—Compared to the 
frequency of infection in other parts of the body infections of the eye after 
injuries and operations are singularly infrequent; undoubtedly the eye shows 
less tendency to infection than other parts of the body. This natural 
resistance is largely due to mechanical cleansing by the tears; these secre- 
tions also possess a feeble but relatively unimportant bactericidal activity. 
Infections of the conjunctiva, lids, and lacrimal gland are usually due to 
direct implantation of organisms; the pneumococcus, gonococcus, staphylo- 
coccus, Koch-Weeks and Morax-Axenfeld bacillus are particularly important 
1 Amer. Jour. Dis. Child., 1913, 6, 117. 
2 Deut. rr.ed. Wchn., 1919, xvi, 266. 
3 Jour. Amer. Med. Assoc., 1918, 70. 213. 
4 St. Barth. Hosp. Jour., 1918-19, 26, 75. TREATMENT OF DISEASES OF THE EYE 
1013 
in this connection and virulent strains are able to overcome the factors of 
natural resistance with the production of purulent inflammation. Even 
in severe streptococcus, pneumococcus, and other blood bacteremias these 
structures are only rarely involved by embolic infections. Why diphtheritic, 
gonococcus, pneumococcus, streptococcus, staphylococcus, and influenzal 
conjunctivitis are not more common in view of the frequent chances for 
direct inoculation of the conjunctiva and cornea from foci in the nose, throat, 
and other mucous membranes cannot be stated, unless the conjunctiva 
enjoys a peculiar tissue immunity acquired especially in adult life, aided 
by the protection afforded by the tears. Minute injuries of the conjunctiva 
and cornea favor infection, but these preliminary breaks in the epithelial 
barrier are not essential for infection. Apparently some organisms are able 
to secure a position on the intact epithelium and by means of toxic products 
induce suppurative changes. In this connection the recent experiments 
of Brown and Pearce1 are of significance; these investigators have succeeded 
in producing syphilis of the eyes of rabbits by dropping Treponema pallidum 
into the apparently normal conjunctival sacs. 
Infections of the cornea—keratitis—may occur by direct implantation 
or by way of the blood-stream. Ulcers—usually pneumococcic or strepto- 
coccic—are probably exogenous infections because the cornea is avascular 
and embolic abscesses can occur only at the margin. However, the T. 
pallidum, tubercle, and lepra bacilli usually infect the cornea through the 
blood and lymphatic channels, producing either diffuse (parenchymatous) 
or punctate keratitis, and it is possible, but not probable, that organisms of 
the pyogenic group (pneumococci and streptococci) may be brought to the 
margin of the cornea in the blood-stream and infiltrate the tissues by way of 
the lymphatics. This is especially true in panopthalmitis. 
The cornea may also be involved from the rear, that is, by infection 
of the aqueous tumor from a pre-existing iritis or iridocyclitis. 
Infections of the iris, choroid, and retina are usually endogenous unless 
directly infected by injuries and operations. Panas originally believed 
that the toxins of bacteria rather than the bacteria themselves, were the 
primary etiologic agents. Most investigators at the present time, however, 
regard microbic infections of these tissues due to the presence of the microbes 
themselves; iritis especially is now known to bear a very important relation 
to focal infection. Streptococci, gonococci, and pneumococci from foci in 
the tonsils, teeth (apices), accessory nasal sinuses, urogenital organs, etc., 
are especially important in this relation. Experimental iritis has been 
produced by several investigators by the intravenous injection of bacteria 
in rabbits, but I do not know of the successful production of these lesions 
by injections of the toxic products of bacteria. However, the possibility 
of inflammation of the iris, choroid, and retina by toxins rather than by the 
bacteria themselves cannot be denied, and especially since we know that 
toxins may selectively locate in other tissues, notably the diphtheria toxins 
in the ganglia of the heart and tetanus toxins in the tissues of the spinal 
cord. The subject is a fertile field for further investigation. 
Why bacteria as the gonococcus, Diplococcus rheumaticus, streptococci, 
gonococci, etc., locate in the iris, choroid, and retina is not definitely known. 
Severe blood infections with these and other bacteria are not infrequent and 
the eye escapes. Indeed, it is the rule for these infections to develop insid- 
iously and one cannot escape the conviction that there is a strong possibility 
of infection by strains possessing a selective affinity for these tissues rather 
than infection on purely mechanical basis, as, for example, by embolic or 
1 Jour. Exper. Med., 1921, 34, 167. 1014 
VACCINES, SERA, BLOOD, PROTEINS IN TREATMENT OF DISEASE 
vascular occlusion. This is especially true of the retina, where the vessels 
are rather large and not at all favorable to embolism. 
For some unknown reason general blood or systemic infections from the 
eye are rare. Even in severe suppurative panopthalmitis blood-cultures 
are usually sterile. Fever, leukocytosis, etc., indicate the absorption of 
toxic substances in various eye infections, but blood infections with metas- 
tases to other organs are quite uncommon. The eye is not the primary 
seat of focal infection, indeed, it is usually the sufferer from primary foci 
situated elsewhere. 
The immunity of the eye is largely cellular and aided by mechanical 
factors. Of course, antibodies in the blood are brought in contact with the 
vascularized tissues, but the humors and avascular cornea do not share in 
the protection afforded by humoral immunity unless vascularity is increased 
by injury followed by increased exudation of serum from hyperemic vessels. 
Furthermore, the lymph is decidedly bactericidal and in this manner tends 
to prevent infections of the cornea and offer some resistance when infection 
has occurred. 
Serum Treatment of Pneumococcus Infections.—Axenfeld1 has stated 
that the subcutaneous injection of 10 to 20 c.c. of antipneumococcus serum 
is of value for prophylactic immunization prior to cataract operations when 
pneumococci are present on the conjunctiva. We now know that the 
pneumococci found on the normal conjunctiva are usually non-virulent 
and belong to Type IV. On the other hand, it is not possible to sterilize 
the conjunctival sac without doing some injury, although something may be 
accomplished in this direction by preliminary instillations of solutions of 
boric acid. In this connection I may state that solutions of mercurophen 
prepared by Schamberg, Raiziss and myself2 are strongly pneumococcidal 
in 1 : 3000 to 1 : 6000 and do not produce irritation in dilutions as low as 
1 : 500. This compound has proved particularly useful in connection with 
preoperative preparation of the eye as well as in the treatment of pneumo- 
coccus conjunctivitis. 
If Type I pneumococci were found in the conjunctiva and particularly 
if associated with inflammatory changes, and operation were imperative, as 
in glaucoma or in penetrating injuries, the subcutaneous or intramuscular 
injection of 20 to 30 c.c. of Type I serum would appear to be a useful pro- 
cedure for prophylactic immunization. 
In the treatment of pneumococcus infections of the cornea (ulcus ser- 
pens), iritis, and in panophthalmitis, pneumococcus serum may prove useful 
in Type I infections; the subcutaneous injection of 10 to 20 c.c. of serum, 
however, is useless, as stated by Axenfeld. If used at all, 50 to 100 c.c. should 
be injected intravenously. In pneumococcus conjunctivitis and dacryocys- 
titis serum therapy is not indicated. The most important pneumococcus 
lesion is corneal ulcer, and in severe Type I infections the intravenous 
injection of serum combined with local instillation may be of value. The 
instillation of 1 : 3000 to 1 : 6000 solutions of mercurophen is of aid in 
infections with all types of pneumococci. 
Serum Treatment in Streptococcus Infections.—Antistreptococcus serum 
has not been widely employed. Small doses (10 to 20 c.c.) by subcutaneous 
injection have been used in corneal ulcerations, but with negative results. 
In severe streptococcus infections, however, antistreptococcus serum may 
prove of value when given intravenously in doses of 50 to 100 c.c. 
1 Serumtherapie infektioses Augenkrankungen, Freiberg, 1905. (Full bibliography to 
early literature.) 
2 Jour. Infect. Dis., 1919, 24, 1547. TREATMENT OF DISEASES OF THE EYE 
1015 
Serum Treatment of Gonococcus Conjunctivitis.—The administration 
of antigonococcus serum may prove of aid in the treatment of both infantile 
and adult infections. For infants 10 c.c. of serum may be injected in the 
muscles of the buttocks when the infection is unusually severe. Corneal 
complications are less improved by serum than the conjunctivitis. In 
adults the serum should be given intravenously in doses of 50 to 100 c.c. 
Serum Treatment of Diphtheric Conjunctivitis and Keratitis.—In acute 
diphtheria of the eye antitoxin should be given in large doses (10,000 to 
20,000 units) by intramuscular injection; intravenous injection is prefer- 
able when conditions permit. Antitoxin may be instilled into the eye if 
a preservative free serum is available, otherwise the serum should be diluted 
with at least 3 parts of sterile saline solution to reduce the irritation caused 
by phenol or tricresol in the serum. 
The administration of serum usually aids greatly in the treatment of the 
conjunctivitis, but the keratitis is much less affected, due in large part to 
secondary infection with pyogenic organisms. 
Diphtheria antitoxin is of no value in Bacillus xerosis infections; on the 
other hand, these are usually chronic infections and serum therapy is not 
indicated. 
Serum Prophylaxis of Tetanus of the Eye.—Tetanus only rarely develops 
from injuries of the eye, but in penetrating wounds contaminated with earth 
and dirt the subcutaneous injection of 500 to 1000 units of antitoxin has 
been advocated as a prophylactic measure. 
Vaccine Treatment of Styes and Ulcerative Blepharitis.—These infec- 
tions are staphylococcic and usually due to Staphylococcus aureus. After 
correction of eye strain, if such exists, the administration of an autogenous 
or stock staphylococcus vaccine is indicated and has generally yielded good 
results as reported by Mayou,1 Medalia,2 and others. The vaccine may con- 
tain 1,000,000,000 cocci per cubic centimeter and treatment begun with 0.1 
c.c. by subcutaneous injection. Subsequent injections are given at intervals 
of five to seven days in increasing amounts until six to ten injections have 
been given. It is sometimes advisable to give an additional one or two 
injections several months later to reinforce immunity. 
Vaccine Treatment of Chronic Dacrocystitis.—These infections in my 
experience are usually pneumococcic, but staphylococcic and streptococcic 
infections occur. Autogenous vaccines are frequently helpful when em- 
ployed with local measures as reported by Bryan3 and others. The vaccine 
may contain 1,000,000,000 cocci per cubic centimeter and treatment begun 
with the subcutaneous injection of 0.1 c.c. Subsequent injections are given 
every five to seven days with increasing amounts. Local applications of 
1 : 3000 mercurophen are frequently very helpful. 
Treatment of Pollen Blennorrhea (Hay-fever).—This condition usually 
develops during the spring months due to sensitization to various spring 
pollens; also in August, due to the pollen of ragweed and other plants. 
Blennorrhea with itching and smarting pain may be among the first symp- 
toms. Cultures of the secretions usually show white staphylococci and 
other organisms of the normal conjunctiva. 
The diagnosis is made by skin tests; biologic treatment consists in 
desensitization by subcutaneous injections of extracts of the pollen or 
pollens responsible for the hay-fever. These methods have been previously 
described. 
1 Ann. Ophthal., 1912, 21, 669. 
2 Boston Med. and Surg. Jour., 1914, clxxi, 621. 
3 Brit. Med. Jour., March 16 and 23, 1912. 1016 
VACCINES, SERA, BLOOD, PROTEINS IN TREATMENT OF DISEASE 
Vernal catarrh may have a similar etiology and has been previously 
discussed. 
Treatment of Conjunctivitis.—In gonococcus, streptococcus, pneumo- 
coccus, and Morax-Axenfeld conjunctivitis vaccines have proved of aid in 
treatment according to the reports of Bryan, Medalia, Rubrecht,1 and others. 
Their administration would appear to be of particular value in the chronic 
infections and should be autogenous. In gonococcus conjunctivitis a 
stock vaccine is usually employed owing to the difficulty of securing the 
gonococcus in culture. 
The vaccine may contain 1,000,000,000 per cubic centimeter and treat- 
ment begun with 0.1 c.c. (100,000,000). Subsequent doses are given at 
intervals of five to seven days in increasing amounts. 
Intramuscular injections of 5 to 10 c.c. of milk, boiled for ten minutes and 
cooled, have been extensively used by European ophthalmologists, and espe- 
cially by v. Pflugh,2 Heineman and Wilke,3 Jendralski,4 Uddgren,5 and 
Jickeli.6 Berneaud7 observed particularly good results in gonorrheal con- 
junctivitis. Liebermann8 found that in gonorrheal infections the milk 
injections were followed by the lessening of the secretion after the first or 
second dose, with marked reduction of numbers of gonococci. The primary 
effect was a chemosis followed by rapid subsidence of inflammation. Corneal 
ulcers were prevented or arrested if already present. 
Klingmuller9 has employed subcutaneous injections of -a few drops of 
20 per cent, solutions of turpentine in sterile oil. 
In trachoma, Friedlander,10 Hiihn,11 Rosenstein,12 Berneaud, Jendralski, and 
others have reported beneficial results from intramuscular injections of 
milk; the reports are so favorable that it would appear that this therapy 
(5 to 10 c.c. of milk boiled for ten minutes and cooled) is worthy of trial 
not only in trachoma, but in follicular conjunctivitis as well. 
Treatment of Keratitis.—Autogenous vaccines may prove of some aid 
in the treatment of corneal ulcers, as reported by Mayou, Medalia, Rubre, 
and others; De Schweinitz13 has reported the successful treatment of a case 
of streptococcus hypopyon ulcer with autogenous vaccine; Medalia has 
reported the successful treatment of 12 cases. These infections are usually 
pneumococcic or streptococcic; staphylococcic and Bacillus xerosis are fre- 
quently secondary organisms. Ulceration of the cornea may also occur as 
a complication of conjunctivitis caused by the gonococcus and other or- 
ganisms. Owing to avascular conditions healing is a slowr process; if vac- 
cines are used, the usual local measures should not be neglected. 
In the treatment of parenchymatous or intestinal keratitis as well as in 
other forms, intramuscular injections of 5 to 10 c.c. of milk, boiled for ten 
minutes and cooled, have been used with marked success by Muller and 
Thanner,14 Jacovides, Pflugh, Jendralski, Uddgren, Jickeli, and others. 
Even in syphilitic keratitis it is believed that milk injections are of value 
in conjunction with antiluetic treatment. Darrier15 has used intramuscular 
1 Bull. d. 1. Soc. Beige d’Ophthal., 1909, 27, 82. 
2 Wchn. f. Hyg. in Therap. d. Anges, 1917, No. 41. 
3 Munch, med. Wchn., 1921, 68. 143. 
4 Berl. klin. Wchn., 1921, 58, 113. 
6 Milchinjection in der Opthalmologie, Stockholm, 1918. 
* Bull, of the Ophth. Soc., Egypt, 1919, 80. 
7 Miinch. med. Wchn., 1919, 66, 1040; Klin. Monatsch. f. Augenh., 1918, 61, 303; Berk 
klin. Wchn., 1919, 56, 887. 
8 Wien. med. Wchn., 1918, 68, No. 27. 12 Med. Klinik, 1917, 13, 185. 
9 Berl. klin. Wchn., 1918, 55, 896. 13 Therap. Gaz., October 15, 1910. 
10 Wien. klin. Wchn., 1916, 29, 1329. 14Ztschr. f. Augenh., 1917, 36, 305. 
'i Med. Klinik, 1916, 12, 1120. 18 Clin. Opth., 1919, 23, 559. TREATMENT OF DESEASES OF THE EYE 
1017 
injections of milk and horse-serum (usually diphtheria antitoxin) by mouth, 
the latter being a favored form of therapy among French clinicians. Jen- 
dralski has cautioned against milk injections if perforation of the cornea is 
threatened. In tuberculous keratitis subcutaneous injections of tuberculin 
(see next chapter) and intramuscular injections of milk have yielded par- 
ticularly good results in the hands of several investigators. Possek1 and 
Haab2 have employed subcutaneous injections of typhoid vaccine, the 
former using a phenol-killed preparation in dose of 500,000,000 repeated the 
following day. Particularly good results were reported in luetic (acquired 
and congenital) infections. 
Treatment of Iritis.—Autogenous vaccines have not been commonly 
employed in the treatment of iritis because cultures cannot be made unless 
iridectomy is conducted for the relief of glaucoma or for some other purpose. 
Autogenous vaccines have been made, however, of streptococci, pneumococci, 
and other organisms from the tonsils and apices of teeth in cases of suspected 
focal infection responsible for iritis. Stock vaccines of streptococci have 
been given in cases of rheumatic iritis and are said to give relief of pain and 
objective improvement in some cases; also stock vaccines in cases of sus- 
pected gonococcus iritis. Vaccine therapy has proved helpful in some 
cases as reported by Reber3 and others; if due to focal infection, care must 
be exercised in the search and treatment of the primary foci of infection. 
Injections are given subcutaneously, the first dose being 100,000,000 or 
0.1 c.c. of a vaccine containing 1,000,000,000 per cubic centimeter. Subse- 
quent injections are given at intervals of five to seven days in gradually 
increasing amounts. 
From the standpoint of biologic therapy most benefit follows the intra- 
muscular injection of 5 to 10 c.c. of milk, boiled for ten minutes and cooled, 
as first reported by Muller and Thanner.4 As a general rule several injec- 
tions are required at intervals of five or more days. Igersheimer,5 Kraupe,& 
Veach,7 v. Pflugh, Bernaud, Jendralski, Uddgren, Darrier, Jickeli, and others 
have rendered favorable reports on this method of treatment and it appears 
well worthy of trial especially in chronic cases. In syphilitic iritis antiluetic 
treatment should be given and is usually sufficient; in tuberculous in- 
fections biologic therapy consists of the injection of both tuberculin and 
milk. 
Vaccine Treatment of Infected Perforating Wounds of the Cornea and 
Infections Following Cataract Extractions.—Maddox,8 Mayou, Bryan, 
Madalia, and others have reported that autogenous vaccines have proved 
helpful in the treatment of infected wounds of the cornea due to perforating 
injuries and cataract operations. Cultures should be made on blood-agar;, 
staphylococci, streptococci, and pneumococci are the usual agents either 
alone or in combination. The vaccine may contain 1,000,000,000 per cubic 
centimeter and the first dose 0.1 c.c. by subcutaneous injection. Subsequent 
injections may be at intervals of five to seven days in gradually increasing 
amounts. Medalia believes that in low-grade conjunctival infections 
preliminary injections of autogenous vaccine may be worth while preliminary 
to cataract operations or simple iridectomy. For this purpose the vaccine 
may be prepared in the same strength and given in four doses at intervals 
of three to four days in doses of 0.1, 0.2, 0.4, and 1 c.c., operation being, 
done about seven to ten days after the last injection. 
1 Wien. klin. Wchn., 1919, 32, 753. 
2 Munch, med. Wchn., 1918, 65, 24. 
3 Ophth. Rec., 1916, 25, 225. 
4 Ophth. Rec., 1916, 25, 225. 
5Therap. Halbmonatsh., 1921, 35, 104. 
6 Ztschr. f. Augenh., 1919, xlii, 105. 
7 Amer. Jour. Ophth., 1920, 3, 93. 
8 Ophthalmoscope, June, 1908. 1018 
VACCINES, SERA, BLOOD, PROTEINS IN TREATMENT OF DISEASE 
Treatment of Iridocyclitis; Choroiditis; Scleritis; Albuminuric Retinitis. 
—Berneaud, Jickeli, Uddgren, Jendralski, and others have reported beneficial 
effects following the intramuscular injection of milk in the treatment of 
these affections. Heine1 treated 17 cases of albuminuric retinitis with 
injections of 5 to 10 c.c. of milk; vision was improved in 11 cases, whereas 
in 4 the disease process was retarded. In 2, however, the disease progressed 
in spite of the injections. 
Treatment of Retinal Hemorrhage.—In the treatment of retinal hemor- 
rhage 10 to 20 c.c. of horse-serum is generally injected subcutaneously or 
intramuscularly. It is better, however, to inject fresh sterile human serum 
because anaphylactic sensitization and reactions are not produced. In 
my experience best results have been secured by the intramuscular injection 
of 20 to 30 c.c. of defibrinated human blood. It is not necessary to type the 
bloods of donor and recipient unless blood is injected intravenously. The 
technic of these injections is very simple and is described in Chapter 
XXXVII. Owing to the stypic action of milk 5 to 10 c.c., boiled for ten 
minutes and cooled, may be injected intramuscularly, but injections of blood 
produce less discomfort and better results. 
TREATMENT OF HEMORRHAGE AND ANEMIA 
Serum and Blood in the Treatment of Melena Neonatorum and Hemo- 
philia.—Subcutaneous and intramuscular injections of normal horse-, rabbit-, 
or human serum have been widely and successfully employed in the treat- 
ment of melena and hemophilia; Welch,2 Reichard,3 Perkins,4 Claybrook,6 
Franz,6 Koch and Klein,7 Myers,8 Freeman,9 and others have reported 
successful results. In melena Berghausen10 has injected citrated human 
blood by way of the superior longitudinal sinus. 
The effects of the serum injections are probably due to the alterations 
in the amount of fibrinogen and thrombokinase that follow injections of 
various non-specific agents, as reported by Moll,11 Modrakowski, and Orator.12 
Wohlegmut13 states that the increase in fibrinogen probably results from 
liver stimulation, but that the thrombokinase arises elsewhere. 
The dose of horse-serum for children is ordinarily 10 to 20 c.c. by sub- 
cutaneous or intramuscular injection; for adults 20 to 30 c.c. are ordinarily 
administered. Subsequent injections are given if hemorrhage continues 
or commences at a later period. 
Human serum is preferable to horse-serum because anaphylactic sensi- 
tization does not occur. The serum is collected by securing blood from a 
healthy adult, allowing it to coagulate, and separating the serum. The 
technic is described in Chapter XXXVII. It is not necessary to type 
the bloods of donor and recipient if the injections are given subcutaneously 
or intramuscularly. The doses are the same as given above. 
1 Munch, med. Wchn., 1920. 57, 1221. 
2 Amer. Jour. Med. Sci., 1910, 139, 800. 
3 Jour. Amer. Med. Assoc., 1912, 59, 1539. 
4 Jour. Amer. Med. Assoc., 1912, 59, 1539. 
6 Jour. Amer. Med. Assoc., 1912, 59, 1540. 
6 Munch, med. Wchn., 1913, 59, 2905. 
7 Gynak. Rundsch., 1912, 6, 597. 
3 Arch. Pediat., 1912, 29, No. 3. 
9 Amer. Jour. Obstet., 1917, 76, 354. 
10 Jour. Amer. Med. Assoc., 1918, 70, 514. 
11 Bert. z. chem. Phys. u. Path., 1904, 4, 578. 
12 Wien. klin. Wchn., 1917, 30, 1093. 
13 Berl. klin. Wchn., 1917, 54, 87. TREATMENT OF HEMORRHAGE AND ANEMIA 
Whole human blood is probably best of all because these injections furnish 
not only serum constituents, but corpuscular elements as well. For sub- 
cutaneous and intramuscular injection a 20 c.c. syringeful of blood may be 
taken from a vein of an adult and before coagulation occurs the syringe is 
emptied in the gluteal muscles of the patient. Or 20 c.c. of blood may be 
collected in a sterile flask containing 2 c.c. of 10 per cent, solution of sodium 
citrate and the mixture injected subcutaneously or intramuscularly. The 
blood may also be collected in a small sterile flask containing glass beads, 
whipped about for defibrination, and subsequently injected. These methods 
are described in more detail in Chapter XXXVII. Blood transfusion has 
also been employed as described in Chapter XLII. 
When a large amount of blood has been lost it may be necessary not only to 
stop bleeding, but restore blood volume; in this case blood transfusion 
should be given. In newborn infants the transfusion is readily conducted 
through the superior longitudinal sinus. Methods are described in Chapter 
XLII. 
Treatment of Hemorrhage in Disease and After Operations.—Hemor- 
rhagic retinitis, intestinal bleeding in typhoid fever and in connection with 
cirrhosis of the liver, in pulmonary tuberculosis, and in some cases of uterine 
disease have been successfully treated by subcutaneous or intramuscular 
injections of horse-serum, human serum, or whole human blood. The 
indications for this therapy are especially clear when the coagulation time 
of the blood is below normal. Intramuscular injections of citrated or 
defibrinated human blood are especially efficacious; the injections should be 
repeated at intervals of four to six hours until bleeding is checked. Blood 
transfusion has also been employed (see Chapter XLIII). 
Kronheimer1 has used subcutaneous injections of human serum in dose 
of 5 c.c. for parenchymatous hemorrhage in large septic wounds; Vonchen2 
has employed injections of whole human blood in the successful treatment 
of a case of persistent bleeding after liver injury. Meyer3 has employed 
subcutaneous injections of human serum in the treatment of hemorrhage of 
icteric patients after operations; Thiesen4 has used injections of serum as a 
preventive of hemorrhage in nose and throat operations when the coagula- 
tion time has been found below normal. Barringer5 has reported the suc- 
cessful treatment by injections of fresh normal human serum of unilateral 
kidney hemorrhage in a hemophiliac, and advises that this simple treatment 
should be tried in similar cases of varicose veins of the renal papilla. Levi- 
son6 has reported the successful checking of hemorrhage from the urinary 
bladder following a simple operation by performing cystostomy, removing 
clots, and filling the bladder with sterile horse-serum. 
For the prevention of hemorrhage in operations upon icteric individuals 
or those having a lowered coagulation time of the blood for any reason, 
20 to 30 c.c. of defibrinated human blood may be injected into the gluteal 
muscles about three hours before operation. Or the blood may be held in 
readiness and the injection given after the operation if bleeding occurs. 
A second injection should be given within four to six hours if bleeding is not 
checked. 
Treatment of Purpura Hsemorrhagica.—Subcutaneous and intramus- 
cular injections of horse- or human sera have been used successfully in the 
1019 
1 Munch, med. Wchn., 1914, 62, No. 1. 
2 Jour. Amer. Med. Assoc., 1920, 75, 307. 
3 Surg., Gyn., and Obst., 1912, 13, 152. 
4 New York Med. Jour., 1914, 100, 849. 
5 Jour. Amer. Med. Assoc., 1912, 59, 1538. 
6 Jour. Amer. Med. Assoc., 1913, 60, 721. 1020 
VACCINES, SERA, BLOOD, PROTEINS IN TREATMENT OF DISEASE 
treatment of purpura, as reported by Class,1 Moss,2 Bodenheimer,3 and 
others. Whole human blood has been used by Emsheimer,4 Cohen,5 and 
others. 
Human serum and defibrinated human blood by intramuscular injection 
are to be preferred. For adults the dose may be 20 to 30 c.c. and repeated 
if necessary. Blood transfusion has also been employed (see Chapter XLII). 
Dollken6 assumes that bleeding in purpura is largely due to increased 
fragility of the blood-vessels and alteration in the coagulability of the blood. 
He has employed intramuscular injections of 5 c.c. of milk, boiled for ten 
minutes and cooled, every three days, with very successful results. Bleeding 
into the tissues usually stopped within five hours after injection and all cases 
having albuminuria before hand showed clear, albumin-free urine after the 
injections. A moderate leukocytosis was observed, but no changes in the 
platelets. 
Treatment of Anemia.—Severe secondary anemia as a result of hemor- 
rhage following injury or operation may be treated with the transfusion of 
blood. Secondary anemia due to repeated losses of small amounts of blood 
may be benefited by intramuscular or subcutaneous injections of normal 
human serum or defibrinated human blood; Miiller7 has employed intra- 
muscular injections of 5 to 10 c.c. of milk, boiled for ten minutes and cooled, 
because of its well-known styptic effects. Of course most attention should 
be given the treatment of the cause for bleeding or blood destruction if 
it can be found and reached therapeutically. 
In pernicious anemia blood transfusion usually gives temporary relief 
and must be repeated at intervals, as discussed in more detail in Chapter 
XLII. Grote8 has employed intramuscular injections of 5 to 10 c.c. of 
milk, with temporary improvement. 
The etiology of the disease is unknown, although very probably a microbic 
infection. Until the cause is discovered specific serum therapy is not 
possible, although the subcutaneous or intravenous injection of horse-serum 
will probably bring about temporary improvement by non-specific agencies 
as has been found for milk and vegetable proteins. 
TREATMENT OF ACUTE ANTERIOR POLIOMYELITIS 
Infection and Immunity in Acute Anterior Poliomyelitis.—Studies con- 
ducted during the great epidemic of this disease in 1916 have added con- 
siderably to our knowledge of its cause and immunity. 
The specific virus is filtrable and infection probably occurs through the 
upper respiratory tract, although Flexner and his associates have succeeded 
in infecting monkeys not only by application of the virus to the nasal mucous 
membrane, but by intracerebral, intravenous, intraperitoneal and sub- 
cutaneous injection, and by the oral administration of large doses as well. 
The first successful transmission of the disease was in 1909 by Land- 
steiner and Popper,9 who succeeded in transmitting the disease to monkeys 
by inoculating them, intraperitoneally, with the spinal cord of a child who 
died of poliomyelitis, but they did not succeed in transmitting the infection 
from monkey to monkey, probably because they used too mild a case. 
Later, in 1909, Flexner and Lewis10 obtained the same result and further 
1 Arch. Int. Med., 1906, 6, 170. 
2 John Hopkins Hosp. Bull., 1911, 22, 272. 
3 New York Med. Jour., 1914, 100, 745. 
4 Jour. Amer. Med. Assoc., 1916, 65, 20. 
5 Arch. med. beiges, 1920, 73, 988. 
6 Berl. klin. Wchn., 1919, 56, 226. 
7 Deutsch. med. Wchn., 1919, 14, 323. 
8 Miinch. med. Wchn., 1919, 66, 307. 
9Ztschr. f. Immunitatsf., 1909, 11, 377. 
10 Jour. Amer. Med. Assoc., 1909, 53, 1639, 1913, 2095. TREATMENT OF ACUTE ANTERIOR POLIOMYELITIS 
1021 
transmitted the infection from monkey to monkey through an indefinite 
number of passages. Landsteiner and Levaditi1 in 1909 also transmitted 
the disease from monkey to monkey and found that the virus remained 
virulent for some time outside of the body; that the degenerated nerve-cells 
are taken up by phagocytes, and that there is an analogy between the lesions 
of poliomyelitis and those produced by rabies. They also demonstrated 
that the virus is filtrable. Leiner and Weisner2 transmitted the infection 
from monkey to monkey, and found that young animals were more sus- 
ceptible to infection than older ones, and that the spinal fluid, blood, and 
spleen were negative. Flexner and Lewis3 transmitted the disease by 
inoculating very large amounts into the blood or peritoneal cavity, also by 
the subcutaneous method, and independently found the virus to be filtrable. 
Landsteiner and Levaditi4 found the virus in the salivary glands, and sug- 
gested the saliva, moist or dry, as a source of infection. They also found 
the spinal fluid negative in monkeys dying from artificial infection. 
Soon after this, in 1913, Noguchi and Flexner5 announced that they had 
obtained cultures in media similar to the medium used by Noguchi in culti- 
vating spirochetes. In such media in about five days the pieces of tissue 
employed become surrounded by an opalescent haze which increases for 
five days more, and a sediment gradually forms. Giemsa’s stain shows the 
presence of minute globoid bodies (0.15 to 0.13 microm. diam.) in pairs, 
short chains, and masses. Cultures were also obtained from the filtered 
virus. Monkeys inoculated with these cultures for a variable number of 
culture generations may die with typical lesions of the disease. The authors 
consider these bodies the cause of the disease. They further report that 
the cultures are filtrable through coarse filters. 
Earlier bacteriologic studies discovered various micrococci in the tissues 
of the spinal cord and brain, sometimes in the spinal fluid. Bulow-Hansen 
and Harbitz6 and Harbitz and Scheel,7,8 found a diplococcus in the cerebro- 
spinal fluid of several cases of acute anterior poliomyeltis and referred to the 
work of Geirsvold,9 who found a diplococcus in the cerebrospinal fluid of 
12 cases and claimed to have produced paralysis and death in experimental 
animals with them. Pasteur, Foulerton, and Maccormac10 reported the 
discovery of a micrococcus in the cerebrospinal fluid of a case during 
life, which produced in rabbits symptoms resembling the disease in human 
subjects. Leiner and von Wiesner11 and Krause and Meinicke12 reported 
that these micrococci were not the etiologic agents of acute poliomyelitis, 
as had Flexner and Lewis, who first showed that the etiologic agent was 
filtrable through dense filters and probably belonged to the filtrable 
viruses13; simultaneous and similar results were observed and reported by 
Landsteiner and Levaditi.14 Dixon, Fox, and Rucker15 also found a diplo- 
1 Compt. rend. Soc. de biol., 1909, 62, 592. 
2 Wiener klin. Wchn., 1909, 22, 1698. 
3 Jour. Amer. Med. Assoc., 1909, 53, 1639, 1913, 2095. 
4 Compt. rend. Soc. de biol, 1909, 67, 787. 
6 Jour. Exp. Med., 1913, 18, 461. 
6 Norsk. Mag. f. Laegevid., 1898, 13, 1170. 
7 Jour. Amer. Med. Assoc., 1908, 1, 281. 
8 Jour. Amer. Med. Assoc., 1907, 49, 1420. 
9 Norsk. Mag. Loegevid., 1905, 3, 1280. 
10 Lancet, 1908, 1, 484. 
11 Wien. klin. Wchn., 1910, 23, 817. 
12 Deutsch. med. Wchn., 1909, 35, 1825. 
13 Jour. Amer. Med. Assoc., 1909, 43, 2095. 
14 Compt. rend. Soc. de Biol., 1909, 57, 592. 
15 Rep. Comm. Health. Penn., 1907, 420. 1022 
VACCINES, SERA, BLOOD, PROTEINS IN TREATMENT OF DISEASE 
coccus in the cerebrospinal fluid, nose, and throat of patients with acute an- 
terior poliomyelitis, and Rucker made a very complete study of this diplococ- 
cus which was apparently identical with that recently described by Mathers1 
and Nuzum and Herzog.2 While the latter have reported the successful 
infection of various laboratory animals with these cocci and the reproduction 
of a disease with clinical symptoms and lesions similar to those of acute 
poliomyelitis, the experiments of the former were negative throughout; an 
inoculated monkey succumbed with a hemorrhage meningitis, but without 
clinical or histologic evidences of anterior poliomyelitis. 
Rosenow, Towne, and Wheeler3 recovered what they called a “peculiar 
polymorphous streptococcus” from the tonsils, brain, cord, mesenteric 
lymph-glands, and once from the blood, but never from the cerebrospinal 
fluid of cases of acute poliomyelitis, which produced lesions and symptoms 
among the lower animals regarded as those of acute poliomyelitis. He has 
described changes in size and staining reaction of these streptococci according 
to the culture-medium employed, the age of the culture, and whether they 
have been grown aerobically or anaerobically. The cocci were said to 
become very small under anaerobic conditions and to approach in size the 
globoid bodies described by Flexner and Noguchi; the small forms were 
found to be filtrable through Berkefeld filters, while the larger forms were 
not. 
There is, therefore, two schools of thought regarding the etiology of 
epidemic poliomyelitis: one regards the globoid bodies as the primary 
etiologic agent of the disease and most experimental evidence is in favor of 
this view. The second regards the stretpococci and micrococci found in the 
tissues as the primary agents. 
Brown, Freese, and myself4 found streptococci, diplococci, and other 
organisms not only in the spinal cord and brain of poliomyelitic cases, but 
in some of the internal organs as well. We were never able to produce the 
disease experimentally with these cultures. Heist, Cohen, and myself5 
succeeded in cultivating the globoid bodies and believe them distinct from 
the aerobic micrococci. The writer is of the opinion that the globoid bodies 
are the cause of epidemic poliomyelitis and that the disease favors secondary 
invasiQn and infection with streptococci, which may possess sufficient 
virulence to add to pathologic lesions just as streptococci do in scarlet fever. 
Whether or not natural immunity to the disease exists is not known;, 
certainly normal human serum cannot neutralize the virus in viiro or protect 
the experimental animal against infection, as does the sera of human con- 
valescents from the disease. Children during the first dentition are the 
most frequent sufferers, but in times of epidemics adults may become in- 
fected. It is highly probable that the majority of children are susceptible, 
but may escape infection by fortuitous circumstances. Some degree of 
natural immunity probably develops with adult age, but it is not humoral, 
that is, neutralizing principles are not present in the blood. 
Individuals recovering from the disease are usually immune for life, 
but second attacks have occurred. The serum of recovered individuals 
contains immunity principles capable of killing the virus in the test tube 
under proper conditions and of protecting monkeys against the disease 
when injected intraspinally. Convalescent serum is, therefore, of value 
in the treatment of the disease. Complement-fixing antibodies and opsonins 
have been found in convalescent sera by Heist, Cohen, and myself. Mathers 
1 Jour. Amer. Med. Assoc., 1916, 67, 1019. 
2 Jour. Amer. Med. Assoc., 1916, 67, 1205. 
3 Jour. Amer. Med. Assoc., 1916, 67, 1202. 
4 Jour. Exper. Med., 1917, 25, 789. 
6 Jour. Infect. Dis., 1918, 22. TREATMENT OF ACUTE ANTERIOR POLIOMYELITIS 
1023 
and Tunnicliff,1 Heist, Cohen, and myself,2 Rosenow and Gray3 also found 
immune opsonins and other specific antibodies in convalescent sera for the 
streptococci of Rosenow and Mathers; this indicates that these micrococci 
are not always saprophytic or agonal, but probably possess sufficient viru- 
lence to stimulate antibody production and possibly add to the pathologic 
changes of the disease. 
Vaccines in the Treatment of Poliomyelitis.—In Chapter XXXV men- 
tion has been made of attempts toward evolving a successful method of 
prophylactic vaccination along lines similar to vaccination against rabies; 
vaccines have not been employed in the treatment of the disease of human 
beings and have generally failed to modify the course of the disease in 
monkeys. 
Treatment of Epidemic Poliomyelitis with Convalescent Serum.—The 
discovery of the filtrable nature of the micro-organism or virus causing the 
disease was quickly followed by the observation by Flexner and Lewis4 that 
recovery from an attack of experimental poliomyelitis afforded protection to 
a second inoculation; and this, in turn, was followed by the detection of 
immunity or neutralizing substances in the blood-serum, first of recovered 
monkeys and then of recovered human beings by Levaditi and Landsteiner,6 
Romer and Joseph,6 Flexner and Lewis,7 and Anderson and Frost.8 
Since recovery from an attack of poliomyelitis was obviously brought about 
through a process of immunization similar to that in other infectious diseases, 
Flexner and Lewis9 endeavored to prevent the development of the infection 
in inoculated monkeys through the administration of blood-serum taken 
(a) from recovered monkeys and (b) from recovered human beings. The 
results, while not constant and regular, were definite. 
These experimental results were at once utilized as a basis of a serum 
therapy in man by Netter and his associates,10 who have reported a total 
of 34 cases of acute poliomyelitis which they have treated by the subdural 
method of injecting immune serum. They have undoubtedly established 
the fact that, as in the monkey, subdural injections intelligently carried 
out in man are safe. They believe, further, that they have proved them 
to be definitely beneficial or curative. They became increasingly con- 
vinced that the period of the disease at which the injections were made 
counted vitally, and they urged that the injections should be made as early 
as possible in the course of the infection. Further reports by Sophian,11 
Alfaro and Hitce,12 Wells,13 Le Boutillier,14 Petty,15 Zingher,16 Draper,17 Acuna,18 
and Amoss and Chesney19 indicate that serum taken from recently recovered 
1 Jour. Amer. Med. Assoc., 1916, 67, 1935. 
2 New York Med. Jour., September 1, 1917. 
3 Jour. Infect. Dis., 1918, 22, 345. 
4 Jour. Amer. Med. Assoc., 1910, liv, 45. 
5 Compt. rend. Soc. de biol., 1910, lxviii, 311. 
6 Munch, med. Woch., 1910, lvii, 568. 
7 Jour. Amer. Med. Assoc., 1910, liv, 1790. 
8 Jour. Amer. Med. Assoc., 1911, lvi, 663. 
9 Jour. Amer. Med. Assoc., 1910, liv, 1780. 
10 Compt. rend. Soc. de biol., 1911, lxx, 625; Bull. Acad, med., 1915, lxxiv, series 3, 403; 
Bull, et mem. Soc. med. hop., Paris, 1916, xl, series 3, 299. 
11 Jour. Amer. Med. Assoc., 1916, lxvii, 426. 
12 Semaine med., 1915, xxii, 211. 
13 Jour. Amer. Med. Assoc., 1916, lxvii, 1211. 
14 Amer. Jour. Med. Sci., 1917, 153, 188. 
15 New York Med. Jour., December 16, 1916. 
16 Jour. Amer. Med. Assoc., 1917, 68, 817. 
17 Jour. Amer. Med. Assoc., 1917, 68, 1153. 
18 Arch. Latino-Amer. d. Ped., 1918, 12, 1. 19 Jour. Exper. Med., 1917, xxv, 581. 1024 
VACCINES, SERA, BLOOD, PROTEINS IN TREATMENT OF DISEASE 
cases of poliomyelitis may be employed in its treatment and probably yields 
the best results. The earlier in the course of the disease the serum is em- 
ployed in suitable doses, the more promise there is of benefit. The action 
of the serum appears to be more precise and definite in arresting paralysis 
than in rapidly bringing about its retrogression. 
Physicians have usually administered the serum by intraspinal injec- 
tion, but the experi rents of Flexner and Amoss1 ascertained that an aseptic 
meningitis set up by an intraspinal injection of any serum permits the 
passage of im lunity principles from the blood into the cerebrospinal fluid. 
Therefore the intravenous injection of serum may influence the course of the 
disease. In the serum treatment of epidemic poliomyelitis Amoss and 
Chesney state that the following conditions should be observed: (1) Early 
and prompt diagnosis and treatment; (2) intraspinal injection of immune 
serum; (3) intravenous or intramuscular injection of immune serum; (4) 
the serum employed should be collected from cases which have recently 
passed through an attack of poliomyelitis, as it is to be supposed that the 
serum will contain a greater amount of immune principles in this early 
period than after the lapse of many years. The use of serum from recently 
recovered persons otherwise healthy involves no risk in transferring the 
micro-organism of poliomyelitis, as the virus has never been detected in 
the circulating blood of human beings even in the first days of the disease. 
Blood should be collected under aseptic precautions and the serum 
separated, submitted to the Wassermann test, cultured for sterility, and 
kept in a cold place, preferably without the addition of a preservative. 
From‘an adult in good condition oroinarily 500 to 700 c.c. of blood may be 
removed; from children of twelve years about 200 to 250 c.c. The addition 
of 0.2 per cent, tricresol does not impair the curative power. In children 
10 to 20 c.c. of serum may be injected intraspinally, and 40 to 80 c.c. intra- 
muscularly or intravenously. Adults should receive larger doses. If 
given before the onset of paralysis one injection may be sufficient; in cases 
in which some degree of paralysis develops soon after the injection, rein- 
jection twelve to twenty-four hours later may be advantageous. 
Treatment of Epidemic Poliomyelitis with Other Sera.—The disease 
has been also treated with intraspinal injections of normal human serum 
and normal horse-serum (or horse-serum diphtheria antitoxin). Some 
physicians believed that good results were obtained, but the New York 
Department of Health2 observed negative results, and Flexner and Amoss 
have found that these sera not only failed to protect monkeys against the 
disease, but actually increased their susceptibility to infection by favoring 
the passage of the virus from the blood to the tissues of the spinal cord. 
These sera should not be employed in treatment. 
Rosenow3 and Nuzum and Willy4 have prepared immune sera by the 
immunization of horses with the streptococci and other micrococci from 
fatal cases of poliomyelitis. Rosenow believes that the intraspinal injection 
of this serum reduced the mortality of the disease from about 35 to approxi- 
mately 6 per cent. Nuzum believed that treatment with his serum reduced 
the mortality from about 32 to 7.5 per cent. Nuzum found that his immune 
serum neutralized the poliomyelitic virus, but Amoss and Eberson5 were 
1 Jour. Exper. Med., 1914, xx, 249; ibid., 1917, xxv, 449. 
2 Monograph on Poliomyelitis, 1917, 245. 
3 Jour. Amer. Med. Assoc., 1917, 69, 261; ibid., 1917, 69, 1074; Jour. Infect. Dis., 1918, 
22 379. 
4 Jour. Infect. Dis., 1918, 22, 301, 258; Jour. Amer. Med. Assoc., 1917, 69, 1247. 
5 Jour. Exp. Med., 1918, 28, 323. TREATMENT OF MALTA FEVER 
1025 
unable to confirm these results and believe that it is inadvisable to employ 
these sera in the treatment of the disease. 
Tsen1 has endeavored to produce an immune serum in rabbits by im- 
lunization with poliomyelitic virus, but with negative results. Up to the 
present time an immune serum for the virus prepared by the immunization 
of horses with emulsions of poliomyelitic brain and spinal cord or cultures 
of the globoid bodies, has not been successfully accomplished. 
In so far as the serum treatment of epidemic poliomyelitis is concerned, it 
would appear that the proper procedure is the intraspinal injection of 10 to 
30 c.c. of human convalescent serum and the intravenous injection of 40 to 100 
c.c. of the same serum, as soon as possible. After these sera have been adminis- 
tered Rosenow serum may be injected intraspinally in dose of 10 to 30 c.c. for 
the treatment of the secondary streptococcus infection. In the absence of human 
convalescent serum, the antistreptococcus serum alone may be employed, but 
normal human and normal horse-sera should not be administered. 
TREATMENT OF LETHARGIC ENCEPHALITIS 
Since lethargic encephalitis was first described by Von Economo2 opin- 
ions have varied in regard to its relation to epidemic poliomyelitis. Strauss, 
Hirshfeld and Loewe3 have demonstrated that the disease is due to a filtrable 
virus, and Loewe and Strauss4 have succeeded in cultivating a small anaerobic 
coccus resembling the globoid bodies of poliomyelitis. These investigators 
regard the disease as etiologically and clinically distinct from poliomyelitis. 
Amoss5 found that the sera of recently recovered human cases of encephalitis 
were unable to neutralize the virus of poliomyelitis—and regards the two 
diseases as separate entities on the basis of these and other laboratory 
experiments and clinical findings; Neustaedter, Larkin and Banzhaf,6 
however, report opposite results, that is, that monkeys were protected 
against poliomyelitis by the administration of sera from human convalescent 
cases of encephalitis. The subject, therefore, is undecided. 
Grunwald7 has injected 80 to 100 c.c. of convalescent serum intraglu- 
teally in the treatment of the disease, with apparently some success. If 
sufficient serum is available it would appear advisable to administer it intra- 
spinally and intravenously in the same manner as described for the treatment 
of poliomyelitis. 
Brill8 reports the successful treatment of several cases by the intraspinal 
injection of 25 to 35 c.c. of'the patient’s own unheated serum. 
Laubie9 reports the rapid recovery of patients following intraspinal 
injection of tetanus antitoxin, but this is virtually normal horse-serum 
in so far as encephalitis is concerned, and the method cannot be endorsed 
unless normal horse-serum is shown to possess neutralizing principles for 
the virus, which so far has not been accomplished. 
TREATMENT OF MALTA FEVER 
Immunity in Malta Fever.—Malta fever or undulant fever is a typhoid- 
like infection caused by the Micrococcus melilensis. The disease is found in 
1 Jour. Immunology, 1918, 3, 213. 
2 Wien. klin. Wchn., 1917, 30, 581. 
3 New York Med. Jour., 1919, cix, 772; Jour. Infect. Dis., 1919, 25, 378. 
4 Jour. Amer. Med. Assoc., 1919, 73, 1056; Proc. New York Path. Soc., 1920, 20, 18. 
5 Jour. Exper. Med., 1921, 33, 187. 
6 Amer. Jour. Med. Sci., 1921, 162, 715. 8 Med. Record, 1920, 97, 1079. 
7 Deutsch. med. Wchn., 1920, xlvi, 45. 9 Bull, de l’Acad. de med., 1920, 83, 246. 1026 
VACCINES, SERA, BLOOD, PROTEINS IN TREATMENT OF DISEASE 
the countries of the Mediterranean basin, India, China, Philippine Islands, 
and in some parts of America (Texas). 
The majority of human beings are probably susceptible, and antibodies 
(opsonins and lysins) are not ordinarily found in normal sera. The majority 
of the lower animals are naturally immune. Monkeys, however, are quite 
susceptible, and may be infected by subcutaneous and oral administration 
of the organisms; guinea-pigs and rabbits are less easily infected. Goats 
suffer with the disease and pass the organisms in the milk, urine, and feces; 
or their milk becomes contaminated by feces and constitutes an important 
means of spread of the disease. The organism exhibits a peculiar affinity 
for the spleen, lymph-nodes, liver, and sex organs, and may persist in these 
tissues for two months. By some bacteriologists the organism is regarded 
as a bacillus rather than a coccus, and Meyer, Shaw, and Fleischner1 have 
recently shown that the lesions produced in the guinea-pig are quite similar 
to those produced experimentally by Bacillus abortus. 
During an attack of Malta fever in both man and the lower animals, 
agglutinins, opsonins, bacteriolysins, and complement-fixing antibodies are 
produced and are useful in the serologic diagnosis of the disease. 
The immunity is very similar to the processes concerned in the resist- 
ance and immunity of typhoid fever. One attack usually protects for the 
balance of life, but second and even third attacks have been known to occur. 
Vaccine and Serum Treatment.—Vaccines of Micrococcus or Bacillus 
melitensis have been used by Eyre2 for prophylactic purposes, but only on 
a small scale and with inconclusive results. A serum prepared by the 
immunization of horses has also been employed in the treatment of human 
and experimental monkey infections, but with indifferent results. 
Owen and Newham3 have reported beneficial effects from the treatment 
of the disease with a vaccine. 
Owing to the usual mildness of the infection with a mortality of only 
2 per cent., specific biologic therapy has not been extensively employed, 
but I surmise that results similar to those observed in typhoid fever may be 
obtained from the employment of vaccines for prophylactic and curative 
purposes. 
TREATMENT OF TUMORS 
Immunity in Tumors in Relation to Biologic Therapy.—One result of 
the discovery of spontaneous malignant tumors in mice, rats, and other 
lower animals and their transmissibility to normal animals of the same 
species by injecting bits of tumor tissue has been a very large amount of 
investigation on the subjects of natural and acquired immunity to these 
growths. Just how much of the knowledge gained by these investigations 
is applicable to tumors of human beings, and especially carcinomata and 
sarcomata, is difficult to state, but it is highly probable that the conditions 
initiating and governing neoplasms among the lower animals are similar to 
those operative among human beings. 
Examples of natural immunity to neoplasms are frequent and striking. 
Some mice cannot be inoculated with a tumor easily transmissible to other 
mice of the same species; mouse tumors may not be transmitted to rats, and 
the reverse. Tumors of one species of mice may not be transmitted to mice 
of a different species. A mouse immune to one transmissible tumor may be 
susceptible to another tumor. Similar transmission or inoculation experi- 
ments, of course, cannot be made among human beings, but it is entirely 
1 Jour. Infect. Dis., 1922, 31, 159. 
2 Kolle and Wassermann’s Handbuch, 1913, 4, 421. 
3 Lancet, 1915, 2, 529. TREATMENT OF TUMORS 
1027 
likely that a similar state of affairs exists, and that some individuals would 
be found with a high natural resistance just as other human beings may 
apparently inherit a predisposition, as shown by the occurrence of cancer 
among individuals of the same family stock. 
Sex has no bearing upon natural immunity to neoplasms among the 
lower animals; among human beings, however, sex does have some influence, 
in so far as the incidence of carcinoma is concerned, being more frequent 
among women than men. Age is an important factor among human beings, 
carcinoma being most frequent after thirty years of age, although sarcoma 
and benign growdhs may occur at any age. Among the lower animals, 
however, age has little or no influence; Loeb,1 Bashford and Murray,2 
Buschke,3 Ehrlich and Apolant,4 and others found mouse tumors transmissi- 
ble to both young and old animals, although half-grown animals are probably 
best for experimental work. It is probable that the influence of age of human 
beings upon the incidence of cancer is more a matter of influencing etiologic 
factors than actual natural resistance. 
As previously stated, race has a great influence upon the transmission 
of tumors among the lower animals; whether a similar state of affairs exists 
among human beings cannot be stated. Neoplasms occur, however, among 
all races of human beings. 
Pregnancy exerts a restraining influence upon tumor growths in the 
lowrer animals according to Haaland,5 Uhlenhuth and Weidanz,6 and others; 
Herzog,7 on the other hand, thought that pregnancy enhanced tumor growth 
among mice. Fichera8 explained these discrepancies with the statement 
that when several embryos were present, specific food-stuffs were absorbed 
by them, leaving but little for the tumor, whereas when few embryos were 
present, sufficient specific pabulum remained for the nourishment of the 
tumor. In women, however, pregnancy does not appear to affect malignant 
tumors in a constant manner, although fibromyomata of the uterus are 
known to have remained stationary in size and even to have retrogressed 
during this state. 
Tyzzer9 found that the mother mouse could transmit her natural resist- 
ance to a neoplasm to her young; Clowes,10 however, had previously shown 
that the young of resistant mothers could be inoculated. Susceptibility to 
inoculation, however, is surely transmissible, as shown by Moran,11 Cuenot 
and Mercier,12 and others. As will be shortly discussed, it is possible to 
produce active immunity against neoplasms in mice, but this induced or 
artificial immunity is not transmissible, as shown by Bashford, Murray, and 
Haalnad.13 
Of even greater interest, and especially from the standpoint of biologic 
therapy, is the peculiar form of active immunity that may be induced 
against malignant tumors among mice and rats. Clowes and Baeslack,14 
1 Jour. Med. Res., 1901, N. S., 1, 36. 
2 Sci. Reports, Cancer Research Fund, London, 1904, 1, 14. 
3 Berl. klin. Wchn., 1911, 48, 215. 
4 Berl. klin. Wchn., 1905, 42, 872. 
6 Berl. klin. Wchn., 1907, 44, 718. 
6 Arb. a. d. k. Gesundh., 1909, 30, 440. 
7 Jour. Med. Res., 1902, 3, 76. 
8 Quoted by Apolant, Jour. Exper. Med., 1911, 14, 320. 
9 Jour. Med. Res., 1909, 16, 519. 
10 Johns Hopkins Hosp. Bull., 1905, 16, 130. 
11 Arch. d. Med. exp. e. d’Anat. path., 1894, 6, 692. 
12 Compt. rend. d. l’Acad. d. Sci., 1910, cl, 1443. 
13 Third Sci. Report, Imperial Cancer Fund, 1908, 395. 
14 Med. News, 1905, 77, 969. 1028 
VACCINES, SERA} BLOOD, PROTEINS IN TREATMENT OF DISEASE 
Flexner and Jobling,1 Lewin,2 and others noted that when tumors underwent 
spontaneous absorption these mice were rendered refractory to reinocula- 
tion. Further investigations showed that in a general way the subcutaneous 
injection of bits of tumor tissue, failing to infect mice the first time, may 
actually render them more resistant than ever; that is, increase their resist- 
ance by a process of active immunization. Ehrlich3 found that this active 
immunity was not specific; that mice inoculated with cancer tissue were 
rendered refractory to sarcomata and the reverse. Bashford4 had pre- 
viously recorded that mice inoculated with one strain of cancer tissue were 
rendered refractory to inoculation with a second and different strain, and 
he and his associates5 later discovered that this immunity was not specific 
and could be engendered by the injection of normal defibrinated mouse 
blood. Numerous investigators later confirmed these findings and showed 
that immunity against inoculable tumors could be engendered in mice by 
preliminary injections of normal mouse spleen, liver, and other tissues. 
Woglom6 found injections of sterile chicken embryo skin particularly 
efficacious, but injections in amounts from 1 to 10 c.c. failed to immunize 
chickens against inoculation with the Rous chicken sarcoma. Fleisher 
and Loeb7 found that mice with spontaneous tumors were slightly more 
refractory to transplantable tumors than normal mice; that its own tumor 
could be transplanted to another part of its body successfully, but rarely 
to other mice. These investigators found that mice with spontaneously 
growing tumors were not immunized thereby to the same extent as the 
immunity engendered by transplantable tumors. 
While the immunity is non-specific in the sense that it may be engendered 
against tumors in general by preliminary injections of normal mouse tissues, 
it was shown that these tissues must be from mice in order to immunize 
mice, from the rat to immunize rats, etc. The injection of normal tissues 
of the rat do not immunize mice, and the reverse. Michaelis8 found that the 
injection of mice with an emulsion of rat carcinoma did not confer an immu- 
nity against mouse carcinoma, and the reverse; Lewin,9 however, thought 
that this was possible, but numerous additional investigations have estab- 
lished fairly well that active resistance against tumors is engendered only 
by intact tumor cells of the same species. 
This active immunity has been shown to be general by Hashford, Murray 
and Cramer10; that is, subcutaneous injections of tumor cells immunize 
against subsequent implants by the intraperitoneal and other routes. Wog- 
lom11 showed that mice made refractory by subcutaneous injections of embryo 
skin were resistant to intrarenal as well as to subcutaneous grafts of cancer. 
Woglom12 and others have also si own that this active immunity cannot be 
engendered by a mouse’s own tissues; that is, if a mouse is injected with an 
emulsion of its own spleen it is not rendered resistant, although this state 
may be induced by injections of the spleen of another mouse. 
Furthermore, Clowes,13 Bride,14 and others have shown that living cells 
are necessary; vaccines of heat and chemically killed cancer cells and of the 
nucleoproteins, do not suffice. 
1 Proc. Soc. Exp. Biol, and Med., 1907-08, 5, 17. 2 Ztschr. f. Krebsf., 1907-08, 6, 306 
3 Ztschr. {. arztl. Fortbildung, 1906, 3, 211. 
4 Brit. Med. Jour., 1906, 2, 209. 
6 Brit. Med. Jour., 1906, 2, 209; Lancet, 1906, 2, 315; Third Scientific Report, Imperial 
Cancer Research, 1908, 333, 369. 10 Proc. Roy. Soc., Series B, 1907, 69, 177. 
6 Jour. Exper. Med., 1915, 22, 154. 11 Lancet, 1911, 2, 92. 
7 four. Med. Res., 1916, 34, 1. 12 Ztschr. f. Immunitatsf., 1911, 11, 683. 
8 Ztschr. f. Krebsf., 1907, 5, 192. 18 Brit. Med. Jour., 1906, 2, 1550. 
9 Berl. klin. Wchn., 1907, 44, 1606. 14 Ann. de l’lnst. Pasteur, 1907, 21, 768. TREATMENT OF TUMORS 
1029 
Of great importance from the standpoint of prevention of metastases 
is the question whether a spontaneously growing tumor in a mouse im- 
munizes the animal against reinoculation with bits of the same tumor. 
Haaland1 found that this did not occur; furthermore, that the spontaneous 
tumor did not immunize against other transplantable growths. In human 
beings there is no adequate immunity reaction to prevent metastases and, 
more unfortunately, no adequate means at hand for bringing about a resist- 
ance. Whether or not the existence of one tumor in a human being im- 
munizes against the development of other tumors of a different kind is doubt- 
ful, and probably does not occur. Certainly it is not infrequent to find 
carcinoma of the cervix and fibromyomata of the uterine body growing 
simultaneously, and hypernephroma of the kidney has progressed at the same 
time as a carcinoma of the stomach, etc. 
Evidences of the refractory state engendered by injections of living 
tumor or other cells first appears, according to Ehrlich,2 in from seven to 
fourteen days, and lasts weeks and even months. .Woglom3 found that the 
immunity engendered by injections of embryo skin reached its maximum 
about the tenth day, remained at a high level until the twenty-fourth day, 
and then declined, to vanish at about the seventy-fifth day. 
Of further interest, and especially from the standpoint of biologic 
therapy, are the investigations bearing upon the possibility of producing 
passive immunity to transplantable tumors. Will the blood of a natural 
resistant mouse protect a susceptible mouse? Is it possible to immunize a 
rabbit with cancer cells and find that its blood will protect animals against 
infection with this cancer? Will the blood of mice or other animals rendered 
refractory or immune by injections of living cancer or normal tissues protect 
susceptible animals against implantations of tumors? Considerable work 
has been done on these questions by Clowes and Baeslack,4 v. Dungern,5 
Apolant,6 Haaland,7 Russel,8 and others, with generally negative results, 
although Gaylord states that passive immunity to cancer does actually 
exist, even though it is not easily or always demonstrable. 
It will be seen, therefore, that malignant tumors of the lower animals 
and presumably of human beings behave in some respects like microbic 
infections, in that they are inoculable and may engender a state of active 
immunity. That this active immunity is not specific is in line with the 
possibility of engendering active immunity against bacterial infections with 
non-specific agents; cancer immunity is peculiar, however, in that it may be 
induced by the injection of cancer or normal cells of the same species (homol- 
ogous immunization), and requires a vaccine of living cells. Failure of 
passive immunization against tumors is not peculiar because passive immuni- 
zation may fail frequently and readily enough in various microbic infections. 
What is the nature or mechanism of natural immunity to neoplasms 
displayed by the lower animals? What is the nature of the artificial or 
induced immunity engendered by injections of homologous living tumor or 
normal cells? In mice and rats transplantable tumors may undergo spon- 
taneous disappearance; what mechanism brings about this condition? These 
are questions that have occupied the attention of numerous investigators 
1 Fourth Sci. Report Imperial Cancer Res., 1911, 79, 83. 
2Ztschr. f. arztl. Fortbildung, 1906, 3, 211. 
3 Jour. Exper. Med., 1912, 16, 629. 
4 Medical News, 1905, 77, 969. 
5 Ztschr. f. Immunitatsf., 1910, 5, 695. 
6 Ztschr. f. Krebsf., 1911-12, 11, 106. 
7 Berl. klin. Wchn., 1907, 44, 717. 
8 Third Scientific Report Imperial Cancer Research, 1908, 357. 1030 
VACCINES, SERA, BLOOD, PROTEINS IN TREATMENT OF DISEASE 
for the past twenty years, and have not yet been answered, although valu- 
able information has been gained.1 
Ehrlich2 has offered an explanation resting on the hypothesis that a 
tumor requires for its growth a certain specific food substance. He found 
that, by inoculating a virulent and rapidly growing tumor, he could render 
impossible the growth of a second tumor, and believed that the first tumor 
had used up the pabulum which creates a condition of aihrepsia or aihrepiic 
immunity. Further support for this theory was found in zigzag inoculations; 
a mouse tumor transplanted to the rat may die owing to a lack of specific 
pabulum, whereas when transplanted back again to the mouse it lives and 
thrives, because of the supply of specific food furnished by the latter. Ehr- 
lich has stated that “whenever cancer cells fail to proliferate it means that 
they fail to obtain the food required either because the normal body cells 
have a greater avidity for this food or else the cells with which the animal 
was immunized anchored all the specific food and the cancer cells inoculated 
subsequently could not obtain it.” 
This theory has been vigorously attacked by Uhlenhuth and his asso- 
ciates,3 Clowes,4 Lewin,5 Bride,6 Bashford and Russel,7 Levin and Sittenfeld,8 
and others; on the other hand, Apolant,9 Schone,10 Borrel,11 and others have 
just as vigorously defended it. The theory appears acceptable to the writer 
in so far as offering an explanation of natural resistance; it is possible that 
these living cells die because of a lack of endocellular nutriment. It may 
even explain spontaneous retrogression of tumors by exhaustion of a vitamin 
required for the growth of these cells. It does not explain, however, the 
mechanism of artificial or induced immunization secured by the injection 
of vaccines of living cancer or normal tissue cells. 
Russel12 has sought to explain natural and acquired resistance on the 
basis that the cancer cell when transplanted is unable to excite the connective 
tissues of the host to produce a “scaffolding” of connective tissue—a failure 
of vascularization and “specific stroma reaction,” which he believes are 
necessary for successful tumor growth. Woglom13 confirmed these observa- 
tions, but Burgess14 believed that the resistance of non-susceptible animals 
■was due to an inflammatory lesion interfering with the nutrition of the 
tumor. Levin15 believes that the primary factor of resistance is inhibitory 
action of the cells of the host upon the tumor cells; that the growth of cancer 
represents a loss of equilibrium between this inhibitory action and the 
proliferative power of cancer cells, and that the connective tissue reaction 
is of secondary importance. 
Da Fano16 has stated that the lymphocytes and plasma cells are especially 
concerned in the mechanism of tumor immunity; this subject has been 
1 For a splendid review of the literature up to 1913 see Studies in Cancer and Allied Sub- 
jects by Woglom, Columbia University Press, New York, 1913. 
2 Verhandl. d. deutsch. path. Gesellsch., 1908, 12, 17. 
3 Centralbl. f. Bakteriol., Ref., 1910, 47, 158. 
4 Brit. Med. Jour., 1906, 2, 1551. 
6 Berl. klin. Wchn., 1907, 44, 1606. 
6 Ann. de l’Inst. Pasteur, 1907, 21, 771. 
7 Proc. Roy. Soc., Series B, 1909, 72, 298. 
8 Jour. Exper. Med., 1911, 13, 511. 
9Ztschr. f. Immunitatsf., 1911, 10, 103; Jour. Exper. Med., 1911, 14, 316. 
10 Deutsch. med. Wchn., 1907, 33, 866. 
11 Bull, de l’Inst. Pasteur, 1907, 5, 594. 
12 Proc. Roy. Soc., Series B, 1909-10, 72, 298. 
13 Fifth Scientific Report Imperial Cancer Research, 1912, 43. 
14 Jour. Med. Res., 1909, 21, 575. 
15 Jour. Exper. Med., 1908, 10, 811; ibid., 1911, 13, 604; ibid., 1912, 15, 163. 
16Ztschr. f. Immunitatsf., 1910, 5, 1. TREATMENT OF TUMORS 
1031 
especially studied by Murphy1 in this country, and Murphy and Taylor2 
have recently summarized the main points of evidence, as follows: 
“(1) the accumulation of lymphocytes about a transplanted cancer 
graft in an immunized animal; (2) the rise in number in the circulating 
lymphocytes during the development of the immune state, irrespective of 
whether the type of immunity induced is artificial or natural; (3) the setting 
aside of the potential immunity by the x-rays where the dosage employed 
is sufficient to destroy a large part of the circulating lymphocytes; and 
finally, (4) the abolition of the potential immunity for a special tumor strain 
by means of the lymphocyte-destroying power of the x-rays. These points 
are further supported by the observations of Loeb3 on the part played by 
the lymphocyte in respect to homoplastic grafts of normal tissue, and by 
those of Murphy4 on heteroplastic tissue grafts.” 
The extensive investigations of Fleischer5 on the histologic reactions 
about transplanted tissues in normal and immunized animals also indicates 
the importance of the leukocyte reaction. 
Failure of passive immunization in tumors of mice suggests that anti- 
bodies for tumor cells are not present in the blood, and the majority of inves- 
tigators support this view. Rous6 joined susceptible mice to non-susceptible 
in parabiosis, but found that nothing developed in the latter to prevent 
successful transplantation. Kross7 has recently failed to confer passive 
immunity with injections of blood from immune animals, and in parabiotic 
experiments failed to increase the susceptibility of the immune or decrease 
the immunity of the susceptible. Tsurum8 and others reported the presence 
of more or less specific complement-fixing antibodies in the sera of immune 
animals, but the consensus of opinion is to the effect that antibodies are 
not demonstrable in the blood of animals possessing natural or acquired 
immunity to transplantable tumors; Lambert and Hanes9 succeeded in 
growing rat sarcoma in immune plasma in vitro as successfully as in normal 
plasma. 
Serums, Vaccines, and Autolysates in the Treatment of Malignant Tu- 
mors.—Investigations in the immunity of mice and other lower animals to 
transplantable carcinomata and sarcomata, and especially the states of in- 
duced resistance to transplantation resulting from the preliminary injection of 
emulsions of tumor cells and normal tissues, naturally aroused strong hopes 
for the development of a form of biologic therapy for the treatment of 
malignant neoplasms in human beings. 
Among the earliest attempts in this field were those by Jensen10 and 
v. Leyden and Blumenthal,11 with serums prepared by the immunization of 
animals with emulsions of tumor cells; Borrel,12 Bride,13 Lewin,14 Uhlenhuth,15 
and others, however, have proved that nothing in the way of a specific 
therapy is to be expected in this direction. Some of these serums undoubted- 
ly contained small amounts of cytolysins for human tissue cells, but these 
were without specific effects upon the neoplastic cells of human beings and 
the lower animals, and too feeble in action to bring about more than the 
1 Jour. Amer. Med. Assoc., 1912, 59, 874; Jour. Exper. Med., 1913, 17, 482; ibid., 1915, 
22 204. 
2 Jour. Exper. Med., 1918, 28, 1. 
3 Jour. Med. Res., 1917, 37, 229. 
4 Jquj* Exper Med j 1914? 513« 
5 Jour! Med. Res., 1917,37,483; ibid., 1918, 38,353; ibid., 1918, 39,1; ibid., 1922, 43,145. 
6 Jour. Exper. Med., 1909, 11, 810. 11 Deutsch. med. Wchn., 1902, 28, 637. 
7 Jour. Cancer Res., 1921, 6, 25, 121. 12 Bull, de l’Inst. Pasteur, 1907, 5, 607. 
8 Jour. Path, and Bacteriol., 1915, 20, 214. 13 Ann. de l’Inst. Pasteur, 1907, 21, 774. 
9 Jour. Exper. Med., 1911, 14, 129, 453. 14 Ztschr. f. Krebsf., 1907-08, 6, 308. 
10 Centralbl. f. Bakteriol., 1903, 34, 30. 15 Arb. a. d. k. Gesundh., 1911, 36, 491. 1032 
VACCINES, SERA, BLOOD, PROTEINS IN TREATMENT OF DISEASE 
usual degree of spontaneous retrogression. Clowes1 thought that the serums 
of mice whose tumors had undergone spontaneous disappearance pos- 
sessed some curative activity, but Bashford2 was unable to confirm these 
results. In this connection mention may also be made of the work of 
Freytag,3 who employed injections of blood and serum from mice and other 
animals in the treatment of mouse cancer with retrogression of the tumors 
in some, but cures in none; also of the experience of Hodenphyl,4 who found 
that the ascites fluid from a case of human cancer in which the primary and 
secondary tumors had undergone spontaneous retrogression had some 
curative influence upon transplantable cancers in mice and in the majority 
of 47 human cancers. Ill and Minningham5 treated 30 additional cases of 
human cancer with the fluid from a similar case, and while no case was cured 
observed relief of pain, hemorrhage, and cachexia in a number. Green 
and Konrad6 also observed some improvement of cases of uterine cancer 
treated with injections of ascites fluid, but believed that the results may 
have been due to non-specific stimulation; Weil7 was unable to determine 
any difference in the effects of fluids from cancer cases from those of other 
origin. It is possible that these fluids contained specific anticancer sub- 
stances, but it is more probable that the temporary beneficial effects were 
due to non-specific protein reactions. 
Vaccines of tumor cells have been used by Blumenthal,8 Coca and Gil- 
man,9 Coca, Dorrance, and Lebredo,10 Graff and Ranzi,11 Risley,12 Vaughan,13 
and others for the treatment of malignant tumors of the lower animals and 
human beings; in some instances beneficial results were observed and espe- 
cially the temporary relief of some of the symptoms, but actual cures and 
prevention of metastases have not been achieved. Lunckenbein14 believed 
that good results were to be obtained by the intravenous injection of a vaccine 
of the patient’s own cancer cells, and more recently Kellock, Chambers, and 
Russ18 have expressed the opinion that the subcutaneous injection of these 
autogenous vaccines of tumor cells may be of value in the prevention of 
metastases after the primary tumor had been removed surgically, but it is 
highly probable that the temporary relief of pain, hemorrhage, and cachexia 
with gain in weight observed among some human cases are to be attributable 
to the non-specific protein reactions that may be elicited by these sub- 
stances. The same is probably true of the beneficial effects following the 
injection of autolysates of tumor cells reported by Bauer, Latzel and Wessely,16 
and Klinger,17 and of autolysates of human embryos, reported by Fichera18 
and others; likewise of injections of the patient’s own blood or serum (auto- 
serum therapy) reported by Krokiewicz,19 Paget,20 and others. 
1 Brit. Med. Jour., 1905, 2, 1552. 
2 Brit. Med. Jour., 1905, 2, 96; ibid., 1906, 2, 209. 
3 Ztschr. f. Krebsf., 1910—11, 10, 157. 
4 Med. Rec., 1910, 67, 359. 
5 Jour. Amer. Med. Assoc., 1912, 49, 497. 
6 Boston Med. and Surg. Jour., 1914, 170, 352. 
7 Jour. Med. Res., 1910, 23, 85. 
8 Med. Klin., 1910, 6, 1982; Berl. klin. Wchn., 1914, 50. 
9 Philippine Jour. Sci., 1909, 4, 391. 
10 Ztschr. f. Immunitatsf., 1912, 13, 543. 
11 Mitt. a. d. Grenzg. d. Med. u. chir., 1912, 25, 211. 
12 Boston Med. and Surg. Jour., 1911, 165; Jour. Amer. Med. Assoc., 1911, 56, 1383. 
13 Jour. Amer. Med. Assoc., 1914, 63, 1258. 
14 Munch, med. Wchn., 1914, 41. 16 Lancet, 1922, 1, 217. 
16 Ztschr. f. klin. Med., 1915, 81, 355. 
17 Cor.-Bl. f. schweiz. Aerzte, 1915, 46, 1217. 
18 Bull, de l’Inst. Pasteur, 1911, 9, 272; Lancet, 1911, 2, 1194; Policlinico, July 3, 1910. 
18 Wien. klin. Wchn., 1912, 25, 1311. 20 Med. Record, 1916, 89, 719. TREATMENT OF TUMORS 
1033 
Treatment of Sarcoma with Coley’s Fluid.—The old clinical observation 
that an attack of erysipelas often causes a decrease in the size of malignant 
tumors, especially sarcomas, received confirmation in the work of Fehleisen, 
who inoculated tumors with living cultures of streptococci. Among 6 pa- 
tients so inoculated, a decrease in the size of the tumor was noted in 5. 
Killed cultures were tried without effect. Coley1 then tried injections of the 
toxins of streptococci and later mixtures of the toxins of streptococci (prefer- 
ably from cases of erysipelas) and Bacillus prodigiosis. 
Since then Coley’s fluid has been rather extensively employed in the 
treatment of sarcomas. There is not a uniformity of opinion in regard to 
its value, but the majority of surgeons believe that it should be tried in 
cases of inoperable sarcoma. Coley has always maintained a conservative 
and scientific attitude and undoubtedly has obtained permanent cures in 
many cases. In a general way cures have been effected by Coley and other 
surgeons in from 4 to 9 per cent, of cases. Hunt,2 Hill,3 Harmer,4 and others 
have reported favorable results; the latter has summarized the results of 
100 inoperable cases proved by microscopic examinations, as follows: 
“The cases were arranged in six groups, determined by the effect of the 
toxins. Group A. Cases in which there was no appreciable effect, 12 cases. 
Group B. Cases in which the growths softened, but did not appreciably 
diminish in size, 5 cases. Group C. Cases in which growths disappeared 
or practically disappeared, but returned, 20 cases. Group D. Cases in 
which growths disappeared, but metastases simultaneously occurred, 10 
cases. Jroup E. Cases in which growths diminished in size, but still per- 
sisted, 14 cases. Group F. Apparent cures in which growths disappeared 
and no metastases have occurred, 73 cases. Spindle-cell sarcoma, 26 cases; 
36 per cent, apparent cures. Round-cell sarcoma, 28 cases; 39 per cent, of 
apparent cures. Melanotic sarcoma, 6 cases, 5 in Group D, one in Group F. 
Giant-cell sarcoma, 14 cases; 11 apparent cures. Mixed-cell sarcoma, 13 
cases; 5 apparent cures.” 
Harmer concludes that mixed toxins of streptococcus and Bacillus pro- 
digiosus are of value in certain cases of inoperable sarcoma. He empha- 
sizes that the treatment of primary or recurrent inoperable sarcoma with 
mixed toxins must be intensive, and that the institution of this treatment 
is unjustifiable in cases in which operative measures of reasonable safety 
offer possible hope for removal. Seventy-three of the cases analyzed after 
most rigorous criticism have been regarded as apparent cures. The small 
round-cell type apparently offers the greatest expectation of benefit, fol- 
lowed closely by the spindle-cell type (excluding fibrosarcoma). Only a 
relatively small number of mixed-cell type have been benefited. The use 
of toxins in cases with multiple melanotic growths does not seem justifiable, 
but their use in single melanotic growths is legitimate. Regarding the 
tissue of origin the greatest number of apparent cures have occurred in bone 
sarcoma (exclusive of giant-cell cases) over 18 per cent, of the total number 
of apparent cures, with an equal division of round-cell and small cell types. 
Giant cell cases furnish about 15 per cent, of the total number of apparent 
cures. Primary, inoperable, round-cell sarcomata of the cervical glands 
compose about 10 per cent, of apparent cures. Sarcomata arising from 
fascia and muscle, which have been apparently cured, have been situated 
in the lower extremity, abdominal wall, and back. They composed about 
16 per cent, of the total number of apparent cures. Nine of 12 are of spindle- 
1 Amer. Jour. Med. Sci., July, 1894; Jour. Amer. Med. Assoc., 1910, 55, 346; Surg., Gyn., 
and Obstet., August, 1911. 3 Arch. Pediat., September, 1912. 
2 Lancet, June 17, 1911. 4 Boston Med. and Surg. Jour., 1915, 172, 321. 1034 
VACCINES, SERA, BLOOD, PROTEINS IN TREATMENT OF DISEASE 
cell type. In a small number of cases the toxins produce striking relief of 
pain. 
The injections are subcutaneous, at first in parts removed from the 
tumor, and later, if possible, directly into the tumor itself. 
The immunity of syphilis and attempts toward active prophylactic 
immunization have been discussed in Chapter XXXV; attempts toward 
passive prophylactic immunization have been reviewed in Chapter XXXVIII. 
Vaccine and Serum Treatment.—Nothing definite has yet been accom- 
plished in the treatment of syphilis by the administration of a vaccine of 
Spirochseta pallida; Noguchi’s luetin is such a vaccine. As stated in Chap- 
ters XXXV and XXXVIII the injection of animals with dead vaccines of 
pallida is followed by very little, if any, antibody production, although I 
am not sure that the results would be the same in man. It is possible that 
the subcutaneous injection of a pallida vaccine, as luetin, in doses of 0.5 to 
1 c. c. may be beneficial, as indicated by the report of Mueller and Planner,1 
who found that the administration of luetin in this way rendered the curative 
activity of neoarsphenamin more pronounced. My colleague, Dr. Scham- 
berg, and myself, treated a series of “Wassermann fast” cases of syphilis 
two years ago with a vaccine of pallida prepared by cultivating the spiro- 
chetes in fluid media, followed by washing, suspension in saline, heating, and 
careful sterility tests; we found in several an apparent increased suscepti- 
bility to arsphenamin with a reduction in the Wassermann reaction, fur- 
thermore some curative effect was produced on the lesions by the vaccine 
alone, so that the subject of vaccine therapy as an adjuvant to the treatment 
of some cases of syphilis refractory to ordinary measures is one worthy of 
further study and interest. 
As discussed in Chapter XXXVIII, potent immune sera for the prophy- 
laxis and treatment of syphilis have not been prepared by the immunization 
of the lowxr animals. Very possibly the sera of human syphilitics in the 
latent stages possess immune principles, but these sera have not been em- 
ployed in the treatment of syphilis, although it is possible that some good 
may be accomplished in both a specific and non-specific way by the intra- 
venous injection of such sera. 
Non-specific Treatment.—Just as the subcutaneous injection of pallida 
vaccine (luetin) may aid in the treatment of syphilis by rendering the 
action of mercury and arsphenamin more effective, so likewise other protein 
agents may have similar effects. 
Uddgren2 has observed that the intramuscular injection of milk may 
provoke a positive Wassermann reaction in syphilitics with previous nega- 
tive reactions; this effect is probably due to some stimulating action upon 
foci of the disease. Stiickgold,3 Schreiner,4 Schacherl,5 Hauber,6 and others 
have employed injections of milk, peptone, tuberculin, and other agents 
as adjuvants to treatment with mercury or arsphenamin, observing a more 
rapid clinical and serologic improvement in syphilitics, especially tertiary, 
nerve, and congenital cases. 
TREATMENT OF SYPHILIS 
1 Berl. klin. Wchn., 1918, lv, 354. 
2 Berl. klin. Wchn., 1918, lv, 354. 
* Ueber den Einfluss von Interkurrenten fieberhaften Krankheiten und von Fieber- 
zustande durch intragluteale Milchinjectionen hervorgerufen sind auf den Verlauf der Syph- 
ilis, mit besonderer Beriicksichtigung der Congenitaler, Berlin, 1919. 
4 Wien. klin. Wchn., 1920, 33, 34. 
6 Jahr. d. Psych, u. Neurol., 1914-15, 35, 27, 207. 
6 Ztschr. f. d. Ges. Neurol, u. Psych., 1914, 24, 1. \ SERUM TREATMENT OF SNAKE-BITES 
1035 
A large number of non-specific agents have been employed in the treat- 
ment of paresis, especially fever-producing agents, as tuberculin, and those 
producing leukocytosis, as nuclein. 
v. Jauregg1 has employed subcutaneous injections of Koch’s old tuber- 
culin every two days, beginning with 0.01 gm. and increasing gradually to 
0.5 gm. He and Pilez,2 Hudovernig,3 Battistessa,4 Jukow,5 and others have 
reported that the results were more favorable when the injections were 
employed along with the administration of mercury and arsphenamin than 
when the latter alone were used. 
Donath,6 Fisher,7 Szedlek,8 and others have employed subcutaneous 
injections of 0.5 to 3 c.c. of 10 per cent, solutions of nuclein every three to 
five days for the purpose of producing leukocytosis, and reported that the 
results were better when employed along with mercury and iodids or 
arsphenamin than when the latter alone were employed; Brown and Ross,9 
however, were not able to confirm these results. 
v. Jauregg has also employed staphylococcus vaccine; Friedlander has 
used typhoid vaccine, and Plant a mixed vaccine of staphylococci and strepto- 
cocci for leukocytosis and other non-specific effects, v. Jauregg has even 
suggested the advisability of inoculating with malaria in order to secure the 
benefit of the febrile reactions; Miihlens, Weygandt, and Kirschbaum10 have 
recently reported on the treatment of cases inoculated with the Spirillum 
obermeieri and malaria plasmodia, the literature being covered by Raecke.11 
In tabes dorsalis Schacherl12 and others have employed injections of old 
tuberculin, beginning with 0.001 gm. and giving 0.1 gm. of salicylate of 
mercury with each third dose. In the treatment of 76 cases the results were 
interpreted as very encouraging. Others have employed injections of milk. 
With both agents the injections may precipitate a gastric crises or lancinating 
pains. 
SERUM TREATMENT OF SNAKE-BITES 
The nature of snake venom is discussed in Chapter VII, and the method 
of preparing antivenomous serum is described in Chapter XIII. 
Calmette’s13 antivenene for cobra venom is useful in the treatment of cobra 
envenomation, but is not serviceable for the treatment of other snake-bites, 
as shown by Martin for Australian serpents and by McFarland for American 
snakes. In the venoms of our snakes, as, for example, the rattlesnake, 
copper-head, and moccasin, the poison is essentially locally destructive, the 
respiratory poison being of secondary importance. McFarland14 failed to 
immunize horses against this locally destructive poison. Later Noguchi 
and Madsden15 succeeded in producing an antiserum, prepared by immunizing 
horses with venom after the toxophorous groups of the molecules had been 
destroyed, capable of neutralizing the hemorrhagin of the Crotalus venom. 
1 Wien. med. Wchn., 1909, 39, 2124; ibid., 1913, 53, 2555. 
2 Wien. med. Wchn., 1912, 52, 2010 and 2083. 
3 Neurol. Centralbl., 1913, 32, 313. 
4Riv. ital. d. neuropatol., etc., 1912, 5, 117. 
6 Russk. Urach., 1913, No. 24, 862. 
8 Wien. klin. Wchn., 1909, 22, 1289; Allg. Ztschr. f. Psych., 1910, 57, 420. 
7 Prag. med. Wchn., 1909, 34, 401. 
8 Neurol. Centralbl., 1916, 35, 57. 
9 Jour. Mental Sci., 1912, 48, 389. 
10 Munch, med. Wchn., 1920, 57, 831. 
11 Therap. Monatsh, 1920, 34, 129. 
12 Jahrb. f. Psych, u. Neurol., 1914-15, 35 207. 
13 Jour. Med. Research, 1909, 21, 47. 
14 Jour. Med. Research, 1909, 21, 51. 
16 Jour. Exper. Med., 1907, 9, 18. 1036 
VACCINES, SERA, BLOOD, PROTEINS IN TREATMENT OF DISEASE 
The serums of Calmette, Noguchi,1 Flexner and Noguchi,2 and others 
are useful in the treatment of their respective envenomations, but aside 
from India, Brazil, and a few other reptile-infested countries, as well as in 
zoological gardens and laboratories where snakes are kept, the serums have 
a very limited sphere of usefulness. 
Houssay, Sordelli, and Negrete3 have recently made extensive studies 
on the preparation and action of various antisera for South American snake 
venoms. In Brazil two specific sera (antibothropic and anticrotalic) have 
yielded excellent results, and a third polyvalent or antiophidian serum has 
recently been announced from the Serotherapic Institute of Brazil, which 
is believed to counteract the effects of all poisonous reptiles in Brazil. Similar 
efforts are to be made in India, where of over 300 species of snakes, at least 
68 are known to be poisonous with an annual death-rate of over 20,000 
human beings and countless nu nbers of animals. 
TREATMENT OF OTHER DISEASES 
Autosera Treatment of Chorea.—Sydenham’s chorea or St. Vitus’ Dance, 
possessing certain features suggestive of focal infection and especially in 
relation to the occurrence of arthritis in about 20 per cent, of cases, endo- 
carditis, as well as to its occasional development after whooping-cough, 
acute septicemia, etc., and in dogs after distemper, has been treated by 
Goodman4 with intraspinal injections of the patient’s own serum. About 
40 to 50 c.c. of blood are drawn aseptically and the serum separated; spinal 
puncture is conducted, and after removal of 15 to 20 c.c. of fluid, an equal 
amount of the serum is injected very slowly. Of 30 cases treated by Good- 
man, 14 received one injection, 8 received two injections, and the balance 
three or more. In the majority of these cases twitching ceased within a 
week or two, but the duration of the improvement has not been stated. 
Porter5 has also reported upon the results of this treatment, and advises 
doses of 5 to 10 c.c. of serum; Brown, Smith, and Phillips6 report the cure 
of 18 patients of a series of 23, with improvement in 4 and failure in 1. Seven- 
teen of these were mild cases and 5 were severe. The average number of 
injections were three. Reactions usually followed the injections. Four 
cases were relieved in one week and 19 in three weeks. Tarr7 has also 
observed good results with this therapy. 
Non-specific Protein Treatment of Epilepsy.—Dollken8 has treated a 
series of cases with injections of milk and a stock vaccine (for bacterial 
proteins) together with the administration of luminal. After four to six 
months of treatment 12 cases were free of attacks for eighteen months and 
60 for one year. In 13 cases an improvement was noted. 
If this therapy is tried, a child under twelve years of age may receive 
by intramuscular injection 1 to 2 c.c. of market milk boiled for ten minutes 
and cooled. Injections are given only after attacks occur, so that the 
intervals are irregular. 
Edgeworth9 has reported favorably upon the intravenous injection of 
a sterilized 5 per cent, solution of Witte’s peptone. The first dose may be 
1 Jour. Exper. Med., 1906, 8, 614. 
2 Jour. Med. Research, 1904, 11, 363. 
3 Rivista del Inst. Bacteriol., 1917,1,15,485, 565, 617; ibid., 1919, 2,151,189,211. 
4 Arch. Pediat., 1916, 33, 649. 
5 Amer. Jour. Dis. Child., 1918, 16, 109. 
6 Brit. Jour. Dis. Child., 1919, 16, 8. 
7 Northwest Med., October, 1917. 
8 Berl. klin. Wchn., 1920, 47, 893, 926. . 
9 Brit. Med. Jour., 1920, 2, 780. TREATMENT OF OTHER DISEASES 
1037 
5 minims, and subsequent doses gradually increased unless the reaction is 
too severe, in which case subsequent doses should be reduced. 
Treatment of Typhus Fever.—Nicolle and Blaizot1 have immunized 
horses with injections of extracts of suprarenal glands and spleens of guinea- 
pigs dying of typhus fever. They report that the treatment of 36 cases 
by subcutaneous injection of this serum in dose of 10 to 20 c.c. per day 
resulted in more prompt defervescence, relief of delirium, stupor, and pros- 
tration, and that but 1 case died. 
Gyori2 has used autoserum treatment with benefit in some cases; Munk3 
has employed injections of normal horse-serum, peptone, and nucleohexyl. 
During the Great War a number of these non-specific agents have been 
employed. Holler4 used daily intravenous injections of deutero-albumose 
with good results when treatment was begun within two days of the onset. 
In a series of 50 cases the mortality was 6 per cent., whereas of 15 untreated 
cases the mortality was about 50 per cent. 
Tagle5 has used intravenous injections of 4 to 10 c.c. of 10 per cent, 
solutions of peptone with a mortality of 5 per cent, in 59 cases; children 
received 4 or 5 c.c. Second injections were given about forty-eight hours 
later, and in some cases a third injection. Opazo6 has also employed injec- 
tions of peptone. 
Schultz and his colleagues7 and Kalberlah8 have employed intravenous 
injections of typhoid vaccine; especially good results have been reported 
by Danielopolu9 with daily intravenous injections of hypotonic salt solution 
(0.065 per cent.). 
Non-specific Proteins in the Treatment of Mumps and Orchitis.—In 
the treatment of mumps of men Salvaneschi,10 Bormamour and Bardin,11 
and Mallie12 have employed subcutaneous injections of 20 c.c. or more 
of diphtheria antitoxin or normal horse-serum. The authors believe that 
these injections shortened the course of the disease, were especially valuable 
for the prevention of orchitis, and frequently yielded relief from pain when 
this complication developed. 
Milk injections have not been employed, but since the intramuscular 
(gluteal) injection of 5 to 10 c.c. of milk is known to afford relief in epididy- 
mitis, it would appear that this simple remedy is worthy of trial in orchitis. 
Serum Treatment of Yellow Fever.—Noguchi13 has immunized horses 
with pure cultures of Leptospira icieroides and found the immune sera very 
active in checking the progress of the infection in guinea-pigs. By means 
of this serum protection test Noguchi and Kigler14 have been able to identify 
the Leptospira. 
Noguchi15 has employed the serum in the treatment of 170 human cases 
of yellow fever, and of 95 treated before the third day the mortality was 
1 Bull, de l’Acad. de med., 1916, 76, 95. 
2 Deut. med. Wchn., 1918, 44, 677. 
3 Munch, med. Wchn., 1916, lxiii, 1239. 
4 Med. Klin., 1917, 13, 1038. 
5 Rev. Med. de Chile, 1919, xlvii, 413. 
6 Rev. Med. de Chile, 1919, xlvii, 433. 
7 Berl. klin. Wchn., 1919, 45, 1226. 
8 Therap. Monatsh., 1918, 32, 328. 
9 Le typhus exanthematique, Bucharest, 1919. 
10 Riforma med., 1917, 33, No. 37. 
11 Presse med., 1920, 28, 929. 
12 Jour, de m£d. de Bordeaux, 1922, 94, 16. 
13 Jour. Exper. Med., 1920, 31, 159. 
14 Jour. Exper. Med., 1920, 32, 601. 
15 Jour. Amer. Med. Assoc., 1921, 77, 181. 1038 
VACCINES, SERA, BLOOD, PROTEINS IN TREATMENT OF DISEASE 
13.6 per cent, as compared with a mortality of 56.4 per cent, in 783 cases 
not treated with serum. After the fourth day of the disease the serum had 
no appreciable effect. 
Treatment of Weil’s Disease (Infectious Jaundice).—The cause of this 
disease is now definitely established as being the Spirochcela iclerohemor- 
rhagice discovered by Inada and Ido1 in 1915. 
Inada, Ido, Kaneko, and Ito2 have found human convalescent serum 
possessing some degree of curative activity, but have succeeded in preparing 
immune sera by inoculation of horses with cultures of the spirochete. These 
sera have proved effectual in the treatment of experimentally infected 
guinea-pigs, and likewise in the treatment of human infections when given 
early in the disease. From 50 to 80 c.c. of serum are injected daily in divided 
doses of about 20 to 30 c.c. each; injections are made subcutaneously or 
intravenously. With rare exceptions the serum destroys all spirochetes 
in the blood and reduces their numbers in the internal organs. Of 35 serum- 
treated cases, the mortality was 20 per cent, as compared with the usual 
mortality in Japan of 30 to 48 per cent. In a subsequent report by Inada3 
the mortality of 86 serum-treated cases varied from 34.7 to 38.5 per cent, 
as compared with a mortality of 57.1 per cent, among 72 cases treated with- 
out serum. 
Dollken4 has employed intramuscular injections of 5 to 10 c.c. of boiled 
milk, and reports that the results have been good in many cases. Owing 
to the styptic effects of milk and its effect upon'the liver metabolism, this 
form of treatment is deserving of consideration. 
Serum Treatment of Hyperthyroidism.—Beebe5 has described a method 
of immunizing sheep with intraperitoneal injections of extracts of human 
thyroid glands. The serum is regarded as antagonistic to the excess toxic 
secretions of the gland regarded by some as present in Graves’ disease. It 
is not thyrotoxic or thyrolytic; certainly not in the amounts administered 
by subcutaneous injection in the treatment of the disease. 
Beebe injects about 0.5 c.c. for the first dose; subsequent injections are 
not given until the local reaction subsides, but are usually in amounts of 
1 c.c. In acute asthenic and sthenic types the serum is injected daily for 
several days if the reactions permit. Beebe states that the serum has 
proved beneficial in at least 50 per cent, of cases. 
Vaccine in the Treatment of Neuritis (Sciatica).—Some cases of sciatica 
are now known to be due to focal infection, and especially abscesses in the 
roots of teeth. In sciatica and other types of neuralgia and neuritis careful 
search should be made for foci of infection: teeth, tonsils, accessory nasal 
sinuses, etc. If a focus is found and is amenable to culture and treatment, 
the preparation and administration of an autogenous vaccine may prove 
beneficial, as reported by Zapffe6 in sciatica. 
Certain non-specific proteins have been used with success. Dollken7 
has treated large numbers of cases of neuralgias (trigeminal, intercostal, 
sciatic) and neuritides of varying etiology (cold, alcoholic, post-typhoidal, 
syphilitic, diphtheric, etc.) with an autolysate of staphylococci and B. 
prodigiosus designated as “vaccinurin.” Pressure neuralgias and so-called 
rheumatic palsies were most easily influenced. Holtzl8 also observed good 
1 Tokyo Ijishinshi, 1915, No. 1908. 
2 Jour. Exper. Med., 1916, 23, 377; ibid., 1916, 24, 485; ibid., 1918, 27, 283. 
3 Japan Med. World, 1922, 2, 189. 4 Berl. klin. Wchn., 1919, 46, 226. 
5 Jour. Amer. Med. Assoc., 1915, 64, 413. 
6 Jour. Amer. Med. Assoc., 1915, 64, 238. 
7 Berl. klin. Wchn., 1914, li, 1807, 1841. 
8 Deutsch. med. Wchn., 1918, xliv, 291. TREATMENT OF DISEASES OF THE LOWER ANIMALS 
1039 
results with this treatment. Cadbury1 and Boyd2 have used typhoid 
vaccine, the latter by intravenous injection. 
I have treated 3 cases of sciatica and 2 of neuritis with intragluteal 
injections of 5 c.c. of market milk boiled for ten minutes and cooled. Marked 
local reactions occurred during the following twenty-four hours, but in all 
instances there was a prompt, but temporary, relief from pain. 
If typhoid vaccine is employed, about 50,000,000 may be injected intra- 
venously. The usual reaction of chills and fever follows, and in some cases 
temporary exacerbation of the neuritis. 
Treatment of Marasmus.—Several reports in literature indicate that the 
subcutaneous injection of various protein agents, as human ascites fluid, 
horse-serum (normal or diphtheria antitoxin), milk, and stock vaccines may 
prove of distinct aid in the treatment of marantic children. The effects 
are purely non-specific, probably improve metabolism and digestion, and the 
leukocytosis may aid in overcoming latent bacterial infections, especially 
of the intestinal tract. 
Leavy and Hastings3 and Carter.4 for example, have given daily sub- 
cutaneous injections of 10 to 20 c.c. of sterile ascites fluid from cases of 
cardiac or renal disease. Slawik5 has employed intramuscular injections of 
1 to 5 c.c. of boiled milk. Ferreira,6 Czerny and Eliasberg7 have used sub- 
cutaneous injections of horse-serum; Plantenga8 has employed injections 
of a serum prepared by immunizing animals against colon bacilli, and Put- 
zig9 subcutaneous injections of diphtheria antitoxin. All of these, however, 
have the objection of producing allergic sensitization to horse-serum proteins 
with the possibility of producing serum sickness. 
If the administration of non-specific agents is employed for their possible 
beneficial effects I believe that preference should be given intramuscular 
injections of boiled milk in dose of 1 or 2 c.c., sterile ascites fluid or sterile 
normal human serum in the same amounts; the injections may be given 
at intervals of one week. Very probably a stock vaccine of colon bacilli 
may yield similar results when injected in dose of 0.1 c.c. of a vaccine con- 
taining about 500,000,000 per cubic centimeter. 
TREATMENT OF DISEASES OF THE LOWER ANIMALS 
Vaccines are not widely employed for the treatment of diseases of the 
lower animals, although as presented in Chapters XXXV and XXXVIII, 
vaccines and sera are employed in the prophylaxis of many diseases. Im- 
mune sera are employed in the treatment of some diseases, notably hog 
cholera, anthrax, navel ill, etc. Doubtless a part of the beneficial results 
are due to non-specific effects, but veterinarians have not employed non- 
specific agents, as injections of milk, peptone, vaccines, etc., as is practised 
in human medicine. 
Serum Treatment of Hog Cholera.—Hog cholera serum is usually em- 
ployed for prophylactic immunization, as discussed in Chapter XXXVIII. 
For curative purposes the serum has yielded good results, providing it is 
administered not later than the fourth day after the animal shows evidences 
of the disease. Several injections may be required, and the intramuscular 
route should be chosen because quicker absorption is thus insured. 
1 China Med. Jour., 1919, 33, 213. 2 Jour. Lab. and Clin. Med., 1919, 5, 88. 
3 Boston Med. and Surg. Jour., 1910, clxiii, 293. 
4 Amer. Jour. Med. Sci., 1911, 142, 241. 
6 Jahr. f. Kinderh., 1919, xc, 231; ibid., 1921, 94, 192. 
6 Arch. Latin-Amer. d. Pediat., 1920, 14, 109. 8 Geneesk. Bladen, 1918, 19, 347. 
7 Monatschr. f. Kinderh., 1920,18, No. 1. 9 Berl. klin. Wchn., 1921, 48, 151. 1040 
VACCINES, SERA, BLOOD, PROTEINS IN TREATMENT OF DISEASE 
The use of the serum for curative purposes is, however, quite limited, 
because in most instances cholera hogs are destroyed in order to prevent the 
spread of the infection. Even though a young hog recovers, it usually 
remains a “runt,” growing poorly and remaining underweight. The serum, 
however, may be used for the treatment of particularly valuable animals 
and those diagnosed very early in the disease and showing nothing but a 
rise of temperature. 
Serum Treatment of Rinderpest.—As mentioned in Chapter XXXVIII, 
this disease (cattle plague) is prevalent among the cattle of European and 
African countries; the etiologic agent is present in the blood and bile, and 
is filtrable. Kolle and Turner have prepared immune serum by injecting 
healthy oxen subcutaneously with increasing amounts of infectious blood 
and bile. The serum is said to be curative when administered by subcu- 
taneous or intramuscular injection at least thirty days after the onset of 
fever. Of 3318 animals treated in this manner, 455, or 13.9 per cent., died, 
while the mortality among non-treated animals is very high, averaging 
85 to 95 per cent. 
Serum Treatment of Anthrax.—The Department of Agriculture1 has given 
the following directions for the administration of anti-anthrax serum for 
curative purposes: 
“In the treatment of anthrax serum should be administered in large 
doses. An animal showing only a high temperature, with no other mani- 
festations of the disease, should be given from 30 to 50 c.c., but if the 
gravity of the disease is pronounced, 100 c.c. should be administered. In 
almost every instance a drop in temperature may be observed and a dimin- 
ishing of the severity of the symptoms. At times, however, a relapse occurs 
about the second or third day following the serum injection, when it becomes 
necessary to administer another and larger dose of serum. It has been 
proved that animals affected with anthrax, even after the bacilli are found 
in the blood circulation, may recover after a large injection of potent serum. 
There is no danger from large doses—the danger is in insufficient dosage or 
delay in administration.” 
In administering the serum for curative purposes injections should be 
made intravenously for quick results, and repeated as often as apparently 
needed. If intravenous injection is impossible, the serum should preferably 
be injected at two or three different places subcutaneously to facilitate 
rapid absorption. 
Serum Treatment of White Scours.—For the treatment of larger animals 
30 to 100 c.c. of immune serum may be given intravenously upon the first 
warning of the disease by the presence of blood in the feces; hogs, sheep, and 
goats may receive 10 to 20 c.c. and cats and dogs about 5 to 10 c.c. When 
intravenous injections cannot be made, the serum should be injected intra- 
muscularly. A second intramuscular injection may be given twelve hours 
later and thereafter at intervals of twenty-four to forty-eight hours until 
the disease is checked. 
Serum Treatment of Joint-ill of Colts.—An immune serum prepared by 
the immunization of horses with Bacillus aborlus-equi and streptococci has 
been used successfully in the treatment of arthritis, joint-ill, or navel-ill of 
colts; the usual dose is 5 to 20 c.c. by subcutaneous injection repeated as 
necessary. M’Fadyean and Edwards2 have observed encouraging results 
with this form of treatment. 
Serum Prophylaxis and Treatment of Hemorrhagic Septicemia.—The 
1 U. S. Dept, of Agricult. Bull. No. 340, December 27, 1915. 
2 Jour. Comparative Path, and Therap., 1917, 30, 321. TREATMENT OF DISEASES OF THE LOWER ANIMALS 
1041 
bacteriology and vaccine prophylaxis of hemorrhagic septicemia or pas- 
teurellosis of animals (horses, cattle, hogs, sheep, etc.) have been discussed 
in Chapter XXXV. Antisera are prepared by the immunization of horses 
with the different varieties of Bacillus bipolaris seplicus and their toxins. 
For prophylactic purposes 20 to 40 c.c. may be injected subcutaneously; 
the immunity, however, disappears after six weeks. 
For curative purposes 30 to 100 c.c. are required, according to the size 
of the animal and severity of the disease. If possible, the serum should be 
injected intravenously; otherwise, intramuscularly. Subsequent injections 
may be made intramuscularly as required. 
Serum Prophylaxis and Treatment of Equine Influenza and Strangles.— 
The bacteriology and vaccine prophylaxis of these diseases have been 
discussed in Chapter XXXV. Antisera are prepared by the immunization 
of horses with pneumostreptococci and Bacillus equisepticus of equine 
influenza and wdth Streptococcus equi (Schutz) of strangles. 
For prophylactic immunization 30 to 50 c.c. may be injected subcutane- 
ously or intravenously; the passive immunity gradually disappears after 
three weeks. 
For curative purposes 50 to 100 c.c. should be injected intravenously 
or intramuscularly; subsequent doses may be given by intramuscular injec- 
tion as required. CHAPTER XLI 
BIOLOGIC THERAPY OF TUBERCULOSIS 
IMMUNITY IN TUBERCULOSIS 
All human beings are susceptible to tuberculosis and the disease is 
always found where opportunity for infection is present. The very wide- 
spread dissemination of the disease throughout the world and the frequency 
with which tubercle is found in necropsies indicates that natural immunity 
to tuberculosis probably does not exist. Not all human beings, however, 
become infected with tubercle bacilli; probably 5 to 10 per cent, escape, 
in so far as finding evidence of the disease after death is concerned, and 
especially in necropsies of individuals living in the rural districts. But 
approximately 90 per cent, of adult human beings are said to show evidences 
of tuberculous infection, although, fortunately, in the majority of these the 
disease remains strictly localized, undiscovered during life, and produces 
no symptoms. 
Tuberculosis also occurs among the lower animals, and especially cattle, 
but these animals apparently possess a greater degree of natural immunity 
than man. Hogs are sometimes infected (2.5 per cent.), but horses, sheep, 
and goats possess a very high natural immunity and are seldom infected. 
Tuberculosis also occurs among the cold-blooded animals, but these bacilli 
do not cause more than local lesions when injected into warm-blooded 
animals, and, indeed, may fail to produce any infection. Likewise tubercle 
bacilli of warm-blooded animals usually fail to infect cold-blooded animals. 
For human beings human tubercle bacilli are most infective; there is a 
greater degree of natural immunity to bovine bacilli, but it is now generally 
agreed that the latter may infect man and especially the lymph-glands of 
children. Likewise the lower animals possess a high natural immunity to 
human tubercle bacilli and are rarely infected spontaneously with these 
organisms. 
Certain families and individuals appear to possess a certain degree of 
natural immunity, but it is always difficult and usually impossible during 
life to absolutely exclude the presence of latent foci capable of engendering 
some degree of acquired immunity. For example, Petruschly1 calls attention 
to “mother immunity,” meaning that a woman from a non-tuberculous 
family apparently escapes infection from her tuberculous husband and yet 
her children succumb to tuberculosis. These mothers usually react to 
tuberculin, indicating the presence of latent infection, and the condition is 
in some respects analogous to women giving birth to syphilitic children 
without themselves showing active lesions of syphilis, but nevertheless 
luetic, and showing the acquired resistance to reinfection with Spirochasta 
pallida as a result of latent syphilis. 
Children of tuberculous parents do not contract tuberculosis in utero; 
the disease is not a prenatal infection, as may occur in syphilis. But these 
children likewise do not inherit an immunity to infection. On the other 
hand, the death-rate among these children may be relatively low because 
they are especially apt to contract the disease early in life and acquire the 
resistance conferred by tuberculous foci. If, however, these children are 
1 Ergebn. d. Immunitatsf., 1914, 1, 189. 
1042 IMMUNITY IN TUBERCULOSIS 
1043 
repeatedly exposed to massive infections in a household, this acquired im- 
munity is readily broken down with a consequently high death-rate. 
Of course, not all individuals exposed to tuberculosis acquire the infec- 
tion. Doubtless in contamination of the mucous membranes of the upper 
respiratory tract with tubercle bacilli of attenuated virulence the organisms 
are removed mechanically by the secretions, and their entrance into the 
tissues and lymphatic channels blocked by ephithelial barriers. Further- 
more, it is highly probable that fixed and wandering cells may destroy the 
bacilli by phagocytosis even though the blood and lymph possess very feeble 
tuberculocidal properties. 
Why human beings possess a high natural immunity for bovine tubercle 
bacilli is not known; likewise the nature of the resistance of the warm-blooded 
animals to tubercle bacilli of the cold-blooded animals and the reverse are 
unsolved. Nothing in the way of immunity principles are found in the 
blood and lymph to account for this natural resistance. The bacilli when 
injected in massive doses may produce local inflammatory lesions, but 
these are similar to lesions produced by any bacterial protein. The bacilli 
simply fail to survive and multiply as if lacking a suitable pabulum and 
suggesting the operation of a mechanism of athreptic immunity, offered by 
Ehlrich as an explanation for the failure of some mouse cancers to grow in 
rats, and the reverse. Probably repeated and persistent cross-infection 
tests would result in the bacilli acquiring the property of surviving in the 
tissues of a heterologous host and resisting the destructive action of phago- 
cytosis. Furthermore, human tubercle bacilli show a marked predilection 
for certaint issues, notably lymphoid tissue, as if finding there favorable 
conditions for survival and multiplication. They seldom infect the muscles 
and may leave no traces of reaction in the tissues of entry, as the lungs or 
intestinal mucosa. This selective affinity may have something to do with 
the question of pabulum or, more probably, with chemical conditions influ- 
encing their growth and metabolism; as long ago as 1889 Weber1 suggested 
that deficiency of carbon dioxid in the tissues favors tuberculosis, while 
accumulation retards the bacilli, and Corper, Gauss, and Rensch2 have 
recently reported experimental evidence in support of this empiric contention. 
Even though our natural defenses against tuberculous infection may be 
insufficient for affording complete protection, yet they are sufficient for 
rendering the initial infection a strictly localized affair and usually of the 
lymph-glands. It is very doubtful that acute overwhelming miliary tuber- 
culosis occurs spontaneously as the primary infection. 
When a few bacilli of moderate virulence gain access to susceptible tissues 
the mechanism of defense is purely cellular. As shown so clearly by Foote3 
the vascular endothelial cells proliferate and migrate, producing the cells 
designated as “epitheliod,” which surround, wall off. and phagocytose the 
bacilli. There may be no other inflammatory vascular phenomena and 
infiltration with leukocytes, and the complete destruction of the bacilli is 
apparently accomplished by these means. 
With larger numbers of bacilli or organisms of greater virulence the 
local reaction is much the same plus inflammatory changes. There may be 
an early and temporary infiltration with polymorphonuclear leukocytes 
followed by proliferation and migration of vascular endothelial cells. The 
latter may multiply and fuse in the lesion to form the giant-cells. The 
1 Therap. Monatsh., 1901, 15, 130. 
2 Jour. Amer. Med. Assoc., 1921, 76, 1216. 
3 Jour. Exper. Med., 1920, 32, 513,533; ibid., 1921, 33, 271; Jour. Med. Research, 1919, 
40, 353. 1044 
BIOLOGIC THERAPY OF TUBERCULOSIS 
cells of the part coming in closest contact with the bacilli are apt to undergo 
destruction. The main struggle is apparently between the invading bacilli 
and the defensive endothelial cells, the latter aiming to wall off the bacilli 
and attempting their destruction to some extent by phagocytosis. Later 
the lymphocytes migrate from the surrounding vessels and the connective- 
tissue cells proliferate into fibroblasts with the production of a wall composed 
of all of these elements. This wall is the chief means of defense against 
the spread of the infections. As stated by Krause,1 “the patient is as resistant 
as the shell of his tubercle.” Within this shell the bacilli may survive for 
long periods of time; their toxins gradually destroy the cells with which they 
come in contact, producing caseation necrosis, which may finally undergo 
enzymic digestion or liquefaction necrosis. The products of the bacilli plus 
those of the necrotic cells constitute the foci poisons; the absorption of these 
is responsible for the general symptoms of tuberculosis, as pyrexia, tachy- 
cardia, anemia, etc. 
These changes may be progressive and effectually resist the walling off 
process. Anything favoring the production of fibrous tissue favors the 
localization of the disease or clinical cure. In this connection mild passive 
hyperemia is probably beneficial as favoring fibrous tissue production, and 
forms the basis for the tuberculin treatment of some forms of tuberculosis 
in order to secure the beneficial effects of the focal hyperemia of the allergic 
reaction. At the same time, however, anything producing excessive hy- 
peremia, as excessive amounts of tuberculin, influences wear and tear upon 
the encapsulating tissue and may favor multiplication of bacilli, and more 
especially facilitate their spread by continuity of tissues, by migrating 
endothelial cells, or by vascular and lymphatic channels. 
To a limited extent the newly formed endothelial cells of a tubercle are 
capable of destroying bacilli of low virulence by means of phagocytosis and 
endocellular ferments; likewise the products of tissue necrosis and especially 
liberated enzymic substances may have a bacteriostatic and bactericidal 
effect and bring about the sterilization of the tubercle in a manner analogous 
to the autosterilization of abscesses produced by gonococci and other pyo- 
genic bacteria. Certainly resistance is cellular and largely mechanical in 
primary infections with tubercle bacilli. Since lymphocytes are found so 
abundantly in tubercles and are increased in the blood it would appear that 
they must be also one of nature’s defensive agencies. Murphy and Ellis2 have 
observed that mice subjected to exposure to x-rays are rendered more sus- 
ceptible to tuberculosis and these effects were ascribed to the destruction of 
lymphocytes and lymphoid tissue. Morton3 has employed x-rays for hasten- 
ing the development of tuberculosis in guinea-pigs, but Weinberg4 was unable 
to hasten the progress of tuberculosis appreciably by exposure of the guinea- 
pig to massive doses of the x-rays, although there occurred a marked reduction 
in the numbers of lymphocytes in the blood. White and Gammon5 have also 
reported that the inhalation of benzene by rabbits reduces their resistance 
to tuberculosis. The investigations of Hektoen, employing x-rays and other 
substances for leukocytic and lymphoid tissue destruction, have indicated 
that these tissues are concerned in antibody production to different antigens 
(Chapter VIII). 
The role of small lymphocytes in resistance to tuberculosis is doubtful, 
but evidence indicates that they probably play a purely secondary part. 
They apparently are not actively phagocytic; they arrive at the scene of 
1 Johns Hopkins Hosp. Bull., 1*917, 28, 191. 
2 Jour. Exper. Med., 1914, 20, 397. 
3 Jour. Exper. Med., 1916, 24, 419. 
4 Arch. Int. Med., 1920, 25, 565. 
6 Tr. Assn. Am. Phys., 1914, 19, 332. IMMUNITY IN TUBERCULOSIS 
1045 
battle rather late and take up a position somewhat to the rear. Doubtless 
they are of most assistance as wall builders and may contribute to the 
production of the fibrous sheath. At all events, the fact that tubercle bacilli 
find lymphoid tissue their choice of location and operation indicates that 
there can be little actual antagonism between the tubercle bacillus and 
small lymphoid cells. The bacillus exercises an attraction for them—a 
positive chemotaxis similar to the positive chemotaxis for these cells dis- 
played in syphilis and other chronic infections. 
It would appear, therefore, that, in primary tuberculosis, when bacilli 
of sufficient virulence have succeeded for the first time in invading our tissues 
and producing tubercle, the resistance offered them is almost purely cellular. 
Glandular secretions may sweep them away mechanically, but are not 
bactericidal. Doubtless the blood is feebly restraining and destructive as 
it is for most organisms, but these activities are of little importance. The 
lymph apparently is even less antagonistic. But after the tubercle has been 
formed an increased resistance to reinfection becomes manifest in which the 
cells again play the major role in resistance, but in which the body fluids 
also participate in a more important manner. 
Now the invasion of tubercle bacilli is followed by an accelerated inflam- 
matory reaction characterized by hyperemia, cellular, and more particularly 
by serous exudation. These changes are a result of the primary infection 
and may occur in tissues remotely removed from the primary lesion; they 
occur only when a focus of living tubercle bacilli has been established, and 
decrease with healing, but seldom disappear entirely because the complete 
destruction of all living bacilli is of infrequent occurrence. 
This reaction to reinfection or the administration of dead bacilli or their 
products (tuberculins) is the allergic reaction of tuberculosis. Sensitiza- 
tion of the tissues of the tubercle and of the tissues in general cannot be 
brought about by anything short of living tubercle bacilli, and hence the 
failure of prophylactic immunization against tuberculosis by vaccines of 
dead bacilli and tuberculins. Indeed, as shown by Romer and Joseph,1 
acquired resistance requires the presence of a focus of infection and may 
not be gained by careful immunization even with vaccines of living bacilli, 
not to mention vaccines of dead bacilli. The mechanism of the reaction 
has been previously discussed in Chapter XXIX and is probably a cellular 
colloidal phenomenon. 
Krause has regarded this reaction as an important means of defense 
against reinfection. In all probability the allergic antibodies in the cells 
are not actually destructive for tubercle bacilli, but the inflammatory reac- 
tion is, and probably by reason of facilitating phagocytosis and to some 
extent by bringing into play plasma of enhanced tuberculocidal activities, 
as at this time small amounts of various antibodies may be found in the 
blood. Krause and Hofer2 have found that bacilli injected into the perito- 
neal cavity of tuberculous guinea-pigs were quickly destroyed by lysis; 
Manwaring and Bronfenbrenner3 have also studied this phenomenon and 
found that whereas tubercle bacilli injected into the peritoneal cavities of 
normal guinea-pigs may be recovered, those injected into tuberculous 
animals sometimes underwent lysis ascribed to acquired changes in the fixed 
peritoneal cells. In so far as resistance and immunity are concerned the 
most important of these antibodies are opsonins and lysins, while agglutinins 
and precipitins may aid in both processes of phagocytosis, intracellular and 
extracellular lysis. 
1 Beitr. z. Klin. d. Tuberk., 1910, 17, 281. 2 Deutsch. med. Wchn., 1912. 
3 Trans. Nat. Assoc. Study and Prevent. Tuberculosis, 1913, 321. 1046 
BIOLOGIC THERAPY OF TUBERCULOSIS 
Antibodies in the blood, however, are probably never the important 
factors of acquired resistance to tuberculosis. Complement-fixation and 
other tests have shown that these seldom reach a high point of production. 
The chief factors are apparently the cells and phagocytosis. 
Probably the majority of human beings after fifteen years owe their 
freedom from clinical tuberculosis to the presence of small latent foci bring- 
ing about this sensitization and accelerated and abortive reactions upon 
reinfection. In 1886 Marfan1 announced the following “law” in tubercu- 
losis: “One almost never finds pulmonary tuberculosis, at least manifest 
and a progressive disease, in people who in infancy have been the subjects 
of scrofula (suppurative tuberculous adenitis of the neck) and who have 
been completely cured of this before the age of fifteen, such cure having 
taken place before any other focus of tuberculosis was discoverable.” 
Children may be overwhelmed before the mechanism is well under way, 
and in both children and adults this resistance is readily enough broken 
down by massive reinfection and auto-infection, as from a broken-down 
focus followed by the discharge of large numbers of virulent bacilli which 
are now probably endowed with more effectual means of resisting the defen- 
sive agencies of the host. Calmette2 has observed that cattle nearly always 
recover from a single infection if carefully isolated, whereas they rarely 
recover, but become actively tuberculous, if they are infected several times 
at short intervals, or if they are left in prolonged contact with tuberculous 
animals. Calmette concludes that the consumptive is one who has received 
these repeated and successive reinfections capable of overcoming the resist- 
ance acquired with the primary infection. 
In syphilis we have seen that when the disease is established there is 
apparently an almost absolute resistance to reinfection until cure is accom- 
plished, when susceptibility is restored, but, unfortunately, this is not so in 
tuberculosis. Instead of nearly absolute acquired resistance the resistance 
is only relative. While the syphilitic is highly resistant to reinfection, he 
is not immune to progressive involvement of his own tissues by Spirochaeta 
pallida; the tuberculous individual is much less resistant to reinfection, 
but his primary lesion may not progress and, indeed, in most individuals 
does not do so to any great extent. In this respect tuberculosis is much 
less dangerous. In both diseases the parasites may survive indefinitely, 
but in tuberculosis are more effectually restrained than occurs in syphilis. 
It is very difficult to determine whether this acquired resistance to 
tuberculous reinfection persists after destruction of all tubercle bacilli in 
the body, and if so, how long it persists. As previously stated, the resistance 
is not transmitted in utero and can be passively transferred by injection of 
serum only to a slight, uncertain, and temporary degree. It is a resistance 
dependent upon the sensitizing activity of some allergen associated in some 
way with living bacilli in the tissues, but it is highly probable that the pro- 
tective tissue sensitization persists in a gradually decreasing degree for some 
time after complete destruction or removal of the living bacilli, and to this 
extent it may be stated that an attack of tuberculosis confers a certain 
measure of temporary immunity after recovery. Since living bacilli have 
been found in calcified foci it is highly probable that complete destruction 
of all tubercle bacilli even in a small isolated lymph-gland is the exception 
rather than the rule, and may never occur at all; we are usually unable to 
destroy the invading enemy, but we may protect ourselves against his 
multiplication and further invasion and against superinfection. 
1 Arch. gen. de med., 1886, 1. 
2 L’Infection Bacillaire et la Tuberculose, Masson and Cie, Paris. PROPHYLACTIC IMMUNIZATION IN TUBERCULOSIS 
1047 
With a disease so wide-spread as tuberculosis it is natural that early and 
persistent attempts should be made for developing a practical and efficient 
means of prophylactic immunization. Without doubt this accomplishment 
would be among the greatest of triumphs in medicine. 
Koch originally stated that only living tubercle bacilli producing a mild 
infection could accomplish immunization, and the sum and substance of a 
great deal of investigation since then, recently summarized by Neufeld,1 
has only served to emphasize the truth of this assertion. 
As stated above in the discussion of the nature of resistance and immunity 
in tuberculosis, the presence of a lesion is required to bring about the sensi- 
tization of the tissues necessary for the enhanced abortive and inflammatory 
reactions regarded as allergic and capable of offering resistance to reinfection 
with bacilli of reduced virulence or within certain small dosage. Even this 
degree of resistance, which is the best obtainable in tuberculosis, is not 
always able to hold the individual’s own infection in check, and may be 
broken down or overcome by infection with “massive” numbers of tubercle 
bacilli or by small numbers of very virulent bacilli. 
Injections of dead bacilli and extracts do not produce this protective 
sensitization of the tissues; they may bring about anaphylactic sensitization 
to the proteins of the tubercle bacillus, but this process is not protective, 
and anaphylactized animals are susceptible and, indeed, may be somewhat 
more susceptible to infection than normal control animals. This subject 
has been discussed in more detail in the chapters devoted to Anaphylaxis 
and Allergy. 
If an actual focus of disease so small as to be symptomless is capable 
of engendering some degree of resistance to tuberculosis, what is to be 
expected of immunization with non-virulent living tubercle bacilli or of 
virulent bacilli in very small numbers? 
Probably the first work done with the living bacillus was that of Dixon2 
with attenuated cultures. This worker found that animals inoculated 
with an old culture containing club-shaped and branching forms of tubercle 
bacilli would resist subsequent inoculation with virulent organisms. Since 
then numerous investigators—Trudeau,3 Pearson,4 de Schweinitz,5 McFad- 
yean,6 Levy,7 Pearson and Gilliland,8 Behring,9 Thomassen,10 Neufeld,11 Theo- 
bald and Smith,12 Webb and Williams,13 and others—have, either directly or 
indirectly, supported this original work, thus indicating that the most 
effectual active immunization is secured by using living cultures. The 
method employed by Webb and Williams is worthy of special mention, inas- 
much as the ascending doses of bacilli are actually counted by an ingenious 
method devised by Barbour.14 These observers used this method quite ex- 
PROPHYLACTIC IMMUNIZATION IN TUBERCULOSIS 
1Ztschr. f. Tuberkulose, 1921, 35, 11. 
2 Medical News, Philadelphia, October 19, 1889. 
3 Amer. Jour. Med. Sci., August, 1906; June, 1907; New York Med. Jour., July 23,1893; 
Medical News, October 24, 1903. 
4 Proc. First Internat. Vet. Congress, 1893. 
5 Medical News, December 8, 1894. 
6 Jour, of Comparative Path, and Therap., June, 1901; March, 1902. 
7 Centralbl. f. Bakt., 1903. 
8 Phila. Med. Jour., November 29, 1902; Univ. of Penna. Med. Bull., 1905. 
9 Beitr. z. exper. Therap., Reft. s. Marburg, 1902. 
10 Recui1 de Med. Vet., January 15, 1903. 
11 Deutsch. med. Wchn., September 1, 1902; April 28, 1904. 
12 Jour. Med. Research, June, 1908. 
13 Trans, of the Sixth Internat. Congress on Tuberculosis, 1908; Jour. Med. Research, 
1911. xix 1. 14 The Kansas Univ. Sci. Bull., 1907, iv, No. 1. 1048 
BIOLOGIC THERAPY OF TUBERCULOSIS 
tensively with lower animals, and have also secured good results with per- 
sons willing and anxious to take all possible risks for the possible good that 
may result. In no case have harmful results followed the injections. 
Naturally, most experiments on prophylactic immunization against 
tuberculosis have been conducted with the lower animals, and notably 
cows, by reason of the special value such experiments possess from the 
standpoint of milk production and the prevention of bovine infection of 
human beings. 
Von Behring has employed on an extensive scale the immunization of 
cattle with intravenous injections of living human bacilli because these 
organisms are usually non-virulent for adult cattle, although calves are 
sometimes infected. The outcome of these experiments showed that some 
degree of resistance was attained, but that it was temporary and apparently 
dependent upon the duration of the living bacilli of the vaccine in the tissues. 
Theobald Smith1 in repeating experiments of this kind also noted the develop- 
ment of a marked degree of resistance to subsequent infections with bovine 
bacilli, but likewise found that the resistance was temporary and that the 
method was not without danger, inasmuch as the living bacilli of the vaccine 
may persist for unknown periods of time in the tissues and in the presence 
of animal parasites, injuries and inflammatory lesions due to other infections 
multiply and produce tuberculous lesions. 
A very complete review of the literature on prophylactic immunization 
of cattle is given by Gilliland2; also an account of the extensive experiments 
conducted by him and his associates in Pennsylvania upon the immunizing 
value of intravenous injections of tubercle bacilli from human sources non- 
virulent for cattle. The conclusions drawn from this extensive and very 
carefully conducted investigation are as follows: 
I. Intravenous injections of tubercle bacilli from human sources non- 
virulent for cattle are capable of conferring an immunity in cattle against 
tuberculosis sufficient to withstand natural infection by association with 
tuberculous cows. 
II. The length of the immunity conferred has not been determined 
definitely, but it is believed to gradually diminish after two and one-half 
years. 
III. The vaccinated animal during the period of vaccination and for 
some weeks afterward is more liable to contract tuberculosis than a normal 
animal. The natural resistance of the animal is apparently lowered during 
the time of vaccination. 
IV. The interval between the vaccinations should be of a sufficient length 
to allow any reaction following the previous vaccination to entirely subside. 
V. The degree of immunity obtained in the animal depends to a certain 
extent upon the number of vaccinations and the amount of vaccine admin- 
istered. 
VI. The vaccine should be prepared so it contains no clumps of bacilli 
and should be administered fresh. 
VII. A number of the vaccinated animals may give a typical tuberculin 
reaction following the vaccinations for a period of twenty months. These 
animals may or may not show lesions of tuberculosis at autopsy. 
VIII. Vaccine administered to animals already infected with tuberculosis 
is capable of retarding or holding in check the progress of the disease. 
IX. The milk from immunized cows when fed over a long period of time 
appears to increase the resistance of calves and pigs. 
1 Jour. Amer. Med. Assoc., 1917, 68, 764. 
2 Circular No. 32 of the State Livestock Sanitary Board of Pennsylvania, Harrisburg, 1915. PROPHYLACTIC IMMUNIZATION IN TUBERCULOSIS 
1049 
X. Vaccination of calves against tuberculosis is of assistance in the 
eradication of tuberculosis from a herd if done under the proper conditions. 
XI. Until further knowledge is obtained in regard to the destruction or 
outcome of the living tubercle bacilli constituting the vaccine, no practical 
method for the immunization of milk-producing animals under ordinary 
conditions can be advocated. 
Recently Calmette and Guerin1 have reported that immunization of 
heifers with a vaccine of living bovine bacilli attenuated by prolonged 
cultivation on media containing bile confers an immunity to infection by 
swallowing living bacilli from cattle or by intravenous injection. Calmette 
believes that human tuberculosis is usually infection with bovine bacilli 
and that the usual route of infection is gastro-intestinal. A great deal of 
work the world over has shown, however, that the majority of cases of 
tuberculosis in human beings are due to human types of bacilli, and that 
infections with bovine bacilli are largely met with in children. However, 
their claim that immunization of cattle with this vaccine of attenuated 
bacilli protects animals against tuberculosis and that if it is used upon a 
sufficiently extensive scale, tuberculosis in cattle will disappear, and that 
with its disappearance tuberculosis in man will diminish, and finally dis- 
appear, cannot be said to be proved by any means. The number of animals 
employed by them was far too small for such sweeping statements, but the 
results are noteworthy and encouraging and should stimulate further work 
along similar lines. 
Within recent years the profession and laity have also been agitated by 
the extravagant claims of Friedmann for a vaccine of living, acid-fast bacilli 
derived from the turtle. This culture is said to be avirulent for human 
beings, and to be capable of stimulating specific antibodies and thus bringing 
about a cure. The unfortunate methods by which these claims have been 
exploited, and the investigations by Anderson and Stimson2 and others 
tending to show that no good has followed its use, and that the vaccine is 
not entirely harmless, preclude any statements at this time except to state 
that the vaccine has not demonstrated powers of prophylactic immunization 
in well-controlled experiments. 
Other non-virulent living vaccines have been advocated for prophylactic 
immunization of human beings and cattle. It is claimed that their adminis- 
tration is followed by the development of antibodies, but none have withstood 
the acid test of preventing infection with virulent bacilli with sufficient 
regularity or for sufficient lengths of time to render them of practical value. 
The principle involved, however, namely, the use of a living vaccine com- 
posed of bacilli so altered by passage through another animal or by chemical 
and physical agents, that they cannot produce tuberculosis, but yet resemble 
the infecting bacillus closely enough to produce protection, is sound, and 
should stimulate further investigation in this direction. Certainly enough 
work has been done to show that very little or nothing of value is to be 
expected of vaccine of dead bacilli and various tuberculins. Just as in 
vaccination against smallpox the living organism is required either in the 
form of a purposely induced mild infection or vaccination with a modified, 
but living virus, the same are apparently required for immunization in 
tuberculosis and syphilis. The first procedure is out of the question because 
the infection cannot be controlled, and the second, up to the present time, 
has been found to confer only temporary resistance and to be not without 
some danger. 
1 Ann. de l’lnst. Pasteur, 1911, 25, 625; ibid., 1920, 34, 553. 
2 Hygienic Lab. Bull. No. 99, 1914. 1050 
BIOLOGIC THERAPY OF TUBERCULOSIS 
THE PRINCIPLES OF TUBERCULIN THERAPY 
Historic.—Within the last twenty years the subject of tuberculin therapy 
has elicited considerable discussion in the diagnosis and treatment of tuber- 
culosis. The wide-spread prevalence of the disease not only in man, but 
in the lower animals as well, the distressing symptoms, the gloomy prognosis, 
and the economic importance it possesses, are a few of the factors that have 
stimulated investigators the world over to zealous and persistent efforts 
directed toward discovering a means of preventing and curing this great 
scourge. Owing to the nature of the infection, which covers relatively long 
periods of time, and the fact that much time is required for the conduct of 
experimental studies, researches are of necessity prolonged, tedious, and 
difficult. Within a period of a few years after the cause of syphilis had been 
discovered and isolated valuable diagnostic reactions and a well-nigh specific 
therapy were discovered. The discovery of an early and specific diagnostic 
and therapeutic measure for tuberculosis will achieve still greater triumphs 
—in fact, few could be greater or more beneficial. 
Koch was the first to note the curative action of tuberculin, and it may 
be well to refer here to the original description of his fundamental experi- 
ments,1 which have been the basis as well as the starting-point of the entire 
study of tuberculins: 
“When one vaccinates a healthy guinea-pig with a pure culture of tubercle 
bacilli the wound, as a rule, closes and in the first few days seems to heal. 
However, in from ten to fourteen days a hard nodule appears which soon 
breaks down, leaving an ulcer that persists to the time of death of the animal. 
There is quite a different sequence of events when a tuberculous guinea-pig 
is vaccinated. For this experiment animals are best suited that have been 
successfully infected four to six weeks previously. In such an animal the 
inoculation wound likewise promptly unites. However, no nodule forms, 
but on the next or second day after a peculiar change occurs. The point of 
inoculation and the tissues about, over an area of from 0.1 to 1 cm. in diame- 
ter, grow hard and take on a dark discoloration. Observation on subsequent 
days makes it more and more apparent that the altered skin is necrotic. 
It is finally cast off, and a shallow ulceration remains, which usually heals 
quickly and permanently without the neighboring lymph-glands becoming 
infected. Inoculated tubercle bacilli act very differently then upon the 
skin of healthy and tuberculous guinea-pigs. This striking action is not 
restricted to living tubercle bacilli, but is equally manifested by dead bacilli, 
whether they be killed by exposure to low temperature for a long time or 
to the boiling temperature, or by the action of various chemicals. 
“After having discovered these remarkable facts, I followed them up 
in all directions and was further able to show that killed pure cultures of 
tubercle bacilli ground up and suspended in water can be injected in large 
amounts under the skin of healthy guinea-pigs without producing any 
effect other than local suppuration. Tuberculous guinea-pigs, on the other 
hand, are killed in from six to forty-eight hours, according to the dose given, 
by the injection of small quantities of such a suspension. A dose which 
just falls short of the amount necessary to kill the animal may produce 
extensive necrosis of the skin about the point of injection. If the suspension 
be diluted until it is just visibly cloudy the injected animals remain alive, 
and if the administration is continued with one- to two- day intervals, a rapid 
improvement in their condition takes place; the ulcerating inoculation 
wound becomes smaller and is finally replaced by a scar, a process that 
1 Deutsch. med. Wchn., 1891, xvii, 101. THE PRINCIPLES OF TUBERCLIN THERAPY 
1051 
never occurs without such treatment; the swollen lymph-glands become 
smaller; the nutrition improves, and the disease process, unless it is too far 
advanced and the animals die of exhaustion, comes to a standstill. 
“Thus was established the basis for a rational treatment of tuberculosis. 
However, such suspensions of killed tubercle bacilli are unsuitable for prac- 
tical use, since they are neither absorbed nor disposed of in other ways, but 
remain a long time unaltered at the point of inoculation and occasion smaller 
or larger abscess.” 
Koch showed further that while the injection of tuberculous guinea-pigs 
with large doses of tubercle bacilli produced rapid death, frequently repeated 
small doses exerted a favorable effect upon the site of infection and the 
general condition of the animals. The same observer also realized that the 
harmful effects of injections of dead tubercle bacilli were due to the non- 
absorbable parts of the bacilli. He attempted to extract the immunizing 
substances, and in this way produced his first or old tuberculin. When 
injected into tuberculous guinea-pigs, old tuberculin produced a rapid 
general reaction without any local necrosis or sloughing, whereas when in- 
jected into a healthy guinea-pig, no reaction, either local or general, was 
produced. The fact that the general results produced by old tuberculin 
were analogous to those obtained by his first vaccine, except that local 
necrosis did not occur, induced Koch, in 1891, to promulgate it as a specific 
cure for tuberculosis in human beings. 
It is hardly necessary to describe the hopeful anticipation with which 
it was received, and the keen disappointment that followed its earlier clinical 
use. Indiscriminate use, extravagant expectations, and excessive dosage 
combined to yield results so discouraging as to swing the pendulum of 
medical opinion so far the other way that even now the very word “tuber- 
culin” suggests to many minds failure and something to be avoided. 
A few earlier followers of Koch continued their studies in the endeavor 
to discover the causes of failure in tuberculin therapy. Their researches 
have led to new principles in treatment and to more exact knowledge of 
its indications, as well as its contraindications. As now employed, its use 
being restricted to suitable patients and administered in safe graduated 
doses, and accepting as evidence only the statements of those who have used 
tuberculin and not of those who believe it to be dangerous and have never 
used it, one deduction is justified: that while tuberculin is not a specific 
“cure” for tuberculosis—any more than hygiene, diet, and climate are 
cures—it helps to arrest the disease and is in general a useful factor in the 
treatment of certain types of the disease. Clinical studies have shown, 
however, that immunization of the tuberculous patient is frequently a 
difficult porcedure, owing to the fact that such patients are prone to develop 
a remarkable state of hypersusceptibility, in consequence of which every 
inoculation will produce a reaction that may be injurious to the patient. 
Nomenclature.—All extracts and suspensions of the tubercle bacillus 
are commonly designated as tuberculins. The term was first employed in 
1884 by Pohl Pincus.1 Koch himself first called his old tuberculin “lymph” 
and it was known for some time as “Koch’s lymph.” Bujwid2 is quoted 
by Reeser3 as being first to apply the term “tuberculin” to Koch’s original 
extract and which Koch later adopted for his preparation. 
Kinds of Tuberculins.—The knowledge that tubercle bacilli and their 
1 Quoted by Riviere and Morland, Tuberculin Treatment, Frowde, Hodder, and Stough- 
ton, London, 1913, 31. 
2 Gazeta lekarska, 1891, No. 4. 
3 Centralbl. f. Bakteriol., orig., 1908, 46, 56. 1052 
BIOLOGIC THERAPY OF TUBERCULOSIS 
secretions as seen in vilro contain both desirable and undesirable substances 
has led Koch and others to adopt different methods of preparing tuberculin 
in the endeavor to obtain the desirable immunizing principles in as pure a 
state as possible. 
As a consequence, a large number of preparations have been advocated 
from time to time, all of which are said to possess some special properties 
and virtues. All tuberculins, whatever their mode of preparation and 
manufacture, are derived from cultures of the tubercle bacillus. So numer- 
ous have the tuberculins become, and so superior are the advantages claimed 
for each new product over the older ones, both for diagnostic and for thera- 
peutic purposes, that only a few of those possessing special interest and 
value can here be described. 
From a practical standpoint the different tuberculins may be classified 
into three main groups as follows: 
1. Those containing only the soluble products of the tubercle bacilli in 
the broth medium in which they are grown. The best known example is 
Koch’s old tuberculin (O. T.). 
2. Those consisting essentially of the insoluble fragments of the bacilli, 
as Koch’s new tuberculin or tuberculin residue (T. R.). 
3. A combination of the first two, as Koch’s bacillen emulsion (B. E.). 
The best known and most widely employed tuberculins are the following; 
they are usually designated by symbols: 
O. T. = Koch’s old or original tuberculin. 
T. R. = Koch’s new tuberculin; Tuberculin Riickstand or Tuberculin 
Residue. 
B. E. = Koch’s Bacillen Emulsion. 
B. F. = Deny’s Bouillon Filtrate. 
T. B. k. = Beraneck’s Tuberculin. 
P. T. O. = Spengler’s Perlsucht Tuberculin. 
A. F. = Albumose-free old tuberculin. 
Preparation of Tuberculins.—1. Old Tuberculin (0. T.).—This is Koch’s 
original tuberculin, and is the variety regarded by many as the most useful 
both in diagnosis and in treatment. Its manufacture was based upon the 
principle that the toxins elaborated by the bacilli into the culture-medium 
or liberated by disintegration of the bodies were chiefly concerned in stimu- 
lating body cells to the formation of antibodies. Since the bacillary bodies 
were regarded as mainly responsible for the production of abscesses at the 
point of inoculation, they are eliminated by a process of filtration. 
Old tuberculin is prepared as follows: Large shallow flasks containing 
5 per cent, of glycerin alkaline broth are inoculated with a culture of human 
tubercle bacilli and grown at body temperature for from six to eight weeks, 
at the end of which time the bacilli have grown into a flat sheet covering 
the surface of the fluid (Fig. 193). The entire contents are then subjected 
to a current of stream over a water-bath for the purposes of sterilization and 
for concentration into one-tenth of the original volume. The glycerin, 
which is not evaporated, thus constitutes 50 per cent, of the resulting mixture. 
The bacilli are removed by filtration through a Berkefeld or Chamberland 
filter. The filtrate is a clear, brown fluid, of a characteristic odor, which 
keeps indefinitely and is ready for use. 
Koch considered the soluble toxins of the bacillus as the desirable im- 
munizing agents, and believed that the endotoxins were responsible for the 
necrotic effects. Since, however, it was accepted that bacteriolytic sub- 
stances would be formed only after the injection of intact or fragmented 
tubercle bacilli—with their contained endotoxins—Koch added T. R. and THE PRINCIPLES OF TUBERCULIN THERAPY 
1053 
later B. E. to his list, in order to make the production of antibacterial sub- 
stances still more complete. Furthermore, in order to obtain as varied a 
supply of antibodies as possible the use of several tuberculins, such as old 
tuberculin and bacillus emulsion, was recommended for use in the same 
patient. 
2. New Tuberculin (Known as T. R. or Tuberculin Residue).—This 
was the next tuberculin to be promulgated by Koch,1 and is prepared as 
follows: Virulent cultures of human tubercle bacilli are grown in flasks 
of nutrient glycerin broth for from four to six weeks, the bacilli being then 
filtered off and dried in a vacuum. One gram of the dried tubercle bacilli 
is ground in an agate mortar until all the bacilli have been broken up. To 
the pulverized mass 100 c.c. of distilled water are added, and the mixture 
is then centrifugalized. The clear supernatant fluid is poured off, and is 
now known as Tuberculin Oberes (T. O., not to be confounded with O. T.). 
It contains substances not precipitable by glycerin. The sediment is again 
dried, powdered, taken up in a small amount of water, centrifuged, the 
supernatant fluid poured off, and the process repeated until no sediment is 
Fig. 193.—Preparation op Tuberculin. 
A flask of bouillon culture of tubercle bacilli (three to four weeks). Note the surface layer of bacilli 
with stalactite formations. 
precipitated except that composed of gross accidental particles. The fluids 
resulting from all the centrifugalizations, except the very first, are poured 
together and the total should not measure more than 100 c.c. This opales- 
cent fluid is preserved with 20 per cent, of glycerin and is known as T. R. 
It should contain in each cubic centimeter 2 milligrams of solids, representing 
10 milligrams of dried tubercle bacilli. 
3. Bacillen Emulsion (B. E.).—This was a still later form of tuberculin 
made by Koch,2 and, as its name indicates, it is an emulsion of tubercle 
bacilli. The culture is grown as for O. T.; the bacilli are filtered off, ground, 
but not washed, and 1 part of the pulverized material emulsified in 100 
parts of distilled water; an equal part of glycerin is then added, making a 
50 per cent, glycerin emulsion, each cubic centimeter of which contains the 
immunizing substance of 5 mg. of dried tubercle bacilli. Since B. E. was not 
washed, it was assumed that it would retain all extractives and the entire 
contents of the bodies of the tubercle bacillus. 
1 Deutsch. med. Wchn., 1897, xxiii, 209. 
2 Deutsch. med. Wchn., 1901, xxvii, 839. BIOLOGIC THERAPY OF TUBERCULOSIS 
1054 
While Koch was preparing these various tuberculins others were being 
made, one of which, prepared by Denys, is used extensively at present in 
the treatment of tuberculosis. 
4. Bouillon Fillrale (B. F.).—This is practically Koch’s old tuberculin 
unheated. It is prepared by Denys1 in the same manner as O. T., except 
that the bacillus-free filtrate—a clear fluid said to contain only the soluble 
secretions of the bacilli plus the metabolized culture-medium—is not heated 
or concentrated and is ready for use without any further modification. 
5. Beraneck’s Tuberculin.2—This is a preparation for which its inventor 
claims only minimal toxicity and a high content of specific substances. 
It is prepared by cultivating the bacilli on a non-peptonized 5 per cent, 
glycerin-bouillon medium, which is not neutralized. The filtrate from this 
culture is known as T. B., or toxin bouillon. The residue is shaken for a 
long time at from 60° to 70° C. with 1 per cent, orthophosphoric acid. Equal 
volumes of the unheated toxin bouillon and of the acid extract of the bacillary 
bodies are combined to form the whole tuberculin. 
6. Spengler’s Perlsucht Tuberculin (P. T. 0.).—This tuberculin is 
prepared by Spengler3 with bovine bacilli from cattle suffering with pearl 
disease in exactly the same manner as Koch’s old tuberculin employing 
bacilli from human lesions. Spengler believes that this tuberculin is indi- 
cated in the treatment of human beings since bovine bacilli are less virulent. 
7. Albumose-free Old Tuberculin (A. F.).—This tuberculin is Koch’s 
old tuberculin from which the albumoses have been removed by precipita- 
ting with six volumes of absolute alcohol. After standing twenty-four hours 
the supernatant is filtered and evaporated at 56° C. over a water-bath to 
the original volume of O. T. employed. The fluid is now sterilized by means 
of candle filtration. 
8. Dixon’s Tubercle-bacilli Extract.4—Dixon has prepared a tuberculin 
by treating cultures of tubercle bacilli with ether and extracting in salt 
solution. This has yielded good results in the treatment of tuberculosis. 
This product is prepared from the living organisms. The tubercle bacilli 
are grown on 5 per cent, glycerin veal broth for a period of from six to eight 
weeks at a temperature of 37.5° C. They are removed from the incubator 
and collected on hard filter-paper. The filtrate of glycerin broth on which 
they are grown is discarded. An equal quantity, by weight, of tubercle 
bacilli of the bovine and human type is used. The mass of organisms is 
partially freed from excess of moisture by placing it between two sterile 
filter-papers, after which it is placed in a dish in the incubator for from 
twenty-four to forty-eight hours. The dried organisms are then washed in 
an excess of ether, which is allowed to act until it has removed all the water 
and glycerin. The organisms are then subjected to an excess of fresh ether 
and washed in this for six hours, to soften the wax of the bacilli. This fat 
separates so that it collects at the bottom of the vessel and is removed with 
a Pasteur pipet. After the second addition of ether has been removed the 
mass is allowed to dry until no ether odor is perceptible. After the mass 
of tubercle bacilli has been thoroughly dried it is ground in a mortar and 
suspended in physiologic salt solution in the proportion of 1 part of the 
ground mass to 5 parts of an 8 per cent, salt solution. This suspension 
is shaken in a shaking machine for from eight to ten hours, and is then 
allowed to stand for several days at room temperature. It is finally passed 
1 Le Bouillon Filtre de Bacille de la Tuberculose, Louvain and Paris, 1905. 
2 Rev. med. de la Suisse rom., 1907, 27, 444. 
3 Deutsch. med. Wchn., 1904, No. 31; ibid., 1905, Nos. 31 and 34. 
4 Penna. Health Bull., No. 28, April, 1914. THE PRINCIPLES OF TUBERCULIN THERAPY 
1055 
through impervious bacteria filters several times and the filtrate examined 
microscopically, bacteriologically, and physiologically. Culture tests are 
made to determine its freedom from contaminating organisms, and guinea- 
pigs are inoculated to ascertain that it contains no living tubercle bacilli. 
One cubic centimeter of this extract represents 0.2 gm. of the organisms, 
and is known as the stock solution, from which serial dilutions are made. 
This solution is sterile, but 0.5 per cent, of phenol (carbolic acid) is added 
as a preservative to prevent subsequent contamination. 
While the tuberculins just described are those mainly used, many others 
have been prepared and advocated in the diagnosis and treatment of tuber- 
culosis. The aim is always to obtain the specific substances with as little 
as possible of the toxic substances—not only those concerned in producing 
necrosis, but likewise the protein constituents responsible for specific sensi- 
tizing action and anaphylactic disturbances. For example, tuberculocidin 
and tuberculol are examples of attempts at isolating the pure immunizing 
principle; endotin, or Moeller’s tuberculin, is an example of an endeavor to 
rid the culture fluid of protein substances.1 
The Effects of Tuberculin.—The toxicity of the different varieties of 
tuberculins vary in some degree, that is, the toxic effects for normal non- 
tuberculous human beings and lower animals due to products of the tubercle 
bacilli, the tubercle-proteins and various substances that may be present 
in the culture-medium employed for the preparation of the tuberculin. 
This subject will be discussed in more detail shortly; here it may be stated 
that old tuberculin (O. T.) is least toxic, while those composed of the bacillary 
fragments are apt to be somewhat more irritating or toxic due to the presence 
of protein substances in common with bacterial vaccines in general. 
But all tuberculins are capable of producing local (at the site of injection), 
focal (at the site of disease), and constitutional reactions in the majority of 
tuberculous individuals when given in sufficient amounts, and few subjects 
in medicine are as perplexing as the nature and mechanism of the action 
of the tuberculins despite a very large amount of clinical and laboratory 
investigation. Brought forward by its great discoverer as an immunizing 
and therapeutic agent, we have since learned that its immunizing power is 
almost negligible and its application in treatment surrounded by definite 
limitations. From the very beginning when it led its discoverer grievously 
astray and induced him to make hasty statements and predictions, it has 
continued to baffle some of the best scientific minds in unraveling the nature 
and mechanism of its action. 
In the first place tuberculin is not a poison or toxin because relatively 
enormous amounts may be administered to non-tuberculous animals with 
impunity. Kraus has injected as much as 30 c.c. intravenously to healthy 
guinea-pigs, and Hamburger has given healthy infants as much as 1 c.c. at 
close intervals repeatedly over a long period without ill effects. But the 
injection of one-millionth of 1 c.c. may make a tuberculous adult ill. For 
the tuberculous, therefore, tuberculin acts like a virulent poison; for the 
non-tuberculous it is relatively inert. 
Why the difference? From whence arise the toxic effects in the tuber- 
culous individual if the tuberculin is not toxic? The evidence quite clearly 
indicates that these toxic substances arise from the tuberculous foci, and 
tuberculin stimulates and facilitates their absorption by producing hyperemia 
and other inflammatory changes in the tissues surrounding these foci. Simi- 
lar effects may be produced by the patient by exercise. The ordinary con- 
1 For a full account of these and other preparations I refer the reader to the book of 
Hamman and Wolman, Tuberculin in Diagnosis and Treatment, 1912, Appleton & Co. 1056 
BIOLOGIC THERAPY OF TUBERCULOSIS 
stitutional symptoms of tuberculosis are quite similar to the general tubercu- 
lin reaction, and both are due to the passage of toxic substances from the 
focus of disease into the general circulation. 
The Allergic Nature of the Tuberculin Reaction.—Why and how does 
tuberculin produce this inflammatory reaction around foci of tuberculous 
infection? This is an unsolved mystery, but evidence indicates that it is an 
allergic phenomenon— local allergic shock as discussed in Chapter XXXI. 
The allergic nature of the tuberculin reaction has been generally held 
since this view was proposed by Wolff-Eisner. It is commonly believed 
that as a result of infection an antibody is produced in the blood of the 
nature of an amboceptor or ferment which acts upon the injected tuberculin 
with the production of a poison (anaphylatoxin) responsible for the general, 
focal, and local (skin) reactions. But there are few or no facts to substan- 
tiate this theory. In the first place the presence of this antibody has not 
been conclusively demonstrated in the blood of tuberculous individuals or 
lower animals. An allergic antibody is produced, but it apparently is 
cellular—that is, upon or within the cells of the body rather than free in the 
blood. Even though it were free in the blood and capable of producing a 
poison by digestion of tuberculin, it is impossible to conceive how sufficient 
poison could be obtained from a thousandth or millionth part of a cubic 
centimeter to produce the profound changes sometimes following injection. 
Anaphylaxis is commonly regarded as caused only by protein substances. 
The chemical nature of tuberculin is biuret free and dialyzable; if it is a 
protein, evidence indicates that it is a polypeptid or proteose, and anaphy- 
laxis of the lower animals cannot be produced regularly with proteoses. 
Many investigators have stated that proteoses could not induce anaphylaxis 
at all, but recent investigations indicate that weak reactions are sometimes 
observed, but granting this, attempts to produce anaphylaxis in animals 
with tuberculin have generally failed. Tuberculin does not appear to 
produce anaphylactic sensitization and reactions. 
How then can the tuberculin reaction be regarded as an allergic phenome- 
non? The tuberculin reaction can only be produced with regularity by 
active tuberculosis. Living tubercle bacilli must be introduced and actual 
lesions result before tuberculin hypersensitiveness is produced. This 
suggests that allergic sensitization is engendered by a substance elaborated 
only by living tubercle bacilli in living tissues. Is this allergen the same 
tuberculin as produced in culture-media in the test-tube? Either it is the 
same or it is a very closely allied combination product of tuberculin and 
tissue cells because the tuberculin reaction is elicited by the injection of 
artificial tuberculin and the allergic reaction is highly specific—that is, the 
substance eliciting the reaction is the same or very nearly so as that engen- 
dering sensitization. If this is true, the failure to sensitize normal animals 
writh artificial tuberculin is due either to some undefined biologic difference 
between focus tuberculin and artificial tuberculin, or we do not know how to 
sensitize animals artificially with tuberculin. The latter assumption, how- 
ever, is less significant than the former because we can with comparative ease 
sensitize and anaphylactize guinea-pigs with proteins of the tubercle bacillus. 
The allergen, therefore, is produced in the foci of tuberculosis; it gradually 
brings about an exquisite degree of active sensitization of the tissue cells 
not only surrounding the focus, but of distant organs as well, including the 
skin and mucous membranes. As a result of sensitization these sensitized 
cells contain the allergic antibody. When tuberculin is injected it unites 
with these sensitized cells, producing cellular allergic shock. This shock is 
characterized by its effect upon the vasomotor system, as is true of other THE PRINCIPLES OF TUBERCULIN THERAPY 
1057 
types of allergy in man, as serum disease, hay-fever, and allergic asthma. 
One effect is hyperemia or dilatation of the vessels accompanied by serous 
and cellular exudation. If tuberculin has been injected subcutaneously 
in sufficient amounts this reaction of hyperemia and exudation takes place 
around the foci of tuberculosis, and, depending upon the degree of reaction 
and the amount and permeability of the fibrous capsule, results in the 
absorption of the focus poisons with the production of general symptoms. 
When applied to the skin the effects are local and characterized by a local 
reaction of hyperemia and edema. 
If the local and focal reaction to tuberculin is allergic shock and the 
constitutional reaction the result of increased absorption of poisons from the 
foci, are these reactions of therapeutic value? In the first place, what is 
the nature of the focus poisons? Krause has worked upon this subject 
extensively and believes them to be non-specific poisons produced by dead 
and decayed cells. These cellular substances are highly toxic and can be 
produced from a variety of cells and other protein substances. They are 
non-specific and much of the nature of Vaughan’s protein poison. 
The Curative Activity of Tuberculin.—Has tuberculin a curative value— 
that is, is the tuberculin reaction curative? Can the allergic tuberculin shock 
reaction redound to the benefit of the tuberculous individual? This has been 
and continues to be a question of vital importance, calling forth an enormous 
amount of clinical and laboratory investigation and a vast literature. 
We have seen that the tuberculin reaction is accompanied by the pro- 
duction of hyperemia and serous and cellular exudation. It is commonly 
believed that we recover from tuberculosis by the production of fibrous tissue 
around the foci. The thicker and denser this capsule, the more complete is 
clinical recovery even though the capsule may enclose living bacilli. The 
cutting off of blood- and lymph- supply to the focus by this fibrous capsule 
gradually results in death and liquefaction of the contents, including the 
bacilli. Anything favoring the development of this capsule is, therefore, 
therapeutically advantageous. 
Has the tuberculin reaction this effect? According to Krause, who has 
studied this problem with particular care, it has. It is a well-known principle 
of pathology that chronic passive congestion is followed by fibrosis. Tuber- 
culin in proper dosage and spacing of injections can elicit a series of mild 
allergic shocks about the focus of disease maintaining a more or less constant 
degree of hyperemia. Evidence indicates that this is followed by fibrosis, 
and fibrosis is greatly desired. Therefore, in chronic tuberculosis where 
the lesions are localized in a certain organ or organs, the administration of 
tuberculin is advantageous. In miliary and disseminated tuberculosis 
it may not be, or its curative effects may be too slow to prove of benefit. 
The great mistake made in the early days of tuberculin therapy was the 
administration of too large amounts. These large doses produced profound 
and injurious focal reactions with the absorption of enormous doses of foci 
poisons, with disastrous effects—the lamentable days of the tuberculin 
delirium, from the impressions of which both the profession and the laity 
have not yet fully recovered. 
I believe that tuberculin has a very definite place in the treatment of 
some cases of tuberculosis and especially those of a chronic and localized 
character. But the doses must be small enough and so spaced that the 
degree of focal reaction is mild, but more or less continuous. The physician 
administering tuberculin should possess a good working knowledge of its 
effects, its potencies for harm and good; the beneficial results are not sudden 
or spectacular, but slow and plodding. Patience and skill are required, 1058 
BIOLOGIC THERAPY OF TUBERCULOSIS 
and especially the patience to give a long series of subcutaneous injections, 
and skill in judging the correct dose. In general terms the latter is judged 
according to the amount of absorption of foci poisons because the hyperemia 
cannot be seen in the internal organs; the degree of consitutional effects 
should be slight, as shown by only a degree or two of febrile reaction and 
minor other effects. Finally, it is very difficult to express the effects by 
percentages of cures; individual cases in the practices of physicians may not 
have been benefited, but the effects of mild allergic tuberculin shocks appear 
to be favorable in that they aid in the production of fibrous tissue, and 
suitable patients should always have the possible benefits of this process. 
Antibody Production from Tuberculin Treatment.—Aside from the possi- 
ble curative effects of the tuberculin reaction about foci of tuberculosis from 
hyperemia, exudation, and fibrosis, do the various tuberculins engender 
antibodies possessing curative effects? 
As previously discussed under Immunity in Tuberculosis antibody 
production in this disease does not progress very far. At least the amounts 
of antibodies (opsonins, agglutinins, lysins, etc.) found in the blood are usually 
small and almost insignificant; the injection of immune serum does not 
confer more than a slight degree of passive immunity, and the curative 
effects of antituberculosis sera are to be ascribed more to non-specific 
protein therapy than specific agents. 
Immunization of the lower animals with the various tuberculins as well 
as with ordinary-heat killed and even living non-virulent vaccines is fol- 
lowed by slow antibody production; months and even years are usually 
required and the sera at best contain relatively small amounts of agglutinins, 
opsonins, and complement-fixing antibodies without being markedly tuber - 
culocidal. This is especially true when old tuberculin is employed for 
immunization. In my experience several months are required with in- 
creasing doses every five to seven days for the production of even small 
amounts of complement-fixing antibodies. With bacillary antigens anti- 
body production is somewhat more rapid, but the tubercle bacillus is but 
poorly antigenic, in so far as the production of serum antibodies in the 
guinea-pig and rabbit are concerned. 
It is doubtful if antibodies are produced during the treatment of tuber- 
culosis with tuberculins in sufficient amounts to exert curative effects. 
Insusceptibility or Tolerance to Tuberculin.—It is commonly observed 
that tuberculous individuals develop a tolerance for tuberculin under treat- 
ment. At the beginning the subcutaneous injection of 0.0001 c.c. of old 
tuberculin may produce local, focal, and constitutional reactions, whereas 
after six months of carefully graded and spaced injections doses of 0.01 c.c. 
and even 0.1 c.c. may produce no reactions at all. 
Formerly this increased tolerance was ascribed to the production of 
antibodies for the tuberculin capable of neutralizing the supposedly toxic 
properties of tuberculin. But it is now quite clear that tuberculin is not 
primarily toxic and does not engender the production of antibodies for itself 
and the tubercle bacillus except to a very limited degree. A more reasonable 
explanation which has the support of experimental and clinical data is that 
the tolerance is due to the progressive development of fibrous tissue about 
the tuberculous lesions preventing the escape of foci poisons. 
Doubtless the products of the tubercle bacilli in the lesions are also 
capable of acting as tuberculins and producing the allergic reactions in the 
periphery of hyperemia and exudation; this explains the focal and constitu- 
tional reactions following any procedure unduly disturbing the foci, as 
exercise or work. Healing, which is fibrous encapsulation, gradually increases THE PRINCIPLES OF TUBERCULIN THERAPY 
1059 
the mechanical barrier to absorption of foci poisons, explaining why the 
tuberculous subject is able to progressively increase exercise and work 
without eliciting reactions. 
In addition to this mechanical factor it is possible that a part of the 
increased tolerance to artificial and foci tuberculins is due to allergic desen- 
sitization of the tissues to these substances. In acute fulminating tubercu- 
losis with little or no encapsulation of the lesions the well-known insuscep- 
tibility or tolerance to injections of tuberculin is probably due to a process 
of desensitization of the tissues by the enormous production and unrestrained 
distribution of foci tuberculins. 
Increased Sensitiveness to Tuberculin.—While the administration of 
tuberculin is usually followed by a gradually increasing tolerance for larger 
and larger doses, just the reverse may occur. A patient may stand the 
injection of 0.01 c.c. of old tuberculin without demonstrable focal and 
constitutional reactions and yet some time later the injection of 0.001 or even 
0.0001 c.c. may elicit well-marked reactions. During the tuberculin treat- 
ment of tuberculosis one may find it necessary to make a sharp drop in the 
dosage by reason of the occurrence of this increased sensitiveness. 
A reasonable explanation is that hyperemia persists about the foci longer 
than expected or is greater than anticipated, and for this reason a much 
smaller dose of tuberculin is capable of increasing the hyperemia and absorp- 
tion of the foci poisons. At one time it was supposed that allergic hyper- 
sensitiveness was produced by the injections of' tuberculin which thereby 
added to the degree of sensitization produced by the disease, but it is now 
well proved that old tuberculin does not sensitize or only to a very mild 
and limited degree, although the administration of tuberculins containing 
the bacillary bodies may result in some degree of anaphylactic sensitization 
to tubercle proteins. But this possibility is remote and does not exist 
practically when old tuberculin is being administered. The purely me- 
chanical factors of increased hyperemia of the capsule and absorption of 
focus products are a more acceptable explanation of increased sensitiveness 
to injected tuberculin and the increased susceptibility to the focus tubercu- 
lins, as a result of exercise or other factors disturbing the lesions. 
Non-specific Reactions in Tuberculosis.—Not only injections of tuber- 
culin may elicit focal and constitutional reactions in tuberculosis, but other 
substances as well. If the reactions are due to the escape of foci poisons 
this is not difficult to understand. Anything increasing hyperemia about 
the lesions may be expected to increase the absorption of these toxic sub- 
stances. In the administration of such non-specific agents as proteoses, 
milk, typhoid, and various other vaccines, in amounts capable of themselves 
of eliciting reactions of chills, fever, and leukocytosis, etc., it is natural to 
expect that foci of infection regardless of the kind of organism concerned 
will show more marked reactions than normal and healthy tissues. Work 
and exercise alone apparently increase the degree of hyperemia about 
tuberculous foci and the absorption of the toxic focus substances, and may 
be regarded as non-specific stimulants. 
But these observations do not question the specific action of the tuber- 
culins which may engender local, focal, and constitutional reactions in 
amounts very much less than required of such non-specific agents as peptone, 
milk, and typhoid vaccine administered by subcutaneous injection. 
Choice of Tuberculin for Treatment.—How do the large number of 
tuberculins compare in their effects and therapeutic value? Is there a 
preference for tuberculin of bovine bacilli in the treatment of some cases of 
tuberculosis? 1060 
BIOLOGIC THERAPY OF TUBERCULOSIS 
In answer to the first question, all tuberculins are essentially the same 
in the general effects produced, although varying greatly in their potency, 
which alters the dose, intensity, and duration of the focal reactions. 
Special claims are made for certain tuberculins, but it is the consensus 
of opinion that the focal reactions are the same and that more depends 
on the method of administration than on the kind of tuberculin. Possibly 
something is to be gained by changing tuberculins during the course of 
treatment. For example, after old tuberculin has been given for some 
time it may be well to use B. E. (bacillen emulsion) or some other tuberculin 
containing the bodies of the tubercle bacilli in addition to their extractives. 
Since the active principle in tuberculin is unknown and has not been 
isolated in a pure state, it is difficult to compare the strengths of different 
tuberculins. One way is by determining the minimal amount required to 
kill tuberculous guinea-pigs; another is by quantitative skin tests. 
Von Behring has found old tuberculin forty-two times stronger than T. R. 
(tuberculin residue). T. R. (tuberculin residue) and B. E. (bacillen emul- 
sion) are generally weaker than O. T. (old tuberculin). 
All tuberculins are relatively feeble in direct toxicity; certainly they do 
not appear to be more toxic for normal individuals than most ordinary 
vaccines. But some tuberculins are less toxic than others. Old tuberculin 
is said to contain about 10 per cent, albumoses, and the removal of these 
appears to reduce the toxicity in some degree (the A. F. or “albumose-free” 
tuberculin of the Farbwerke Hochst Company). Baraneck’s tuberculin (T. 
Bk.) is said to be least toxic, followed in order by A. F. (albumose-free tu- 
berculin), B. F. (Deny’s bouillon filtrate), and O. T. (Koch’s old tuberculin). 
A good deal of difference exists among tuberculins in regard to the rate 
of absorption. Extract tuberculins, as O. T., B. F., and A. F., are more 
rapidly absorbed and produce their effects more quickly than bacillary 
suspensions, as B. E. The latter are less likely to produce immediate reac- 
tions, are milder in action, but more difficult to gage in dosage. 
Bovine tubercle bacilli are apparently less virulent for human beings 
than human tubercle bacilli and human bacilli, are less virulent for cattle 
than bovine bacilli. Undoubtedly tuberculosis in human beings is some- 
times due to bovine infections and especially tuberculosis of the mesenteric 
glands of children. For these reasons some observers have advised the use 
of bovine tuberculin for the treatment of tuberculosis of human beings as 
representing a tuberculin of less virulent bacilli. 
Koch, however, saw no difference in the effects of tuberculins made of 
human and bovine bacilli, although one or the other may be somewhat 
stronger depending more upon quantitative cultural differences than a vital 
or essential difference in the active principle of tuberculin. Other observers 
with extensive experience have reached the same conclusion, and for the 
treatment of tuberculosis of human beings tuberculin is usually made of 
human bacillus, and for veterinary purposes bovine bacillus is employed. 
Autogenous Tuberculins.—Owing to the great amount of time and labor 
required for the isolation of individual strains of tubercle bacilli, only a small 
amount of investigation has been conducted in the treatment of tuberculosis 
with autogenous tubercle vaccines or tuberculins. Some observers believe 
that the results have been better, but all tuberculins are similar in their 
effects upon tuberculous lesions bringing about hyperemia, exudation, 
increased absorption of foci tuberculin or other toxic substances and facili- 
tating encapsulation. If tuberculin itself has virtues in the treatment of 
tuberculosis aside from bringing about these allergic focal reactions, the 
patient probably derives more benefit from absorption of his own foci THE PRINCIPLES OF TUBERCULIN THERAPY 
1061 
tuberculin (autotuberculinization) than from the relatively small amounts 
of artificial tuberculin administered. 
Indication for Tuberculin Treatment.—Tuberculin is not primarily toxic 
in the small amounts usually administered, and does not itself add to the 
symptoms of toxemia, but a sufficiently large dose may augment the toxe- 
mia by increasing the absorption of foci poisons. For this reason a distinc- 
tion is to be drawn between those cases with and without auto-intoxication. 
1. Patients afflicted with incipient tuberculosis are proper subjects for 
receiving tuberculin treatment, since it tends to protect them from open 
lesions and relapse, and insures, to a greater degree, their ability to con- 
tinue work. 
2. Advanced and moderately advanced cases may be given tuberculin 
if the nutrition is fair, the febrile reaction mild, the pulse not very rapid, 
and if the treatment is controlled by rest. Old fibroid cases with fair nutri- 
tion are especially suitable, as such patients become capable of moderate 
activity and are much less likely to suffer from relapses. 
3. Cases of tuberculosis of the lymphatic glands, skin, and special organs 
may be benefited by prolonged and careful tuberculin therapy. 
4. Cases of latent tuberculosis, especially the children of infected families 
who are below par physically and show tuberculin hypersensitiveness, with 
indefinite physical signs, are proper subjects for receiving tuberculin treat- 
ment. 
5. The question arises as to whether tuberculin may be administered to 
ambulant patients. Tuberculin should be regarded as but one factor in the 
treatment of tuberculosis, and, as such, should be combined with the best 
therapeutic measures available. Therefore the tuberculin treatment is 
supplementary to rest, hygiene, and fresh air, and the benefit of the sana- 
torium should not be denied to patients, especially to the poorer ones. The 
treatment of a mildly progressive ambulant case should be undertaken only 
when prolonged rest in bed has had no visible effect, and when no measures 
can be devised for administering the tuberculin while the patient is in bed, 
as, e. g., patients who have been at the sanatorium and have returned to 
work, or those who cannot be persuaded to enter a sanatorium or for whom 
no place can be found. The tuberculin treatment of more chronic or local- 
ized tuberculosis may be successfully undertaken in the clinic or office. 
Contraindications to Tuberculin Therapy.—Owing to the increased 
focal hyperemia that follows the injection of tuberculin, hemoptysis has been 
considered a contraindication to its use. In such cases it is well to wait 
for some time at least and begin the injections with very small doses, as the 
ultimate effect, namely, the production of fibrous tissue, may be of great 
aid in prolonging life. 
Various authorities have expressed different views regarding other con- 
traindications, such as marked general weakness, fever, cardiac disease, 
nephritis, epilepsy, syphilis, hysteria, etc. As was stated by Hamman 
and Wolman, these are not contraindications, but unfortunate complications 
that would embarrass any form of treatment. Tuberculin may be given to 
any patient whose resisting powers have not been too much depressed as the 
result of complications. For the beginner in this form of therapy, however, 
it is advisable that he acquire experience by undertaking the treatment of 
uncomplicated cases before assuming the responsibility of treating the more 
difficult ones. 
The fact that the ophthalmic test has been made is no contraindication 
to treatment by tuberculin if the reaction has subsided, since a flare-up 
rarely occurs except after large diagnostic doses (Hamman and Wolman). 1062 
BIOLOGIC THERAPY OF TUBERCULOSIS 
Tuberculin Reactions.—Of most importance in this connection is the 
attitude of the therapeutist toward the question of reactions following the 
administration of tuberculin. He must know whether he does or does not 
wish to obtain symptoms of a tuberculin reaction during the treatment; 
the size of the initial and particularly of subsequent doses will depend upon 
his desire to obtain a reaction or upon his anxiety to avoid it. 
At the present time tuberculin is never used for the purpose of obtaining 
strong reactions, such as Koch originally insisted upon getting. Koch 
administered a dose large enough to elicit a strong constitutional reaction, 
and repeated it at intervals of one or more days until that dose no longer 
produced a reaction, after which a still larger dose was given and the former 
procedure repeated. Many—too many—were unable to pass through this 
therapeutic furnace unsigned, and, in fact, the results obtained led to the 
period known as the “tuberculin delirium,” ending, as Hamman and Wolman 
stated, “to the consequent downfall of the arrogant therapy to an humble 
position, whence it is but just emerging, chastened and refined, to assert its 
modest but now truthful claims to a therapy less spectacular, but more 
healing, less forceful, but more gently persuasive: healing a few, helping 
many, and hurting none.” 
While there is this general unanimity of opinion regarding the harmful 
effects of strong reactions, yet tuberculin therapeutists may be divided into 
those who scrupulously avoid all reaction, those who are a little bolder and 
do not object to a very mild reaction, and to an intermediate class. In the 
first group are Sahli, Denys, and Trudeau; the former claims that it is 
essential, first of all, to do no harm, and that cases treated cautiously attain 
a tolerance for high doses as soon as, and even sooner than, those that are 
rushed. Petruschky, Spengler, Lowenstein, Moller, Pickert, White, and 
others are exponents of the bolder method, believing that, by proceeding 
very cautiously, much time is wasted and not enough focal reaction is pro- 
duced to promote healing. To the intermediate class, and approaching 
rather the timid class, are Bandelier and Roepke, Hamman and Wolman, 
M. S. Cohen, and many others. These observers adopt no scheme of fixed 
dosage, but study the individual patient, remembering what is to be avoided, 
rather than the high dose to be reached. 
The constitutional symptoms of a reaction are: temperature, loss of weight, 
rapid pulse, and general symptoms, such as malaise, headache, chilliness, 
arthritic pains, gastric or intestinal disturbances, nausea, loss of appetite, 
insomnia, and skin eruptions. 
Of these general signs, the most important are fever, loss of weight, and 
symptoms of general depression. The temperature should be taken for 
several days preceding the initial dose, so that the patient’s “normal” limits 
are known before the tuberculin is given. When the usual maximum 
temperature is reached, it is especially necessary to watch closely for addi- 
tional signs of reaction. Without some constitutional disturbance a slight 
pyrexia is of less significance, and it is to be remembered that tuberculous 
patients may have a flare-up of fever when they are not being treated with 
tuberculin. Denys refuses to consider any temperature reaction as due to 
tuberculin that comes on more than forty-eight hours after the injection has 
been given. It is characteristic of tuberculin reactions that the rise is abrupt, 
and not in step-like progression. Hamman and Wolman would hesitate to 
ascribe an elevation coming on suddenly in the midst of a perfectly smooth 
course of tuberculin treatment, and unaccompanied by a local reaction to 
the injections, when the dose has not been unduly large. 
ADMINISTRATION OF TUBERCULIN ADMINISTRATION OF TUBERCULIN 
1063 
Loss of weight is a delicate sign—more valuable as a symptom of over- 
dosage late in the treatment than as a protection against the suddenly 
appearing reactions. 
Bandelier and Roepke regard an increase in the pulse-rate as a sign of 
great importance. Hamman and Wolman and Lawrason Brown have not 
been able to observe this sign very frequently. 
The local signs are: pain, tenderness, and swelling at the site of injec- 
tion. These may consist of all gradations from simple thickening of the 
skin to a wide, deep, hard, and painful node, with or without involvement 
of the neighboring lymphatics. 
The local reaction has assumed great importance in recent years, espe- 
cially since Denys drew attention to it as serving as a warning of the approach 
of a general reaction. It is characterized by the development of pain, 
tenderness, and redness about the site of a former injection. It is more 
often solitary than any other sign, and, in the absence of a temperature 
record, it is safe to proceed, the sole guidance being the local reaction, both 
subjective and objective. A large dose of tuberculin may give a local 
reaction, due merely to its bulk and concentration (500 to 600 mg.), simu- 
lating a true reaction; this may be avoided by dividing the dose into two 
injections, which are given at the same time, but not into neighboring areas 
of skin. 
The focal signs in pulmonary tuberculosis are: increased cough, dyspnea, 
expectoration, thoracic pain, hemoptysis, and extension of the physical 
signs. 
Focal signs of a reaction are an indication that the treatment has been 
conducted too rapidly. 
The Favorable and Desirable Tuberculin Reaction.—Unless one belongs 
to the ultraconservative class of tuberculin therapeutists, it is not a slight 
focal reaction that is to be avoided, but those reactions that are large enough 
to manifest themselves by changes in the physical signs or by decided 
symptoms. On the contrary, it is the production of such slight hyperemia 
about the focus of disease that constitutes the most valuable result of tuber- 
culin therapy. It is well to start with a minute dose, and push the dosage 
rapidly until a slight focal reaction at the site of injection or mild pyrexia 
is observed. When a mild focal reaction is not produced, the patient is not 
receiving his due amount. 
As stated by Cohen1 success in tuberculin treatment can be expected 
only when the individual patient is given his appropriate dose, whether by 
chance, accident, in the course of systematic gradual increases, or by first 
determining this dose by intracutaneous tests. 
As stated above, I believe that it is desirable to produce mild focal reac- 
tions, but more importantly, a favorable constitutional reaction defined by 
Cohen2 as an improvement in appetite, increase of strength, a feeling of 
well being; lessening of symptoms, flattening of the temperature curve, and 
all favorable phenomena occurring on the day tuberculin is given or on the 
following day. 
The Choice of Tuberculin and the First Dose.—As previously stated, all 
tuberculins are essentially the same in their action, but the rate of absorption 
and consequently the rapidity of action varies among them. The extract 
tuberculins, like O. T., are more quickly absorbed than the bacillary suspen- 
sions or vaccines, as B. E. and T. R. My own preference is for O. T., 
Koch’s old or original tuberculin; for febrile patients where slower absorp- 
1 Arch. Pediat., November, 1920. 
2 Jour. Amer. Med. Assoc., 1914, 63, 1386; Med. Record, 1914, 86, 756. 1064 
BIOLOGIC THERAPY OF TUBERCULOSIS 
tion may be desirable, I use T. R. (tuberculin riickstand) or B. E. (bacillen 
emulsion). Whatever tuberculin is employed the principles governing Us 
administration are the same, and the technic described below is adapted for the 
administration of any tuberculin. 
These directions are for subcutaneous injections; other routes of ad- 
ministration may be employed, to be described later. 
The directions for administration are likewise adaptable for tuberculosis 
of different organs, but more especially for pulmonary disease. 
It is very important not to give an excessive dose at any time including 
the first dose. Since each patient is a law unto himself, the initial dose 
should be so small that no harm can result from its use. Children and 
adults who exhibit a slight fever or are not in good condition require smaller 
doses than those patients in good condition. 
The proper dose is that capable of eliciting a mild allergic reaction about 
the tuberculous foci. One may either start with an arbitrarily small dose 
and rather rapidly increase it until a favorable reaction is secured, or one 
may first conduct skin tests and determine by this means the sensitiveness 
of the patient. The latter is preferable, but laborious and not essential for 
success. 
In the arbitrary method I believe that an initial dose of any tuberculin 
may be 0.0001 c.c. for adults by subcutaneous injection. In cases of marked 
auto-intoxication with fever of over 100° F. and tachycardia, the dose may 
be 0.00001 c.c. Children under twelve years require less according to body 
weight. 
White and von Norman1 use a cutaneous or von Pirquet test with 0.0001 
c.c. of O. T. (never with T. R.) and reduce or increase this amount until 
they produce a minimal reaction, namely, one that shows redness and 
swelling measuring 4 to 6 millimeters in diameter within seventy-two hours. 
They then inject intracutaneously the exact amount that produced this 
reaction every two weeks for a period of three months. Another test is 
made at this time and treatment continued as heretofore. 
My own method is slightly modified after that employed by Cohen,2 
who gives a series of intracutaneous injections using the skin of the forearm. 
At the first time four injections are made at least 1| inches apart as follows: 
0.1 c.c. of 1 : 1,000,000 (0.0000001 c.c.) below the elbow. 
0.1 c.c. of 1 : 100,000 (0.000001 c.c) next below. 
0.1 c.c. of 1 : 10,000 (0.00001 c.c.) next below. 
0.1 c.c. of 1 : 1000 (0.0001 c.c.) above the wrist. 
The method of preparing these dilutions is described below. The unit 
of measure of tuberculin is commonly expressed in milligrams (mg.), but 
inasmuch as all tuberculins are fluid and not solid, the measure should be 
in fractions of the cubic centimeter or c.c. (milliliter or ml. is now being 
used by many instead of cubic centimeter or c.c.). 
The reactions are read forty-eight hours later and a papule and indura- 
tion regarded as positive. If no reactions occur, a second series of injections 
are given in the skin of the opposite arm using these doses: 
0.1 of 1 : 100 (0.001 c.c.) below the elbow. 
0.1 of 1 : 10 (0.01 c.c.) next below. 
In the majority of instances, however, the first set of injections shows 
the smallest amount of tuberculin giving a distinct reaction and this amount 
is the first dose to be given by subcutaneous injection. 
1 Jour. Med. Res., 1909, 20, 347; ibid., 1909, 21, 225; Arch. Int. Med., 1910, 6, 449; ibid., 
1912, 9, 114. 
2 Jour. Infect. Dis., 1917, 20, 223; New York Med. Jour., 1913, 98, 268. ADMINISTRATION OF TUBERCULIN 
1065 
Subsequent Doses and Intervals Between Injections.—The interval be- 
tween injections is usually from three to four days—i. e., two injections a 
week. One of these injections should be given on Saturday, in order that 
the patient, if ambulant, may have the advantage of rest over Sunday. 
This interval is merely tentative, and while it should not be shortened, it 
may be necessary to prolong it. While most reactions set in within from 
twenty-four to thirty-six hours, some may begin as late as from forty-eight 
to sixty hours (L. Brown). By waiting at least three days we may be 
assured that, if no reaction has occurred, none will take place. 
The usual interval is maintained as long as the patient is doing well. 
After a while the patient becomes intolerant and exhibits slight reactions, 
depression, and loss of weight. In such instances the interval may be 
increased to a week and the injections continued. Hamman and Wolman 
find this occurrence so frequent that they advise increasing the dose interval 
to one week, when the dose of 0.1 c.c. of O. T. is reached, 0.2 c.c. of T. R. 
and B. E., and 50 mg. of B. F. 
The aim should be to avoid the unfavorable reactions and gain the 
favorable. Both physician and patient should be thoroughly familiar with 
these. If any dose produces an unfavorable reaction the next dose should 
not be given until this subsides and should then be not more than one-half 
or even less than the former dose depending upon the degree of reaction. 
Practice among the most experienced phthisiotherapists varies greatly 
in regard to the size of subsequent injections. Probably the majority follow 
an arbitrary plan of beginning with a small dose and increasing upward in a 
gradual manner, dropping back when unfavorable reactions occur. 
Cohen aims to give the dose that produces a favorable reaction and 
continues that dose unchanged as long as the patient continues to do well; 
when progress becomes stationary he increases the dose accordingly. This 
appears to be a sensible and safe plan. 
Method of Preparing Dilutions and Doses.—In sanatoria this is a simple 
matter, but for the general practitioner treating a few cases may prove a 
troublesome feature. The technic, however, is simple, but the dilutions 
should he prepared at least once a month and preferably once every two weeks, 
in order to guard against employing an inert product. The tuberculin 
dilutions should be kept in a cold place. The stock tuberculins are prepared 
by various manufacturers of biologic products and are less subject to 
deterioration than the dilutions. For the method described below the 
stock tuberculin may be purchased in ampules of 1 or 5 c.c. and diluted as 
required. 
If purchased already diluted and ready for use due care must be exercised 
to use a sufficiently fresh product. 
The initial dose given arbitrarily may be 0.0001 c.c. as stated above; 
this is also the largest amount given intracutaneously, if the skin tests are 
conducted for determining the dose. This is equivalent to 0.1 c.c. of a 
1 :1000 dilution. 
1. One should have on hand dark glass bottles, holding at least 10 c.c. 
and furnished with rubber-dam caps. These are sterilized in boiling water 
just before use. Also 100 c.c. of sterile physiologic salt solution to which 
0.3 c.c. of pure fluid phenol or tricresol is added and well shaken (= 0.3 per 
cent.). 
2. With a sterile pipet place 9.9 c.c. of the phenolized saline solution in a 
bottle and add 0.1 c.c. of the pure tuberculin; mix well. This gives a dilu- 
tion of 1 : 100, of which 0.1 c.c. carries 0.001 c.c. tuberculin. Label, 1 : 100. 
3. In a second bottle place 9 c.c. of phenolized saline and 1 c.c. of the 1066 
BIOLOGIC THERAPY OF TUBERCULOSIS 
1 : 100 dilution. This gives a 1 : 1000 dilution, of which 0.1 c.c. carries 
0.0001 c.c., the usual initial dose. Label, 1 : 1000. 
4. For the skin tests or in case a smaller initial dose is to be given, higher 
dilutions may be prepared. In a third bottle place 9 c.c. of the phenolized 
saline and add 1 c.c. of the 1 : 1000 dilution. This gives a 1 : 10,000 dilution, 
of which 0.1 c.c. carries 0.00001 c.c. The next higher dilution prepared in 
the same manner is 1 : 100,000, of which 0.1 c.c. carries 0.000001 c.c. 
5. As previously stated, treatment is usually begun with 0.1 c.c. of 
1 : 1000 dilution representing 0.0001 c.c. of tuberculin. The next dose may 
be 0.2 c.c. and each succeeding dose increased (unless contraindicated) by 
0.1 c.c. until a dose of 1 c.c. of this dilution has been given. 
The next dose would be 0.2 c.c. of the 1 : 100 dilution and increased by 
0.1 c.c. each time until a dose of 1 c.c. is reached. 
This would be followed by a dose of 0.2 c.c. of 1 : 10 dilution prepared 
by diluting 1 c.c. of the tuberculin with 9 c.c. of the phenolized saline solu- 
tion. The last dose would be 1 c.c., which carries 0.1 c.c. of tuberculin. 
If still larger doses were required the injections may be continued by 
using 0.2 c.c. of undiluted tuberculin followed by increases of 0.1 c.c. until 
a dose of 1 c.c. undiluted tuberculin was being given. 
As a general rule, the succeeding doses of a tuberculin may be increased 
by 0.1 c.c. as previously stated, due care being exercised when the dilutions 
are changed. If the patient is quite sensitive, it may be well to repeat the 
first dose of each stronger solution once or more often as it is reached, and, 
instead of proceeding to 0.2 c.c., give only 0.15 c.c., thus tiding over the gap. 
If the patient is not so sensitive, a few leaps may be taken with the weaker 
dilution, so as to test the patient’s tolerance, and, if it is good, the next 
higher dilution is given at once. 
Beraneck’s tuberculin is marketed, ready diluted, in a series of syringes, 
A/128, A/64, A/32, to A/4, A/2, A, B, C, to H. H is the pure tuberculin. 
Each solution is one-half the strength of the next stronger one. The increase 
of dosage is usually by 0.1 c.c. until 0.5 c.c. is given. Then 0.1 c.c. of the 
next stronger dilution is given, and so on. 
Dixon's tuberculin, employed by the Pennsylvania Health Department, 
is supplied in syringes in the series of dilutions given in the following table, 
so that the doses may be increased as is found desirable: 
Dilution 
Amount (in Grams) or 
Extract or Tubercle 
Dilution 
Amount (in Grams) or 
Extract or Tubercle 
No. 
Bacilli Contained in 
No. 
Bacilli Contained in 
1 
0.000001 
9 
0.00008 
2 
0.00001 
10 
0.00009 
3 
0.00002 
11 
0.00010 
4 
0.00003 
12 
0.00011 
5 
0.00004 
13 
0.00012 
6 
0.00005 
14 
0.00013 
7 
0.00006 
15 
0.00014 
8 
0.00007 
16 
0.00015 
ADMINISTRATION OF DIXON’S TUBERCULIN 
It is suggested that in ordinary cases the injection of each dilution be 
repeated at least five times before changing to the next number or next 
stronger dilution. Larger doses may be given by using two injections of 
No. 16. 
Site of Injection.—Injections are probably best given in the back, at 
the lower angle of the scapula. Local reactions are more reliable here than 
when the injections are given in the arm, although injections at about the 
insertion of the deltoid muscle are quite satisfactory. All injections should ADMINISTRATION OF TUBERCULIN 
1067 
be given under aseptic precautions and with a sterilized syringe in exactly 
the same manner as any bacterial vaccine is given. All injections should 
be subcutaneous; intramuscular injections are to be avoided, and no great 
advantage is to be gained from using intravenous injections. 
Time of Injection.—There is some difference of opinion as to the best 
time of the day for administering tuberculin therapeutically. If given 
in the morning, a slight febrile reaction may occur during the evening which 
would otherwise be overlooked. Brown is in favor of the afternoon as the 
most suitable time, because it affords an opportunity for omitting the dose 
in case there is an accidental rise of temperature on that day. On the other 
hand, it is contended that the rest at night would tend to prevent the occur- 
rence of the reactions that might appear if the patient were up and about. 
While it is not essential that the patient rest for a few hours after a dose 
has been administered, this is advisable, and where absolute rest can be 
enforced, the dosage may be increased with greater rapidity than in ambu- 
lant patients. 
The Duration of Treatment.—The length of the course of tuberculin 
treatment varies considerably according to individual circumstances. The 
first three months are the most important and require the closest co-operation 
between physician and patient because of the difficulties that may occur. 
The administrations should continue until the patient can stand without 
reaction a dose of at least one hundred times larger than the first reacting 
dose. For example, if the first dose were 0.001 c.c. the injections should 
continue until a dose of at least 0.1 c.c. has been reached. Riviere and 
Morland state that injections of any tuberculin may stop at 0.1 c.c. except 
in advanced cases where symptoms have not improved up to this dose and 
when the patient cannot be kept under observation. It is never necessary 
to exceed a dose of 1 c.c. undiluted tuberculin, although this is the custom of 
some, as Gotsch and Schlossmann. 
Some physicians, as Jochmann, employ the skin tests as a guide and 
continue treatment until the v. Pirquet cutaneous test yields a negative 
reaction. This practice is based upon the assumption that desensitization 
of the skin occurs as tolerance is increased, but it would appear that tolerance 
may be built up to a high degree by progressive encapsulation of the focus 
of tuberculous disease without a commensurate degree of desensitization 
of the skin. 
It is a safe rule to continue the injections as long as the symptoms continue 
to improve, and the time varies from three months to one or two years. 
Even after sputum ceases and tubercle bacilli are no longer found, it is well 
to continue for a few months after this gain. 
Repetition of Tuberculin Treatment.—After treatment ceases sensitive- 
ness to tuberculin continues to decrease as tolerance increases in the favorable 
cases. Sensitiveness to tuberculin, however, may remain and even increase 
without recrudescence of the disease, due to the absorption of sufficient 
tubercle products to maintain and increase allergic sensitiveness of the skin 
and other tissues. Furthermore, the disease may suddenly spring into ac- 
tivity and prove destructive without an apparent return or increase of 
sensitiveness; this condition may be due to accidental rupture of a gland 
or other focus and hypersensitiveness is lost during acute infections. 
However, a return of sensitiveness is usually a sign of activity of the 
disease and Petruschky states that this occurs in about three months and 
that the series of injections should be recommenced. The return of sputum 
or the reappearance of tubercle bacilli, cough, or any symptoms which had 
disappeared, should suggest a repetition of the series of injections. As long 1068 
BIOLOGIC THERAPY OF TUBERCULOSIS 
as the patient is gaining in weight and conditions are generally favorable 
and satisfactory there is no indication for repetition, but upon the first 
evidences of backsliding the course should be recommenced. 
Other Routes for the Administration of Tuberculin.—Oral Route.— 
It has been shown that reactions may follow the oral administration of 
tuberculin. But absorption is so irregular that a quantity of tuberculin 
may be absorbed suddenly and cause unexpected reactions. Much depends, 
apparently, upon the state of digestion and upon the condition of the ali- 
mentary tract. The oral route also deprives the physician of the benefits 
to be obtained from using the local reaction as a guide. Otherwise the 
method is simple and the tuberculin may be administered in the form of 
tablets or in capsules. It is important, however, to exercise supervision 
over the patient. S. Solis Cohen1 has used a modification of Latham’s 
method, and reports favorable results: “Tuberculin residue (T. R.) tritu- 
rated with milk-sugar is given with skimmed milk, whey, or beef-juice. The 
initial dose is 0.000001 mg. Both subjective and objective symptoms of 
reaction are watched for. The dose is repeated once or twice weekly, 
according to results. It is gradually increased by increments of 0.000001 mg. 
to the reaction point, and then dropped one point lower, and so continued for 
some weeks. Later, a further increase is attempted, and if reaction is not 
shown, is proceeded with in a similar gradual way. The arbitrary increment 
of 0.000001 mg. is maintained during this remittent progression until 0.0001 
mg. has been reached. After that the increment may be raised to 0.00001 
mg. Thus, by successive stages, a maximum dose is attained at a point 
determined for each individual by all the factors in the case, including the 
rapidity of increase, character and intensity of reaction, and maintenance 
of tolerance, as well as the focal and general signs of improvement. The 
treatment is continued with intermissions for many months, and may be 
resumed, if necessary, from time to time over a period of years.” 
Tuberculin has also been administered intrabronchiatty and by the 
rectum without good results. 
Koch was the first to administer tuberculin intravenously, but this route 
has not come into general favor, owing to the fact that even greater control 
over dosage is necessary; there is, besides, no local reaction to serve as a 
guide, and technically the administration is more difficult than is sub- 
cutaneous injection. 
The Inlrafocal Route.—The use of tuberculin intrafocally, that is, a 
method by which the tuberculin is brought into immediate contact with 
the diseased area, has been advocated in the treatment of tuberculous pleu- 
risy, tuberculous peritonitis, and tuberculosis of other serous membranes, 
such as those lining joints, the tunica vaginalis of the testicle, etc. Senger, 
Crocker, and Pernet advise the intrafocal use of tuberculin in the treatment 
of lupus; others have applied it directly to broken-down glands and to 
sinuses. William Egbert Robertson2 has reported very good results in 2 
cases of tuberculous pleurisy as a result of withdrawing a small amount of 
fluid and injecting 5 mg. of old tuberculin through the needle, which is left 
in situ for that purpose. After some hours there was a slight reaction; the 
remainder of the fluid was rapidly absorbed, and convalescence was promptly 
established, though, of course, a damaged lung remained. In a case of 
hydrocele the injection of old tuberculin was followed by a sharp local and 
a moderate general reaction, followed by absorption, and without the slightest 
evidence of recurrence to date, now about a year since the injection was 
made. 
1 Amer. Jour. Med. Sci., 1915, 149, 81. 
2 Personal communication. TUBERCULIN TREATMENT OF PULMONARY TUBERCULOSIS 
1069 
TUBERCULIN TREATMENT OF PULMONARY TUBERCULOSIS 
Ambulant Treatment.—The tuberculin treatment of ambulant cases of 
pulmonary tuberculosis is becoming more prevalent. These cases may, 
or may not, be able to work, but come from their homes to the physician’s 
office or to some institution at regular intervals for observation and treat- 
ment. Experience has shown that the practice is without danger and the 
administrations may be carefully controlled by instructing the patient to 
take the temperature and observe symptoms. Elaborate and very satis- 
factory charts have been devised for recording the progress of individual 
cases and a great deal can be done toward securing the confidence of the 
patient and his close co-operation, which are so essential for the successful 
treatment of this disease. 
A distinct drawback to ambulant treatment is the danger of mixed infec- 
tion due to the inhalation of dust and dirt. Whenever the disease is both 
open and recent, every effort should be made to place the patient in a sana- 
torium where the air is pure and better hygienic treatment procurable, until 
the lesion has again become closed. 
A second drawback to ambulant treatment is that rest cannot be pre- 
scribed after the injections as well as in institutions. But much can be done 
in this direction by giving the injections late in the afternoons and particu- 
larly on Saturdays. Of course, in febrile cases where more or less continuous 
rest is essential, ambulant treatment should not be given unless for sym- 
tomatic relief. 
Ambulant treatment, therefore, is safe for closed cases when prolonged 
rest in bed is not required Also for cases returned home from sanatoria 
when the course of tuberculin treatment has not been completed. It is the 
next best thing for those who are unable to secure sanatorium treatment for 
social or family reasons (makeshift ambulant treatment). Distinct advan- 
tages are that the patient may remain at home and probably at work, free 
of financial anxiety, strange surroundings, and home-sickness. 
Treatment of Children.—Tuberculosis in infants is highly fatal and the 
morbidity and mortality keep closely parallel during the first year of life. 
The glands are usually first involved and, if superficial with softening, recov- 
ery may take place; if the mesenteric and mediastinal group are involved 
the disease usually extends to neighboring organs and tuberculin offers little 
or nothing in the way of treatment. The resistance of very young children 
is very low with but slight tendency for encapsulation of the lesions. Tuber- 
culin may be administered, but the chances for success are remote. 
In children of school age the glands are usually involved and the tuber- 
culin treatment of these will be shortly considered. In hilum phthisis and 
apical infections resistance is more marked and something may be gained 
by the administration of tuberculin. 
The method of administration is exactly as previously described, except 
that the initial dose must be smaller. An excellent plan is to determine the 
proper initial dose by the intracutaneous tests. If these cannot be made, an 
arbitrary initial dose for children three to twelve years of age may be 0.2 c.c. 
of 1 : 10,000 dilution of O. T., T. R., or B. E.; this amount corresponds to 
0.00002 c.c. which is one-fifth the usual initial dose for adults. 
Treatment During Pregnancy.—It is common clinical experience to 
observe an increased sense of well being and a loss of some of the symptoms 
of pulmonary tuberculosis during pregnancy; on the other hand, a minority 
of cases become steadily worse. Petruschky was the first to advise tuber- 
culin in the treatment of pulmonary tuberculosis in pregnancy; there appears 1070 
BIOLOGIC THERAPY OF TUBERCULOSIS 
to be no contraindication to its careful administration, and Riviere and 
Morland state “there appears very good ground for a thorough trial of 
tuberculin during pregnancy in women with active tubercle, or in whom a 
lesion has recently been active and fears are entertained for the effect of 
labor.” 
Treatment of Mixed Infections.—True mixed infections are those in 
which other bacteria in addition to the tubercle bacilli have gained access 
to the tissues and are producing pathologic changes. Other bacteria are 
always to be found in the sputum of all cases, but this does not necessarily 
mean mixed infection. 
Various organisms may produce mixed infections in pulmonary tubercu- 
losis, the most important being in the order named; streptococci, Bacillus 
mucosus capsulatus (B. friedlander), pneumococci when abundant, staphylo- 
cocci of the aureus variety and Micrococcus catarrhalis. 
The diagnosis of mixed infection is made when it is apparent from physical 
signs that a cavity exists in communication with a bronchus; also when 
there is a hectic type of temperature, a history of secondary infection and 
particularly influenza or grip, and when the washed sputum shows the 
pyogenic bacteria in intimate admixture with tubercle bacilli. 
Prophylaxis of mixed infection is of the utmost importance. The patient 
with an open lesion should be scrupulously guarded against colds and the 
like; also against dust and dirt in so far as this is possible, and best by 
sojourning in the country, and especially in sanatoria, where the air is purer 
than in cities and towns. Every effort of this kind should be made to guard 
against mixed infection. 
In the treatment of mixed infections autogenous vaccines may prove 
very helpful in addition to the usual hygienic measures. Cultures of the 
sputum should be made with particular attention to the isolation of strepto- 
cocci and pneumococci. If streptococci greatly predominate in direct 
smears of the sputum, they alone should be employed. The vaccine may 
contain 1,000,000,000 per cubic centimeter. 
Extra care must be exercised in the dosage. If tuberculin is given, the 
doses should be small enough to produce none or but the very slightest focal 
reaction. The first dose of the vaccine may be 0.1 c.c. by subcutaneous 
injection. It is my custom to give one dose of vaccine and one of tuberculin 
per week (vaccine on Tuesday and tuberculin on Saturday). The doses of 
each are cautiously increased from week to week, as possible, until a dose of 
1 c.c. of vaccine is reached. I never give more than ten doses of any one 
vaccine, finding that better results are to be obtained by preparing fresh 
vaccines from time to time, not only because they possess better vaccinogenic 
activities, but likewise because of the chances for changes in the bacterial 
flora of the lesions. 
Results of Tuberculin Therapy.—The early and disastrous results 
obtained with tuberculin in various types of tuberculosis cannot be used at 
the present day as a measure for determining the therapeutic value of 
tuberculin. 
So much depends upon the individual case, the duration and activity 
of the disease, the possibility of supplementing the active immunization 
with sanatorium treatment, the skill and patience of the physician, etc., 
that, from a prognostic standpoint, every case must be judged upon its own 
merits. The results of the modern use of tuberculin show quite clearly 
that it is not a “cure” for tuberculosis, but, rather, a rational and useful 
therapeutic aid, the best results being secured when the treatment is carried 
out in special institutions or by specially trained physicians who have a TUBERCULIN TREATMENT OF PULMONARY TUBERCULOSIS 
practical knowledge of the difficulties, dangers, and possibilities of tuberculin 
therapy. 
The value of this or of any therapy can be judged according to various 
standards: (1) Working ability; (2) duration of life; (3) the presence of 
tubercle bacilli in the sputum; (4) physical signs, and (5) the symptoms. 
The first three are especially valuable; duration of life is, after all, the 
most important criterion, as anything that prolongs life is, of course, wel- 
comed. 
Kremser chose 110 patients expectorating tubercle bacilli and treated 
55 unselected cases with tuberculin; of these, 22, or 40 per cent., lost the 
bacilli from their sputum; of those treated without tuberculin only 16, or 
29 per cent., lost their bacilli. Phillippi found that 58 per cent, of his second 
stage cases were rid of bacilli in the sputum under tuberculin treatment, as 
against 19 per cent, without. Brown reports from Saranac Lake that, in 
the incipient class, 67 per cent, of the tuberculin patients were rid of bacilli; 
of the others, 64 per cent. In the moderately advanced the figures are 
respectively 44 and 24 per cent. Bandelier gives the reports of 500 cases, 
of whom 202 had tubercle bacilli in the sputum. In the following table he 
compares the working capacity and sputum examinations of these patients 
under tuberculin treatment: 
1071 
Total. 
Stage I, 
Per cent. 
Stage II, 
Per cent. 
Stage III, 
Per cent. 
Complete earning capacity on discharge. 
500 cases (69.8 per 
cent.) 
90.4 
80.7 
32.8 
Sputum changed from positive to nega- 
202 cases (63.9 per 
tive 
cent.) 
100.0 
87.3 
44.0 
The parallelism between the bacillary content of the sputum and the 
working capacity is close and shows the value, from a statistical point of 
view, of sputum examinations. 
Healing with and without tuberculin is qualitatively the same, although, 
in the opinion of Ziegler, Petruschky and Rohmer, Pearson and Gillilan, 
Jurgens, Neumann, and others, quantitatively the results are different, all 
authors agreeing on the presence of more fibrosis about the lesions than is 
usual in the untreated cases. Autopsy findings necessarily point with less 
favor to tuberculin therapy than do clinical facts bearing on the rapidity 
and permanency with which a lesion heals. 
The influence of tuberculin cannot at present be judged from the presence 
of antibodies in the serum, as data bearing upon the presence or absence of 
antitoxins, opsonins, agglutinins, and bacteriolysins are insufficient. 
General symptoms and signs, such as fever, cough, loss of weight, accel- 
erated pulse, digestive disturbances, dyspnea, and pain, if they are due to 
tuberculosis, as a rule improve under tuberculin treatment. Under proper 
conditions the incidence of hemoptysis is not usually increased, and the 
quantity of sputum and number of expectorated bacilli gradually decrease. 
Riviere and Morland have stated the following in reference to the value 
of tuberculin in the treatment of pulmonary tuberculosis: 
“Certain results may be said to be well established by clinical experience. 
The first and most striking of these is that phthisis treated with tuberculin 
before it has become open—i. e., before it has been exposed to the risk of 
secondary infection—remains closed. The importance of this fact, on which BIOLOGIC THERAPY OF TUBERCULOSIS 
1072 
there is practically unanimous opinion, can hardly be exaggerated. It is 
true that the same result has been claimed for hygienic treatment. Bande- 
lier regards the fact as being so well established that he refrains from giving 
tuberculin to these patients because it is unnecessary. It is also true that 
the vis medicairix malurce unfettered by art would have had the same result 
in a large proportion of cases—the Paris Morgue (quoted by Huggard) gives 
68 per cent, of cures without the bias of any pet remedy; but there remains 
a proportion, it may be small, of closed pulmonary tuberculosis which will 
not get well, and with these tuberculin has been shown to be competent to 
deal. Early diagnosis—that is to say, really early diagnosis, before tubercle 
bacilli appear in the sputum—combined with specific treatment insures 
completely against a breakdown. 
“We believe that statistics have already shown the ability of tuberculin 
to increase the percentage of those who lose their sputum, or the tubercle 
bacilli contained in it, during hygienic treatment, and to extend the expecta- 
tion of working efficiency after hygienic treatment; but we are content to 
leave this to a more rigid demonstration. Of all these matters the tubercular 
patient is the final judge, and misled as he was by the disasters of 1890-91 
there is no doubt that his experience of tuberculin under the new conditions 
is making him willing, and sometimes even anxious, to submit himself to 
treatment with the remedy.” 
TREATMENT OF TUBERCULOUS ADENITIS 
These infections usually occur in childhood and may involve the super- 
ficial glands of the neck (scrofula) or the deeper glands, as those in the medias- 
tinum and mesentery. 
In enlarged but not caseous glands tuberculin treatment may be of some 
assistance and aid in bringing about resolution. When softening has occurred 
tuberculin is of no aid in the absorption of the caseous material, surgical 
removal of the gland or group of glands followed by tuberculin treatment 
being the method of choice. When sinuses exist tuberculin may prove of 
distinct aid in treatment if surgical measures are refused or impossible. 
These cases are usually mixed infections, the pus showing the presence of 
staphylococci, diphtheroid bacilli, and the like. 
The doses should be such as to maintain a state of progressive healing 
or bring about a mild focal reaction of slightly increased discharge with not 
more than a rise of 1° F. in temperature. As a general rule injections should 
be given once in seven to ten days. If a vaccine of the secondary pyogenic 
bacteria is employed, it should be autogenous and given every ten to twelve 
days alternating with the injections of tuberculin. Under this plan the 
patient is inoculated every five or six days. 
The initial dose of tuberculin for the treatment of localized tuberculosis 
of children five to fifteen years of age may be 0.0002 c.c., which is larger than 
usually given in pulmonary tuberculosis. This amount corresponds to 0.2 
c.c. of a 1 : 1000 dilution. The doses are gradually increased as described 
in the treatment of pulmonary tuberculosis. My own practice is not to 
increase the dose as long as there are evidences of healing; when the condition 
becomes especially sluggish and at a standstill the doses of both tuberculin 
and vaccine are cautiously increased. 
The duration of treatment is necessarily prolonged, being a matter of 
several months and even a year or more. Even after complete healing has 
become apparent, it is well to continue the injections of tuberculin for several 
months beyond this point. TREATMENT OF TUBERCULOSIS OF THE SKIN 
1073 
It is to be carefully remembered, however, that tuberculin and vaccine 
therapy have little to offer as long as a sinus is maintained by necrotic bone; 
ordinary medical and surgical treatment should not be neglected as is some- 
times the tendency when biologic therapy is being employed. 
Tuberculous dactylitis of children, tuberculosis of the spinal vertebra 
(Pott’s disease), and tuberculosis of the hip and other large joints are the most 
familiar examples of tuberculous infections of these tissues. When extensive 
destruction has occurred, healing is a very prolonged and tedious process. 
Tuberculin may prove of some aid and when properly given can do no harm. 
In early cases of dactylitis the results are frequently very good, but in the 
usual case both patient and physician must expect prolonged treatment 
and slow progress. 
In addition to placing the patient under the best hygienic conditions 
obtainable, fixing the part in order to reduce motion and auto-intoxication 
to a minimum, aspiring pus and cleaning up necrotic tissues as far as possible, 
tuberculin may be administered at intervals of five to seven days in gradually 
increasing doses to secure very mild focal reactions. The initial dose varies 
greatly with different patients, but for children may be 0.2 c.c. of a 1 : 1000 
dilution (0.0002 c.c.) and for adults 0.5 c.c. of a 1 : 1000 dilution (0.0005 c.c.). 
These amounts are not likely to produce focal reactions (slight increase of 
discharge and slight increased pain and tenderness), but only the mildest of 
reactions are desirable, and if healing is occurring under one dose it should be 
continued unchanged as long as progress is satisfactory. Sometimes hyper- 
sensitiveness develops, in which case the dose must be decreased. 
In mixed infections an autogenous vaccine may be of aid. It may be 
prepared to contain 1,000,000,000 per cubic centimeter, the initial dose being 
0.1 c.c. followed by gradually increasing amounts. The vaccine and tuber- 
culin may be given alternately every ten to twelve days, which means an 
injection every five or six days. 
TREATMENT OF TUBERCULOSIS OF BONES AND JOINTS 
Tuberculosis of the skin usually occurs as lupus vulgaris, but also as warty 
or tuberculosis verrucosa cutis, scrofuloderma, or tuberculosis of the skin around 
discharging tuberculous lymph-glands and sinuses of other tissues, and 
tuberculosis cutis orificialis or tuberculous ulcers of the skin around orifices 
(anal, vulvar, nasal, and oral). 
Tuberculin has been rather extensively used in former years in treatment 
but with negative or indifferent results. At the present time, however, 
some dermatologists are securing good results and especially with T. R. and 
B. E. 
The doses should be sufficiently large to produce mild focal reactions of 
increased hyperemia. Rather large amounts may be required for this result, 
and for this reason the patient should be carefully examined for pulmonary 
lesions which may preclude the injection of amounts sufficient for producing 
these focal reactions. There should be no disturbance of the general health, 
which should receive attention in the way of fresh air, good food, and other 
hygienic conditions. The initial dose of T. R. or B. E. for adults may be 
0.5 c.c. of a 1 : 1000 dilution, subsequent doses being raised to secure very 
mild focal reactions as previously described under the administration of 
tuberculin. 
The injections may be given every seven to ten days, and the treatment 
TREATMENT OF TUBERCULOSIS OF THE SKIN 1074 
BIOLOGIC THERAPY OF TUBERCULOSIS 
should be accompanied by the usual local measure with special reference to 
the use of the Finsen or z-rays in proper dosage. 
Exceptionally good results have been observed in the treatment of 
tuberculosis of the eye and ear. Both organs, and particularly the former, 
are adaptable for the observation of focal reactions. 
Some cases of chronic uveitis and iridocyclitis are apparently due to 
tubercle toxins rather than to the presence of tubercle bacilli, but in the 
majority of cases the eye lesions are due to the presence of the bacilli. In 
many cases of ocular and oral tuberculosis pulmonary lesions are likewise 
present, although these may be latent. The possibility of their presence, 
however, should always be kept in mind and especially in relation to dosage 
of tuberculin. 
Preference should be given the bacillary tuberculins, as T. R. or B. E., in 
view of the excellent results reported from their use. The initial dose of 
either may be 0.0001 c.c., which corresponds to 0.1 c.c. of a 1 : 1000 dilution. 
The method of preparing the dilutions has been previously described. The 
injections may be given every four or five days in gradually increasing 
amounts, as described for the administration of tuberculin, until a dose of 
0.010 or 0.1 c.c. of the tuberculin is reached. Very mild focal reactions are 
desirable, v. Hippel1 starts with 0.0002 c.c. (0.2 c.c. of 1 : 1000) and gives 
an injection every other day in gradually increasing amounts until a dose 
of 0.100 (1 c.c. of 1 : 10) is reached. 
TREATMENT OF TUBERCULOSIS OF THE EYE AND EAR 
TREATMENT OF TUBERCULOSIS OF THE KIDNEYS AND BLADDER 
In tuberculosis of the kidney surgeons are in general accord regarding 
the advisability of removal of the organ as soon as possible, providing the 
patient has two kidneys of wdiich the other is apparently not involved or 
involved so lightly that on the basis of functional tests there is good evi- 
dence for believing that it is capable of caring for elimination. 
Tuberculin has been employed in the expectant plan of treatment, but 
with negative or indifferent results. When operation is refused or is im- 
possible, as in bilateral infections, it may be tried; likewise after operation, 
although it is to be clearly borne in mind that tuberculin has no prophylactic 
properties as previously discussed. 
In bladder tuberculosis the results of tuberculin treatment have been 
somewhat better; if both ureters are involved, however, great care must be 
exercised in dosage in order to avoid occlusion due to focal reactions. 
Very mild focal reactions are desirable, but severe reactions of pain, 
increased frequency of micturition, and hematuria are to be carefully avoided. 
Dosage is also to be controlled by the state of the pulmonary lesions, which 
are present in many cases. 
The initial dose may be 0.2 c.c. of 1 : 1000 dilution (0.0002 c.c.); subse- 
quent doses are gradually increased at intervals of four to five days according 
to the plan previously outlined for the administration of tuberculin in general. 
In tuberculous peritonitis the results of tuberculin treatment are sometimes 
strikingly good. In the ascitic form drainage alone may be sufficient owing 
to the healing qualities of the resulting hyperemia. In the plastic form, 
TREATMENT OF TUBERCULOUS PERITONITIS 
1 Arch. f. Ophthal., 1904, lix, 1. SERUM TREATMENT OF TUBERCULOSIS 
1075 
however, this procedure is not available, and fortunately tuberculin is 
sometimes of aid if there is no tuberculous ulcerative enteritis. 
In practically all cases tuberculous peritonitis of both children and adults 
is secondary; bovine bacilli have been found in from 50 to 70 per cent, of 
cases. 
The doses must be small and there is no way for observing focal reactions. 
Since the patients are bed-ridden, an accurate record of the temperature is 
possible. A dose of tuberculin should not produce an increase of tempera- 
ture of more than 10 or 2° F. 
For a child the initial dose may be 0.1 c.c. of 1 : 1000 dilution (0.0001 
c.c.); an adult may receive 0.5 c.c. of 1 : 1000 (0.0005 c.c.) unless there is an 
acute primary lesion indicating a smaller dose. Skin tests are especially 
useful, the initial dose being the smallest amount producing a mild intra- 
cutaneous reaction. 
The injections may be given every five or six days in gradually increasing 
amounts. When a dose has been reached under which the patient is gaining, 
it should be continued until progress becomes stationary, at which time it 
may again be increased. If hypersensitiveness supervenes during treatment 
the dose should be reduced. 
Inasmuch as so many cases of tuberculous peritonitis are infections due 
to bovine bacilli, it may be a good plan to employ a bovine tuberculin and 
especially if human tuberculin has not proved beneficial. Raw,1 however, 
believes that the tuberculins should be opposite to the infection, that is, 
children should receive tuberculin of human bacilli and adults, tuberculin 
of bovine bacilli. 
SERUM TREATMENT OF TUBERCULOSIS 
Various immune sera have been employed from time to time for the 
treatment of tuberculosis. As stated in the discussion on Immunity in 
Tuberculosis, it is exceedingly difficult to immunize the lower animals to the 
extent of producing sera containing sufficient amounts of antibodies to be 
even hopeful from the therapeutic standpoint. Even the immunized animal 
may develop the disease if too large a dose of living bacilli is given for 
immunizing purposes. Probably the best known sera are those of Marag- 
liano and Marmorek: 
Maragliano’s serum is prepared by immunizing horses for from four to 
six months with a mixture of a toxin prepared by the filtration of cultures 
only a few days old and concentrated in vacuo at a temperature of 30° C., 
mixed with that obtained by aqueous extraction of killed virulent cultures, 
and concentrated by heating on a water-bath at 100° C. for three or four 
days. Maragliano assumes that the antiserum possesses antitoxic, bac- 
tericidal, and agglutinating properties. One cubic centimeter of this serum 
is injected every other day for one and a half months.. The favorable action 
of the serum is reported on, especially by Mircoli and other Italian physicians, 
but in Germany and France proof of its value could not be established. 
Marmorek’s serum is now prepared by immunization of horses with 
young tubercle bacilli, whose acid-fast character is still very slight or entirely 
absent. When the horses have attained a high degree of immunity they 
receive injections of various strains of pure cultures of streptococci obtained 
from the sputum of tuberculous patients. The serum of these animals is, 
therefore, antituberculous and also antistreptococcic, and is serviceable 
against a mixed infection. 
1 Brit. Med. Jour., April 17, 1920, 538. 1076 
BIOLOGIC THERAPY OF TUBERCULOSIS 
The serum is administered daily either by subcutaneous injection, in 
doses of from 5 to 10 c.c., or by the rectum, in doses of from 10 to 20 c.c. 
The latter form of administration is quite objectionable to most patients, 
but is the one least likely to produce serum sickness. 
While this serum has been used quite extensively, the evidence at present 
is too conflicting to permit definite conclusions to be drawn as to its value 
in treatment. It would, however, seem to be worthy of further trial in 
cases of localized bone and joint tuberculosis and in the incipient stage of 
pulmonary tuberculosis. Citron recommends its use in patients who 
evince persistent rise of temperature, and in the very severe but not hopeless 
cases where tuberculin therapy cannot be undertaken. In some of these 
cases he has obtained very encouraging results. Citron occasionally begins 
with the serum treatment, and later combines tuberculin administration 
with it, finally omitting the serum altogether. 
Porter,1 Eversole and Lowman2 have reported somewhat favorably on 
the use of Spengler’s “I. K.” serum, and Cavasse3 more recently upon 
Jousset’s serum. 
1 Amer. Jour. Orthoped. Surg., 1911, 9, 346. 
2 Amer. Jour. Orthoped. Surg., 1912, 10, 234. 
3 Medicine, Paris, 1920, 1, 490. CHAPTER XLII 
BLOOD TRANSFUSION 
INDICATIONS AND METHODS 
Historic.-—While the ancients placed considerable importance upon the 
blood in philosophic and theologic discussions and in relation to disease, 
and, of course, could not have failed observing the direct consequences of 
hemorrhage, it is curious that no authentic records of blood transfusion are 
available until about 1660. There are earlier indefinite references to the 
subject, and it is very probable that earlier attempts were made, but without 
success, owing to a lack of knowledge regarding the circulation of the blood 
and a lack of suitable instruments and surgical skill for overcoming the 
difficulties arising from coagulation. 
Most historians give credit to Christopher Wren, the later famous archi- 
tect, for suggesting transfusion experiments following some intravenous 
injections which he had done in the Oxford laboratory. According to 
Viets,1 an account of Wren’s experiments with the assistance of Robert 
Boyle, the physicist, is found in No. 7 of the Philosophical Transactions of 
the Royal Society for December 4, 1665. It seems certain that as early as 
1660 Boyle had given a dog an intravenous injection of opium by means 
of a quill, and that later Wren, Lower, and Boyle carried out blood transfu- 
sion experiments. 
In February, 1665, Lower, who was an “expert anatomist” assistign 
Willis, transfused blood from one sheep to another, first from vein to vein 
and later from artery to vein, employing quills for the connections. His 
first demonstrations were made at Oxford, and on May 17, 1665, before the 
Royal Society in London, an account being recorded in the Transactions 
under date of December 17, 1666 (No. 20). During that same year Denis 
and Bayant, in France, performed similar experiments recorded in a letter 
from Denis published in No. 25, May 6, 1667, of the Transactions. 
Samuel Pepys on November 14, 1666 recorded in his diary that “at a 
meeting of Greshan College, the experiment of transfusing the blood of one 
dog into another was made before the Society by Mr. King and Mr. Thomas 
Coxe upon a little mastiff and a spaniel, with very good success.” On 
November 21st he records that “the spaniel was produced and found very 
well.” 
After Lower’s public demonstration on May 17, 1665, the Society dis- 
cussed the subject of priority, noting that “a Benedictine Fryer discoursed 
of it at M. de Monmours, ten years ago,” but without further evidence to 
strengthen this French claim, and wisely concluding, “But whoever the 
Parent be, that is not so material as all that lay claim to this child should 
joyn together their efforts and cares to breed it up for the service and relief 
of human life, if it be capable of it; and this is the main thing aimed at and 
solicited in this Discourse.”2 
Lower, having blazed the trail by his animal experiments, followed this 
sound advice, and with the assistance of King performed the first transfusion 
1 See letter by Dr. Henry Viets in Jour. Amer. Med. Assoc., 1918, 68, 480. 
2 Gotch, Two Oxford Physiologists, Lower and Mayon, Oxford. 1908. 
1077 1078 
BLOOD TRANSFUSION 
of sheep blood to a human being on November 23, 1667, at a meeting of the 
Royal Society at Arundel House, London, in the presence of “considerable 
and intelligent persons.” A silver tube was used for connecting the carotid 
of the sheep with a vein in the subject’s arm. The patient was one Arthur 
Coga, according to Pepys, a “poor and debauched man that the College 
had hired for twenty shillings to have some of the blood of a sheep let into 
his body, . . . their purpose to let in about 12 ounces, which they compute 
is what will be let in in a minute’s time by the watch.” The Society reports 
(Transactions, No. 30, December 9, 1667) that “the man, after this operation 
as well as in it, found himself very well, and hath given his own Narrative 
under his own hand enlarging more upon the benefit he thinks he hath 
received by it than we think fit to own as yet.” 
From this time on many attempts at blood transfusion were made, but 
with far more failures than successes, owing to coagulation of the blood in 
the cannulas and lack of sufficient skill for direct artery-to-vein anastomoses. 
Finally, Bischoff in 1835 introduced the method of defibrination, and from 
1863 to 1884-transfusion was supposed to be a “cure-all,” and the claims 
made for it were preposterous until Cohnheim in 1883 showed the danger of 
injecting defibrinated blood intravenously. 
Human blood was generally employed, but sheep blood was likewise used 
in some instances until Landois in 1875 demonstrated that the corpuscles 
of these and other of the lower animals wrere unsuitable. Needless to state 
reactions were the rule and frequently fatal, but little attention was given 
to any of these which, at the present day, command our serious attention. 
For the following twenty-five years blood transfusion commanded but 
little interest, the pendulum of medical opinion having swung far in the 
opposite direction, until Crile in 1898 revived interest in the subject by 
successful arteriovenous suture perfected and popularized a few years later 
by the work of Carrel and others. Owing to the difficulties attending 
vascular anastomoses it was not long before other methods employing 
cannulas and paraffined tubes and syringes were introduced, so that today 
various devices of this sort are being commonly employed for direct transfu- 
sion instead of the suture of vessel to vessel. 
In the meantime the occurrence of hemolysins and agglutinins in the 
serum of one human being for the corpuscles of another had been discovered, 
the latter by Landsteiner and Shattock in 1900. The significance of these 
in relation to the dangers and untoward effects of transfusion were quickly 
appreciated by Hektoen, Crile, Ottenberg, Libman, and practically all 
who have written on the subject; methods were soon devised for the pre- 
liminary testing of the blood of patient and donor in order that compatible 
blood may be transfused, these precautions having contributed in a very 
important manner to the safety and comfort of the operation. 
Finally, a distinct advance in blood transfusion was made by the use of 
sodium citrate for the prevention of coagulation almost simultaneously by 
D’Agote1 of Buenos Aires, Hustin2 of Brussels, and Weil3 and Lewisohn4 of 
New York. Hirudin and other anticoagulants had been previously em- 
ployed, but the work of these men placed the citrate method on a firm basis 
and opened up blood transfusion as a therapeutic procedure within the reach 
of a great number of physicians lacking skill and experience for vascular 
anastomoses and other direct methods. 
1 An. d. Inst. mod. clln. med., Buenos Aires, 1915, Nos. 1 and 3. 
2 Jour. med. d. Bruxelles, 1914, 12, 436. 
3 Jour. Amer. Med. Assoc., 1915, 64, 425. 
4 Surg., Gynec., and Obst., 1915, 21, 37. SIMPLE HEMORRHAGE AND SECONDARY ANEMIA 
1079 
Since the technic of blood transfusion has become progressively more 
simple and our knowledge of some of the causes of untoward effects with 
means for their prevention advanced, the field of therapeutic application of 
transfusion and indications for its use have been greatly widened, and the 
operation no longer used only as a last resort for life-saving purposes. 
TRANSFUSION FOR SIMPLE HEMORRHAGE AND SECONDARY ANEMIA 
Blood Transfusion and Infusion of Other Fluids in Severe Injuries and 
Postoperative Hemorrhage.—Quite naturally, severe injuries accompanied 
by the loss of large amounts of blood were among the first conditions treated 
with the transfusion of blood. A large literature has accumulated on this 
subject, which need not be reviewed at this time, except to state that practi- 
cally every writer on the subject of blood transfusion has reported upon the 
treatment of simple hemorrhage. 
The literature records some striking examples of success, even in cases of 
almost complete exsanguination with apparent cessation of respiratory and 
cardiac activity. During the Great War blood transfusion was conducted 
on numerous occasions and proved a life-saving procedure with some almost 
miraculous recoveries, as indicated by the experiences of Berard and Lu- 
miere,1 Primrose,2 Harrison,3 Robertson and Watson,4 Crabtree,5 and numer- 
ous others. 
The question when to transfuse for acute hemorrhage depends, of course, 
upon individual circumstances. The drop in blood-pressure consequent to 
hemorrhage is sometimes life-saving, and one should hesitate raising it by 
transfusion until bleeding has been controlled by the tourniquet or operative 
procedures. As a general rule, operation is required, and especially in 
injuries of internal organs when no other means is available for securing 
the bleeding points; the accompanying shock is no contraindication to trans- 
fusion, which should be done on the table. It is well to transfuse the border- 
line cases with pale, cold, cyanotic, and clammy skins; weak, rapid, and 
thready pulse with yawning or dyspnea and a systolic pressure of 80 or less. 
On the other hand, care must be exercised against “pushing the patient 
over” by any procedure likely to add to shock. For this reason due care 
must be taken to reduce the chances and degree of reaction following trans- 
fusion, with special reference to the selection of a compatible donor, the 
injection of whole blood by one of the direct methods being preferable to 
citrated blood, providing the facilities are at hand for rapid and successful 
work. Otherwise citrated blood may be used. The amount of blood to be 
transfused depends upon circumstances. Usually 500 to 1000 c.c. are 
sufficient, although a second transfusion may be required a few hours later 
in extreme cases. Ravdin and Glenn6 report the recovery of an extreme 
case receiving 2100 c.c. distributed over twelve hours. 
If it is largely a question of restoring blood volume, as in hemorrhages of 
moderate degree, the infusion of warm sterile 0.9 per cent, saline solution 
may be suRcient, and has the added advantage of being less likely to produce 
a reaction. However, as shown by Crile7 and others, simple saline solution 
is not as efficacious as whole blood in severe hemorrhage. The latter not 
1 Presse m6d., 1915, 23, 333. 
2 Ann. Surgery, 1918, 68, 118. 
3 Lancet, 1918, 2, 455. 
4 Ann. Surgery, 1918, 67, 1. 
5 Boston Med. and Surg. Jour., 1919, July 17, 60. 
6 Amer. Jour. Med. Sci., 1921, 161, 705. 
7 Hemorrhage and Transfusion, Appleton & Company, 1909. 1080 
BLOOD TRANSFUSION 
only maintains the nutrition of the vasomotor centers essential for the 
maintenance of normal blood-pressure but, being colloidal, exerts a me- 
chanical effect in maintaining pressure because slowly diffusible, whereas 
simple saline solution gives an immediate rise of pressure which is not long 
sustained because of its non-colloidal and rapidly diffusible character. 
During the Great War, when the demand for transfusion in acute hemor- 
rhage became urgent, other substitutes for blood more efficacious and 
sustaining than saline solution were eagerly sought. Bayliss1 advocated the 
intravenous injection of 5 to 7 per cent, solutions of acacia, and Hogan,2 
a 2.5 per cent, solution of gelatin; both of these were regarded better than 
plain saline solution by Hurwitz,3 Rous and Wilson,4 Barthelemy,5 and 
others. However, acacia solutions are not as harmless as was originally 
thought to be the case. Both Kruse,6 and Karsner and Hanzlik7 have 
shown that acacia in concentrations in which it is used intravenously may 
agglutinate human corpuscles not only in the test-tube, but in the living 
animal as well, producing emboli and anaphylactoid symptoms. Foster 
and Whipple8 have shown that acacia may actually reduce the coagulation 
time of the blood by interfering with the prompt return of fibrin to its normal 
value. Henderson and Haggard9 state that while the immediate effects 
from infusion of acacia are good, it does not improve the chances of recovery 
of animals subjected to the so-called “standard hemorrhage.” These reports 
are in harmony with the unfavorable experiences of some surgeons during 
the war,10 and fatalities from its injection have been recorded by Olivecrona11 
and Lee.12 Farrar and Ward,13 however, have used the acacia solution in 
some 400 odd cases, and with properly prepared solutions have had no bad 
results. Bayliss14 has also answered the criticisms advanced against acacia 
solutions and is convinced of the value of 6 per cent, solutions for intravenous 
infusion; he draws particular attention to the necessity of using a good 
brand of acacia. 
As shown by Henderson, Haggard, and their colleagues, the deleterious 
effects of severe hemorrhage are due more to the loss of red corpuscles with 
consequent disturbance of the acid-alkali balance of the blood and the 
development of a form of asphyxia, than to the effects of fall of blood- 
pressure, and in their opinion transfusion of blood is the method of choice 
in treatment. 
Transfusion for Hemorrhage Due to Ruptured Tubal Pregnancy; Placenta 
Praevia; Child Birth; Rupture of the Spleen and Liver, and other Injuries 
of Abdominal Organs.—When large blood-vessels in communication with 
cavities as the pelvis and abdomen, stomach and intestines and pleural sacs, 
or the external world, are ruptured by injury or disease, the resulting hemor- 
rhage is usually severe enough to require blood transfusion. In ruptured 
tubal pregnancy the patient may become almost exsanguinated before the 
1 Brit. Med. Jour., 1918, 2, 553. 
2 Jour. Amer. Med. Assoc., 1915, 64, 721. 
3 Jour. Amer. Med. Assoc., 1917, 68, 699. 
4 Jour. Amer. Med. Assoc., 1918, 70, 219. 
8 Rev. d. Clin., 1920, 39, 271. 
6 Amer. Jour. Physiol., 1919, 49, 137. 
7 Jour. Pharmacol, and Exper. Therap., 1920, 14, 379, 425, 449, 479. 
3 Amer. Jour. Physiol., 1922, 58, 393. 
9 Jour. Amer. Med. Assoc., 1922, 78, 697. 
10 Central. Med. Dept. Lab. Div. Surg. Research, A.P.O., 721, October 27, 1918. 
11 Acta chir. Scandin., 1921, 45, 1. 
Jour. Amer. Med. Assoc., 1922, 79, 726. 
13 Amer. Jour. Obstet., 1920, 1, 1; Surg., Gynec., and Obst., 1921, 32, 328. 
11 Jour. Amer. Med. Assoc., 1922, 78, 1885. SIMPLE HEMORRHAGE AND SECONDARY ANEMIA 
1081 
pressure drops to a life-saving degree because of the capacity of the pelvis 
and abdomen for extravasated blood. 
As a general rule there are no abnormalities in coagulation and bleeding 
time, except possibly in some cases of liver injury with cholemia in which 
coagulation may be delayed owing to deficient calcium for prothrombin 
inactivation. The problem is chiefly one of maintaining a sufficient number 
of corpuscles in circulation. When the bleeding point can be tied or con- 
stricted by operative procedures, as in ruptured tubal pregnancy, rupture 
or injury of abdominal organs, placenta prsevia, and postpartum hemorrhage, 
large transfusions should be done on the table; if operation is not done, the 
transfusions should be smaller (250 to 500 c.c.) and repeated in order not to 
raise the blood-pressure too high or suddenly. 
Doderlein1 and others have stated that if the blood in the peritoneal 
cavity is not contaminated, as is usually the case in ruptured tubal preg- 
nancy and rupture of the spleen, the extravasated blood may be collected 
into citrate solution, filtered through sterile gauze, and immediately trans- 
fused into an ovarian vein if exposed, or a vein at the elbow. If there is 
doubt regarding its sterility, he injects the blood subcutaneously or by 
rectum. As a general rule, however, surgeons prefer a donor and assistant 
to transfuse during the operation with the least possible delay. Titus2 has 
advocated the routine typing of the blood of all pregnant women within 
six or eight weeks of expected delivery, with provisions for a compatible 
donor in case of necessity at a later time. This would appear to be excellent 
advice, and certainly a list of grouped donors should be available in maternity 
hospitals for call with the least possible delay in order to meet these emer- 
gencies. 
Transfusion for Hemorrhage Due to Gastric and Duodenal Ulcers, 
Typhoid Fever, Dysentery, and Pulmonary Tuberculosis.—Ottenberg and 
Libman3 have shown that in the hemorrhages of gastric and duodenal 
ulcers, typhoid fever, and dysentery blood transfusion may not only prove 
a life-saving procedure, but aid in the treatment of the general state of 
lowered nutrition. The coagulation time is not usually increased, but they 
direct attention to the flabby state of the clots incapable of producing firm 
coagula for checking the bleeding. These investigators report that in 
almost all their cases of gastric and duodenal ulcers the hemorrhages stopped 
after transfusion, and especially those with repeated or prolonged bleeding. 
Transfusion should not be done during actual bleeding in order not to 
raise blood-pressure, unless it is necessary to save life, in which case repeated 
trnasfusions of 250 to 300 c.c. of blood may be given. In typhoid fever 
cases Ottenberg and Libman have advised that upon the first appearance of 
blood in the stools preparations should be made for transfusion, so that 
if the necessity arises there may be no unusual delay. 
Likewise in pulmonary tuberculosis and vascular erosions so situated 
that the bleeding point cannot be reached surgically, small and repeated 
transfusions are usually better than large transfusions in order that a low 
blood-pressure shall be maintained. Unfortunately, the administration 
of calcium chlorid does not increase the coagulability of the blood to any 
therapeutic degree, except possibly in obstructive jaundice, in which some 
effect has been noted by Schloessmann,4 Lee and Vincent,5 when administered 
in doses of 100 grains per day for several days. 
1 Deutsch. med. Wchn., 1920, 46, 449. 
2 Boston Med. and Surg. Jour., 1920, 183, 438. 
3 Amer. Jour. Med. Sci., 1915, 150, 36. 
4 Beitr. z. klin. Chir., 1912, 66, 492. 5 Arch. Int. Med., 1915, 16, 59. 1082 
BLOOD TRANSFUSION 
Transfusion Preliminary to Surgical Operations and for the Treatment 
of Shock.—Cases of severe injuries with the loss of large amounts of blood, 
cases of gastric and duodenal ulcers, carcinoma of different organs, uterine 
fibroids with hemorrhage, etc., characterized by grave anemia and con- 
stituting bad operative risks, may be benefited and brought safely through 
operation by preliminary transfusion. In some instances this is accomplished 
by a single large transfusion or the transfusion of 600 to 700 c.c. about three 
days before operation, with a second transfusion on the operating-table or 
at the completion of the operation. 
In shock due to injury and hemorrhage transfusion has proved a life- 
saving procedure and particularly in the experience of surgeons during the 
recent war; shock due to other causes may not be benefited by this measure. 
Harrison1 states that transfusion in the early stages of hemorrhage shock, 
before the “gray-blue” stage is reached, is useful and the most efficient 
treatment. In the terminal or “gray-blue” stage, however, transfusion is 
of little use; likewise other forms of treatment. According to Cannon, shock 
is due to a diminution in the normal alkalinity of the blood caused by de- 
ficient oxidation, secondary to loss of red corpuscles, cardiac weakness, and 
low blood-pressure. In such cases the intravenous injection of simple saline 
solution may be followed by immediate slight improvement, but is apt to be 
followed by collapse within an hour later; glucose and acacia solutions have 
produced more lasting benefit, but in the opinion of most army surgeons 
transfusion and especially transfusion of whole blood by a direct method, as 
with Vincent’s modification of the Kimpton-Brown tube, has proved the 
best method of treatment. 
TRANSFUSION FOR HEMORRHAGIC DISEASES AND DISEASES OF THE 
BLOOD 
The hemorrhagic diseases may be divided into those that are primary, 
including a form of bleeding of the newborn (morbus maculosis neona- 
torum), hemophilia and some of the purpuras, and those that are secondary 
to such blood diseases as pernicious anemia, aplastic anemia and leukemia, 
to hepatic disease, and various acute bacterial infections. In the primary 
hemorrhagic diseases the injection of blood is required primarily for pro- 
moting the coagulation of blood and to stop bleeding, and secondarily, for 
the correction of the resulting anemia. In the second group comprising the 
blood diseases, transfusion is given primarily for the temporary relief of the 
symptoms due to the grave anemia consequent to increased blood destruction 
and partial failure of regeneration, and secondarily, for stopping the hemor- 
rhages that sometimes occur in these diseases. 
Transfusion in Melena Neonatorum.—In hemorrhages of the newborn 
intramuscular and intravenous injections of whole blood have frequently 
yielded excellent results and proved a life-saving procedure. The hemor- 
rhages may come from the nose, mouth, and gastro-intestinal tract, um- 
bilicus, etc. Accidental hemorrhage may be due to faulty tying off of the 
umbilical cord. 
In true hemorrhagic disease of the newborn (morbus maculosus neona- 
torum) the bleeding usually occurs in the gastro-intestinal tract with profuse 
and frequent vomiting and tar-like stools. The coagulation time of the 
blood is delayed, and, according to Whipple,2 there is a deficiency in pro- 
thrombin or, at least, the elaboration of this substance is unstable. 
Rabbit-, horse-, and human sera have been employed in treatment by 
1 Jour. Amer. Med. Assoc., 1918, 71, 1403. 
2 Arch. Int. Med., 1912, 9, 365. HEMORRHAGIC DISEASES AND DISEASES OF THE BLOOD 
1083 
subcutaneous or intramuscular injection, but, as stated in Chapter XL, 
much better results have followed the administration of whole blood. 
Treatment should be given on the first appearance of hemorrhage. An 
efficient and simple procedure is to draw a 20 c.c. syringeful of blood from a 
vein of the father or some other healthy adult and immediately inject into 
the muscles of the buttocks of the child before coagulation occurs. By 
using a No. 18 needle for aspirating the blood a syringe may be filled in a 
few seconds; the injection may be given with the same or a slightly smaller 
needle. It is well to repeat this injection two or three hours later. 
Defibrinated blood may be injected intramuscularly or subcutaneously 
in dose of 10 to 20 c.c. every two or three hours until bleeding has been 
checked; a method is described later in this chapter. 
If the hemorrhages have been very copious and frequent, with signs of 
approaching exsanguination and collapse, transfusion of blood will be 
required. The injections are usually given by way of the superior longitu- 
dinal sinus. Helmholtz1 has described a special apparatus for holding the 
head and transfusing whole blood by a syringe method. 
Great care must be exercised against giving the injectionsinto the sub- 
stance of the brain; the method is by no means simple or devoid of danger. 
At least 50 to 75 c.c. of blood should be given, depending on circumstances; 
larger amounts may embarass the heart muscle already weakened by anemia. 
Citrated blood should not be given unless means for giving whole blood are 
not available. I have transfused defibrinated blood filtered through gauze 
with complete success on several occasions. The Kimpton-Brown tube 
may be used for transfusing whole blood or a syringe method employing a 
battery of three syringes. Vincent2 has described the successful transfusion 
of 7 cases by means of the external jugular vein and paraffin-coated tubes 
of proper length and size. 
In none of these methods is it absolutely necessary to group the bloods 
of the patient and donor, inasmuch as agglutinins and hemolysins are not 
present in the baby’s plasma. As a matter of caution, however, the writer 
tests the donor’s serum against the baby’s corpuscles before sinus transfu- 
sions (see chapter on Hemagglutinins). 
Transfusion in Hemophilia.—In hemophilia blood transfusion produces 
prompt effects and acts as a specific agent for the treatment of hemorrhage. 
Unfortunately the effects are not lasting, and no means for the cure of this 
baffling disease have as yet been discovered. As advised by Ottenberg 
and Libman, every hemophiliac ought to have at his call several persons whose 
blood by previous tests is known to be compatible and acceptable, and who are 
willing to give blood for transfusion when the necessity arises. 
According to Howell3 the hemophiliac condition, both the true hereditary 
type and in those instances in which the disease appears without a definite 
history of heredity, is due to a diminution of prothrombin with a relative 
excess of antithrombin; the calcium content is normal. Transfusion of 
normal blood supplies the deficient prothrombin for a varying period of 
time during which the patient is protected. Ottenberg and Libman have 
suggested that it may be worth while injecting a syringeful of whole blood 
(25 c.c.) every one to three months for prophylactic purposes. 
Injections of serum have been employed in the treatment of hemophilia 
and apparently with success in some cases, but well-controlled studies in this 
disease have proved that serum, and especially old serum, is much less 
1 Amer. Jour. Dis. Child., 1915, 10, 194. 
2 Boston Med. and Surg. Jour., 1912, 166, 627. 
3 Arch. Int. Med., 1914, 13, 76. 1084 
BLOOD TRANSFUSION 
efficacious than blood in hemorrhagic diseases like hemophilia due to pro- 
thrombin deficiency. Lewisohn,1 Losee,2 and practically all writers on the 
subject of transfusion have recorded successful results. If serum is adminis- 
tered, it should be fresh human serum free of agglutinins for the corpuscles 
of the patient and given intravenously, as advised by Weil.3 
Inasmuch as the hemophiliac may require several transfusions during 
the course of his life, the operation should be done with the least possible 
pain and without cutting down on the vein; the vein should be entered, if 
possible, by simple passage of a needle. Citrated blood has given just as 
good results as whole blood by a direct method, even though the chances 
for a reaction are greater. In slight hemorrhages one may first try the intra- 
muscular injection of 20 to 40 c.c. of defibrinated blood, and if this fails, 
100 c.c. of citrated blood should be given intravenously. As a general rule, 
these small amounts suffice for checking the hemorrhage, but if a large amount 
of blood has been lost with signs of approaching collapse, 500 c.c. should be 
transfused. 
In necessary operations upon hemophiliacs, as the extraction of teeth, 
a prophylactic injection of blood should be given, preferably 50 to 100 c.c. of 
citrated blood by intravenous injection. Or 20 to 40 c.c. of blood may be 
injected intramuscularly with provisions made for transfusion if bleeding 
occurs after the operation. The intravenous injection, however, can be 
made less painful, although one naturally hesitates to puncture the vein of 
a hemophiliac. A small needle should be used (No. 20 is large enough for 
citrated blood), and it is my custom, to pass the needle for at least % inch 
in the subcutaneous tissues before entering the vein. 
Transfusion in Purpura Haemorrhagica and Chronic Purpura.—As shown 
by Duke,4 bleeding in acute and chronic idiopathic purpura haemorrhagica 
and some other forms of purpura is due primarily to a deficiency in blood- 
platelets; other factors concerned in coagulation appear to be present in 
normal amounts. Duke has laid emphasis upon the mechanical action of 
the platelets and their ability to stop hemorrhage by the role they play in 
the formation of thrombi; others believe that the platelets are an important 
source of prothrombin and a thromboplastic substance which hasten coagula- 
tion by neutralizing the antithrombin present in normal blood. Whatever 
may be the mechanism of their action, and it is probable that both processes 
are concerned, as suggested by Hurwitz,6 bleeding in purpura appears to be 
due in most cases to a deficiency in their numbers. 
For treatment fresh human serum in dose of 10 to 20 c.c. may be injected 
subcutaneously, but better results have followed the intramuscular injection 
of 20 to 40 c.c. of defibrinated human blood. In severe cases it may be 
necessary to transfuse 100 to 250 c.c. of citrated blood. 
Transfusion in Pernicious Anemia.—No other disease has excited as 
much discussion relative to the value of blood transfusion as pernicious 
anemia. A large literature has accumulated in which the experiences of 
different observers have varied widely,6 but recent reviews of the subject 
have served to crystallize our knowledge of the subject and correlate expe- 
riences. 
1 Amer. Jour. Med. Sci., 1919, 157, 253. 
2 Amer. Jour. Med. Sci., 1919, 158, 717. 
3 Trib. med., Paris, 1907, 21. 
4 Arch. Int. Med., 1913, 11, 100; Jour. Amgr. Med. Assoc., 1915. 65, 1600. 
5 Amer. Jour. Med. Sci., 1917, 154, 689. 
6 For review of literature up to 1910 see article by Schultz in Crawitz, Klin. Path. d. 
Blutes, Leipsic, 1911, 381. For literature between 1910 and 1920 see article by Anders in 
Amer. Jour. Med. Sci., 1920, 158, 659. HEMORRHAGIC DISEASES AND DISEASES OF THE BLOOD 
1085 
The cause of pernicious anemia is still unknown; therefore our therapeutic 
efforts are purely empirical. The most important change appears to be the 
presence of a hemolytic agent capable of destroying the red blood-corpuscles 
more quickly than they can be produced. This factor, however, has not 
been identified, although it may be a living microparasite or its products. 
The plasma does not appear to possess increased hemolytic activity, and 
the location of the hemolytic agent is unknown. One naturally suspects 
the spleen in this relation, because it shows evidences of excessive erythro- 
cytic activity in pernicious anemia, but the evidence is not clear or conclusive, 
although, as will be discussed shortly, some observers believe that removal 
of this organ constitutes an important therapeutic measure. 
In addition to this important factor of increased blood destruction, 
evidence at hand indicates that there may be an acquired fault in hemato- 
genesis on the part of the bone-marrow. In view, however, of the histologic 
evidences usually presented of rather strenuous efforts in this direction, it 
would appear to be a less important etiologic factor, and especially in the 
early stages of the disease. 
The purpose of treatment is to prolong life by relieving anemia and pro- 
ducing frequent' periods of remission. While cures have been reported, it 
is likely that these are explicable on the ground of faulty diagnosis or insuf- 
ficient observation. Much in the way of symptomatic relief can be offered, 
however, except in those instances of acute virulent, rapidly progressirg 
cases with none, or very short, remissions. 
The general plan of treatment consists in a rigid and painstaking search 
for and treatment of foci of infection as emphasized by Barker and Sprunt,1 
and the administration of a full roborant diet, full doses of hydrochloric acid 
after meals to correct the anacidity usually present, etc. Iron may be 
administered by subcutaneous injection, and neoarsphenamin in small 
doses by irtravenous injection for the effects these may have on the hemo- 
globin and its functions. If neoarsphenamin is administered for its gereral 
and hematinic effects, 0.1 or 0.2 gm. dissolved in 10 c.c. of sterile water and 
injected intravenously once a week for four or five weeks after transfusion 
followed by an interval, would appear to be sufficient. Of course, if syphilis 
is known to be present, more rigorous treatment may be required. 
Unquestionably, however, the most important single therapeutic agent 
is blood transfusion. 
The beneficial effects and value of blood transfusion in pernicious anemia 
are ascribed to a variety of possibilities: (a) The blood is probably func- 
tionally active as a transplanted tissue at least for a few weeks in the average 
case. (b) The transfused blood may act as a stimulant of the bone-marrow 
and revive for a time its flagging hematogenic properties, (c) Probably 
in pernicious anemia an effort is made to produce a substance in the 
nature of an antihemotoxin capable of neutralizing the hemotoxic agent, 
and blood transfusion may stimulate this process. (d) Finally, transfused 
blood may have a direct antagonistic effect upon the hypothetic hemolytic 
agent. 
Whatever may be the mechanism of the beneficial effects of blood trans- 
fusion in pernicious anemia, the accumulated experiences of most observers 
indicate that the procedure is capable of initiating periods of remission even 
though the duration of these periods may not be longer than naturally occurs 
in the disease. But when transfusion is given early in the disease before the 
recuperative powers of the bone-marrow are exhausted and the degenerative 
changes in the spinal cord and other organs due to the anemia become appar- 
3 New York Med. Jour., June 9, 1917. 1086 
BLOOD TRANSFUSION 
ent, these remissions may add years to life, aside from the relief from “anemic 
fever,” increased appetite, and other general improvement that may follow. 
In the statistical study made by Anders remissions were induced by trans- 
fusion in 56.3 per cent, of 362 cases, and this agrees very closely with the 
experience of Ottenberg and Libman to the effect that transfusion initiates 
remissions in fully one-half of cases. 
When to transfuse is usually an important question. Many patients 
naturally fear the operation and the possibility of disagreeable reactions. 
The tendency has been to delay transfusion too long; most reported cases 
have been transfused when the hemoglobin has reached 30 per cent, or less. 
This is unwarranted and uncalled for; even with the citrate method which 
gives a larger number of reactions than direct methods, attention to technical 
details and the injection of small amounts of blood at frequent intervals 
leaves little to be feared, and the technic is essentially simple. 
Remissions are more easily induced in the early than in the late stages 
of the disease, and the relief from anemia will retard the degenerative effects 
and general symptoms directly referable to the anemia. Even when the 
hemoglobin falls to 20 per cent, transfusion is apt to be worth while and a 
means for delaying a fatal outcome, but it would appear that best results 
are to be expected when transfusion is initiated in the earlier stages, as soon 
as the diagnosis is certain and general treatment is found to be unsatisfactory, 
and before the hemoglobin falls below 50 per cent. 
How much blood to transfuse and the frequency of transfusions are also 
questions of practical importance. The tendency has been to give large 
transfusions at irregular intervals. As previously stated, transfusion should 
be resorted to earlier in the disease than has been the general custom, and 
when the hemoglobin is not below 40 per cent, the injection of 200 to 250 c.c. 
of blood is sufficient, as the induction of the remission bears no relation to 
the amount of blood transfused. This amount of blood is less likely to 
produce a reaction than 500 c.c. or more, and renders the patient more 
willing for subsequent injections. It is likewise usually sufficient for the 
control of hemorrhage. 
However, in cases of grave anemia, it may be necessary to transfuse 500 
to 1000 c.c. of blood at least for the first time; under these conditions it is 
better not to use citrated blood if a direct method that does not produce too 
much injury to the veins of patient and donor can be employed. 
The frequency of transfusion depends upon individual circumstances. 
After transfusion the number of erythrocytes and percentage of hemoglobin 
are usually increased, and an effort should be made to prevent their dropping 
back to pretransfusion levels as long as possible. Hemoglobin estimations, 
erythrocyte counts, and an inspection of the erythrocytes should be made 
at least once a week in periods of uncertainty and once a month under any 
condition. Vogel and McCurdy1 advise that these routine examinations 
include a study of reticulated red blood corpuscles by means of vital staining; 
according to these investigators the presence of reticulated and nucleated 
erythrocytes are an indication of hemopoiectic activity of the bone-marrow, 
and drop in their percentage constitutes an indication for transfusion. 
The method of transfusion to be employed has already been briefly men- 
tioned. The injection of defibrinated blood so popular in Germany up to 
recent years is not a method of choice, although if the donor is properly 
selected, the blood allowed to stand for an hour and filtered before injection, 
transfusion with defibrinated blood serves a useful purpose, and especially 
if only 100 to 200 c.c. are required. The intramuscular injection of 20 to 
1 Arch. Int. Med., 1913, 12, 707. HEMORRHAGIC DISEASES AND DISEASES OF THE BLOOD 
1087 
40 c.c. of defibrinated blood will usually check hemorrhage if this is the chief 
object of treatment. 
Citrated blood has generally been employed, and when the amount of 
citrate is 0.2 to 0.25 per cent, the method is quite acceptable when not more 
than 250 to 300 c.c. of blood are required. Reactions are more likely, 
however, than when blood is transfused by the Kimpton-Brown, Lindemann, 
Unger, and such methods, but these should not be employed unless the 
physician possesses the necessary assistance, experience, and skill for com- 
plete success. Direct vascular anastomosis is not advisable, becuase the 
patient and donor alike may refuse subsequent transfusions by this method. 
Extra care should be exercised in the selection of a donor. Since the 
plasma of the patient may possess increased hemolytic properties it is 
always well to match the bloods of patient and donor direct for both agglu- 
tinins and hemolysins, in addition to selecting a donor belonging to the 
same blood group. Every effort should be made to reduce the chances for 
reactions after the transfusion and to render the operation as painless and 
agreeable as possible. Ottenberg and Libman have advised using the same 
donor as long as his blood induces a remission; if the blood of a donor fails 
to bring about this result his blood should not be used for subsequent, 
transfusions. 
The value of splenectomy in the treatment of pernicious enamia is still 
uncertain. Ottenberg and Libman believe that this operation should be 
done if good remissions are no longer obtainable by transfusion alone and 
if the patient can be brought up to a fair standard of operative risk. For 
this purpose one or more transfusions may be required at intervals of three 
or four days before operation with another transfusion on the table. 
Krumbhaar,1 in a statistical study of 153 cases collected from literature 
found that the mortality was about 20 per cent, and the improvement only 
transient. Operative risks are increased when the hemoglobin is below 
35 per cent. Lindemann,2 McClure,3 Giffin,4 and others, however, have 
endorsed splenectomy as part of the treatment, and especially in young and 
middle-aged patients in fair general condition and in whom the spleen is 
moderately enlarged. There is no evidence, however, of cure, although a 
gain in weight and general improvement have been observed in at least 
75 per cent, of cases. 
In general terms, it would appear that the question of splenectomy is 
worthy of serious consideration and especially early in the disease when the 
operation risks are almost negligible. The spleen is known to show changes 
of increased blood destruction in pernicious anemia, and it is possible, but 
not proven, that splenectomy may remove at least some of the primary 
hemotoxic factors and processes responsible for the disease. The operation, 
however, should be preceded and followed by transfusions; aside from opera- 
tive risks, the removal of this organ does not appear to be injurious. 
Transfusion in Aplastic Anemia.—In the aplastic anemias the coagulation 
time of the blood is usually prolonged, and hemorrhages may require the 
injection of serum or blood for their control. Bleeding is apparently due to 
a hypofunctionating marrow as shown by the studies of Hurwitz and Drinker5 
and the Drinkers,6 with a deficiency in platelets and prothrombin. Ac- 
cording to Ravdin and Glenn, blood transfusion does not cause a remission 
in these cases, but rapidly repeated transfusions (twice a week) may relieve 
symptoms and enable the patient to conduct his affairs for a variable period, 
1 Jour. Amer. Med. Assoc., 1916, 67, 723. 
2 Jour. Amer. Med. Assoc., 1915, 64, 613. 
3 Jour. Amer. Med. Assoc., 1916, 67, 793. 
4 Jour. Amer. Med. Assoc., 1917, 68, 429. 
5 Jour. Exper. Med., 1915, 21, 401. 
6 Amer. Jour. Physiol., 1916, 41, 5. 1088 
BLOOD TRANSFUSION 
though never a long one. Dorrance has not observed good results from 
transfusion in this disease. 
Transfusion in Leukemia.—Hemorrhages in the leukemias are due to 
different factors. The coagulation time may or may not be prolonged. 
In the lymphatic leukemias the platelets are usually diminished, but in- 
creased in the splenomelogenous type. Antithrombin is apt to be increased 
in both types, while fibrinogen is absent or diminished in the lymphatic 
types. 
Injections of blood are palliative and not curative. If hemorrhages 
require treatment, whole blood rather than serum should be given in order 
to obtain the effects of the platelets and other coagulating principles men- 
tioned. Intramuscular injections cf 20 to 40 c.c. of defibrinated blood or 
whole blood in a syringe before coagulation occurs may be tried; severe and 
recurring hemorrhages require transfusion of about 250 c.c. of blood. 
Hemorrhage in Cirrhosis of the Liver; Jaundice; Acute Yellow Atrophy 
of the Liver.—Hemorrhages associated with acute or chronic injury of the 
liver are generally due to a reduction in fibrinogen. The blood coagulates, 
but tardily, and the clots are apt to be soft and flabby. Low fibrinogen is 
especially associated with cirrhosis of the liver, and Erben1 has described 
the presence of a ferment in the blood in cirrhosis and certain types of 
myeloid leukemia, capable of producing hemorrhage by dissolving the 
fibrin of the clot. As previously stated, calcium is not actually diminished 
in obstructive jaundice, but it may be bound by the bile pigment with a 
functional decrease. 
When surgical operations are required in the presence of jaundice and 
hemolytic ictero-anemia, it may be well to prepare the patient by the intra- 
muscular injection of 20 to 40 c.c. of defibrinated human blood. Injections 
of serum and especially of old serum are much less efficacious. If bleeding 
occurs after operation or injury, or spontaneously, intramuscular injections 
may suffice, but if not, transfusion of 250 to 300 c.c. of blood may be required. 
It is possible that the beneficial results are due in part to the calcium ad- 
ministered in the transfused blood. Walters2 has reported that the daily 
intravenous injection of 5 c.c. of sterile 10 per cent, solutions of calcium 
chlorid for three injections, combined with the administration of large 
quantities of carbohydrates and fluids by mouth and glucose solution by 
proctoclysis, enabled operations upon 34 patients with extreme jaundice 
without a death from hemorrhage. 
• Transfusion in Hemorrhages Secondary to Severe Infections in the 
Treatment of Septicemia, Influenzal Pneumonia, etc.—Hemorrhages may 
occur in the course of some of the acute infectious diseases. According to 
Duke3 the platelets may be diminished. In septicemia the coagulation time 
may be prolonged, due to an excess of anti thrombin. Prothrombin and 
fibrinogen may be diminished. Intramuscular injections of defibrinated 
blood generally suffice in treatment; in severe hemorrhages blood transfusion 
may be required, and owing to the necessity for avoiding reactions a direct 
method is preferable to citrated blood if the means are available for this 
kind of transfusion. 
Many reports have been made on the treatment of septic infections with 
blood transfusions, with particular reference to septic wounds and ulcerative 
endocarditis. Most surgeons are of the opinion that the transfusion of plain 
normal blood is of no value in the treatment of sepsis, although it may 
1 Wien. klin. Wchn., 1902, 15, 276. 
2 Jour. Amer. Med. Assoc., 1922, 79, 1793. 
3 Jour. Amer. Med. Assoc., 1910, 45, 1185. TRANSFUSION IN OTHER CONDITIONS 
1089 
relieve for a time the associated anemia so apt to develop in prolonged 
staphylococcus and streptococcus infections. Repeated small transfusions 
of 30 to 50 c.c. of blood may, however, aid in tiding a patient over a critical 
period, as indicated by the reports of Moncamy1 and Polak.2 As stated 
in Chapter XL transfusion of blood from donors previously immunized by 
few injections of vaccine may prove worth while, as indicated by the reports 
of Hooker,3 Wekesser,4 and Fry5 in the treatment of chronic wound infections, 
of Levison6 in the treatment of ulcerative endocarditis, and of Neuman7 
in the treatment of typhoid fever. 
In Chapter XL mention was also made of transfusions of citrated blood 
from convalescents in the treatment of influenzal pneumonia and other 
acute infections. 
Transfusion in Acute Poisoning.—In illuminating-gas poisoning transfu- 
sion after the removal of a large amount of blood is now regarded as the 
best method of treatment. Phlebotomy may be done on one arm, and 
after the escape of 500 c.c. of blood, transfusion is started on the other arm. 
As much as 1000 c.c. of blood may be removed and an equal amount trans- 
fused. In severe cases a second bleeding and transfusion, or transfusion 
alone, may be required within twelve hours. Artificial respiration may be 
required. 
Similar treatment may be employed in severe cases of poisoning due to 
other substances having their primary effect upon the blood, as benzol, 
nitrobenzol, etc. 
Transfusion in Pregnancy; Acidosis; Diabetic Coma.—Mention has 
already been made of the usefulness of transfusion in the treatment of 
severe anemia from hemorrhage in ruptured tubal pregnancy, placenta 
praevia, and postpartum hemorrhage. 
Bell8 has recently advised transfusion in the treatment of eclampsia, and 
Gettler and Lindemann,9 in the treatment of acute and severe acidosis 
complicating pregnancy, the donor first being prealkalinized by the adminis- 
tration of 20 gm. of sodium carbonate every two hours for eight doses, 
transfusion being done about one-half hour after the last dose. 
Garnett10 has reported favorably upon the treatment of 2 cases of per- 
nicious vomiting of pregnancy by transfusion with the blood of women donors 
about ten days after delivery. 
Ottenberg and Libman have reported upon the results of transfusion of 
4 cases of diabetes; temporary improvement was noted in some of these, but 
was without effect upon the ultimate course of the disease. 
Apparently transfusion has not been used in the treatment of uremia, 
but phlebotomy with transfusion may be worthy of trial, although only 
temporary effects would be expected. 
Transfusion for Debilitated States.—Mention has already been made of 
the usefulness of transfusion for preparing cancerous individuals for opera- 
tion; also for the relief of anemia and cachexia due to hemorrhages or blood 
destruction in other conditions requiring operative measures. Blood trans- 
fusion is thought to possess nutritive value in addition to correcting at least 
temporarily the faults of deficient numbers of red corpuscles and hemoglobin. 
TRANSFUSION IN OTHER CONDITIONS 
1 Surg., Gyn., and Obstet., 1918, 27, 221. 
2 Amer. Jour. Obstet., 1919, 80, 291. 
3 Ann. Surg., 1917, 66, 513. 
4 Jour. Amer. Med. Assoc., 1917, 69, 2182. 
5 Brit. Med. Jour., February 28, 1920, 290. 
6 Jour. Lab. and Clin. Med., 1921, 6, 191. 
7 Amer. Jour. Med. Sci.,l;1922, 164, 690. 
8 Brit. Med. Jour., 1920, May 8, 625. 
9 Jour. Amer. Med. Assoc., 1918, 68, 594. 
10 Amer. Jour. Obst., 1917, 76,*303. 1090 
BLOOD TRANSFUSION 
Fischer1 has found transfusion useful in atrophic infants and children, and 
Kerley2 has likewise reported increase of weight in a series of such children 
following transfusion with “the change from sickly, whiny infants into 
happy, apparently well infants. The patients were transformed from those 
with a digestive capacity barely able to maintain existence into those that 
took on the normal constructive processes of early life.” Marriott3 has 
also recommended transfusion as a valuable measure in the treatment of 
“athrepsia” because of the diminished protein concentration of the plasma 
and the lowerea blood volume found in this condition. 
REACTIONS FOLLOWING BLOOD TRANSFUSION 
Reactions frequently follow blood transfusion; and in severely sick indi- 
viduals, especially cases of shock following injuries and hemorrhage, due 
account must be taken of this possibility in order to avoid fatal consequences. 
The symptoms of these reactions when compatible blood is .transfused, 
vary from a slight rise of temperature that may not be detected at all or a 
fever of 2° to 3° F. with malaise, nausea, and perspiration, to high fever, 
chills, vomiting, and urticaria. Cardiac dilatation may occur if transfusion 
is too rapid. The symptoms following transfusion of incompatible blood may 
have the added effects of intravascular agglutination and hemolysis. The 
reaction may occur during the operation or shortly afterward. Tingling 
body pains, precordial oppression and extreme anxiety, cyanosis, dyspnea, 
slow thready pulse, cold clammy skin and unconsciousness develop followed 
by chills, fever, and delirium. Jaundice and hemoglobinuria develop and 
death may occur. 
The causes of these reactions are not well understood. Very probably 
more than one factor may be responsible, and the possible causes may be 
summarized as follows: 
1. Incompatibilities of the blood of patient and donor; that is, agglutinins 
and hemolysins in the blood of the patient for the corpuscles of the donor or 
in the blood of the donor for the corpuscles of the patient. Undoubtedly 
gross errors in the preliminary tests or transfusion without these tests at 
all, may lead to accidents of this kind. In carcinoma cases, the serum of the 
patient may be hemolytic and result in intravascular hemolysis, despite 
careful preliminary tests. 
2. Changes in the blood and especially of the platelets are believed 
important by Drinker and Brittingham.4 These changes resulting in an 
increased toxicity of the transfused blood, are believed tc occur even though 
coagulation of the blood does not actually take place, and especially in the 
citrate method. As shown by the studies of Novy and De Kruif,5 simple 
defibrination of rabbit blood, or merely withdrawal of rabbit or rat blood 
and reinjection into the same animal or animals of the same species, may 
result in the acquisition of toxic properties designated by them as “anaphyla- 
toxins” because of the kind of lesions and symptoms produced. These toxic 
substances may be derived from preclotting changes involving the platelets, 
or they may be the result of disturbed colloidal conditions. At any rate, 
only direct vessel-to-vessel transfusion appears to remove the possibility of 
their production. 
3. In the citrate method Drinker and Brittingham believe that the 
anticoagulant promotes hemolysis of the corpuscles, even though there may 
1 Med. Record, 1916, 89, 223. 
2 Amer. Dis. Children, 1917, 14, 470. 
3 Amer. Jour. Dis. Child., 1920, 20, 461. 
4 Arch. Int. Med., 1919, 23, 133. 
5 Jour. Infect. Dis., 1917, 20, 566, 589. REACTIONS FOLLOWING BLOOD TRANSFUSION 
1091 
be no evidence of this in the test-tube. Bernhein1 has also emphasized the 
dangers due to hemolysis, but according to Bayliss2 hemoglobin itself is not 
toxic and serves as a carrier of oxygen. 
4. Toxic substances may be derived from new rubber tubing in the citrate 
method, as shown by Bussman, unless precautions are taken to first soak it 
overnight in 5 per cent, sodium hydroxid followed by thorough washing. 
5. In the citrate method enough of the anticoagulant alone may be 
injected in large transfusions to produce some untoward symptoms. The 
amount of citrate advised is between 0.2 and 0.3 gm. per 100 c.c. of blood. 
If of blood are transfused, from 2 to 3 gms. of the citrate will be 
given. These amounts are regarded as safe in so far as severe toxic reactions 
are concerned, and as much as 5 gm. may be given. For white rats I have 
found that 02. gm. per kilo of body weight represents the maximum tolerated 
dose, corresponding to 12 grams for an individual weighing about 130 pounds. 
While the intravenous injection of 2 or 3 gm. of citrate may be without 
danger to life, yet it is entirely likely that untoward effects may be produced. 
6. The blood of one individual may act as a foreign protein when trans- 
fused to another and initiate the changes and reaction described in Chapter 
XXXIX. 
7. Finally, some reactions may be anaphylactoid due to some effect of 
foreign protein, as suggested by Bayliss and also by Bowcock.3 It is possible 
that with intravascular hemolysis the hemoglobin may not be toxic, but the 
corpuscular fragments may produce embolic or colloidal anaphylactoid 
lesions and symptoms. The possibility of acquired sensitization and true 
anaphylactic reactions is present, if the blood proteins of the donor undergo 
sufficient change to act as a foreign protein. 
This possibility is suggested by clinical observations, showing that after 
repeated transfusions and especially in pernicious anemia, the patient 
acquires a tendency for reactions, and particularly if the same donor is 
employed. Apparently agglutinins and hemolysins capable of demon- 
stration in the test-tube do not develop, but an increased tendency toward 
reactions may become apparent, as reported by Sydenstricker, Mason, and 
Rivers.4 Bowcock has also described the occurrence of these reactions in 
cases of pernicious anemia, and states that transfusion may become self- 
limited in this disease because of the inadequacy of methods for selecting 
suitable donors. Meleny, Stearns, Fortuine, and Ferry5 have likewise 
observed this acquired tendency, and believe that the blood of one donor 
may be more likely to produce reactions than others. 
The effects of transfusion upon the blood of the patient are variable, and 
probably due to the mechanical results in relation to the quantity of blood 
transfused as well as to reactive effects upon the patient’s blood-making 
organs. As shown by the studies of Huck,6 transfusion is usually followed 
by an immediate increase of red blood-corpuscles, but not in proportion to 
the amount of blood injected. In some cases this increase lasts only for 
a few hours, in others for twenty-four hours followed by a drop to the pre- 
transfusion level, and in others the increase reaches its maximum in twenty- 
four to forty-eight hours and remains higher than before transfusion for a 
variable time. The hemoglobin is generally increased, reaching the maxi- 
mum in twenty-four hours, but these changes are not necessarily in propor- 
'Lancet-Clinic, 1915, 113, 253. 
2Brit. Jour. Exper. Path., 1920, 1, 1. 
3 Bull. Johns Hopkins Hosp., 1921, 32, 83. 
4 Jour. Amer. Med. Assoc., 1917, 68, 1677. 
5 Amer. Jour. Med. Sci., 1917, 154, 733. 
6 Bull. Johns Hopkins Hosp., 1919, 30, 63. 1092 
BLOOD TRANSFUSION 
tion to the red corpuscle changes. The leukocytes are almost always in- 
creased with an increase of the polymorphonuclear neutrophils; in some 
cases, however, there are no changes or even a decrease. As stated by 
Huck, it may be said that transfusion raises the count, but the effects are 
essentially biologic, involving the redistribution of blood in the body, and 
the exact nature of the changes are not understood. 
CHOICE OF METHODS FOR BLOOD TRANSFUSION 
Until about ten years ago blood transfusion was an exceedingly difficult 
procedure involving artery-to-vein anastomoses and requiring refined surgical 
technic and considerable skill and experience; even under these conditions 
the operation frequently failed, and the ordeal was one which most patients 
refused to undergo a second time. 
With the introduction of syringe-cannula and paraffin-coated tube 
methods the technic of transfusion of whole blood was greatly simplified, 
but the introduction of anticoagulants and especially sodium citrate, brought 
the operation within the reach of many more physicians. 
All methods have certain advantages and disadvantages which may be 
summarized as follows: 
1. Direct Artery-to-vein or Vein-to-vein Anastomosis.—The chief and 
only, but important, advantage is the transfusion of blood practically 
unchanged from its condition in the donor. Reactions are almost eliminated, 
or at least reduced to less than 10 per cent. 
The disadvantages are the chances of failure due to clotting, leakage, 
etc. Also the fact that cutting is required, with pain, fright, and nervousness 
on the part of both patient and donor. The method is not at all adapted 
for cases requiring repeated transfusion. 
Furthermore, it is impossible to measure the amount of blood transfused, 
although Libman and Ottenberg1 have described a method for approximating 
the amount. Dorrance2 has recorded one instance where a back flow from 
patient to donor must have occurred, with consequent infection of the donor. 
This method has, therefore, little to recommend it, and has been largely 
discarded, as, likewise, modifications of the method whereby the vessels are 
united by buttons or tubes. 
2. Syringe, syringe-cannula, paraffin tube, and special methods of a 
similar kind for the transfusion of unmodified blood have the advantage of 
producing fewer reactions and injecting nothing but blood free of anti- 
coagulant. When it is especially advisable to avoid reactions, as in cases 
of severe hemorrhage with shock and cases of primary and secondary anemia 
in an extreme state of exhaustion, when a reaction may prove to be the 
“final push,” these methods are especially valuable, as recently stated by 
Bernheim.3 Lindemann,4 Losee,5 Unger,6 Lundblad,7 and others prefer these 
methods, and the Vincent modification of the Kimpton-Brown tube was 
used quite extensively in the American, French, and British armies during 
the World War. 
As stated by Unger, when the indications are to replace diseased with 
normal blood for the purpose of deriving the benefits of the latter as a 
transplanted tissue, whole unmodified blood is to be preferred to citrated 
blood. 
1 Jour. Amer. Med. Assoc., 1914, 62, 764. 
2 Penn. Med. Jour., 1922, 25, 245 (discussion). 
3 Jour. Amer. Med. Assoc., 1921, 77, 275. 
4 Jour. Amer. Med. Assoc., 1916, 66, 624. 6 Jour. Amer. Med. Assoc., 1921, 77, 2107. 
5 Amer. Jour. Med. Sci., 1919, 158, 711. 7 Penn. Med. Jour., 1922, 25, 245. CHOICE OF METHODS FOR BLOOD TRANSFUSION 
There can be no doubt, however, that all of these methods are more 
complicated than the citrate method. They require experience, skill, and 
trained assistance, and especially the Ziemssen syringe method and its 
modification by Lindemann, and the Unger method. The Kimpton-Brown 
tube can be worked by one operator, and with a properly prepared tube is 
an excellent method. 
3. The citrate method has the great advantage of simplicity. Any 
physician practised in intravenous therapy can make it work satisfactorily. 
No special apparatus is required; any apparatus suitable for intravenous 
infusion may be employed, although special apparatus is advisable for 
preventing accidental contamination of the blood during its collection and 
administration. 
Since the introduction of this method in 1915, and especially by Lewisohn, 
it has easily become the most popular, although its value is reduced by 
certain disadvantages. 
In addition to the ease with which blood is collected and administered, 
this method has the additional advantage of producing little or no pain, as 
it is unnecessary to cut down on the vein of either patient or donor, except, 
possibly, in children and obese adults. Furthermore, the transfusion can 
be done with the minimum of fuss and excitement, and the donor need not 
be brought in contact with the patient. Blood may be collected at one 
place and infused some hours later, or even next day at another place. It is 
especially useful for the transfusion of nervous and apprehensive patients, 
as the injection of blood can be given at the bedside. 
The method is well suited for the treatment of hemorrhage unless the 
patient is shocked, when the effects of a reaction are to be borne in mind. 
Bernheim states that the citrate method produces reactions in 20 to 40 
per cent, of cases, as compared with 5 per cent, or less reactions when un- 
modified blood is transfused. Taking a rise of temperature of 2.5° F. as 
constituting a reaction, Drinker and Brittingham observed 53.3 per cent, 
reactions, with chills in 40 per cent. Pemberton, however, observed reactions 
in but 21 per cent, of 1036 transfusions. Meleny and his associates observed 
reactions in 70 per cent, of cases transfused with citrated blood, as compared 
with 26.4 per cent, of reactions when unmodified blood was transfused by 
the Lindemann syringe-cannula method. There can be no doubt, there- 
fore, that citrated blood is more likely to produce reactions than unmodified 
blood; in addition, Unger states that citrate renders the corpuscles more 
fragile, diminishes complement in the donor’s serum, and reduces the phago- 
cytic activities of leukocytes. However, according to Mellon, Hastings and 
Casey,1 Unger’s statements concerning the destructive influence of sodium 
citrate upon complement, phagocytic activity, and opsonins have not been 
confirmed by their experiments, and an explanation for the reactions attend- 
ing the use of citrate must be sought along other lines. According to Lewis- 
ohn2 a mixture of sodium citrate with blood in the proportion of 0.25 per cent, 
does not affect the erythrocytes or leukocytes and does not increase the per- 
centage of post transfusion chills. 
Other anticoagulants have been employed, especially hirudin by Satterlee 
and Hooker,3 and others, but this substance is apt to be more toxic, due, 
according to Marshall,4 to products of putrefaction. Potassium oxalate, so 
commonly employed for the prevention of coagulation of blood for chemical 
1093 
1 Proc. Soc. Exper. Biol, and Med., 1922, 19, 344. 
2 Jour. Amer. Med. Assoc., 1923, 80, 247. 
3 Jour. Amer. Med. Assoc., 1914, 62, 1781; ibid., 1916, 66, 618. 
4 Jour. Pharmacol, and Exper. Therap., 1915, 7, 157. 1094 
BLOOD TRANSFUSION 
analysis, is too toxic, and should never be used. In my experiments it was 
found that the maximum tolerated dose of this substance for white rats by 
intravenous injection was about 0.025 gm. per kilo as compared to 0.200 gm. 
of sodium citrate. It is, therefore, six to ten times more toxic than sodium 
citrate, although slightly more anticoagulating. It is generally necessary 
to use 0.2 per cent, potassium oxalate for the prevention of coagulation and 
0.2 to 0.3 per cent, sodium citrate, but potassium oxalate is so much more 
toxic and dangerous than sodium citrate that mention is made of it only as 
a matter of precaution against its use. 
4. The method of defibrination is not now generally employed for large 
transfusions, although it is very useful for the intramuscular injection of 
small amounts of blood for the control of hemorrhage. 
Defibrination is usually conducted by whipping the blood with sterile 
glass rods or collecting about 100 c.c. in each of a series of flasks and shaking 
with sterile glass beads. It is always advisable to filter the blood before 
intravenous injection. 
The method has the advantage of simplicity, although it is cumbersome 
and apt to be messy. Reactions may follow, and many investigators, 
including O’Connor,1 Brodie,2 Sollmann,3 and Stewart and Harvey,4 have 
shown that pressor and other toxic substances develop after defibrination 
and clotting. Hiirter5 states that the therapeutic effects of whole blood are 
distinctly better than those following the use of defibrinated blood. 
SELECTION OF DONORS 
Donors should be preferably large, strong, muscular men with prominent 
veins at the elbow. Nervous, apprehensive, and obese individuals should 
not be used, and especially in syringe-cannula methods for the transfusion 
of unmodified blood, as they may be unable to keep quiet at critical times. 
If a professional donor is employed, due care must be exercised against 
employing one with anemia due to too frequent transfusions. If doubt 
exists regarding the state of the blood, the erythrocytes should be counted 
and hemoglobin estimated. In transfusion for hemophilia, if a relative is 
employed, extra care must be exercised in the choice of the donor. 
In case of professional donors it is well or have them sign a legal form 
relieving patient and physician of further responsibility, and stating the 
stipend received. 
The donor should be free of tuberculosis, syphilis, or other communicable 
disease; there should be no evidences of cardiac disease. If time permits 
the Wassermann reaction should be conducted. Sydenstriker, Mason, and 
Rivers have recorded the probable transmission to patients of syphilis and 
malaria by blood transfusion. 
During the transfusion the donor should be watched for sweating, yawn- 
ing, and pallor, as collapse may follow. In the citrate method very little 
injury is done the vein, usually nothing more than a needle puncture, and 
the donor may resume his usual activities the following day. If the vein 
is exposed by incision, a week or more may be required for healing. 
For infants, the mother or father, or practically any healthy adult may 
be used as a donor. 
1 Arch. f. exper. Path. u. Pharmakol., 1912, 67, 195. 
2 Jour. Physiol., 1900, 26, 48. 
3 Amer. Jour. Physiol., 1905, 13, 291. 
4 Jour. Exper. Med., 1912, 16, 103. 
5 Med. Klinik, 1911, No. 12. PRELIMINARY COMPATIBILITY BLOOD TESTS 
1095 
As previously stated, the blood of a proposed donor may contain aggluti- 
nins and hemolysins for the corpuscles of the patient, and if his blood were 
tranfused, dangerous reactions would very probably follow the operation. 
Or these substances may not be in the blood of the donor, but in the blood 
of the patient for the corpuscles of the donor, in which case the donor would 
be unsatisfactory for the transfusion. 
In former years severe reactions were due to the transfusion of such 
incompatible bloods; even today some physicians minimize the danger and 
go ahead with transfusion regardless of the presence or absence of these 
substances. In grave emergencies this is justifiable when time for the 
matching tests cannot be spared because the good that may be done by 
prompt transfusion under the circumstances offsets the possible harm. 
But the tests require very little time and should always be done if at all 
possible. Surgeons of wide experience, like Bernheim and Dorrance, who 
at first ignored these tests and who gave many transfusions without untoward 
results, now regard them as absolutely essential after having had some 
unfortunate experiences. 
Many observers have stated that agglutination tests alone are sufficient, 
and that if these substances are absent in the blood of the donor for the 
corpuscles of the patient and in the blood of the patient for the corpuscles 
of the donor, the bloods may be regarded as compatible, because hemolysins 
are not present if agglutinins are absent. In a general way this is true 
and undoubtedly the agglutination tests are more important than the 
hemolysin tests and may be relied upon alone; but, as shown by Ottenberg 
and Kaliski,1 hemolysins may develop in disease. Losee states that in the 
treatment of hemorrhage due to injury, the patient being otherwise healthy, 
agglutination tests alone are sufficient, but that in disease both agglutination 
and hemolysin tests should be conducted. As a matter of fact, both tests 
can be done together in about the same time required for the agglutination 
tests alone, and when the patient is suffering with pernicious anemia or 
carcinoma, the hemolysin tests should always be included. According to 
Levine and Segall2 prolonged etherization may cause a temporary change in 
the blood grouping. 
The subject of hemagglutinins is discussed in Chapter XVI, and methods 
described for conducting these preliminary agglutinin and hemolysin tests. 
These tests should be conducted with great, care and both Dykes3 and 
Ottenberg4 have recently emphasized the importance of using only active 
Group II and III serums, since error may be due to the use of old and weak 
serums; Dyke recommends using only such serums as are able to cause defi- 
nite agglutination when diluted 1 : 10. 
As is now well known, the blood of human beings is divisible into four 
groups on the basis of agglutination tests, and it is easily possible to determine 
to which group any individual belongs. After childhood the group does 
not change; if an adult is found to belong to Group I his corpuscles will 
always remain in that group throughout life. 
Unfortunately, two methods of grouping are now in vogue, namely, that 
of Jansky and that of Moss. Each recognizes four groups; Jansky’s Group I 
corresponds to Moss’s Group IV and Jansky’s Group IV to Moss’s Group I. 
Groups II and III correspond in both. It is essential that the method is 
PRELIMINARY COMPATIBILITY BLOOD TESTS 
1 Jour. Amer. Med. Assoc., 1913, 61, 2138. 
2 Surg., Gyn., and Obstet., 1922, 35, 313. 
3 Brit. Jour. Exper. Path., 1922, 3, 146. 
4 Jour. Amer. Med. Assoc., 1922, 79, 2137. 1096 
BLOOD TRANSFUSION 
known and recorded when an individual’s blood is grouped. For example, a 
patient may be found belonging to Group I by the Moss classification; the 
donor must be Group I by the same classification, as a Group I donor by 
the Jansky method would be incompatible. If the grouping of both patient 
and donor is done in the same laboratory, confusion is not apt to occur; but 
if the patient is grouped and a donor selected on the basis of a grouping done 
elsewhere, great care must be exercised against the possibility of error. 
It is generally regarded as sufficient to select a donor belonging to the 
same group as the patient. But if time permits il is always well to directly 
match the bloods of patient and donor before transfusion by mixing the serum 
of the patient with the corpuscles of the donor and the serum of the donor 
with the corpuscles of the patient. The technic is described in Chapter 
XVI. Thalhimer1 and others have emphasized the necessity of conduct- 
ing these direct tests, and especially when the patient has been transfused 
on a former occasion. Mention has already been made of observations by 
McClure and Dunn2 and others, indicating that in pernicious anemia re- 
peated transfusions sometimes result in more and more difficulty in finding 
a perfectly compatible donor. While there is no evidence to show that 
injections of human corpuscles into human beings lead to the formation of 
agglutinins and hemolysins, yet this has been found to occur by Robertson 
and Rous3 in the lower animals, and changes may be brought about in 
human beings, although their nature is not at all understood. Furthermore, 
subgroups to the main groups of hemagglutinins are now known to occur, and 
these may be delected and minor incompatibilities prevented by the direct tests 
in addition to the group tests. Guthrie and Huck4 have recently studied this 
important subject of subgroups with particular care and have presented 
evidence that the popular belief in the existence of four and only four iso- 
agglutinin groups is incorrect. By direct tests and by absorption experi- 
ments they believe that they have demonstrated the existence of three 
instead of two iso-agglutinins and three instead of two iso-agglutinogens, 
with these three iso-agglutinins and three iso-agglutinogens there are pos- 
sible at least 27 biologic combinations; 8 of which they believe to have 
observed, instead of but the four groups or combinations recognized at 
present in the Jansky and Moss classifications. For this reason some of the 
post-transfusion reactions occurring after the injection of bloods of the 
same groups may be due to the presence of undetected agglutinins; more 
than ordinary care is required in the blood-matching tests, and Guthrie 
and Huck have suggested that the grouping should be determined on a basis 
of the agglutinin-agglutinogen content, since only bloods with the same 
agglutinin-agglutinogen content belong to the same group. 
Individuals belonging to Group I (Jansky) or Group IV (Moss) are 
commonly regarded as universal donors suitable in case of emergency for 
the transfusion of anyone. Under these conditions reliance is based upon 
the fact that the high dilution to which the incompatible plasma is brought 
may nullify its toxic effect, but during the World War the Interallied Surgi- 
cal Conference5 stated that fatal accidents have followed, although the 
chances of eliciting these were slight, and that such donors may be used in 
emergencies. 
1 Jour. Amer. Med. Assoc., 1921, 76, 1345. 
2 Bull. Johns Hopkins Hosp., 1917, 28, 99. 
3 Jour. Exper. Med., 1922, 35, 141. 
4 Bull. Johns Hopkins Hosp., 1923, 34, 125. 
5 Presse med., 1918, 26, 193. FATE OF THE TRANSFUSED BLOOD 
1097 
PREPARATION OF THE PATIENT AND DONOR 
No special preparation is required. Losee recommends that the patient 
be given 900 c.c. of milk in divided amounts twenty-four hours before trans- 
fusion in order that the heart may become accustomed to increased volume 
of fluid. It is well that the bowels of both patient and donor be empty and 
likewise the bladder of each, in view of the nervousness of both during the 
operation. A nervous patient may require | gram of morphin about one- 
half hour before operation. 
The donor should not have eaten within two hours of operation; I believe 
that reactions are sometimes caused by the transfusion of chylous plasma. 
It is not necessary to prepare the arms beforehand, and every effort should 
be made to prevent undue confusion and fuss; it is easily possible to protect 
the patient and donor against nervous strain and fear. 
POSTTRANSFUSION TREATMENT 
During the transfusion both patient and donor should be watched closely, 
especially the patient during the injection of the first 100 c.c. of blood, and 
the donor after 500 c.c. of blood have been removed. 
If the patient shows signs of distress, as cyanosis, dyspnea, slow pulse, 
clammy skin, etc., as previously described, the transfusion should be inter- 
rupted and 1 c.c. of 1 : 1000 adrenalin and 1/100 grain of atropin sulphate 
administered subcutaneously. If these effects pass off, the transfusion may 
be carefully and slowly resumed. If they do not pass away or recur, the 
transfusion should be stopped, as the donor is very probably incompatible. 
After transfusion the patient usually develops fever and perspires; this 
does not require special treatment. If a severer reaction occurs, the treat- 
ment is symptomatic. Urticaria may be relieved by local applications or 
the subcutaneous injection of adrenalin, as described in Chapter XXXII for 
the treatment of serum sickness. 
The donor usually feels little or no effects if 500 c.c. or less blood are 
removed. He should be carefully watched, however, for yawning, pallor, 
nausea, cold clammy skin, and a weak pulse. These effects are usually 
psychic, and a kind, sympathetic attitude on the part of the operator with 
words of encouragement, are usually all that is required for their prevention. 
When larger amounts of blood are removed, the donor may feel faint; and 
if so, removal of blood should be stopped at least for a few minutes and 
aromatic spirits of ammonia or whisky offered. It is well for the patient 
to rest for a few hours after removal of blood, although the majority refuse 
to do so. 
FATE OF THE TRANSFUSED BLOOD 
As previously stated, the effects of transfusion on the patient are due to 
the mechanical presence of so much new blood, to stimulation of the patient’s 
blood-making organs, and a redistribution of blood through the body. As 
shown by Ashby,1 the length of time the transfused blood remains in circu- 
lation varies greatly and elimination varies greatly and takes place in cyclic 
crises rather than as a continuous process. 
Rous and Robertson2 have shown that the transfused corpuscles are 
gradually broken up and the fragments filtered out in the blood-stream. 
Doubtless some of the cells are phagocytosed by fixed and wandering cells 
of the patient; this undoubtedly occurs when incompatible bloods are trans- 
fused, as shown by Ottenberg and others. 
1 Jour. Exper. Med., 1921, 34, 127. 
2 Jour. Exper. Med., 1917, 25, 651. 1098 
BLOOD TRANSFUSION 
Just how long the transfused corpuscles continue to function in their 
new host is not possible to state. In pernicious anemia and other hemolytic 
diseases their duration is probably much shorter than in the blood of a 
healthy host treated for hemorrhage. Even before they are removed from 
the circulation the cells may lose in hemoglobin content and be functionally 
inactive. But if there is no immediate break down of the donated corpus- 
cles, they may be demonstrated in the blood for more than thirty days, and 
according to the results obtained by Ashby employing an agglutination 
method, for as long as three months even in pernicious anemia. This leads 
Ashby to conclude that the primary beneficial effects of transfusion are not 
due to a stimulating effect upon the bone-marrow, but more probably to the 
functioning of the transfused red blood-corpuscles. These results have re- 
cently been confirmed by Wearn, Warren, and Ames,1 who found that red 
blood-corpuscles from donors in Group IV transfused into patients in Group 
II with pernicious anemia and anemia secondary to nephritis, remained in the 
circulation longer than has been generally believed to be the case. The last 
of the transfused red blood-cells disappeared from the circulation in from 
fifty-nine to one hundred and thirteen days, with an average of eighty-three 
days. No difference was noted in a series of observations in the duration 
of the stay of the transfused red blood-corpuscles in the circulation between 
patients with primary anemia and. secondary anemia (due to nephritis). 
In a single observation red blood-corpuscles from a patient with pernicious 
anemia transfused into another patient with pernicious anemia, behaved 
as did the corpuscles from normal donors. 
TRANSFUSION WITH PRESERVED RED BLOOD-CORPUSCLES 
Rous and Turner2 have shown that the red blood-corpuscles of rabbits 
may be kept in mixtures (of 3 parts of blood in 2 parts of 3.8 per cent, sodium 
citrate in water with 5 parts of 5.4 per cent, saccharose in water) for fourteen 
days; the mixtures were kept in a refrigerator, and before transfusion the 
supernatant fluid was pipeted off and the corpuscles suspended in fresh 
sterile Locke’s solution. When used to replace normal blood the transfused 
corpuscles remained in circulation and functioned as well as fresh corpuscles. 
Cells kept for longer periods, though intact and apparently unchanged when 
transfused, soon left the circulation, although without producing harm. 
Robertson3 used this method for the preservation of human corpuscles 
and transfused 20 individuals suffering with hemorrhage, shock, and sepsis, 
chiefly hemorrhage and shock. Considering the type of cases given these 
transfusions, the results were good and equal in Robertson’s opinion to 
transfusions of fresh blood. Four weeks represented the limit to which hu- 
man blood could be kept in the above solution in a refrigerator. The ad- 
vantages were the simplicity of technic, having blood on hand at the minute 
it was needed, and that the transfusion could be given without assistance. 
A large number of methods for blood transfusion have been described. 
Some are quite complicated and others simple, doubtless all have proved 
satisfactory in the hands of their inventors. 
(A) Suture and Cannula Methods for Direct Transfusion.—The Carrell 
suture, Crile and Elsberg cannula methods are difficult surgical operations. 
Pope4 has described a simple cannula method composed of two glass tips 
METHODS FOR BLOOD TRANSFUSION 
1 Arch. Int. Med., 1922, 29, 527. 
* Jour. Exper. Med., 1916, 23, 239. 
3 Brit. Med. Jour., 1918, 1, 691. 
4 Jour. Amer. Med. Assoc., 1913, 60, 1284. TECHNIC OF BLOOD TRANSFUSION 
1099 
connected with a short piece of rubber tubing, the whole being boiled in 
equal parts of petrolatum and paraffin before use for sterility and coating. 
A similar apparatus has been more recently described by Fullerton, Dreyer, 
and Bazett1 for direct transfusion. It is necessary in these methods to cut 
down and expose the vessels of both donor and patient in order to insert and 
tie in the cannulas. 
(B) Syringe and Syringe Cannula Methods for Direct Transfusion.— 
The oldest syringe method is that of Ziemssen; the Lindemann2 syringe- 
cannula method is practically a refinement of this method and is described 
later on. Abelmann,3 Brown,4 and Lintz5 have described special syringe 
methods; and Unger8 has likewise devised an apparatus for drawing blood 
from the donor and at once injecting it into the patient by the operation of 
a system of stop-cocks. 
Cooley and Vaughan7 and Howard8 have described very simple syringe 
methods for transfusing children with whole or citrated blood, the technic 
being described later on. 
(C) Paraffined Tube Methods for Direct Transfusion.—The method of 
Kimpton and Brown9 is best known and yields excellent results; it will be 
described later under Technic. Percy10 and Kreuscher11 have also described 
similar methods. 
(D) Citrate Methods.—Lewisohn12 has described a very simple method, 
but it is open to the objection of the blood being subject to contamination. 
The apparatus devised independently by Hoffmann and Habein13 and Robert- 
son14 are closed and remove this objection. My own apparatus, described 
later, is almost identical with these. Hartman15 has also described a special 
jar for transfusion of citrated blood. 
TECHNIC OF BLOOD TRANSFUSION 
Since in the practice of blood transfusion some cases may be satisfactorily 
transfused with citrated blood, while others should receive unmodified blood, 
two methods for each with which the writer has had best results, are described. 
Lewisohn’s Citrate Method.—Obtaining the Blood from the Donor.—A 
tourniquet is applied to the donor’s arm, and one of the larger veins in the 
elbow region (usually the median cephalic vein) is punctured. A cannula 
of large diameter (13 gage) is used in order that the blood may flow out 
rapidly through the needle. For guiding the flow of blood a Luer adapter 
with short rubber tubing is inserted in the needle hub. The blood is col- 
lected in a graduated glass jar which contains at its bottom the 2.5 per cent, 
citrate solution. If 450 c.c. of blood is to be given, add 50 c.c. of this solu- 
tion, thus affecting a 0.25 per cent, mixture. Smaller and larger amounts 
of blood are treated proportionately. This technic is illustrated in Fig. 194. 
Infusion of the Blood Into the Recipient.—The recipient is usually so 
anemic that the vein must be exposed by a small incision in about 80 per 
cent, of the cases. A cannula 15 gage is inserted, and the latter is attached 
to the gravity tube which contains from 20 to 30 c.c. of physiologic sodium 
1 Lancet, 1917, 1, 715. 5 Jour. Amer. Med. Assoc., 1916, 66, 506. 
2 Amer. Jour. Dis. Child., 1913, 6, 28. 6 Jour. Amer. Med. Assoc., 1915, 65, 1027. 
3 Jour. Amer. Med. Assoc., 1916, 67, 142. 7 Jour. Amer. Med. Assoc., 1913, 60, 435. 
4 Jour. Exper. Med., 1918, 28, 623. 8 Jour. Amer. Med. Assoc., 1915, 65, 1365. 
9 Jour. Amer. Med. Assoc., 1913, 61, 171; Boston Med. and Surg. Jour., 1918, 178, 351. 
10 Surg., Gyn., and Obstet., 1915, 65, 269. 
11 Jour. Amer. Med. Assoc., 1918, 70, 223. 
12 New York State Jour. Med., 1919, 19, 431. 
11 Jour. Amer. Med. Assoc., 1921, 76, 358. 
14 Brit. Med. Jour., 1918, 1, 477. 15 Jour. Amer. Med. Assoc., 1918, 71, 1658. 1100 
BLOOD TRANSFUSION 
chlorid solution. The blood is then poured into this apparatus and allowed 
to run into the punctured vein by gravitation (exactly like an ordinary saline 
solution). In newborn infants a Kaliski cannula, 18 gage represents the 
proper size (Fig. 195). 
The apparatus required for this method may be purchased as a complete 
outfit, but can be improvised with perfectly satisfactory results. A dis- 
advantage is that there are chances for air contamination of the blood and 
especially during its collection. For that reason the following method was 
devised: 
Author’s Citrate Method.—As previously stated, this method is fashioned 
after that of Robertson and Hoffmann and Habein. 
The apparatus is shown in Fig. 196 and serves both for the collection of 
blood and its injection under aseptic conditions. One operator can do the 
work, but it is better done with an assistant. 
Fig. 194.—Taking Blood from the Donor. (Lewisohn’s Method.) 
The apparatus is readily assembled and consists of a 1000 c.c. filter 
flask graduated into divisions of 100 c.c. each and fitted with a two-holed 
rubber stopper, glass tubes, rubber tubing, gage No. 18 needles, and a 
Potain aspirating syringe. The assemble and dimensions of all parts are 
shown in Fig. 196 and need not be further described. Tubes B and C with 
attached needles are enclosed in test-tubes, and the whole sterilized by 
autoclaving. The 50 c.c. graduated cylinder (H) carries a sterile 5 per cent, 
solution of sodium citrate in water. This is added to the blood in the proportion 
of 5 c.c. per 100 c.c. of blood, giving a concentration of 0.25 per cent, citrate. 
An advantage of the apparatus is that citrate is added under aseptic conditions, 
as required, in order to avoid an excess. Even with large transfusions of 
900 c.c. the total amount of citrate would be 45 c.c. of the 5 per cent, solution 
or 2.25 gm., which is well within the limits of safety. 
The technic is as follows: 
1. A tourniquet is applied around the arm of the donor, and he is re- 
quested to vigorously open and close the hand for the purpose of rendering TECHNIC OF BLOOD TRANSFUSION 
1101 
Fig. 195.—Infusion of Citrated Blood Into the Recipient. (Lewisohn’s Method.) 
Fig. 196.—Apparatus Employed by the Author eor the Collection, Citrating, and Infusion 
of Blood. 
A, A 1000 c.c. filter flask; B, tubing and No. 18 needle for collection of blood (enclosed in a 
test-tube for sterlization); D, tubing with filter (I) and syringe (G) for exhausting air during collection 
and pumping air for the infusion; C, tubing with needle (No. 18) for citrating blood (H) during collec- 
tion and later for infusion of the mixture into the patient; E and F are stop-cocks. 1102 
BLOOD TRANSFUSION 
the superficial veins full and tense. The skin is cleansed and one or two 
coats of tincture of iodin applied. 
2. Tube C with its needle are removed from the test-tube and the end 
placed in the cylinder of 5 per cent, citrate solution. Stop-cock E is left 
open, but F is closed. The syringe G is applied to tube D and suction made, 
drawing 10 c.c. of the citrate solution into the flask, which is sufficient for 
the first 200 c.c. of blood. 
Stop-cock E is now closed, but during the collection of blood it is opened, 
as required, and measured amounts of the citrate solution permitted to enter 
the flask. 
3. Tube B with its needle is removed from the test-tube and the needle 
inserted into a prominent vein of the donor. Stop-cock F is open, but E is 
Fig. 197.—Collection and Citration of Blood. 
closed. Suction is made with the pump to keep up a negative pressure in 
the flask, but continuous pumping is not required, leaving the hands free to 
gently shake the flask from time to time to mix the blood and citrate solution. 
Blood flows in a steady stream (Fig. 197). As the 200 c.c. mark is reached, 
stop-cock E is opened and 5 c.c. of citrate solution permitted to enter at the 
same time, which is sufficient for 100 c.c. more of blood. In this manner 
citrate is gradually added beforehand as required. 
4. The arm of the patient is prepared in the same manner. Stop-cock F 
is closed throughout. The syringe G is reversed to pump air into the flask, 
which is filtered through I. Stop-cock E is opened and the pump worked 
until tube C and its needle are filled with blood. Cock E is then closed 
until the needle has been entered into the vein, when it is opened, the tourni- TECHNIC OF BLOOD TRANSFUSION 
1103 
quet released, and the blood gently pumped in (Fig. 198). Care must be 
exercised against too vigorous pumping, and it is advisable to wire the 
stopper into the flask to avoid loosening during the infusion. As the flask 
is about emptied, due care must be exercised against admission of air by closing- 
cock E and removing the needle from the vein. 
Some of the advantages of this apparatus are as follows: (a) The 
negative pressure in the flask during the collection of blood prevents coagula- 
tion in the needle and tubing until the required amount of blood is obtained. 
This also permits the use of a smaller needle than ordinarily employed. 
(5) The blood is collected and citrate added under closed and aseptic con- 
ditions. (c) Citrate is added as required in order to avoid an excess, as 
sometimes happens if enough is placed in a flask for 500 c.c. of blood, but 
only 200 to 300 c.c. of blood actually obtained, (d) It provides a uniform 
Fig. 198.—Infusion of the Citrated Blood. 
steady flow of blood. (e) With a little practice the whole operation may 
be done by one person. 
Kimpton-Brown Method.—This is a satisfactory method for the transfu- 
sion of unmodified blood and especially for adults. It depends upon the 
principle of using a coating of paraffin to prevent contact of blood with glass 
and thereby preventing coagulation for a sufficient time to permit transfu- 
sion. I have applied this method as follows: 
1. The tube is shown in Fig. 199 and usually comes in two sizes, 250 c.c. 
and 150 c.c., although larger or smaller ones may be prepared. I have found 
it convenient to slightly enlarge the tip and attach about f-inch of rubber 
tubing, carrying an adapter for the Record syringe needles. 
2. The tube is wrapped in a towel and autoclaved. The paraffin (melt- 
ing-point 54° C.) or better, Vincent’s mixture of 1 part stearin, 2 parts BLOOD TRANSFUSION 
1104 
of paraffin, and 2 parts vaselin, is sterilized and melted by boiling in a 
water-bath or autoclave. 
The tube is carefully and moderately heated over two alcohol lamps. 
The stopper is removed and the tube half-filled with the hot fluid paraffin 
mixture, which is rolled around inside the tube and over the surface of the 
cork, and a little permitted to run out the cannula’s end (needle detached). 
The tube is now emptied through the side tube, leaving a thin but complete 
coat of paraffin mixture inside. The tube is now cooled by sponging with 
alcohol, and is ready for use. Two or three such tubes should be in readiness, 
and may be prepared beforehand and kept sterile wrapped up in sterile 
towels. 
3. It is worth while to attach a piece of rubber tubing to the side arm and 
carrying a glass window filled with cotton for a filter. This tubing is 
sterilized by autoclaving. During the collection of blood the operator can 
apply suction if necessary to aid filling by negative pressure. During the 
This shows a 100 c.c. size with tubing (C) and mouth-piece attached with air filter (D) for suction 
during collection of blood if necessary. A short piece of tubing (A) has been attached in such manner 
that the glass tip of the tube almost touches the adapter B fitting a 1-inch No. 16 Record needle. 
Fig. 199.—Kimpton-Brown Tube. 
injection of blood a bulb syringe may be attached and filtered air pumped 
into the tube to facilitate emptying. 
4. Donor and patient should be lying on adjoining tables or beds, and 
their arms prepared. 
A tourniquet is applied to the arm of the donor and a vein at the elbow 
rendered full and tense by opening and closing the hand. A Record needle 
(gage No. 16, 1 inch in length) is inserted into a vein, and the adapter end 
of the tube attached. The tube fills with blood by venous pressure, care 
being taken not to have the tourniquet too tight or too loose. A blood- 
pressure cuff is best for keeping the blood-pressure at about 60 to 65. Filling 
may be facilitated as necessary by suction on the rubber tubing attached to 
the side tube. If a needle is not employed, the skin over the vein should 
be infiltrated with sterile novocain, an incision made, the vein exposed, and 
the usual ligatures applied. The vein is opened and the cannula end of the 
tube inserted for about § to 1 inch, and held upright until it fills. I prefer TECHNIC OF BLOOD TRANSFUSION 
1105 
the needle arrangement as obviating the necessity for cutting in most in- 
stances and facilitating the filling of two or three tubes with least loss of 
blood. 
5. The tube is now detached from the needle and a second tube imme- 
diately attached and filled. The first tube may be handed to an assistant, 
who proceeds with the injection, or it is held in such a way as to prevent loss 
of blood until the second and third tubes have been filled, when the tourni- 
quet is released and the needle withdrawn. 
6. The tourniquet is now applied to the patient, and a vein at the elbow 
rendered full and tense if this is possible. If the anemia does not permit 
this, it may be necessary to cut down on the vein under local anesthesia, 
but this should be determined beforehand and done if necessary, so that 
there may be no delay in injecting the blood after the tube or tubes have 
been filled. I have used Lindermann’s cannula with excellent results for 
insertion in the vein of the patient, although a plain needle of the above- 
mentioned dimensions is satisfactory for the donor. 
Fig. 200.—Injection of Blood with the Kihpton-Brown Tube. 
If a vein is full and sufficiently tense a No. 16 needle (1 inch length) or 
Lindemann cannula is inserted direct, the tourniquet released, a tube at- 
tached, and the blood infused, the flow being aided by gentle pumping with 
a cautery bulb or some other suitable bulb (Fig. 200). After one is nearly 
emptied, the succeeding ones are attached, the needle being left in place 
until the infusion is completed. 
The Lindemann Syringe Cannula Method.—This method is for the 
transfusion of unmodified blood. Its special feature is the needle-cannula 
(Fig. 201) by means of which cannulas are inserted into the veins of donor 
and patient without exposing the veins. In addition to two of these cannu- 
las, three or four 25 c.c. Record syringes are required, and all sterilized by 
boiling for a few minutes just before use. 
1. One operator draws the blood, a second injects, and a third cleanses 
the syringes. Patient and donor should be on adjoining tables, with an 1106 
BLOOD TRANSFUSION 
instrument table between upon which the outstretched arms are placed; 
the third assistant stands between with a basin of warm sterile saline solu- 
tion, washing the syringes and handing them forward and backward between 
the operators (Fig. 202). 
Fig. 201.—The Lindemann Cannula. 
2. The arm of the donor is prepared and a tourniquet applied. A cannula 
is gently inserted, the trocar (A) being sharp, until cannulas B and C are 
in the vein for at least a length of f to 1 inch. Trocar A and cannula B are 
Fig. 202.—Blood Transfusion with Lindemann Cannulas and Syringes. 
removed, leaving cannula C in place. A 25 c.c. Record syringe filled with 
warm sterile saline solution is attached, and the assistant maintains a very 
slow injection to prevent escape of blood and coagulation. 
3. The second cannula is inserted into the vein of the patient, the tourni- TECHNIC OF BLOOD TRANSFUSION 
1107 
quet released, and a syringe containing warm saline attached to prevent 
loss of blood and coagulation. 
4. The first operator now fills his syringe with blood, detaches, and 
immediately connects a second syringe. The second operator detaches his 
syringe carrying saline, attaches the syringe filled with blood, and injects. 
The third assistant washes out the syringes and hands them to the first 
operator, etc., until the required amount of blood has been transfused. 
I have sometimes used a sterile 5 per cent, solution of sodium citrate in 
water for washing the syringes instead of plain saline solution; this tends to 
prevent coagulation in the syringes, and the very small amounts of citrate 
injected do not materially modify the method. 
In adults and most children over two years of age the median basilic vein 
may be employed. 
Transfusion of Infants.—The Lindemann method may be employed, the 
cannula being inserted into the external jugular or internal saphenous veins. 
Not more than 50 to 100 c.c. of blood are usually required. According to 
Halbertsma1 an injection of 15 c.c. of blood per kilogram of body weight 
produces a rise of 1,000,000 corpuscles and that this affords a means for cal- 
culating beforehand the amount of blood to be transfused. 
A citrate syringe method may be employed as follows: 
1. Three or four 25 c.c. Record syringes are sterilized and 2 c.c. of a 
sterile 5 per cent, solution of sodium citrate in water placed in each. 
2. The arm of the donor is prepared and a tourniquet applied. A needle 
(gage No. 18 is usually large enough) is inserted into a vein and the syringe 
almost filled with blood. A second syringe is quickly attached and filled, 
followed by a third and fourth if required. The tourniquet is now released 
and the needle removed. This method is similar to that described by 
Howard.2 
3. Coagulation is prevented if the citrate and blood in each syringe are 
mixed. 
4. Or the blood may be citrated by placing 5 c.c. of 5 per cent, citrate 
solution in a sterile bottle or small Erlenmeyer flask and emptying the 
syringes as they are filled. This amount of citrate is sufficient for 100 c.c. 
of blood. If less blood is required, 1 c.c. of the citrate solution is sufficient 
for each 20 c.c. of blood. As a general rule, there should be three syringes, 
so that an assistant may empty a syringe, wash it out once or twice by 
drawing up and expelling warm citrate solution, and place it beside the 
operator, who fills and detaches without undue haste. Or the method 
described by Zingher3 may be employed. 
5. The injections are now given through the superior longitudinal sinus 
or external jugular vein, a smaller needle being employed. The technic is 
described in Chapter II and illustrated in Figs. 21 and 23. 
6. Recently Siperstein and Sansby4 have advocated intraperitoneal 
injections of citrated blood in the transfusion of infants. In experiments 
employing rabbits they have found that blood is rapidly absorbed from 
the peritoneal cavity and that intraperitoneal transfusion of freshly citrated 
blood acts as a true transfusion and not merely as the absorption of a 
nutrient material. Siperstein5 has subsequently reported upon the treat- 
ment of 5 cases, favorable results being obtained in 3. This route of 
1 Amer. Jour. Dis. Child., 1922, 24, 269. 
2 Jour. Amer. Med. Assoc., 1915, 65, 1365. 
3 Medical Record, March, 1915, 13. 
4 Amer. Jour. Dis. Child., 1923, 25, 107. 
5 Ibid., 1923, 25, 208. 1108 
BLOOD TRANSFUSION 
transfusion is simple, practical, and efficient according to these investigators, 
and while it is not the route of choice, it is suggested as a therapeutic pro- 
cedure of merit when the anterior fontanel is closed and surgical exposure 
of a vein is difficult. 
Intramuscular Injections of Blood.—Several methods may be employed; 
all are quite simple and efficient. 
1. According as 20 or 40 c.c. of blood are required, one or two 25 c.c. 
Record syringes are sterilized, 2 c.c. of 5 per cent, solution of citrate placed 
in each, the syringe or syringes filled with blood from the donor, the contents 
mixed, and the intramuscular injections given as described in Chapter 
XXXVII and illustrated in Fig. 181. 
2. Or the required amount of citrate solution (2 c.c. of a 5 per cent, 
solution for each 20 to 25 c.c. of blood) is placed in a sterile bottle, blood 
aspirated from a vein of the donor, and expelled into the bottle. As a general 
rule only one syringe is required, as it may be easily filled twice without 
washing between fillings, a finger closing the end of the needle to prevent 
bleeding while the syringe is being emptied. After the blood is secured the 
tourniquet is released, the syringe washed out once or twice with saline 
solution, and the injection given. 
3. Or one, two, or more syringefuls of blood may be collected and expelled 
into a sterile bottle or flask containing glass beads, shaken for a few minutes 
for defibrination, and the blood injected intramuscularly. 
4. Or, if the patient is placed close at hand and only 20 to 25 c.c. of blood 
are required, a syringeful may be collected, using a gage No. 16 or 18 needle 
to facilitate quick filling, the tourniquet released, needle and syringe with- 
drawn, and the blood injected at once (but not too rapidly) before it has had 
time to coagulate in the syringe. Part V 
EXPERIMENTAL INFECTION AND IMMUNITY 
INTRODUCTORY 
Methods.—The exercises and experiments herein outlined are for the 
purpose of teaching the principles of infection and immunity by actual 
laboratory work, whereby the student performs the experiments and is 
taught to observe the results. At the same time a knowledge of the technic 
of immunologic methods is obtained. The instructor in charge of such an 
experimental course may choose certain experiments in outlining a course 
according to the allotment of time and purpose of the instruction. 
In all instances I have outlined the experiments according to average 
conditions. It is readily understood that differences in the virulence of a 
certain culture or the weight and physicial condition of an experimental 
animal will require that the doses advised be changed to meet the conditions. 
As a general rule, a course should be concentrated, with exercises at 
least three times a week in order that the student may follow his work 
closely and have an opportunity for making adequate and accurate ob- 
servations. 
An attempt is made to bring out the important points of an experiment 
by a few pertinent questions. It should be impressed upon the student 
that the mind should be held open for observation and that unexpected and 
untoward results may be obtained which, however, are always of interest 
and always instructive when the experiment is conducted in a careful, 
methodical, and conscientious manner. Frequent general discussions should 
be held for a general review of the subject and correlation of facts and 
observations. In my experience students are eager for the work and seldom 
fail to suggest additional work in the nature of original research. 
The Student.—1. The student should work protected by an apron or 
gown with short sleeves; he should be careful of the hands and avoid abra- 
sions and cuts, and carefully wash and disinfect the hands after the work of 
each day has been completed. 
2. The working table should be set in order after each day’s work; 
pipets and soiled glassware should be properly disposed of, instruments 
thoroughly cleansed, and the table wiped off with 1 per cent, formalin 
solution. It should be impressed upon the student that good and accurate 
work is seldom done amid disorderly and dirty surroundings. 
3. The student must never sacrifice accuracy for speed; painstaking 
and accurate work is always to be desired; speed is gained only with practice. 
Records.—Each worker should record in writing in a suitable note- 
book his observations of the various tests and experiments. Not infre- 
quently unexpected results are obtained, and it is important to understand 
and explain these as correctly as possible. Note-books should be subject 
to frequent inspection; it is not necessary to write out a description of the 
technic, but the results of the experiment and answers to the questions 
should be set forth clearly and with sufficient detail. 
Animal Experiments and Autopsies.—In all experiments calling for 
operative procedure an anesthetic is to be used in order that unnecessary 
pain be avoided. Ordinarily, ehter is to be employed, or in rabbits the 
1109 1110 
EXPERIMENTAL INFECTION AND IMMUNITY 
rectal injection of 0.5 to 1 gram of chloral hydrate dissolved in 5 to 10 c.c. 
of water. Autopsies are to be carefully conducted, and the lesions de- 
scribed in writing. After autopsy the table is to be scrubbed and cleansed 
with a solution of formalin, and the carcass disposed of by incineration or 
placing in a solution of formalin until removed and otherwise disposed of. 
ACTIVE IMMUNIZATION OF ANIMALS 
Experiment 1.—Production of Agglutinins, Bacteriolysins, and Op- 
sonins: 
1. Secure a culture of Bacillus typhosus, an old laboratory culture of cholera bacilli, 
and a culture of Staphylococcus aureus. 
2. Proceed to immunize two rabbits with each culture by intravenous injections accord- 
ing to the technic described in the text. 
3. One week after the last injection the animals are to be bled and the serums secured 
after the methods described in the text. 
(a) Define the meaning of antigen. 
(b) What are antibodies? 
(c) What is meant by active immunization? 
(d) What precautions should be observed in handling these antigens? 
(e) What precautions are to be observed in giving intravenous injec- 
tions? 
(/") What antibodies do you suspect are present in these immune serums? 
Experiment 2.—Production of Precipitins: 
1. Immunize two rabbits with sterile horse-serum by intravenous injections after the 
methods described in the text. 
2. Immunize two rabbits with sterile human serum. 
3. Bleed the animals from the carotid artery one week after the last injection and pre- 
serve the serums in a sterile condition. 
Experiment 3.—Production of Hemolysins: 
1. Immunize a rabbit with washed sheep corpuscles by intravenous injections after 
methods described in the text. 
2. Immunize a second rabbit with sheep cells. 
3. Immunize a third rabbit with washed human cells. 
4. Immunize a fourth rabbit with washed human cells by intraperitoneal injections. 
5. Bleed the animals in four days to a week after the last injection and separate the 
serums. 
6. Prepare a portion of human amboceptor dried on paper as described in the text. 
7. Preserve the balance of the serum and the other serums with equal parts of neutral 
glycerin. 
Experiment 4.—Production of Cytotoxin: 
1. Immunize two rabbits with dog kidney after the method described in the text. 
2. After the immunization has been completed, preserve the serum with 0.2 per cent, 
phenol in sterile ampules. 
While these various immune serums are being prepared the student is en- 
gaged with the work on injection, or if the subject of immunity is taken up at 
once, immune serums should be furnished by the instructor. 
EXPERIMENTAL INFECTION 
Experiment 5.—Experimental Pneumonia: 
1. Grow a virulent culture of pneumococcus in glucose serum broth for forty-eight to 
seVenty-two hours at 37° C. until a good rich growth is secured. Prepare and stain smears 
by Gram’s method. EXPERIMENTAL INFECTION 
1111 
2. Pass a large catheter which has its tip cut off into the trachea of a dog until it has 
passed into one of the primary bronchi. By means of a syringe inject quickly 15 c.c. of 
culture; remove the catheter and mouth-gag. Take the rectal temperature and leukocyte 
count previous to inoculating. 
3. Inject a rabbit with 5 c.c. of culture by tracheotomy under anesthesia. 
4. Inject a second rabbit with 5 c.c. of culture intravenously. 
5. Observe all animals for forty-eight hours, taking the rectal temperature night and 
morning. Make leukocyte counts every four hours during the day. Make physical ex- 
amination of the chest. 
6. Autopsy the animals under aseptic precautions and with complete anesthesia. 
7. Culture the heart’s blood of each in serum bouillon. 
8. Culture the pulmonary lesions in serum bouillon. 
9. Prepare smears of the heart’s blood and lesions and stain with methylene-blue and 
Gram’s stain. 
10. Remove consolidated portions of lung and place in 5 per cent, formalin. After 
twenty-four hours cut sections which are passed through by the paraffin method and stained 
with hematoxylin and eosin, methylene-blue, and Gram’s stain. 
(a) Did any of the animals show evidences of infection? 
(b) Are there evidences of pneumonia? How do these lesions compare 
with those of human pneumonia? 
(c) How do you explain their production? 
(d) Does the rabbit receiving the dose of pneumococci intravenously 
show evidences of pneumonia? If not, why not? 
(e) Were the temperature changes similar to those observed in human 
lobar pneumonia? 
(f) Did leukocytosis occur, and if so, why? 
(g) Are there any evidences of pleuritis, and if so, how do you explain 
its production? 
(h) Are the pneumococci seen in the smears of the lesions and blood 
encapsulated? If so, what is the significance of these capsules? Compare 
these cocci with those shown in the smear of the culture before injection? 
Are the capsules lost in the artificial culture-media? 
(i) Discuss the question of the possible modes of infection in human 
lobar pneumonia. 
Experiment 6.—Relation of Numbers of Microparasites to Infection: 
1. Secure a white rat infected with T. equiperdum and showing large numbers of the micro- 
parasites in tail blood examined microscopically. 
2. With a red corpuscle pipet, a diluting fluid (2 c.c. glacial acetic acid, 2 c.c. formalin, 
2 c.c. carbolfuchsin, and 96 c.c. of water), and a counting chamber, determine the number of 
parasites per cubic centimeter of blood. 
3. Place 10 c.c. of 4 per cent, sodium citrate in water in a 25 c.c. graduated cylinder. 
Etherize the rat and bleed from the vessels of the neck, collecting the blood by means of a 
small funnel into the citrate. Mix well and calculate amount of blood secured and approxi- 
mate number of trypanosomes per cubic centimeter. 
4. Prepare a dilution with warm saline solution in such manner that 1 c.c. contains 
100,000,000 trypanosomes; a second solution to contain 1,000,000 per cubic centimeter; a 
third to contain 100,000 per cubic centimeter; a fourth to contain 10,000 per cubic centimeter; 
a fifth to contain 1000 per cubic centimeter; and a sixth to contain 100 per cubic centimeter. 
Inject 1 c.c. of each intraperitoneally into 6 rats respectively of approximately equal weights. 
5. Examine 1 drop of blood from the tail of each animal eighteen and twenty-four hours 
later and daily thereafter. 
6. When the animals succumb, remove the brain, liver, and spleen of each; fix in formalin, 
prepare and stain sections. 
(a) Does the time of appearance of trypanosomes in the peripheral blood 
bear a relation to numerical infection? Does the duration of life? 
(b) Discuss the probable causes of death in these animals. EXPERIMENTAL INFECTION AND IMMUNITY 
1112 
BACTERIAL, PLANT, AND ANIMAL TOXINS 
Experiment 7.—Diphtheria Toxin and Toxoid: 
1. Secure diphtheria toxin or prepare by inoculating tubes of neutral or slightly alkaline 
bouillon with a virulent culture of diphtheria bacilli (Park-Williams Bacillus No. 8 being 
especially desirable). 
2. Grow at 35° C. for five days and filter the culture through a Berkefeld filter. 
3. Inject a 250- to 300-gram guinea-pig in the median abdominal line with 0.5 c.c. of the 
filtrate (toxin). 
4. Heat 1 c.c. of toxin at 60° C. for an hour and inject 0.5 c.c. into a second guinea-pig 
subcutaneously. 
5. Observe the animals for at least four days, especially for the development of a char- 
acteristic edema about the site of injection. 
6. After death perform a careful autopsy, paying particular attention to the bloody 
edema at the site of injection and marked hyperemia of the suprarenal glands. Make cul- 
tures on Loffler’s blood-serum medium of edematous area, peritoneum, and heart blood. 
(a) To what has death been due? 
(,b) Has diphtheria toxin a selective affinity for any particular tissue? 
(c) Has heat any effect upon diphtheria toxin? 
(d) Is there a period of incubation before symptoms develop, and why? 
Experiment 8.—Method of Testing the Virulence and Toxicity of 
Diphtheria Bacilli: 
1. Make a culture of a patient harboring the bacilli on a tube of Loffler’s serum medium. 
Inoculate at 35° C. for from eighteen to twenty-four hours; prepare a smear and stain with 
Loffler’s methylene-blue. If diphtheria bacilli are present, they must be isolated in pure 
culture. Never attempt a guinea-pig test with an impure culture. 
2. Isolate by the “streak” method, on plates of blood-serum. 
3. Inoculate a tube of 1 per cent, glucose bouillon, which is neutral or slightly alkaline, 
with several different colonies. 
4. Incubate at 35° C. for three days, keeping the tube in a slanted position in order to 
give the culture as much oxygen as possible. If a good growth does not appear in twenty- 
four hours, transplant to another tube of bouillon until the bacilli have been “educated” 
to grow on the medium. 
,5. Examine for purity. Select a 250- to 300-gram guinea-pig and inject 2 c.c. of the un- 
filtered culture in the median abdominal line. Animals over the weight specified are more 
resistant and less reliable for test. The unfiltered culture is used, since toxin is but one 
element of the disease-producing power of diphtheria bacilli, and toxin production in bouillon 
may not be a true index of the toxin production in mucous membranes. 
6. Carefully observe the animal for at least four days. Even slight toxemia, especially 
if accompanied by edema at the site of injection, should be regarded as a positive result. 
7. After death perform a careful autopsy. Make cultures of the edematous area, per- 
itoneum, and heart blood. Observe whether acute hyperemia of the suprarenal glands is 
present. 
8. Not infrequently animals showing mild or even an absence of the symptoms of tox- 
emia develop paralysis of the hindquarters two or three weeks later. According to Ehrlich, 
this paralysis is due to the action of “toxon,” a toxic substance secreted by the bacillus or, 
as believed by others, a modified form of toxin. 
9. To prove that diphtheria was the cause of the toxemia or death, mix 2 c.c. of the 
the culture in a test-tube with 1 c.c. of diphtheria antitoxin (500 units). After standing 
aside for an hour at room temperature, inject the mixture subcutaneously in the median 
abdominal line of a 250- or 300-gram guinea-pig. 
(a) Is the diphtheria bacillus aggressive? 
(b) What evidence have you that the lesions and death are due to a 
toxin? 
(c) Why does the use of diphtheria antitoxin make the test conclusive? 
Would tetanus antitoxin be capable of neutralizing diphtheria toxin? 
Experiment 9.—Tetanus Toxin; Tetanospasmin: 
Tetanus toxin is composed of two distinct poisons of different prop- 
erties. One, tetanospasmin, has a great affinity for the central nervous BACTERIAL, PLANT, AND ANIMAL TOXINS 
1113 
system, and is largely responsible for the symptoms of tetanus infection 
(neurotoxic); the second, tetanolysin, is thermolabile and is hematoxic. 
Tetanus toxin is very labile, and when in solution soon becomes attenuated. 
For these experiments it is necessary to use either fresh toxin or that which 
has been recently precipitated and dried. 
1. Secure some dried tetanus toxin and dissolve in sterile salt solution. The toxin 
may be secured from an antitoxin laboratory and preserved indefinitely in the refrigerator. 
Since the strength varies with different products, the lethal dose for a 300-gram guinea-pig 
should be ascertained from the laboratory furnishing the toxin. 
2. Secure 2 grams of fresh normal guinea-pig brain and liver. Crush in separate sterile 
mortars. 
3. Add a lethal dose of fresh tetanus toxin and 5 c.c. sterile salt solution to each. Mix 
thoroughly and place in the incubator for an hour. Remove and carefully transfer to sterile 
centrifuge tubes. Centrifuge thoroughly. 
4. Inject two 300-gram guinea-pigs subcutaneously in the median abdominal line with 
the supernatant fluids. 
5. Inject a third guinea-pig with a similar dose of toxin and 5 c.c. salt solution (control). 
6. Mark pigs carefully and observe for several days. 
(а) What are the symptoms of tetanus? 
(б) To what constituents of tetanus toxin are the chief symptoms due? 
(c) Does the pig which received the toxin-brain mixture show symptoms? 
How do you explain the result? 
(d) Does the pig which received the toxin-liver mixture show symptoms? 
How do you account for the result? 
(e) Is the tetanus bacillus characterized by its toxicity or aggressive- 
ness? 
(/) Is there a period of incubation before symptoms develop, and why? 
(g) Autopsy the animal. Are there any definite lesions of the tissues 
and internal organs? 
Experiment 10.—Action oe Tetanus Antitoxin; Specificity of Anti- 
toxins: 
1. In a test-tube or syringe place 50 M. L. D. of tetanus toxin and 2 c.c. of saline solution. 
2. In a second test-tube or syringe place 50 M. L. D. of tetanus toxin and 2 c.c. of tetanus 
antitoxin. 
3. In a third test-tube or syringe place 50 M. L. D. of tetanus toxin and 2 c.c. of diph- 
theria antitoxin. 
4. After one hour inject the mixtures subcutaneously into three young guinea-pigs 
respectively. 
5. Observe the animals for four or five days. 
(a) Discuss the mechanism of neutralization of tetanus toxin by anti- 
toxin. 
(,b) Discuss the specificity of antitoxins. 
(c) Define the units of tetanus and diphtheria antitoxins. 
Experiment 11.—Tetanus Toxin; Tetanolysin:, 
1. Secure some fresh tetanus toxin. 
2. Prepare a 1 per cent, emulsion of washed sheep corpuscles (washed three times in an 
excess of normal salt solution). 
3. Into a series of six test-tubes place 1 c.c. of the corpuscle emulsion and increasing 
doses of toxin: 0.1, 0.2, 0.4, 0.8,1, and 2 c.c. Add sufficient normal salt solution to bring the 
total volume to 3 c.c. 
4. Place 1 c.c. of the corpuscle suspension in 2 c.c. salt solution as a control to make sure 
that the salt solution is isotonic. 
5. Heat a portion of toxin at 60° C. for an hour in a water-bath and set up a similar set 
of tubes. 
6. Incubate all tubes for two hours at 37° C. 1114 
EXPERIMENTAL INFECTION AND IMMUNITY 
7. Read results at the end of this time and again after the tubes have settled in the 
refrigerator overnight. 
(a) Has hemolysis occurred in any of the tubes? 
(b) What is meant by hemolysis? 
(c) What constituent of tetanus toxin is responsible for hemolyzing these 
cells? 
(d) Does this experiment show the selective affinuty of a toxin for certain 
cells? 
(e) How is tetanus toxin produced? 
(/") Does heat destroy the hemotoxic agent? 
Experiment 12.—Antitetanolysin: 
1. Having determined the hemolytic unit of the tetanus toxin, place 2 units in each of 
four small test-tubes. 
2. Add tetanus antitoxin to each of the first three in the following amounts: 0.1, 0.5, 
and 1 c.c. To the fourth tube add 1 c.c. of saline solution. 
3. In a fifth tube place 2 c.c. of saline (to be the corpuscle control). 
4. Place tubes in a water-bath for one-half hour. 
5. To each of the five tubes add 1 c.c. of a 1 per cent, suspension of washed sheep cor- 
puscles; mix and reincubate for one hour. 
(a) In which tube has hemolysis occurred, and why? 
(b) Is tetanus antitoxin capable of neutralizing tetanolysin? 
Experiment 13.—Botulism Toxin: 
1. Prepare toxin by cultivating the Bacillus botulinus in an alkaline bouillon made in 
the form of an infusion from ham with the addition of 1 per cent, of glucose, 1 per cent, of 
peptone, and 1 per cent, of sodium chlorid. Strict anaerobic cultures should be grown for 
four weeks and filtered through a Berkefeld filter. The toxin may be preserved in brown, 
sealed vials, and kept on ice, or in a dried form in vacuum. 
2. Dilute 0.2 c.c. of the toxin with 3 or 4 c.c. salt solution and administer, per os, to a 
cat by means of a small catheter passed into the stomach. 
3. Inject 0.1 c.c. of the toxin subcutaneously in the abdominal wall of a rabbit. 
4. Observe animals closely for several hours following administration of the toxin, for 
some products are so highly toxic that symptoms may appear within a few hours. 
(a) What are the symptoms of botulism infection? 
(b) Do these symptoms show any selective action of botulism toxin for 
certain tissues? 
(c) Autopsy the animals. Are there any gross tissue changes? 
Experiment 14.—Dysentery Exotoxin and Endotoxin: 
1. To plain meat infusion broth add one-third volume of a 10 per cent, solution of egg- 
albumen. Adjust to Ph = 7.6 to 7.8 and autoclave for forty-five minutes at a pressure of 
15 pounds. 
2. Inoculate heavily with Shiga dysentery bacillus and incubate at 37° C. for five days. 
Filter through a sterile Berkefeld N candle. The filtrate is to be tested for exotoxin (Olitsky 
and Kligler). 
3. Inject two young rabbits intravenously with 0.5 and 1 c.c. of the toxin. Heat toxin 
at 70° C. for one hour and inject a third rabbit with 1 c.c. 
4. Cultivate the bacillus on Blake bottles of agar for twenty-four hours. Wash off 
each growth with 15 c.c. of saline solution and incubate the suspension at 37° C. for two 
days. Filter through a Berkefeld N candle (endotoxin). 
5. Inject two young rabbits intravenously with 2 and 5 c.c. of the toxin. Heat toxin 
at 70° C. for one hour and inject a third rabbit with 5 c.c. 
(a) Observe the animals for evidences of muscular weakness, prostration, 
paralysis, loss of weight, and diarrhea. 
(b) Carefully examine rabbits postmortem and particularly for patho- 
logic changes in the peritoneum, intestines, brain, and spinal cord. BACTERIAL, PLANT, AND ANIMAL TOXINS 
1115 
(c) Do the two toxins act alike? Which produces lesions and symp- 
toms referable to the nervous system? Which produces lesions and symptoms 
referable to the intestinal tract? 
(d) Do the toxins vary in resistance to heat? 
Experiment 15.—Bacteremia; Embolic Abscesses: 
Inject 5 c.c. of a twenty-four-hour bouillon culture of a virulent strain of Staphylococcus 
aureus in the ear vein of a rabbit. Take the temperature before and every twelve hours 
after injection; also leukocyte counts. Autopsy the animal seventy-two hours later under 
aseptic precautions. 
(a) What lesions are found? 
(b) Where are these chiefly situated, and why? 
(c) Culture the heart’s blood on tubes of agar, also several of the lesions. 
(d) After twenty-four hours’ incubation examine your cultures. 
(e) What was death due to? 
(/") Define bacteremia. 
(g) Prepare and stain sections of the kidney, including a lesion. What 
are the histologic changes? How do you explain the production of these 
tissue changes? 
(h) What is meant by focal infection and selective localization of bac- 
teria? What physical factors may influence the localization of bacteria in 
the tissues? 
Experiment 16.—Staphylolysin: 
1. Inoculate a flask of neutral bouillon with a virulent culture of Staphylococcus aureus 
and grow for several days at 37° C. 
2. Prepare a 1 per cent, suspension of washed rabbit erythrocytes in normal salt solution. 
3. Into a series of six test-tubes place 1 c.c. of the corpuscle suspension and increasing 
amounts of staphylococcus culture: 0.1, 0.2, 0.4, 0.8, 1, and 2 c.c. Add normal salt solution 
to bring the total volume to 3 c.c. 
4. Prepare a control by placing 1 c.c. of corpuscle suspension in 2 c.c. salt solution; also 
a second control by mixing 1 c.c. of sterile bouillon and 1 c.c. of corpuscle suspension. 
5. Incubate tubes for two hours at 37° C. and read results. 
6. Repeat, using human, rabbit, and guinea-pig erythrocytes. 
7. This test will be referred to again in the technic for determining the antilysin in blood- 
serum. 
(a) Has hemolysis occurred in any of the tubes? 
(b) To what constituents of the culture are these results due? 
(c) Does this experiment show the selective action of a bacterial toxin? 
(d) Do the results have any bearing upon the anemia and jaundice of 
severe staphylococcus infections? 
(e) What other bacteria contain or produce hemolysins? 
Experiment 17.—Streptotoxin; Streptolysins: 
1. Cultivate a strain of virulent hemolytic streptococci in tubes of slightly alkaline serum 
or ascites bouillon for forty-eight hours at 37° C.; also on blood-agar plates. 
2. Inoculate a guinea-pig intraperitoneally with 1 or 2 c.c. of the unfiltered culture. 
3. Test for the presence of streptolysins, using sheep corpuscles and increasing amounts 
•of unfiltered culture as in Experiment 16. Repeat, using human, rabbit, and guinea-pig 
corpuscles. 
4. Autopsy aseptically eighteen or twenty-four hours later. 
(a) What are the gross features of the exudate? 
(b) Prepare smears of the exudate and stain with methylene-blue. 
Are there many cells present? How do you explain the cellular content? 1116 
EXPERIMENTAL INFECTION AND IMMUNITY 
(c) Are streptococci present? Are these inclosed by any of the cells? 
How do you explain the condition? 
(d) Is the exudate bloody? If so, why? 
(e) Examine your blood-agar plates. Are there any peculiar changes 
around the colonies? Do any of the test-tubes show hemolysis? To what 
are these changes due? Does this have any connection with the bloody 
character of the exudate? 
(f) Why are streptococcus infections so virulent and spreading in char- 
acter? 
(g) Is the streptococcus an aggressive micro-organism? 
(,h) Do streptococci contain an endotoxin? 
(i) How are streptococci classified? 
Experiment 18.—Phytotoxins: 
1. Secure 0.01 gm. ricin or abrin and dissolve in 10 c.c. normal salt solution or distilled 
water. 
2. Inject 1 c.c. intravenously into a rabbit. Place the animal in a metabolic cage and 
collect urine, or catheterize every twelve hours or at death. 
3. Examine the urine for hemoglobin and erythrocytes. 
4. Prepare a 1 per cent, suspension of washed rabbit and guinea-pig corpuscles. 
5. Into a series of six small test-tubes place increasing doses of the ricin or abrin solu- 
tion as follows: 0.1, 0.2, 0.3, 0.4, 0.5, and 0.8 c.c. Add 1 c.c. of rabbit-cell emulsion to each 
and sufficient normal salt solution to make the total volume in each tube equal to 2 c.c. 
A seventh tube is the corpuscle control and contains 1 clc. of the erythrocyte suspension and 
1 c.c. of salt solution. 
6. Prepare a similar series of tubes with the guinea-pig erythrocyte suspension. 
7. Shake the tubes gently and incubate for two hours. 
(a) Do any of the tubes show hemolysis or hemagglutination? 
(b) Is the action the same with both bloods? 
(c) Does the plant toxin show a selective affinity? 
(id) Does the rabbit show any evidences of hemolytic jaundice? Are 
there blood elements in the urine? 
(e) Autopsy the animal, paying particular attention to the kidneys and 
gastro-intestinal tract? What changes have occurred? 
Experiment 19.—Zootoxin (Cobra Venom): 
1. Collect about 2 c.c. of normal human blood and place in 5 c.c. of 2 per cent, sodium 
citrate in 0.85 per cent, sodium chlorid solution. This mixture must not be shaken and the 
cells should be washed at least four times with normal salt solution, after which they are 
made up to a 4 per cent, suspension. 
2. Prepare a 2 per cent, suspension of sheep corpuscles in the same manner. 
3. Prepare a stock 1 : 5000 dilution of cobra venom by dissolving 0.005 gm. dried venom 
in 10 c.c. of normal salt solution. This solution will keep about one week in the refrigerator. 
4. Prepare a solution of venom 1 : 10,000 by diluting 2 c.c. of the stock solution with 8 
c.c. salt solution. Prepare a 1 : 20,000 dilution by diluting 2 c.c. of dilution 1 : 10,000 with 
2 c.c. salt solution. 
5. Place 1 c.c. of corpuscle suspension into each of three small test-tubes and add 1 c.c. 
of each dilution of venom respectively. Shake the tubes gently and place in the incubator 
for an hour and then in the refrigerator overnight. 
6. Repeat, using rabbit and guinea-pig corpuscles. 
7. Inject 2 c.c. of the 1 :5000 dilution intravenously into a rabbit and 1 c.c. intraven- 
ously into a guinea-pig. 
8. Observe the animals closely, particularly the urine. 
(a) Does the venom show a hemotoxic action? Does it act the same 
on rabbit and guinea-pig corpuscles? If not, why not? 
(b) How do you explain venom hemolysis? 
(c) What are the evidences of venom hemolysis in vivo? 
(,d) Is the hemotoxic poison the most dangerous constituent of venom? BACTERIAL PROTEIN; PTOMAINS 
1117 
ENDOTOXINS AND AGGRESSINS 
Experiment 20.—Endotoxins and Artificial Aggressins: 
1. Inoculate eight agar slants with Bacillus typhosus and cultivate at 37° C. for seventy- 
two hours. 
2. Wash off the growths with 3 c.c. distilled water for each tube. 
3. Place the emulsion in a sterile bottle with glass beads and shake for twenty-four 
hours at room temperature. At the same time inoculate six more slants of agar and wash 
off the growths at the end of twenty-four hours with 3 c.c. of normal salt solution for each 
tube. 
4. Filter the emulsion, which has been shaken, through a Berkefeld or candle filter. 
5. Inject two rabbits intravenously with 5 c.c. each of the filtrate (endotoxin) and un- 
filtered culture. 
6. Observe influence on body weight and temperature and for symptoms of toxemia. 
7. Inject three pigs intraperitoneally as follows: 
First pig: 4 c.c. of emulsion of typhoid bacilli. 
Second pig: 4 c.c. of filtrate. 
Third pig: 2 c.c. each of emulsion and filtrate. 
(a) Are there symptoms of toxemia in the rabbits? Are these symp- 
toms alike in both animals? Which is more toxic, the filtrate or the whole 
culture? 
(b) Does the filtrate enhance the effect of the whole culture? 
(c) If any of the animals succumb, autopsy under aseptic precautions. 
Prepare cultures of the peritoneum and heart’s blood on agar slants or in 
neutral bouillon. Is the typhoid bacillus markedly aggressive? 
(<d) If there is a peritoneal exudate, prepare smears and stain with 
carbolfuchsin (1 ; 10). What type of cell predominates? Are any of the 
bacilli engulfed in the cells? 
(e) Do any of the bacilli show signs of disintegration? If so, do you 
know what these changes are due to? 
(/) Is the filtrate alone toxic? 
(g) Does the filtrate appear to enhance the effects of the unfiltered 
culture? 
BACTERIAL PROTEIN; PTOMAINS; MECHANICAL ACTION OF BACTERIA 
Experiment 21.—Bacterial Protein (Vaughan): 
1. Inoculate 10 slants of 2 per cent, neutral agar in large bottles with a culture of Bacillus 
coli and grow at 37° C. for four days. 
2. Add about 10 c.c. of distilled water to each bottle and gently remove the growth with 
a sterilized platinum wire, being particularly careful not to remove fragments of agar. 
3. Pipet the heavy emulsion to a large centrifuge tube. Another cubic centimeter or 
two of water is added to each bottle to remove the balance of the growth. 
4. Centrifuge at high speed until the bacilli are thoroughly settled. Decant off the super- 
natant fluid and add SO per cent, alcohol; mix and centrifuge. Decant and add 95 per cent, 
alcohol; mix and centrifuge. 
5. Remove the sediment of bacteria to a small flask with 50 c.c. of absolute alcohol and 
set aside at room temperature for a day. Decant off the alcohol and add 50 c.c. of ether. 
Mix and set aside for another twenty-four hours. Decant off the ether that remains and 
remove sediment to a porcelain or agate mortar. Place in the incubator for a few hours 
until thoroughly dry. 
6. Grind the dry mass very thoroughly, the operator wearing a mask, until a fine powder 
is produced. Place the powder of bacterial substance in a wide-mouthed dark glass bottle 
and preserve in a dark closet. A portion will be needed later for experiments in anaphylaxis. 
7. Mix 0.02 gm. of the dry powder with 10 c.c. salt solution. 
8. Inject 2 c.c. of this emulsion intraperitoneally in a 300-gram guinea-pig. 
9. Heat 2 c.c. of the emulsion at 60° C. for two hours and inject intraperitoneally in a 
guinea-pig. 
10. Inject a third pig intraperitoneally with 2 c.c. of a four-day bouillon culture of Ba- 
cillus coli. 1118 
EXPERIMENTAL INFECTION AND IMMUNITY 
(a) Could endotoxins withstand these various manipulations? 
(b) Does the heated extract kill the animal more quickly than the 
unheated, and if so, why? 
(c) What are the symptoms produced? 
(,d) Autopsy the animals. Are there evidences of peritonitis? Are 
there differences in the lesions of the three animals? If so, how do you 
explain them? 
(e) The bacterial split portion will be studied later under Anaphylaxis. 
Experiment 22.—Ptomains: 
1. Procure 4 ounces of beef and mince. 
2. Place in a flask with 250 c.c. tap-water and inoculate with a culture of Bacillus coli. 
Fit the flask with a rubber stopper with a glass tube to carry off gases. 
3. Inoculate a flask of neutral bouillon at the same time with the same culture. 
4. Cultivate both at 37° C. for several days. 
5. Filter both through a coarse Berkefeld filter. 
6. Concentrate both filtrates to one-half their volume at a low temperature on a water- 
bath. 
7. Inject two guinea-pigs with each preparation, giving 10 c.c. subcutaneously and 
5 c.c. intraperitoneally. 
8. Observe the four animals closely. 
(a) What symptoms develop in both sets of animals? 
(b) What are ptomains? What role do they play in the production of 
disease? 
(c) Can antibodies be produced for true ptomains? 
Experiment 23.—Mechanical Action of Bacteria: 
1. Inject a 300-gram pig intraperitoneally with 5 c.c. of a forty-eight-hour-old bouillon 
culture of virulent Bacillus anthracis. Great care must be exercised in handling and injecting 
this culture. 
2. Autopsy the animal at the end of forty-eight hours and make cultures on agar slants 
of the heart’s blood, liver, and spleen. Remove the heart, lungs, liver, spleen, and kidneys, 
and place in 2 per cent, formalin. Prepare smears of the blood and stain with Gram’s method. 
3. After fixing for twenty-four hours, sections of these organs are to be prepared and 
stained with methylene-blue and with Gram’s stain. 
4. Examine the cultures at the end of twenty-four hours. 
5. Examine the sections. 
(a) Is the anthrax bacillus aggressive? 
(b) Does the anthrax bacillus produce much toxin? 
(c) Did the animal show any symptoms of its infection? 
(d) Are there evidences of peritonitis? 
(e) How do you explain the probable causes of death in human anthrax? 
KINDS OF IMMUNITY 
Experiment 24.—Phagocytosis in Natural Immunity: 
1. Inject a healthy guinea-pig intraperitoneally with 1 c.c. of a twenty-four-hour or 
forty-eight-hour glucose bouillon culture of streptococci of low or moderate virulence. A 
culture of staphylococci may be used instead, although the former is preferable. 
2. At the end of eighteen hours study the peritoneal fluid. Prepare smears and stain with 
methylene-blue or other suitable stain. Prepare cultures in glucose bouillon of the heart’s 
blood. 
(a) Did the animal show any evidences of infection? 
Cb) Has phagocytosis occurred in the peritoneal cavity? 
(c) What constituent of normal serum aids phagocytosis? 
(d) Did a bacteremia develop? KINDS OF IMMUNITY 
1119 
Experiment 25.—Natural Antibacterial Immunity: 
1. Inject a 250-gram guinea-pig subcutaneously with 5 c.c. of a twenty-four-hour bouillon 
culture of Bacillus anthracis. 
2. Inject a large albino rat with the same dose and in the same manner. 
3. Observe the animals for twenty-four to forty-eight hours. At autopsy prepare smears 
and cultures of the heart blood of each. Stain the smears with methylene-blue and according 
to the method of Gram. 
(a) Do the animals present symptoms of infection? Are there any differ- 
ences between the two? 
(b) Are anthrax bacilli found in the smears and cultures of both animals? 
Which animal is immune? Could phagocytosis play a role in this immunity? 
Experiment 26.—Relative Factors in Natural Immunity: 
1. Place a frog in a shallow vessel of warm water in the incubator at 37° C. 
2. After twenty-four hours give a subcutaneous injection of tetanus toxin equal to one- 
tenth the fatal dose for a 300-grani guinea-pig. 
3. Inject a second frog with an equal amount of toxin and keep it at ordinary room 
temperature in cold water. 
4. Observe both animals. 
(a) Do symptoms of tetanus develop in any of the animals? 
(b) Why does heat favor tetanus infection? 
(c) Give further examples of relative natural immunity. 
Experiment 27.—Influence of Temperature Upon Natural Immunity: 
1. Procure dried tetanus toxin and dissolve sufficient in 6 c.c. normal salt solution equiv- 
alent to 6 lethal doses for a 350-gram guinea-pig. 
2. Inject 2 c.c. into the pectoral muscles of a young hen. Take the rectal temperature. 
3. Inject 2 c.c. into a second hen. Take the rectal temperature and place her in cold 
water until the temperature has dropped several degrees. 
4. Place hen No. 1 in a cage and keep her at ordinary laboratory temperature. 
5. Place hen No. 2 in a cage and keep her in a cold place or renew the bath several times 
in order to keep her temperature subnormal. 
(a) What is the normal temperature of the hen? 
(b) Do both animals present symptoms of tetanus? 
(c) How do you explain the difference? 
6. If hen No. 1 does not show symptoms of tetanus by the second day, reinject her with 
an amount of toxin equal to 10 lethal doses for a guinea-pig. 
(a) Does this hen present evidences of tetanus? 
(b) If so, how do you explain the result? 
Experiment 28.—Natural Immunity to Diphtheria: 
1. Secure diphtheria toxin in which the M. L. D. and L+ doses are known. 
2. Inject each of two white rats intraperitoneally with 1000 M. L. D. Inject a third 
rat with 10,000 M. L. D. 
3. Next morning bleed one of the first two rats and inject the serum into a guinea-pig 
(intraperitoneally). 
4. Dilute a portion of the toxin with saline solution in such manner that 0.2 c.c. carries 
1/50 of the L+ dose. 
5. Secure blood from additional normal rats and defibrinate; centrifuge and secure the 
serum. 
6. In a small test-tube place 0.2 c.c. of the diluted toxin and 0.8 c.c. of the serum; in a 
second tube place 0.2 c.c. of the toxin and 0.8 c.c. of saline solution. Allow mixtures to stand 
thirty minutes. 
7. Secure a 10- to 12-ounce guinea-pig; pluck the hairs away at three places on the abdo- 
men. Inject 0.1 c.c. of each of the three mixtures intracutaneously. 
8. Observe all of the animals daily for evidences of inflammatory reactions and necroses. 1120 
EXPERIMENTAL INFECTION AND IMMUNITY 
(a) Is the rat naturally immune to diphtheria toxin? Is the guinea-pig? 
(b) Is the immunity of the rat relative or absolute? 
(c) Is natural diphtheria antitoxin present in rat serum? 
(d) Upon what does the natural immunity of the rat to diphtheria 
depend? 
Experiment 29.—Natural Immunity to Diphtheria; The Schick Test: 
1. Secure diphtheria toxin of known titer and dilute for the Schick test as described in 
the text. 
2. Inject the proper amount intracutaneously; it is well for each member of the class to 
take an injection. 
3. Also secure the control fluid and inject intracutaneously. 
4. Read the reactions at the end of twenty-four and forty-eight hours; if positive make 
daily observations on the course and gradual subsidence of the reaction. 
(a) To what is natural immunity of human beings to diphtheria toxin 
due? 
(b) Discuss the relation of age to the Schick reaction. 
(c) Describe the appearance, mechanism, and significance of the positive 
reaction; the negative reaction. 
(d) Discuss the appearance, mechanism, and significance of the pseudo- 
positive reaction. 
(e) Discuss the practical value of the Schick test. 
ACQUIRED IMMUNITY 
Experiment 30.—Acquired Active (Antibacterial) Immunity: 
1. Immunize a rabbit with typhoid bacilli. 
2. Inject this immune rabbit intravenously with 6 loopfuls of a twenty-four-hour culture 
of Bacillus typhosus thoroughly emulsified in 2 c.c. sterile salt solution. 
3. Inject a normal rabbit of equal weight with an equal dose and in the same manner. 
4. Both animals are weighed and their temperatures recorded before the injections are 
given. If possible, take temperature every four hours. Observe both animals for symptoms .. 
of infection. If one or both succumb, autopsy aseptically and culture the heart blood in 
neutral bouillon or bile. 
(a) Are there differences in the clinical condition of the two animals? 
(b) To what do you ascribe these differences? 
(c) What chief antibody is responsible for the destruction of the bacilli 
in the immune animal? 
(d) Has typhoid bacteremia developed? 
Passive Acquired Immunity 
Experiment 31.—Acquired Passive (Antitoxic) Immunity: 
1. Inject a 250- to 300-gram guinea-pig subcutaneously with 1 c.c. (500 units) of diph- 
theria antitoxin (pig No. 1). 
2. Secure diphtheria toxin in which the L+ dose is known. 
3. Place this dose of toxin in a small test-tube or, better, in the barrel of a precision 
syringe (Hitchens), and add 1 c.c. antitoxin (500 units). Mix thoroughly and keep at room 
temperature for an hour. 
4. Inject pig No. 1 with an amount of toxin equal to the L-(- dose. 
5. Inject the toxin and antitoxin mixture into a second pig of 250 to 300 grams. 
6. Inject the same amount of toxin into a third pig of equal weight as a control. 
7. Observe animals closely for at least four or five days. 
(a) What are the chief symptoms of diphtheria intoxication of the 
guinea-pig? What are the chief pathologic changes? 
(b) Do all animals show these symptoms? 
(c) How do you explain the absence of symptoms in pig No. 1? How 
in pig No. 2? Is the mechanism of protection the same in both? PHAGOCYTOSIS AND CHEMOTAXIS 
1121 
(d) What bearing has this experiment upon the treatment of diph- 
theria? 
Experiment 32.—Acquired Passive (Antitoxic) Immunity: 
The above experiment may be conducted in exactly the same manner, using tetanus 
toxin and antitoxin. 
Experiment 33.—Acquired Passive (Antibacterial) Immunity: 
1. Secure a culture of Type I pneumococcus, and if its lethal dose for white mice is un- 
known, determine this by injecting a series of mice with decreasing doses of a twenty-four-hour 
serum bouillon culture. 
2. Secure a few cubic centimeters of Type I antipneumococcus serum. 
3. Inject three mice intraperitoneally with 0.1, 0.5, and 1 c.c. of the serum. Then inject 
them intraperitoneally with 1000 minimal lethal doses of pneumococci. Inject a fourth 
mouse with culture alone (control). 
4. Observe the animals for several days, and at autopsy culture the heart blood in tubes 
of serum bouillon and prepare smears which are to be stained according to Gram. 
(a) Has the serum served to protect any of the mice? 
(b) Is this protection due to bacteriolysins, bacteriotropins, or both? 
Do antitoxins play any part? 
(c) Why is it preferable to use homologous culture and immune serum? 
What bearing has this upon the treatment of human pneumococcus infec- 
tions? 
(d) Do you know a method for quickly isolating and identifying pneu- 
mococci from sputum? 
Note.—This experiment may be conducted with rabbits; also a culture 
of streptococcus and its immune serum may be used. 
PHAGOCYTOSIS AND CHEMOTAXIS 
Experiment 34.—Phagocytosis (Macrophages) : 
1. Secure pigeons’ blood and defibrinate. Wash the corpuscles several times and prepare 
a 5 per cent, suspension. 
2. Inject a guinea-pig intraperitoneally with 3 c.c. of this corpuscle suspension. 
3. After three hours withdraw a small amount of peritoneal exudate by means of a 
capillary pipet. Examine with hanging-drop preparations and prepare smears and stain 
with Wright’s stain. 
4. Make similar preparations twelve, eighteen, twenty-four, and forty-eight hours 
after injection. 
(a) Has phagocytosis occurred? 
(,b) Which cells have become phagocytes? 
(c) Do the pigeon cells appear as if undergoing digestion? 
(d) Which portion of the pigeon cell resists digestion? 
(e) Explain mechanism of intracellular digestion. 
Experiment 35.—Phagocytosis (Microphages): 
1. Place a drop of blood in the hollow cell of a ground-out slide such as is used for hanging- 
drop preparations. Cover with a clean slide, seal with a ring of vaselin, and place in a large 
Petri dish containing pieces of filter-paper moistened with water (moist chamber). Place in 
the incubator for fifteen minutes. 
2. Remove the cover-glass and carefully wash cover-glass and cell with normal salt 
solution. The erythrocytes are removed and the leukocytes left adherent to the cell and 
cover-glass. 
3. Fill the cell with fresh serum and add a quantity of culture, preferably a loopful of a 
twenty-four-hour bouillon culture of non-virulent anthrax bacilli. Apply the cover-glass 
and vaselin the margins to prevent evaporation. 
4. If at all possible, employ a warm stage and watch the process under the microscope. 1122 
EXPERIMENTAL INFECTION AND IMMUNITY 
(a) Does phagocytosis occur? 
(b) Which cells are acting as phagocytes? 
(c) Can you make out the phase of fixation and phase of ingestion? 
(d) By what agencies do leukocytes kill engulfed bacteria? 
Experiment 36.—Phagocytosis; Positive Chemotaxis: 
1. Inject a guinea-pig intraperitoneally with 5 c.c. of finely divided cinnabar suspension 
and 1 c.c. subcutaneously on each side in the region of the inguinal lymph-glands. 
2. Autopsy the animal at the end of twenty-four hours. Prepare hanging-drop prepara- 
tions and smears of the peritoneal exudate (stained lightly with methylene-blue). Prepare 
sections of the inguinal and abdominal lymphatic glands. 
(a) Has phagocytosis occurred in the peritoneal cavity? Which cells 
have become phagocytic? 
(b) Do you find phagocytes in the lymph-glands? Where are the 
leukocytes situated? Which cells have become phagocytes? 
(c) How did the material reach the glands? 
(d) How do you explain the presence of cinnabar in the endothelial cells 
of the gland? 
1. Inject a guinea-pig intraperitoneally with 10 c.c. of a twenty-four-hour culture of 
Staphylococcus aureus. 
2. Autopsy the animal eighteen hours later and prepare smears of the exudate. Fix 
with methyl alcohol for five minutes, dry, and stain with carbol-thionin, Wright’s, or Loffler’s 
methylene-blue, counterstained with eosin. 
3. Prepare cultures of the peritoneal exudate. 
(a) Describe the exudate. 
(b) Has phagocytosis occurred? 
(c) Which cells are actively phagocytic? 
(d) Are all the cocci engulfed? 
(e) Are there any evidences of the cocci undergoing intracellular diges- 
tion? 
(/) Could a sterile substance call forth an exudation of leukocytes when 
injected into a serous cavity? 
Experiment 37.—Negative Chemotaxis: 
1. Inject a guinea-pig intraperitoneally with 5 c.c. of a forty-eight-hour serum bouillon 
culture of virulent streptococci. 
2. Autopsy at the end of eighteen to twenty-four hours if death has not already occurred. 
3. Prepare cultures in serum bouillon and a number of smears. Stain the latter with 
methylene-blue or according to Gram. 
(a) Describe the exudate. How does it differ from that found in the 
preceding experiment? 
(b) How do you explain the serous character of the exudate? 
(e) Has phagocytosis occurred? 
id) Do you think the streptococci have multiplied in the peritoneal 
cavity? 
(e) What means could you suggest for overcoming this action of virulent 
streptococci? 
OPSONINS 
Experiment 38.—Normal Opsonins: 
1. Prick the finger and secure 1 c.c. of blood in a small test-tube. Also 1 c.c. in a centri- 
fuge tube containing 2 c.c. of sodium citrate solution. After coagulation remove the serum 
from the first tube. OPSONINS 
1123 
2. Divide the serum into two portions and heat one portion at 56° C. for thirty minutes. 
3. Prepare an emulsion of leukocytes by centrifuging the blood collected in the citrate 
solution, removing the supernatant fluid, adding normal salt solution, and centrifuging 
again. Repeat this step once more in order to wash the cells thoroughly and after the last 
centrifuging remove the supernatant fluid and add sufficient salt solution to make the total vol- 
ume 1 c.c. and mix thoroughly. 
4. Prepare an emulsion of staphylococci which is homogeneous and free of clumps. 
5. Mark two capillary tubes with a wax pencil about 1 inch from the tip; fit rubber teats 
to the other end. 
6. With pipet No. 1 take up a volume of blood-cells; allow a bubble of air to enter and 
then an equal volume of bacterial emulsion; bubble of air and an equal volume of the fresh 
unheated serum. Mix well by alternate expulsion and aspiration on a clean slide. Then 
draw the whole into the stem of the pipet and seal the tip in a flame. 
7. Repeat with pipet No. 2, using the heated serum. 
8. Incubate both pipets at 37° C. for half an hour. 
9. Remove the pipets from the incubator, break off the tips, mix the contents, and 
prepare smears. 
10. Fix the smears with a saturated solution of bichlorid of mercury for one minute; wash 
in water and stain with carbol-thionin for two minutes; wash in water and dry. 
11. Examine with oil-immersion lens. 
(a) What are the requisites of a satisfactory opsonic preparation? 
(b) Is there any difference in the amount of phagocytosis between the 
heated and unheated serums? If so, how do you explain the result? 
(c) Do normal opsonins play any role in natural immunity? 
(d) Give the properties of normal opsonins. 
Experiment 39.—Immune Opsonins (Bacteriotropins): 
1. Secure 1 c.c. of serum from a rabbit immunized with staphylococci and heat 0.5 c.c. 
at 56° C. for thirty minutes. 
2. Using the same blood and bacterial emulsion as prepared in the preceding experiment 
prepare two opsonic preparations with the heated and unheated immune serum. 
(a) Is phagocytosis more marked than in the preceding experiment? 
If so, why? 
(b) Is there any difference in the degree and extent of phagocytosis 
with the heated and unheated serum? 
(c) Give the properties of immune opsonin or bacteriotropin. 
Experiment 40.—Quantitative Estimation of Bacteriotropins: 
1. Secure a culture of normal meningococci and a small amount of polyvalent antimenin- 
gococcus serum. 
2. Determine the tropin titer according to the technic described in the text. 
(a) Define the meaning of opsonic index versus phagocytic index. 
(b) Discuss the relation of bacteriotropins to mechanism of therapeutic 
activity of antimeningococcus sera. 
(c) Define opsonins versus bacteriotropins. 
(d) What relation do bacteriotropins bear to immunity? 
(ie) In what diseases are bacteriotropins active? 
Experimen t 41.—Hemopsonin : 
1. Secure 0.5 c.c. serum from a rabbit immunized with sheep cells and heat at 56° C. for 
a half an hour to destroy hemolytic complement. 
2. Prepare a 5 per cent, suspension of sheep erythrocytes in normal salt solution after 
washing them three times. 
3. Prepare an emulsion of rabbit leukocytes; the emulsion prepared in the preceding 
experiments may be used. 
4. Take a capillary pipet; make a mark about 1 inch from the tip and fit the other end 
with a rubber teat. 1124 
EXPERIMENTAL INFECTION AND IMMUNITY 
5. Draw up equal volumes of leukocytic emulsion, sheep cell suspension, and serum. 
Mix well, seal the pipet, and inoculate for an hour at 37° C. 
6. Prepare smears and stain with Wright’s blood-stain. 
(a) Has phagocytosis occurred? 
(b) Which cells have become phagocytes? 
(c) Has hemolysis occurred in the mixtures, and if not, why not? 
Experiment 42.—Mechanism of Action of Opsonins and Bacteriotro- 
pins: 
1. Secure serum from a rabbit immunized with Staphylococcus pyogenes aureus. 
2. Prepare a suspension of rabbit leukocytes. 
3. Prepare an emulsion of eighteen-hour culture of Staphylococcus aureus. 
4. Make the following mixtures in capillary pipets: 
No. 1: Equal parts of leukocytes, serum, and bacterial emulsion. 
No. 2: Equal parts of leukocytes and bacterial suspension. 
5. In two small test-tubes mix: 
No. 3: 0.5 c.c. each of leukocytic suspension and serum. 
No. 4: 0.5 c.c. each of serum and bacterial emulsion. 
6. Incubate pipets 1 and 2 and tubes 3 and 4 for fifteen minutes at 37° C. 
7. Prepare smears of capillary tubes 1 and 2 and stain with carbol-thionin. 
8. To tubes 3 and 4 add 15 c.c. normal salt solution, mix, and centrifuge thoroughly. 
Decant off the supernatant fluid and restore volume in each tube to 1 c.c. 
9. Prepare capillary pipets as follows: 
No. 3: Equal parts of contents of tube 3 and bacterial emulsion. 
No. 4: Equal parts of contents of tube 4 and leukocytic mixture. 
10. Incubate both pipets for fifteen minutes, prepare smears, and stain with carbol- 
thionin. 
(a) Carefully compare the degree and extent of phagocytosis in the 
four mixtures. 
(b) Has leukocytosis occurred in mixture No. 2? If so, what special 
term is applied to phagocytosis in the absence of serum? 
(c) What role does serum play in phagocytosis? Explain the mechanism 
involved. 
(d) Discuss the question of “stimulin” and opsonin as demonstrated 
by the results in preparations Nos. 3 and 4. 
Experiment 43.—Specificity of Opsonins and Bacteriotropins: 
1. Secure 1 c.c. of the serum of a rabbit immunized with staphylococci and heat at 56° 
C. for thirty minutes. Divide into two portions in separate small test-tubes. 
2. To No. 1 add 1 c.c. of normal salt solution and 4 or 5 loopfuls of a twenty-four-hour 
agar slant culture of staphylococci. Incubate for thirty minutes; centrifuge thoroughly and 
remove the supernatant fluid (diluted serum) to a separate tube. Call this “treated” serum. 
3. To the untreated serum add 1 c.c. salt solution, so that both are diluted equally. 
4. Prepare an emulsion of rabbit or human leukocytes. 
5. Prepare an emulsion of staphylococci, homogeneous and free of clumps. 
6. Prepare two opsonic mixtures: No. 1 containing equal parts of blood suspension, bac- 
terial emulsion, and untreated serum; No. 2 containing equal parts of blood, bacterial emul- 
sion, and treated serum. 
7. Incubate for thirty minutes. During this interval proceed as follows: 
8. Prepare an emulsion of typhoid bacilli, homogeneous and free of clumps. 
9. Secure 0.5 c.c. of serum from a rabbit immunized with typhoid bacilli and heat at 36° 
C. for thirty minutes. 
10. Prepare the following mixtures in capillary pipets: 
No. 3 blood-cells + typhoid serum + typhoid bacterial emulsion. 
No. 4 blood-cells + typhoid serum + staphylococcus bacterial emulsion. 
No. 5 blood-cells + staphylococcus serum + typhoid emulsion. 
11. Incubate for fifteen minutes. 
12. Prepare smears of all and stain with carbol-thionin. 
(a) Is there any difference in the degree of phagocytosis in mixtures 
No. 1 and 2? If so, why? ANTITOXINS 
1125 
(b) Are opsonins and bacteriotropins specific? 
(c) Has phagocytosis occurred in the cross-mixtures of staphylococci 
with typhoid serum, and typhoid bacilli with staphylococcus serum? If 
so, how do you explain this apparent lack of specificity? 
(d) What may happen in a mixture of fresh unheated typhoid serum, 
typhoid bacilli, and leukocytic emulsion? 
OPSONIC INDEX 
Experiment 44.—Determining the Opsonic Index: 
1. Secure a small quantity of blood from a guinea-pig or rabbit which has been immu- 
nized with Staphylococcus pyogenes aureus. Also two specimens from normal animals to 
serve as control serums. Remove the serums, being careful to keep them marked and sepa- 
rate. The two normal serums may be mixed in a watch-glass or small test-tube (pooled). 
2. Prepare a leukocyte mixture, using normal guinea-pig or rabbit blood according to 
the immune animal used. 
3. Prepare a bacterial emulsion of an eighteen-hour culture of Staphylococcus pyogenes 
aureus. 
4. Proceed to determine the phagocytic and opsonic index as per technic given in the 
text. 
(a) What constitutes a satisfactory bacterial emulsion? 
(b) What constitutes a satisfactory leukocytic suspension? Name 
several methods of obtaining leukocytes for this technic. 
(c) What constitutes a satisfactory phagocytic film? 
(id) Should the serum be fresh? 
(e) How do you determine the phagocytic index? 
(f) How do you determine the opsonic index? 
(g) What is the relation between the opsonic and phagocytic indices? 
(h) Give the practical value of the opsonic index in disease. 
(i) Give the practical value of the opsonic index as a measure of immunity. 
Note.—If any of the students have received typhoid vaccine, the opsonic 
index with his serum may be determined. 
BACTERIAL VACCINES 
Experiment 45.—Preparation of Typhoid Vaccine : 
1. Prepare two to six agar slant cultures of Bacillus typhosus and grow for forty-eight 
hours at 37° C. 
2. Remove cultures, prepare a suspension, count by the method of Wright, sterilize, 
dilute, and prepare vaccine for administration as given in the text. 
3. Prepare two emulsions: one to contain about 500,000,000 bacilli in each cubic centi- 
meter (first dose), and the second 1,000,000,000 per cubic centimeter (second and third doses). 
Experiment 46.—Preparation of Staphylococcus Vaccine: 
1. Prepare two to six agar slant cultures of Staphylococcus aureus and grow for twenty- 
four hours at 37° C. If a patient with furunculosis is available, make cultures of pus and 
secure staphylococcus, of which a vaccine is prepared. 
2. Proceed in the preparation of the vaccine as given in the text, placing each dose in 
separate ampules, and so diluting that each dose is of 1 c.c. and contains 1,000,000,000 of cocci. 
Count by the method of Wright and by the counting chamber method. 
ANTITOXINS 
Experiment 47.—Standardizing Diphtheria Antitoxin: 
1. Prepare a strong diphtheria toxin with the Park-Williams Bacillus No. 8, after the 
technic given in the text. 
2. Secure some of the dried Standard Antitoxin and dilute so that 1 c.c. is equal to one 
immunity unit. 1126 
EXPERIMENTAL INFECTION AND IMMUNITY 
3. With this antitoxin determine the L-f dose of the toxin prepared according to the 
technic given, using 6 guinea-pigs and the following doses of toxin; 0.1, 0.12, 0.15, 0.18, 
0.2, and 0.25 c.c. 
4. Secure a sample of diphtheria antitoxin in the open market containing about 4 c.c. 
serum and 2000 units of antitoxin. If the titration given on the label is still correct, one would 
expect about 500 units of antitoxin per cubic centimeter of serum. Carefully remove 1 c.c. 
of serum and dilute with 19 c.c. salt solution (1 ; 20). From this stock dilution prepare the 
following dilutions (taken frQm Bulletin No. 21, Hygienic Laboratory, M. J. Rosenau): 
1 c.c. + 14 c.c. NaCl solution 1 c.c. = .00333 or 1/300 = 300 units per c.c. 
1 c.c. + 16 c.c. NaCl solution 1 c.c. = .00294 or 1/340 = 340 units per c.c. 
1 c.c. + 18 c.c. NaCl solution 1 c.c. = .00263 or 1/380 = 380 units per c.c, 
1 c.c. + 20 c.c. NaCl solution 1 c.c. = .00238 or 1/420 = 420 units per c.c, 
1 c.c. + 22 c.c. NaCl solution 1 c.c. = .00217 or 1/460 = 460 units per c.c. 
1 c.c. -j- 24 c.c. NaCl solution 1 c.c. = .002 or 1/500 = 500 units per c.c. 
5. Mix 1 c.c. of these various dilutions with the L-f- dose of toxin; stand aside for an 
hour and inject subcutaneously in median abdominal line of 250- to 300-gram guinea-pig as 
per the technic already given. 
6. Carefully observe all animals for a period of four days at least. Autopsy those that 
succumb, paying particular attention to the condition of the abdominal wall and suprarenal 
glands. If the serum should contain less than 300 units of antitoxin per cubic centimeter of 
serum, the test should be repeated with lower dilutions. 
7. Inject 1 c.c. of the serum subcutaneously into a white mouse to test for excess of 
preservative. It requires 1 c.c. of a 0.5 per cent, solution of tricresol or 0.5 c.c. of a 0.5 per 
cent, phenol solution to kill a medium-sized mouse. If the mouse shows trembling, it would 
indicate that the serum contains nearly this percentage of tricresol (Bulletin No. 21, Hy- 
gienic Laboratory). 
8. Inoculate 1 c.c. of the serum in a flask containing 100 c.c. sterile neutral bouillon. 
Incubate at 37° C. for at least four days. This will test the sterility of the product. 
(a) Define the unit of diphtheria antitoxin. 
(,b) Of what practical value is the measurement of diphtheria antitoxin? 
(c) Is there any practical method of determining the quantity of toxin 
in the blood of a diphtheric patient? 
(id) How would you determine the amount of natural antitoxin in the 
blood of a person? 
Experiment 48.—Standardizing Tetanus Antitoxin: 
The technic of standardizing tetanus antitoxin may be carried out in a similar manner 
with dried Standard Toxin and an antitoxin purchased in the open market. 
Experiment 49.—Specificity of Antitoxins; 
1. Secure small quantities of fresh diphtheria and tetanus toxins; the L+ dose of each 
should be known. 
2. Secure small quantities of diphtheria and tetanus antitoxins; the number of units per 
cubic centimeter of serum should be known. 
3. Into four precision syringes place the following mixtures. After standing an hour at 
room temperature, inject subcutaneously into 300-gram guinea-pigs. 
No. 1: L-f- dose of diphtheria toxin + 100 units of diphtheria antitoxin. Inject 
into pig No. 1. 
No. 2: L+ dose of diphtheria toxin -f- 100 units of tetanus antitoxin. Inject into 
pig No. 2. 
No. 3: L-}- dose of tetanus toxin -f- 100 units of tetanus antitoxin. Inject into pig 
No. 3. 
No. 4: L-f- dose of tetanus toxin + 100 units of diphtheria antitoxin. Inject into 
pig No. 4. 
(a) What do you observe regarding the specificity of antitoxins? 
(b) What are the main symptoms of diphtheria and tetanus in the 
guinea-pig? 
(c) Even though a pig injected with diphtheria toxin shows no general 
symptoms of intoxication, what local sign may be present? 
(d) What is the nature of the toxin-antitoxin reaction? FERMENTS AND ANTIFERMENTS 
1127 
Experiment 50.—Nature of the Toxin-antitoxin Reaction. Action 
of Antitetanolysin: 
1. Secure fresh tetanus toxin and determine the dose producing complete hemolysis 
of 1 c.c. of a 1 per cent, suspension of rabbit corpuscles in two hours at 37° C. 
2. Place double this dose of toxin in a series of six small test-tubes and add increasing 
doses of fresh tetanus antitoxin: 0.001, 0.005, 0.01, 0.05, 0.1, and 0.2 c.c. Add salt solu- 
tion to bring the total volume to 1 c.c.; incubate at 37° C. for one hour. Add of 1 per 
cent, suspension of rabbit corpuscles to each tube. Prepare two controls, one with the 
dose of toxin and corpuscles, but to which no serum is added, the second with 0.2 c.c. serum 
and dose of corpuscles. Shake tubes and incubate for two hours. Make a preliminary read- 
ing and again after tubes have settled twenty-four hours in the refrigerator. 
3. The first control should be completely hemolyzed, indicating that a sufficient lytic 
dose of toxin was employed. 
(a) What constituent of tetanus toxin has a marked affinity for erythro- 
cytes? 
(b) Is this agent thermostabile? How can you determine this? 
(c) What role does it play in tetanus intoxication? 
(d) How is this hemotoxic agent neutralized by tetanus antitoxin? 
(e) Would anti-tetanospasmin neutralize the hemotoxic activity of 
tetanus toxin? 
(f) Would diphtheria antitoxin neutralize this hemotoxic agent? If 
not, why not? 
(g) Explain the mechanism of neutralization of a toxin by antitoxin 
in vitro. Is it the same as that occurring in vivo? 
Experiment 51—Antistaphylolysin: 
1. Prepare a staphylolysin by growing a culture of Staphylococcus aureus in bouillon 
for two or three days. Pass through a Berkefeld filter and preserve the filtrate with 0.5 
phenol. Determine the lytic dose for 1 c.c. of a 1 per cent, suspension of rabbit corpuscles. 
2. Secure serum from a rabbit immunized with staphylococci. Heat at 56° C. for thirty 
minutes. 
3. Secure normal horse-serum. Heat it at 56° C. for thirty minutes. 
4. Secure normal rabbit-serum. Heat at 56° C. for thirty minutes. 
5. Into a series of six small test-tubes place the lytic dose of staphylolysin and increasing 
amounts of rabbit immune serum as follows: 0.001, 0.005, 0.01, 0.05, 0.1, 0.2 c.c. Arrange 
a similar series, using normal horse-serum and normal rabbit-serum. Prepare a control con- 
taining the lytic dose of toxin. Add 1 c.c. of a 1 per cent, suspension of rabbit cells to each 
tube and sufficient normal salt solution to make the total volume equal 2 c.c. Shake each 
tube gently and incubate at 37° C. for two hours. 
6. Inspect the tubes. The smallest amount of normal horse-serum inhibiting hemolysis 
is taken as 1, or the unit. Compare the values of the immune and normal rabbit-serums with 
this unit. 
(a) Why is normal horse-serum adopted as the standard? 
(b) What is antistaphylolysin? 
(c) What are the various agents produced by staphylococci and re- 
sponsible for the lesions and symptoms of staphylococcus infections? 
(d) Would the antilysin neutralize the leukocidin? 
(e) Explain the mechanism of lysin-antilysin action. 
(/) Which role does the lysin play in staphylococcus infections? 
(g) Of what value would be the titration of antistaphylolysin in the 
serum of the patient? 
FERMENTS AND ANTIFERMENTS 
Experiment 52.—Tryptic Ferment oe Leukocytes: 
1. Collect 0.5 c.c. of blood in each of two small test-tubes by puncture of a finger; set 
aside until coagulation has occurred. Add several changes of warm distilled water until the 
red corpuscles are hemolyzed and colorless clots of leukocytes, fibrin platelets, and detritus 
are secured. 1128 
EXPERIMENTAL INFECTION AND IMMUNITY 
2. Place one clot in the bottom of a tube of sterile and slanted Loffier blood-serum 
medium which is fairly dry and firm. 
3. Place the second clot in a second tube of this medium and add 0.2 to 0.4 c.c. of fresh 
serum. 
4. Plug these tubes firmly, paraffin the stoppers to prevent evaporation, and incubate 
along with a non-inoculated control at 50° C. for two days. 
(a) In which tube do you note evidences of digestion? 
(b) Describe the appearance of the digested medium. 
(c) Does the tube containing serum show digestion? How do you 
explain the result? 
(d) What role may this ferment play in infection? 
(e) Why do not the leukocytes digest themselves? 
(/) What is the nature of the ferment? 
(g) May this ferment play a role in the disposal of old and dead leu- 
kocytes? 
Experiment 53.—Testing the Antitryptic Power of Blood-serum 
(After the Marcus Modification of the Method of Muller 
and Jochmann): 
1. Prepare a solution of trypsin by thoroughly shaking 0.1 gm. of Kahlbaum’s trypsin 
with 5 c.c. glycerin and 5 c.c. distilled water. Incubate at 55° C. for half an hour, shake 
thoroughly, filter, and preserve in the refrigerator. 
2. Secure six Petri dishes of Loffier blood-serum culture-media which have been suffi- 
ciently dried to drive off the water of condensation and with a firm elastic surface. 
3. Collect 1 c.c. of blood from a patient three hours after last meal; separate the serum. 
4. In a watch-glass or hanging-drop slide mix 1 drop of serum with an equal sized drop 
of trypsin solution. Mix well and transfer five loopfuls to an area on a Petri dish of medium. 
With a blue-wax pencil draw a circle on the cover of the plate to include the site of planting 
and mark No. 1. 
5. Prepare serial dilutions with 1 drop of serum with 2, 3, 4, 5, 6, 7, 8, 9, and 10 drops 
of trypsin solution. Mix well and plant five loopfuls of each dilution on the medium in five 
dishes. Two plantings may be made in each plate and with care they will not become con- 
fluent. Mark each plate. 
6. Inoculate the sixth plate with 5 drops of trypsin solution without serum (control). 
7. Incubate all tubes at 37° C. for six to twelve hours until the control shows well-marked 
digestion. 
8. Wherever the trypsin has not been neutralized by the serum, shallow liquefied dimples 
appear on the surface of the Loffier medium. The greater the amount of trypsin required to 
cause digestion, the higher the titer of antitrypsin in the blood. By conducting a control 
series with the two normal pooled serums one may determine whether in a given case the 
antitryptic power of the blood is normal, increased, or decreased. 
(a) Does this test possess any practical value? 
(b) Discuss the presence of antitrypsin in the blood-serum. 
BACTERIAL AGGLUTININS 
Experiment 54.—The Gruber-Widal Reaction in Typhoid Fever (“Wet” 
Method). Microscopic Reaction: 
1. Collect blood in a Wright capsule or small test-tube from a typhoid convalescent 
patient. Rabbit immune serum may be used instead. Separate the serum from the clot. 
2. Prepare two dilutions with hanging-drop slides, 1 : 50 and 1 :100, and a culture con- 
trol. Use a twenty-four-hour bouillon culture of Bacillus typhosus—one free of clumps and 
in which the bacilli are long and motile. 
3. It is well to prepare a similar test with a known positive typhoid serum and a normal 
negative serum. 
4. Dilution and slides are prepared according to the technic given in the text. A second 
set of tests in dilutions of 1 :40 and 1 : 80 may be prepared with the aid of the white cor- 
puscle pipet. 
5. Place slides away from direct sunlight and examine at the end of an hour. Examina- 
tions are readily made with the £ objective, the light being well cut off. Examine the cul- 
ture control first, then the higher and lower dilutions. BACTERIAL AGGLUTININS 
1129 
(a) Describe the phenomenon of agglutination. 
(b) Is it necessary for all bacilli to be agglutinated to constitute a positive 
reaction? 
(c) What are the features of a doubtful reaction? Of a negative re- 
action? 
(d) Does the normal serum contain agglutinin for the typhoid bacillus? 
(e) Why is it necessary to use high dilutions? What dilution is the 
lower limit of practical safety? 
(/) Does the control show agglutination? If so, by what term is this 
agglutination known? 
(g) What are the characteristics of a satisfactory culture for the micro- 
scopic agglutination test? 
(h) Would a dead culture be serviceable in this reaction? 
(i) What practical value has the Widal reaction in the diagnosis of 
typhoid fever? 
(j) What value has the agglutination reaction in determining the degree 
of immunity following typhoid immunization? 
Note.—If the students are immunized with typhoid vaccine, they may 
use one another’s serum in this and following experiments. 
Experiment 55.—Gruber-Widal Reaction in Typhoid Fever (“Dry” 
Method) : 
1. Prepare smears of blood of a typhoid convalescent patient on clean glass slides or 
collect a few drops on partially glazed paper, as prescription blanks. Allow blood to dry 
and do not apply heat. 
2. Using a good twenty-four-hour-old culture of Bacillus typhosus, prepare an aggluti- 
nation test after the technic given in the text. Be particularly careful not to transfer 
paper fiber to the slide—apply the salt solution and dissolve the blood by gently rubbing with 
a small platinum loop (2 mm.). Mix the blood in the loopful of culture with the slide held 
or placed over a white surface so that the proper delicate orange tint is secured. 
3. Prepare culture control as usual; also similar tests with a known positive and negative 
serum. 
4. Examine at the end of an hour. 
(a) What special precautions are to be observed in this method? 
(b) What is the particular practical value of this reaction? 
(c) Are accurate dilutions possible, and if so, by what technic? 
(d) Are agglutinins resistant to deleterious influences? 
(e) Compare the value of this method with the one used in the preceding 
experiment. 
Experiment 56.—Macroscopic Agglutination Reaction: 
1. Secure serum from a rabbit which has been immunized with Bacillus typhosus. Serum 
of a typhoid convalescent may be used instead. 
2. Prepare a series of serum dilutions in small narrow test-tubes in amounts of 1 c.c., 
ranging from 1 :10 up to 1 : 640. 
3. Add to each 1 c.c. of the bacillary emulsion of a good twenty-four- to forty-eight-hour 
bouillon culture of Bacillus typhosus or an emulsion prepared by washing twenty-four-hour 
growths from agar slant cultures with normal salt solution, according to the technic given in 
the text. This doubles the dilutions, which now range from 1 : 20 up to 1 : 1280. Prepare 
the culture control. 
4. Shake gently and incubate for two hours at 37° C. and then record results after tubes 
have been standing at room temperature for six hours. Re-examine tubes with the agglutino- 
scope and note the higher delicacy of such readings. 
5. If a typhoid immune serum of unknown titer is used and agglutination is complete 
in the highest dilution, the test must be repeated with still higher dilutions in order to deter- 
mine the agglutinin titer of the serum. 1130 
EXPERIMENTAL INFECTION AND IMMUNITY 
(a) Has agglutination occurred in the control tube? Why is this control 
so important in this and all agglutination reactions? 
(b) Is agglutination as complete in the lower as in the higher reaction? 
If not, how do you explain this result? Of what practical import is this 
phenomenon? 
(c) Describe the appearance of macroscopic agglutination. 
(d) Are the agglutinated bacilli dead? To determine this, pipet off the 
supernatant fluids of several tubes into a germicidal solution; then add an 
excess of sterile normal salt solution to the sediment of agglutinated bacilli, 
stir up the sediment, and transfer with a sterile pipet to a sterile centrifuge 
tube; centrifuge thoroughly; remove the supernatant fluid with a sterile 
pipet and plant several loopfuls of bacteria on slants of agar and in neutral 
bouillon. Why is it advisable to wash the sediment? 
(e) What advantages has the macroscopic over the microscopic method? 
Experiment 57.—Macroscopic Agglutination Reaction: 
1. Prepare dilutions in amounts of 1 c.c. of a typhoid immune serum in proper test-tubes, 
ranging from 1 : 20 up to 1 : 1280. 
2. Add one loopful of a twenty-four-hour culture of Bacillus typhosus to each tube, 
being careful to emulsify thoroughly on the side of the test-tube according to the technic 
given. This does not materially alter the degree of dilution. Prepare the culture control 
as usual. 
3. Incubate and examine tubes as in the previous experiment. 
What are the advantages and disadvantages of this method? 
Experiment 58.—Macroscopic Agglutination Reaction Employing 
Killed Culture: 
1. Prepare a formalized culture of Bacillus typhosus as described in the text after the 
Dreyer method. 
2. Prepare a series of dilutions of typhoid immune serum and conduct this experiment 
according to the macroscopic technic, using the dead instead of a living bacillary emulsion. 
(a) What are the advantages of using a killed culture? 
(b) Is this method as delicate as when living cultures are used? 
(c) Is spontaneous agglutination likely to occur? 
(d) What constitutes a satisfactory killed culture for this reaction? 
Experiment 59.—Group Agglutination: 
1. Prepare five series of test-tubes containing 1 c.c. of dilutions ranging from 1 : 20 to 
1 : 640 of a typhoid immune serum. 
2. To the first series add 1 c.c. of a thirty-six-hour bouillon culture of Bacillus typhosus; 
to the second series, Bacillus paratyphosus “B”; to the third, Bacillus paratyphosus “A”; 
to the fourth, Bacillus enteritidis (Gartner); to the fifth, Bacillus coli. The dilutions are thus 
doubled. Prepare culture controls of all five cultures, and be careful that tubes are properly 
labeled. 
3. Incubate for one hour and record the reactions. 
4. This experiment may be repeated with the serum of a typhoid convalescent patient. 
(a) In which series of tubes is agglutination most complete? 
(b) Discuss the question of specificity of the agglutinins. 
(c) How do you explain group agglutination? 
(d) Are group agglutinins present to the same degree as the main agglu- 
tinin. How may the group agglutinins be eliminated? 
(g) Of what value is the agglutination reaction in showing the biologic 
relationship of bacteria? 
(f) How would the agglutination reaction be used in the diagnosis of 
an unknown micro-organism? BACTERIAL AGGLUTININS 
1131 
Experiment 60.—Pro-agglutination—Agglutinoids : 
1. This very important phase of agglutination may have been noted in the previous 
experiments, especially if old immune serums were used, when there is a possibility tha 
agglutinin has degenerated in agglutinoids. Agglutinoids having a stronger affinity than 
agglutinin for the agglutinogen, and having lost the agglutinophore group, produce little or 
no agglutination and prevent the action of agglutinin until diluted out of action in the higher 
serum dilutions. In this way agglutination may be poor or absent in low and present in 
higher dilutions, an important practical fact to be remembered. 
2. The action of agglutinoids may be seen by conducting a macroscopic test with an old 
immune serum. 
3. Repeat the tests with old and fresh immune sera in dilutions of 1 : 40, 1 :80, 1 :60, 
1 :320, using the macroscopic and microscopic technic, as this phenomenon of pro-agglutina- 
tion is not infrequently found in the routine Widal reaction in typhoid fever. 
Experiment 61.—The Effect of Heat Upon Agglutinins: 
1. Dilute 0.1 c.c. of typhoid immune serum possessing a titer of 1 :100 or higher, with 
4.9 c.c. of saline solution (1 : 50). 
2. Place 1 c.c. in each of 5 small test-tubes. 
3. Heat No. 1 in a water-bath at 55° C. for one-half hour; No. 2 at 60° C. for one-half 
hour; No. 3 at 70° C. for one-half hour, and No. 4 at 80° C. for one-half hour. No. 4 is not 
heated. 
4. After cooling add 1 c.c. of typhoid suspension to each tube, mix, and incubate for an 
hour. 
5. In a series of 6 small test-tubes place 1 c.c. of 1 : 50 dilution of the antityphoid serum. 
6. Place five of the tubes in a water-bath at 70° C.; remove No. 1 after ten minutes and 
Nos. 2, 3, 4, and 5 after fifteen, thirty, forty-five, and sixty minutes respectively. 
7. After cooling add 1 c.c. of typhoid emulsion to each tube and No. 6 (unheated control). 
8. In tube No. 7 place 1 c.c. of saline solution and 1 c.c. typhoid emulsion (control). 
9. Place all tubes in water-bath at 38° C. for one hour and read results. 
(a) Does heat influence typhoid agglutinin? 
Cb) Are the bacterial agglutinins equally affected by heat? 
(c) Are natural agglutinins more vulnerable to heat than immune 
agglutinins? 
(d) Are immune agglutinins easily destroyed by heat, desiccation, 
chemical germicides, etc.? 
Experiment 62.—The Influence of Electrolytes on Agglutination: 
1. Place 10 c.c. of formalized typhoid culture for agglutination tests and 10 c.c. of typhoid 
serum 1 : 50 in a centrifuge tube; incubate one hour and centrifuge very thoroughly. 
2. Discard the supernatant fluid and suspend the bacilli in 5 c.c. of distilled water, break- 
ing up the clumps as thoroughly as possible. 
3. Set up the following in small test-tubes: 
No. 1: suspension 1 c.c. + 1 c.c. distilled water. 
No. 2: suspension 1 c.c. + 0.9 c.c. distilled water + 0.1 c.c. of 10 per cent, solu- 
tion of sodium chlorid. 
No. 3: suspension 1 c.c. + 0.9 c.c. distilled water + 0.1 c.c. of 0.2 per cent, solu- 
tion of copper sulphate. 
No. 4: suspension 1 c.c. + 0.9 c.c. distilled water + 0.1 c.c. of 0.8 per cent, solu- 
tion of copper sulphate. 
4. Mix each tube and incubate at 37° C. for one hour. 
(a) In which tubes has agglutination taken place, and why? 
(b) Discuss the relation of electrolytes to the mechanism of agglutina- 
tion. 
Experiment 63.—Absorption of Agglutinins: 
1. Set up macroscopic agglutination tests with a rabbit antityphoid serum and Bacillus 
typhosus; the titer should be known beforehand so that the range of dilutions may be satis- 
factory. 1132 
EXPERIMENTAL INFECTION AND IMMUNITY 
2. Repeat, using Bacillus paratyphosus B. 
3. Repeat, using Bacillus coli. . 
4. In 2 and 3 lower dilutions of serum will be required, the final dilutions being 
1 :10, 1 : 20, 1 : 30, 1 :40, etc. 
5. Prepare a thick suspension of typhoid bacilli by washing off three agar slants. Place 
5 c.c. of saline in the first and remove the growth; pour suspension into the second, etc. Heat 
at 60° C. for one hour. 
6. Prepare similar suspension of Bacillus paratyphosus B and Bacillus coli. 
7. Place 1 c.c. of typhoid serum in each of three small centrifuge tubes. To No. 1 add 4 
c.c. of the typhoid suspension; to No. 2, 4 c.c. of the paratyphoid suspension, and to No. 3, 
4 c.c. of the colon suspension Mix each tube, incubate one hour, and stand in refrigerator 
until next day. 
8. Centrifuge each tube very thoroughly and secure the supernatant fluids which repre- 
sent 1 :5 dilutions of the absorbed serum. Discard the sediments. 
9. Repeat the agglutination tests with the three micro-organisms and compare the 
results observed with plain serum. 
(a) Does absorption of typhoid serum with Bacillus coli remove any of 
the typhoid agglutinin? Does absorption with B. paratyphosus A remove 
any, and if so, why? 
(b) What are “major” and “minor” agglutinins? 
(c) Did absorption with typhoid bacilli remove all of the agglutinins, 
and if not, why? 
(d) Of what practical value is this method? 
Experiment 64.—Agglutination in Vivo: 
1. Cultivate Type I pneumococci in suitable fluid medium for eighteen to twenty-four 
hours. Centrifuge 200 c.c. very thoroughly and suspend the cocci in 10 c.c. of saline solu- 
tion, endeavoring to obtain an even suspension free of clumps. 
2. Inject 5 c.c. of the suspension intravenously into a rabbit. 
3. Immediately remove a little blood from the heart with a syringe and prepare smears. 
Remove specimens again five, ten, and fifteen minutes later and prepare smears. Stain 
and examine for pneumococci. 
4. Inject 5 c.c. of the suspension in a second rabbit and immediately thereafter 2 c.c. 
of Type I antipneumococcus serum; both injections intravenously. 
5. Immediately remove blood from the heart (within less than one minute after the in- 
jection of serum) and prepare films. Repeat two, three, five, ten, and fifteen minutes later. 
Stain and examine the films. 
(a) Are there evidences of agglutination in the blood of the first rabbit? 
In the second? 
(b) Are there any evidences of phagocytosis? 
(c) Discuss the agglutinins in relation to immunity. 
Experiment 65.—Agglutination for the Differentiation of Pneumo- 
cocci: 
1. Sputum should be secured from a patient with Type I or Type II pneumonia. In the 
absence of these , sputa may be prepared by adding 1 c.c. of Type I pneumococci to 1 c.c. of 
saliva and injecting a mouse intraperitoneally with 1 c.c. of the mixture. A second mouse 
should be injected with a similar preparation of Type II pneumococci. 
2. The balance of the tests, including mouse necropsies, collecting peritoneal exudates, 
etc., should be conducted as described in the text. 
(a) Does the mouse possess natural immunity to pneumococci? 
(b) Why is the mouse employed in this test? 
(c) Why is the bile control tube included? 
(d) What other micro-organisms have been divided into types on the 
basis of immunologic reactions? 
(e) Of what practical value is the agglutination test for the differentia- 
tion of pneumococci? PRECIPITINS 
1133 
Experiment 66.—Hemagglutinins: 
1. Secure 0.1 c.c. of antihuman hemolysin prepared by giving a rabbit a series of injec- 
tions of washed human erythrocytes. Inactivate the serum by heating to 55° C. for a half- 
hour. Dilute 1 :40 by adding 3.9 c.c. salt solution. 
2. Prepare 10 c.c. of a 1 per cent, suspension of washed human erythrocytes. 
3. Into a series of six small test-tubes place 0.1, 0.2, 0.4,0.6, 0.8, and 1 c.c. of the diluted 
immune serum; add 1 c.c. of suspension of blood-cells and 1 c.c. of salt solution to each tube. 
As a control, place 1 c.c. of corpuscle suspension and 1 c.c. of normal salt solution into a 
seventh tube. 
4. Shake gently and place in the incubator for an hour. 
5. Prepare a series of hanging-drop preparations for microscopic examination. 
• 
(a) Describe agglutination of red corpuscles. 
(b) Are the clumps easily broken up? 
(c) What would have occurred if complement were present? 
(d) Why was it necessary to heat the fresh immune serum? Is it neces- 
sary to heat an old immune serum? 
Experiment 67.—Agglutination Tests for the Grouping of Human 
Corpuscles Before Blood Transfusion: 
1. Collect in test-tubes containing 1 c.c. of citrate solution a few drops of blood from 
each of several persons. 
2. Working with known Group II and Group III sera determine the group of which each 
person’s corpuscles belong according to a method described in the text, employing a micro- 
scopic technic. 
(a) Into how many groups may human corpuscles be divided? 
(b) What is the difference between the Jansky and Moss classifications? 
(c) How may human sera for these tests be secured and preserved? 
(<d) Of what practical importance are these tests? What lesions and 
symptoms may develop in agglutination and hemolysis of corpuscles in 
vivo? 
(e) How is hemolysis detected in these microscopic tests? 
(/) Discuss the influence of age in .relation to the isohemagglutinins. 
Experiment 68.—Direct Agglutination and Hemolysin Tests Pre- 
liminary to Blood Transfusion: 
1. Secure a small amount of blood in a small dry test-tube and an additional few drops 
in a little citrate solution from each of 4 individuals. 
2. Designate one person as a patient and the others as donors. Separate the sera by 
centrifuging. 
3. Test the serum of the patient for agglutinins and hemolysins for the corpuscles of the 
donors using both a microscopic and macroscopic technic. 
4. Test the serum of each donor against the corpuscles of the patient in the same manner. 
(a) What are the advantages of this direct matching of bloods? 
(b) May hemolysins be present in serum in the absence of demonstrable 
amounts of agglutin? 
(c) In what diseases are isohemolysins particularly likely to be found? 
(d) What are auto-agglutinins and autohemolysins? 
Experiment 69.—Titration of a Precipitin Serum: 
PRECIPITINS 
1. Secure 1 c.c. of antihorse immune serum prepared by immunizing a rabbit with normal 
horse-serum. 
2. Secure 1 c.c. of normal-horse serum and place 2 c.c. of the following dilutions made 
with normal salt solution into a series of six narrow test-tubes: 1 :100, 1 :500, 1 :1000, 
1 :2000, 1 : 5000, and 1 :10,000. 1134 
EXPERIMENTAL INFECTION AND IMMUNITY 
3. To each tube add 0.1 c.c. of the immune serum. The tubes must not be shaken. 
For exact technic consult the text. 
4. Place tubes in an incubator at 37° C. and observe every thirty minutes for two hours 
and again after standing in a refrigerator overnight. 
(a) Describe the phenomenon of precipitation. 
(b) What are the requisites of a satisfactory precipitin serum? 
(c) What are the requisites of a satisfactory preparation of the antigen 
for this reaction? 
(d) Discuss the delicacy of this reaction. 
% 
Experiment 70.—Titration of a Precipitin Serum: 
Repeat this titration, using antihuman immune serum and normal human serum. 
Experiment 71.—Specificity of Precipitins: 
1. Secure 1 c.c. of each clear antihorse and antihuman serums. 
2. Secure 1 c.c. each of normal horse, normal human, and normal guinea-pig serum. 
Prepare 1 : 100 dilutions with normal salt solution; set up the following in long narrow test- 
tubes: 
Tube 1: 2 c.c. of normal horse serum (1 :100) + 0.1 c.c. antihorse serum. 
Tube 2: 2 c.c. of normal horse serum (1 :100) +0.1 c.c. antihuman serum. 
Tube 3: 2 c.c. of normal human serum (1 :100) +0.1 c.c. antihuman serum. 
Tube 4: 2 c.c. of normal human serum (1 :100) + 0.1 c.c. antihorse serum. 
Tube 5: 2 c.c. of normal guinea-pig serum (1 : 100) +0.1 c.c. antihorse serum. 
Tube 6: 2 c.c. of normal guinea-pig serum (1 : 100) +0.1 c.c. antihuman serum. 
Tube 7: 2 c.c. of normal salt solution + 0.1 c.c. antihorse serum (control). 
Tube 8: 2 c.c. of normal salt solution + 0.1 c.c. antihuman serum (control). 
3. Do not shake the tubes; place them in the incubator at 37° C.; inspect every thirty 
minutes for two hours. 
(a) Discuss the question of specificity of precipitins. 
(b) Of what practical value are precipitin reactions? 
(c) Enumerate the chief points in the technic of a precipitin reaction 
in the differentiation of proteins. 
Experiment 72.—Forensic Blood Test: 
1. Secure from the instructor two pieces of muslin or gauze containing respectively a 
stain of human and horse blood. These are to be numbered and the source of each stain known 
only to the instructor. Secure 1 c.c. each of normal human and horse serum and dilute 
1 :1000. 
2. Prepare extracts of each stain as described in the text. 
3. Secure 1 c.c. of clear antihuman and antihorse immune serum. 
4. Set up the following in long narrow test-tubes. 
Tube 1: 2 c.c. of extract No. 1 in dilution of 1 :1000 + 0.1 c.c. of antihuman serum. 
Tube 2: 2 c.c. of extract No. 1 in dilution of 1 : 1000 + 0.1 c.c. of antihorse serum. 
Tube 3: 2 c.c. of extract No. 2 in dilution of 1 :1000 + 0.1 c.c. of antihorse serum. 
Tube 4: 2 c.c. of extract No. 2 in dilution of 1 :1000 + 0.1 c.c. of antihuman serum. 
Tube 5: 2 c.c. of normal human serum in dilution 1 :1000 + 0.1 c.c. antihuman 
serum (control). 
Tube 6: 2 c.c. of normal horse serum in dilution of 1 :1000 + 0.1 c.c. antihorse 
serum (control). 
Tube 7: 2 c.c. of normal salt solution + 0.1 c.c. of antihuman serum. 
Tube 8: 1 c.c. of normal salt solution + 0.1 c.c. of antihorse serum. 
5. Do not shake tubes; place in the incubator at 37° C. and inspect every thirty minutes 
for two hours and after standing in the refrigerator overnight. 
(a) Inspect the controls. Are the antiserums potent and satisfactory? 
Why is it advisable to have these controls? 
(b) Are you able to diagnose the source of each stain? PRECIPITINS 
1135 
(c) In a medicolegal test, if these reactions were negative, how would 
you proceed further in your efforts to establish the identity of a particular 
stain? 
Experiment 73.—Bacterial Precipitins: 
1. Inoculate two flasks each containing 50 c.c. of sterile neutral bouillon with cultures 
of Bacillus typhosus and Bacillus coli and cultivate at 37° C. for three weeks. Filter cultures 
through a sterile Berkefeld filter until clear. 
2. Into a series of four small test-tubes place 2 c.c. of the following dilutions of the 
typhoid filtrate (precipitinogen) undiluted—1 : 2, 1 :4, and 1 : 10. Prepare a second series 
of tubes with similar amounts of the same dilutions of Bacillus coli filtrate. 
3. Add 0.1 c.c. of potent typhoid immune serum to all tubes of both series. A serum 
with high agglutinin titer will be satisfactory. 
4. Place 2 c.c. of the undiluted typhoid and coli filtrates in separate tubes as controls. 
Prepare an additional control by placing 0.1 c.c. of the typhoid immune serum in 2 c.c. 
normal salt solution. 
5. Observe the tubes at the end of a half-hour, and after one, two, and six hours. 
(a) Describe a positive bacterial precipitin reaction. 
(b) Has a precipitate formed with the Bacillus coli filtrate? With what 
micro-organism would typhoid immune serum be likely to produce a pre- 
cipitate? 
(c) Discuss the relative delicacy of precipitation and agglutination 
reactions in the differentiation of bacteria. 
Experiment 74.—Bacterial Precipitinogens in Inflammatory Exudate: 
1. Repeat the inoculation of mice with sputum containing Type I pneumococci as in 
Experiment 65. 
2. After securing the peritoneal exudate centrifuge thoroughly and conduct precipitin 
tests with Type I antipneumococcus serum as described in the text. 
3. If urine from a case of pneumonia due to Type I pneumococci is available, repeat the 
tests as described in the text. 
(a) What is the source of the precipitinogens in exudate and urine? 
(b) Discuss the practical applications. 
Experiment 75.—Thermoprecipitinogens : 
1. Centrifuge 50 c.c. of a broth culture of Type I pneumococci. Suspend the cocci in 
4.5 c.c. distilled water and add 0.5 c.c. of antiformin. Boil for several minutes or until the 
cocci are dissolved as shown by translucency. Add several drops of 1 per cent, alcoholic 
solution of phenolphthalein and then just enough N/l hydrochloric acid to discharge the 
color. Add 15 c.c. of 95 per cent, alcohol, mix, and centrifuge a half-hour later. Add about 
2 c.c. of saline solution to the sediment and boil for five minutes. Centrifuge and use the 
supernatant fluid (precipitinogen). 
2. In a small test-tube place 0.5 c.c. of Type I antipneumococcus serum and carefully 
overlay with 0.5 c.c. of the precipitinogen. Observe for the formation of a ring at the line 
of contact. 
(a) Discuss the influence of heat upon precipitinogens in general with 
special reference to precipitin tests for anthrax, pneumonia, and the differen- 
tiation of meats. 
Experiment 76.—Noguchi Globulin Reaction: 
1. Secure 1 c.c. each of several cerebrospinal fluids from cases of paresis, tuberculous men- 
ngitis, serous meningitis, and normal human subjects. 
2. Conduct this floccule-forming reaction after the technic given in the text. 
3. Make total cell counts on each fluid. 
(a) Explain the appearance of a positive reaction. 
(b) What constitutes a negative reaction? 1136 
EXPERIMENTAL INFECTION AND IMMUNITY 
(c) What is the diagnostic value of this reaction in syphilis? 
(d) What other value has this reaction in diagnosis? 
(e) What relation does this reaction bear to total cell counts? 
(f) Explain the mechanism of the reaction. 
Experiment 77.—Flocculation Reactions in Syphilis: 
1. Secure 1 c.c. of a cholesterolized alcoholic extract of beef heart and quickly add 3 c.c. 
of saline solution. Invert the mixture several times to secure an opalescent emulsion. 
2. Secure sera from several syphilitic individuals yielding strongly positive Wassermann 
reactions; also serum from a healthy non-syphilitic individual. 
3. Place 0.3 c.c. of each serum in small test-tubes and add 0.05 c.c. of the diluted extract. 
4. Mix the contents of each tube gently and incubate at 38° C. for sixteen to eighteen 
hours, making a preliminary inspection at the end of one hour for evidences of flocculation 
(see Chapter XXIV). 
(a) Have flocculi developed in any of the tubes? 
(b) Discuss the probable mechanism of flocculation in syphilis. 
Experiment 78.—Resistance of Red Blood-corpuscles to Salt Solution 
of Various Tonicities. Non-specific Hemolysis: 
1. Arrange a series of dilutions of sodium chlorid in distilled water ranging from 0.7 
to 0.2 per cent, as described in Chapter XX. 
2. Secure human blood as described and conduct the tonicity tests. 
3. Repeat at the same time, using defibrinated sheep blood. 
(a) What is the appearance of hemolyzed blood? 
(b) What is the meaning of the term hemolysis? 
(c) Why is this called non-specific hemolysis? 
(,d) What does a normal or physiologic salt solution mean? 
(e) What is the importance of using a normal salt solution in hemolytic 
experiments? 
(f) How is hemolysis produced by hypotonic solutions? 
(g) What would be the objections of using a hypertonic salt solution? 
(h) How is an isotonic salt solution prepared? 
(i) What is the meaning of the terms minimal and maximal resistance 
of red blood-corpuscles? 
(J) Of what practical value is this test? 
(k) What are the average points of minimal and maximal resistance of 
red blood-corpuscles of healthy human beings. 
(/) What influence has defibrination on these values? 
AMBOCEPTORS AND COMPLEMENTS—HEMOLYSINS AND HEMOLYSIS 
Experiment 79.—Saponin Hemolysis: 
1. Prepare a 1 : 1000 solution of Merck’s saponin in physiologic saline solution. 
2. Collect human blood and conduct the tests as described in Chapter XX using un- 
washed and washed corpuscles. Employ 16 dilutions of saponin varying from 1 :20,000, 
1 : 22,000, 1 : 24,000, etc., to 1 : 50,000. 
3. Repeat, using defibrinated sheep blood. 
4. Repeat, using citrated guinea-pig blood. 
(a) Is saponin hemolysis specific or non-specific? 
(b) What is the mechanism of saponin hemolysis? 
(c) Does the presence of serum protect the corpuscles against saponin? 
Experiment 80.—Bacterial Hemolysis: 
1. Secure an eighteen- to twenty-four-hour broth culture of Staphylococcus aureus; also 
one of a hemolytic streptococcus. AMBOCEPTORS AND COMPLEMENTS—HEMOLYSINS AND HEMOLYSIS 
1137 
2. Place 0.1, 0.2, 0.4, 0.6, 0.8, 1, and 2 c.c. of each culture in two sets of small test-tubes. 
3. Arrange a third set of tubes carrying similar amounts of sterile broth. 
4. To each tube add 1 c.c. of a 1 per cent, suspension of washed sheep corpuscles and 
sufficient saline solution to make the total volume 3 c.c. in each tube. 
5. Put up a control with 1 c.c. of corpuscle suspension and 2 c.c. of saline solution. 
6. Mix well and incubate in a water-bath at 38° C. for one hour. 
7. Read the results; then place tubes in a refrigerator overnight and record the results. 
(a) Has hemolysis been produced? 
(b) Are the bacterial hemolysins specific or non-specific? 
(c) May the culture-medium itself be hemolytic, and if so, what con- 
stituents may produce hemolysis? 
Experiment 81.—The Hemolytic Activity of Acids and Alkalies: 
The object of this experiment is to demonstrate the hemolytic activity 
of acids and alkalies as bearing upon the question of clean glassware in 
hemolytic work, and particularly complement-fixation tests. 
1. Into a series of six test-tubes place increasing amounts of a solution of hydrochloric 
acid prepared by diluting 1 c.c. of the normal solution with 99 c.c. of normal salt solution: 
0.2, 0.4, 0.6, 0.8, 1, and 2 c.c. respectively. 
2. Into a second series of six test-tubes place increasing amounts of a solution of sodium 
hydroxid prepared by diluting 1 c.c. of the normal solution with 99 c.c. of normal salt solu- 
tion: 0.2, 0.4, 0.6, 0.8, 1, and 2 c.c. respectively. 
3. Make the total volume in each tube equal 2 c.c. with the addition of normal salt 
solution. 
4. To each tube and a normal salt solution control add 1 c.c. of a 2| per cent, suspension 
of washed sheep cells. Mix and stand aside for an hour or two. 
(a) Has hemolysis occurred in any of the tubes? 
(b) What are the smallest amounts of acid and alkali producing complete 
hemolysis in this experiment? 
(c) Of what practical significance are these results? 
id) How should test-tubes be prepared for hemolytic work? 
Experiment 82.—Serum Hemolysis in Vitro: 
1. Secure 2 c.c. of blood from the ear of a rabbit which has received at least two intra- 
venous injections of sheep cells. Separate the serum and divide into two portions. Inactivate 
one portion (A) by heating in a water-bath for a half-hour at 56° C. 
2. Place 0.2 c.c. of the fresh unheated serum of portion (B) in a test-tube. Likewise the 
same amount of the heated serum (A) in a tube. Add 1 c.c. of a 1 per cent, suspension of 
washed sheep cells to each tube and sufficient salt solution to bring the total volume to 3 c.c. 
As a control, place 1 c.c. of corpuscle suspension and 3 c.c. salt solution in a tube. Shake 
gently, and incubate for an hour in a water-bath at 38° C. 
(a) Which tube shows hemolysis? 
(b) What substances are concerned in serum hemolysis? 
(c) Why did hemolysis occur with the unheated and not with the heated 
serum? What is the meaning of inactivation of a serum? 
(d) Why is this called serum hemolysis? 
(e) What is the appearance of the control tube? Why is this control 
included? What would have happened if a hypotonic salt solution had 
been used? 
Experiment 83.—Serum Hemolysis in Vivo: 
1. Immunize a rabbit with three intravenous injections of 5 c.c. each of a 10 per cent, 
suspension of washed cat erythrocytes in sterile salt solution. Give the injections each day; 
four days after the last injection the rabbit serum is titrated and usually contains a fair amount 
of anticat hemolysin. 
2. Heat 5 c.c. of the immune serum at 56° C. for thirty minutes and inject into the ex- 
ternal jugular vein of a cat. 1138 
EXPERIMENTAL INFECTION AND IMMUNITY 
3. Place the animal in a metabolic cage and collect the urine. 
4. After two or three days, autopsy, removing the spleen, liver, and kidneys. Place in 
5 per cent, formalin and prepare and stain sections with hematoxylin and eosin and Giemsa 
solution. 
(a) Has blood destruction occurred? What are the evidences? 
(b) Would hemolysis occur in the test-tube with a heated immune 
serum? If not, why not? 
(e) How then do you explain hemolysis in vivo with a heated serum? 
(d) Examine sections of the spleen, liver, and kidneys. Are there any 
evidences of phagocytosis of blood-cells, focal necrosis, and nephritis? 
Explain the probable mechanism of the production of these changes. 
Experiment 84.—Titration of Antisheep Hemolysin: 
1. Secure 1 or 2 c.c. of blood from the ear of a rabbit which has been immunized with 
injections of sheep cells. Separate the serum and inactivate by heating to 56° C. for a half- 
hour. Dilute 1 :100 by adding 0.1 c.c. serum to 9.9 c.c. physiologic saline solution. 
2. Prepare 40 c.c. of a 5 per cent, dilution of fresh guinea-pig serum to be used for com- 
plement. Prepare 40 c.c. of a 2\ per cent, suspension of washed sheep cells by adding 1 c.c. 
of corpuscles to 39 c.c. of physiologic saline solution. 
3. Proceed with the titration as given in the text on page 408. If the smallest dose, 
viz., 0.1 c.c. of the 1 : 100 dilution, completely hemolyzes the corpuscles, it will be necessary 
to retitrate with a dilution of 1 :1000. 
(a) Is it necessary to use exactly the same amounts of complement and 
corpuscles in all tubes, and if so, why? 
(b) If hemolysis did not occur at all in this experiment, what factors 
may be at fault? 
(c) If the corpuscle control were completely hemolyzed, what deduc- 
tion would you draw? 
(d) What is the hemolytic unit of this serum? 
Experiment 85.—Quantitative Factors in Serum Hemolysis: 
1. Having determined the hemolytic unit of the above antisheep immune serum, proceed 
as follows: 
2. To a series of four test-tubes add an amount of immune serum equaling one, two, 
three, and five hemolysin units respectively. To each tube add 0.5 c.c. of the 1 : 20 dilution 
of complement serum. This amount is just half the dose of complement used in titrating the 
hemolysin. Add 1 c.c. of a 2\ per cent, suspension of sheep cells to each tube and sufficient 
salt solution. 
3. In a second series of four test-tubes place one-half an amboceptor unit and the follow- 
ing amounts of diluted complement serum: 1, 2, 3, and 4 c.c. Add 1 c.c. of the suspension 
of sheep corpuscles and sufficient salt solution to make the total volume in each tube about 
equal. Incubate for one hour and read the results. 
4. In a third series of four test-tubes place one amboceptor unit, 1 c.c. of complement 
serum (1 : 20), and the following amounts of corpuscle suspension: 1, 2, 3, and 4 c.c. Add 
sufficient salt solution to make the total volume in each tube about equal. 
5. Prepare a corpuscle control with 1 c.c. of suspension and 4 c.c. normal salt solution. 
6. Shake all tubes gently and incubate for two hours at 37° C. 
(a) Can an excess of hemolysin make up for a deficiency in complement? 
(b) Can an excess of complement make up for a deficiency in hemolysin? 
(c) What happens when an excess of corpuscle suspension is used? 
(<d) Discuss the importance of quantitative factors in serum hemolysis. 
Experiment 86.—Role of Hemolysin amd Complement in Hemolysis: 
1. Prepare a 2| per cent, suspension of washed sheep corpuscles. Bleed a healthy 
guinea-pig under ether anesthesia, separate the serum, and dilute 1 :20 with normal salt 
solution (complement). Secure antisheep hemolytic serum (inactivated) whose hemolytic 
titer is known (consult instructor). A MBOCEP TORS A ND COMP LEM EN TS—HEMOL YSINS A ND HEMOL YSIS 
1139 
2. Proceed to set up a series of four test-tubes as follows: 
Tube 1: 1 c.c. corpuscle suspension + sufficient normal salt solution to make the 
total volume about 3 c.c. 
Tube 2: 1 c.c. corpuscle suspension + 1 c.c. complement serum (1 : 20) + sufficient 
salt solution. 
Tube 3: 1 c.c. corpuscle suspension + two hemolytic units of antisheep hemolysin 
+ salt solution. 
Tube 4: 1 c.c. corpuscle suspension + 1 c.c. complement serum + two units of 
hemolysin + salt solution. 
3. Shake each tube gently and incubate at 37° C. for one or two hours. 
(a) In which tube has hemolysis occurred? Explain the results. 
(b) What does inactivation of a serum mean? 
(c) Could some hemolysis occur, using the complement serum and 
corpuscles without immune hemolysin? 
(d) Could hemolysis occur in the absence of complement? 
Experiment 87.—Specificity of Hemolysins: 
1. Prepare a 2| per cent, suspension of washed sheep corpuscles and a 1 per cent, sus- 
pension of washed human corpuscles. Bleed a guinea-pig under ether, separate serum, and 
dilute 1 : 10 with normal salt solution. Secure antisheep and antihuman hemolytic serums 
(inactivated) whose hemolytic titers are known. 
2. Proceed as follows: 
Tube 1: 1 c.c. sheep corpuscles + 1 c.c. complement (1 : 10) + two units of anti- 
sheep hemolysin + salt solution up to 3 c.c. 
Tube 2: 1 c.c. sheep corpuscles + 1 c.c. complement (1 :10) + two units of anti- 
human hemolysin. 
Tube 3: 1 c.c. human corpuscles + 1 c.c. complement (1 :10) + two units of anti- 
human hemolysin. 
Tube 4: 1 c.c. human corpuscles + 1 c.c. complement (1 :10) + two units of anti- 
sheep hemolysin. 
Tube 5: 1 c.c. sheep corpuscles + 2 c.c. salt solution (control). 
Tube 6: 1 c.c. human corpuscles + 1 c.c. salt solution (control). 
3. Shake tubes gently and incubate for an hour or two. 
(a) In which tubes has hemolysis occurred? 
(b) What does this experiment teach as to the specificity of these ambo- 
ceptors? 
(c) Discuss the specificity of hemolytic and bacteriolytic amboceptors. 
(d) If hemolysin occurs in tube No. 2, how do you explain? 
Experiment 88.—Thermostability oe Hemolysins: 
1. Secure antisheep hemolysin so diluted that 1 c.c. contains one unit. 
2. Prepare a 25 per cent, suspension of sheep corpuscles. 
3. Secure fresh guinea-pig complement serum and dilute 1 : 20. 
4. Place two units of hemolysin into each of six test-tubes. Place these in a water-bath 
and heat at 60° C. 
5. Remove a tube from the water-bath after five, ten, fifteen, thirty, forty-five minutes, 
and one hour. After allowing them to cool, add 1 c.c. corpuscle suspension and 1 c.c. com- 
plement serum (1 :20). 
6. Place two units of hemolysin in each of four test-tubes. Heat No. 1 at 56° C. for 
thirty minutes; No. 2 at 60° C.; No. 3 at 70° C., and No. 4 at 80° C. After cooling add 1 c.c. 
of 1 : 20 complement and 1 c.c. of corpuscle suspension to each tube. 
7. Set up a tube containing two units of amboceptor (not heated) + 1 c.c. corpuscle sus- 
pension -f 1 c.c. complement serum (control). Shake each tube gently and incubate for one 
hour at 37° C. 
(a) Discuss the stability of hemolysins to heat, age, and germicides. 
(b) Discuss the relative thermostability of hemolysins and complements. 1140 
EXPERIMENTAL INFECTION AND IMMUNITY 
Experiment 89.—Resistance of Hemolysins and Complement to 
Desiccation: 
1. Secure 1 c.c. of antisheep hemolytic serum and evenly distribute in an appropriate- 
sized piece of filter-paper. 
2. Repeat, using 1 c.c. of guinea-pig serum complement. 
3. Permit both papers to dry at room temperature. Place both in an incubator over- 
night. 
4. Next day or several days later, titrate the papers. Cut each into strips 5 mm. wide. 
5. Place increasing lengths of the hemolysin paper in six test-tubes: 2, 5, 10, 15, 20, and 
30 mm. Add 1 c.c. of 1 :20 complement and 1 c.c. of 2\ per cent, suspension of sheep cells 
in each tube. Mix and incubate in a water-bath for one hour. 
6. Arrange a second series of tubes carrying similar amounts of complement paper. To 
each tube add the smallest hemolytic unit of hemolysin in paper. (Note: If the 2 mm. length 
produces complete hemolysis, it may be used as the dose of hemolysin.) Add 1 c.c. of 2\ 
per cent, corpuscle suspension and 1 c.c. of saline solution to each tube. Mix and place 
in a water-bath for one hour. 
Discuss the relative stability of hemolysins and complement to 
desiccation. 
Experiment 90.—Mechanism of Hemolysin Action: 
1. Secure antisheep hemolysin so diluted that 1 c.c. contains one hemolytic unit. 
2. Prepare a 2\ per cent, suspension of sheep corpuscles. 
3. Secure fresh guinea-pig serum and dilute 1 : 20. 
4. Into two centrifuge tubes place two units of hemolysin and 1 c.c. of corpuscle sus- 
pension. Mix and place one tube in a glass of cracked ice, leaving the other at room tem- 
perature. After one hour centrifuge both. Pipet the supernatant fluids into two separate 
test-tubes. 
5. To the sedimented corpuscles in both centrifuge tubes add 1 c.c. diluted comple- 
ment serum and 2 c.c. of saline solution. Mix well. Incubate tubes for one hour at 37° C. 
6. To the supernatant fluid of each tube add 1 c.c. corpuscle suspension and 1 c.c. of di- 
luted complement serum. Mix. Incubate tubes for an hour. 
(a) In which tubes has hemolysis occurred? 
(b) How do you explain the results? 
(c) Discuss the prevailing views of hemolysin action. 
Experiment 91.—A Further Study of the Mechanism of Hemolysin 
Activity: 
1. In a centrifuge tube place two units of antisheep hemolysin and 2 c.c. of fresh guinea- 
pig complement diluted 1 : 20 . Place in a glass of cracked ice until the mixture is thoroughly 
chilled. Then add 2 c.c. of a 2J per cent, suspension of sheep cells (also chilled). Mix and 
keep at a low temperature for an hour, centrifuge rapidly, and pipet the supernatant fluid 
to a separate test-tube. 
2. In a second centrifuge tube place 2 c.c. of diluted complement and 2 c.c. corpuscle 
suspension. Keep at room temperature for one hour, centrifuge, and pipet the supernatant 
fluid into a test-tube. 
3. Proceed as follows: To 2 c.c. of the supernatant fluid from the first centrifuge tube 
add 1 c.c. corpuscle suspension; to the remaining 2 c.c. add 1 c.c. corpuscle suspension and 
two units of hemolysin; to the sedimented corpuscles add 2 c.c. of diluted complement serum. 
4. To 2 c.c. of the supernatant fluid from the second centrifuge tube add 1 c.c. corpuscle 
suspension; to the remaining 2 c.c. add 1 c.c. corpuscle suspension and two units of hemolysin. 
To the sedimented corpuscles add 1 c.c. of complement and 1 c.c. of saline solution. 
5. Shake all tubes gently and incubate for one hour at 37° C. 
(a) In which tubes has hemolysis occurred? 
(b) Does complement unite directly with corpuscles? 
(c) What evidence have you that hemolysin unites directly with cor- 
puscles? 
(d) What is meant by the term “sensitizing corpuscles”? 
(e) Does hemolysis occur at a low temperature? 
(/) What temperature best favors hemolysis? AMBOCEPTORS AND COMPLEMENTS—HEMOLYSINS AND HEMOLYSIS 
1141 
Experiment 92.—Natural Hemolysins: 
1. Place 0.2 c.c. of serum from each of 4 normal rabbits in four small test-tubes. To each 
add 1 c.c. of 1 per cent, sheep corpuscle suspension. 
2. Secure six fresh human sera. 
3. Place 0.2 c.c. of each in six test-tubes and add 1 c.c. of 1 per cent, sheep corpuscle sus- 
pension to each. 
4. Repeat, using a suspension of washed guinea-pig corpuscles. 
5. Incubate the three sets of tubes in a water-bath for one hour and read the results. 
(a) What are natural or normal hemolysins? 
(b) Are they specific? 
(c) What natural hemolysins may be formed in human sera? 
(d) What influence may natural hemolysins have on complement-fixa- 
tion tests? 
(e) Are natural hemolysins for human corpuscles present in human sera? 
What are they called? 
(/) Are there any evidences of hemagglutination? 
Experiment 93.—Resistance of Natural Hemolysins to Heat: 
1. Secure 1 c.c. of each of six fresh human sera employed in Experiment 92 and place 
0.2 c.c. of each in six small test-tubes respectively. Place these in a water-bath at 55° C. for 
one-half hour. 
2. Place 0.2 c.c. of each serum in a second set of small tubes and heat in the same manner. 
3. To each tube of the first set add 1 c.c. of 1 : 20 complement and 1 c.c. of 2\ per cent, 
sheep corpuscle suspension. 
4. To each tube of the second set add 1 c.c. of 1 :20 complement and 1 c.c. of 2\ per 
cent, guinea-pig corpuscle suspension. 
5. Mix well and place in a water-bath at 38° C. for one hour. 
Discuss the resistance of different natural hemolysins to heat. 
Experiment 94.—Removal of Natural Hemolysins from Human 
Sera: 
1. Secure 1 c.c. of each of four human sera known to contain natural antisheep hemolysin. 
2. Heat each serum at 55° C. for fifteen minutes. 
3. Place 0.2 c.c. of each in test-tubes and add 1 c.c. of 1 : 20 complement and 1 c.c. of 2\ 
per cent, sheep corpuscle suspension. Incubate one hour. 
4. To the balance of each serum add 2 drops of washed undiluted sheep corpuscles after 
the last centrifuging; mix well and centrifuge after standing thirty minutes. 
5. Place 0.2 of each supernatant serum in small test-tubes and add 1 c.c. of 1 : 20 com- 
plement and 1 c.c. of sheep-corpuscle suspension to each. Mix and incubate for one hour. 
(a) What are the evidences of removal of hemolysin? 
(b) Why were the sera heated? 
(c) How would you remove hemolysin from an active or unheated 
serum? 
Experiment 95.—Rapidity of Absorption of Hemolysin by Corpuscles: 
1. In each of a series of four small test-tubes or centrifuge tubes place 5 units of anti- 
sheep hemolysin (so diluted that the unit is 0.2 c.c.). 
2. To each tube add 1 c.c. of 2\ per cent, sheep-cell suspension and mix well. 
3. Immediately and rapidly centrifuge tube No. 1; centrifuge No. 2 at the end of five 
minutes, No. 3 after ten minutes, and No. 4 after fifteen minutes. 
4. Place the supernatant fluids in test-tubes and add 1 c.c. of 1 :20 complement and 1 
c.c. of a 21 per cent, sheep corpuscle suspension to each. 
5. Prepare a control carrying 2 units of hemolysin, 1 c.c. of 1 : 20 complement, and 1 c.c. 
of corpuscle suspension. Also a second control carrying 1 c.c. of corpuscle suspension and 
2 c.c. of saline solution. 
6. Mix contents of each tube and place in a water-bath at 38° C. for one hour. 1142 
EXPERIMENTAL INFECTION AND IMMUNITY 
(a) Discuss the rapidity of absorption of hemolysin by hemologous 
corpuscles. 
(b) Would antisheep hemolysin be absorbed by human corpuscles? 
(c) Discuss the influence of temperature upon the rate of hemolysin 
absorption. 
Experiment 96.—Amount of Hemolysin Absorbed by Corpuscles: 
1. Titrate antisheep hemolysin with 1 c.c. of 1 :20 complement, 1 c.c. of 2\ per cent, 
sheep cells, 1 c.c. of saline solution, and water-bath incubation for one hour. 
2. After the unit has been determined (and the hemolysin should be so diluted that the 
unit is not more than 0.2 c.c.) place 2, 4, 6, 8, 10, 15, and 20 units in six centrifuge tubes 
respectively. 
3. To each tube add 1 c.c. of 2| per cent, sheep cell suspension and mix well. 
4. After standing one-half hour in the incubator, centrifuge each tube and pipet the 
supernatant fluids to test tubes. Discard the sediments of corpuscles. 
5. To each tube add 1 c.c. of 1 : 20 complement and 1 c.c. of 2\ per cent, suspension of 
sheep cells. Mix and place in a water-bath for one hour. 
(a) How much hemolysin was absorbed? 
(b) Discuss the mechanism of hemolysin absorption. 
Experiment 97.—Dissociation of Hemolysin: 
1. In a test-tube place 1 c.c. of 5 per cent, sheep corpuscles and 20 units of hemolysin; 
mix and place in a water-bath for thirty minutes. 
2. Centrifuge and set aside the supernatant fluid. This is Tube A. 
3. Wash the cells three times, using 3 c.c. of saline solution for each washing, and set 
aside the supernatant fluid of the last washing. This is Tube B. 
4. To the washed corpuscles add 3 c.c. of saline solution; mix well and place in water- 
bath for one hour. Centrifuge and set aside supernatant fluid, which is Tube C. 
5. To the corpuscles add 1 c.c. of 1 : 10 complement and 4 c.c. of salt solution; Tube D. 
6. To Tubes A, B, and C add 1 c.c. of 5 per cent, corpuscles and 1 c.c. of 1 : 10 com- 
plement. 
7. Place all tubes in water-bath for one hour. 
Are there any evidences of dissociation of hemolysin? 
Experiment 98.—Hemolytic Complement: 
1. After giving a rabbit three intravenous injections of 5 c.c. of a 10 per cent, suspension 
of washed sheep cells at intervals of three days, remove 3 c.c. of blood from an ear and sepa- 
rate the serum. Dilute the serum 1 : 10 with salt solution. 
2. Into four test-tubes place 1 c.c. of a 2§ per cent, suspension of sheep corpuscles and 
increasing amounts of the above dilution of rabbit serum (must be fresh—not over twelve 
hours old) as follows: 0.4, 0.8, 1, and 2 c.c.; add sufficient salt solution. Shake and incubate 
for one hour. 
(a) Has hemolysis occurred? 
(b) How do you explain the reaction? 
(c) From where was the complement derived? 
(d) Are complements found in the bloods of all animals? 
(e) Discuss the question of the multiplicity of complements. 
Experiment 99.—Inactivation and Reactivation of Complement: 
1. Heat 1 c.c. of the 1 :10 dilution of fresh rabbit immune serum used in the preceding 
experiment to 56° C. for a half hour (in a water-bath). 
2. In a test-tube place 1 c.c. of this heated serum and 1 c.c. of a 2§ per cent, suspension 
of sheep cells and sufficient salt solution. Shake gently and incubate for one hour. 
(a) Does hemolysis occur? 
(b) How do you explain the result? 
(c) What is meant by inactivation of complement? 
(d) Is complement thermostabile? AMBOCEPTORS AND COMPLEMENTS—HEMOLYSINS AND HEMOLYSIS 
1143 
3. Add to the tube 1 c.c. of fresh guinea-pig serum diluted 1 : 10 and reincubate for one 
hour. 
(a) Has hemolysis occurred? 
(b) How do you explain the result? 
(c) What is meant by reinactivation? 
Experiment 100.—Thermolability of Complement: 
1. Secure antisheep amboceptor the hemolytic unit of which is known by previous titra- 
tion. 
2. Prepare a 2| per cent, suspension of washed sheep corpuscles. 
3. Into a series of four test-tubes place the following: 
Tube 1: 1 c.c. of fresh guinea-pig serum diluted 1 : 20. 
Tube 2: 1 c.c. of guinea-pig serum (1 : 20) which has been kept at room temperature 
for two days. 
Tube 3: 1 c.c. of guinea-pig serum (1 : 20) three days old. 
Tube 4: 1 c.c. of guinea-pig serum (1 : 20) five days old. 
Add 1 c.c. of a per cent, suspension of sheep corpuscles, two hemolytic units of anti- 
sheep amboceptor, and sufficient normal salt solution to each tube. Shake gently and 
incubate at 37° C. for one hour. 
4. Into a series of six test-tubes place 1 c.c. of fresh guinea-pig complement serum diluted 
1 : 20. Place these in a water-bath at 56° C. At intervals of ten, twenty, thirty, forty, fifty, 
and sixty minutes remove a tube, add 1 c.c. corpuscle suspension, two units of amboceptor, 
and sufficient salt solution. Shake gently each tube and incubate for one hour at 37° C. 
(a) Record the results. Does hemolytic complement deteriorate readily 
at room temperature? 
(b) What are complementoids? 
(c) What practical significance has this experiment? Should a comple- 
ment serum be fresh when used in hemolytic work? 
(<d) Is complement thermolabile? How long does it take to destroy a 
diluted complement at 56° C.? 
(ie) In inactivating a serum why do we not use a higher temperature? 
Experiment 101.—Effect of Acids, Alkalies, and Filtration Upon Com- 
plement: 
1. In each of three small test-tubes place 1 c.c. of 1 :20 complement serum. 
2. To No. 1 add 1 c.c. of 1 :300 dilution in saline solution of N/l solution of sodium 
hydroxid; to No. 2 add 1 c.c. of a 1 :300 dilution of N/l solution of hydrochloric acid. To 
No. 3 add 1 c.c. of saline solution. Mix and place in a water-bath for thirty minutes. Then 
add 2 units of antisheep hemolysin and 1 c.c. of 2\ per cent, sheep corpuscle suspension to 
each tube; mix and reincubate for one hour. 
3. Filter 5 c.c. of 1 :20 complement through a small Kitasato or other earthen filter. 
Place 1 c.c. of the filtrate in a test-tube; add 2 units of hemolysin and 1 c.c. of corpuscle sus- 
pension. Mix and incubate one hour. 
4. Set up a corpuscle control. 
(a) Does sodium hydroxid and hydrochloric acid destroy complement? 
(b) What would be the effect of using larger amounts? 
(c) What relation do these results bear to the preparation of glassware 
for complement fixation tests? 
(d) Is complement-filtrable? 
Experiment 102.—Titration of Hemolytic Complement: 
1. Prepare 20 c.c. of complement 1 : 20; prepare a 2| per cent, suspension of sheep cells, 
2. To a series of 10 test-tubes add increasing doses of diluted complement serum: 0.1. 
0.2, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9, 1, and 1.5 c.c.; add 2 units of hemolysin (determined by pre- 
vious titration), 1 c.c. of corpuscle suspension, and sufficient salt solution to bring the total 
volume in each tube to 3 c.c. 
3. Shake gently and incubate for one hour at 37° C. 1144 
EXPERIMENTAL INFECTION AND IMMUNITY 
(a) Record the results. What is the complement unit of this serum? 
(b) Is the complement content of the serums of different animals con- 
stant? Does it vary in animals of the same species? In the same animal 
at different times? 
Experiment 103.—Complement Activity of Different Animal Sera: 
1. Secure 0.5 c.c. of fresh human, guinea-pig, and rabbit serum. Dilute each with 4.5 
c.c. of saline solution (1 :10). 
2. Test each serum for complement and natural antisheep hemolysin by placing 0.5 c.c. 
in test-tubes with 1 c.c. of 2| per cent, sheep-cell suspension. Mix and incubate one hour. 
3. Arrange three series of six test-tubes each. Into the first set place 0.1, 0.2, 0.4, 0.6, 
0.8, and 1 c.c. of the diluted human serum. Place similar amounts of the guinea-pig serum in 
the second set and rabbit serum in the third set. 
4. To each tube add 2 units of antisheep hemolysin (based upon titration with 1 c.c. of 
1 :20 guinea-pig complement and 1 c.c. of 2| per cent, sheep corpuscle suspension) and 1 
c.c. of 2 5 per cent, sheep corpuscle suspension 
5. Set up a corpuscle control. 
6. Mix contents of each tube and place in a water-bath for one hour. 
(a) Are the different sera of equal hemolytic activity? 
(b) Are the differences due to variation in amounts of complement or 
natural antisheep hemolysin, or both? 
(c) What animal serum is best adapted for complement in complement- 
fixation tests? 
Experiment 104.—Preservation of Complement: 
1. Dissolve 2 grams of chemically pure sodium chlorid in 10 c.c. of fresh guinea-pig 
serum. Keep in a refrigerator. 
2. Immediately after the solution has been prepared, dilute 1 c.c. with 19 c.c. of distilled 
water, which gives 20 c.c. of a 1 : 20 dilution of complement in isotonic solution. 
3. Titrate the diluted complement with 2 units of hemolysin, 1 c.c. of 2 5 per cent, sheep- 
cell suspension, and sufficient saline, to make 3 c.c. Water-bath incubation one hour. 
4. Repeat the titration three days later, again at the end of a week, and thereafter at 
weekly intervals. Wassermann tests may also be performed. 
(a) Discuss the influence of temperature upon the preservation of 
complement serum. 
(b) What other methods may be employed? 
(c) How long may complement serum be preserved? 
(d) What are the signs of deterioration? 
(e) Does hemolytic activity deteriorate sooner or later than fixability 
by syphilis antigen and antibody? 
(/)What are the practical advantages of using preserved complement? 
Experiment 105.—Phenomenon oe Complement Fixation: 
This experiment is introduced here to show the specific Bordet-Gengou 
phenomenon of complement fixation. The exact technic of complement- 
fixation reactions as conducted for the diagnosis of syphilis and other infec- 
tions will be given in subsequent exercises. 
1. Dilute 1 c.c. of guinea-pig serum complement with 9 c.c. of saline solution (1 :10) 
and titrate in doses of 0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9, and 1 c.c. against 2 units of 
hemolysin and 1 c.c. of 2§ per cent, sheep corpuscle suspension. 
2. Secure 1 c.c. of an antigonococcus and a normal serum and heat each at 56° C. for 
thirty minutes. 
3. Secure an emulsion of gonococci which is now called the antigen; for the proper dose 
to use consult the instructor, or a typhoid immune serum and typhoid antigen may be em- 
ployed. ANTIGENS FOR THE WASSERMANN REACTION 
1145 
4. Proceed as follows: 
Tube 1: 0.2 c.c. antigonococcus serum + dose of antigen + 2 units of complement 
+ normal salt solution. 
Tube 2: 0.2 c.c. antigonococcus serum + 2 units of complement + normal salt 
solution (serum control). 
Tube 3: 0.2 c.c. normal serum + dose of antigen + 2 units of complement + 
normal salt solution. 
Tube 4: 0.2 c.c. normal serum + 2 units of complement + normal salt solution 
(serum control). 
Tube 5: dose of antigen + 2 units of complement + normal salt solution (antigen 
control). 
Tube 6: 2 units of complement + normal salt solution (hemolytic control). 
Tube 7: 1 c.c. corpuscle suspension + normal salt solution (corpuscle control). 
Plug tube with cotton as it is finished. 
5. Shake all tubes gently and incubate for one hour at 37° C. 
6. Add 2 units of hemolysin and 1 c.c. corpuscle suspension to all tubes except corpuscle 
control. Shake gently and reincubate for one hour. 
(a) Examine tubes and record your results. 
(b) Why was the serum control on each serum necessary? 
(c) Why was the hemolytic system control necessary? 
(d) Why was the antigen control necessary? 
(e) What is meant by non-specific complement fixation? 
(/) What is meant by specific complement fixation? Explain the 
phenomenon. Which tube of this series shows specific complement fixation, 
and why? Is there any evidence of non-specific fixation in any of the tubes? 
If so, what bearing would this have on the results of this test? 
(g) What is meant by complement deviation? By complement fixation? 
(h) Upon what does the specificity of complement fixation depend? 
ANTIGENS FOR THE WASSERMANN REACTION 
Experiment 106.—Preparation of Antigens for the Wassermann Re- 
action: 
1. Prepare 50 c.c. of an alcoholic extract of beef heart. 
2. Prepare an extract of acetone-insoluble lipoids from 20 grams of beef heart. 
3. Prepare 50 c.c. of a cholesterinized alcoholic extract of human heart. 
4. Prepare 50 c.c. of the author’s new antigen. 
(a) Are antigens for the Wassermann syphilis reaction specific? 
(b) Upon what does the high specificity of the Wassermann reaction 
depend? 
(c) What constitutes a specific antigen for the Wassermann reaction? 
(d) Discuss the important relation of antigen to the Wassermann re- 
action. 
Experiment 107.—Methods of Titration of Antigens: 
While the above extracts are in the course of preparation, stock extracts 
may be used for titration. After the student has finished the above four 
extracts, they are titrated in a similar manner. 
1. Dilute 1 c.c. of an alcoholic extract of beef heart in a test-tube 1 :10 by slowly adding 
9 c.c. of normal saline solution. Dilute a cholesterinized extract (1 :20) by slowly adding 
9.5 c.c. salt solution to 0.5 c.c. of extract. 
2. Conduct the anticomplementary titration of each (page 473). 
3. Conduct the antigenic titration of each (page 475). 
(a) What is meant by the anticomplementary dose of an antigen? 
(b) What is meant by the antigenic dose of antigen? 
(c) Why are these titrations so important? 1146 
EXPERIMENTAL INFECTION AND IMMUNITY 
(d) In a complement-fixation test what would be the result if the antigen 
were used in the anticomplementary dose? What would be the result if 
the antigen were used in less than the antigenic dose? 
(e) What constitutes a satisfactory antigen for any complement-fixation 
test? 
(/) In conducting a diagnostic reaction, should the antigen be used in 
exactly one antigenic dose or double or treble this amount? Why and 
under what conditions would this be a safe procedure? 
(g) Why should the antigen be diluted slowly with salt solution? 
(h) Why is a serum control so important in these titrations? What 
other controls are used, and why? 
(i) Which is more important, the anticomplementary or antigenic 
titration, and why? 
Experiment 108.—Titration of Antigen (Continued): 
Titrate the writer’s new antigen for hemolytic, anticomplementary, and antigenic 
activities according to the new technic (page 483). 
What substances in tissue extracts are largely responsible for hemoly- 
tic activity? For anticomplementary activity? For antigenic activity? 
Experiment 109.—Anticomplementary Action of Sera: 
1. Secure 1 c.c. of five different specimens of human sera which have been standing in the 
laboratory for one, two, four, six, and ten days respectively. The last specimen should be 
intentionally contaminated with a culture of staphylococcus. 
2. Divide each serum into two portions and heat portion “B” to 56° C. for a half-hour. 
3. Place 0.2 c.c. of each specimen (unheated) in a series of five test-tubes. 
4. Place 0.2 c.c. of each specimen (heated) in a second series of five test-tubes. 
5. To each tube add 1 c.c. of complement 1 : 20 and sufficient salt solution to bring the 
total volume to 3 c.c. Shake gently and incubate for a half-hour. Add 2 units of hemolysin 
and 1 c.c. corpuscle suspension to each tube and incubate for one hour. 
6. A hemolytic system control (complement, corpuscles, and amboceptor) should be in- 
cluded; also a corpuscle control. 
(a) Record your results. Are any of the sera anticomplementary? 
How do you determine this? 
(b) What is meant by the terms anticomplementary action of a serum? 
(c) What significance has this phenomenon in complement-fixation 
work? 
(d) Are anticomplementary bodies thermolabile or thermostabile? 
May both exist? Under what conditions are each most likely to be present? 
\e) How are thermolabile anticomplementary bodies removed or their 
influence overcome? When a serum is three or more days old, what is the 
chief object of heating it before it is used in a complement-fixation test? 
(/) Does complement keep for three days? 
(g) Is it possible to remove the thermostabile anticomplementary action 
of a serum? Could such a serum be used in a complement-fixation reaction? 
How is this condition of the serum detected? 
(,h) Under what conditions is a serum likely to become anticomple- 
mentary in action? Is a sterile serum likely to become anticomplementary? 
Experiment 110.—Wassermann Reaction (First Method): 
1. Secure four specimens of blood: one from a known syphilitic person; the second from 
a known normal person, the third and fourth specimens for diagnosis. Also a specimen of 
cerebrospinal fluid. 
2. Conduct a Wassermann reaction with each serum after I\\z first method, using an antigen 
of alcoholic extract of syphilitic liver or beef heart. ANTIGENS FOR THE WASSERMANN REACTION 
1147 
(a) Record your results. In case the hemolytic system control was 
not completely hemolyzed, what may be the reasons and what influence 
would this result have on interpreting the reactions? 
(b) In case the antigen control was not completely hemolyzed, what 
influence would this result have on the reactions? 
(c) If a serum control were incompletely hemolyzed, what influence 
would this have on the results of the test? 
(d) Is this method a quantitative reaction? 
(e) What is the nature of the antibody in a syphilitic serum? 
(j) How are these reactions recorded? How should they be reported 
to a clinician? 
Experiment 111.—Author’s Modification of the Wassermann Reaction 
Employing Multiple Antigens: 
Conduct a Wassermann reaction with each of five specimens of serum after the second 
method, using an alcoholic extract of syphilitic liver or beef heart, an extract of acetone- 
insoluble lipoids, and an alcoholic extract of heart reinforced with cholesterin. One of these 
sera should be from a syphilitic person (positive control) and one from a normal person (nega- 
tive control). 
(a) What are the advantages of using more than one antigen? 
(b) Discuss the relative value of these three antigens. 
(c) Under what conditions may a cholesterinized extract be used? 
Experiment 112.—Non-specific Complement Fixation: 
1. Secure specimens of sera from normal rabbits and heat a portion of each at 55° C. for 
thirty minutes. 
2. Prepare an antigen of staphylococci or typhoid bacilli by washing off twenty-four-hour 
agar slant cultures with saline solution; heat for one hour and titrate for anticomplementary 
activity. Use an amount equal to one-third the anticomplementary unit in these tests. 
3. Conduct Wassermann tests after the second method, employing three tissue and 
the bacterial antigens. 
4. Heat a fresh portion of each serum at 60° to 62° C. for thirty minutes and repeat the 
tests. 
(a) Discuss non-specific complement fixation versus anticomplementary 
.activity. 
(,b) Were any of these sera or antigens anticomplementary? 
(c) What animal sera are known to give non-specific reactions? 
(d) What influence does heating the sera have on the phenomenon? 
Experiment 113.—Non-specific Complement Fixation by Human Sera; 
Proteotropic Reactions: 
1. Secure the sera of 12 healthy non-syphilitic individuals. 
2. Use each serum unheated in dose of 0.1 c.c. and conduct Wassermann tests with three 
tissue antigens including a cholesterolized extract, after the Noguchi method. 
3. Heat a portion of each serum at 55° C. for fifteen minutes and repeat the tests. 
(a) May unheated non-syphilitic human sera yield non-specific com- 
plement-fixation reactions? 
(,b) What relation does that kind of antigen employed bear to these 
reactions? 
(c) Does heating human serum increase or decrease the tendency for 
non-specific reactions? 1148 
EXPERIMENTAL INFECTION AND IMMUNITY 
Experiment 114.—The Influence of Temperature and Duration of 
Primary Incubation Upon Complement-fixation Reactions: 
1. Prepare a 2\ per cent, suspension of sheep corpuscles; also a 1 : 10 dilution of com- 
plement. Titrate the hemolysin in doses of 0.1, 0.15, 0.2, 0.25, 0.3, 0.35, 0.4, 0.45, and 0.5 
c.c., using 0.5 c.c. of corpuscle suspension, 0.3 c.c. of complement, and sufficient saline solution 
to make 2 c.c. 
2. Now titrate the complement in dose of 0.1, 0.15, 0.2, 0.25, 0.3, 0.35, 0.4, 0.45, and 0.5 
c.c., using two units of hemolysin, 0.5 c.c. corpuscle suspension, and saline solution to make 
2 c.c. 
3. Determine the influence of temperature upon complement destruction as follows: Arrange 
a series of five test-tubes. Place 1, 2, 3, 4, and 5 units of complement in each respectively, 
with saline to make 1.5 c.c. Place in a water-bath for one hour for primary incubation. Add 
2 units of hemolysin and 0.5 c.c. of corpuscle suspension and reincubate one hour. Record 
the unit. 
Repeat, except that the primary incubation is conducted by placing the tubes in a re- 
frigerator for eighteen hours at 6° to 10° C. Next day add 2 units hemolysin and 0.5 c.c. of 
the corpuscle suspension (carried over in the refrigerator) and incubate one hour. Record 
the unit. 
4. Determine the influence of temperature upon the anticomplementary action of antigen as 
follows: Repeat the complement titrations in exactly the same manner with both kinds of 
primary incubation, except that 5 units of antigen are added to each tube. 
5. Determine the influence of temperature upon the anticomplementary action of serum: 
Repeat the complement titrations in exactly the same manner with both kinds of primary 
incubation, except that 0.1 c.c. of a mixture of strongly positive syphilitic sera (heated to 
55° C. for fifteen minutes) is added to each tube. 
6. Determine the influence of temperature upon the specific fixation of complement: Ar- 
range a series of eight test-tubes; place 0.1 c.c. of the syphilitic serum and 5 units of the 
antigen in each. Add increasing amounts of complement to the tubes respectively: 2, 4, 6, 
8, 10, 12, 15, and 20 units. Add sufficient saline solution to equalize the tubes. Give a pri- 
mary incubation of one hour in a water-bath. Add 2 units of hemolysin and 0.5 c.c. of cor- 
puscle suspension and reincubate one hour. 
Repeat, using a primary incubation of eighteen hours at 6° to 10° C. 
7. Now subtract the total amount of complement destroyed, the amount fixed by antigen 
alone and by serum alone from that fixed by the mixture of serum and antigen in the water- 
bath primary incubations. The difference expresses the amount of complement fixed 
specifically. 
8. Make the same calculations for the titrations employing the cold or refrigerator method 
of incubation. 
(a) What influence has temperature and duration of primary incubation 
upon the destruction of complement? Upon the anticomplementary 
activity of antigen alone and serum alone? Upon the specific fixation of 
complement by mixtures of serum and antigen? 
(.b) Why may a hemolytic system and complement-fixation test satis- 
factory for a warm primary incubation be unsatisfactory with a prolonged 
cold incubation? 
Experiment 115.—The Author’s New Complement-fixation Test Based 
Upon the Results of Studies in the Standardization of Technic: 
Conduct tests with each of four specimens of serum after the third method, using the 
author’s new antigen. One of these serums should be a positive control and one a normal 
or negative control. 
(a) Why is a mixture of guinea-pig sera employed as complement? 
(b) Why are the specimens of guinea-pig blood placed in the incubator 
for one hour or in a refrigerator overnight, before separation of the sera? 
(c) What is the advantage of titrating the complement in the presence 
of the dose of antigen? 
(<d) What is the advantage of titrating the hemolysin each time the 
tests are done? 
(<e) Why are the doses of various reagents made 0.5 c.c. instead of 
employing smaller amounts? ANTIGENS FOR THE WASSERMANN REACTION 
1149 
(/) Why are the sera heated at 55° C. for fifteen minutes? 
(g) Why are the mixtures of serum and antigen allowed to stand before 
the addition of complement? 
(h) What are the advantages of a primary incubation of eighteen hours 
at 6° to 8° C. followed by ten minutes in a water-bath at 38° C.? What is the 
disadvantage? 
(i) What are the advantages of a quantitative test? 
(j) What is the advantage of using a color scale? 
Experiment 116.—Author’s Modification of the Wassermann Reaction 
Employing Varying Amounts of Complement: 
Conduct a Wassermann reaction with a known positive, a known negative, and two 
unknown serums after the fourth method, using an extract of acetone-insoluble lipoids. 
What are the main advantages of this technic? 
Experiment 117.—Titration of Antihuman Hemolysin: 
1. Secure 0.1 c.c. of inactivated antihuman rabbit hemolytic serum and dilute 1 :40 
by adding 3.9 c.c. normal saline solution. 
2. To a series of six small test-tubes add increasing amounts of diluted hemolysin as 
follows: 0.1, 0.2, 0.4, 0.6, 0.8, and 1 c.c., add 0.1 c.c. of 40 per cent, complement, 1 c.c. of 1 
per cent, human corpuscle suspension, and sufficient saline solution to bring the total volume 
to 2 c.c. 
3. Incubate for one to two hours, shaking the tubes once or twice during this time. 
(a) Are there any evidences of hemagglutination? 
(b) What constitutes the unit of hemolysin? 
(c) Is an antihuman hemolytic system more delicate than an antisheep 
system in complement-fixation work? 
(d) What are the advantages and disadvantages of using an antihuman 
hemolytic system? 
Experiment 118.—Technic oe the Noguchi Modification: 
1. Secure five specimens of blood not over twenty-four hours old, including at least one 
positive and one negative serum. 
2. Conduct the Noguchi reaction with each serum in the unheated or active state, using 
an antigen of acetone-insoluble lipoids. (See page 496). 
3. Conduct the reactions with the same antigen, using the serums after heating to 56° C. 
for a half-hour. 
(a) Record your results. Are they equal in both series? 
(b) What is meant by proteotropic reaction? 
(c) What effect has heat upon syphilis reagin? 
(d) Does an active serum react more delicately than an inactivated 
one? 
(e) In performing this test with unheated serum what precautions are 
to be observed? 
Experiment 119.—Preparation and Titration of Bacterial Antigens: 
1. Cultivate typhoid bacilli upon slants of agar for forty-eight to seventy-two hours. 
2. Wash these off with sufficient saline solution to give a cloudy suspension. Shake with 
glass beads for one hour and heat at 60° C. for one hour. Add 0.05 c.c. of a 5 per cent, solu- 
tion of phenol for each cubic centimeter of antigen. 
3. Titrate the antigen for anticomplementary activity and use it in dose corresponding 
to one-quarter this amount. 
4. Conduct complement-fixation tests with the human sera if sera from typhoid fever 
patients are available. If not, employ rabbit typhoid immune serum heated to 60° to 62° C. 
for thirty minutes. If the author’s new technic is employed the maximum dose should be 
not more than 0.025 c.c. to avoid non-specific reactions. 1150 
EXPERIMENTAL INFECTION AND IMMUNITY 
(a) Describe different methods for the preparation of bacterial antigens. 
(b) In what diseases of human beings and the lower animals are comple- 
ment-fixation tests of practical diagnostic value. 
(c) Why does complement fixation fail in some bacterial diseases? 
(d) Why is complement fixation generally stronger in syphilis than in 
other bacterial diseases? 
(e) Why is a sensitive technic especially desirable in bacterial comple- 
ment-fixation tests? 
Experiment 120.—Technic of the Gonococcus Reaction: 
1. Secure six specimens of serum from a genito-urinary clinic, particularly of men suf- 
fering with chronic gonococcus infections. Also a known positive and a known normal serum 
for controls. 
2. Conduct the reactions after one of the methods described. 
(a) Is the gonococcus reaction specific? 
(ft) In what cases of gonococcus infection is the reaction likely to be 
negative? Likely to be positive? 
(c) How soon after infection is the reaction likely to become positive? 
(d) Discuss the practical value of the gonococcus complement-fixation 
test. 
Experiment 121.—Tuberculosis Complement-fixation Test: 
Secure an antigen of tubercle bacilli and conduct complement-fixation tests with the 
following types of sera: 
(a) From individuals with acute pulmonary tuberculosis. 
Cb) From cases of chronic fibroid tuberculosis. 
(c) From healthy individuals giving negative v. Pirquet tuberculin reactions. 
(d) From syphilitic individuals giving negative v. Pirquet tuberculin reactions. 
(a) Is the tuberculosis complement-fixation test specific? 
(b) May the sera of syphilitic non-tuberculous individuals give positive 
reactions, and if so, why? 
(c) Discuss the practical value of the complement-fixation reaction in 
human tuberculosis. 
Experiment 122.—Complement Fixation in the Differentiation of 
Protein; Titration of Immune Sera: 
1. Secure 1 c.c. each of antihuman and antihorse-serum. Inactivate both. 
2. Secure 0.1 c.c. each of fresh human and horse-serum and dilute 1 :1000. 
3. Conduct an antigenic titration of each immune serum as described on page 561. 
(a) Explain the mechanism of this reaction. 
(b) Are protein amboceptors specific? 
(c) Discuss the question of inhibition of hemolysis due to a precipitin 
reaction. 
Experiment 123.—Technic oe the Forensic Complement-fixation 
Blood Test: 
1. Secure from the instructor two pieces of gauze stained respectively with human and 
horse blood. These are to be numbered and their identity known only to instructor. 
2. Secure 1 c.c. each of antihuman and antihorse-serum. 
3. Secure 0.1 c.c. of known human and horse-serum for controls. 
4. Proceed with the test for identifying these bloods as described in the text. 
5. Conduct precipitin tests at the same time. 
(a) Discuss the specificity of this test. 
(b) Is this test more delicate than the precipitin reaction? ANTIGENS FOR THE WASSERMANN REACTION 
1151 
(c) Which is the more liable to error in technic? 
(d) Discuss the applicability of the complement-fixation technic to the 
differentiation of proteins in general, as animal, vegetable, and bacterial 
proteins. 
Experiment 124.—Venom Hemolysis in Syphilis: 
1. Secure 1 c.c. of venom solution 1 :1000. 
2. Secure four specimens of blood (for technic see page 416) from cases of late secondary 
or tertiary syphilis and one specimen of normal blood. Also a known normal blood for 
control. 
3. Prepare the subdilutions of venom and conduct the tests after the technic described 
on pages 416 and 417. 
(a) Discuss the prevailing views regarding the mechanism of venom 
hemolysis. 
(b) Discuss the value of the venom test in the diagnosis of syphilis. 
(c) Discuss the main points in the technic. 
Experiment 125.—Pfeiffer Bacteriolytic Test: 
1. Secure 1 c.c. of serum from a rabbit which has been immunized with cholera bacilli. 
By working with cholera and a highly potent serum better results are secured, but as there 
is probably more danger connected with the handling of cholera cultures, typhoid may be 
substituted with fairly good results. Inactivate by heating to 56° C. for half an hour. 
2. Prepare a twenty-four-hour-old agar culture of a virulent strain of Bacillus cholera. 
3. Prepare a 1 :100 dilution of the immune serum and place 3 c.c. in a small test-tube. 
Add three loopfuls of culture and emulsify thoroughly. Inject a young guinea-pig intra- 
peritoneally with 2 c.c. of the emulsion. 
4. At intervals of ten, twenty, forty, and sixty minutes withdraw small amounts of exu- 
date and study bacteriolysis; prepare hanging-drop preparations which may be compared 
with a similar control on the culture; prepare smears of the culture and peritoneal exudate 
and stain with carbolthionin or carbolfuchsin. 
5. If desirable, the bacteriolytic titer of the serum may be determined after the method 
given in the text. 
(a) Describe the phenomenon of bacteriolysis. 
(b) Are bacteriolysis and hemolysis similar processes? Give the source 
of complement in this reaction. 
(c) Discuss the specificity of bacteriolytic reactions. 
(d) Discuss the practical value of the bacteriolytic test in the diagnosis 
of cholera. 
Experiment 126.—Microscopic Method op Measuring the Bacterio- 
lytic Power of the Blood 
This method may be employed for a rapid estimation of the bacteriolysin 
produced in the blood in response to an inoculation of typhoid vaccine. 
1. Secure a small quantity of the patient’s serum by collecting blood aseptically in a 
Wright capsule. About 1 c.c. of serum will be sufficient. Secure a control serum, preferably 
a “pooled serum,” in the same manner. Prepare dilutions of the patient’s serum in the fol- 
lowing manner: Place a series of six small test-tubes (sterile) in a rack; add 1 c.c. sterile broth 
to each tube; into the first tube place 1 c.c. of the fresh serum from the patient (1:2) dilu- 
tion); mix well and transfer 1 c.c. to the second tube; mix and transfer 1 c.c. to the third, and 
so on to the last tube, when 1 c.c. is discarded. 
2. Secure a twenty-four-hour culture of typhoid bacilli in neutral bouillon. 
3. Take a simple capillary pipet with a long stem, plugged with cotton and sterilized, 
and make a mark about 3 cm. from the end. Draw up the highest dilution of serum to the 
pencil mark, then a bubble of air, and finally an equal volume of culture. Thoroughly mix 
by carefully aspirating and driving out the contents on a hollow ground slide or in a small 
tubule. Aspirate the mixture into the middle portion of the stem and seal the pipet in a 
flame. Label with the final dilution (1 : 128). 
4. Proceed in the same manner with the remaining five dilutions of the patient’s serum, EXPERIMENTAL INFECTION AND IMMUNITY 
1152 
which then, mixed with an equal quantity of culture, equals 1 :64, 1 :32, 1 :16, 1 :8, and 
1 :4. Place these pipets in a large test-tube and label with the patient’s name and time when 
placed in the incubator. 
5. Prepare a similar set of pipets using the control serum. 
6. Prepare a culture control by mixing equal volumes of culture and sterile broth. 
7. Place all pipets in the incubator at 37° C. for two hours. 
8. Prepare hanging-drop preparations of each pipet by breaking off the tip and placing 
a drop of the contents (after mixing) on a cover slide and suspending in the usual manner. 
First examine the culture control and then each of the various dilutions, noting in each case 
the dilution in which there is the first bacteriolytic effect, and the dilution in which there is 
a complete effect; arrive at the bacteriolytic index by comparing the patient’s and the control 
blood just as the opsonic index is stated. 
(a) What are the first evidences of bacteriolysis? 
(b) Are endotoxins neutralized when bacteriolysis occurs? 
Experiment 127.—Method of Measuring the Bactericidal Activity 
of the Blood in Vitro (Method of Stern and Korte): 
With a culture of typhoid bacilli, rabbit typhoid immune serum, and rabbit complement 
serum, a test may be carried out after the method given in the text. 
(a) What main precautions are to be followed in this method? 
(b) Discuss complement deviation. 
(c) Discuss the practical value of this method. 
Experiment 128.—Bactericidal Power of Normal Serum: 
1. Secure 5 c.c. of blood from the vein of a person aseptically and separate the serum. 
2. Place 0.5 c.c. serum in a sterile test-tube and add 0.5 c.c. of 1 :1000 dilution of twenty- 
four-hour broth culture of Bacillus typhosus. 
3. Repeat, using 1 : 5000 dilution of culture. 
4. Repeat with the serum of an individual recently immunized against typhoid fever, 
using both dilutions of culture. 
5. Place 0.5 c.c. of the typhoid serum in a sterile test-tube and heat at 56° C. for a half 
hour; after cooling add 0.5 c.c. of 1 : 5000 culture. 
6. Place 0.5 c.c. sterile salt solution in a sterile test-tube and add 0.5 c.c. of 1 : 5000 cul- 
ture (culture control). 
7. Place 0.5 c.c. sterile salt solution in a sterile test-tube and add 0.5 c.c. of unheated serum 
(serum control). 
8. Place all tubes in a water-bath at 38° C. for two hours; then pour into each a tube 
(about 10 c.c.) of agar which has been boiled and cooled to 42° C. Mix well and pour into 
sterile Petri dishes. 
9. Incubate in an inverted position for twenty-four to forty-eight hours and count 
colonies. 
(a) What bacteriolysins may be found in normal human serum? 
(b) Does active immunization with typhoid vaccine increase typhoid 
bacteriolysin? 
(c) What relation may agglutinins bear to bacteriolysis? 
(d) What relation does complement bear to bacteriolysis? 
Experiment 129.—Bactericidal Activity of Whole Coagulable Blood 
(Heist-Lacy Method): 
Test the bactericidal activity of the whole blood of a chicken or pigeon for pneumococci; 
at the same time repeat the tests, employing human blood. The technic is described in the 
text. 
(a) Is natural immunity sometimes due to bactericidal substances in 
the blood? 
(b) Does the bactericidal activity of serum always require the presence 
of a complement? 
(c) What are the advantages and disadvantages of this technic? CYTOTOXINS 
1153 
Experiment 130.—Neisser-Wechsberg Phenomenon of Complement 
Deviation: 
1. Dilute 0.1 c.c. of a twenty-four-hour broth culture of Bacillus cholerae with 99.9 c.c. 
of sterile saline solution (1 : 1000). 
2. In a series of twelve small sterile test-tubes place 1 c.c. of sterile saline solution. To 
No. 1 add 1 c.c. of sterile anticholera serum; mix and transfer 1 c.c. to No. 2, and so on to 
No. 12, from which discard 1 c.c 
3. To each tube add 0.5 c.c. of the diluted culture and 0.3 c.c. of sterile guinea-pig serum. 
4. Set up Tube No. 13 with 0.5 c.c. culture and 1.3 c.c. of saline solution (culture control). 
5. Set up Tube No. 14 containing 0.5 c.c. of culture, 0.3 c.c. of guinea-pig serum, and 1 
c.c. of saline solution (complement control). 
6. Set up Tube No. 15 containing 1 c.c. of 1 :100 anticholera serum, 0.5 c.c. of culture, 
and 0.3 c.c. of saline solution. 
7. Mix all tubes well and incubate three hours in a water-bath at 38° C. 
8. Boil 15 tubes of agar and cool to 45° C.; arrange 15 sterile Petri dishes. 
9. Place 5 drops from each tube in a tube of agar, pour into Petri dishes, and incubate 
twenty-four to forty-eight hours. Count the colonies. 
(a) In which tubes occurred the greatest degree of bacteriolysis? 
(b) Does the degree of bacteriolysis bear a relation to the amount of 
immune serum employed? 
(c) Why is bacteriolysis less marked with the larger amounts of serum? 
(d) Is normal guinea-pig serum bacteriolytic for cholera bacilli? Is 
normal rabbit serum? 
Experiment 131.—Action of Nephrotoxin: 
CYTOTOXINS 
1. Prepare an emulsion of guinea-pig kidney as described in the text and immunize a 
rabbit. 
2. Inject a guinea-pig intravenously with 5 c.c. of rabbit immune serum per kilo of body 
weight. 
3. Place the animal in a metabolism cage, collect all urine, and carefully examine for 
albumin, casts, and hemoglobin; make quantitative albumin determinations. 
4. After three days autopsy and examine the kidneys and liver histologically. 
5. Heat the immune serum to 56° C. for thirty minutes and place increasing amounts in 
a series of test-tubes as follows: 0.01, 0.05, 0.08, 0.1, and 0.2 c.c.; add 1 c.c. of fresh guinea-pig 
complement serum 1 : 20, and 1 c.c. of 2\ per cent, suspension of washed guinea-pig erythro- 
cytes. Incubate for two hours. 
(a) Describe the changes occurring in the liver and kidneys. Give the 
prevailing views regarding the mechanism of their production. 
(b) Are there any evidences of a hemolytic action of the serum in vivo? 
In vitro? 
(c) How do you explain the presence of hemolysin in this immune serum? 
(d) Discuss the specificity of cytotoxins in general. 
(e) How many a hemolysin be removed from a nephrocytotoxic serum? 
(/) Have the cytotoxins any practical value in the treatment of disease? 
Experiment 132.—Spermatotoxin: 
1. Secure the testes of a large guinea-pig immediately after death and prepare a saline 
extract. Centrifuge very briefly if necessary for the removal of large particles. 
2. Examine a drop under the microscope for spermatozoa, many of which should be 
present. 
3. Inject a rabbit intraperitoneally at intervals of five days with increasing amounts, 
beginning with 2 c.c. 
4. After five or six doses secure fresh serum from this rabbit; also from a normal rabbit. 
5. Prepare a fresh suspension of guinea-pig testes containing many motile spermatozoa. 
6. On cover-glasses mix two loopfuls of immune serum with one loopful of suspension of 
spermatozoa. Suspend in a hanging-drop slide and vaselin the edges. Repeat, using the 
normal serum. Repeat, using saline solution instead of serum. 
7. Observe the slides at intervals over a period of several hours. 1154 
EXPERIMENTAL INFECTION AND IMMUNITY 
8. In a series of six small test-tubes place varying dilutions of immune serum in amounts 
of 1 c.c.; undiluted, 1 : 2, 1 :4, 1 : 8,1 :16, and 1:32. Add 1 c.c. of sprematozoa suspension 
to each. Put a control tube using saline solution in place of serum. Mix each tube and 
place in a water-bath for two hours. Observe for evidences of agglutination. Examine 
drops microscopically for loss of motility. 
Do the normal and immune sera affect the spermatozoa, and if so, in 
what manner? 
Experiment 133.—Technic of the Miostagmin Reaction: 
1. Secure a portion of dry pulverized cancer tissue and prepare an antigen after the 
technic described in the text. 
2. Titrate this antigen. 
3. Secure four specimens of serum: one from a known case of cancer; one from a normal 
person, and two for diagnosis. 
4. Proceed with the reaction as described in the text. 
(a) Discuss the principles of this reaction. 
(b) Discuss the practical value of this reaction in the diagnosis of cancer. 
Experiment 134.—Colloidal Gold Reaction: 
1. Prepare a reagent as described in the text. 
2. Secure four or more specimens of spinal fluid from the following types of cases: 
(a) From a case of paresis. 
(b) From a case of locomotor ataxia or some other type of tertiary syphilis. 
(c) From a non-syphilitic individual. 
(d) From a case of suppurative meningitis. 
3. Conduct tests with each fluid after the method given in the text. 
(a) What constitutes a satisfactory reagent? 
(b) Why must 0.4 per cent, sodium chlorid be employed for the dilutions 
of spinal fluid? 
(c) Would the presence of blood in the spinal fluid spoil the reaction? 
(d) What types of curves of precipitation of gold may be observed? 
(e) Discuss the practical diagnostic value of the test. 
ANAPHYLAXIS 
For experiments in anaphylaxis sensitize a series of animals as follows: 
(a) Give 7 young guinea-pigs an intraperitoneal injection of 0.01 c.c. 
horse-serum (1 c.c. of a 1 : 100 dilution). 
Cb) Give 3 young guinea-pigs an intraperitoneal injection of 0.01 c.c. 
of human serum. 
(c) Give 2 young guinea-pigs an intraperitoneal injection of 0.1 c.c. 
egg-albumen (1 c.c. of 1 : 10 dilution). 
(d) Give 2 rabbits 1 c.c. of horse-serum intravenously every three 
days for three doses. 
(e) Give a dog 10 c.c. of horse-serum subcutaneously. 
Experiment 135.—Anaphylaxis in the Guinea-pig. Specificity of Ana- 
phylaxis: 
1. Twelve days after the sensitizing injections proceed as follows: 
(a) Inject a guinea-pig sensitized to horse-serum with 1 c.c. horse-serum intra- 
peritoneally. 
(jb) Inject a guinea-pig sensitized to human serum with 0.1 c.c. of human serum 
intravenously. 
(c) Inject a guinea-pig sensitized to egg-albumen with 1 c.c. of diluted (1 :4) 
albumin intraperitoneally. 
(id) Inject a guinea-pig sensitized to horse-serum with 1 c.c. of diluted (1:4) egg- 
albumen. ANAPHYLAXIS 
1155 
(e) Inject a guinea-pig sensitized to human serum with 0.1 c.c. horse-serum intra- 
venously. 
(/) Inject a guinea-pig sensitized to horse-serum with 1 c.c. human serum intra- 
peritoneally. 
2. Carefully study the symptoms of anaphylaxis. After death autopsy the animals. 
(a) Describe acute anaphylactic shock in the guinea-pig. Does death 
always occur? 
(b) How soon after the intraperitoneal injection of the intoxicating dose 
do symptoms develop? After the intravenous? Why the difference? 
(c) Is anaphylaxis a specific reaction? 
(d) What are anaphylactogens? 
(e) Discuss the prevailing views regarding the mechanism of the anaphy- 
lactic reaction. 
(/) Describe the anatomic changes occurring in acute and fatal anaphy- 
laxis in the guinea-pig. Discuss the cause and nature of these changes. 
Experiment 136.—Anaphylaxis in the Dog: 
1. Five weeks after sensitization anesthetize a dog with ether and connect the carotid 
or femoral artery with a mercurial manometer and arrange for a kymographic tracing. 
2. After recording the normal blood-pressure tracing, inject 5 c.c. of homologous serum 
(horse) into a jugular vein. 
3. Record the blood-pressure. 
4. Study the coagulation time of the blood. 
5. Autopsy. 
(a) Describe the blood-pressure changes. To what may they be ascribed? 
(b) Is blood coagulation delayed? Give a reason for this change. 
(c) Do anatomic changes occur in anaphylaxis of the dog? 
Experiment 137.—Passive Anaphylaxis: 
1. Secure 1 c.c. of serum from a rabbit sensitized to horse-serum five weeks previously 
and inject 0.5 c.c. intraperitoneally into each of two guinea-pigs. 
2. Twenty-four and forty-eight hours later inject both animals intravenously with 0.2 
c.c. of horse-serum. 
(a) Are anaphylactic symptoms in evidence? 
(b) Discuss the nature of passive anaphylaxis. 
(c) If allowed to live, would these animals become sensitized to rabbit 
serum? 
Experiment 138.—Antianaphylaxis: 
1. Eight days after sensitizing a guinea-pig with horse-serum, inject 1 c.c. of horse-serum 
subcutaneously. 
2. Fifteen days after sensitizing a guinea-pig with horse-serum, inject 2 c.c. of serum into 
the rectum. 
3. Fifteen days after sensitizing a guinea-pig with horse-serum, inject 0.0001 c.c. serum 
subcutaneously every fifteen minutes for six doses. 
4. On the sixteenth day after sensitizing, inject a guinea-pig with 1 c.c. of horse-serum. 
If this animal develops anaphylactic shock, inject the other 3 guinea-pigs in a similar 
manner. 
(a) Do anaphylactic symptoms develop? If not, why not? 
(b) Discuss the importance of quantitative factors in producing anti- 
anaphylaxis. 
(c) Discuss the importance of anaphylaxis and antianaphylaxis in 
serum therapy. 
(d) What method is generally employed in the effort to induce an anti- 
anaphylactic state in serum therapy? 1156 
EXPERIMENTAL INFECTION AND IMMUNITY 
(e) What drug has been found experimentally of value in ameliorating 
or preventing the pulmonary changes of anaphylaxis? 
(/) Do anesthetics prevent anaphylaxis? 
(g) Under what conditions would you expect anaphylaxis in human 
beings? 
Experiment 139.—Local Anaphylactic Reactions; Anaphylaxis in the 
Rabbit: 
1. Shave the abdomen of the rabbit sensitized to horse-serum five weeks after the time of 
sensitization. 
2. Make two superficial abrasions over the upper portion of the abdomen; into one rub 
horse-serum, and in the other normal salt solution. 
Does a local reaction occur? If so, describe the lesion. 
3. Then inject 0.01 c.c. of serum subcutaneously. If acute anaphylactic death does 
not follow, observe the animal during the next twenty-four hours. 
(a) Does a local Arthus reaction follow this injection of serum? If 
so, describe the reaction. 
(b) Explain the mechanism of a local anaphylactic reaction. 
(c) Are the tuberculin, luetin, and mallein reactions anaphylactic? 
Explain their mechanism and discuss their practical value in diagnosis. 
4. Inject 1 c.c. horse-serum intravenously. If anaphylaxis occurs, record the symptoms. 
Autopsy with particular attention to the heart. 
Experiment 140.—Anaphylaxis in Vitro: 
1. Sensitize a young female guinea-pig with horse-serum. 
2. Twelve to sixteen days later remove the uterus and suspend in a saline bath according 
to the Schultz-Dale method. Arrange for kymographic tracing of contracting uterus. 
3. After securing a tracing of normal contractions add horse-serum to the bath and ob- 
serve the contractions. 
(a) Discuss the prevailing views on the humoral and cellular theories 
of anaphylaxis. 
(b) Discuss the nature of anaphylactogens. 
Experiment 141.—Chemotherapy: Toxicity or Organotropism: 
1. Prepare a 2 per cent, alkalinized solution of arsphenamin by dissolving 0.2 gm. in 
8.4 c.c. of hot water and adding 1.6 c.c. of N/l solution sodium hydroxid. 
2. Inject 4 white rats weighing 100 to 150 grams intravenously with the following amounts 
per 100 grams of weight: 0.4, 0.5, 0.6, and 1 c.c. which administers 80, 100, 120, and 200 mg. 
of the drug per kilo of body weight. 
3. Prepare a 1 : 1000 dilution of mercury bichlorid by dissolving 0.1 gm. in 100 c.c. of 
saline or distilled water. 
4. Inject 4 white rats intravenously with 0.2, 0.4, 0.6, and 1 c.c. which administers 2, 
4, 6, and 10 mg. per kilo of body weight. 
5. Observe the animals for ten days. 
(a) What is the maximum tolerated dose of arsphenamin? Of mercuric 
bichlorid? 
(b) What is meant by toxicity or organotropism? 
(c) What is meant by do sis lethalis and dosis tolerata? 
Experiment 142.—Chemotherapy: Parasitotropism: 
1. Infect 4 white rats by injecting each intraperitoneally with 0.2 c.c. of the citrated 
blood of rat infected with T. equiperdum and showing large numbers of the parasites in the 
tail blood when examined microscopically. 
2. Twenty-four hours later inject No. 1 intravenously with 0.2 c.c. of a 0.5 per cent, al- 
kalinized solution of arsphenamin (freshly prepared by dissolving 0.1 gm. of the powder in ANAPHYLAXIS 
1157 
19 c.c. of hot water and adding 1 c.c. of N/I solution sodium hydroxid) per 100 grams of 
weight, which corresponds to 10 mg. per kilo of body weight. This amount corresponds to 
about one-tenth the toxic dose. 
3. Inject the second rat with 0.2 c.c. of a 1 per cent, solution of neo-arsphenamin (freshly 
prepared by dissolving 0.1 gm. in 10 c.c. of sterile water) per 100 grams of weight, which 
corresponds to 20 mg. per kilo of body weight. * 
4. Inject the third rat with 1 c.c. of a 1 : 5000 dilution of bichlorid of mercury per 100 
grams of weight which corresponds to 2 mg. per kilo of body weight. This amount corre- 
sponds to about one-third the toxic dose. The fourth rat is kept as a control. 
5. Examine the tail blood of the four rats a few hours later and again at daily intervals. 
(a) What is meant by parasitotropism? 
(b) Is arsphenamin trypanocidal? Is mercury bichlorid? 
(c) What is the purpose of chemotherapy? 
(id) Discuss the question of chemoreceptors. 
(e) Discuss “drug fastness.” INDEX 
Abderhalden’s ferment tests, 237 
Bronfenbrenner’s modification, 243 
dialyzation method, 237, 238 
blood-serum, 242 
dialyzing cylinder for, 238 
glassware, 238 
ninhydrin, 239 
precautions, 239 
preparation of placental tissue, 
241 
reading reaction, 243 
reagents, 239 
sources of error in, 243 
testing dialyzing shell, 238 
shell for non-permeability to 
albumin, 239 
for permeability to peptone, 
240 
the test, 242 
mechanism of, 234 
Bronfenbrenner’s theory, 234 
methods, 237 
optical method, 238 
practical value of, 244 
in cancer, 245 
in mental diseases, 245 
in pregnancy, 244 
in syphilis, 245 
in tuberculosis and acute infec- 
tions, 245 
Abdominal organs, injuries of, hemorrhage 
due to, transfusion for, 1080 
pain, allergic, 655 
Abin, 103 
Abortin reaction, 711 
Abortion, contagious, complement fixation in, 
548 
infectious, of cows, vaccination, 832 
of mares, vaccination against, 833 
serum prophylaxis of, 877 
of cattle, infectious, 711 
Abrus precatorius, 103 
Abscess, apical, vaccine treatment, 1008 
embolic, experimental work, 1115 • 
non-specific treatment of, 890 
serum treatment of, 889 
treatment of, 888 
vaccine treatment of, 889 
Absorption agglutination test in mixed infec- 
tion, technic, 289 
methods for differentiating between mixed 
and single infection, 261 
of bacterioprecipitins, 332 
removal of hemolysins from serum by, 399 
Abwehrfermente, 232 
Accelerated reaction in infection, 89 
Accessory sinuses of nose, diseases of, treat- 
ment, 1002 
Acetone-insoluble lipoids for Wassermann 
reaction, 457 
Acid agglutination, 259 
technic, 289 
butyric, reaction in syphilis, Noguchi’s, 518 
nitric, reaction in syphilis, B ruck’s, 519 
Acid-fast bacilli., fixing and staining, 183 
Acidosis, blood transfusion in, 1089 
Acids, effect of, on complement, 343 
experimental work, 1143 
hemolysis by, 383 
hemolytic activity, experimental work, 1137 
Acne, milk injections in, 993 
non-specific treatment, 993 
treatment of, 992 
turpentine injections in, 993 
typhoid vaccine in, 993 
vaccine treatment, 992 
Acquired predisposition to infection, 73 
susceptibility, 74 
Actinocongestin, 595 
Actinomycosis, vaccine treatment, 998 
Active acquired immunity, 119 
immunization, 748. See also Immuniza- 
tion, active. 
Activity, antigenic, of microparasites, varia- 
tion in, 752 
specific, of vaccines, 755 
of sera, anticomplementary, 405 
protective or antihemolytic, 404 
Acupuncture, multiple, method of vaccina- 
tion, 775 
Acute yellow atrophy of liver, hemorrhage 
in, transfusion for, 1088 
Addiment, 338 
Addisonian anemia, 405 
Adenitis in serum disease, 661 
tuberculous, tuberculin treatment, 1072 
Adenoid tissue as foci of infection, 79 
Adnexal disease, turpentine in, 984 
Adrenalin chlorid as preventive of serum 
shock, 724 
Adrenotoxins, 571 
Adulteration, meat, detection of, by anaph- 
ylaxis reaction, 642 
precipitin test for, 329 
Aerogenes capsulatus bacillus, 102, 872 
A. F. tuberculin, preparation of, 1054 
Age, bearing of, upon immunity to tumors, 
1027 
Agglutinating strength of serum, variation in, 
262 
Agglutination, 254 
acid, 259 
1159 1160 
INDEX 
Agglutination acid, technic, 289 
bacterial, test, absorption, in mixed infec- 
tion, 289 
Dreyer standardized, 282 
modified, 284 
in differentiation of meningococci, 288 
of pneumococci, 286 
macroscopic, advantages of, 273 
bacterial suspension, 277 
Dreyer’s method, 278 
disadvantages, 273 
Kolle and Pfeiffer’s method, 279 
slide methods, 284 
technic, 277 
microscopic, 272 
advantages, 272 
disadvantages, 272 
dry method, 276 
wet method, 273 
precautions, 275 
technic, 272 
degrees of, terms employed in describing, 
284 
for differentiation of pneumococci, experi- 
mental work,1132 
in vivo, experimental work, 1132 
influence of electrolytes on, experimental 
work,1131 
mechanism of, 259 
Bordet’s theory, 260 
Gruber’s theory, 259 
Paltauf’s theory, 260 
mil, 284 
standard, 283, 284 
test after typhoid-paratyphoid immuniza- 
tion, 266 
for detection of typhoid carriers, 266 
for grouping of human corpuscles before 
blood transfusion, experimental work, 
1133 
group, experimental work, 1130 
in bacillary dysentery, 268 
in cerebrospinal meningitis, 269 
in cholera, 269 
in diagnosis of disease, 264 
of typhoid and paratyphoid fevers in 
vaccinated persons, 267 
production of agglutinins by vaccines 
in relation to use of, 266 
in dourine, 271 
in glanders, 268 
in identification of micro-organism, 264 
in inoculated persons, 266 
in leprosy, 271 
in malaria, 271 
in Malta fever, 270 
in measuring immunizing response, 264 
in mixed infections, 268 
in paratyphoid fever, 264 
in pertussis, 270 
in plague, 270 
in pneumonia, 270 
in single or mixed infection, 264 
in sporotrichosis, 270 
in syphilis, 270 
in tuberculosis, 269 
in typhoid fever, 264 
in typhus fever, 268 
Agglutination test in whooping-cough, 270 
macroscopic, experimental work, 1129, 
1130 
microscopic, experimental work, 1128 
practical applications, 264 
preliminary to blood transfusion, experi- 
mental work, 1133 
total, 284 
trace, 284 
Agglutinin unit, standard, 283 
bacterial, 140, 254 
absorption of, experimental work, 1131 
and agglutinoids, 256 
theoretic structure of, 257 
definition, 254 
effect of heat on, experimental work, 
1131 
experimental work, 1110, 1128 
formation of, 141 
group or partial, 260 
historic, 254 
immune, 255 
production of, 271 
intraperitoneal method, 272 
intravenous method, 271 
in immunity, role of, 263 
kinds of, 255 
macroscopic tests for, 298 
pipet method, 299 
normal, 255 
origin and distribution, 257 
properties and nature of, 258 
relation of lipoids to, 583 
specificity of, 260 
theoretic formation of, 256 
Agglutinogen, 257 
Agglutinoids and agglutinins, 256 
theoretic structure of, 257 
experimental work, 1131 
Agglutinoscope, 281 
Aggressins, 75, 85, 107, 108 
anti-, 110 
artificial, 76, 109 
experimental work, 1117 
experimental work, 1117 
natural, 109 
nature of, 109 
Aggressiveness of bacteria, 68 
Albuminolysin, 599, 626 
relation of, to allergin, 627 
of precipitins to, 315 
to complement-fixing antibody, 427 
Albuminuric retinitis, milk injections in, 1018 
Albumose solution, intravenously, in broncho- 
pneumonia of influenza, 920 
Albumose-free old tuberculin, preparation of, 
1054 
Albumoses intravenously in treatment of 
typhoid fever, 904 
Alcohol, influence of, on antibodies and im- 
munity, 156 
Alcoholic extract of heart muscle cholesterol- 
ized and lecithinized for Wassermann 
reaction, 457 
of normal organs for Wassermann reac- 
tion, 455 
reinforced with cholesterin for Was- 
sermann reaction, 456 INDEX 
1161 
Alcoholic extract of syphilitic livers for Was- 
sermann reaction, 455 
Alexins,, 133, 135, 336, 338, 357, 358. See also 
Complement. 
Alexofixagens, 425 
chemical nature, 426 
nature, 427 
specificity of, 429 
Alkalies, effect of, on complement, experi- 
mental work, 1143 
hemolytic activity, experimental work, 1137 
Allergens, 599, 606 
food, method for preparing, 682 
kinds of, 607 
pollen, preparation of, 729 
Allergic asthma, groups of, 653 
skin tests as guide to treatment by 
desensitization, 733 
for bacteria, fungi and protozoa in, 
684 
treatment of, 733 
bronchial asthma, 651 
dermatitis, treatment, 735 
nature of tuberculin reaction, 1056 
reactions, production of, 621 
rhinitis, 653 
shock, prevention of, 620 
skin lesions, 653 
reactions, 671. See also Anaphylactic re- 
action. 
vomiting, diarrhea, and abdominal pain, 
655 
Allergin, 599, 626 
relation of albuminolysins to, 627 
of ferments to, 628 
of precipitin to, 626 
of proteolysins to, 627 
Allergization, 616 
Allergy, 594, 596 
acquired and natural, 650 
cellular colloidal nature of, 632 
theory of, 624 
clinical, 650 
definition, 598 
historic, 594 
Arthus phenomenon, 595 
Richet’s studies, 595 
Theobald Smith phenomenon, 596 
von Pirquet’s early studies, 595 
human, general lesions and symptoms of, 
651 
treatment of, 721 
humoral theories of, 622 
Besredka’s, 623 
Friedberger’s, 623 
Gay and Southard’s, 623 
Hamburger and Moro’s, 622 
Nolf’s, 624 
Richet’s, 622 
Vaughan and Wheeler’s, 623 
in man, bronchial asthma as symptoms of, 
652 
importance and frequency of, 650 
in relation to infection and immunity, 643 
local, 637 
symptoms and lesions of, 639 
mechanism of, analysis of data bearing 
upon, 625 
Allergy, natural and acquired, 650 
pathologic physiology of, 604 
pathology, 599 
pollen, 662. See also Hay-fever. 
prevention of, 620 
prophylactic treatment of, 721 
relation of antibodies to, 626 
of humoral poisons to, 629 
specificity of, 640 
symptoms, 599 
theories of mechanism of, 622 
to animal substances, 666 
to bacteria, 667 
fungi and protozoa, skin test for, 684 
to dandruffs, skin tests for, 681 
to drugs, 652, 669 
incubation period, 670 
skin tests for, 684 
symptoms, 670 
treatment, 738 
to feathers, skin tests for, 681 
to foods, 668 
desensitization by feeding, 737 
diagnosis, 669 
foods commonly producing, 669 
sensitzation in, 668 
skin tests for, 682 
symptoms, 669 
treatment, 669, 736 
by desensitization, 737 
by elimination of foods, 736 
to hairs, skin tests for, 681 
to plant substances, 666 
to pollens, 662 
asthma due to, 652 
skin test for, 680 
treatment of, 725. See also Hay-fever. 
to serum, 653 
prophylactic treatment of, 722 
cases requiring desensitization, 722 
methods for specific desensitization, 
723 
prevention of sensitization, 722 
skin tests for, 680 
to stings of insects, 666 
Alveolitis, vaccine treatment, 1008 
Amanita phalloides, 103 
Amboceptoids, 340 
Amboceptor, 133, 141, 143, 338, 172 
and complement, 336 
mechanism of action of, 353 
Baumgarten’s theory, 353 
Bordet’s theory, 354 
Buchner’s theory of, 353 
Ehrlich-Morgenroth theory, 354 
Metchnikoff’s theory, 355 
antigenophore group of, 142 
bacteriolytic, 142, 338 
theoretic structure of, 362 
Bordet, 428 
complementophile group of, 142 
formation of, 340 
haptophore group of, 142 
hemolytic, 338 
historic, 339 
native, 341 
natural, 341 
properties of, 340 1162 
INDEX 
Amboceptor, specificity of, 340 
structure of, 141, 339 
Amebadiastase, 131 
Amorphous colloids, 576 
Amoss and Flexner method for rapid produc- 
tion of antidysenteric serum, 221 
Amoss and Wollstein’s rapid method of pre- 
paring antimeningococcus serum, 948 
Ampule, vaccine, 199 
conversion of test-tube into, 7 
Anaphylactic asphyxia, 600 
reactions, amount of serum administered in 
relation to, 657 
as indices of infection and hypersuscep- 
tibility, 673 
cutaneous versus intracutaneous tests, 
673 
danger of, 657 
diagnostic applications, 642 
differentiation of blood-stains by, 642 
of proteins by, 642 
effect of drugs upon, 679 
in blastomycosis, 717 
in cancer, 719 
in chick enpox, 717 
in coccidiosis, 717 
in cowpox, 717 
in detection of meat adulteration, 642 
in diagnosis of cancer, 642 
of disease, 642 
in diphtheria, 715 
in echinococcus disease, 718 
in favus, 718 
in gonococcus infections, 714 
in meningococcus carriers, 716 
in pertussis, 717 
in pneumonia, 715 
in pregnancy, 719 
in ringworm, 718 
in smallpox, 717 
in sporotrichosis, 717 
in typhoid fever, 712 
influence of age, 671 
of disease, 671 
of non-specific factors, 672 
intracutaneous tests more likely than 
cutaneous tests to elicit general reac- 
tions, 674 
local, experimental work, 1156 
mechanism, 639 
of specific reactions, 672 
non-specific and specific, 671 
skin reactions, 672 
prevention of, 721 
reactivity of skin in, 671 
route of administration of serum in rela- 
tion to, 656 
skin tests for allergies to bacteria, fungi, 
and protozoa, 684 
to drugs, 684 
for food allergies, 682 
for pollen and plant allergies, 680 
for serum allergy, 680 
in focal and mixed infections, 685 
specific and non-specific, 671 
symptoms of unusual reaction, 658 
technic of cutaneous test, 674 
of intracutaneous tests, 677 
Anaphylactic serum shock, curative influence 
of, 649 
immediate and severe, symptoms of, 
658 
medicinal prophylaxis of, 724 
Anaphylactin, 599, 626 
Anaphylactogenic activity and specificity, 
structural basis of, 611 
of protein derivatives, 613 
for lower animals, 614 
of racemized proteins, 613 
Anaphylactogens, 599, 607 
bacterial, 609 
food, 608 
hair, 608 
non-protein, 613 
physical state of, 610 
pollen, 608 
protein, chemistry of, 611 
protozoan, 609 
relation of precipitinogens to, 615 
serum, 608 
toxin, 609 
Anaphylactoid phenomena, 606 
reactions, 606 
Anaphylatoxin, 113, 627 
Anaphylaxis, 594 
acquired, 598 
bacterial, 643 
relation of, to disease, 646 
chronic, 603 
classification of phenomena, 597 
Coca’s, 598 
definition, 598 
experimental work, 1154 
in cats, 603 
in cattle, 604 
in dogs, 603 
experimental work, 1155 
in guinea-pig, 600 
experimental work, 1154 
in hogs, 604 
in horses, 604 
in mice and rats, 604 
in rabbit, 602 
experimental work, 1156 
in serum therapy, 840 
in sheep, 604 
in vitro, experimental work, 1156 
indirect, 607 
natural, 598 
passive, 610 
experimental work, 1155 
production of, 633 
pseudo-, 606 
relation of, to immunity, 647 
of lipoids to, 584 
true, criteria of, 606 
Anemia, Addisonian, 405 
aplastic, blood, transfusion in, 1087 
blood injections in, 1020 
hemolytic, 405 
pernicious, blood transfusion in, 1084-1087 
splenectomy in, 1087 
relation of protective activity of sera to, 
404 
secondary, blood transfusion in, 1079 
treatment of, 1018 INDEX 
1163 
Anesthesia for intraspinal injection of serum, 
852 
in lumbar puncture, 26, 852 
Angioneurotic edema, 653, 654 
Animal and human blood, precipitin test for 
differentiation, 321 
blood. See Blood, animal. 
conjunctival tuberculin test in, 703 
cutaneous tuberculin test in, 705 
experiments, 1110 
higher, toxins of, 104 
immunization with diphtheria antitoxin, 
213 
inoculation, 42 
intracardial, 50 
of guinea-pig, 50 
intramuscular, 44 
intraperitoneal, 52 
of guinea-pig, 52 
of rabbit, 52 
intravenous, of dog, 49 
of goats, 49 
of guinea-pig, 46 
Roth’s method, 46 
of horse, 48 
of mice, 47 
of rabbit, 44 
of rats, 47 
of sheep, 49 
subcutaneous, 43 
with fluid inoculum, 43 
with solid inoculum, 44 
technic, 42 
general rules, 42 
intracutaneous tuberculin test in, 705 
living, preservation of serum in, 57 
lower, anaphylactogenic activity of pro- 
teins for, 614 
diseases of, treatment, 1039 
isohemagglutinins in, 295 
parasites, aggressiveness of, 82 
infection with, 81 
modes of, 82 
production of disease by, 86 
selective affinity of, 82 
Pasteurellosis of, 834 
prophylactic immunization of, 829 
secretions, hemolysis by, 386 
serum prophylaxis in diseases of, 876 
subcutaneous tuberculin test in, 700 
substances, allergy to, 666 
toxins, experimental work, 1112 
vaccination in, 829 
Anthrax bacteremia, treatment of, 970 
bacterial precipitins in, 316 
immunity in, 966 
in animals, serum treatment, 1040 
infection in, 966 
internal, treatment of, 970 
serum prophylaxis of, 878 
treatment, 967, 969 
symptomatic, vaccination, 831 
treatment of, 966, 968 
vaccination, 829 
dosage of vaccine, 831 
preparation of vaccines, 830 
technic, 830 
value, 831 
Anthrax vaccination with vaccine and serum, 
831 
vaccine for, 830 
Anti-aggressins, 110 
Antiamboceptors, 161, 342 
Anti-anaphylaxis, 634, 721 
by antibody protection, 634 
experimental work, 1155 
mechanism of, 637 
Anti-anthrax serum, 967 
in malignant pustule, 969 
Anti-antibodies, 161 
Anti-autolysin, 388 
Antibacterial immunity, experimental work, 
1119, 1120, 1121 
immunization, 838 
Antiblastic immunity, 162, 228 
Antibody, 118, 137 
and antigen, dissociation of, 151 
and immunity, influence of alcohol on, 156 
of drugs on, 154 
of nutrition on, 154 
of x-rays on, 154 
transmission of, 152 
chemical nature of, 150 
complement-fixing, kinds of antigens in- 
ducing production of, 425 
nature of, 427 
relation to precipitins, cytolysins, and 
albuminolysins, 427 
specificity, 429 
spirochetic, nature and specificity, 429 
definition, 148 
distribution of, in body fluids, 149 
formation, influence of temperature upon, 
157 
immune, 161 
local production of, 149 
normal, 161 
of first order, 137 
of second order, 140 
of third order, 140, 336, 337 
production from tuberculin treatment, 1058 
influence of non-specific agents upon, 883 
relation of, to allergy, 626 
solution, Huntoon’s, in treatment of pneu- 
mococcus pneumonia, 914 
specificity of, 157 
tissues concerned in production of, 148 
transmission of, in colostrum, 152 
vs. ferments in disease, 231 
Antichicken hemolytic system in qualitative 
and quantitative complement fixation, 491 
Anticholera serum, 973 
administration, 974 
dosage, 974 
value of, 974 
Anticomplementary action of sera and an- 
tigens, importance of, in relation to 
complement-fixation tests and reac- 
tions, 437 
of serums, experimental work, 1146 
activity of antigens and sera, 433 
of sera, 405 
serum, 399 
substances, chemical nature, 435 
influence of heat upon, 435 
kinds, 404 1164 
INDEX 
Anticomplementary substances, mechanism of 
action, 436 
properties, 435 
titration of antigens for Wassermann reac- 
tion, 462 
unit, 437 
Anticomplements, 344 
Anticytotoxic serums, 569 
Antidiphtheric serum, production of, 210 
Antidysenteric serum, 220, 940 
administration and uses, 940 
collecting and testing, 222 
production of, 220 
culture, 220 
immunizing animals, 221 
rapid method of Flexner and Amoss, 
221 
Antiferments, 227 
experimental work, 1127 
influence of non-specific agents upon, 882 
nature of, 229 
relation of lipoids to, 582 
simple, 137 
Antigen, 146 
and antibody, dissociation of, 151 
and sera, anticomplementary activity of, 
433 _ 
bacterial, identification of, complement 
fixation for, 564 
preparation and titration, experimental 
work, 1149 
definition, 146 
determination of, by complement fixation, 
560 
drugs as, 148 
for Wassermann reaction, 453 
experimental work, 1145 
method of diluting, 461 
new, Kolmer’s, 457 
titration of, principles for, 462 
heterologous, production of hemolysins by, 
389 
heterophile, 159 
importance of anticomplementary action 
of, in relation to complement-fixation 
tests and reactions, 437 
inducing production of complement-fixing 
antibodies, kinds, 425 
lipoids as, 582 
nature of, 146 
non-protein, 147 
non-specific, 146 
specific, 146 
Antigen-antibody equilibrium, 151 
reaction, 151 
Antigenic activity of microparasites, varia- 
tion in, 752 
specific, of vaccines, 755 
titration of antigens for Wassermann reac- 
tion, 462 
unit of syphilitic serum, 454 
values of various extracts, comparative, in 
Wassermann reaction, 459 
Antigonococcus serum in acute epididymitis, 
978, 979 
gonorrhea, 979 
in arthritis, 978, 979 
in gonococcus septicemia, 978 
Antigonococcus serum in orchitis, 978, 979 
in prostatitis, 978, 979 
in salpingitis, 979 
Antihemolysins, 161 
Antihemolytic activity of sera, 404 
Antihuman hemolysin, production of, 401 
titration, experimental work, 1149 
hemolytic system in qualitative and quanti- 
tative complement fixation, 491 
Anti-influenza serum, 961 
administration of, 962 
Antilactase, 230 
Antilysins, kinds, 434 
thermolabile, influence of heat upon, 435 
Antimeningococcus serum, action of, 951 
Amoss and Wollstein’s rapid method of 
preparing, 948 
dosage, 953 
Flexner and Jobling’s method of pre- 
paring, 947 
in tuberculous meningitis, 965 
intraspinal administration, 953 
intravenous administration, 954 
monovalent, 952 
polyvalent, 952 
preparation of, 947 
preservation of, 949 
reactivation of, 954 
repeating doses, 955 
serum sickness from, 957 
standardization of, 949 
Antiox hemolytic system in quantitative and 
qualitative complement fixation, 490 
Antipepsin, 230 
Antipertussis serum, 975 
Antipest serum, 970 
administration, 971 
dosage of, 971 
Antiplague serum, 970 
administration, 971 
dosage, 971 
Antipneumococcus serum, action, 911 
administration, 912 
and ethylhydrocuprein hydrochlorid in 
pneumococcus meningitis, 965 
in treatment of bronchopneumonia of 
influenza, 919 
intravenous injections, in pneumococcus 
meningitis, 964 
Kyes’, 913 
in pneumococcus meningitis, 964 
preparation, 910 
sodium oleate, and boric acid in pneu- 
mococcus meningitis, 962 
standardization, 910 
Antiprecipitins, 307 
Antirennin, 230 
Antisensitization, 620 
Antiseptics, preservation of serum by, 53 
Antisheep hemolysin, production of, 402 
titration, experimental work, 1138 
Antistaphylococcus serum, 223 
antilysin test for, 223 
preparation of, 223 
Antistaphylolysin, experimental work, 1127 
Antisteapsin, 230 
Antistreptococcus serum, administration of, 
895 INDEX 
1165 
Antistreptococcus serum in treatment of 
bronchopneumonia of influenza, 920 
of scarlet fever, 899 
mode of action of, 894 
preparation of, 894 
standardization of, 895 
value of, 896 
Antitetanic serum, production of, 218 
Antitetanolysin, action of, experimental 
work,1127 
experimental work, 1114 
Antitoxic immunity, experimental work, 1120, 
1121 
immunization, 838 
sera, monovalent vs. polyvalent, 208 
polyvalent vs. monovalent, 208 
Antitoxin, 137, 203 
Bacillus perfringens preparation of, 872 
botulinus, 220 
definition, 203 
diphtheria, 921. See also Diphtheria anti- 
toxin. 
dysentery, administration and uses, 941 
results, 941 
experimental work, 1125 
for therapeutic purposes, production of, 
210 
for toxin of Bacillus welchii, 102 
formation of, 204 
globulins, isolating, method of concentrat- 
ing serum by, 214 
haptophore group of, 205 
historic, 203 
immunity to diphtheria, tests for, 740 
immunizing animals with, 213 
measure of, 225 
method of concentrating, 214 
natural, 207 
diphtheria, Schick test for, 208 
in human serum, Kellogg’s method for 
determining as substitute for Schick 
test, 217 
pollen, production of, 225 
production of, by horses, 213 
for therapeutic purposes, 210 
properties of, 205 
relation of, to proteins, 206 
specificity of, 208 
experimental work, 1113 
structure of, 205 
tetanus, 218, 993. See also Tetanus anti- 
toxin. 
theoretic formation of, 138, 204 
unit,225 
Antitrypsin, 238 
test, 236 
Fuld and Goss’s, 236 
Antityrosinase, 230 
Antiurease, 230 
Anti vaccinationists, 780, 781 
Antivenin, 204 
preparation of, 225 
production of, 224 
Apical abscesses, vaccine treatment, 1008 
Apices of teeth as foci of infection, 79 
Aplastic anemia, blood transfusion in, 1087 
Apophylactic phase of vaccine therapy, 765 
Apotoxin, 622 
Apparatus for aseptic collection of large 
amounts of blood, 859 
of small amounts of blood, 858 
Appendicitis, vaccine treatment, 1010 
Aqueous extracts of pallidum culture for 
Wassermann reaction, 459 
of syphilitic livers for Wassermann reac- 
tion, 455 
Arthritis, antigonococcus serum in, 978, 979 
autoserum treatment, 987 
deformans, vaccine treatment, 987 
gonococcus and mixed vaccines in, 980, 981 
immunity in, 985 
in serum disease, 661 
infection in, 985 
infectious, of colts, vaccination against, 833 
intravenous vaccine in, 981 
milk injections in, 983 
non-specific agents in, 981 
treatment, 988 
rheumatic, chronic, vaccine treatment, 987 
serum treatment, 986 
treatment of, 985 
typhoid vaccine in, 988 
contraindications, 989 
dosage and administration, 989 
results and practical value, 990 
vaccine treatment, 987 
Artificial agressins, 76, 109 
Asiatic cholera, serum treatment, 973 
treatment of, 973 
vaccine treatment, 975 
Asphyxia, anaphylactic, 600 
Asthma, allergic, groups of, 653 
skin tests as guide to treatment by 
desensitization, 733 
for bacteria, fungi, and protozoa in, 
684 
treatment of, 733 
bacterial, treatment, 735 
bronchial, allergic, 651 
symptom of allergy in man, 652 
Danysz’s enterovaccine in, 1006 
due to pollen allergy, 652 
dust, 652 
food, treatment, 734 
horse, cat, dog and fowl, treatment, 734 
horse-serum in, 722 
horse-serum, treatment, 734 
in serum therapy, 841 
patient’s defibrinated blood in, 1006 
pollen, treatment, 734 
treatment of, 1005 
typhoid bacilli in, 1006 
vaccine treatment, 1005 
Witte’s peptone in, 1006 
Asthmatics, horse, 657 
Ataxia, locomotor, treatment of, 1035 
Athrepsia, 166, 1030 
blood transfusion in, 1090 
'\threptic immunity, 166, 1030 
Atmosphere, bacteria in, 62 
Atrophy, receptoric, 57 
in active immunization, 168 
Atropin sulphate as preventive of serum 
shock, 724 
Auditory canal, external, furunculosis of, 
treatment, 1000 1166 
INDEX 
Auto-anticomplements, 344 
Autocytotoxins, 568 
Autogenous tuberculins, 1061 
versus stock bacterial vaccines, 755 
Autohemagglutinins, 255, 292 
Autohemolysin, 388, 390 
Autolysates in malignant tumors, 1031 
Autolysins, 387 
Autolysis, 114 
Autolytic products, hemolysis by, 384 
Automatic pipet, Comer’s, 199 
Autoprecipitin, 308 
Autoserum, human serum for, methods of 
preparing, 857 
treatment of arthritis, 987 
of chorea, 1036 
of eczema, 994 
of hepatitis with ascites, 1010 
of pleuritis, 1006 
of psoriasis, 995 
of tuberculous pleurisy, 1006, 1007 
Auto vaccination, 759 
Autumnal catarrh, 225 
Auxilysins, 355, 356 
Avery and Dochez’s precipitin reaction with 
urine in pneumonia, 332 
Babcock needles for spinal puncture, 30 
Bacillary dysentery, agglutination reaction 
in, 268 
non-specific treatment, 944 
serum prophylaxis of, 871 
treatment of, 940 
treatment of, 940 
vaccine treatment of, 943 
Bacillen emulsion tuberculin, preparation of, 
1053 
Bacillus abortus, 711 
equi, 548, 712, 833 
of Bang, 548, 832 
acid-fast, fixing and staining, 183 
aerogenes capsulatus, 102, 872 
avisepticus, 834 
botulinus, 100, 878 
bovisepticus, 834 
bronchisepticus, 833 
carriers, milk injections for, 905 
chauveaui, 102 
diphtheria, 97 
toxicity and virulence of, method of 
testing, 97, 1112 
dysentery, 100, 810, 871, 940 
Shiga, 940 
edema tis maligni, 102 
equisepticus, 834 
Friedlander, 812 
gas, 102 
influenza, 102, 811, 812 
mallei, 710 
of blackleg, 102 
ovisepticus, 834 
ozaena-foetidae, 1004 
perfringens, 102, 872 
pestis, 805 
in vaccine for immunization against 
bubonic plague, 795 
Pfeiffer’s, 811 
Bacillus, Plotz, 801 
proteus, 801 
pyocyaneus, 102 
Shiga dysentery, 940 
suisepticus, 834 
toxic-producing, types of, 208 
typhi exanthematici, 801 
tetanus, 99, 102, 868 
typhoid, 102, 713 
and paratyphoid, triple vaccine of, 795 
paratyphoid and cholera, tetravaccine 
of, 795 
in asthma, 1006 
welchii, 102, 872 
antitoxin, 102 
Bacteremia, 67 
anthrax, treatment of, 970 
experimental work, 1115 
meningococcus, 945 
streptococcus, vaccine treatment, 892 
Bacteria, active sensitization by, 634 
aggressiveness of, 68 
allergy to, 667 
skin test for, 684 
anaphylaxis reaction in differentiation of, 
642 
Bail’s classification of, 108 
complement-fixation test, Kolmer’s new 
method, for identification, 565 
conglutination test with, technic of, 285 
disease-producing, 60 
effect of opsonins on, 174 
fixing and staining, 183 
identification and differentation of, bac- 
terial precipitins for, 317 
in atmosphere, 62 
in food, 61 
in milk, 61 
in placenta, 62 
in water, 61 
invasion of, mechanism of, 66 
normal defenses against, 65 
living, 200 
mechanical action, 85, 116 
experimental work, 1117, 1118 
non-agglutinable species, 261 
non-pathogenic, 60 
on skin, 62, 63 
on unclean objects, 62 
pathogenic, 60 
production of disease by, 84 
recovered from feces, water-supplies, etc., 
bacteriolytic test in vivo for identifica- 
tion of, 365 
saprophytic, 67 
selective affinity in, 71 
sensitization of, 200 
susceptibility to opsonification, 174 
toxicity of, 68 
transmission by suctorial insects, 62 
virulence of, 68 
decrease in, 69 
increase in, 69 
by addition of animal fluids to culture- 
medium, 70 
by use of collodion sacs, 70 
Bacteria-free filtration, preservation of serum 
by, 55 INDEX 
Bacterial agglutination reactions, technic, 
272. See also Agglutination, bacterial, 
reactions. 
agglutinins, 254. See also Agglutinins, 
bacterial. 
anaphylactogens, 609 
anaphylaxis, 643 
relation of, to disease, 646 
antigens, identification of, complement 
fixation for, 564 
preparation and titration, experimental 
work, 1149 
asthmas, treatment, 735 
complement-fixation tests, technic, 539 
diseases, specific complement fixation in, 538 
emulsion in opsonic index, 178 
ferments, immunity to, 228 
in relation to infection, 227 
hemagglutinins, 102 
hemolysins, 102 
method for determining, 413 
hemolysis, experimental work, 1136 
hemotoxins, 102 
infections, complement fixation in, 538 
precipitin reactions in, 333 
invasion, avenues of entrance, 62 
factors preventing, 65 
mechanism of, 66 
normal defenses against, 65 
poisons, hemolysis by, 385 
precipitinogens in inflammatory exudates, 
316 
experimental work, 1135 
in urine in pneumonia, 315 
preparation of, 331 
precipitins, experimental work, 1135 
for identification and differentiation of 
bacteria, 317 
in diseases, 316 
tests, 331 
products, effect of, on phagocytosis, 131 
proteins, 85, 110 
action, 111 
digested, toxicity of, 114 
experimental work, 1117 
Friedberger’s theory, 113 
nature, 110 
Vaughan’s theory, 112 
ptomains, Breiger’s method of isolation, 116 
toxemia, 85 
toxins, 92 
experimental work, 1112 
extracellular, 92 
definition, 92 
nomenclature, 92 
other, 102 
vaccines, 189, 749. See also Vaccines, 
bacterial. 
Bactericidal activity of blood, Stern and 
Korte’s experimental work, 1152 
of coagulable blood, Lacy-Heist method 
of determining, 378 
of normal blood and serum, 357 
of whole blood versus plasma and serum, 
361 
coagulable blood, Heist-Lacy method, 
experimental work, 1152 
without complement, 362 
Bactericidal power of blood, capillary pipet 
method of measuring, 375 
Neisser and Wechsberg’s method of 
measuring, 370 
plate culture method of measuring, 370 
Stern and Korte’s method of measur- 
ing, 370 
Topfer and Jaffe’s method of measur- 
ing, 372 
of serum, experimental work, 1152 
Bactericidins, 358 
Bacterin, 750 
therapy, definition, 750 
Bacteriology of gas gangrene, 871 
Bacteriolysant, 250 
dysentery, D’Herelle’s, 943 
Bacteriolysins, 118, 140, 338, 357 
definition, 358 
experimental work, 1110 
historic, 357 
immune, 364 
in immunity, role of, 364 
leukocytic, relationship of, to complement, 
360 
natural, 363 
origin, 358 
practical application, 364 
properties, 363 
specificity, 364 
Bacteriolysis, 135 
and hemolysis, analogy between, 393 
mechanism of, 362 
Twort-d’Herelle phenomenon of, 246 
d’Herelle’s method of securing bac- 
teriolytic agent, 247 
nature of bacteriolytic agent, 250 
properties of bacteriolytic agent, 249 
sources and specificity of bacteriolytic 
agent, 247 
Bacteriolytic agent in bacteriolysis, d’Herelle’s 
method of securing, 247 
nature of, 250 
properties of, 249 
sources and specificity of, 247 
amboceptors, 142, 338 
theoretic structure of, 362 
power of blood, microscopic method of 
measuring, experimental work, 1151 
serums in treatment of disease, 380 
tests in vivo for identification of bacteria 
recovered from feces, water-supplies, 
etc., 365 
method of testing virulence of culture, 
366 
of titrating bacteriolytic serum, 366 
Pfeiffer’s, 365 
experimental work, 1151 
in vivo in diagnosis of disease, 369 
technic, 369 
preparing immune serum, 366 
technic of, 365 
Bacterio-opsonins, 172 
Bacteriophage, 250, 251 
intestinal, d’Herelle’s, in relation to infec- 
tion, 72 
Bacteriophagum intestinale, 72, 251 
Bacterioprecipitins, 305 
absorption of, 332 
1167 1168 
INDEX 
Bacteriotropins, 134, 171 
experimental work, 1123, 1124 
quantitative estimation, experimental work, 
1123 
hygienic laboratory method, 186 
Neufeld’s technic, 184. See also Neu- 
feld’s quantitative estimation. 
Simon’s method, 186 
specificity of, experimental work, 1124 
Bacterium pneumosintes, 813, 917 
Bacteriuria, vaccine treatment, 1011 
Bail’s classification of bacteria, 108 
hypothesis, 108 
Bandi and Terni’s plague vaccine, 806 
Bang, bacillus abortus of, 832 
Basedow’s disease, complement fixation in, 
559 
Bass and Watkin’s slide method for Widal 
test for typhoid fever, 285 
Baumgarten’s theory of mechanism of action 
of complement and amboceptor, 353 
Beraneck’s tuberculin, 1066 
preparation, 1054 
Berkefeld filter, 54 
cleaning and care of, 56 
Besredka’s theory of allergy, 623 
B. E. tuberculin, preparation of, 1053 
B. F. tuberculin, preparation of, 1054 
Biologic therapy, kinds of streptococci in 
relation to, 891 
non-specific, 879 
of tuberculosis, 1042 
Bites, snake-, serum treatment of, 1035 
Blackfan’s apparatus for collecting blood, 23 
Blackleg bacillus, 102 
vaccination, 831 
preparation of aggressin, 831 
of vaccine, 832 
vaccine, 831, 832 
Bladder, diseases of, vaccine treatment, 1011 
tuberculosis of, tuberculin treatment, 1074 
Blake’s precipitin reaction with mouse 
peritoneal exudate in pneumonia, 332 
Blastomycosis, anaphylactic reaction in, 717 
complement fixation in, 559 
Blennorrhea, pollen, treatment of, 1015 
Blepharitis, ulcerative, vaccine treatment, 
1015 
Blood and serum, convalescent, in treatment 
of scarlet fever, 899 
human convalescent, in treatment of 
bronchopneumonia of influenza, 919 
animal and human, obtaining, 11 
obtaining large amounts, 33 
from dog, 40 
from guinea-pig, 37 
from hog, 40 
from horse, 41 
from monkey, 40 
from rabbit, 33 
Nuttall’s method, 33 
from rat, 38 
from sheep, 38 
small amounts, 32 
from dog, 40 
from guinea-pig, 32 
from horse, 40 
from monkey, 40 
Blood, animal, obtaining small amounts from 
rabbit, 32 
from rats, 37 
from sheep, 33, 39 
bactericidal activity, Stern and Korte’s 
experimental work, 1152 
power, capillary pipet method of meas- 
uring, 375 
Neisser and Wechsberg’s method of 
measuring, 370 
plate culture method of measuring, 370 
Stern and Korte’s method of measur- 
ing, 370 
Topfer and Jaffe’s method of measur- 
ing, 372 
Wright’s capillary pipet method of 
measuring, 375 
bacteriolytic power, microscopic method of 
measuring, experimental work, 1151 
Blackfan’s apparatus for collecting, 23 
capsules, Wright’s, making of, 6 
coagulable, bactericidal activity of, Lacy- 
Heist method of determining, 378 
collecting, for Wassermann reaction, 444 
large amounts, apparatus for, 859 
small amounts, apparatus for, 858 
diseases of, blood transfusion in, 1082 
film, method of preparing, 182 
Heist-Lacy method of bactericidal activity, 
experimental work, 1152 
human and animal, obtaining, 11 
precipitin test for differentiation, 321 
grouping for medicolegal identification of 
blood-stains, 298 
medicolegal application of, 297 
heterologous hemagglutinins in, 295 
obtaining, from anterior jugular vein, 
21 
large amounts of, 18 
phlebotomy, 18 
wet cupping, 22 
small amounts, 15 
from infants and children, 17 
influence of non-specific agents on, 882 
injections in anemia, 1020 
in hemophilia, 1018 
in melena neonatorum, 1018 
in retinal hemorrhage, 1018 
intramuscular injections, 1108 
Keidel tube for collecting, 20 
normal, and serum, bactericidal activity of, 
357 # 
obtaining, from superior longitudinal sinus, 
23 
placental, method of obtaining, 23 
plasma and serum, bactericidal activity of, 
versus whole blood, 361 
obtaining, 14 
precipitin in examination and identifica- 
tion of, 317 
specimens, collection of, 190 
test, forensic, 321 
complement fixation, technic, experi- 
mental work, 1150 
experimental work, 1134 
transfusion, 1077 
agglutination test preliminary to, ex- 
perimental work, 1133 INDEX 
1169 
Blood transfusion by superior longitudinal 
sinus in infants, 23 
cannula method, 1098 
choice of methods for, 1092 
citrate methods, 1093, 1099 
defibrination method, 1094 
direct artery-to-vein anastomosis, 1092 
vein-to-vein anastomosis, 1092 
fate of transfused blood, 1097 
hemolysin test preliminary to, experi- 
mental work, 1133 
history, 1077 
in acidosis, 1089 
in acute poisoning, 1089 
in aplastic anemia, 1087 
in athrepsia, 1090 
in chronic purpura, 1084 
in debilitated states, 1089 
in diabetic coma, 1089 
in diseases of blood, 1082 
in eclampsia, 1089 
in hemophilia, 1083 
in hemorrhage, 1079-1081 
due to childbirth, 1080 
due to dysentery, 1081 
due to gastric and duodenal ulcers, 
1081 
due to injuries of abdominal organs, 
1080 
due to placenta praevia, 1080 
due to pulmonary tuberculosis, 1081 
due to rupture of liver and spleen, 1080 
due to ruptured tubal pregnancy, 1080 
due to typhoid fever, 1081 
in acute yellow atrophy of liver, 1088 
in cirrhosis of liver, 1088 
in jaundice, 1088 
secondary to severe infections in treat- 
ment of septicemia, etc., 1088 
in hemorrhagic diseases, 1082 
in leukemia, 1088 
, in melena neonatorum, 1082 
in morbus maculosus neonatorum, 1082 
in pernicious anemia, 1084-1087 
vomiting of pregnancy, 1089 
in postoperative hemorrhage, 1079 
in pregnancy, 1089 
in purpura haemorrhagica, 1084 
in secondary anemia, 1079 
in shock, 1082 
in treatment of streptococcus septicemia, 
896 
indications, 1077 
intramuscular injections, 1108 
Kimpton-Brown method, 1103 
Kolmer’s citrate method, 1100 
Lewisohn’s citrate method, 1099 
Lindemann syringe cannula method, 1105 
methods, 1077, 1098 
of infants, 1107 
paraffined tube methods, 109 
posttransfusion treatment, 1097 
preliminary compatibility blood tests, 
1095 
to surgical operations, 1*082 
preparation of patient and donor, 1097 
reactions following, 1090 
selection of donors, 1094 
Blood transfusion, suture method, 1098 
syringe method, 1092, 1099 
syringe-cannula method, 1092, 1099 
technic of, 1099 
with preserved red blood-corpuscles, 1098 
whole, bactericidal activity of, versus 
plasma and serum, 361 
Zingher’s method, in treatment of scarlet 
fever, 900 
Blood-corpuscles and blood-serum, obtaining, 
14 
red. See Erythrocytes. 
Blood-pressure as guide in intraspinal injec- 
tion of serum, 850 
Blood-serum. See Serum. 
Blood-stains, complement fixation for identi- 
fication of, 561 
Kolmer’s first method, 561 
new method, 565 
differentiation of, by anaphylaxis reaction, 
642 
medicolegal identification of, human blood 
grouping for, 298 
precipitins in examination and identifica- 
tion of, 317 
Teichmann hemin crystal test for, 321 
Body fluids, distribution of antibodies in, 149 
relation of, to phagocytosis, 133 
immune, 338 
reaction, 626 
Bones, identification of, precipitin test in, 333 
precipitins for examination and identifica- 
tion of, 320 
tuberculosis of, tuberculin reaction in diag- 
nosis of, 691 
treatment, 1073 
Bordet-Gengou phenomenon of complement 
fixation, 351, 425. See also Complement 
fixation. 
Bordet’s amboceptor, 428 
theory of interaction of hemolysin, com- 
plement and corpuscles, 392 
of mechanism of action of amboceptor 
and complement, 354 
Boric acid, sodium oleate, and antipneumo- 
coccus serum in pneumococcus meningitis, 
963 
Botulinus antitoxin, 220 
bacillus, 100 
serum prophylaxis of, 878 
Botulism, 100 
bacillus, 878 
toxin, 100 
experimental work, 1114 
special properties, 100 
Bouillon filtrate tuberculin, preparation of, 
1054 
Bovine colloid, 262, 335 
Box for drying serums, 57 
Breiger’s method of isolating bacterial 
ptomains, 116 
Bronchi as foci of infection, 79 
Bronchial asthma, allergic, 651 
Bronchitis, chronic, vaccine treatment, 1004 
skin tests for bacteria, fungi, and protozoa 
in, 684 
Bronchopneumonia, etiology of, 917 
of influenza, treatment of, 917 INDEX 
1170 
Bronchopneumonia of influenza, treatment, 
with antipneumococcus serum, 919, 
920 
with human convalescent serum and 
blood, 919 
with non-specific agents, 920 
streptococcus, treatment, 917 
Bronfenbrenner’s modification of Abderhal- 
den’s ferment tests, 243 
theory of protective ferments in preg- 
nancy, 234 
Bruck’s nitric acid reaction in syphilis, 519 
Buboes, non-specific agents in, 984 
vaccine treatment, 984 
Bubonic plague, Bacillus pestis in vaccine for 
immunization against, 795 
bacterial precipitins in, 316 
Buchner’s theory of mechanism of action of 
complement and amboceptor, 353 
Butyric acid reaction, Noguchi’s, in syphilis, 
518 
Cadaver serum in Wassermann reaction, 
testing, 446 
Cadaverin, 116 
Calf cholera, serum prophylaxis of, 877 
Callison’s method of counting bacterial vac- 
cines, 194 
Calmette’s conjunctival tuberculin test, 689, 
698 
Cancer. See Carcinoma. 
Canine distemper, vaccination against, 833 
rabies, vaccination against, 834 
Cannula, Lindemann, 1106 
Capillary permeability, increased, theory of 
Starkenstein, 886 
pipet for counting bacterial vaccine, 192 
for opsonic index determination, 179 
method of measuring bactericidal power 
of blood, 375 
of sealing, 181 
Capsules, blood, Wright’s, making of, 6 
Carbuncles, non-specific treatment, 890 
serum treatment, 889 
treatment, 888 
vaccine treatment of, 889 
Carcinoma a contraindication to active im- 
munization, 761 
anaphylactic reactions in, 719 
autolysates in, 1031 
complement fixation in, 560 
diagnosis of, anaphylaxis reaction in, 642 
cytolytic, of Freund and Kaminer, 572 
serum treatment, 1031 
vaccine treatment, 1031 
value of Abderhalden’s ferment tests in, 245 
venom hemolysis in, 419 
Carriers, bacillus, milk injections for, 905 
diphtheria, antitoxin treatment of, 930 
non-specific agents in treatment of, 933 
relation of Schick test to, 745 
treatment with vaccines, 933 
meningococcus, 716 
typhoid, agglutination reaction for detec- 
tion of, 266 
vaccine treatment of, 903 
Castellani, saturation test of, 289 
Cataract extractions, infections following, 
vaccine treatment, 1017 
Catarrh, autumnal, 225 
Cats, active sensitization of, 617 
anaphylaxis in, 603 
asthma, treatment, 734 
symptoms of rabies in, 785 
Cattle, anaphylaxis in, 604 
Bang’s disease among, abortin test, 711 
plague, serum prophylaxis of, 878 
treatment, 1040 
Cells, functions of, 136 
of excreting glands, normal defenses of, 
against bacterial invasion, 66 
types, relation to infection and phagocy- 
tosis, 125 
Cellular colloidal nature of allergy, 632 
resistance, theory of, 123 
stimulation theory of Weichardt and Doll- 
ken, 885 
theory of allergy, 624 
of immunity, 122 
Cellulitis, streptococcus, vaccine treatment, 
892 
Centrifuge, 1 
care of, 1 
electric, 2 
Cerebral syphilis, Wassermann reaction in, 
510 
Cerebrospinal fever. See Meningitis, meningo- 
coccus. 
fluid, disposal of, after spinal puncture, 30 
examination chart for recording results, 
31 
for Wassermann test for syphilis, 444 
collecting, 446 
measuring pressure of, in spinal punc- 
ture, 26 
method of securing, 25 
normal pressure of, 27 
pressure of, conditions in which in- 
creased, 27 
meningitis, agglutination reaction in, 269 
vaccination, 817 
Certificates of vaccination, 779 
Cervicitis, gonococcus vaccine in, 981 
Chart, opsonic index, 188 
Chauveau’s retention theory of immunity, 122 
Chemical coagulating reactions in syphilis, 
517 
properties of exotoxins, 93 
substances, hemolysis by, 383 
Chemistry, colloidal, reactions of immunity 
and, analogy between, 579 
Chemotaxis, 127 
experimental work, 1121 
immunity, theory of, 123 
negative, 127, 129, 760 
experimental work, 1122 
positive, 127 
experimental work, 1122 
Chemotherapy, experimental work, 1156 
Chickenpox, anaphylactic reaction in, 717 
complement fixation in, 556 
immunity in* 820 
vaccination, 820 
Child, legitimacy of, human blood grouping 
for determining, 297 INDEX 
1171 
Childbirth, hemorrhage due to, transfusion 
for, 1080 
Cholecystitis, treatment of, 1009 
vaccine treatment, 1009 
Cholera, agglutination reaction in, 269 
Asiatic, serum treatment, 973 
treatment of, 973 
vaccine treatment, 975 
calf, serum prophylaxis of, 877 
hog, complement fixation in, 556 
serum prophylaxis of, 876 
treatment, 1039 
immunity in, 808 
typhoid, and paratyphoid bacilli and micro- 
coccus melitensis pentavaccine of, 
795 
tetravaccine of, 795 
vaccination, 807 
dosage of vaccine, 809 
effects, 809 
Haffkine’s vaccine, 808 
Kolle’s vaccine, 808 
preparation of vaccine, 808 
results, 809 
Strong’s vaccine, 808 
vaccine, preparation of, 809 
Cholesterolized and lecithinized alcoholic 
extract of heart muscle for Wassermann 
reaction, 457 
Chorea, autosera treatment, 1036 
Choroiditis, milk injections in, 1018 
Chosky’s serum treatment of plague, 972 
Cirrhosis of liver, hemorrhage in, transfusion 
for, 1088 
Citrate methods of blood transfusion, 1093, 
1099 
Kolmer’s, 1100 
Lewisohn’s, 1100 
Claypoole and Gay’s sensitized typhoid vac- 
cine, 794 
typhoidin, 713 
Cleaning of glassware, 7 
Clinical allergy, 650 
Coagulable blood, bactericidal activity of, 
Lacy-Heist method of determining, 378 
Coagulating reactions in syphilis, 517 
Coaguloreaction in syphilis, Hirschfeld and 
Klinger’s, 520 
Cobra hemotoxin, 105 
lecithid, 349 
venom, experimental work, 1116 
tests, technic, 416 
Cobralecithinase, 415 
Coccidiodal granuloma, anaphylactic reac- 
tion in, 717 
Coccidiosis, anaphylactic reaction in, 717 
Coctoprecipitins, 309 
Cold, common, 815. See also Common 
cold. 
rose, etiology of, 726 
Cole’s method of demonstrating endotoxin in 
pneumococci, 106 
Coley’s fluid in sarcoma, 1033 
Colitis, vaccine treatment, 1010 
Collapse during intraspinal injection of 
serum, 851 
Colles’ law, 824 
Wassermann reaction and, 511 
Collodion sacs for increasing virulence of 
bacteria, 70 
Colloidal chemistry, reactions of immunity 
and, analogy between, 579 
flocculation reactions in syphilis, 517 
gold reaction, Lange’s, 585 
experimental work, 1154 
precipitation reactions in syphilis, 520 
reactions, action of agglutinins based on 
580 
of antitoxins based on, 579 
of hemolysins based on, 581 
of precipitins based on, 580 
complement-fixation test as, 581 
technic of, 585 
solutions, 576 
suspensions, 576 
Colloids, absorption by, 578 
amorphous, 576 
bovine, 262, 355 
electric conductivity of, 576 
inorganic, hemolysis by, 384 
irreversible, 576 
nature and properties of, 575 
non-diffusible, 576 
osmotic pressure, 576 
physical structure, 577 
precipitation of, 577 
by colloids, 578 
by electrolytes, 577 
relation of, to immunity, 575 
reversible, 576 
shock, 606 
size, 577 
solutions, 576 
surface tension, 576 
varieties of, 575 
Colostrum, transmission of antibodies in, 152 
Colts, infectious arthritis of, vaccination 
against, 833 
joint-ill of, serum treatment, 1040 
Coma, diabetic, blood transfusion in, 1089 
Comer’s automatic pipet, 199 
Common cold, etiology and immunity, 815 
vaccination, 815, 816 
vaccine for, 816, 1002 
Complement, 133, 135, 141, 143, 338 
activity of different animal sera, experi- 
mental work, 1144 
and amboceptor, mechanism of action of, 
353 
Baumgarten’s theory, 353 
Bordet’s theory, 354 
Buchner’s theory, 353 
Ehrlich-Morgenroth theory of, 354 
Methnikoff’s theory, 355 
bactericidal activity without, 362 
definition, 342 
deflection of, 374 
destruction of, or interference with its 
activity, 434 
deviation, Neisser-Wechsberg phenomenon 
of, 352 
dominant, 339, 392 
effect of acids on, 343 
experimental work, 1143 
of alkalies on, experimental work, 1143 
of filtration on, 343 1172 
INDEX 
Complement, effect of filtration on, experi- 
. mental work, 1143 
of heat upon, 343 
of salts upon, 343 
of shaking upon, 344 
endocellular, 348 
fixation, 351, 420 
a colloidal reaction, 432, 580 
bacterial, technic, 539 
Kolmer’s first method, 541 
method based upon studies in 
standardization of technic, 543 
preparation of bacterial antigens, 
539 
by precipitates, 315 
clinical applications, 560 
collecting blood for, 444 
determination of antigen by, 560 
for differentiation of proteins, 538, 560 
Kolmer’s new method, 565 
for identification of bacterial antigens, 
564 
of blood-stains, 561 
bacteria and other proteins, Kol- 
mer’s new method, 565 
Kolmer’s first method, 561 
of meats, 563 
hemolysins in relation to, 405 
history of applications, 422 
importance of anticomplementary action 
of sera and antigens in relation to, 437 
in acute anterior poliomyelitis, 556 
in bacterial infections, 538 
in Basedow’s disease, 559 
in blastomycosis, 559 
in cancer, 560 
in chickenpox, 556 
in contagious abortion, 548 
in differentiation of microparasites, 538 
of protein, experimental work, 1150 
in dourine, 548 
in echinococcus disease, 558 
in exophthalmic goiter, 559 
in glanders, 547 
in gonococcus infections, 544 
antigen for, 545 
historic, 544 
practical value, 546 
specificity, 545 
in hog-cholera, 556 
in hydatid disease, 558 
in intestinal helminthiasis, 558 
in paratyphoid fever, 549 
in pertussis, 554 
in relation to natural hemolysins, 400 
in smallpox, 556 
in sporotrichosis, 559 
in standardization of immune serums, 560 
in syphilis, 442. See also Wassermann 
reaction in syphilis. 
in tuberculosis, 551 
experimental work, 1150 
non-specific reactions in syphilis, 554 
practical value, 554 
preparation of antigen, 553 
in typhoid fever, 549 
in vaccinia, 556 
in whooping-cough, 554 
Complement fixation, influence of duration of 
primary incubation on, experimental 
work,1148 
of temperature on, experimental work, 
1148 
mechanism of, 432 
natural hemolysins in relation to, 400 
non-specific, 437 
by human serums, experimental work, 
1147 
by normal human sera, 438 
rabbit-, dog-, and mule-sera, 438 
experimental work, 1147 
original, 421 
phenomenon of, experimental work, 1144 
practical applications, 440 
principles, 420 
proteotropic, 438 
qualitative, 488 
quantitative, 486 
and qualitative, relative value of, 489 
with antihuman and antichicken 
hemolytic systems, 491 
with antiox hemolytic system, 490 
factors in, 439 
importance of proper and accurate 
adjustment of hemolytic system, 
440 
necessity for proper adjustment of 
hemolytic system, 439 
for using optimum amount of an- 
tigen, 439 
relation of mechanism of precipitation in 
syphilis to, 533 
specific, in bacterial diseases, 538 
technic, 541 
for Wassermann reaction for syphilis, 447 
hemolysin and corpuscles, interaction of, 
390 
theory of Bordet, 392 
of Ehrlich and Morgenroth, 390 
of Metchnikoff, 393 
quantitative relationship between, 406 
hemolytic, experimental work, 1142 
titration of, experimental work, 1143 
historic, 342 
in serum, variation in amount of, 347 
inactivation of, 343 
experimental work, 1142 
multiplicity of, 346 
nature of, 348 
non-dominant, 339, 392 
preservation of, experimental work, 1144 
properties, 342 
reactivation of, experimental work, 1142 
relationship of leukocytic bacteriolysins 
to, 360 
resistance of, to desiccation, experimental 
work, 1140 
role of, in hemolysis, experimental work, 
1138 
source of, 345 
splitting, 350 
structure, 342 
thermolability of, experimental work, 1143 
Complement-fixing antibodies, kinds of an- 
tigens inducing production of, 425 
nature, 427 Complement-fixing antibodies, relation to 
precipitins, cytolysins, and albumin- 
olysins, 427 
specificity of, 429 
spirochetic, nature and specificity, 429 
Complementoids, 343 
Complications and relapses in typhoid fever, 
effect of vaccine therapy upon, 903 
Congenital mental deficiency, Wassermann re- 
action in, 512 
syphilis, Wassermann reaction in, 511 
Congestin, 622 
Conglutination, 262 
test with bacteria, technic of, 285 
Conglutinin, 262, 285, 355 
in relation to theories of cytolysis, 355 
Conjunctival tuberculin test of Calmette, 
689, 698 
dangers of, 692 
in animals, 703 
of Wolff-Eisner, 689, 698 
dangers of, 692 
in animals, 703 
Conjunctivitis, diphtheric, serum treatment, 
1015 
gonococcus, serum treatment, 1015 
milk injections in, 1016 
treatment of, 1016 
turpentine in, 1016 
vaccine treatment, 1016 
vasomotor, 653 
vernal, 653 
Contagious abortion, complement fixation in, 
548 
diseases, 61 
Contamination and invasion by micropara- 
sites, 58 
Convalescent serum and blood in treatment 
of scarlet fever, 899 
human serum for, methods for preparing, 
857 
Cornea, infected perforating wounds of, vac- 
cine treatment, 1017 
Corneal ulcers, vaccine treatment, 1016 
Corpuscles, complement and hemolysin, in- 
teraction of, 390 
quantitative relationship between, 406 
Cow disease, 749 
Cowpox, 749 
anaphylactic reaction in, 717 
vaccine, preparation of, 768 
animals, 769 
collection of virus, 769 
keeping, 771 
Noguchi’s method, 771 
seed virus, 768 
testing, 770 
vaccination, 769 
Cows, infectious abortion of, vaccination 
against, 832 
Craig’s modification of Wassermann reaction, 
500 
Crotin, 103 
Croton tiglium, 103 
Cryptogenic infections, 66 
Crystalloids, 575 
Culture-medium, addition of animal fluids to, 
for increasing virulence of bacteria, 70 
INDEX 
1173 
Curative immunization, 837 
influence of anaphylactic shock, 649 
Curcin, 103 
Cutaneous allergic test, technic of, 674 
reaction in typhoid fever, 712 
tuberculin reaction in animals, 705 
of Ligniere, 689, 699 
of von Pirquet, 689, 696 
versus intracutaneous tests for allergy, 673 
Cystitis, vaccine treatment, 1011 
Cytase, 131, 143, 338 
Cytolysins, 336, 566 
definition, 337 
formation of, 142, 337 
general considerations, 336 
in immunity, role of, 356 
nomenclature, 338 
relation to complement-fixing antibody, 427 
varieties of, 338 
Cytolysis, conglutinins in relation to theories 
of, 355 
mechanism of, 353 
Cytolytic cancer, diagnosis of Freund and 
Kaminer, 572 
Cytotoxic reactions, 572 
Cytotoxins, 140, 338, 566 
experimental production, 1110 
work, 1153 
in diagnosis, 572 
in immunity, role of, 572 
in therapeutics, 572 
methods of studying, 567 
natural, 566 
nature and general properties of, 566 
nomenclature, 566 
practical applications, 572 
preparation of, 566 
production of, 573 
specificity of, 568 
varieties of, 569 
Cytozyme, 534 
Dacrocystitis, chronic, vaccine treatment, 
1015 
Dandruffs, allergy to, skin tests for, 681 
Dangerous wounds from standpoint of tet- 
anus, 868 
Danysz’s enterovaccine in asthma, 1006 
in eczema, 994 
in psoriasis, 995 
Debilitated states, blood transfusion in, 1098 
Defibrination method of blood transfusion, 
1094 
Definitions, 59 
Deflection of complement, 374 
Delirium, tuberculin, 649 
Depression immunity, 162 
Dermatitis, allergic, treatment, 735 
herpetiformis, treatment of, 996, 997 
venenata, 103, 666 
Dermatoses, pyogenic, milk injections in, 993 
turpentine in, 993 
vaccine treatment, 993 
Desensitization, 634, 718 
in ivy poison, 735 
non-specific, 634, 636 
preseasonal, in hay-fever, value of, 727 1174 
INDEX 
Diphtheria anitoxin, sequelae, 930 
serum sickness from, 930 
standardizing, 99, 215 
experimental work, 1125 
subcutaneous administration of, 924 
treatment of carriers, 930 
of relapses, 930 
value of, 931 
bacilli, 97 
virulence and toxicity, method of testing, 
97, 1112 
carriers, antitoxin treatment of, 930 
non-specific agents in treatment of, 933 
relation of Schick test to, 745 
treatment with vaccines, 933 
diagnosis of, relation of Schick test to, 745 
immunity in, 860 
nasal, dosage of antitoxin in, 928 
natural antitoxin immunity to, 739 
immunity to, experimental work, 1119, 
1120 
nature of, 922 
non-specific agents in treatment of, 933 
normal horse-serum in treatment of, 933 
of eye, dosage of antitoxin in, 928 
of vulva, dosage of antitoxin in, 928 
of wounds, dosage of antitoxin in, 928 
Schick test for immunity to, 739. See also 
Schick test. 
serum prophylaxis of, 860 
dosage and administration, 860 
duration of immunity, 861 
indications, 861 
results in, 861 
Schick test in relation to, 861 
serum treatment of, 921. See also Diph- 
theria antitoxin. 
tests for antitoxin immunity to, 740 
tonsillar, dosage of antitoxin in, 928 
toxin, classification, 99 
experimental work, 1112 
limes death dose, 215 
zero dose, 216 
minimum lethal dose, 98 
production of, 210 
reaction in horses, 213 
special properties of, 97 
standardizing, 98 
testing, 213 
toxin-antitoxin vaccination against, von 
Behring method, 862 
age in relation to immuniza- 
tion, 864 
dangers and contraindications, 
865 
development of immunity, 865 
dose and method of administra- 
tion, 864 
duration of immunity, 865 
historical, 862 
mechansim of, 863 
practical value, 866 
preparation of toxin-antitoxin, 
863 
reactions, 864 
recommendations, 866 
test for immunity, 865 
toxoid, 99 
Desensitization, preseasonal, Kolmer’s method 
of, 730 
Walker’s method of, 730 
specific, 634 
mechanism of, 635 
methods for, 723 
production of, 634 
to serum, cases requiring, 722 
Desiccation, resistance of complement to, 
experimental work, 1140 
of hemolysins to, experimental work, 
1140 
D’Herelle’s dysentery bacteriolysant, 943 
intestinal bacteriophage in relation to in- 
fection, 72 
method of securing bacteriolytic agent, 247 
D’Herelle-Twort phenomenon of bacteriol- 
ysis, 246. See also Tworl-d'Herelle. 
Diabetes a contraindication to active immun- 
ization, 761 
mellitus, Wassermann reaction in, 504 
Diabetic coma, blood transfusion in, 1089 
•Diarrhea, allergic, 655 
Diathesis, pneumocatarrhal, 1005 
Diet as predisposing to infection, 74 
Digested bacterial protein, toxicity of, 114 
Digestive tract as portal of entrance for bac- 
teria, 64 
diseases of, treatment, 1008 
vaccine treatment, 1008 
Diphtheria, anaphylactic reactions in, 715 
antitoxin, 921 
action of, 923 
dosage, 927 
age influencing, 929 
day of illness influencing, 929 
general condition of patient influ- 
encing, 929 
in diphtheria of eye, 928 
of vulva, 928 
of wounds, 928 
in general treatment, 927 
in nasal diphtheria, 928 
in tonsillar diphtheria, 928 
large, importance of, 929 
repeating, 929 
•early use, importance of, 926 
for prophylaxis, 836 
immunizing animals with, 213 
intramuscular administration of, 924 
intravenous administration of, 924 
local application, 926 
method of concentrating, 214 
mortality after use, 931 
before use, 931 
in diphtheria influenced by, 926 
natural, Schick test for, 208 
oral administration, 925 
serum sickness after, 926 
preparation of, 921 
production of, 210 
collecting serum, 214 
immunizing animals, 213 
testing, 213 
purposes of, 922 
Tectal injection, 925 
Romer and Sames’ method of determin- 
ing small amounts of, 217 INDEX 
1175 
Diphtheria toxoid, experimental work, 1112 
treatment of, 921. See also Diphtheria 
antitoxin. 
Diphtheric conjunctivitis, serum treatment, 
1015 
keratitis, serum treatment, 1015 
Diphtherin, 715, 743 
Diplococcus pneumoniae, 908 
Direct method of obtaining hemagglutinins, 
300 
Disease, active immunization in treatment of, 
758 
anaphylaxis reaction in diagnosis of, 642 
bacterial anaphylaxis in relation to, 646 
bacteriolytic serums in treatment of, 380 
contagious, 61 
endemic, 59 
enzootic, 59 
epidemic, 59 
epizootic, 59 
infectious, 61 
active sensitization in, 644 
course of, 87 
definition, 59 
leukocytic extracts in treatment of, 881 
pandemic, definition, 59 
parasitic, causes of, 58 
production of, 84 
by animal parasites, 86 
by bacteria, 84 
by higher plants, 86 
prophylaxis of, active immunization in, 757 
protective ferments in, 231 
sera in prophylaxis and treatment of, 835 
sporadic, definition, 59 
Distemper, canine, vaccination against, 833 
Dixon’s tubercle-bacilli extract, 1054, 1066 
preparation of, 1054 
Dochez and Avery’s precipitin reaction with 
urine in pneumonia, 332 
Dog, active sensitization of, 618 
anaphylaxis in, 603 
asthma, treatment, 734 
immunization of, against rabies, 791 
intravenous injection of, 49 
obtaining large amounts of blood from, 40 
small amounts of blood from, 40 
serum, normal, non-specific complement 
fixation by, 438 
symptoms of rabies in, 785 
Doganoff and Moro’s tuberculin reaction, 689 
Dogwood, poison of, 103 
Dolken and Weichardt, cellular stimulation 
theory of, 885 
Dominant complements, 339, 392 
Dosis lethalis minimus, 98 
Dourine, agglutination test in, 270 
complement fixation in, 548 
Dreyer and Ward, sigma reaction of, in 
syphilis, 531 
Dreyer standardized agglutination test, 282 
modified, 284 
Dried food allergens, preparation of, 682 
paper form, preservation of serum in, 57 
Drill method of vaccination, 775 
Drugs, allergy to, 652, 669 
incubation period, 670 
skin tests for, 684 
Drugs, allergy to, symptoms, 670 
treatment, 738 
as antigens, 148 
effect of, upon phagocytosis, 129 
upon skin reactions, 679 
hypersensitiveness, incubation period, 618 
influence of, on antibodies and immunity, 
154 
Dunbar’s antiserum, pollantin, 732 
Duodenal ulcer, hemorrhage due to, trans- 
fusion for, 1081 
Dust asthmas, 652 
Dysentery antitoxin, administration and uses, 
941 
results, 941 
bacillary, agglutination reaction in, 268 
non-specific treatment, 944 
serum prophylaxis of, 871 
treatment of, 940 
results, 941 
treatment of, 840 
results, 941 
vaccine treatment, 943 
bacillus, 100, 871, 940 
bacteriolysant, D’Herelle’s, 943 
exotoxin, experimental work, 1114 
hemorrhage due to, transfusion for, 1081 
immunity in, 810 
toxin, 100 
special properties of, 100 
vaccination, 810 
Ear, diseases of, treatment, 1000 
secretions, collection of, 189 
tuberculosis of, tuberculin in, 1002 
reaction in diagnosis of, 691 
treatment, 1074 
Echinococcus disease, anaphylactic reactions 
in, 718 
bacterial precipitins in, 316 
complement fixation in, 558 
Eclampsia, blood transfusion in, 1089 
Ecphylaxis of Wright, 760 
Ecthyma, vaccine treatment, 993 
Eczema, 653, 654 
autoserum treatment, 994 
Danysz’s enterovaccine in, 994 
milk injections in, 994 
treatment of, 993 
turpentine in, 994 
vaccine treatment, 993 
Edema, angioneurotic, 653, 654 
malignant, bacillus of, 102 
Edestin, 231 
Eggs, precipitins for examination and identi- 
fication of, 320 
skin tests for allergy to, 683 
Ehrlich and Morgenroth’s theory of inter- 
action of hemolysin, complement, 
and corpuscles, 390 
of mechanism of action of complement 
and amboceptor, 354 
Ehrlich’s side-chain theory of immunity, 122, 
135 
theory of immunity and Metchnikoff’s 
theory, compatibility of, 142 
of tumor immunity, 1030 1176 
INDEX 
Electric centrifuge, 2 
Electrolytes, influence of, on agglutination, 
experimental work, 1131 
Embolic abscesses, experimental work, 1115 
Empyema, treatment of, 1006 
Emulsion, 576 
Encephalitis, lethargic, serum treatment, 1025 
Endemic disease, 59 
Endocarditis, acute streptococcus, vaccine 
treatment, 892 
Endocellular complement, 348 
Endocomplement, 348, 414 
Endolysins, 131, 361 
Endometritis, puerperal streptococcus, vac- 
cine treatment, 892 
Endotoxins, 84, 92, 106 
action of, 107 
definition, 106 
dysentery, experimental work, 1114 
experimental work, 1117 
methods of obtaining, 106 
nature of, 107 
properties of, 107 
toxicity of, 107 
Enteritis anaphylactia, 603 
Enterovaccine, Danysz’s, in asthma, 1006 
in eczema, 994 
in psoriasis, 995 
Enzootic disease, 59 
Enzymes and toxins, difference between, 95 
Epidemic disease, 59 
parotitis, serum prophylaxis of, 875 
Epididymitis, acute, antigonococcus serum 
in, 978, 979 
intravenous vaccine in, 981 
milk injections in, 983 
non-specific agents in, 981 
normal horse-serum in, 983 
Epididymo-orchitis, gonococcus and mixed 
vaccines in, 980 
Epilepsy, non-specific protein treatment of, 
1036 
Wassermann reaction in, 512 
Witte’s peptone in, 1036 
Epiphanin reaction, 588 
in cancer, 590 
in malignant disease, 590 
in pregnancy, 590 
in sarcoma, 590 
in syphilis, 590 
practical value, 590 
principle, 588 
reading results, 590 
specificity, 588 
technic, 588 
Epiphylactic response of Wright, 760 
Epitheliotoxin, 570 
Epizootic disease, 59 
Equilibrium, antigen-antibody, 151 
Equine influenza, vaccination against, 834 
Ereptase, 229 
Erysipelas, immunity in, 897 
leukocytic extract in treatment of, 898 
non-specific agents in treatment of, 898 
serum treatment of, 897 
treatment of, 897 
vaccine treatment of, 897 
Erythema, fixed pigmented, 670 
Erythema multiforme, 653, 655 
Erythrocytes for Wassermann reaction, 451 
method for measuring resistance to saponin, 
412 
obtaining, 11 
preservation of, 13 
preserved, transfusion with, 1098 
resistance of, to salt solution, experimental 
work, 1136 
variations in, 382 
tonicity test for measuring resistance to 
hypotonic saline solutions, 411 
washing, 11 
Erythrocytolysis, 381 
Erythroprecipitins, 314 
Ether as preventive of serum shock, 725 
Ethylhydrocuprein hydrochlorid and anti- 
pneumococcus serum in pneumococcus 
meningitis, 965 
Excision of malignant pustule, 968 
Exciting or toxic injection, 599 
Exhaustion theory of immunity, 122 
Exogenous infections, 61 
from micro-organisms in atmosphere, 62 
infecting placenta, 62 
from microparasites in water and foods, 61 
from skin wounds, 62 
from suctorial insects, 62 
toxins, 92 
Exophthalmic goiter, complement fixation in, 
559 
Exophylaxis, 167 
Exotoxins and toxinemia, 84 
chemical properties of, 93 
dysentery, experimental work, 1114 
nature of, 94 
precipitation of, 94 
selective action of, 96 
structure of, 94 
Experimental immunity, 1109 
infection, 1109, 1110 
Experiments, animal, 1110 
Exposure as predisposing to infection, 75 
Extracellular bacterial toxins, 92 
definition, 92 
toxins, 92 
Extracts, leukocytic, 359 
in treatment of disease, 881 
of erysipelas, 898 
of streptococcus infections, 897 
preparation of, 360 
tissue, hemolysis by, 384 
Exudates in serum therapy, 838 
inflammatory, bacterial precipitinogens in, 
316 
experimental work, 1135 
toxic, and ptomains, 85, 115 
Eye, diphtheria of, dosage of antitoxin in, 928 
diseases of, infection and immunity in, 1012 
treatment, 1012 
pneumococcus infections, serum treatment, 
1014 
streptococcus infections, serum treatment 
1014 
tetanus of, serum prophylaxis, 1015 
tuberculosis of, tuberculin reaction in diag- 
nosis of, 691 
treatment, 1074 INDEX 
1177 
Factor, reduction, 282 
Familial susceptibility, 73 
Fastigium period of infection, 90 
Favus, anaphylactic reaction in, 718 
Feathers, allergy to, skin tests for, 681 
Ferments, 113, 227 
and toxins, similarity between, 95 
bacterial, immunity to, 228 
in relation to infection, 227 
experimental work, 1127 
influence of non-specific agents upon, 882 
leukocytic and serum, immunity to, 229 
in relation to infection, 229 
protective, in disease, 231 
in pregnancy, Bronfenbrenner’s theory, 
234 * 
relation of lipoids to, 582 
to allergin, 628 
tryptic, 230 
vs. antibodies in disease, 231 
Fever period of infection, 90 
protein, 91, 114 
Film, blood, method of preparing, 182 
Filter, 55 
Berkefeld, 54 
cleaning and care of, 56 
Filtration, effect of, on complement, 343 
experimental work, 1143 
Fixateur. See Amboceptor. 
Fixative, 338. See also Amboceptor. 
Fixator, 355. See also Amboceptor. 
Flasks, sterilization of, 8 
Flexner and Amoss method for rapid pro- 
duction of antidysenteric serum, 221 
Flexner and Jobling’s method of preparing 
antimeningococcus serum, 947 
Flocculation reactions in syphilis, experi- 
mental work,1136. 
Fluid, Coley’s, in sarcoma, 1033 
Fluids, body, distribution of antibodies in, 149 
relation of, to phagocytosis, 133 
Focal and mixed infections, allergic skin tests 
in, 685 
infection, 78 
primary, 79 
etiology, 79 
secondary, 80 
vaccine treatment, 1008 
reactions, 641 
following injection of non-specific agents, 
884 
specific and non-specific, 641 
Foci of infection, primary, 79 
etiology, 79 
secondary, 80 
Fomites, 61 
Food allergens, method for preparing, 682 
allergy, 668 
desensitization by feeding, 737 
diagnosis, 669 
foods commonly producing, 669 
sensitization in, 668 
symptoms, 669 
skin tests for, 682 
treatment, 669, 736 
by desensitization, 737 
by elimination of foods, 736 
anaphylactogens, 608 
Food asthma, treatment, 734 
bacteria in, 61 
Forage poisoning, 100 
Force and Gay, typhoidin reaction of, 712 
Force and Stevens’ typhoidin, 713 
Forensic blood test, 321 
experimental work, 1134 
complement-fixation blood test, technic, 
experimental work, 1150 
Formol reactions in syphilis, 520 
Fowl asthma, treatment, 734 
Fraction, lecithin, 160 
Frambesia, Wassermann reaction in, 504 
Freezing, preservation of serum by, 56 
Freund and Kamerer, cytolytic cancer, diag- 
nosis of, 572 
Friedberger’s theory of allergy, 623 
of bacterial proteins, 113 
Frigo, 56 
Fruit allergens, preparation of, 683 
Fuld and Goss’s antitrypsin test, 236 
Fulminating infection, 91 
Fungi, allergy to, skin test for, 684 
infection with, 80 
Furunculosis, non-specific treatment of, 890 
of external auditory canal, treatment, 1000 
serum treatment of, 889 
treatment of, 888 
vaccine treatment of, 888 
Gat.eotti and Lustig’s plague vaccine, 806 
Gall-bladder infections, vaccine treatment, 
1009 
Gangrene, gas, bacteriology of, 871 
immunity in, 871 
serum prophylaxis of, 871, 873 
preparation of immune sera, 872 
treatment of, 939 
Gas bacillus, 102 
gangrene, bacteriology of, 871 
immunity in, 871 
serum prophylaxis of, 871, 873 
preparation of immune sera, 872 
treatment of, 939 
Gastric juice, germicidal powers of, 65 
ulcer, hemorrhage due to, transfusion for, 
1081 
Gastrotoxin, 570 
Gauze method of vaccination, 776 
Gay and Claypool’s sensitized typhoid vac- 
cine, 794 
typhoidin, 713 
and Force, typhoidin reaction of, 712 
and Minaker’s meningococcin, 716 
and Southard’s theory of allergy, 623 
Gel, definition, 576 
reaction, McDonagh, in syphilis, 519 
Genital organs as portal of entrance for bac- 
teria, 64 
Genito-urinary system, tuberculosis of, tuber- 
culin reaction in diagnosis of, 691 
Georgi-Sachs’ reaction in syphilis, 529 
Germicides, local injection in malignant 
pustule, 968 
Gingivitis, chronic, vaccine treatment, 1008 
Glanders, agglutination reaction in, 268 
bacterial precipitins in, 316 1178 
INDEX 
Glanders, complement fixation in, 547 
immunity in, 998 
mallein in, 999 
reaction for, 710. See also Mallein 
reaction. 
treatment of, 998 _ 
vaccine prophylaxis in, 998 
treatment, 998 
Glands, tuberculosis of, tuberculin reaction in 
diagnosis of, 691 
Glassware, cleaning of, 7 
Globulins, antitoxin, isolating, method of 
concentrating serum by, 214 
Noguchi butyric acid test for, 518 
Glucosids, 104 
Goats, intravenous injection of, 49 
Goiter, exophthalmic, complement fixation 
in, 559 
Gonococcin, 715 
Gonococcus conjunctivitis, serum treatment, 
1015 
infections, anaphylactic reactions in, 714 
complement fixation in, 544 
antigen for, 545 
historic, 544 
practical value, 546 
specificity of, 545 
immunity in, 977 
serum treatment, 977 
treatment of, 977 
turpentine in, 983 
reaction, technic, experimental work, 1150 
septicemia, antigonococcus serum in, 978 
vaccines in acute urethritis, 980 
in arthritis, 980, 981 
in cervicitis, 98i 
in chronic urethroprostatitis, 980 
in epididymo-orchitis, 980 
in vaginitis, 981 
Gonorrhea, acute, antigonococcus serum in, 
979 
bacterial precipitins in, 316 
Gonorrheal complications, normal horse- 
serum in, 983 
urethritis, acute, gonococcus and mixed 
vaccines in, 980 
Goodal’s subcutaneous method of vaccina- 
tion, 776 
Goss and Fuld’s antitrypsin test, 236 
Graduated pipets, 4 
Graminol, Weichhardt’s, in treatment of 
hay-fever, 732 
Granuloma, coccidioidal, anaphylactic reac- 
tion in, 717 
Gravity method for intraspinal injection of 
serum, 852 
for intravenous injection of serum, 847 
Grippe, 815 
Group immune bodies, 340 
Gruber-Widal reaction, 255 
in typhoid fever, experimental work, 
1128, 1129 
Guinea-pig, active sensitization of, 617 
anaphylaxis in, 600 
intracardial injection, 50 
intraperitoneal injection, 52 
intravenous inoculation, 46 
obtaining small amounts of blood from, 32 
Guinea-pig, securing large amounts of blood 
from, 37 
Gums as foci of infection, 79 
Haffkine’s cholera vaccine, 808 
plague vaccine, 806 
dosage, 807 
effects, 807 
results, 807 
Hair, allergy to, skin tests for, 681 
anaphylactogens, 608 
Half-parasites, 109 
Hamburger and Moro’s theory of allergy, 622 
Haptines, 136 
Haptophore group of toxin, 139 
Harris method of preparing rabies vaccine, 
789 
Hay-fever, 608, 662 
diagnosis, 665 
etiology of, 726 
historical, 663 
nature of, 663 
pollens causing, 664 
. preseasonal desensitization in, Kolmer’s 
method, 730 
value of, 727 
Walker’s method, 730 
prevention of, 727 
seasonal treatment, 732 
seasons, 727 
skin test for pollens causing, 680 
specific treatment, 665 
symptoms, 665 
treatment of, 725, 734 
with bacterial vaccines, 732 
with sera, 732 
vaccine treatment, 1003 
Heart muscle, alcoholic extract of, cholesterol- 
ized and lecithinized for Wassermann re- 
action, 457 
Heat, effect of, upon agglutinins, experimental 
work, 1131 
upon anticomplementary substances, 435 
upon complement, 343 
Hecht-Weinberg-Gradwohl modification of 
Wassermann reaction, 502 
Heist-Lacy method of determining bacteri- 
cidal activity of coagulable blood, 378 
Helminthiasis, intestinal, complement fixa- 
tion in, 558 
Hemagglutination, influence of temperature 
upon, 296 
substances causing, 291 
Hemagglutinins, 255, 291 
bacterial, 102 
experimental work, 1133 
heterologous, 255, 292 
in human blood, 295 
homologous, 255 
immune, 292, 296 
microscopic test for, 300 
direct method, 300 
grouping of blood, 301 
Vincent’s open slide method, 303 
natural, 292 
normal, 292 
serum, 292 INDEX 
1179 
Hemagglutinins, serum, kinds of, 292 
Hematoprecipitins, 306, 317 
Hemin crystal test, Teichmann’s, for blood- 
stains, 321 
Hemoglobinuria, 102 
paroxysmal, 398 
serum diagnosis of, 410 
Hemolysinogens, nature of, 388 
Hemolysins, 140, 172, 338, 381 
action, mechanism of, experimental work, 
1140 
activity, mechanism of, experimental work, 
1140 
amount absorbed by corpuscles, experi- 
mental work, 1142 
antihuman, production of, 401 
titration of, experimental work, 1149 
antisheep, production of, 402 
titration of, experimental work, 1138 
bacterial, 102 
method for determining, 413 
complement and corpuscles, interaction of, 
390 
theory of Bordet, 392 
of Ehrlich and Morgenroth, 390 
of Metchnikoff, 393 
quantitative relationship between, 406 
dissociation of, experimental work, 1142 
distribution of, 394 
experimental production, 1110 
for Wassermann test for syphilis, 450 
immune, 387 
method of titration, 407 
in relation to complement fixation, 405 
macroscopic tests for, 298 
pipet method, 299 
methods for removing from serum, 409 
natural, 394 
experimental work, 1141 
in relation to complement-fixation reac- 
tions, 400 
method for removal of, from active sera, 
409 
from inactivated sera, 409 
of determining in serum, 410 
removal of, experimental work, 1141 
resistance of, to heat, experimental work, 
1141 
normal, 387, 394 
practical applications, 407 
production of, 400 
by heterologous antigens, 389 
properties of, 403 
rapidity of absorption by corpuscles, ex- 
perimental work, 1141 
removal from serum by absorption, 399 
role of, in hemolysis, experimental work, 
1138 
resistance of, to desiccation, experimental 
work, 1140 
serum, 387 
definition, 387 
historic, 387 
nature, 389 
nomenclature, 387 
source of, 404 
specific, 387 
specificity of, 394 
Hemolysins, specificity of, experimental work, 
1139 
structure of, 390 
test preliminary to blood transfusion, ex- 
perimental work, 1133 
thermolabile, 403 
thermostability of, experimental work, 1139 
Hemolysis, 381 
and bacteriolysis, analogy between, 393 
bacterial, experimental work, 1136 
by acids, 383 
by animal secretions, 386 
by autolytic products, 384 
by bacterial poisons, 385 
by chemical substances, 383 
by hypertonic saline solutions, 381 
by hypotonic saline solutions, 381 
by inorganic colloids, 384 
by mechanical agitation, 381 
by photodynamic substances, 384 
by salts, 383 
by serum, 386 
by temperature changes, 381 
by tissue extracts, 384 
by vegetable poisons, 384 
by water, 381 
non-specific, experimental work, 1136 
relation of lipoids to, 583 
r6le of complement in, experimental work, 
1138 
of hemolysin in, experimental work, 1138 
saponin, experimental work, 1136 
serum, in vitro, experimental work, 1137 
in vivo, experimental work, 1137 
mechanism of, 390 
quantitative factors in, experimental 
work, 1138 
venom, 414. See also Venom hemolysis. 
Hemolytic activity of acids, experimental 
work, 1137 
of alkalies, experimental work, 1137 
agents, kinds of, 381 
amboceptor, 338 
anemia, 405 
antihuman and antichicken system in 
qualitative and quantitative complement 
fixation, 491 
antiox system in quantitative and qualita- 
tive complement fixation, 490 
complement, experimental work, 1142 
titration of, experimental work, 1143 
titration of antigens for Wassermann re- 
action, 462 
Hemophilia, blood injections in, 1018 
transfusion in, 1083 
serum treatment, 1018 
Hemopsonins, 172 
experimental work, 1123 
Hemorrhage after operations, treatment of, 
1019 
blood transfusion in, 1079-1081 
due to childbirth, transfusion for, 1080 
to dysentery, transfusion for, 1081 
to gastric and duodenal ulcers, trans- 
fusion for, 1081 
to injuries of abdominal organs, trans- 
fusion for, 1081 
to placenta prsevia, transfusion in, 1081 1180 
INDEX 
Hemorrhage due to pulmonary tuberculosis, 
transfusion for, 1081 
to rupture of spleen and liver, trans- 
fusion for, 1080 
to ruptured tubal pregnancy, transfusion 
for, 1080 
to typhoid fever, transfusion for, 1081 
in acute yellow atrophy of liver, trans- 
fusion for, 1088 
in cirrhosis of liver, transfusion for, 1088 
in disease, treatment of, 1019 
in jaundice, transfusion for, 1088 
postoperative, blood transfusion in, 1079 
retinal, blood injections in, 1018 
treatment of, 1018 
saline solution in, 1079 
secondary to severe infections in treatment 
of septicemia, etc., transfusion for, 1088 
standard, 1080 
treatment of, 1018 
Hemorrhagic diseases, blood transfusion for, 
10.82 
septicemia, serum prophylaxis, 1040 
treatment, 1040 
vaccination, 834 
Hemorrhagin, 105 
Hemo toxin, 101 
bacterial, 102 
cobra, 105 
Hepatitis, treatment of, 1009 
vaccine treatment, 1009 
with ascites, autoserum treatment, 1010 
Hepatotoxin, 338, 570 
Herman-Perutz reaction in syphilis, 522 
Heterologous antigens, production of he- 
molysins by, 389 
hemagglutinins, 255, 292 
in human blood, 295 
Heterolysins, 387 
Heterophile antigen and antibody, 159 
Heterovaccines intravenously in treatment of 
typhoid fever, 904 
Hirschfeld and Klinger’s coaguloreaction in 
syphilis, 520 
Hitchens’ syringe, 216 
Hog, anaphylaxis in, 604 
cholera, complement fixation in, 556 
serum prophylaxis of, 876 
treatment, 1039 
obtaining large amounts of blood from, 40 
Holzmann and Much’s psychoreaction in 
venom hemolysis, 418 
Homologous hemagglutinins, 255 
Hopkins’ method of counting bacterial vac- 
cines, 194 
tube for standardizing bacterial vaccine, 
197 
Horse, anaphylaxis in, 604 
and mule, strangles of, vaccination against, 
834 
asthma, horse-serum in, 722 
treatment, 734 
asthmatics, 657 
diphtheria toxin reaction in, 213 
intravenous injection of, 48 
obtaining large amounts of blood from, 41 
small amounts of blood from, 40 
Horse-serum asthma, treatment, 734 
Horse-serum in horse asthma, 722 
in psoriasis, 995 
normal, in epididymitis, 983 
in gonorrheal complications, 983 
in treatment of diphtheria, 933 
sensitization by, prevention of, 722 
serum disease from, 655 
syphilis, complement fixation in, 548 
Human allergy, general lesions and symp- 
toms of, 651 
treatment of, 721 
convalescent influenza serum in treatment 
of influenza of child, 906 
serum in treatment of meningitis, 960 
Humoral and phagocytic theories of immun- 
ity, compatibility of, 142 
poisons, relation of, to allergy, 629 
theories of allergy, 622 
of immunity, 135 
Huntoon’s antibody solution in treatment of 
pneumococcus pneumonia, 914 
Hydatid disease, complement fixation in, 558 
Hydrophobia, 782. See also Rabies. 
Hygienic laboratory method of quantitative 
estimation of bacteriotropins, 186 
of Wassermann reaction in syphilis, 494 
Hypersensitiveness, 594 
acquired, 599 
definition, 599 
drug, incubation period, 618 
natural, 599 
pollen, 639 
practical importance of, 597 
toxin, 645 
Hypersusceptibility, allergic skin reactions as 
indices of, 673 
Hyperthyroidism, serum treatment, 1038 
Hypothesis, Bail’s, 108 
of Welch, 76, 109 
Hypotonic salt solutions, tonicity test for 
measuring resistance of red blood-corpus- 
cles to, 411 
Idiosyncrasies, 608 
Illegitimacy, human blood grouping for de- 
termining, 297 
Immediate reaction in infection, 89 
Immune agglutinins, 255 
production of, 271 
intraperitoneal method, 272 
intravenous method, 271 
antibodies, 161 
bacteriolysins, 364 
bodies, 338 
group, 340 
partial, 340 
hemagglutinins, 292, 296 
hemolysins, 387 
method of titration, 407 
isohemagglutinins, 296 
opsonins, experimental work, 1123 
production of, 175 
properties of, 174 
specificity of, 173 
precipitins, 306 
serums, complement-fixation test in stand- 
ardization of, 560 1181 
INDEX 
Immune serums, preservation of, 53 
Immunity, 118 
acquired, 162, 167 
active, 119, 162, 167 
experimental work, 1120 
gained by accidental infection, 168 
by inducing an attack of disease by 
artificial inoculation, 168 
by vaccination, 168 
antibacterial, 168 
antitoxic, 168 
experimental work, 1120 
passive, 162, 169 
antitoxic, 169 
experimental work, 1120, 1121 
receptoric atrophy in, 168 
Vaughan’s theory, 170 
allergy in relation to, 643 
and antibodies, influence of alcohol on, 156 
of drugs on, 154 
of nutrition on, 154 
of x-rays on, 154 
transmission of, 152 
antibacterial, 838 
experimental work, 1119-1121 
antiblastic, 162, 228 
antitoxic, 838 
experimental work, 1120, 1121 
antitoxin, to diphtheria, tests for, 740 
athreptic, 166, 1030 
bacteriolysins in, rdle of, 364 
cellular theory of, 122 
chemotaxis, theory of, 123 
cytolysins in, role of, 356 
cytotoxins in, role of, 572 
definition, 119 
depression, 162 
exhaustion theory of, 122 
experimental, 1109 
history, 120 
how transmitted, 153 
humoral theory of, 135 
in acute anterior poliomyelitis, 1020 
in anthrax, 966 
in arthritis, 985 
in chickenpox, 820 
in cholera, 808 
in common cold, 815 
in diphtheria, 860 
Schick test for, 739. See also Schick 
test. 
in diseases of eye, 1012 
in dysentery, 810 
in erysipelas, 897 
in gas gangrene, 871 
in glanders, 998 
in gonococcus infections, 977 
in influenza, 811 
in leprosy, 999 
in Malta fever, 1025 
in measles, 821 
in meningococcus meningitis, 817 
in mumps, 875 
in paratyphoid fever, 792 
in pertussis, 816 
in plague, 805 
in pneumonia, 802 
in rabies, 783 
Immunity in scarlet fever, 818 
in smallpox, 766 
in staphylococcus infections, 888 
in streptococcus infections, 892 
in syphilis, 822 
in tetanus, 867 
in tuberculosis, 1042 
in tumors in relation to biologic therapy, 
1026 
in typhoid and paratyphoid fever, 792 
in typhus fever, 801 
in whooping-cough, 816 
in yellow fever, 822 
infection, 162 
kinds of, 162, 170 
mother, in tuberculosis, 1042 
natural, 119, 162, 163 
antitoxin, to diphtheria, 739 
definition, 163 
individual, 164 
influence of temperature on, experi- 
mental work, 1119 
mechanism of, 164 
phagocytosis in, 165 
experimental work, 1118 
racial, 164 
relative factors in, experimental work, 
1119 
skin in relation to, 167 
species, 163 
to diphtheria, experimental work, 1119, 
1120 
Vaughan’s theory, 170 
non-specific, 159 
passive, duration of, 839 
phagocytic theory and humoral theory, 
compatibility of, 142 
reaction, 90 
colloidal chemistry and, analogy be- 
tween, 579 
in smallpox vaccination, 779 
in vaccination, 777 
relation of anaphylaxis to, 647 
of colloids to, 575 
of infection to, 60 
of lipoids to, 582 
of microfungi toxins to, 103 
of phytotoxins to, 1Q3 
to Twort-d’Herelle phenomenon, 252 
retention theory of, 122 
revised theory and rdle of phagocytosis in, 
134 
r61e of bacterial agglutinins in, 263 
of opsonins in, 174 
of precipitins in, 315 
science of, 119 
side-chain theory of, Ehrlich’s, 122, 135 
skin in relation to, 991 
summary of kinds, 170 
theories of, 118 
obsolete, 122 
to bacterial ferments, 228 
to leukocytic and serum ferments, 229 
types of, 162 
Vaughan’s theory, 170 
Immunization, active, contraindications to, 
761 
history of, 748 1182 
INDEX 
Immunization, active, in rabies, 787 
in prophylaxis of disease, 757 
in treatment of disease, 758 
negative phase in, 765 
nomenclature of, 749 
principles of, 748 
prophylactic, 766. See also Vaccination. 
antibacterial, 838 
antitoxic, 838 
curative, 837 
of animals with diphtheria antitoxin, 213 
of dogs against rabies, 791 
passive, 860 
definitions, 835 
diphtheria, 860 
dosage and administration, 860 
duration of immunity, 861 
indications, 861 
results in, 861 
Schick test in relation to, 861 
duration of, 839 
for prevention of disease, 835 
for treatment of disease, 837 
principles of, 835 
varieties of, 837 
Pasteur on, 120 
prophylactic, in lower animals, 829 
in meningococcus meningitis, 873 
in tuberculosis, 1047 
toxin-antitoxin, 862 
typhoid-paratyphoid, agglutination reac- 
tion after, 266 
with typhoid-paratyphoid vaccine, prac- 
tical value of, 798 
Immuno-transfusion, 896 
Impetigo contagiosa, vaccine treatment, 993 
Inactivation of complement, 343 
Incision in malignant pustule, 968 
Increased capillary permeability theory of 
Starkenstein, 886 
Incubation period of infection, 87 
Incubator, opsonic, 181 
Index, opsonic, 175, 756. See also Opsonic 
index. 
phagocytic, 183 
Individual immunity, 164 
predisposition to infection, 73 
Infantile paralysis. See Poliomyelitis, acute 
anterior. 
Infants, blood transfusion of, 1107 
Infected material, procuring, for preparation 
of bacterial vaccines, 189 
Infection, 58 
accelerated reaction in, 89 
acute, 91 
vaccine therapy of, 759 
allergic skin reactions as indices of, 673 
allergy in relation to, 643 
avenues of, 62 
digestive tract, 64 
genital organs, 64 
placenta, 65 
respiratory organs, 63 
skin and mucous membranes, 63 
tissue susceptibility and, 70 
bacterial ferments in relation to, 227 
precipitin reactions in, 333 
chronic, 91 
Infection, complications of, 91 
cryptogenic, 66 
defensive mechanism of micro-organism in 
relation to, 75 
definition, 59 
d’Herelle’s intestinal bacteriophage in re- 
lation to, 72 
diet as predisposing to, 74 
endogenous, 61, 62 
exogenous, from micro-organisms in at- 
mosphere, 62 
by micro-organisms infecting placenta, 
62 
from microparasites in water and foods, 
61 
from skin wounds, 62 
from suctorial insects, 62 
experimental, 1109, 1110 
exposure as predisposing to, 75 
factors to be considered, 59 
focal, 78 
and mixed, allergic skin tests in, 685 
primary foci, 79 
secondary foci, 80 
vaccine treatment, 1008 
foci of, primary, 79 
etiology, 79 
secondary, 80 
following cataract extractions, vaccine 
treatment, 1017 
fulminating, 91 
general susceptibility in relation to, 73 
gonococcus, anaphylactic reactions in, 714 
immunity in, 977 
serum treatment, 977 
treatment of, 977 
turpentine in, 983 
grades of, 91 
immediate reaction in, 89 
immunity, 162 
in acute anterior poliomyelitis, 1020 
in anthrax, 966 
in arthritis, 985 
in diseases of eye, 1012 
intoxications as predisposing to, 75 
jaundice, serum treatment, 1038 
leukocytic and serum ferments in relation 
to infection, 229 
malignant, 91 
malnutrition as predisposing to, 74 
mechanism of, 68 
mixed, 77 
absorption agglutination test in, technic, 
289 
and single, absorption methods for dif- 
ferentiating between, 261 
tuberculin treatment, 1070 
numeric relationship of parasites to, 71 
overwork as predisposing to, 74 
period of convalescence, 90 
of decline, 90 
of fastigium, 90 
of high fever, 90 
of incubation, 87 
factors determining, 90 
of prodromal symptoms, 90 
pneumococcus, immunity in, 802 
of eye, serum treatment, 1014 INDEX 
1183 
Infection, prenatal, 65 
previous, as predisposing to infection, 74 
relapse in, 91 
relation of cell types to, 125 
of microfungi toxins to, 103 
of phytotoxins to, 103 
to immunity, 60 
remittent, 91 
secondary, 77 
selective tissue affinity in, 71 
sequelae of, 91 
single and mixed, absorption methods for 
differentiating between, 261 
skin in relation to, 991 
sources of, 60 
stages of, 87 
staphylococcus, immunity in, 888 
non-specific treatment of, 890 
serum treatment, 889 
vaccine treatment of, 889 
streptococcus, immunity in, 892 
leukocytic extracts in treatment of, 897 
non-specific agents in treatment of, 897 
of eye, serum treatment, 1014 
peptone in treatment of, 897 
serum treatment, 893 
treatment, 891 
vaccine treatment, 892 
systemic reaction to, 91 
torpid early reaction in, 89 
trauma as predisposing to, 75 
with animal parasites, 81 
with fungi, 80 
Infectious abortion of mares, vaccination 
against, 833 
serum prophylaxis of, 877 
arthritis of colts, vaccination against, 833 
diseases, 61 
active sensitization in, 644 
course of, 87 
definition, 59 
jaundice, vaccination against, 828 
Infestation, definition, 59 
Inflammatory exudate, bacterial precipitin- 
ogens in, experimental work, 1135 
Influenza bacillus, 102 
bronchopneumonia of, treatment of, 917 
with antipneumococcus serum, 919 
with antistreptococcus serum, 920 
with human convalescent serum and 
blood, 919 
with non-specific agents, 920 
equine, vaccination against, 834 
etiology of, in relation to vaccination, 
. 812 . 
immunity in, 811 
in animals, serum prophylaxis, 1041 
treatment, 1041 
of horses, vaccination against, 834 
serum, human convalescent, in treatment 
of influenza of child, 906 
serum treatment of, 906 
treatment of, 906 
vaccination, 811 
etiology in relation to, 812 
kinds of vaccines, 813 
value of, 814 
vaccine, 814 
Influenza, vaccine treatment of, 906 
Influenzal meningitis, 961 
serum treatment of, 960 
pneumonia, etiology, 917 
Infusoriotoxins, 572 
Inherited predisposition to infection, 73 
Inoculated persons, agglutination reaction in, 
266 
Inoculation, animal, 42. See also Animal 
inoculation. 
Inorganic colloids, hemolysis by, 384 
Insects, stings of, allergy to, 666 
suctorial, transmission of bacteria by, 62 
Interbody, 141, 338 
Intestinal bacteriophage, d’Herelle’s, in rela- 
tion to infection, 72 
helminthiasis, complement fixation in, 558 
toxemia, vaccine treatment, 1010 
Intoxicating dose, 599 
Intoxications as predisposing to infection, 75 
Intracardial injection of animals, 50 
of guinea-pigs, 50 
Intracellular toxins, 92 
Intracutaneous method of vaccination, 
Wright’s, 776 
tests for allergy more likely than cutaneous 
tests to elicit general reactions, 674 
technic, 677 
tuberculin test, 695 
in animals, 705 
of Mantoux, 689 
of Mendel, 689 
urine test, Wildbolz, 695 
versus cutaneous tests for allergy, 673 
Intramuscular injection of animals, 44 
of blood, 1108 
of serum, technic, 843 
Intraperitoneal inoculation of animals, 52 
of guinea-pig, 52 
of rabbit, 52 
method of production of immune agglu- 
tinins, 272 
Intraspinal injection of serum, anesthesia for 
852 
blood-pressure as guide, 850 
collapse during, 851 
gravity method, 852 
methods, 850 
syringe method, 855 
technic, 850 
Intravenous injection of animals, 44 
of children in serum treatment, 850 
of dog, 49 
of goats, 49 
of guinea-pig, 46 
of horse, 48 
of mice, 47 
of rabbit, 44 
of rats, 47 
of serum, gravity method, 847 
syringe method, 845 
technic, 845 
of sheep, 49 
method of production of immune agglutin- 
ins, 271 
vaccine in arthritis, 981 
in epididymitis, 981 
in prostatitis, 981 1184 
INDEX 
Invasion and contamination by micropara- 
sites, 58 
of bacteria, mechanism of, 66 
normal defenses against, 65 
Iridocyclitis, milk injections in, 1018 
Iritis, treatment of, 1017 
vaccine treatment, 1017 
Irreversible colloid, 576 
Iso-agglutinins, 291 
human, relation of different classifications, 
294 
Isocytotoxins, 569 
Isohemagglutinins, 255, 292, 293 
classification and grouping, 293 
immune, 296 
in lower animals, 295 
tests for, technic, 298 
transmission and influence of age, 293 
Isohemolysins, 397 
tests for, technic, 298 
Isolysins, 387 
Isoprecipitin, 306, 308 
Ivy poison, hypersensitiveness to, 666 
poisoning, desensitization in, 735 
Jaffe and Topfer’s method of measuring bac- 
tericidal power of blood, 372 
Jansky’s classification of human iso-agglu- 
tinins, 294 
Jatropha curcus, 103 
Jaundice, hemorrhage in, transfusion for, 1088 
infectious, serum treatment, 1038 
vaccination against, 828 
Wassermann reaction in, 504 
Jenner on vaccination, 120, 767 
sudden cuticular inflammation of, 717 
Jobling and Flexner’s method of preparing 
antimeningococcus serum, 947 
Joint-ill of colts, serum treatment, 1040 
Joints, tuberculosis of, tuberculin reaction in 
diagnosis of, 691 
treatment, 1073 
Jugular vein, anterior, obtaining blood from, 
21 
Kahn reaction in syphilis, 531 
Kaminer and Freund, cytolytic cancer, diag- 
nosis of, 572 
Kataphylaxis, 760 
Keidel tube for collecting blood, 20 
Kellogg’s method for determining presence of 
natural antitoxin in human serum as sub- 
stitute for Schick test, 217 
Keratitis, diphtheritic, serum treatment, 1015 
milk injections in, 1016 
treatment of, 1016 
vaccine treatment, 1016 
Kidney, diseases of, treatment, 1011 
tuberculosis of, tuberculin treatment, 1074 
Kimpton-Brown method of blood transfusion, 
1103 
tube, 1104 
Klausner’s water reaction in syphilis, 520 
Klinger and Hirschfeld’s coaguloreaction in 
syphilis, 520 
Koch’s lymph, 1051 
subcutaneous tuberculin test, 689, 693 
Kolle and Otto’s plague vaccine, 806 
Kolle’s cholera vaccine, 808 
method of counting bacterial vaccines, 194 
of preparing plague serum, 971 
plague vaccine, 806 
Kolmer and Weiss’s pneumotoxin, 716 
Kolmer’s bacterial complement-fixation test, 
based upon studies in standard- 
ization of technic, 543 
first method, 541 
citrate method of blood transfusion, 1100 
complement-fixation test based on results 
of studies in standardization of 
technic, experimental work, 1148 
for blood-stains, bacteria and other 
proteins, new method, 565 
for identification of blood-stains, first 
method, 561 
of proteins, new method, 565 
method of preseasonal desensitization, 730 
of testing virulence and toxicity of diph- 
theria bacilli, 97, 98 
modification of Wassermann reaction em- 
ploying multiple antigens, ex- 
perimental work, 1147 
varying amounts of complement, 
technic, 468, 491, 1149 
with multiple antigens, technic, 463, 
473 
with varying amounts of serum 
based upon studies in standard- 
ization of technic, 463, 478 
new antigen for Wassermann reaction, 457 
Korte and Stern’s method of measuring bac- 
tericidal power of blood, 370 
Kretz, paradox of, 610 
Kyes’ antipneumococcus serum, 913 
in pneumococcus meningitis, 964 
Lactoprecipitins, 306, 319 
Lactoserum, 304, 306, 319 
Lacy-Heist method of determining bac- 
tericidal activity of coagulable blood, 378 
Lamar and Simon’s modification of Wright’s 
method for opsonic index, 184 
Landon mercurial manometer, 27 
Landsteiner’s classification of human iso- 
agglutinins, 294 
Lange’s colloidal gold reaction, 585 
experimental work, 1154 
practical value of, 587 
preparation of colloid gold, 585 
principles, 585 
types of reactions, 586 
Laryngitis, chronic, vaccine treatment, 1004 
Larynx, tuberculosis of, tuberculin reaction 
in diagnosis of, 691 
Latent syphilis, Wassermann reaction in, 510 
Law, Colles’, 824 
Marfan’s, in tuberculosis, 1046 
Profeta’s, 824 
Lecithid, 583 
cobra, 349 
Lecithin fraction, 160 
Lecithinized and cholesterolized alcoholic ex- 
tract of heart muscle for Wassermann reac- 
tion, 457 INDEX 
1185 
Legitimacy of child, human blood grouping 
for determining, 297 
Leistenkern, 136 
Leistungskern, 136 
Leitenketter, 136 
Leprosy, agglutination test in, 271 
immunity in, 999 
nastin in, 1000 
serum treatment, 999 
treatment of, 999 
vaccine treatment, 999 
Wassermann reaction in, 504 
Leptospira icteroides, 822 
Lethargic encephalitis, serum treatment, 
1025 
Leukemia, blood transfusion in, 1088 
Leukocidin, 101, 102, 359 
Leukocytes, phagocytic activities of, varia- 
tions in, 126 
washed, in opsonic index, 179 
Leukocytic and serum ferments, immunity to, 
229 
in relation to infection, 229 
bacteriolysins, relationship of, to comple- 
ment, 360 
extracts, 359 
in treatment of disease, 881 
of erysipelas, 898 
of streptococcus infections, 897 
in tuberculous meningitis, 966 
preparation of, 360 
Leukocytotherapy of pneumococcus pneu- 
monia, 917 
Leukoprecipitins, 314 
Leukoproteases, 229 
Leukotoxins, 338, 570 
Lewisohn’s citrate method of blood trans- 
fusion, 1099 
Lichen planus, treatment of, 996, 997 
Ligniere’s tuberculin reaction, 689, 699 
Limes death dose of diphtheria toxin, 215 
zero dose of diphtheria toxin, 216 
Lindemann cannula, 1106 
syringe cannula method of blood trans- 
fusion, 1105 
Lipoid precipitin, 309 
Lipoids, acetone-insoluble, for Wassermann 
reaction, 457 
as antigens, 582 
relation of, to agglutinins, 583 
to anaphylaxis, 584 
to antiferments, 582 
to ferments, 582 
to hemolysis, 583 
to immunity, 582 
to opsonins, 582 
to phagocytosis, 582 
to Wassermann reaction, and precipita- 
tion, 584 
Lipovaccines, preparation of, 200 
Rosenow and Osterberg’s method, 201 
Whitemore and Fennel’s method, 201 
Liver, acute yellow atrophy of, hemorrhage 
. in, transfusion for, 1088 
cirrhosis of, hemorrhage in, transfusion for, 
1088 
rupture of, hemorrhage due to, transfusion 
for, 1080 
; Liver, syphilitic,- alcoholic extracts of, for 
Wassermann reaction, 455 
aqueous extracts of, for Wassermann re- 
action, 455 
Lobar pneumonia, mortality in, 911 
nature of, 908 
pneumococcus, treatment of, 907. See 
also Pneumonia, lobar, pneumococcus. 
Lobinol, 667 
Lockjaw. See Tetanus. 
Locomotor ataxia, treatment of, 1035 
Wassermann reaction in, 510 
Looped pipets, 3 
Loops, platinum, standardization of, 195 
Luetin, preparation of, 705 
reaction, 705 
method of application, 706 
negative, 706 
non-specific, 708 
normal, 706 
papular form, 706 
positive, 706 
practical value, 708 
preparation of luetin, 705 
pustular form, 707 
results, 707 
torpid form, 707 
Lumbar puncture, 25 
after-treatment, 30 
anesthesia in, 26, 852 
chart for recording results of examina- 
tion, 31 
contraindications, 25 
difficulties in, 856 
disposal of fluid, 30 
dry tap in, 856 
failure to enter subarachnoid space, 856 
hemorrhage in, 856 
measuring pressure of spinal fluid in, 26 
needles for, 28 
Babcock’s, 30 
obstruction of needle, 856 
preparation of patient, 25 
sequelae of, 857 
technic, 27 
value of, 25 
Lung puncture, 189 
Lupus vulgaris, tuberculin treatment, 1073 
Lustig and Galeotti’s plague vaccine, 806 
Lustig’s method of preparing plague serum, 
971 
Lymph, Koch’s, 1051 
Lymph-glands, tuberculosis of, tuberculin 
treatment, 1072 
Lymphadenitis, streptococcus, vaccine treat- 
ment, 892 
treatment, 890 
Lysins, 142 
Machine, shaking, 191 
Macrocytase, 131, 144 
Macrophages, 124 
experimental work, 1121 
Macroscopic tests for agglutinins 
molysins, 298 
pipet method, 299 
Malaria, agglutination test in, 271 1186 
INDEX 
Malaria, Wassermann reaction in, 504 
Malignant infection, 91 
pustule, serum treatment, 969 
treatment, 968 
tumors, autolysates in, 1031 
serum treatment, 1031 
vaccine treatment, 1031 
Mallein, 998 
in glanders, 999 
preparation of, 710 
reaction, 710 
ophthalmic, 711 
preparation of mallein, 710 
subcutaneous, 710 
Malnutrition as predisposing to disease, 74 
Malta fever, agglutination reaction in, 270 
immunity in, 1025 
micrococcus of, and typhoid, paraty- 
phoid, and cholera bacilli, pentavac- 
cine of, 795 
serum treatment, 1026 
treatment of, 1025 
vaccine treatment, 1026 
Manometer, mercurial, Landon, 27 
Mantoux’s tuberculin reaction, 689 
Maragliano’s serum in tuberculosis, 1075 
Marasmus, serum treatment, 1039 
treatment of, 1039 
Marcus modification of Muller and Joch- 
mann’s method of testing blood-serum, ex- 
perimental work, 1128 
Mares, infectious abortion of, vaccination 
against, 833 
Marfan’s law in tuberculosis, 1046 
Markl’s method of preparing plague serum, 
971 
Marmorek’s serum in tuberculosis, 1075 
Mastoid, diseases of, treatment, 1000 
Mastoiditis, acute, vaccine treatment, 1002 
chronic, vaccine treatment, 1002 
Maternal transmission of passive sensitiza- 
tion, 619 
McCampbell’s technic in opsonic index, 184 
McDonagh gel reaction in syphilis, 519 
McFarland nephelometer, 195, 196 
Measles, immunity in, 821 
serum prophylaxis of, 874 
vaccination, 821 
Measure of antitoxins, 225 
Meat adulteration, detection by anaphylaxis 
reaction, 642 
precipitin test for, 329 
poisoning, 100 
Meats, identification of, complement fixation 
for, 563 
precipitins in examination and identifica- 
tion of, 319 
Medicinal prophylaxis of anaphylactic serum 
shock, 724 
Medicolegal application of human blood 
grouping, 297 
identification of blood-stains, human blood 
grouping for, 298 
Meier-Porges reaction in syphilis, 522 
Meinicke’s reactions in syphilis, 522 
Melena neonatorum, blood injections in, 1018 
transfusion in, 1082 
serum treatment, 1018 
Melitensis, micrococcus of, and typhoid, para- 
typhoid and cholera bacilli, pentavaccine 
of, 795 
Membranes, mucous, and skin, active sen- 
sitization of, 638 
passive sensitization of, 637 
serous, tuberculosis of, tuberculin reaction 
in diagnosis of, 692 
Mendel’s tuberculin reaction, 689 
Meninges, mechanism of infection of, 945 
Meningitis, cerebrospinal, vaccination against, 
817 
influenzal, 961 
serum treatment of, 960 
meningococcus, immunity in, 817 
prophylactic immunization in, 873 
serum treatment of, 946 
cases of posterior basal meningitis, 
956 
with dry canal, 956 
with thick plastic exudate, 956 
chronic cases, 957 
early diagnosis and treatment, im- 
portance of, 952 
general plan for, 957 
results, 958 
serum sickness, 957 
subacute cases, 957 
types of meningococci causing, 944 
vaccine treatment of, 960 
pneumococcus, antipneumococcus serum 
and ethylhydrocuprein hydro- 
chlorid in, 965 
sodium oleate, and boric acid in, 
963 
diagnosis of, 962 
intravenous injections of serum in, 964 
Kyes’ antipneumococcus serum in, 964 
nature of, 962 
serum treatment, 962 
treatment of, 962 
importance of early diagnosis and 
drainage, 963 
precipitin test with spinal fluid in, 333 
serous, tuberculous and, Noguchi’s butyric 
acid reaction in, 518 
serum prophylaxis of, 873 
streptococcus, serum treatment, 965 
treatment with human serum, 960 
tuberculous, antimeningococcus serum in, 
965 
leukocytic extracts in, 966 
serum treatment, 965 
treatment of, 965 
tuberculin reaction in diagnosis of, 692 
Meningococcic vaccine, 818 
Meningococcin of Gay and Minaker, 716 
Meningococcus bacteremia, 945 
carriers, 716 
differentiation of, agglutination test in, 288 
meningitis, immunity in, 817 
prophylactic immunization in, 873 
serum treatment of, 944. See also 
Meningitis, meningococcus. 
vaccine treatment of, 960 
types of, 944 
Menstruation contraindication to typhoid 
vaccination, 796 INDEX 
1187 
Mental deficiency, congenital, Wassermann 
reaction in, 512 
diseases, value of Abderhalden’s ferment 
tests in, 245 
Mercurial manometer, Landon, 27 
Metabolism, influence of non-specific agents 
upon, 884 
Metchnikoff’s humoral theory of immunity, 
122 
theory of immunity and Ehrlich’s theory, 
compatibility of, 142 
theory of interaction of hemolysin, com- 
plement and corpuscles, 393 
of mechanism of action of amboceptor 
and complement, 355 
of phagocytosis, 122 
Mice and rats, anaphylaxis in, 604 
intravenous injection of, 47 
Micrococcus catarrhalis, 715, 812, 816 
melitensis and typhoid, paratyphoid, and 
cholera bacilli pentavaccine of, 795 
Microcytase, 131, 144 
Microfungi, toxins of, 103 
relation to infection and immunity, 103 
Micro-organism, defensive mechanism of, in 
relation to infection, 75 
morphologic and physiologic change in, 75 
Microparasites, 58, 60 
antigenic activity of, variation in, 752 
complement fixation in differentiation of, 
538 
invasion and contamination by, 58 
relation of numbers of, to infection, experi- 
mental, 1111 
Microphages, 124 
Microscopic test for hemagglutinins, direct 
method, 300 
grouping of blood, 301 
Vincent’s open slide method, 303 
Milk, bacteria in, 61 
casein allergen, preparation of, 683 
identification of, precipitin test in, 334 
injections for bacillus carriers, 905 
in acne, 993 
in albuminuric retinitis, 1018 
in arthritis, 983 
in bronchopneumonia of influenza, 920 
in choroiditis, 1018 
in conjunctivitis, 1016 
in eczema, 994 
in epididymitis, 983 
in iridocyclitis, 1018 
in iritis, 1017 
in keratitis, 1016 
in neuritis, 1039 
in orchitis, 983, 1037 
in prostatitis, 983 
in psoriasis, 995 
in pyelitis in children, 1012 
in pyogenic dermatoses, 993 
in scleritis, 1018 
in skin diseases, 997 
in trachoma, 1016 
in typhoid fever, 905 
in Weil’s disease, 1038 
precipitins in examination and identifica- 
tion of, 319 
Minaker and Gay’s meningococcin, 716 
Minimum lethal dose, 98 
Miostagmin reaction, 590 
methods for preparing antigens, 592 
practical value, 593 
principles, 590 
technic, 591 
experimental work, 1154 
Mixed and focal infections, allergic skin tests 
in, 685 
infection, 77 
tuberculin treatment, 1070 
Molecule of toxin and toxoid, theoretic struc- 
ture of, 139 
Monkey, obtaining large amounts of blood 
from, 40 
small amounts of blood from, 40 
Monofatty-acid-lecithin, 415 
Monovalent vs. polyvalent antitoxic sera, 208 
Morbid conditions as predisposing to infec- 
tion, 75 
Morbus maculosus neonatorum, blood trans- 
fusion in, 1082 
Morgenroth and Ehrlich’s theory of inter- 
action of hemolysin, complement and cor- 
puscles, 390 
Moro and Doganoff’s tuberculin reaction, 689 
Moro and Hamburger’s theory of allergy, 622 
Moro’s tuberculin reaction, 699 
Morphologic changes in micro-organisms, 75 
Moss’s classification of human iso-agglutin- 
ins, 294 
Mother, immunity of, in tuberculosis, 1043 
Much and Holzmann’s psychoreaction in 
venom hemolysis, 418 
Mucous membranes and skin, active sensi- 
tization of, 638 
passive sensitization of, 637 
Mules and horses, strangles of, vaccination 
against, 834 
Mule-serum, normal, non-specific comple- 
ment fixation by, 438 
Muller and Jochmann’s method of testing 
blood-serum, Marcus modification, experi- 
mental work, 1128 
Multiform rashes in serum disease, 660 
Mumps, immunity in, 875 
non-specific proteins in, 1037 
. serum prophylaxis of, 875 
Muscarin, 116 . 
Musculoprecipitins, 319 
Mushroom poisons, 104 
Myalgia in serum disease, 661 
Myco-agglutinins, 255 
Mycoprecipitins, 305 
Nasal diphtheria, dosage of antitoxin in, 928 
secretion, collection of, 189 
Nastin, 147 
in leprosy, 1000 
Natural aggressins, 109 
antitoxin immunity to diphtheria, 739 
immunity, 119 
Needle for lumbar puncture, 28 
Babcock, 30 
Negative chemotaxis, 127, 129, 760 
phase of vaccine therapy, 765 
Negri bodies, 783 1188 
INDEX 
Neisser-Wechsberg method of measuring bac- 
tericidal power of blood, 370 
phenomenon of complement deviation, 352 
experimental work, 1153 
Nephelometer, McFarland, 195, 196 
methods of counting bacterial vaccines, 195 
Nephritis, advanced, a contraindication to 
active immunization, 761 
contraindication to typhoid vaccination, 
796 
Nephrotoxin, 338, 570 
experimental work, 1153 
Neufeld’s quantitative estimation of bac- 
teriotropins, 184 
controls, 186 
culture, 185 
leukocytes, 185 
precautions, 186 
readings, 186 
serum, 185 
Simon’s method, 186 
test, 186 
Neuritis in serum disease, 661 
milk injections in, 1039 
vaccine treatment, 1038 
Neurorrhyctes hydrophobise, 783 
Neurotoxin, 338, 571 
Nitric acid reaction, Bruck’s, in syphilis, 519 
Noguchi’s butyric acid reaction in syphilis, 
518. 
globulin reaction, experimental work, 1135 
method of preparing cowpox virus, 771 
modification of Wassermann reaction, 496 
anticomplementary titration, 498 
antigen for, 498 
antigenic titration, 499 
complement for, 496 
experimental work, 1149 
fluid to be tested, 499 
hemolytic amboceptor, 496 
human corpuscles, 496 
test, 499 
titration of antigen, 498 
of dried amboceptor paper, 497 
of serum hemolysin, 497 
Nolf’s theory of allergy, 624 
Non-agglutinable species of bacteria, 261 
Non-diffusible colloids, 576 
Non-dominant complements, 339, 392 
Non-pathogenic bacteria, 60 
Non-protein antigens, 147 
biologic therapy, general reaction in, 881 
non-specific agents for, 880 
Non-specific agents, focal reaction following 
injection of, 884 
for non-protein biologic therapy, 880 
in treatment of bronchopneumonia of in- 
fluenza, 920 
of diphtheria, 933 
carriers, 933 
of erysipelas, 898 
of scarlet fever, 901 
of streptococcus infections, 897 
influence of, upon antibody production, 
883 
upon blood, 882 
upon ferments and antiferments, 882 
upon metabolism, 884 
Non-specific antigens, 146 
biologic therapy, 879 
general reaction, 881 
protein therapy, principles of, 879 
reaction, mechanism of, 885 
relation of, to therapeutic activity, 886 
treatment of abscesses, 890 
of bacillary dysentery, 944 
of carbuncle, 890 
of furunculosis, 890 
of pneumococcus pneumonia, 916 
of staphylococcus infections, 890 
of tetanus, 939 
of typhoid fever, 904 
albumoses intravenously, 904 
contraindications, 906 
heterovaccines intravenously, 904 
mechanism of action of agents in, 
905 
milk intramuscularly, 905 
precautions in, 905 
Normal serum in treatment of scarlet fever, 
901 
Nose, accessory sinuses, diseases of, treat- 
ment, 1002 
Numeric relationship of parasites to infection, 
71 
Nutrition, influence of, on antibodies and 
immunity, 154 
Nuttall’s method of obtaining large amounts 
of blood from animals, 33 
Oak, poison of, 103 
Oliver’s precipitin reaction with sputum in 
pneumonia, 333 
Omnicellular plasma-activation theory, 885 
Open slide method, Vincent’s, in microscopic 
test for hemagglutinins, 303 
Operations, hemorrhage after, treatment of, 
1019 
surgical, blood transfusion preliminary to, 
1082 
Ophthalmic mallein reaction, 711 
reaction in typhoid fever, 712 
Opsonic incubator, 181 
index, 175, 176, 756 
as diagnostic procedure, 187 
as guide to size and frequency of doses of 
bacterial vaccines, 187 
chart, 188 
definition, 176 
determining, experimental work, 1125 
in prognosis, 871 
limitation of method, 176 
McCampbell’s technic, 184 
practical value of, 187 
principle, 176 
purpose of method, 176 
Simon and Lamar’s modification of 
Wright’s method, 184 
technic, 178 
bacterial emulsion in, 178 
collection of patient’s and control 
serums, 178 
precautions in, 177 
the test, 179 
washed leukocytes in, 179 
Veitch’s technic, 184 INDEX 
1189 
Opsonification, susceptibility of bacteria to, 
174 
Opsonins, 134, 171 
action of, mechanism, experimental work, 
1124 
definition, 172 
effect of, on bacteria, 174 
experimental work, 1110, 1122 
historic, 171 
immune, experimental work, 1123 
production of, 175 
properties of, 174 
specificity of, 173 
natural, properties of, 174 
specificity of, 173 
nature of, 172 
normal, experimental work, 1122 
properties of, 172 
relation of lipoids to, 582 
role of, in immunity, 174 
source of, 174 
specificity of, experimental work, 1124 
Opsono, 171 
Orchitis, antigonococcus serum in, 978, 979 
milk injections in, 983, 1037 
non-specific proteins in, 1037 
Organ virulence, 77 
Organotropism, experimental work, 1156 
Original theory of phagocytosis, 124 
Osteoprecipitins, 320 
Otitis media, acute, vaccine treatment, 1000, 
1001 
chronic, vaccine treatment, 1000, 1001 
Otto and Kolle’s plague vaccine, 806 
O. T. tuberculin, preparation of, 1052 
Overproduction theory, Weigert’s, 137, 204 
Overstrain as predisposing to disease, 74 
Overwork as predisposing to disease, 74 
Ovoprecipitins, 320 
Ozena, vaccine treatment, 1003 
Pain, abdominal, allergic, 655 
Pandemic disease, definition, 59 
Paradox of Kretz, 610 
Paralysis, general, Wassermann reaction in, 
510 
infantile. See Poliomyelitis, acute anterior. 
Parasites, 50, 60 
animal, aggressiveness of, 82 
infection with, 81 
modes of, 82 
production of disease by, 86 
selective affinity of, 82 
half, 109 
numeric relationship of, to infection, 71 
partial, 109 
true, 109 
Parasitic causes of disease, 58 
fungi, infection with, 80 
Parasitotropism, experimental work, 1156 
Parasyphilitic diseases, Wassermann reaction 
in, 510 
Paratyphoid and typhoid bacilli, triple vac- 
cine of, 795 
fever, agglutination test in, 264 
in diagnosis of, in vaccinated per- 
sons, 267 
Paratyphoid fever, complement fixation in, 
549 
immunity to, 792 
vaccination, 792 
contraindications to, 795 
duration and degree of, 797 
frequency and number of inoculations, 
795 
method of inoculation, 794 
reactions in, 796 
recommendations, 800 
revaccination, 797 
value of, 798 
typhoid, and cholera bacilli and micrococ- 
cus melitensis pentavaccine of, 795 
tetravaccine of, 795 
vaccine, preparation of, 793 
sensitized, 793 
Paresis, treatment of, 1035 
Wassermann reaction in, 510 
Parotitis, epidemic, serum prophylaxis of, 875 
postoperative, 875 
Paroxysmal hemoglobinuria, 398 
serum diagnosis of, 410 
Partial parasites, 109 
Passive immunization. See Immunization, 
Passive. 
Pasteur on immunization, 120 
Pasteur’s exhaustion theory of immunity, 122 
method of preparing anthrax vaccine, 830 
rabies vaccine, 788 
treatment of rabies, 787 
Pasteurellosis of animals, 834 
Pathogenic bacteria, 60 
Pathogens, 60 
Pellagra, treatment of, 996, 997 
Pelvic tuberculosis, tuberculin reaction in 
diagnosis of, 691 
Pemphigus, treatment of, 996 
Pentamethylendiamin ,116 
Pentavaccine of typhoid, paratyphoid, and 
cholera bacilli and micrococcus melitensis, 
795 
Peptidase, serum, 229 
Peptone, 105 
in treatment of streptococcus infections, 897 
Witte’s, in asthma, 1006 
in epilepsy, 1036 
Percutaneous tuberculin reaction of Moro, 699 
of Moro and Doganoff, 689 
Perennial vasomotor rhinitis, 653 
Perethynol, 523, 524 
preparation of, 525 
Peritoneal exudate from mouse, precipitin 
reaction with, in pneumonia, 332 
Peritonitis, tuberculous, tuberculin reaction 
in diagnosis of, 692 
treatment, 1074 
Pernicious anemia, blood transfusion in, 1084- 
1087 
splenectomy in, 1087 
vomiting of pregnancy, blood transfusion 
in, 1089 
Pertussis, agglutination test in, 270 
anaphylactic reaction in, 717 
complement fixation in, 554 
immunity in, 816 
serum treatment, 975 1190 
INDEX 
Pertussis, treatment of, 975 
vaccine treatment, 816, 817, 975 
dosage and administration, 976 
Perutz-Herman reaction in syphilis, 522 
Pfaundler’s reaction, 254 
Pfeiffer’s bacillus, 811 
bacteriolytic test, 365 
experimental work, 1151 
in vivo in diagnosis of disease, 369 
technic, 369 
Phagocyte, 124 
varieties of, 124 
Phagocytic activities of leukocytes, variations 
in, 126 
and humoral theories of immunity, com- 
patibility of, 142 
count, 176 
index, 183 
mixture, 176 
Phagocytosis, effect of bacterial production, 
131 
of drugs on, 129 
experimental work, 1121, 1122 
in natural immunity, 165 
experimental work, 1118 
mechanism of, 132 
Metchnikoff’s theory of, 122 
relation of body fluids to, 133 
of cell types to, 125 
of lipoids to, 582 
results of, 131 
revised theory and r61e of, in immunity, 134 
spontaneous, 171, 185 
theory of, and cellular resistance and im- 
munity chemotaxis, 123 
historic, 123 
original, 124 
Phallin, 103 
Pharyngitis, acute, vaccine treatment, 1004 
Phenomena, anaphylactoid, 606 
Sach’s-Friedberger, 399 
Phlebotomy, 18 
in children, 18 
Phogosin, 127 
Photodynamic substances, hemolysis by, 384 
Phylactic power, 760 
Physiologic changes in micro-organisms, 75 
Phytagglutins, 255 
Phytoprecipitin, 304, 305, 313 
Phytosensitinogens, 608 
Phytotoxins, 83, 92, 103 
action of, 103 
experimental work, 1116 
nature of, 104 
relation to infection and immunity to, 103 
Pipets, 18 
automatic, Comer’s, 199 
capillary, for counting bacterial vaccine, 192 
for measuring bactericidal power of 
blood, 375 
for opsonic index determination, 179 
making of, 3 
method of sealing, 181 
rubber teats for, 3 
graduated, 4 
looped,3 
essential features of, 4 
making of, 2, 4 
Pipets, many stemmed, Wright’s, 379 
soiled, jar for holding, 6 
sterilization of, 6, 8 
Placenta as portal of entrance for bacteria, 65 
bacteria in, 62 
praevia, hemorrhage due to, transfusion for, 
1080 
Placental blood, method of obtaining, 23 
Placentin, 719 
Placentotoxin, 571 
Plague, agglutination test in, 270 
cattle, serum prophylaxis of, 878 
serum treatment, 1040 
immunity in, 805 
serum, administration of, 971 
dosage of, 971 
treatment, 970 
vaccination, 805 
dosage of vaccine, 807 
duration, 807 
Haffkine’s vaccine, 806 
dosage, 807 
effects, 807 
results, 807 
Kolle and Otto’s vaccine, 806 
Kolle’s vaccine, 806 
Lustig and Galeotti’s vaccine, 806 
preparation of vaccine, 806 
results, 807 
Temi and Bandi’s vaccine, 806 
vaccine, preparation of, 806 
Plants, allergy to, 666 
higher, production of disease by, 86 
toxins of, 103 
relation to infection and immunity, 103 
toxins, experimental work, 1112 
Plasma, blood, obtaining, 14 
Plasma-activation of Weichardt, 641, 686, 
884 
omnicellular theory, 885 
Plate culture method of measuring bacteri- 
cidal power of blood, 370 
Platinum loops, standardization of, 195 
Pleurisy, treatment of, 1006 
tuberculous, autoserum therapy, 1006, 1007 
tuberculin reaction in diagnosis, 692 
Plotz bacillus, 801 
Pneumocatarrhal diathesis, 1005 
Pneumococcus, agglutination test in differen- 
tiation of, 286 
infections, immunity in, 802 
of eye, serum treatment, 1014 
lobar pneumonia, leukocytotherapy in, 917 
non-specific treatment of, 916 
serum treatment of, 909. See also 
Pneumonia, lobar, serum treatment of. 
treatment of, 907 
with Huntoon’s antibody solution, 
914 
vaccine treatment of, 915 
meningitis, antipneumococcus serum and 
ethylhydrocuprein hydrochlorid 
in, 965 
sodium oleate, and boric acid in, 963 
diagnosis of, 962 
intravenous injections of serum in, 964 
Kyes’ antipneumococcus serum in, 964 
nature of, 962 INDEX 
1191 
Pneumococcus meningitis, serum treatment, 
962 
treatment of, 962 
importance of early diagnosis and 
drainage, 963 
pneumonia, 715 
vaccine treatment, 915 
protein, 715 
types of, in relation to serum treatment and 
mortality of pneumonia, 911 
in serum treatment of pneumococcus 
pneumonia, 909. 
vaccination, experimental, 803 
vaccines, dosage and administration, 804 
kinds of, 803 
preparation of, 804 
Pneumonia, agglutination test in, 270 
anaphylactic reactions in, 715 
experimental, 1110 
immunity in, 802 
influenzal, etiology, 917 
lobar, pneumococcus, leukocytotherapy in, 
917 
mortality in, 911 
nature of, 908 
non-specific treatment of, 916 
serum treatment, 909 
administration of serum, 912 
indications, 909 
results in, 912 
types of pneumococci in relation 
to, 911 
treatment of, 907 
with Huntoon’s antibody solution, 
914 
vaccine treatment of, 915 
of horses, vaccination against, 834 
pneumococcus, 715 
vaccine treatment, 915 
precipitin reaction with mouse peritoneal 
exudate in, 332 
with sputum in, 333 
Staphylococcus aureus, 919 
streptococcus, etiology, 918 
urine in, bacterial precipitinogens in, 315 
Dochez and Avery’s precipitin reaction 
with, 332 
vaccination, 802 
contraindications, 804 
dosage and administration, 804 
experimental, 803 
indications, 805 
kinds of vaccine, 803 
reactions, 804 
results, 804 
Pneumoniae, diplococcus, 908 
Pneumotoxin of Weiss and Kolmer, 716 
Poison ivy, 103 
hypersensitiveness to, 666 
Poisoning, acute, blood transfusion in, 1089 
forage, 100 
ivy, desensitization in, 735 
meat, 100 
ptomain, 100 
commonest sources of, 116 
venom, 225 
Poisons, bacterial, hemolysis by, 385 
humoral, relation of, to allergy, 629 
Poisons, mushroom, 104 
vegetable, hemolysis by, 384 
Poliomyelitis, acute anterior, complement 
fixation in, 556 
immunity in, 1020 
infection in, 1020 
serum treatment, 1023, 1024 
treatment, 1020 
vaccination, 820, 1023 
Pollantin, 663 
Dunbar’s, in treatment of hay-fever, 732 
Pollen allergens, preparation of, 729 
allergy, 662. See also Hay-fever. 
asthma due to, 652 
skin tests for, 680 
treatment of, 725. See also Hay-fever. 
anaphylactogens, 608 
antitoxin, production of, 225 
asthma, treatment, 734 
blenorrhea, treatment of, 1015 
causing hay-fever, 664 
collecting, Wodehouse method, 729 
hypersensitiveness, 639 
preseasonal desensitization to, Kolmer’s 
method, 730 
value, 727 
Walker’s method, 730 
Polyceptor, 339 
Polyvalent vs. monovalent antitoxic sera, 208 
Porges-Meier reaction in syphilis, 522 
Positive chemotaxis, 127 
Postoperative hemorrhage, blood transfusion 
in, 1079 
parotitis, 875 
Posttransfusion treatment in blood trans- 
fusion, 1097 
Powder form, preservation of serum in, 56 
Precipitate, 305 
complement fixation by, 315 
origin of, 311 
Precipitation in syphilis, mechanism of, and 
relation to complement fixation, 533 
mechanism of, 310 
of exotoxins, 94 
origin of precipitate, 311 
reactions, colloidal, in syphilis, 520 
in syphilis, 517 
advantages of, 535 
classification, 517 
complement fixation and, comparison 
of, 536 
disadvantages of, 536 
practical value of, 537 
sensitiveness, 536 
specificity, 537 
and practical value of, 535 
relation of lipoids to, 584 
Precipitin, 140, 304, 305 
bacterial, experimental work, 1135 
for identification and differentiation of 
bacteria, 317 
in anthrax, 316 
in bubonic plague, 316 
in echinococcus, 316 
in glanders, 316 
in gonorrhea, 316 
in syphilis, 316 
in tuberculosis, 316 1192 
INDEX 
Precipitin, definition, 304 
experimental production, 1110 
work, 1133 
for examination and identification of bones, 
320 
of eggs, 320 
formation of, 141 
group, 308 
historic, 304 
immune, 306 
in examination and identification of bloods, 
blood-stains and sera, 317 
of meats, 319 
of milks, 319 
of seminal stains, 318 
in immunity, r61e of, 315 
kinds, 305 
lipoid, 309 
nature and properties, 306 
nomenclature, 305 
normal, 306 
origin of, 308 
production of, 308 
reactions in syphilis, specific, 517 
relation to albuminolysins, 315 
to allergin, 626 
to complement-fixing antibody, 427 
to sensitizin, 626 
serum, titration experimental work, 1133, 
1134 
specificity, 311 
experimental work, 1134 
tests, bacterial, 331 
for blood-stains, apparatus, 325 
capillary tube method, 328 
collection and storage of sera, 324 
microscopic and chemical tests pre- 
ceding, 321 
opalescent antisera in, 328 
preliminary titrations, 325 
preparation of extract of stain, 322 
of immune serum, 323 
sources of error in, 328 
technic, 327 
for detection of meat adulteration, 329 
preparation of immune serum, 330 
of meat extract, 329 
technic, 330 
for differentiation of human and animal 
blood, 321 
for identification and differentiation of 
bacteria, 317 
in identification of bone, 333 
in identification of milk, 334 
of semen, 334 
in other bacterial infections, 333 
in syphilis, 317 
practical applications, 315 
quantitative, 332 
technic, 321 
uses of, 333 
with bacterial precipitins for diagnosis of 
bacterial infections, 316 
precipitinogens in inflammatory exu- 
dates, 316 
in urine in pneumonia, 315 
with mouse peritoneal exudate in pneu- 
monia, Blake’s, 332 
Precipitin tests with spinal fluid in meningitis, 
333 
with sputum in pneumonia, Oliver’s, 333 
with urine in pneumonia, Dochez and 
Avery’s, 332 
yeast, 317 
Precipitinogen, 305 
bacterial, in inflammatory exudates, 316 
experimental work, 1135 
in urine in pneumonia, 315 
production of, 331 
chemistry of, 308 
relation of, to anaphylactogens, 615 
Precipitoid, 305, 307 
Predisposition, 73. See also Susceptibility. 
Pregnancy, anaphylactic reaction in, 719 
bearing of, upon immunity to tumors, 1027 
blood transfusion in, 1089 
contraindication to typhoid vaccination, 
796 
late, Wassermann reaction in, 504 
pernicious vomiting of, blood transfusion 
in, 1089 
protective ferments in, 231 
Bronfenbrenner’s theory, 234 
pulmonary tuberculosis in, tuberculin treat- 
ment, 1069 
ruptured tubal, hemorrhage due to, trans- 
fusion for, 1080 
value of Abderhalden’s ferment tests in, 244 
Prenatal infection, 65 
syphilis, Wassermann reaction in, 511 
Preparator, 338 
Preservation of erythrocytes, 13 
Primary foci of infection, 79 
etiology, 79 
syphilis, Wassermann reaction in, 508 
Pro-agglutination, 281 
experimental work, 1131 
Pro-agglutinoids, 257 
Prodromal symptoms of infection, 90 
Profeta’s law, 824 
Wassermann reaction and, 512 
Prophylactic immunization. See Vaccina- 
tion. 
serum, 860 
surgical treatment of wounds, 870 
treatment of allergy, 721 
of serum allergy, 722 
Prophylaxis and treatment of disease, sera in, 
835 
medicinal, of anaphylactic serum shock, 724 
of disease, active immunization in, 757 
serum, 860. See also Serum prophylaxis. 
Prostatitis, antigonococcus serum in, 978,979 
intravenous vaccine in, 981 
milk injections in, 983 
non-specific agents in, 981 
Protease, serum, 229 
Protective activity of sera, 404 
ferments in pregnancy and disease, 231 
Bronfenbrenner’s theory, 234 
substance, 342 
Protein, anaphylactogenic activity of, for 
lower animals, 614 
anaphylactogens, chemistry of, 611 
bacterial, 85, 110 
action of, 111 INDEX 
Protein, bacterial, digested, toxicity of, 114 
experimental work, 1117 
Friedberger’s theory, 113 
. nature of, 110 
Vaughan’s theory of, 112 
complement fixation for identification, 
Kolmer’s new method, 565 
in differentiation, 538, 560 
experimental work, 1150 
derivatives, anaphylactogenic activity of, 
613 
differentiation of, by anaphylaxis reaction, 
642 
fever, 91, 114 
non-specific, in mumps, 1037 
in orchitis, 1037 
pneumococcus, 715 
racemized, anaphylactogenic activity of, 
613 
sensitinogens, 607 
shock therapy, 649 
treatment, non-specific, of epilepsy, 1036 
principles of, 879 
Proteolysins, relation of, to allergin, 627 
Proteotropic complement fixation, 438 
reactions, 438 
Protozoa, active sensitization by, 643 
allergy to, skin test for, 684 
toxins of, 104 
Protozoan anaphylactogens, 609 
Prurigo, treatment of, 996, 997 
Pruritus ani, vaccine treatment, 997 
of scrotum, vaccine treatment, 997 
treatment of, 996, 997 
vulvae, vaccine treatment, 997 
Pseudo-anaphylaxis, 606 
Pseudopodia, 123 
Psoriasis, autoserum treatment, 995 
Danysz’s enterovaccine in, 995 
horse-serum in, 995 
milk injections in, 995 
treatment of, 994 
turpentine injections in, 995 
vaccine treatment, 994 
Psychoreaction of Much and Holzmann in 
Venom hemolysis, 418 
P. T. O. tuberculin, preparation of, 1054 
Ptomains, 115 
and toxic exudates, 85, 115 
bacterial, Breiger’s method of isolation, 116 
experimental work, 1117, 1118 
poisoning, 100 
commonest sources of, 116 
Puerperal sepsis, streptococcus infection in, 
vaccination against, 819 
vaccine treatment, 892 
Pulmonary tuberculosis, ambulant cases, 
tuberculin treatment, 1069 
during pregnancy, tuberculin treatment, 
1069 
hemorrhage due to, transfusion for, 1081 
in children, tuberculin treatment, 1069 
tuberculin reaction in diagnosis of, 691 
treatment, 1069 
Pump, suction, 12 
Puncture, lumbar. See Lumbar puncture. 
lung, 189 
spinal. See Lumbar puncture. 
Purpura, chronic, blood transfusion in, 1084 
hsemorrhagica, blood transfusion in, 1084 
serum treatment, 1019 
Pustule, malignant, serum treatment, 969 
treatment of, 968 
Pyelitis in children, milk injections in, 1012 
terpichin in, 1012 
treatment, 1012 
turpentine injections in, 1012 
vaccine treatment, 1012 
Pyelonephritis, treatment of, 1012 
vaccine treatment, 1012 
Pyemia, 67 
Pyocyanase, 227 
Pyocyaneus bacillus, 102 
Pyodermia, vaccine treatment, 993 
Pyogenic dermatoses, milk injections in, 993 
turpentine injections in, 939 
vaccine treatment, 993 
Pyorrhea alveolaris, vaccine treatment, 1008 
Quincke’s disease, 654 
Rabbit, anaphylaxis in, 602 
intraperitoneal injection of, 52 
inoculation, method of production of im- 
mune agglutinins in, 272 
intravenous inoculation, 44 
method of production of immune ag- 
glutinins in, 271 
obtaining large amounts of blood from, 33 
by carotid artery, 34, 35 
Nuttall’s method, 33 
small amounts of blood from, 932 
active sensitization of, 617 
Rabbit-serum, normal, non-specific comple- 
ment fixation by, 438 
Rabies, administration of vaccine, 789 
canine, vaccination against, 834 
contraindications to vaccination, 788 
diagnosis of, 785 
duration of immunity after vaccination, 791 
etiology of, 782 
immunity in, 783 
immunization of dogs for, 791 
incubation period of, 784 
management of, 785 
mixed serum and vaccine in, 788 
Negri bodies, 783 
diagnostic value of, 786 
Pasteur treatment in, 787 
principle of, 787 
reactions to injections, 790 
results of vaccination, 791 
skepticism regarding, 783 
susceptible animals, 783 
symptoms of, in dog and cat, 785 
vaccination against, 782 
discovery of, 784 
vaccine, administration of, 789 
Harris method of preparing, 789 
Pasteur’s method of preparing, 788 
Terrell’s method of preparing, 789 
virus fixe of, 787, 788 
Race, bearing of, upon immunity to tumors,. 
1027 
1193 1194 
INDEX 
Racemized proteins, anaphylactogenic activ- 
ity of, 613 
Races, resistant, 145 
Rachicentesis, 25. See also Lumbar puncture. 
Racial immunity, 164 
predisposition to infection, 73 
Rashes, multiform, in serum disease, 660 
scarlatiniform, in serum disease, 660 
urticarial, in serum disease, 659 
Rats, active sensitization of, 617 
and mice, anaphylaxis in, 604 
intravenous injection of, 47 
obtaining large amounts of blood from, 38 
small amounts of blood from, 38 
Reaction, Abderhalden’s ferment, 237. See 
also Abderhalden’s ferment test. 
abortion, 711 
accelerated, in infection, 89 
after administration of sera, 839 
agglutination. See also Agglutination, bac- 
terial, test. 
allergic, production of, 621 
skin, 671. See also Anaphylactic reac- 
tion. 
anaphylactic, 671. See also Anaphylactic 
reaction. 
anaphylactoid, 606 
antigen-antibody, 151 
antitrypsin, 236 
Fuld and Goss’s, 236 
bacteriolytic, 365. See also Bacteriolytic 
tests. 
blood, forensic, experimental work, 1134 
body, 599,626 
Bruck’s nitric acid, in syphilis, 519 
colloidal gold, Lange’s, 585 
experimental work, 1154 
precipitation, in syphilis, 520 
complement fixation, 420. See also 
Bordet-Gengou phenomenon of comple- 
ment fixation. 
conglutination with bacteria, 285 
cutaneous, allergic, technic of, 674 
versus intracutaneous for allergy, 673 
Dreyer standardized agglutination, 282 
modified, 284 
epiphanin, 588 
focal, 641 
following injection of non-specific agents, 
884 
for antitoxin immunity to diphtheria, 740 
for isohemagglutinins, 298 
for isohemolysins, 298 
forensic blood, 321 
gonococcus, technic of, experimental work, 
1150 
Gruber-Widal, in typhoid fever, experi- 
mental work, 1128, 1129 
Herman-Perutz, in syphilis, 522 
Hirschfeld and Klinger’s coaguloreaction 
in syphilis, 520 
immediate, in infection, 89 
immunity, 90 
intracutaneous urine, of Wildbolz, 695 
Kahn, in syphilis, 531 
Klausner’s water, in syphilis, 520 
Lange’s colloidal gold, 585 
luetin, 705. See also Luetin reaction. 
Reaction, macroscopic, for agglutinins and 
hemolysins, 298 
mallein, 710. See also Mallein reaction. 
McDonagh gel, in syphilis, 519 
Meinicke’s, in syphilis, 522 
microscopic, for hemagglutinins, 300 
miostagmin, 590 
technic of, experimental work, 1154 
Noguchi’s butyric acid, in syphilis, 518 
non-specific, relation of, to therapeutic 
activity, 886 
of immunity, colloidal chemistry and, 
analogy between, 579 
in smallpox vaccination, 779 
in vaccination, 777 
Pfeiffer’s bacteriolytic, experimental work, 
1151 
Porges-Meier, in syphilis, 522 
precipitin, 321. See also Precipitin tests. 
qualitative complement fixation, 488 
quantitative complement fixation, 486 
Sachs-Georgi, in syphilis, 529 
saturation, of Castellani, 289 
Schick, experimental work, 1120 
for immunity to diphtheria, 739. See 
also Schick test. 
Sigma, of Dreyer and Ward in syphilis, 531 
skin allergic, mechanism of, 639 
strep totrichin, 699 
systemic, to infection, 91 
Teichmann’s hemin crystal, for blood- 
stains, 321 
tonicity, for measuring resistance of red 
blood-corpuscles to hypotonic saline 
solutions, 411 
torpid early, in infection, 89 
toxin-antitoxin, nature of, 208 
experimental work, 1127 
tuberculin, 685. See also Tuberculin re- 
action. 
typhoidin, of Gay and Force, 712 
Vemes, in syphilis, 523 
Wassermann. See Wassermann reaction. 
Weil-Felix, in typhoid fever, 268 
Widal, in typhoid fever, 273 
Reactivity of skin in allergic skin reactions, 
671 
Reagin, syphilis, 429 
Receptoric atrophy, 57 
in active immunization, 168 
Receptors, 136 
of first order, 137, 205 
of third order, 336, 337 
sessile, 625, 645 
three orders, 137 
Reduction factor, 282, 283 
Relapses and complications in typhoid fever, 
effect of vaccine therapy upon, 903 
Relapsing fever, Wassermann reaction in, 504 
Resistance, 119. See also Immunity. 
Resistant races, 145 
Respiratory organs as portal of entrance for 
bacteria, 63 
diseases of, treatment, 1002 
Retention theory of immunity, 122 
Retinal hemorrhage, blood injections in, 1018 
treatment of, 1018 
Retinitis, albuminuric, milk injections in, 1018 INDEX 
1195 
Revaccination, smallpox, 779 
Reversible colloid, 576 
Revised theory and role of phagocytosis in 
immunity, 134 
Rheumatic arthritis, chronic, vaccine treat- 
ment, 987 
fever, acute, vaccine treatment, 987 
Rhinitis, acute, vaccine treatment, 1002 
allergic, 653 
chronic, vaccine treatment, 1003 
perennial vasomotor, 653 
skin tests for bacteria, fungi, and protozoa 
in, 684 
vasomotor, skin tests as guide to treatment 
by desensitization, 733 
treatment of, 733 
Rhus toxicodendron, 103 
venenata, 103 
Richet’s theory of allergy, 622 
Ricin, 103 
Ricinus communis, 103 
Rickettsia prowazeki, 801 
Rigg’s disease, vaccine treatment, 1008 
Rinderpest, serum prophylaxis of, 878 
serum treatment, 1040 
Ringworm, anaphylactic reaction in, 718 
treatment of, 995 
vaccine treatment, 995 
Robin, 103 
Robinia pseudo-acacia, 103 
Romer and Sames’ method of determining 
small amounts of diphtheria antitoxin, 217 
Rose cold, etiology of, 726 
Rosenow’s method of securing endotoxin from 
pneumococci, 106 
Roth’s method of intravenous inoculation of 
guinea-pig, 46 
Rubber teats for capillary pipets, 3 
Rupture of spleen and liver, hemorrhage due 
to, transfusion for, 1080 
Ruptured tubal pregnancy, hemorrhage due 
to, transfusion for, 1080 
Russel’s theory of tumor immunity, 1030 
SACHS-Friedberger phenomenon, 399 
Sachs-Georgi reaction in syphilis, 529 
Saline solution in hemorrhage, 1079 
Saliva, antiseptic and germicidal properties 
of, 65 
Salpingitis, antigonococcus serum in, 979 
non-specific agents in, 984 
vaccine treatment, 984 
Salt solution, 9 
hypertonic, hemolysis by, 381, 382 
hypotonic, hemolysis by, 381, 382 
tonicity test for measuring resistance 
of red blood-corpuscles to, 411 
variation in resistance of red blood- 
corpuscles to, 382 
preparation of, 9 
resistance of erythrocytes to, experi- 
mental work, 1136 
Salts, effect of, upon complement, 343 
hemolysis by, 383 
Sames and Romer’s method of determining 
small amounts of diphtheria antitoxin, 217 
Saponin hemolysis, experimental work, 1136 
Saponin method for measuring resistance of 
red blood-corpuscles to, 412 
substances, 384 
Sapremia, 67 
Saprophytes, 60, 108 
Saprophytic bacteria, 67 
Sarcoma, Coley’s fluid in, 1033 
serum treatment, 1031 
vaccine treatment, 1031 
Saturation test of Castellani, 289 
Scar of smallpox vaccination, 779 
Scarlatiniform rashes in serum disease, 660 
Scarlet fever, antistreptococcus serum in 
treatment, 899 
complications of, vaccine treatment, 898 
convalescent serum and blood in treat- 
ment of, 899 
technic, 900 
value of, 901 
Weaver’s method, 900 
whole blood, method of Zingher, 
900 
immunity in, 818 
non-specific agents in treatment of, 901 
normal serum in treatment of, 901 
streptococcus infection in, vaccination 
against, 819 
treatment, 898 
vaccination, 818 
Wassermann reaction in, 504 
Schick test, 739 
control fluid for, 741 
dose for, 740, 742 
experimental work, 1120 
for immunity to diphtheria, 739 
for natural diphtheria antitoxin, 208 
in relation to passive immunization 
against diphtheria, 861 
influence of age on, 746 
of disease upon, 745 
Kellogg’s method for determining pres- 
ence of natural antitoxin in human 
serum as substitute for, 217 
method of conducting, 742 
of dilution of toxin, 741 
necessity of a control in, 744 
negative reaction, significance of, 742 
positive reactions, percentage of, influ- 
ence of age, race, and inheritance 
on, 744 
significance of, 743 
pseudopositive reaction, significance of, 
743 
pseudoreaction in relation to toxin- 
antitoxin immunization, 744 
relation of, to diphtheria carriers and to 
diagnosis of diphtheria, 745 
toxin for, 740 
value of, 745 
Schultz-Dale in vitro anaphylaxis reaction, 
642 
Schiitz, streptococcus equi of, 834 
Sciatica, vaccine treatment, 1038 
Scleritis, milk injections in, 1018 
Scours, 153 
white, serum treatment, 1040 
Scratch method of vaccination, 775 
test for allergic sensitization, 673 1196 
INDEX 
Scrofuloderma, tuberculin treatment, 1073 
Scrotum, pruritus 6f, vaccine treatment, 997 
Secondary foci of infection, 80 
infection, 77 
syphilis, Wassermann reaction in, 509 
Secretions, animal, hemolysis by, 386 
Selective action of exotoxins, 96 
Semen, identification of, precipitin test for, 
334 
Seminal stains, precipitins in examination and 
identification of, 318 
Sensibiligen, 599 
Sensibilisin, 599, 626 
Sensibilisinogen, 599 
Sensitinogens, 599, 606 
kinds of, 607 
non-protein, 615 
protein, 607 
Sensitization, 599 
active, 616 
amounts of protein sensitinogens, 617 
by bacteria and protozoa, 643 
by toxins, 645 
duration of, 618 
in infectious diseases, 644 
incubation period of, 618 
of dogs, 618 
of guinea-pigs, 617 
of rabbits, cats, and rats, 617 
of skin and mucous membranes, 638 
route of administration, 616 
by horse-serum, prevention of, 722 
in food allergy, 668 
inherited tendency to acquire, 650 
passive, 618 
heterologous, 619 
historic, 618 
homologous, 619 
maternal transmission, 619 
mechanism of, 620 
of skin and mucous membranes, 637 
with human sera, 619 
prevention of, 620 
Sensitized vaccines, 753 
bacterial, preparation of, 199 
subcutaneously, in treatment of typhoid 
fever, 902 
Sensitizers, 336. See also Amboceptors. 
Sensitizin. See Allergin. 
Sensitizing dose, 599 
injection, 599 
substance, 133, 135 
Sepsis, puerperal, streptococcus infection in, 
vaccination against, 819 
vaccine treatment, 892 
Septic wounds and sinuses, treatment, 890 
vaccine treatment of, 890 
Septicemia, 67 
gonococcus, antigonococcus serum in, 978 
hemorrhagic, serum prophylaxis, 1040 
serum treatment, 1040 
vaccination against, 834 
streptococcus, blood transfusion in treat- 
ment of, 896 
vaccine treatment of, 892 
Sero-anaphylactic reactions, accelerated, 656 
delayed, 656 
immediate, 656 
Sero-anaphylactic reactions, types of, 656 
Serophytes, 891 
Serosaprophytes, 891 
Serous membrane, tuberculosis of, tuberculin 
reaction in diagnosis of, 692 
meningitis, tuberculous and, Noguchi’s- 
butyric acid reaction in, 518 
Serum, active, method for removal of natural 
hemolysins from, 409 
agglutinating strength of, variation in, 262 
allergy, 653 
prophylactic treatment of, 722 
cases requiring desensitization, 722 
methods for specific desensitizationr 
723 
prevention of sensitization, 722 
skin tests for, 680 
amount administered in relation to anaph- 
ylactic shock, 657 
anaphylactogens, 608 
and antigens, anticomplementary activity 
of, 433 
and blood, human convalescent, in treat- 
ment of bronchopneumonia of 
influenza, 919 
of scarlet fever, 899 
technic, 900 
value, 901 
Weaver’s method, 900 
Zingher’s method, 900 
and corpuscles, obtaining, 14 
and leukocytic ferments, immunity to, 229 
in relation to infection, 229 
antianthrax, 967 
in malignant pustule, 969 
anticholera, 973 
administration of, 974 
dosage of, 974 
value of, 974 
anticomplementary, 399 
activity of, 405 
experimental work, 1146 
anticytotoxic, 569 
antidiphtheric, production of, 210 
antidysenteric, 220, 940 
administration and uses, 940 
collecting and testing, 222 
production of, 220 
culture, 220 
immunizing animals, 221 
rapid method of Flexner and Amoss, 
221 
antigonococcus, in acute epididymitis, 
979 
gonorrhea, 979 
in arthritis, 978, 979 
in gonococcus septicemia, 978 
in orchitis, 978, 979 
in prostatitis, 978, 979 
in salpingitis, 979 
antihemolytic activity of, 404 
anti-influenza, 961 
administration of, 962 
antimeningococcus, 947 
in tuberculous meningitis, 965 
antipertussis, 975 
antipest, 970 
administration of, 971 INDEX 
1197 
Serum, antipest, dosage of, 971 
antiplague, 970 
administration of, 971 
dosage of, 971 
antipneumococcus, action of, 911 
administration of, 912 
and ethylhydrocuprein hydrochlorid in 
pneumococcus meningitis, 965 
in treatment of bronchopneumonia of 
influenza, 919 
intravenous injections, in pneumococcus 
meningitis, 964 
Kyes’, 913 
in pneumococcus meningitis, 964 
preparation of, 910 
sodium oleate, and boric acid, in pneu- 
mococcus meningitis, 963 
standardization of, 910 
antistaphylococcus, 223 
antilysin test for, 223 
preparation of, 223 
antistreptococcus, administration of, 895 
in treatment of bronchopneumonia of 
influenza, 920 
of scarlet fever, 899 
mode of action of, 894 
preparation of, 894 
standardization of, 895 
value of, 896 
antitetanic, production of, 218 
antitoxic, monovalent vs. polyvalent, 208 
antitryptic power, testing, experimental 
work, 1128 
bactericidal power, experimental work, 1152 
bacteriolytic, in treatment of disease, 380 
blood, and plasma, versus whole blood, bac- 
tericidal activity of, 361 
cadaver, testing of, in Wassermann reac- 
tion, 446 
complement in, variation in amount of, 347 
convalescent, human serum for, methods for 
preparing, 857 
diagnosis of paroxysmal hemoglobinuria, 
410 
disease, 596, 600, 653, 655. See also 
Serum sickness. 
drying, box far, 57 
for Wassermann test, 444 
collecting, 445 
hemagglutinins, 292 
kinds of, 292 
hemolysins, 387. See also Hemolysins, 
serum. 
hemolysis, 386 
in vitro, experimental work, 1137 
in vivo, experimental work, 1137 
quantitative factors in, experimental 
work, 1138 
hog cholera, production and standardiza- 
tion of, 876 
horse-, in psoriasis, 995 
normal, in epididymitis, 983 
in gonorrheal complications, 983 
human convalescent influenza, in treatment 
of influenza of child, 906 
for autoserum, methods for preparing, 
857 
in serum therapy, 838 
Serum, human, in treatment of meningitis, 
960 
method for securing large amounts, 859 
small amounts, 857 
passive sensitization with, 619 
immune, for gas gangrene, preparation of, 
872 
horse-, in serum therapy, 838 
preservation of, 53 
standardization of, complement-fixation 
test in, 560 
importance of anticomplementary action 
of, in relation to complement-fixation 
tests and reactions, 437 
in prophylaxis and treatment of disease, 835 
methods for administration, 842 
inactivated, method for removal of natural 
hemolysins from, 409 
intramuscular injection, technic, 843 
intraspinal injection, technic, 850. See also 
Intraspinal injection. 
intravenous injection, gravity method, 847 
of children, 850 
syringe method, 845 
technic, 845 
kinds employed in serum therapy, 838 
Kyes’ antipneumococcus, in pneumococ- 
cus meningitis, 964 
Maragliano’s, in tuberculosis, 1075 
Marmorek’s, in tuberculosis, 1075 
method of concentrating by isolating anti- 
toxin globulins, 214 
of determining natural hemolysins in, 410 
of removing hemolysins from, 409 
normal blood and, bactericidal activity of, 
357 
human, non-specific complement fixation 
by, 438 
in serum therapy, 838 
in treatment of scarlet fever, 901 
preservation of, 53 
obtaining of, 15 
peptidase, 229 
precipitins in examination and identifica- 
tion of, 317 
titration of, experimental work, 1133, 
1134 
preservation of, 53 
in dried paper form, 57 
in fluid form by bacteria-free filtration, 55 
by freezing, 56 
with antiseptics, 53 
in living animal, 57 
in powder form, 56 
prophylactic, 860 
prophylaxis, 835, 860 
in diseases of lower animals, 876 
of anthrax, 878 
of bacillary dysentery, 871 
of botulinus, 878 
of calf cholera, 877 
of cattle plague, 878 
of diphtheria, 860 
dosage and administration, 860 
duration of immunity, 861 
indications, 861 
results in, 861 
Schick test in relation to, 861 1198 
INDEX 
Serum prophylaxis of gas gangrene, 871, 873 
preparation of immune sera, 872 
of hog cholera, 876 
of infectious abortion, 877 
of influenza in animals, 1041 
of measles, 874 
of meningitis, 873 
of mumps, 875 
of rinderpest, 878 
of strangles, 1041 
of syphilis, 876 
of tetanus, 867 
dose, 870 
of eye, 1015 
value of, 869 
protease, 229 
protective activity of, 404 
relation to anemia, 404 
reactions after administration of, 839 
removal of hemolysins from, by absorp- 
tion, 399 
route of administration of, in relation to 
anaphylactic reactions, 656 
shock, anaphylactic, medicinal prophylaxis 
of, 724 
sickness, 596, 600, 653, 655 
a true anaphylactic phenomenon, 656 
adenitis in, 661 
after lumbar puncture, 857 
arthritis in, 661 
detection of, 662 
eruption in, 659 
fever in, 659 
frequency, 658 
from antimeningococcus serum, 957 
general symptoms, 661 
incidence of, 659 
incubation period, 659 
leukocyte changes in, 661 
multiform rashes in, 660 
myalgia in, 661 
nature of, 655 
neuritis in, 661 
prevention of, 662 
recurrent, 661 
scarlatiniform rashes in, 660 
symptoms, 658 
treatment of, 662 
urticarial rashes in, 659 
subcutaneous injection, technic, 842 
treatment, 837 
anaphylaxis in, 840 
asthma in, 841 
contraindications to, 840 
dangers of, 840 
definition, 750 
intramuscular injection, technic, 843 
intraspinal injection, technic, 850. See 
also Intraspinal injection. 
intravenous injection, gravity method, 
847 
of children, 850 
syringe method, 845 
technic, 845 
kinds of sera employed in, 838 
non-specific and specific, 838 
of abscesses, 889 
of acute anterior poliomyelitis, 1023,1024 
Serum treatment of anthrax, 967 
bacteremia, 970 
in animals, 1040 
of arthritis, 986 
of Asiatic cholera, 973 
of bacillary dysentery, 940 
results, 941 
of carbuncles, 889 
of cattle plague, 1040 
of chorea, 1036 
of diphtheria, 921. See also Diphtheria 
antitoxin. 
of diphtheritic conjunctivitis, 1015 
keratitis, 1015 
of erysipelas, 897 
of furunculosis, 889 
of gas gangrene, 939 
of gonococcus conjunctivitis, 1015 
infections, 977 
of hay-fever, 732 
of hemophilia, 1018 
of hemorrhage in disease and after opera- 
tion, 1019 
of hemorrhagic septicemia, 1040 
of hog cholera, 1039 
of hyperthyroidism, 1038 
of infectious jaundice, 1038 
of influenza, 906 
in animals, 1041 
of influenzal meningitis, 960 
of internal anthrax, 970 
of joint-ill of colts, 1040 
of leprosy, 999 
of lethargic encephalitis, 1025 
of malignant pustule, 969 
tumors, 1031 
of Malta fever, 1026 
of marasmus, 1039 
of melena neonatorum, 1018 
of meningococcus meningitis, 946 
cases of posterior basal meningitis, 
956 
with dry canal, 956 
with thick plastic exudate, 956 
chronic cases, 957 
results, 958 
serum sickness in, 957 
subacute cases, 957 
of pertussis, 975 
of plague, 970 
of pneumococcus infections of eye, 1014 
meningitis, 962 
pneumonia, 909. See also Pneumonia, 
lobar, pneumococcus. 
of pneumonia, results in, 912 
of purpura haemorrhagica, 1019 
of retinal hemorrhage, 1018 
of rinderpest, 1040 
of sleeping sickness, 1025 
of snake-bites, 1035 
of staphylococcus infections, 889 
of strangles, 1041 
of streptococcus infections, 893 
of eye, 1014 
meningitis, 965 
of syphilis, 1034 
of tuberculosis, 1075 
of tuberculous meningitis, 965 INDEX 
1199 
Serum treatment of typhoid fever, 903 
of typhus fever, 1037 
of Weil’s disease, 1038 
of white scours, 1040 
of#yellow fever, 1037 
origin, 835 
specific and non-specific, 838 
status lymphaticus in, 841 
subcutaneous injection, technic, 842 
Torrey’s antigonococcus, 978 
Yersin’s plague, 971 
Sessile receptors,'625, 645 
Sex, bearing of, upon immunity to tumors, 
1027 
Shaking, effect of, upon complement, 344 
machine, 191 
Sheep, anaphylaxis in, 604 
intravenous injection of, 49 
obtaining large amounts of blood from, 38 
small amounts of blood from, 35, 39 
Shell-fish allergen, preparation of, 683 
Shiga dysentery bacillus, 871, 940 
Shock, allergic, prevention of, 620 
anaphylactic, curative influence of, 649 
immediate and severe, symptoms of, 
658 
blood transfusion in, 1082 
colloid, 606 
serum, anaphylactic, medicinal prophylaxis 
of, 724 
treatment, protein, 649 
Sickness, serum, 596, 600, 653, 655. See also 
Serum sickness. 
sleeping, serum treatment, 1025 
Side-chain theory of immunity, 122 
Ehrlich’s, 135 
Sigma reaction of Dreyer and Ward in 
syphilis, 531 
Simon and Lamar’s modification of Wright’s 
method for opsonic index, 184 
Simon’s method of quantitative estimation of 
bacterio tropins, 186 
Sinus, superior longitudinal, obtaining blood 
from, in infants, 23 
Sinuses, accessory, of nose, diseases of, treat- 
ment, 1002 
Sinusitis, acute, vaccine treatment, 1004 
chronic, vaccine treatment, 1004 
Skin allergic reaction, mechanism of, 639 
and mucous membranes, active sensitiza- 
tion of, 638 
passive sensitization of, 637 
anthrax of, serum treatment, 969 
treatment, 968 
as portal of entrance for bacteria, 63 
bacteria on, 62, 63 
diseases, milk injections in, 997 
treatment, 991 
turpentine in, 997 
typhoid bacilli in treatment of, 997 
in relation to infection, immunity, and 
biologic therapy, 991 
to natural immunity, 167 
lesions, allergic, 653 
normal defenses of, against invasion of 
bacteria, 65 
reactions, allergic, 671. See] also Anaphy- 
lactic reaction. 
Skin reactions, allergic, as guide to treatment 
by desensitization in allergic asthma 
and vasomotor rhinitis, 733 
as indices of infection and hyper- 
susceptibility, 673 
cutaneous versus intracutaneous tests, 
673 
effects of drugs upon, 679 
for bacteria, fungi, and protozoa, 684 
for drugs, 684 
for food, 682 
for hairs, dandruffs, and feathers, 681 
for pollen and plant allergies, 680 
for serum, 680 
in focal and mixed infections, 685 
influence of age, 671 
of disease, 671 
of non-specific reactions, 672 
intracutaneous tests more likely than 
cutaneous tests to elicit general re- 
actions, 674 
mechanism of specific reactions, 672 
non-specific, 672 
reactivity of skin in, 671 
specific and non-specific reactions in, 
671 
technic of cutaneous test, 674 
of intracutaneous tests, 677 
reactivity of, in allergic reactions, 671 
tuberculosis of, tuberculin reaction in 
diagnosis of, 692 
treatment, 1073 
Sleeping sickness, serum treatment, 1025 
Smallpox, 118 
anaphylactic reaction in, 717 
complement fixation in, 556 
immunity in, 766 
vaccination, 766. See also Vaccination, 
smallpox. 
vaccinia and, relationship, 767 
Snake venoms, 105, 224 
nature of, 105 
properties of, 105 
Snake-bites, serum treatment of, 1035 
Sodium carbonate as preventive of serum 
shock, 725 
chlorid as preventive of serum shock, 725 
intravenously in bronchopneumonia of 
influenza, 920 
citrate solution, 10 
oleate, boric acid, and antipneumococcus 
serum in pneumococcus meningitis, 963 
Sohxlet extraction apparatus with vacuum, 
524 
Sols, 576 
Soluble toxins, 92 
general properties of, 93 
Solutions, 9 
colloidal, 576 
salt, 9 
sodium citrate, 10 
Southard and Gay’s theory of allergy, 623 
Species susceptibility, 73 
virulence, 77 
Specific antigens, 146 
Spengler’s perlsucht tuberculin, preparation 
of, 1054 
Spermatoprecipitins, 318 1200 
Spermatotoxin, 569 
experimental work, 1153 
Spinal fluid, precipitin test with, in menin- 
gitis, 333 
puncture, 25. See also Lumbar puncture. 
Spirochaeta icterohaemorrhagicae, 828 
pallida, 822 
Spirochetic complement-fixing antibody, na- 
ture and specificity, 429 
Spleen, rupture of, hemorrhage due to, trans- 
fusion for, 1080 
Splenectomy in pernicious anemia, 1087 
Spontaneous phagocytosis, 171, 185 
Sporadic disease, definition, 59 
Sporotrichosis, agglutination reaction in, 270 
anaphylactic reaction in, 717 
complement fixation in, 559 
Sprue, vaccine treatment, 1010 
Sputum, collection of, 189 
precipitin reaction with, in pneumonia, 
Oliver’s, 333 
Staining, acid-fast bacilli, 183 
bacteria, 183 
Stains, blood-, blood grouping for identifica- 
tion of, 298 
precipitins in examination and identifica- 
tion of, 317 
seminal, precipitins in examination and 
identification of, 318 
Stalagmometer, 591 
Standard hemorrhage, 1080 
Standardization of diphtheria antitoxin, 99 
experimental work, 1125 
toxin, 98 
of platinum loops, 195 
of tetanus antitoxin, experimental work, 
1126 
Staphylococcus albus, 888 
aureus, 888 
pneumonia, 919 
citreus, 888 
infections, immunity in, 888 
non-speciflc treatment of, 890 
serum treatment, 889 
vaccine treatment of, 889 
pyogenes albus, 101 
aureus, 101 
toxins, 101 
vaccine, preparation of, experimental work, 
1125 
Staphylolysin, 101 
experimental work, 1115 
Starkenstein’s theory of increased capillary 
permeability, 887 
Status lymphaticus in serum treatment, 841 
Stegomyia fasciata, 822 
Sterilization of bacterial vaccines, 196 
testing, 196 
of glassware, 7 
of pipets, 6 
of syringes, 9 
Stern and Korte’s method of measuring bac- 
tericidal power of blood, 
370 
experimental work, 1152 
Stevens and Force’s typhoidin, 713 
Stichreaktion, 688, 689 
Stimulins, 134, 171 
INDEX 
Stings of insects, allergy to, 666 
Stock versus autogenous bacterial vaccines, 
755 
Strangles, serum prophylaxis, 1041 
treatment, 1041 
vaccination, 834 
Streptococcus bronchopneumonia, treatment 
of, 917 
endocarditis, acute, vaccine treatment, 892 
equi of Schiitz, 834 
hemolysans, 892 
hemolyticus, 892 
infections and septicemia, treatment of, 
with special reference to cellulitis, 
puerperal sepsis, and endocarditis, 
891 
immunity in, 892 
in scarlet fever and puerperal sepsis, vac- 
cination against, 819 
leukocytic extracts in treatment of, 897 
non-specific agents in treatment of, 897 
of eye, serum treatment, 1014 
peptone in treatment of, 897 
serum treatment, 893 
treatment, 891 
vaccine treatment, 892 
kinds of, in relation to biologic therapy, 891 
meningitis, serum treatment, 965 
non-hemolyticus, 892 
pneumonia, etiology, 918 
pyogenes, 888 
septicemia, blood transfusion in treatment 
of, 896 
toxins, 101 
vaccine, 819 
dose, 891 
viridans, 892 
Streptolysin, 102 
experimental work, 1115 
Strep to toxin, experimental work, 1115 
Streptotrichin reaction, 699 
Strong’s cholera vaccine, 808 
Strophulus treatment of, 996, 997 
St. Vitus’ dance, autosera treatment, 1036 
Styes, vaccine treatment, 1015 
Subcutaneous injection of animals, 43 
with fluid inoculum, 43 
with solid inoculum, 44 
of serum, technic, 842 
mallein reaction, 710 
method of vaccination, Goodal’s, 776 
tuberculin test, 689, 793 
in animals, 700 
Subdural injection of serum, 850. See also 
Intraspinal injection. 
Subinfection, 64 
definition, 59 
Substance sensibilisatrice, 133, 135, 172, 338, 
358, 387 
saponin, 384 
Suction pump, 12 
Suctorial insects, transmission of bacteria by, 
62 
Sumac, poison of, 103 
Superior longitudinal sinus, obtaining blood 
from, in infants, 23 
Surface discharges, normal defenses of, 
against invasion of bacteria, 65 INDEX 
1201 
Surgical operations, blood transfusion pre- 
liminary to, 1082 
treatment of wounds, prophylactic, 870 
Susceptibility, acquired, 73, 74 
familial, 73 
general, in relation to infection, 73 
individual, 73 
inherited, 73. 
of bacteria to opsonification, 174 
racial, 73 
species, 73 
tissue, avenue of infection and, 70 
Suspensions, 576 
colloidal, 576 
Sycosis vulgaris, vaccine treatment, 993 
Sydenham’s chorea, autosera treatment, 1036 
Symptomatic anthrax, vaccination, 831 
Synocytotoxin, 571 
Syphilimetry, 524 
Syphilis after smallpox vaccination, 781 
agglutination reaction in, 270 
bacterial precipitins in, 316 
Brack’s nitric acid reaction in, 519 
cerebral, Wassermann reaction in, 510 
coagulating reactions in, 517 
colloidal flocculation reactions in, 517 
complement fixation in, 442. See also 
Wassermann reaction in syphilis. 
chemical coagulating reactions in, 517 
colloidal precipitation reactions in, 520 
congenital, Wassermann reaction in, 511 
Craig’s modification of Wassermann reac- 
tion, 500 
flocculation reactions in, experimental 
work, 1136 
formol reactions in, 520 
Hecht-Weinberg-Gradwohl modification of 
Wassermann reaction, 502 
Herman-Perutz reaction in, 522 
Hirschfeld and Klinger’s coaguloreaction 
in, 520 
horse, complement fixation in, 548 
immunity in, 822 
Kahn reaction in, 531 
Klausner’s water reaction in, 520 
latent, Wassermann reaction in, 510 
luetin reaction for, 705. See also Luetin 
reaction. 
McDonagh gel reaction in, 519 
Meinicke’s reactions in, 522 
Noguchi’s butyric acid reaction in, 518 
modification of Wassermann reaction, 
496. See also Noguchi’s modification. 
non-specific treatment, 1034 
Porges-Meier reaction in, 522 
precipitation in, mechanism of, and rela- 
tion to complement fixation, 533 
reaction in, 517 
advantages of, 535 
classification, 517 
complement fixation and, comparison 
of, 536 
disadvantages of, 536 
practical value of, 537 
sensitiveness of, 536 
specificity and practical value of, 535, 
537 
prenatal, Wassermann reaction in, 511 
Syphilis, qualitative complement-fixation test 
in, 488 
quantitative complement-fixation test in, 
486 
reagin, 429 
Sachs-Georgi reaction in, 529 
serum prophylaxis of, 876 
treatment, 1034 
Sigma reaction of Dreyer and Ward in, 531 
specific precipitin reactions in, 517 
treatment of, 1034 
vaccine treatment, 822, 828, 1034 
value of Abderhalden’s ferment tests in, 245 
venom hemolysis in, 415 
experimental work, 1151 
practical value, 417 
technic, 416 
Vernes’ reactions in, 523 
direct method, 523 
technic, 525 
indirect method, 523 
technic, 525 
Syphilitic livers, alcoholic extracts of, for 
Wassermann reaction, 455 
aqueous extracts of, for Wassermann re- 
action, 455 
serum, antigenic unit of, 454 
Syringe, 8 
Hitchens’, 216 
method for intravenous injection of serum, 
845, 855 
sterilization of, 9 
Systemic reaction to infection, 91 
TAB vaccine, 795 
Tabes dorsalis, treatment of, 1035 
Wassermann reaction in, 510 
Teats, rubber, for capillary pipets, 3 
Technic, general, 1 
Teeth, apices of, as foci of infection, 79 
Teichmann hemin crystal test for blood- 
stains, 321 
Temperature changes, hemolysis by, 381 
influence of, on antibody formation, 157 
on complement-fixation reactions, experi- 
mental work, 1148 
on hemagglutination, 296 
on natural immunity, experimental work, 
1119 
Terni and Bandi’s plague vaccine, 806 
method of preparing, 971 
Terpichin in pyelitis in children, 1012 
Terrell’s method of preparing rabies vaccine, 
789 
Tertiary syphilis, Wassermann reaction in, 510 
Test. See Reaction. 
Test-tubes, 7 
conversion of, into ampules for holding 
vaccines, 7 
sterilization of, 8 
Tetanolysin, 96, 99, 1112 
Tetanospasmin, 96, 99, 1112 
Tetanus, acute, incubation period in, 935 
after smallpox vaccination, 780 
antitoxin, 933 
action of, 935 
experimental work, 1113 1202 
Tetanus antitoxin, administration of, 937 
for prophylaxis, 836 
indications for, 936 
intramuscular injection, 936 
intraneural injection, 936 
intravenous injection, 936 
methods of administering, 935 
of titrating, 220 
preparation and standardization of, 933 
production of, collecting serum, 219 
immunizing animals, 219 
results of treatment, 938 
standardization, 219 
experimental work, 1126_ 
subcutaneous injection, 935 
subdural injection, 936 
value of, in treatment of tetanus, 935 
bacillus, 99, 102, 868 
chronic, incubation period in, 935 
dangerous wounds from standpoint of, 868 
. general treatment, 936 
immunity in, 867 
incubation period in, 934 
indications for treatment, 936 
mortality in, 868 
non-specific treatment of, 939 
of eye, serum prophylaxis, 1015 
serum prophylaxis of, 867 
dose, 870 
value of, 869 
treatment, 933. See also Tetanus anti- 
toxin. 
surgical treatment, 937 
toxin, 99 
action of, 933 
experimental work, 1112, 1113 
production of, 218 
special properties of, 99 
treatment of, 933. See also Tetanus anti- 
toxin. 
value of antitoxin in treatment of, 935 
prophylaxis, 869 
Tetravaccine of typhoid, paratyphoid, and 
cholera bacilli, 795 
Theories of immunity, 118 
Theory, cellular stimulation, of -Weichardt 
and Dollken, 885 
of cellular resistance, 123 
of immunity, cellular, 122 
chemo taxis, 123 
exhaustion, 122 
humoral, 135 
obsolete, 122 
retention, 122 
side-chain, 122 
of phagocytosis, and cellular resistance and 
immunity chemotaxis, 123 
historic, 123 
Metchnikoff’s, 122 
original, 124 
omnicellular plasma-activation, 885 
Weigert’s overproduction, 137 
Therapeutic activity, relation of non-specific 
reaction to, 886 
purposes, production of antitoxins for, 210 
Thermocautery in malignant pustule, 968 
Thermolabile antilysins, influence of heat 
upon, 435 
INDEX 
Thermolabile hemolysins, 403 
Thermolability of complement, experimental 
work, 1143 
Thermoprecipitinogens, 309, 320 
experimental work, 1135 
methods of Ascoli, Krumwiede, and Noble, 
331 
Thermostabile antilysins, influence of heat 
upon, 435 
hemolysins, 403 
Thermostability of hemolysins, experimental 
work, 1139 
Thyrotoxins, 571 
Tissue affinity, selective, 71 
concerned in production of antibodies, 148 
extracts, hemolysis by, 384 
susceptibility, avenue of infection and, 70 
Tonicity test for measuring resistance of red 
blood-corpuscles to hypotonic saline solu- 
tions, 411 
Tonsillar diphtheria, dosage of antitoxin in, 
928 
Tonsillitis, acute, vaccine treatment, 1004 
Tonsils as foci of infection, 79 
as source of entry of bacteria, 66 
Topfer and Jaffe’s method of measuring bac- 
tericidal power of blood, 372. 
Torpid early reaction in infection, 89 
Torrey’s antigonococcus serum, 978 
Toxalbumins, 104 
vegetable, 384 
Toxemia, 67 
bacterial, 85 
intestinal, vaccine treatment, 1010 
Toxic exudates and ptomains, 85, 115 
or exciting injection, 599 
Toxicity and virulence of diphtheria bacilli 
method of testing, 97 
of bacteria, 68 
of digested bacterial protein, 114 
Toxic-producing bacilli, types of, 208 
Toxin, 92 . ' 
active sensitization by, 645 
anaphylactogens, 609 
and enzymes difference between, 95 
and ferments, similarity between, 95 
animal, experimental work, 1112 
bacterial, 92 
experimental work, 1112 
nomenclature, 92 
other, 102 
botulism, 100 
experimental work, 1114 
special properties of, 100 
diphtheria. See Diphtheria toxin. 
dysentery, 100 
special properties of, 100 
exogenous, 92 
extracellular, 92 
bacterial, 92 
definition, 92 
for Schick test, 740 
haptophon group, 139 
hypersensitiveness, 645 
intracellular, 92 
of higher animals, 104 
plants, 103 
relation to infection and immunity, 103 INDEX 
1203 
Toxin of microfungi, 103 
relation to infection and immunity, 103 
of protozoa, 104 
plant, experimental work, 1112 
principal, special properties of, 97 
producers, 84 
soluble, 92 
genera] properties of, 93 
staphylococcus, 101 
streptococcus, 101 
tetanus, 99, 218. See also Tetanus toxin. 
toxophore group, 139 
true, 92, 93 
Toxin-antitoxin for diphtheria vaccination, 
preparation of, 863 
immunization, 862. See also Diphtheria, 
toxin-antitoxin vaccination. 
reaction, nature of, 208 
experimental work, 1127 
vaccination against diphtheria, von Behring 
method, 862. See also Diptheria, toxin- 
antitoxin vaccination. 
Toxinemia, 84 
and exotoxins, 84 
Toxogen, 599, 626 
Toxogenin, 622 
Toxoid, 94, 139 
diphtheria, 99 
experimental work, 1112 
Toxon, 96 
Toxophore group of toxin, 139 
Trachoma, milk injections in, 1016 
Transfusion, blood, 1077. See also Blood 
transfusion. 
Transmission of antibodies and immunity, 152 
Transudates in serum therapy, 838 
Trauma as predisposing to infection, 75 
Treponema pallidum, 71, 706 
aqueous extracts of, for Wassermann re- 
action, 459 
Trichophytuq 718 
Trichophytosis, treatment of, 995 
vaccine treatment, 995 
Triketohydrindenhydrate, 239 
Triple vaccine of typhoid and paratyphoid 
bacilli, 795 
T. R. tuberculin, preparation of, 1053 
True parasites, 109 
toxins, 92, 93 
Trypanosoma equiperdum, 549 
Tryptic ferment, 230 
Tubal pregnancy, ruptured', hemorrhage due 
to, transfusion for, 1080 
Tuberculin, 1051 
administration of, 1062 
intrabronchially, 1068 
intrafocal route, 1068 
intravenous, 1068 
oral route, 1068 
A. F., preparation of, 1054 
albumose-free old, preparation of, 1054 
autogenous, 1061 
bacillen emulsion, preparation of, 1053 
B. E., preparation of, 1053 
Beraneck’s, 1066 
preparation of, 1054 
B. F., preparation of, 1054 
bouillon filtrate, preparation of, 1054 
Tuberculin, choice of, 1063 
curative activity, 1057 
delirium, 649, 1062 
Dixon’s, 1054, 1066 
effects of, 1055 
first dose, 1063 
in tuberculosis of ear, 1002 
intervals between injections, 1065 
kinds of, 1051 
Koch’s old, 693 
method of preparing dilutions and doses, 
1065 
new, preparation of, 1053 
old, Koch’s, 693 
preparation of, 1052 
O. T., preparation of, 1052 
preparation of, 1052 
P. T. O., preparation of, 1054 
reaction, 685, 1062 
allergic nature, 1056 
among lower animals, 700 
Calmette’s, 689, 698 
dangers of, 692 
comparative delicacy and relation, 689 
conjunctival, in animals, 703 
of Calmette, 689, 698 
dangers of, 692 
of Wolff-Eisner, 689, 698 
dangers of, 692 
cutaneous, in animals, 705 
of Ligniere, 689, 699 
of von Pirquet, 696, 689 
dangers of, 692 
delicacy, comparative, 689 
error in, sources, 688 
false negative reactions, 688 
positive reactions, 688 
favorable and desirable, 1063 
in diagnosis of pulmonary tuberculosis, 
691 
of tuberculosis of bones, joints, and 
glands, 691 
of eye, ear, and larynx, 691 
of genito-urinary system, 691 
of serous membrane, 692 
of skin, 692 
of tuberculous meningitis, 692 
peritonitis, 692 
pleurises, 692 
value, 690 
in prognosis, value, 692 
intracutaneous, 695 
in animals, 705 
of Mantoux, 689 
of Mendel, 689 
Ligniere’s, 689, 699 
methods of conducting, 689 
Moro and Doganoff’s, 689 
Moro’s, 699 
nature of, 685 
percutaneous, of Moro, 699 
of Moro and Doganoff, 689 
relation, comparative, 689 
specificity of, 686 
subcutaneous, 689, 693 
dangers of, 692 
in animals, 700 
von Pirquet’s, 689, 696 1204 
INDEX 
Tuberculin reaction, Wolff-Eisner’s, 689, 698 
dangers of, 692 
residue, preparation of, 1053 
site of injection, 1066 
Spengler’s perlsucht, preparation of, 1054 
subcutaneous injection, 693 
subsequent doses, 1065 
time of injection, 1067 
T. R., preparation of, 1053 
treatment, antibody production from, 1058 
contraindications to, 1061 
curative activity, 1057 
duration of, 1067 
history of, 1050 
increased sensitiveness to, 1059 
indication for, 1061 
insusceptibility or tolerance to, 1058 
of ambulant cases of pulmonary tuber- 
culosis, 1069 
of lupus vulgaris, 1073 
of mixed infections, 1070 
of pulmonary tuberculosis, 1069 
during pregnancy, 1069 
in children, 1069 
of scrofuloderma, 1073 
of tuberculosis, choice of tuberculin, 1060 
of bladder, 1074 
of bones, 1073 
of ear, 1074 
of eye, 1074 
of joints, 1073 
of kidneys, 1074 
of skin, 1073 
of tuberculous adenitis, 1072 
peritonitis, 1074 
principles of, 1050 
repetition of, 1067 
results of, 1070 
Tuberculonastin, 147 
Tuberculosis, active, contraindication to ty- 
phoid and paratyphoid fever vaccina- 
tion, 796 
immunization in, contraindications, 761 
and acute infections, value of Abderhalden’s 
ferment tests in, 245 
agglutination reaction in, 269 
bacterial precipitins in, 316 
biologic treatment, 1042 
complement fixation in, 551 
experimental work, 1150 
non-specific reactions in syphilis, 554 
preparation of antigen, 553 
practical value, 554 
immunity in, 1042 
Maragliano’s, in tuberculosis, 1075 
Marfan’s law in, 1046 
Marmorek’s, in tuberculosis, 1075 
mother immunity in, 1042 
non-specific reactions in, 1059 
of bladder, tuberculin treatment, 1074 
of bones, joints, and glands, tuberculin re- 
action in diagnosis, 691 
tuberculin treatment, 1073 
of cervical lymph-glands, tuberculin treat- 
ment, 1072 
of ear, tuberculin in, 1002, 1074 
of eye, ear, and larynx, tuberculin reaction 
in diagnosis of, 691 
Tuberculosis of eye, tuberculin treatment, 
1074 
of genito-urinary system, tuberculin reac- 
tion in diagnosis of, 691 
of joints, tuberculin treatment, 1073 
of kidneys, tuberculin treatment, 1074 
of lungs, Wassermann reaction in, 504 
of serous membrane, tuberculin reaction in 
diagnosis of, 692 
of skin, tuberculin reaction in diagnosis of, 
692 
treatment, 1073 
prophylactic immunization in, 1047 
pulmonary, ambulant cases, tuberculin 
treatment, 1069 
hemorrhage due to, transfusion for, 1081 
in children, tuberculin treatment, 1069 
in pregnancy, tuberculin treatment, 1069 
tuberculin reaction in diagnosis of, 691 
treatment, 1069 
serum treatment, 1075 
tuberculin treatment, 1050. See also 
Tuberculin treatment. 
venom hemolysis in, 419 
Tuberculous adenitis, tuberculin treatment, 
10.72 
meningitis, antimeningococcus serum in, 
965 
leukocytic extracts in, 966 
serous and, Noguchi’s butyric acid reac- 
tion in, 518 
serum treatment, 965 
treatment of, 965 
tuberculin reaction in diagnosis of, 692 
peritonitis, tuberculin reaction in diagnosis 
of, 692 
treatment, 1074 
pleurisy, autoserum therapy, 1006, 1007 
tuberculin reaction in diagnosis of, 692 
Tumors, immunity in, in relation to biologic 
therapy, 1026 
relation of age to, 1027 
of pregnancy to, 1027 
of race to, 1027 
of sex to, 1027 
malignant, autolysates in, 1031 
serum treatment, 1031 
vaccine treatment, 1031 
treatment of, 1026 
Turpentine injections in acne, 992 
in adnexal disease, 984 
in bronchopneumonia of influenza, 921 
in conjunctivitis, 1016 
in eczema, 994 
in gonococcus infections, 983 
in psoriasis, 995 
in pyelitis in children, 1012 
in pyogenic dermatoses, 993 
in skin diseases, 997 
Twort-d’Herelle phenomenon of bacteriol- 
ysis, 246 
d’Herelle’s method of securing bac- 
teriolytic agent, 247 
nature of bacteriolytic agent, 250 
. properties of bacteriolytic agent, 249 
sources and specificity of bacteriolytic 
agent, 247 
relation of, to immunity, 252 INDEX 
1205 
Typhoid and paratyphoid bacilli, triple vac- 
cine of, 795 
bacillus, 102 
in asthma, 1006 
in treatment of skin diseases, 997 
carriers, agglutination reactions for detec- 
tion of, 266 
vaccine treatment of, 903 
fever, agglutination test in, 264 
in diagnosis of, in vaccinated per- 
sons, 267 
anaphylactic reactions in, 712 
complement fixation in, 549 
cutaneous reaction in, 712 
Gruber-Widal reaction in, experimental 
work, 1128, 1129 
hemorrhage due to, transfusion for, 1081 
immunity to, 792 
non-specific treatment of, 904 
albumoses intravenously, 904 
contraindications, 906 
heterovaccines intravenously, 904 
mechanism of action of agents in, 905 
milk intramuscularly, 905 
precautions in, 905 
ophthalmic reaction in, 712 
relapses and complications in, effect of 
vaccine therapy upon, 903 
serum treatment of, 903 
treatment of, 901 
comparative value of different vaccines 
by subcutaneous and intravenous 
injections, 903 
effect of vaccine on relapses and com- 
plications in, 903 
with plain vaccine subcutaneously, 901 
with sensitized vaccine subcutaneously, 
902 
with vaccine intravenously, 902 
vaccination, 792, 901 
contraindications to, 795 
dosage of vaccine, 795 
duration and degree of, 797 
frequency and number of, 795 
method of inoculation, 794 
precautions in, 905 
reactions in, 796 
recommendations, 800 
revaccination, 797 
value of, 798 
Widal reaction in, 273 
Bass and Watkins’ slide method, 285 
paratyphoid, and cholera bacilli and Micro- 
coccus melitensis, pentavaccine of, 
795 
tetravaccine of, 795 
vaccine, dosage, 795 
in acne, 993 
in arthritis, 988 
contraindications, 989 
dosage and administration, 989 
results and practical value, 990 
preparation of, 793 
experimental work, 1125 
sensitized, 793 
of Gay and Claypool, 794 
Typhoidin, 647, 712 
of Force and Stevens, 713 
Typhoidin of Gay and Claypoole, 713 
test of Gay and Force, 712 
Typhoid-paratyphoid immunization, agglu- 
tination reaction after, 266 
vaccine, immunization with, value of, 798 
Typho-protein, 712 
Typhus fever, agglutination reaction in, 268 
immunity in, 801 
prophylactic immunization, 801 
serum treatment, 1037 
treatment of, 1037 
vaccination, 801 
Weil-Felix reaction in, 268 
Tyrotoxicon, 116 
Ulcerative blepharitis, vaccine treatment, 
1015 
Ulcers, corneal, vaccine treatment, 1016 
duodenal, hemorrhage due to, transfusion 
for, 1081 * 
gastric, hemorrhage due to, transfusion for, 
1081 
Umstimmung, 671, 705 
Undulant fever, treatment of, 1025 
Unit, agglutinin, standard, 283 
anticomplementary, 437 
antigenic, of syphilitic serum, 454 
antitoxin, 225 
Unitarian theory of Bordet of interaction of 
hemolysin, complement and corpuscles, 392 
Uremia, Wassermann reaction in, 504 
Urethritis, acute, gonococcus and mixed vac- 
cines in, 980 
mixed vaccines in, 980 
gonorrheal, acute, gonococcus and mixed 
vaccines in, 980 
Urethroprostatis, chronic, gonococcus and 
mixed vaccines in, 980 
Urine, collection of, 190 
in pneumonia, bacterial precipitinogens in, 
315 
Dochez and Avery’s precipitin reaction 
with, 332 
test, intracutaneous, of Wildbolz, 695 
Urticaria, 653 
treatment of, 996, 997 
Urticarial rashes in serum disease, 659 
Vaccination, 749, 766 
acute anterior poliomyelitis, 820 
anthrax, 829 
dosage of vaccine, 831 
preparation of vaccines, 830 
technic, 830 
value, 831 
with vaccine and serum, 831 
blackleg, 831 
preparation of aggressin, 831 
vaccine, 832 
canine distemper, 833 
rabies, 834 
cerebrospinal meningitis, 817 
chickenpox, 820 
cholera, 807 
dosage of vaccine, 809 
effects, 809 1206 
INDEX 
Vaccination, cholera, Haffkine’s vaccine, 808 
Kolle’s vaccine, 808 
preparation of vaccine, 808 
results, 809 
Strong’s vaccine, 808 
common cold, 815, 816 
diphtheria, 862. See also Diphtheria, 
toxin-antitoxin vaccination. 
dysentery, 810 
equine influenza, 834 
for curative immunization, 750 
for prophylactic purposes, 750 
hemorrhagic septicemia, 834 
history of, 748 
in lower animals, 829 
infectious abortion of cows, 832 
of mares, 833 
influenza, 811 
etiology in relation to, 812 
kinds of vaccines, 813 
value of, 814 
Jenner on, 120, 767 
measles, 821 
meningococcus meningitis, 873 
paratyphoid fever, 792. See also Para- 
typhoid fever vaccination. 
pertussis, 816, 817 
phenomena of, 778 
plague, 805 
dosage of vaccine, 807 
duration, 807 
Haffkine’s vaccine, 806 
dosage, 807 
effects, 807 
results, 807 
Kolle and Otto’s vaccine, 806 
Kolle’s vaccine, 806 
Lustig and Galeotti’s vaccine, 806 
preparation of vaccine, 806 
results, 807 
Terni and Bandi’s vaccine, 806 
pneumococci, experimental, 803 
pneumonia, 802. See also Pneumonia vac- 
cination. 
rabies, 782. See also Rabies. 
scarlet fever, 818 
smallpox, 766 
age for, 777 
certificates for, 779 
drill method, 775 
failure of, 777 
gauze method, 776 
history of, 766 
immunity reaction in, 779 
internal administration of vaccine, 777 
intracutaneous method of Wright, 776 
mechanism of, 778 
multiple acupuncture method, 775 
precautions, 777 
preparation of skin, 772 
of vaccine, 773 
protective value of, 781 
revaccination, 779 
risks of, 780 
scar in, 779 
scratch method, 772 
subcutaneous method of Goodal, 776 
subsequent care of wound, 775 
Vaccination, smallpox, subsequent examina- 
tions and reactions, 777 
syphilis after, 781 
technic, 772 
tetanus after, 780 
strangles, 834 
streptococcus infection in scarlet fever and 
puerperal sepsis, 819 
syphilis, 822, 828 
tuberculosis, 1047 
typhoid fever, 792. See also Typhoid fever 
vaccination. 
typhus fever, 801 
varicella, 820 
Weil’s disease, 828 
whooping-cough, 816, 817 
yellow fever, 822 
Vaccine, 749, 750 
ampule, 199 
conversion of test-tubes into, 7 
anthrax, 830 
antigenic activity of, specific, 755 
bacterial, 187, 749, 750 
adding a preservative, 198 
administration of, 761 
inoculation, method of making, 761 
autogenous versus stock, 755 
capillary pipet for counting, 192 
counting, 192 
Callison’s method, 194 
Hopkins’ method, 194 
Kolle’s method, 194 
nephelometer methods, 195 
Wright’s method, 193 
definition, 189 
deterioration of, 199 
diluting, 198 
dosage of, 197 
opsonic index as guide to, 187 
experimental work, 1125 
in treatment of hay-fever, 732 
inoculation with, 761 
constitutional reaction, 764 
dosage, 763 
effects of, 762 
focal reactions, 764 
frequency, 763 
intervals for, 763 
local reactions, 763 
method of making, 761 
mixed, preparation of, 198 
non-specific activity, 756 
of one micro-organism, preparation of, 198 
preparation of, 189 
sensitized, preparation of, 199 
standardization, 192 
stock versus autogenous, 755 
technic for preparing,_ 189 
counting suspension, 192 
emulsion, 191 
procuring infected material, 189 
pure cultures, 190 
sterilization, 196 
testing, 196 
Bandi and Terni’s plague, 806 
blackleg, 831 
by intravenous injection in pneumococcus 
pneumonia, 916 INDEX 
1207 
Vaccine by subcutaneous injection in pneu- 
mococcus pneumonia, 915 
cholera, preparation of, 809 
cowpox,-preparation of, 768 
animals, 769 
collection of virus, 769 
keeping, 771 
Noguchi’s method, 771 
testing, 770 
seed virus, 768 
vaccination, 769 
for common cold, 816 
for infectious abortion of cows, 832 
Galeotti and Lustig’s plague, 806 
gonococcus in acute urethritis, 980 
in arthritis, 980, 981 
in cervicitis, 981 
in chronic urethroprostatitis, 980 
in epididymo-orchitis, 980 
in vaginitis, 981 
Haffkine’s plague, 806 
dosage, 807 
effects, 807 
results, 807 
immunizing activity of, relation of method 
of preparation to, 751 
influenza, 813 
internal administration of, 777 
intramuscular injection of, mechanism of 
reaction following, 763 
intravenous, in arthritis, 981 
in epididymitis, 981 
in prostatitis, 981 
killed versus living, 751 
Kolle and Otto’s plague, 806 
Kolle’s cholera, 808 
plague, 806 
living versus killed, 751 
Lustig and Galeotti’s plague, 806 
meningococcic, 818 
mixed, 754 
in acute urethritis, 980 
in arthritis, 980, 981 
in chronic urethroprostatis, 980 
in epididymo-orchitis, 980 
nomenclature, 749 
of fungi, 196 
Otto and Kolle’s plague, 806 
paratyphoid, preparation of, 793 
sensitized, 793 
plague, preparation of, 806 
plain, intravenously, in treatment of ty- 
phoid fever, 902 
subcutaneously in treatment of typhoid 
fever, 901 
pneumococcus, dosage and administration, 
804 
kinds of, 803 
preparation of, 804 
preparation of, method, 750 
relation of method of, to immunizing 
activity of, 751 
prophylaxis in glanders, 998 
rabies, administration of, 789 
Harris’ method of preparing, 789 
Pasteur’s method of preparing, 788 
reactions to, 790 
Terrell’s method of preparing, 789 
Vaccine, sensitized, 753 
intravenously, in treatment of typhoid 
fever, 902 
subcutaneously, in treatment of typhoid 
fever, 902 
specific antigenic activity of, 755 
staphylococcus, preparation of, experi- 
mental work, 1125 
streptococcus, 819 
Strong’s cholera, 808 
subcutaneous injection of, mechanism of 
reaction following, 763 
TAB, 795 
Terni and Bandi’s plague, 806 
treatment, apophylactic phase of, 765 
causes of failure in, 760 
definition, 750 
ecphylaxis of Wright in, 760 
epiphylactic response of Wright in, 760 
kataphylaxis in, 760 
negative chemotaxis in, 760 
phase of, 765 
of abscesses, 888 
of acne, 992 
of actinomycosis, 998 
of acute anterior poliomyelitis, 1023 
infections, 759 
negative phase in, 760 
laryngitis, 1004 
mastoiditis, 1002 
otitis media, 1000, 1001 
pharyngitis, 1004 
rheumatic fever, 987 
rhinitis, 1002 
sinusitis, 1004 
tonsillitis, 1004 
of alveolitis, 1008 
of apical abscesses, 1008 
of appendicitis, 1010 
of arthritis, 987 
deformans, 987 
of Asiatic cholera, 975 
of asthma, 1005 
of bacillary dysentery, 943 
of bacteriuria, 1011 
of buboes, 984 
of carbuncles, 889 
of cholecystitis, 1009 
of chronic bronchitis, 1004 
dacro cystitis, 1015 
gingivitis, 1008 
mastoiditis, 1002 
otitis media, 1000, 1001 
rheumatic fever, 987 
rhinitis, 1003 
sinusitis, 1004 
of colitis, 1010 
of common cold, 1002 
of complications of scarlet fever, 898 
of conjunctivitis, 1016 
of corneal ulcers, 1016 
of cystitis, 1011 
of dermatitis herpetiformis, 996,r997 
of diphtheria carriers, 933 
of diseases of bladder, 1011 
of digestive tract, 1008 
of kidney, 1011 
of ecthyma, 992 1208 
INDEX 
Vaccine treatment of eczema, 993 
of erysipelas, 897 
of focal infection, 1008 
of furunculosis, 888 
of gall-bladder infections, 1009 
of gingivitis, 1008 
of glanders, 998 
of hay-fever, 1003 
of hepatitis, 1009 
of impetigo contagiosa, 993 
of infected perforating wounds of cornea, 
ip17 
of infections following cataract extrac- 
tions, 1017 
of influenza, 906 
of intestinal toxemia, 1010 
of iritis, 1017 
of keratitis, 1016 
of leprosy, 999 
of lichen planus, 996, 997 
of malignant tumors, 1031 
of Malta fever, 1026 
of mastoiditis, 1002 
of meningococcus meningitis, 960 
of neuritis, 1038 
of otitis media, 1000, 1001 
of ozena, 1003 
of pemphigus, 996 
of pertussis, 975 
dosage and administration, 976 
of pleuritis, 1006 
of pellagra, 996, 997 
of pneumococcus pneumonia, 915, 916 
of prurigo, 996, 997 
of pruritus, 996, 997 
ani, 997 
of scrotum, 997 
vulvae, 997 
of psoriasis, 994 
of pyelitis in children, 1012 
of pyelonephritis, 1012 
of pyodermia, 993 
of pyogenic dermatoses, 993 
of pyorrhea alveolaris, 1008 
of Rigg’s disease, 1008 
of ringworm, 995 
of salpingitis, 984 
of scarlet fever complications, 898 
of sciatica, 1038 
of sinusitis, 1004 
of sprue, 1010 
of staphylococcus infections, 889 
of streptococcus infections, 892 
of strophulus, 996, 997 
of styes, 1015 
of sycosis vulgaris, 992 
of syphilis, 1034 
of trichophytosis, 995 
of typhoid carriers, 903 
fever, 901 
of ulcerative blepharitis, 1015 
of urticaria, 996, 997 
phylactic power in, 760 
requisites for success in, 760 
triple, of typhoid and paratyphoid bacilli, 
795 
typhoid, dosage, 795 
in acne, 993 
Vaccine, typhoid, in arthritis, 988 
contraindications, 989 
dosage and administration, 989 
results and practical value, 990 
preparation of, 793 
experimental work, 1125 
sensitized, 793 
of Gay and Claypool, 794 
typhoid-paratyphoid immunization with, 
value, 798 
Vaccinia, 89, 748, 749, 778 
complement fixation in, 556 
smallpox and, relationship, 767 
Vaccinoid, 90, 778, 779 
Vaccinurin, 1038 
Vaginitis, gonococcus vaccine in, 981 
Varicella vaccination, 820 
Variola. See Smallpox. 
Varioloid, 780 
Vasomotor conjunctivitis, 653 
rhinitis, perennial, 653 
skin tests as guide to treatment by de- 
sensitization, 733 
treatment of, 733 
Vaughan and Wheeler’s theory of allergy, 623 
Vaughan’s theory of bacterial proteins, 112 
of immunity, 170 
Vegetable allergens, preparation of, 683 
poisons, hemolysis by, 384 
toxalbumins, 384 
Vein, anterior jugular, obtaining blood from, 
21 
Veitch’s technic in opsonic index, 184 
Venesection, 18 
in children, 18 
Venom, cobra, experimental work, 1116 
tests, technic, 416 
hemolysis, 414 
in cancer, 419 
in syphilis, 415 
experimental work, 1151 
practical value, 417 
technic, 416 
in tuberculosis, 419 
nature, 414 
psychoreaction of Much and Holzmann 
in, 418 
poisoning, 224 
snake, 105, 224 
nature of, 105 
properties of, 105 
solution, preparation of, 416 
Vernal conjunctivitis, 653 
Vernes’ reactions in syphilis, 523 
antigen for, 524 
direct method, 523 
technic, 525 
indirect method, 523 
technic, 525 
main test, 529 
readings, 529 
sheep corpuscles for, 526 
swine-serum for, 527 
Vincent’s open slide method in microscopic 
test for hemagglutinins, 303 
Virulence and toxicity of diphtheria bacilli, 
method of testing, 97 
of bacteria, 68 INDEX 
1209 
Virulence of bacteria, decrease in, 69 
increase in, 69 
by addition of animal fluids to culture- 
medium, 70 
by passage through animals, 69 
by use of collodion sacs, 70 
organ, 77 
Virus fix6, 121 
Vomiting, allergic, 655 
pernicious, of pregnancy, blood transfusion 
in, 1089 
von Behring’s method of toxin-antitoxin vac- 
cination against diphtheria, 862 
von Pirquet’s cutaneous tuberculin test, 689, 
696 
Vulva, diphtheria of, dosage of antitoxin in, 
928 
pruritus of, vaccine treatment, 997 
Walker’s method of preseasonal desensitiza- 
tion, 730 
Ward and Dreyer sigma, reaction of, in 
syphilis, 531 
Washing erythrocytes, 11 
Wassermann reaction, 442 
and precipitation, relation of lipoids to, 
584 
Colles’ law and, 511 
demand for standardization of technic, 
442 
discovery, 422 
of non-specific character, 423 
early work on, 423 
in cerebral syphilis, 510 
in congenital mental deficiency, 512 
in diabetes mellitus, 504 
in epilepsy, 512 
in frambesia, 504 
in general paralysis, 510 
in jaundice, 504 
in late pregnancy, 504 
in leprosy, 504 
in locomotor ataxia, 510 
in malaria, 504 
in non-syphilitic conditions, 503 
biologic and technical reasons for 
positive reactions in, 505 
in parasyphilitic diseases, 510 
in relapsing fever, 504 
in scarlet fever, 504 
in syphilis, 442 
acetone-insoluble lipoids for, 457 
alcoholic extracts of normal organs for, 
455 
reinforced with cholesterin 
for, 456 
of syphilitic livers for, 455 
antigens for, 453 
experimental work, 1145 
Kolmer’s new, 457 
method of diluting, 461 
preparation of, 454 
antigenic unit in, 454 
aqueous extracts of pallidum culture 
for, 459 
of syphilitic livers for, 455 
biologic and technical reasons for 
falsely negative reactions in, 505 
Wassermann reaction in syphilis, biologically 
non-specific nature of, 503 
cadaver serum in, testing, 446 
cerebrospinal fluid for, 444, 446 
cholesterolized and lecithinized alco- 
holic extract of heart muscle for, 
457 
clinical significance of, 503 
collecting blood for, 444 
serum, 445 
comparative antigenic values of vari- 
ous extracts, 459 
complement for, 447 
amount to employ, 449 
preservation of, 448 
congenital, 511 
Craig’s modification, 500 
doses of serum, 446 
effect of treatment upon, 512 
first method, 463 
technic, 469 
flocculation reaction and, comparison, 
536 
fourth method, 468 
reading results, 494 
technic, 491 
general technic, 444 
glassware, 444 
guinea-pig complement serum for, 
447 
heating serum, 446 
Hecht-Weinberg-Gradwohl modifica- 
tion, 502 
hemolysis for, 450 
hygienic laboratory method, 494 
Kolmer’s modification with multiple 
antigens, 463 
technic, 473 
with varying amounts of comple- 
ment, 468, 491 
of serum based upon studies 
in standardization of tech- 
nic, 463, 478 
latent, 510 
necessity for proper understanding of, 
503 
Noguchi’s modification, 496. See also 
Noguchi’s modification. 
original method, 463 
technic, 469 
pipets for, <444 
positive spinal fluid and negative 
blood reactions, 506 
practical specificity of, 503 
prenatal, 511 
primary, 508 
microscopic vs. serum tests, 508 
serum vs. microscopic tests, 508 
provocative stimulation in, 514 
qualitative and quantitative tests with 
antihuman and antichicken 
hemolytic systems, 491 
with antiox hemolytic system, 
490 
complement fixation, 488 
quantitative complement fixation, 486 
reading and recording, 471 
red blood corpuscles for, 451 1210 
INDEX 
Wassermann reaction in syphilis, relative 
value of qualitative and quantita- 
tive complement-fixation tests in, 
489 
second method, 463 
reading results, 477 
technic, 473 
secondary, 509 
serum for, 444 
collecting, 445 
standard, requirements of and how 
fulfilled by new technic, 464 
tertiary, 510 
test-tubes, 444 
third method, 463 
technic, 578 
titration of antigens, principles for, 462 
various methods for conducting, 463 
Wechselmann’s technic with serum, 
446 
in tabes dorsalis, 510 
in tuberculosis of lungs, 504 
in uremia, 504 
in various stages of syphilis, 507 
in yaws, 504 
negative, does not exclude possibility of 
syphilis, 506 
occurrence in non-syphilitic conditions, 
503 
positive, is not always an indication that 
lesion is syphilitic, 506 
unexpectedly, biologic reasons for, 507 
practical value of, 515 
Profeta’s law and, 512 
standardization of technic, Kolmer’s in- 
vestigations on, 443 
technical sources of error in, 507 
variation in results in different labora- 
tories, 507 
weakly positive, significance of, 507 
Water, bacteria in, 61 
hemolysis by, 381 
reaction in syphilis, Klausner’s, 520 
through straw during lumbar puncture, 852 
Watkins and Bass’s slide method for Widal 
test for typhoid fever, 285 
Weaver’s method of serum treatment in 
scarlet fever, 900 
Wechberg and Neisser’s method of measuring 
bactericidal power of blood, 370 
Weichardt and Dollken, cellular stimulation 
theory of, 884 
Weichardt’s cattle serum, graminol, 732 
plasma-activation, 641, 686, 884 
Weigert’s overproduction theory, 137, 204 
Weil-Felix reaction in typhus fever, 268 
Weil’s disease, milk injections in, 1038 
serum treatment, 1038 
vaccination, 828 
Weiss and Kolmer’s pneumotoxin, 716 
Welch, bacillus of, 102 
hypothesis of, 109, 176 
Wet cup for securing blood from children, 21 
cupping, obtaining blood by, 22 
Wheeler and Vaughan’s theory of allergy, 623 
White scours, serum treatment, 1040 
Whole blood, Zingher’s method, in treatment 
of scarlet fever, 900 
Whooping-cough, agglutination test in, 270 
anaphylactic reaction in, 717 
complement fixation in, 554 
immunity in, 816 
serum treatment, 975 
treatment of, 975 
vaccination, 816, 817, 975, 976 
Widal reaction in typhoid fever, 273 
Bass and Watkins’ slide method, 285 
experimental work, 1128, 1129 
Wildbolz intracutaneous urine test, 695 
Witte’s peptone in asthma, 1006 
in epilepsy, 1036 
Wodehouse method of collecting pollen, 729 
Wolff-Eisner’s conjunctival tuberculin test, 
689, 698 
dangers, 633 
in animals, 703 
Wollstein and Amoss’s rapid method of pre- 
paring antimeningococcus sferum, 948 
Wounds, dangerous, from standpoint of 
tetanus, 868 
diphtheria of, dosage of antitoxin in, 928 
infected perforating, of cornea, vaccine 
treatment, 1017 
prophylactic surgical treatment of, 870 
septic, vaccine treatment of, 890 
Wright’s blood capsules, making of, 6 
method of sealing, 17 
removing serum from, 17 
capillary pipet method for measuring bac- 
tericidal power of blood, 375 
ecphylaxis, 760 
epiphylactic response, 760 
intracutaneous method of vaccination, 776 
many stemmed pipet, 379 
method of counting bacterial vaccines, 193 
£-Rays, influence of, on antibodies and im- 
munity, 154 
Yaws, Wassermann reaction in, 504 
Yeast precipitins, 317 
Yellow fever, immunity in, 822 
serum treatment, 1037 
vaccination, 822 
Yersin’s plague serum, 971 
Zingher’s method of whole blood treatment 
in scarlet fever, 900 
Zooagglutinins, 255 
Zoo-anaphylactogens, 608 
Zooprecipitin, 304, 305, 317 
Zoosensitinogens, 608 
Zootoxins, 92, 104 
experimental work, 1116