DIAGNOSTIC METHODS WEBSTER DIAGNOSTIC METHODS CHEMICAL, BACTERIOLOGICAL AND MICROSCOPICAL A Text-book for Students and Practitioners BY RALPH W. WEBSTER, M. D„ Ph. D. ASSISTANT PROFESSOR OF PHARMACOLOGICAL THERAPEUTICS AND INSTRUCTOR IN MEDICINE IN RUSH MEDICAL COLLEGE, UNIVERSITY OF CHICAGO; DIRECTOR OF CHICAGO CLINICAL LABORATORY THIRD EDITION, REVISED AND ENLARGED WITH 37 COLORED PLATES AND 164 OTHER ILLUSTRATIONS PHILADELPHIA P. BLAKISTON'S SON & CO. 1012 WALNUT STREET 1913 First Edition, Copyright, 1909, By P. Blakiston's Son & Co. Second Edition, Copyright, 1912, By P. Blakiston's Son & Co. Third Edition, Copyright, 1913, By P. Blakiston's Son & Co. THE. MAPLE . PRESS • YORK- PA TO THE MEMORY OF MY FATHER DR. JOHN RANDOLPH WEBSTER THIS VOLUME IS LOVINGLY DEDICATED PREFACE TO THE THIRD EDITION Owing to the short time which has elapsed since the appearance of the second edition of this book, the author has found considerable difficulty in collecting and collating new material for this revision. It has been his policy to include only those tests which have become established by standing the fire of criticism or which have opened up somewhat new fields of investigation. He has, therefore, been more conservative than usual in this revision. Yet, not- withstanding this, many changes and inclusions may be noted. Among these one may find a discussion of streptococcic sore throat; Darling's method of staining amebas; Rose's method for creatin; Jolles' new modification of Spiegler's test for albumin; Tollens' naphtho-resorcin test for glycuronic acid; Hart's qualitative test for /?-oxybutyric acid; Pappenheim's beautiful panoptic blood stain; Dohle's "inclusion bodies" in scarlet fever; antityphoid vaccination; Tunnicliff's stain for treponema pallidum; Abderhalden's Sero-Diagnosis of Pregnancy has been given in great detail; and Collignon and Pilod's test for meningococcic meningitis. The subject-matter of the text has been brought thoroughly up to date by the insertion of a large number of references to the recent literature of the various subjects treated. The author wishes to express his extreme gratification at the very flattering reception of the second edition of his book and hopes that this third edition may prove even more acceptable to the special and general worker than have those which preceded it. 8 North State St., Chicago. Ralph W. Webster. VII PREFACE TO THE SECOND EDITION. The flattering reception which this book has received from the profes- sion has greatly encouraged the author in his work of revision. In the prep- aration of this second edition, he has endeavored to profit by the suggestion- and criticisms offered, so that he might increase the working-value of the book. Many paragraphs have been rewritten and the general subject-matter has been thoroughly revised and brought up to date by inclusion of the recent researches along lines already discussed in the first edition. In this edition much new matter has been added and some deletions have, necessarily, been made. The additions include only those methods which seem to promise most in diagnostic work, those which are still sub judice being reserved for later editions. Among these new insertions one finds the Antiformin method for tubercle bacilli; Much's method for staining tubercle bacilli; Neubauer & Fischer's test for gastric carcinoma; Gross' method for trypsin; Kendall and Day's method of isolation of typhoid bacilli from feces; Folin's newer methods for urinary sulphur compounds; the formalin method for ammonia in urine; colloidal nitrogen; Benedict's and Bang's tests for glucose in urine; the phenolsulphonephthalein test for functional activity of the kidneys; Wright and Kinnicutt's method of counting the blood plates; sulphhemoglobinemia; Noguchi's method of cultivating the treponema palli- dum; the tuberculin and luetin reactions; Ghoreyeb's and Burri's methods of staining the Spirochaeta pallida; and Noguchi's butyric acid test. The sections dealing with Bacteriology of the Blood and with Serum Pathology have been greatly enlarged. The Wassermann test has been discussed in great detail. The author wishes to express his deep obligation to all those who have aided him by their kind suggestions and criticisms. He is especially in- debted to his colleagues Drs. E. C. Rosenow, T. L. Dagg, and C. C. Croy for their advice regarding changes in the subject-matter in various parts of the work. 8 North State St., Chicago. Ralph W. Webster. PREFACE TO THE FIRST EDITION. In the present work the writer has attempted to bring together, for the use of the student and practitioner, the generally accepted facts regarding the various phases of clinical medicine, which may be rather more closely studied by the application of laboratory methods than without their use. It is to be especially emphasized that laboratory work must go hand in hand with the more direct clinical examination of the patient, as the former can be inter- preted only in the light of the latter. While it is true that, in some cases, the laboratory findings may be of even more value than those of physical examina- tion, yet it is to be understood that the function of the Clinical Laboratory is, more largely, perhaps, an accessory one, the application of methods of physical diagnosis usually pointing out the way toward a successful solution of the clinical problem by special laboratory methods, which yield numerous confirmatory or differential points not at all clearly defined by the methods of direct clinical examination. The aim of the author has been to present the direct bearing of the various methods outlined upon the clinical history of the case and to point out the special interpretation of the findings in any given examination. Particular attention has been directed to the selection of methods, both for the simpler clinical and more complex scientific requirements. As no method, no matter how exact it may be or how sound its basic principle, will yield reliable re- sults in the hands of the inexperienced, the writer has paid much attention to the details of such methods and has endeavored to direct the thought of the worker to the possible obstacles to be overcome before he is able properly to perform these examinations and interpret his results. The writer makes no claim for originality except, possibly, in the matter of arrangement of subject-matter and selection of methods and ideas, which have been established by others after years of earnest research. His endeavor has been to be as catholic as possible in his reading and as selective as his judgment permitted, to the end that the student or practitioner be saved the burden of sifting the wheat from the chaff. Numerous text-books, monographs, and special articles have been freely used in the preparation of the text, the writer attempting in each case to give due credit for such reference. It is possible that some direct use of material has been made which has not received deserved recognition. If so, the writer XI XII PREFACE. here acknowledges his indebtedness to them as well as to those whom he has directly quoted. It has seemed desirable to omit extensive reference to the literature, as a bib- liography to be of working value must be much more extensive and complete than is possible within the scope of this book. The writer has, however, in- serted at the end of each chapter a list of the more important larger works, which he has found useful in correlating the general subjects included in the several sections. In conclusion, the writer wishes to express his deep obligation to his col- leagues for their many valuable suggestions as to the subject-matter of the text: to Prof. W. S. Haines for assistance in revising the section on Urine; to Dr. J. M. Washburn for many additions and revisions of the section on Blood; to Dr. O. J. West for many practical points throughout the whole work; to Dr. W. A. Pusey for photographs of Blastomycetes under various conditions; to Dr. Brown Pusey for slides showing various organisms in the conjunctival exudates; and to Dr. N. Gildersleeve, of the University of Pennsylvania, for il- lustrations of megalosporon and microsporon. Further, the writer wishes to express his thanks to Miss Hill for the excellent work done by her in preparing the original drawings which appear throughout the work. Ralph W. Webster. ioo State St., Chicago. TABLE OF CONTENTS. CHAPTER I. The Sputum. Page I. General Considerations i II. Physical and Chemical Characteristics 2 Amount 2 Consistency 2 Reaction ,. . . 3 Color 3 Odor 4 Character 4 Chemistry 5 III. Macroscopic Examination 6 Cheesy masses 6 Dittrich's plugs 7 Curschmann's spirals 7 Fibrinous casts 8 Concretions 8 Bronchioliths 8 Pneumoliths 9 Echinococcus membranes . 9 Foreign bodies 9 IV. Microscopic Examination 9 Pus-cells 10 Red blood-cells 11 Epithelial cells 11 Elastic tissue 12 Crystals 13 Bacteria 15 Saprophytes 15 Pathogenic types 18 Tubercle bacillus 18 Lepra bacillus 23 Smegma bacillus 23 Timothy bacillus 24 Pneumococcus 24 Friedlander's bacillus 25 Influenza bacillus 25 Bacillus pertussis 25 Bacillus typhosus 26 Staphylococcus and streptococcus pyogenes .... 26 Bacillus pestis 26 Bacillus anthracis 26 Bacillus mallei 26 Actinomyces hominis 27 XIII XIV TABLE OF CONTENTS. Page Animal parasites 27 Amebae 27 Flagellates 28 Cestodes 28 Trematodes 29 V. The Sputa in Disease 29 Pulmonary tuberculosis 29 Croupous pneumonia 30 Broncho-pneumonia 30 Acute bronchitis 31 Chronic bronchitis 31 Simple type 31 Putrid type 31 Fibrinous type 32 Bronchial asthma 32 Influenza 32 Gangrene of the lung 32 Abscess of the lung 33 Perforating empyema 33 Pneumonoconioses 33 CHAPTER II. Oral, Nasal, Aural and Conjunctival Secretions. I. Oral Secretion 34 General considerations 34 Microscopic examination 35 Pathologic changes 36 Pharyngomycosis leptothrica 37 Diphtheria 37 Vincent's angina 39 Streptococcic sore throat 39 Gonorrheal stomatitis 39 Thrush 40 II. Nasal Secretion 4° General considerations . . 40 Pathologic changes 41 Hay fever 41 Meningitis 41 III. Aural Secretion 42 General considerations 42 Pathologic changes 42 IV. Conjunctival Secretions 42 General considerations 42 Pathologic changes 43 Diphtheritic conjunctivitis 43 Infectious conjunctivitis 43 Gonorrheal conjunctivitis 43 Trachoma 43 Vernal conjunctivitis 44 TABLE OF CONTENTS XV CHAPTER III. Gastric Contents. Page I. General Considerations 45 II. Methods of Obtaining Gastric Contents 47 Stomach-tube 47 Test meals ' 4g Ewald meal 49 Boas meal . . 50 Riegel meal 50 Fischer meal 50 Salzer meal 51 Sahli meal 51 III. Macroscopic Examination 52 Amount 52 Color 54 Odor 54 Consistency 54 Contents from fasting stomach 55 Vomitus 55 Contents after test meals 57 IV. Microscopic Examination 57 General 57 Food remnants 57 Boas-Oppler bacillus 57 Sarcinae ventriculi 58 Protozoa 58 Tissue fragments 58 Crystals 59 V. Chemical Examination 59 General 59 Total acidity 59 Free hydrochloric acid 60 Qualitative tests 61 Topfer's test . .■ . 61 Gunzburg's test 61 Boas' test 62 Tropeolin test 62 Quantitative tests 62 Mintz's method 63 Topfer's method 63 Amount of free hydrochloric acid 64 Euchlorhydria 65 Hypochlorhydria 65 Ana-chlorhydria 66 Hyperchlorhydria 66 Combined hydrochloric acid 66 Method of Martius and Liittke 66 Method of Tdpfer 67 Hydrochloric acid deficit 68 Organic acids 68 Total organic acids 68 XVI TABLE OF CONTENTS. Page Lactic acid 69 Uffelmann's test 70 Kelling's test 70 Strauss' method 71 Butyric acid 72 Acetic acid 72 Gastric ferments 73 Pepsin 73 Qualitative methods 73 Quantitative examination 74 Hammerschlag's method 75 Mett's method 75 Method of Thomas and Weber 76 Chymosin 76 Leo's method 77 Riegel's method 77 Lipase 77 Products of protein digestion 78 Products of carbohydrate digestion 78 Blood 78 Gases 78 Function of the stomach and contents 79 VI. Motility of the Stomach 80 Leube's method 81 Boas' method 81 Method of Ewald and Sievers 81 Winternitz' test 82 VII. Absorptive Power of the Stomach 82 Potassium iodide test 82 VIII. Indirect Examination of the Stomach Contents 83 Gunzburg's method .... 83 Sahli's desmoid reaction 83 IX. Gastric Juice in Disease 84 Hyperchlorhydria 84 Hypersecretion 85 Achylia gastrica 85 Acute gastritis 86 Chronic gastritis 86 Nervous dyspepsia 86 Ulcer of the stomach 87 Carcinoma of the stomach 87 Salomon's test 89 Neubauer and Fischer's test 89 CHAPTER IV. The Feces. I. General Considerations 92 Normal feces 92 Diet of Schmidt and Strasburger 93 Diet of Folin 94 TABLE OF CONTENTS. XVII Page Obtaining intestinal juice 94 Functions of the intestinal juice 95 Estimation of intestinal digestion 96 II. Macroscopic Examination 97 Method 97 Amount 97 Consistency and form 98 Odor 99 Color 99 Blood 101 Mucus 104 Pus 106 Food remnants 106 Protein residues 108 Fat residues 109 Carbohydrate residues no Biliary constituents no Intestinal sand and concretions 112 Tissue fragments 112 III. Microscopic Examination 112 Technic 112 Morphological elements 113 Crystals 114 IV. Chemical Examination . . 114 Reaction 115 Total solids 115 Total nitrogen 116 Fat 117 Carbohydrates 117 V. Bacteriology of the Feces 119 Technic 119 Cholera spirillum 121 Typhoid bacillus 121 Method of Drigalski and Conradi 122 Method of Kendall and Day 123 Bacillus of Dysentery 124 Tubercle bacillus • 124 VI. Parasitology of the Feces 125 Technic 125 Protozoa 125 Rhizopoda 126 Amebina 126 Ameba coli 126 Entameba coli 128 Sporozoa 128 Coccidium hominis 128 Flagellata 129 Trichomonas intestinalis 129 Cercomonas hominis 129 Megastoma entericum 129 Infusoria 130 Balantidium coli 130 XVIII TABLE OF CONTENTS Page Entozoa 130 Platodes 130 Cestodes 130 Taeniidae 132 Taenia solium 132 Taenia saginata 132 Taenia cucumerina 133 Taenia nana 133 Taenia diminuta 134 Taenia echinococcus 134 Bothriocephaloidea 135 Bothriocephalus latus 135 Dibothriocephalus cordatus 136 Bothriocephalus sp. Ijima et Kurimoto . . 136 Trematodes 136 Nematodes 137 Ascaridae *. 137 Ascaris lumbricoides 137 Ascaris mystax 138 Oxyuris vermicularis 138 Angiostomidae 139 Strongyloides intestinalis 139 Trichotrachelidae 141 Trichiuris trichiura 141 Trichinella spiralis 141 Strongylidae 143 Uncinaria duodenalis 143 Uncinaria Americana 144 Pseudo-parasites 146 CHAPTER V. Parasites. I. General Considerations 147 II. Trematodes 147 Fasciolidae • 148 Fasciola hepatica 148 Fasciolopsis Buski 149 Opisthorchis felineus 150 Opisthorchis sinensis 150 III. Nematodes 151 Eustrongylus gigas 151 IV. Parasites of the Skin 152 Arthropoda 152 Arachnoidea 153 Sarcoptes scabiei 153 Demodex folliculorum 154 Leptus autumnalis 154 Insecta 154 Hemiptera 154 Pediculus capitis 154 TABLE OF CONTENTS, XIX Page Pedi cuius vestimenti 154 Pediculus pubis 154 Cimex lectularius 155 Diptera 156 Pulex irritans 156 Pulex penetrans 156 Vegetable Parasites 156 Achorion Schonleinii 156 Trichophyton megalosporon endothrix 157 Microsporon Audouini 158 Microsporon furfur 160 Microsporon minutissimum 160 Blastomycetes 160 Sporothrix Schenckii 162 CHAPTER VI. The Urine. I. General Considerations 163 Collection and preservation of the urine 164 II. Physical Properties 165 Quantity 165 Polyuria 166 Oliguria 167 Anuria 168 Appearance 168 Color 169 Odor 172 Reaction 172 Folin's method for total acidity 173 Free mineral and organic acidity 174 Specific gravity 175 Technic 176 Rough estimate of total solids 177 Optical activity 179 III. Chemical Properties 179 Normal composition 179 Total solids and total ash 180 Inorganic constituents 181 Chlorids 181 Estimation of the chlorids 184 Quantitative determination 184 Volhard's method 184 Purdy's centrifugal method 187 Phosphates 187 Estimation of phosphates -190 Quantitative determination 191 Uranium method 191 Total phosphoric acid 194 Purdy's centrifugal method 194 XX TABLE OF CONTENTS Page Sulphur compounds. 195 Preformed sulphates 196 Ethereal sulphates 196 Neutral sulphur 197 Estimation of total sulphur 198 Folin's method 198 Determination of total sulphates 199 Folin's method 199 Determination of inorganic sulphates 199 Purdy's centrifugal method 200 Carbonates 200 Sodium and potassium 200 Calcium and magnesium 201 Iron 202 Organic constituents 203 Nitrogenous bodies 203 Total nitrogen 203 Kjeldahl's method 206 Urea 209 Determination of urea 211 Knop-Hiifner method 211 Doremus ureometer 212 Folin's method 213 Mdmer-Sjdqvist method 214 Ammonia 215 Quantitative determination 216 Method of Schldsing 216 Folin's method 218 Formalin method 219 Uric acid 219 Quantitative determination 223 Folin's method 223 Salkowski-Ludwig method 224 Method of Rudisch and Kleeberg . . 225 Ruhemann's method 227 Purin bases 228 Creatinin 229 Qualitative tests 231 Weyl's test 231 Jaffa's test 231 Quantitative determination 231 Folin's method 231 Undetermined nitrogen 233 Amino acids 233 Hippuric acid 234 Oxyproteic and alloxyproteic acids . . . .235 Allantoin 235 Fatty acids 236 Oxalic acid 236 Quantitative determination 237 Baldwin's method 237 TABLE OF CONTENTS XXI Ferments 238 Pepsin 238 Lipase 238 Mucin-like bodies 239 Mucin 239 Nucleo-albumin 239 Pigments and chromogens 241 Urochrome 241 Uroerythrin 241 Urobilin . . 241 Indican 243 Tests for indican 244 Jaffe's test 244 Obermayer's test 244 Rosenbach's test 245 Quantitative determination 245 Wang's method 245 Folin's method 246 Uroroseinogen 247 Abnormal Composition 247 Proteins 247 Serum-albumin 247 Albuminuria 247 Functional 248 Febrile 251 Traumatic 251 Hematogenous 252 Toxic 252 Neurotic 252 With definite renal lesions 252 Qualitative tests 253 Heat and acid test 254 Heller's nitric acid test 255 Ferrocyanide test 258 Sulpho-salicylic acid test 258 Spiegler's test 258 Quantitative methods 259 Scherer's method 259 Esbach's method 259 Method of Tsuchiya 260 Purdy's centrifugal method 261 Removal of albumin 261 Serum-globulin 261 Qualitative tests 262 Quantitative method 263 Proteoses 263 Primary proteoses 263 Bence-Jones' protein 263 Secondary proteoses 265 Tests 266 Bang's method 266 Clinical significance 267 XXII Page Peptone 268 Hemoglobin 268 Heller's test 269 Donogany's test 270 Fibrin 270 Histon and nucleo-histon 271 Carbohydrates 271 Glucose 271 Glycosuria 271 Qualitative tests 275 Trommer's test 276 Benedict's test 277 Fehling's test 278 Haines' test 279 Almen-Nylander's test 279 Fermentation test 280 Phenyl-hydrazin test 281 Quantitative tests 283 Feeling's method 283 Bang's method 284 Purdy's method 286 Haines' method ' 286 Polariscopic method 288 Fermentation method 291 Roberts' method 292 Levulose 293 Levulosuria 293 Seliwanoff's test 294 Phenyl-methyl-hydrazin test 294 Pentose 296 Pentosuria 296 Qualitative tests 297 Tollen's reaction 297 Orcin test 298 Quantitative test 298 Diphenyl-hydrazin method 298 Cammidge's reaction 299 Lactose 300 Lactosuria 300 Rubner's test 301 Maltose 301 Maltosuria 301 Glycuronic acid 302 Neuberg's quantitative method 305 Acetone bodies 305 Acetone 308 Qualitative tests 308 Legal's test 308 Lieben's test 309 Gunning's test 309 Frommer's test 309 TABLE OF CONTENTS. TABLE OF CONTENTS XXIII Page Quantitative methods 310 Huppert-Messinger method 310 Folin's method 312 Diacetic acid 312 Qualitative tests 312 Gerhardt's test 313 Arnold's test 313 Lipliawsky's test 313 ^-oxybutyric acid 314 Quantitative determination 315 Black's method 315 Shaffer's method 316 Abnormal pigments 317 Blood pigments 317 Hemoglobin 317 Hematoporphyrin 317 Biliary pigments 317 Qualitative tests 318 Smith's test 318 Gmelin's test 318 Rosenbach's test 319 Nakayama's test 319 Hammarsten's test . . 319 Bile acids 320 Hay's test 320 Oliver's test 320 Melanin 321 Phenol derivatives 321 Alkapton 321 Ehrlich's diazo reaction 322 Russo's reaction 324 Dimethyl-amino-benzaldehyd reaction 325 Drug reactions 325 IV. Microscopic Examination 325 Unorganized sediments 328 Those appearing in acid urine 328 Uric acid 328 Sodium acid urate 329 Potassium acid urate 329 Xanthin 330 Calcium oxalate 330 Cystin 331 Cystinuria 331 Leucin 332 Tyrosin 333 Calcium sulphate 334 Bilirubin 334 Hippuric acid 33 5 Neutral calcium phosphate . . .335 Fat 335 Chyluria 336 XXIV TABLE OF CONTENTS Page Those appearing in alkaline urine 336 Ammonium urate 336 Calcium tri-phosphate 337 Magnesium phosphate 337 Magnesium-ammonium phosphate 337 Calcium carbonate 338 Organized sediments 339 Mucoid material 339 Epithelial cells 339 Pus-cells 341 Pyuria 341 Vitali's test 343 Donna's test 343 Enumeration of pus-cells 343 Red blood-cells 344 Hematuria 344 Casts 345 True casts 345 Hyaline casts 345 Granular casts 348 Waxy casts 348 Fibrinous casts 349 Epithelial casts 349 Fatty casts 349 Blood-casts 350 Pus-casts 350 Cylindroids 350 Pseudo-casts 351 Cylindruria 352 Spermatozoa 353 Tissue fragments 353 Bacteria 353 Bacilluria 356 Parasites 356 V. Calculi 357 Heller's table for analysis 358 Uric acid calculi 359 Calcium oxalate calculi 359 Phosphatic calculi 359 Calcium carbonate calculi 359 Cystin calculi 360 Xanthin calculi 360 Urostealith calculi 360 VI. Functional DiaGNOsis 360 Cryoscopy 361 Electric conductivity -....361 Chlorid excretion 362 Methylene blue test 362 Phloridzin test 362 Phenolsulphonephthalein test 363 TABLE OF CONTENTS. XXV CHAPTER VII. Secretions of the Genital Organs. Page I. Male Secretions 366 General considerations 366 Microscopic examination 366 Pathologic variations 368 Medico-legal aspects 369 Florence's test 369 Barberio's test 370 II. Female Secretions 370 Vaginal secretions 370 Microscopic examination 370 Pathology 371 Blenorrhea 371 Purulent secretions 372 Fetid secretions 372 Uterine secretions 372 Menstruation 373 The lochia 373 Amniotic fluid 373 Abortion ..' 374 Vesicular mole 374 Carcinoma 375 CHAPTER VIII. The Blood. I. General Considerations 376 II. Physiology and Chemistry 377 Blood formation and blood-forming organs 377 Total volume of blood 378 Volume relations of cells to plasma 381 Methods of obtaining blood 383 Physical properties 384 Color 385 Odor 386 Reaction 386 Specific gravity 390 Viscosity 392 Coagulation 393 Osmotic pressure and cryoscopy 396 Electric conductivity 398 Chemical properties 399 Total solids 4°i Blood pigments 4O1 Hemoglobin . 401 Pseudo-hemoglobin 4° 3 Oxy-hemoglobin 4°3 Met-hemoglobin 4°3 Carbon monoxid-hemoglobin 4°4 TABLE OF CONTENTS XXVI Page Carbon-dioxid-hemoglobin 404 Sulph-hemoglobin 404 Decomposition products 405 Hematin 405 Hematoporphyrin 406 Hematoidin 406 Hemosiderin 407 Malarial pigment 407 Estimation of hemoglobin 407 Direct methods 408 Indirect methods 408 Hemometer of Fleischl-Miescher 409 Hemoglobinometer of Dare 412 Hemometer of Sahli 413 Hemoglobinometer of Oliver 414 Hemoglobinometer of Tallqvist 416 Variations in amount of hemoglobin 418 Oligo-chromemia 418 Color index : . 418 Proteins of the blood 419 Other nitrogenous constituents 422 Total nitrogen 422 Urea 422 Uric acid 423 Xanthin bases 424 Ammonia 424 Carbohydrates 425 Fats and fatty acids 426 Acetone 427 Biliary constituents 427 Inorganic constituents 428 Blood gases 430 Ferments of the blood 430 Enumeration of the cells 431 Hemocytometer of Thoma-Zeiss 432 Hemocytometer of Durham 443 Hemocytometer of Oliver 444 III. Morphology of the Blood 445 Examination of fresh blood 446 Preparation of smears 447 Fixation of smears '..450 Staining methods 453 Erythrocytes 462 Appearance and structure 462 Size and shape 464 Nucleation 466 Number 469 Normal variations 469 Pathological variations 471 Oligo-cythemia 471 Poly-cythemia 472 Staining properties 473 TABLE OF CONTENTS. XXVII Page Degenerations 475 Isotonicity and resistance 476 Variations in childhood and old age 478 Functions 478 Leucocytes 479 Appearance 479 Types in normal blood 479 Lymphocytes . . . • 479 Large mononuclears 480 Polymorphonuclear neutrophiles 481 Polymorphonuclear eosinophiles 483 Polymorphonuclear basophiles 484 Types in pathological blood 485 Myelocytes 485 Irritation forms 486 Degeneration forms 486 Differential counting 487 Number 488 Leucocytosis 489 Physiological 490 Pathological 492 Mixed leucocytosis . 495 Lymphocytosis 496 Eosinophilia 496 Mast-cell type 498 Leucopenia 498 Variations in infancy and childhood 499 Functions 499 Blood-plates 501 Appearance ..501 Size 501 Number 501 Staining properties 502 Function 502 Hemoconien 502 Morphology of the blood-forming organs 503 IV. Pathology of the Blood 505 Special 505 Anemia 506 Primary 506 Simple primary anemia 506 Chlorosis 506 Progressive pernicious anemia 508 Splenic anemia 510 Anemia infantum pseudo-leukemica 511 Leukanemia 511 Aplastic anemia 512 Secondary 512 Acute hemorrhage 513 Chronic hemorrhage 514 Inanition 514 Intestinal parasites 515 XXVIII TABLE OF CONTENTS. Page Fever 515 Blood poisons 516 Leukemia 517 Spleno-myelogenous type 517 Lymphatic type 520 Acute type 521 Pseudo-leukemia 522 Hodgkin's disease 522 Tuberculosis of the lymph glands 523 Lympho-sarcoma 523 Gummatous lymphoma 523 General 523 Blood changes following surgical intervention 523 Constitutional diseases 525 Diabetes mellitus 525 Gout 526 Addison's disease . . 52$ Rickets 526 Myxedema 527 Acute infections 527 Pneumonia 528 Typhoid fever 53° Scarlet fever 531 Measles 532 Variola 532 Diphtheria 533 Pertussis 534 Rheumatism 534 Chronic infections 535 Tuberculosis 535 Syphilis 53 $ Leprosy 537 Carcinoma 537 Effects of splenectomy 53$ V. Parasitology of the Blood 539 Malaria 539 Examination of fresh blood 54* Tertian parasite 541 Quartan parasite 543 Estivo-autumnal parasite 544 Stained specimens 54$ Sporogony 548 General hematological changes 55° Relapsing fever 55 2 Sleeping sickness 553 Kala-Azar 554 Filariasis 555 Syphilis 55$ Cultivation of treponema pallidum 559 Yellow fever S^1 Rocky mountain spotted fever 5$2 Distomiasis 5^3 TABLE OF CONTENTS XXIX VI. Bacteriology of the Blood 563 Technic ' 563 Organisms found in the blood 565 VII. Serum Pathology 571 Ehrlich's side-chain theory 571 Phagocytosis 576 Opsonins 576 Diagnostic use of vaccines 579 Tuberculin reactions 579 Luetin reactions 581 VIII. Sero-Diagnosis 583 Agglutination reactions 583 Gruber-Widal test 584 Method of Bass and Watkins 588 Diseases other than typhoid 590 Precipitin reaction 590 Complement-fixation test 594 Wassermann reaction 596 Noguchi method 605 Diseases other than syphilis 612 Sero-diagnosis of Pregnancy 613 IX. Medico-legal Aspects 622 Red cells 622 Guaiac test 623 Schaer's test 623 Phenolphthalin test ....624 Teichmann's test .... 4 625 Spectroscopic examination 626 Precipitin test 626 X. Value and Limitations of Blood Examinations 627 CHAPTER IX Transudates and Exudates. I. General Considerations 630 II. Physical and Chemical Properties 631 Serous exudates 632 Chylous exudates 633 Chyloid exudates 633 Hemorrhagic exudates 633 Purulent exudates 634 Putrid exudates 634 III. Bacteriology 635 Tubercle bacilli 635 Inoscopy 635 Gonococci 635 Smegma bacilli 637 Ducrey's bacilli 637 Spirochete pallidae 637 IV. Cytology 639 Technic 639 Cytology of normal fluids 640 XXX TABLE OF CONTENTS. Page Cytology of pathological fluids 640 Pleural exudates 641 Primary tubercular pleurisy 641 Secondary tubercular pleurisy ..641 Pneumococcus pleurisy 641 Streptococcus pleurisy 642 Typhoid pleurisy 642 Malignant pleurisy 642 Nephritic and cardiac pleurisy 642 Peritoneal exudates 642 V. Cyst Fluids 643 Ovarian cysts 643 Serous cysts 643 Myxoid or colloid cysts 643 Papillary cysts 644 Dermoid cysts 644 Parovarian cysts 644 Hydrocele 644 Spermatocele 645 Hydronephrosis 645 Hydatid cysts 645 Pancreatic cysts 645 VI. Cerebrospinal Fluid 645 Lumbar puncture 646 Microscopic examination 648 Epidemic cerebrospinal meningitis 649 Tubercular meningitis 650 Acute anterior poliomyelitis 650 Cerebrospinal syphilis 651 Noguchi's butyric acid test 651 CHAPTER X. Secretion of the Mammary Glands. I. General Considerations 652 II. Physical and Chemical Properties 653 Appearance and color 654 Specific gravity 654 Reaction 655 Coagulation 655 Total solids 655 Ash 655 Protein 655 Total protein 656 Method of Sebelien 656 Method of Boggs 656 Casein 657 Albumin and globulin 657 Fat 657 Babcock's method 657 Extraction method 658 TABLE OF CONTENTS. XXXI Page Lactose 659 Preservatives in cow's milk 659 Sodium carbonate 659 Salicylic acid 660 Formaldehyde 660 Boric acid and borax 660 III. Bacteriological Examination of Milk . • 660 Index 663 LIST OF ILLUSTRATIONS. PLATES. To Face Page I. Tubercle Bacilli in Sputum 20 II. Streptococcus Pyogenes 26 III. Leptothrix and Spirocheta Buccalis (Unstained) 35 IV. Diphtheria Bacilli Showing Polar Staining 38 V. Koch-Weeks Bacillus 43 Morax-Axenfeld Diplobacillus . . . 43 VI. Trachoma Bodies of Prowazek-Greeff 44 VII. Vegetable Cells found in Feces 146 VIII. Osazons 281 IX. Ammonium Urate Crystals 337 X. Waxy Casts 348 XI. Mucous Threads in Urine (Unstained) 351 XII. Cystitis Due to Colon Bacillus 354 XIII. Staphylococcus Cystitis 356 XIV. Absorption Spectra 402 XV. Absorption Spectra 403 XVI. Fresh Normal Blood 447 XVII. Types of Red Cells 465 XVIII. Ring Bodies in Red Cells 476 XIX. The Leucocytes 479 XX. lodophilia 487 XXL Polynuclear Leucocytosis 489 XXII. Eosinophilia 497 XXIII. Chlorotic Anemia 507 XXIV. Blood in Pernicious Anemia 509 XXV. Blood in Leukanemia 511 XXVI. Blood in Spleno-myelogenous Leukemia 518 XXVII. Lymphatic Leukemia 521 XXVIII. The Tertian Parasite (Unstained) 541 XXIX. The Quartan Parasite (Unstained) 543 XXX. The Estivo-autumnal Parasite (Unstained) 545 XXXI. Tertian Parasite (Stained) 546 XXXII. Estivo-autumnal Parasite (Stained) 548 XXXIII. Gonococci in Urethral Discharge 636 XXXIII XXXIV LIST OF ILLUSTRATIONS. To Face Page XXXIV. Spirochetae Pallidae in Tissue 638 XXXV. Exudate from Tubercular Pleurisy 640 XXXVI. Exudate in Pneumonic Pleurisy 641 XXXVII. Exudate in Malignant Pleurisy 642 FIGURES. Page i. Curschmann's Spirals 7 2. Objects Found in the Sputum 10 3. Aspergillus Fumigatus r6 4. Micrococcus Catarrhalis 16 5. Budding Forms of Blastomycetes 17 6. Diplococcus Pneumoniae 24 7. Friedlander's Bacillus 25 8. Bacillus Influenzae ' 25 9. Actinomyces 27 10. Paragonimus Westermanii 28 11. Ovum of Paragonimus Westermanii 29 12. Vincent's Spirillum and Bacillus 39 13. Oidium Albicans 40 14. Stomach Tube 47 15. Turek's Aspiration Apparatus 48 16. Boas-Oppler Bacillus 58 17. Strauss'Separatory Funnel 71 18. Sahli's Desmoid Bag 83 19. Normal Feces 93 20. Boas' Stool-Sieve 97 21. Schmidt's Fermentation Apparatus 118 22. Cholera Spirilla 121 23. Bacillus Typhosus 122 24. Amoeba Coli 127 25. Coccidium Hominis 128 26. Trichomonas Intestinalis 129 27. Cercomonas Hominis 129 28. Megastoma Entericum 130 29. Balantidium Coli 130 30. Taenia Solium 132 31. Taenia Saginata 133 32. Taenia Cucumerina 133 33. Taenia Nana 134 LIST OF ILLUSTRATIONS. XXXV Page 34. Taenia Diminuta 134 35. Ovum of Taenia Diminuta 134 36. Taenia Echinococcus . 135 37. Hydatid Cyst 135 38. Bothriocephalus Latus 136 39. Dibothriocephalus Cordatus 136 40. Ascaris Lumbricoides * x 137 41. Ascaris Mystax .' 138 42. Oxyuris Vermicularis 139 43. Strongyloides Intestinalis 140 44. Trichiuris Trichura . . . 141 45. Trichinella Spiralis 142 46. Tail of Uncinaria Duodenalis 143 47. Anterior End of Uncinaria Duodenalis 143 48. Tail of Uncinaria Americana 144 49. Anterior End of Uncinaria Americana 144 50. Parasitic Bodies, Ova, and Larvae 145 51. Fasciola Hepatica 148 52. Fasciolopsis Buski 149 53. Opisthorchis Felineus 150 54. Opisthorchis Sinensis 151 55. Eustrongylus Gigas . . 152 56. Acarus Scabiei 153 57. Demodex Folliculorum 154 58. Leptus Autumnalis 154 59. Pediculus Capitis 155 60. Pediculus Vestimenti 155 61. Pediculus Pubis 155 62. Pulex Irritans 156 63. Pulex Penetrans 157 64. Achorion Schbnleinii 158 65. Normal Hair 159 66. Trichophyton Endo-ectothrix 159 67. Microsporon Audouini 160 68. Mycelial Threads of Blastomycetes 161 69. Urinometer and Cylinder 177 70. Volumetric Flasks 192 71. Kjeldahl's Nitrogen Apparatus 207 72. Doremus' Ureometer - 212 73. Doremus-Hinds Ureometer 213 74. Folin's Urea Apparatus 214 75. Schlbsing's Ammonia Apparatus 217 76. Folin's Ammonia Apparatus 218 77. Folin's Absorption Bulb 219 XXXVI LIST OF ILLUSTRATIONS. Page 78. Ruhemann's Uricometer 227 79. Sargent's Colorimeter 232 80. Conical Test-glass 256 81. Horismascope 256 82. Esbach's Albuminometer 259 83. Laurent Polariscope 288 84. Diagrammatic Representation of the Course of Light through the Laurent Polariscope : 289 85. Einhorn's Saccharometer 290 86. Lohnstein's Fermentation Tube for Undiluted Urine 292 87. Lohnstein's Fermentation Tube for Diluted Urine 292 88. Purdy Electric Centrifuge 325 89. Sediment Tube 326 90. Percentage Centrifuge Tube 326 91. Various Forms of Uric Acid 328 92. Acid Sodium Urate » 329 93. Xanthin 329 94. Calcium Oxalate 330 95. Cystin 331 96. Pure Leucin 333 97. Impure Leucin 333 98. Tyrosin 334 99. Calcium Sulphate 334 100. Bilirubin 334 101. Cholesterin 336 102. Magnesium-Ammonium Phosphate 338 103. Calcium Carbonate 338 104. Urinary Epithelium 339 105. Pus Corpuscles 341 106. Hyaline Casts 346 107. Granular Casts 347 108. Epithelial Casts 348 109. Fatty Casts 349 no. Blood, Pus, Hyaline, and Epithelial Casts 350 in. Cylindroids 351 112. Scolex and Hooklets of Taenia Echinococcus in Urine 356 113. Ova and Miracidium of Schistosomum Hematobium 357 114. Normal Semen 367 115. Chorionic Villi 374 116. Daland's Hematocrit 381 117. Hematocrit Tube 382 118. Blood Needle 384 119. Dare's Hemoalkalimeter 387 120. Pycnometer 391 LIST OF ILLUSTRATIONS. XXXVII Page 12i. Wright's Coagulometer 394 122. Boggs' Coagulometer 396 123. Beckmann Apparatus 397 124. Direct-vision Spectroscope 402 125. Hemometer of Fleischl-Miescher 409 126. Hemoglobinometer of Dare 412 127. Method of Filling the Dare Blood Pipet 413 128. Hemometer of Sahli 414 129. Hemoglobinometer of Oliver 415 130. Tallqvist's Hemoglobinometer 416 131. Thoma-Zeiss Counting Chamber 432 132. Diluting Pipets 433 133. Ruled Surface of Thoma-Zeiss Counting Chamber 434 134. Turk's Ruling of Counting Chamber 434 135. Plan of Counting the Cells 440 136. Cross-section of Durham's Blood Pipet 444 137. Oliver's Hemocytometer 444 138. Preparation of Blood-smears with Glass Slides 448 139. Preparation of Blood-smears with Cigarette Paper 448 140. Ehrlich Forceps 449 141. Pinch Forceps 449 142. Oven for Fixing Blood-films 451 143. Normal Blood Showing Rouleaux Formation and Fibrin Network . 463 144. Cycles of the Malarial Parasite 549 145. Spirillum of Obermeier 552 146. Trypanosoma Gambiense 553 147. Filaria Bancroft! 555 148. Spirochetae Pallidae and Refringens 558 149. Ultra-condenser of Reichert 559 150. Schistosomum Hematobium 562 151. Illustrating the Mechanism of Toxin-cell Union 572 152. Illustrating the Elaboration and Action of Antitoxin 573 153. Illustrating the Mechanism of Hemolysis 574 154. Illustrating the Mechanism of Antihemolysis 575 155. Bacillus Typhosus at Beginning of Widal Test 583 156. A Pseudo-Widal Reaction 584 157. A Positive Widal Reaction 584 158. Hemin Crystals from Human Blood 625 159. Lumbar Puncture 646 160. Diplococcus Intracellularis Meningitidis 649 161. Normal Milk and Colostrum 652 162. Babcock Milk and Cream Bottles 658 163. Bottle for Human Milk 658 164. Soxhlet Apparatus 659 DIAGNOSTIC METHODS CHAPTER I. THE SPUTUM. I. General Considerations. At the present time the examination of the sputum is more or less lim- ited to the search for various specific organisms, especially the bacilli of tuber- culosis and of pneumonia. This is very much to be regretted, as frequently the appearance, amount, consistency, color, and other characteristics are of great aid and have led our older brothers to correct diagnoses before the days of microscopic examination. The sputum, strictly speaking, should be considered as the material which comes from the respiratory passage anywhere along its course. It may be, therefore, of laryngeal, bronchial, or alveolar origin. More commonly, how- ever, we find the sputum considered by the general practitioner and by the laity as anything which is expectorated, so that specimens which consist of nothing more than salivary secretion are frequently sent to laboratories for examination. In some cases, especially of inflammation of the naso-pharynx or perforation from neighboring organs, this buccal secretion may be mixed with material from the nose, mouth, ear, or esophagus. For these reasons, if for no others, it is absolutely essential that frequent examinations of sputum, in the strict sense, be made before a negative diagnosis of a suspected condition may be given. As a rule, any sputum at all should be considered pathological, as normal persons raise little or nothing from the lungs at any time. By this is not meant that serious disease may obtain when a small amount of sputum exists, as patients suffering with catarrhal conditions of the naso-pharynx frequently have an accumulation of material, which has settled in the bronchial tubes over night and is raised in the morning. Those of us who are unfortunate enough to live in atmospheres which are loaded with soot and dirt frequently raise a certain amount of sputum which arises from an increased activity of the mucous mem- brane of the respiratory passages to compensate for the dryness and irritation which these foreign substances have caused. This morning sputum is small in amount and is in the form of large, tough, elastic masses which very much resemble boiled sago. These masses are at times extremely dark in color, due 2 diagnostic methods. to the dirt which has been taken in with the inhaled air. In such sputum we find much mucus, degenerated epithelium, pus cells and various micro-organ- isms. These organisms are rarely of pathologic significance, although many pathogenic types may be present. The sputum should be collected in receptacles which may be completely and easily disinfected. The best receivers for the sputum are the ordinary paper spit-cups which can be burned as occasion may demand. The practice of ex- pectorating upon cloths is only to be advised when these cloths are immediately burned. In the spit-cup should always be contained a certain amount of disin- fectant, such as dilute carbolic acid or corrosive sublimate solution, so that no chance of transference of infection may arise. Neglect of such precautions has frequently led to serious consequences in the case of the healthy members of the family. When sputum is to be collected for examination, the material raised by coughing should be received in a wide-mouthed bottle which contains no disinfectants, as these agents coagulate the protein material of the infecting organism and in some cases change its staining characteristics. Immediately after examination of such material it should, of course, be treated with the dis- infecting solution. II. Physical and Chemical Characteristics. Amount. Some general idea of the amount of sputum expectorated is always advis- able. It is rarely necessary, however, to make any collection of the material during the 24 hours' period, as one can usually gain the information by question- ing the patient or nurse as to the amount passed from time to time. The amount of sputum expectorated in 24 hours depends very much upon the nature of the pathologic condition. In cases of so-called dry bronchitis, diffuse bronchitis, early tuberculosis, and occasionally of lobar pneumonia, the spu- tum is so viscid that there is practically none obtained. In cases of chronic bronchitis, tuberculosis with cavity formation, and bronchiectasis we find large amounts; while in cases of lung abscess or of perforating pleurisy, blood or pus may flow from the mouth in very large quantities. The absolute amount of sputum may, therefore, vary from a cubic centimeter to a liter or more. An extensive expectoration will naturally have more or less serious effect upon the patient's general nutrition, so that we are not surprised to find cases in which as high as 5 per cent, of the total nitrogen eliminated passes from the system in the sputum (Lenz). Consistency. As a rule, the consistency of the sputum varies from that of a liquid to a highly tenacious material, inversely as the amount of sputum. This latter statement holds in most cases with the exception of pneumonia, in which we have an extremely tenacious sputum and, also, a very abundant one. Just what substance induces this extreme tenacity is uncertain, but it may be gener- THE SPUTUM. 3 ally said that mucin is the causative factor, although the tenacious pneumonic sputum shows very little mucin. According to Kossel, the tenacity may be due to the presence of nuclein derivatives. In the early stages of acute bron- chitis, or bronchial asthma, and in whooping-cough the sputum is usually very tenacious and ropy; while in edema of the lungs, pulmonary abscess, putrid bronchitis, and pulmonary gangrene it is very watery and contains large num- bers of pus-cells. Reaction. The fresh specimen of sputum is usually alkaline in reaction. However, in cases in which the sputum has remained in the lungs for some time, as in cavity formation, the reaction is acid. Color. The color of the sputum may range from that of a colorless material to one showing any of the tints of the rainbow. These colors are due to admixtures of various abnormal products with the sputum. A bright red sputum is sig- nificant of the presence of blood, according to Traube, unchanged red blood- cells necessarily being present. The amount of blood may vary from a light streaking of the sputum to one showing a deep red, rusty, or prune-juice color. Such bloody sputa are found after trauma, pneumonia, gangrene, hemorrhagic infarction of the lungs, chronic passive congestion, as well as adventitious mix- ture of sputum with nasal or pharyngeal material containing blood. The blood may be due to rupture of a vessel and may then constitute the condition known as pulmonary hemorrhage. This condition of hemoptysis is frequently confounded with that of hematemesis and is differentiated by the fact that the blood in hemoptysis is frothy, bright red in color, alkaline in reaction, and usu- ally associated with mucopurulent material; while the blood in hematemesis is frequently dark and grumous, usually clotted and acid in reaction. In sputa which are tinted by changed hemoglobin, the color is most varied. Many oxidation products of hemoglobin are found in the sputum, depending upon the time which the sputum has lain in the lungs. Thus, for instance, in pneu- monia we find a rusty, prune-juice colored sputum, whose color seems to be due to an unknown derivative of hemoglobin. The shade of color in pneumonia may, however, range from red through brown to green. In cases of mitral disease associated with passive congestion of the lungs we find, frequently, a light brown color, due to the presence of hematoidin granules in the epithelial cells. In many cases of abscess of the liver which perforates into the lung or in catarrhal jaundice, bile pigments may be found in the sputum which may give rise to various colored sputa from red through blue to green. Although, as shown in a later table, the bile pigments and hemoglobin are very closely re- lated chemically, yet the clinical significance of the appearance of bile pigment in such cases as the above is very important. In many cases of true jaundice the sputum, in case any exists, may show a distinct grass-green color due to 4 DIAGNOSTIC METHODS. the presence of oxidized bile pigments. This same green color frequently appears in the sputum, however, in cases of croupous pneumonia during lysis. In this latter case the color is due, probably, to the same pigment which is oxi- dized before expectoration. In these cases of pneumonia with green sputum and no jaundice, a fresh involvement of the lung is usually associated with a rusty sputum. Besides jaundice and pneumonia we may have, as causes of green sputum, certain chloromata of the lungs as well as the development of chromogenic bacteria within the lungs. This action of chromogenic bacteria is not always observed when the sputum is expectorated, but may appear only after the sputum has stood. The presence of the bacillus pyocyaneus may give a sputum which is brilliant blue or greenish in tint. Sputa very frequently show changes in color as well as consistency due to various substances inhaled. Thus we find a distinctly black sputum, in cases of anthracosis, in coal-miners and in many city residents. So frequent is this occurrence that the lung tissue may even be invaded by the coal pigment. The writer has been told by one of our most prominent pathologists that he has never seen, postmortem, a lung of a city resident which did not contain large amounts of coal pigment. This would partly substantiate the statement made by some writers that the pigment in the sputum is only the dust which had been taken up by the phagocytic cells as this dust is on its way to the lungs. Workers in bronze and brass as well as other metals frequently show a spu- tum tinged red with ferric oxide, arising from a condition of the lung known as siderosis. Stone-cutters frequently show much stone-dust in the sputum which is characteristic of the condition of chalicosis, stone-cutters'phthisis, or grinders' rot. Workers in flour mills and in bakeries frequently expectorate doughy masses, while those in cotton mills show the presence of cotton fibers in their sputum. Finally it may be said that the color of the sputum may be changed by the presence of certain foods, such as milk, eggs, and chocolate, while tobacco users frequently have a sputum tinged dark brown. Odor. Ordinarily the sputum has no odor unless it has stagnated either in the receiving cup or in the system. Such old sputum sometimes has a very distinct putrefactive odor. The odor of sputum in tuberculosis and bronchiectasis is peculiarly heavy and sweet, while that in putrid bronchitis is often extremely offensive. In cases of perforating empyema a peculiar cheese-like odor is observed. While these odors are not characteristic in themselves, they are usually more or less significant of the condition with which they are commonly associated. Character of the Sputum. The character of the sputum has reference more to the apparent com- position of the sputum than to its consistency. Air is usually present in the sputum in various amounts, so that one may judge, from the size of the air- bubbles, of the size of the bronchi from which the sputum came. The sputum THE SPUTUM. 5 from cavities and large bronchi contains no air and, therefore, sinks in water. This is the so-called sputum fundum petens. Sputum known as mucoid sputum is glairy, transparent, and tenacious, becoming cloudy on the addition of acetic acid due to the precipitation of mucin. This type of sputum is found particularly in acute bronchitis and in asthma. A mucopurulent sputum is one containing both pus and mucoid material. Small amounts of pus give a whitish color, either to the whole or to portions of the sputum, the pus being observed in masses or in streaks through the mucoid material. Larger quantities of pus give a yellowish or occasionally a yellowish- green tinge to the sputum. In this latter type the pus and mucus seem to be mixed homogeneously. In the sputum from cavities we find the mucopuru- lent material arranging itself flatly like a coin, constituting the so-called "num- mular" sputum. Purulent sputum is found in cases of ruptured empyema, abscess of the lung, and in some cases of bronchiectasis. This purulent sputum differs from the mucopurulent type in the fact that the pus is much more abundant and is almost in the pure state, being mixed with a small amount of tenacious mucus. In some cases, especially in edema of the lungs, a sputum is obtained, known as serous sputum, which is colorless and quite frothy. This sputum resembles very closely the ordinary salivary secretions and should not be confused with it. In many cases, especially in putrid bronchitis, gangrene of the lung, and bronchiectasis, the sputum on being voided into a cylinder will separate into three distinct layers, occasionally into four. The upper layer is of frothy mucus; a second layer, which is not always present, consists of certain albumin- ous material which hangs in long shreds down into the third layer, which con- sists of the sero-pus and is usually opaque and watery. The bottom layer contains the morphological elements, pus, tissue shreds, and bacteria. Besides the varieties of sputum named above, the admixture of blood may give rise to sputum which is known as sanguinous sputum or as sanguino-muco- purulent or purulent sputum. Chemical Properties. The chief chemical examination is applied to the detection and estimation of the amount of albumin and of mucin in the sputum. These methods are discussed in the section on urine, to which the reader is referred. It goes, without saying, that a large amount of albumin indicates such conditions as pneumonia, pulmonary edema or perforating empyema. Of greater diag- nostic importance, however, are the relatively small amounts of albumin. Wanner1 regards anything more than a slight opalescence as pathological, a trace of precipitate indicating a pulmonary tuberculosis in distinction from a chronic bronchitis, which latter condition shows considerable mucin. Roger and Levy-Valensi,2 Geeraerd,3 and Raymond4 show that albumin is present in 1 Deutsch. Arch. f. klin. Med., Bd. 75, 1903, S. 347. 2Presse Med., Tm. 18, 1910, p. 289; Ibid., Tm. 19, 1911, p. 409. 3 Jour. med. Bruxelles. Tm. 15, 1910, p. 505. 4 Presse Med., Tm. 19, 1911, p. 675. 6 DIAGNOSTIC METHODS. the sputum whenever there is active exudation into the alveoli as well as into and under the pleura and believe that any appreciable amount of albumin is a reliable means of differentiating tuberculosis from bronchitis, pulmonary emphysema and conditions such as syphilis, gangrene and cancer. Goodman,1 Fishberg and Felberbaum,2 Fishberg3 and Works4 however, have shown that albumin may not be present in the sputum of true pulmonary tuberculosis, but is frequently seen in benign affections and when present is due, in the major- ity of cases, to occult blood. These methods should, therefore, be used only in conjunction with other and, perhaps, more reliable diagnostic tests and must be given merely a presumptive interpretation.5 Besides these albuminous principles, the sputum, especially in cases of gangrene and putrid bronchitis, contains a ferment very much resembling in its action the trypsin of the pancreatic juice (Stolnikow).6 This ferment seems to indicate a highly destructive process in the lung tissue. Other chemical substances, such as glycogen and fatty acids, are frequently found in the sputum, but may be passed with mere mention. The so-called myelin granules or globules, which appear in the alveolar cells of the sputum, consist largely of fatty principles, such as protagon, lecithin and cholesterin. III. Macroscopic Examination. While much that is included in the previous section would come properly under the head of the macroscopic examination of the sputum, the writer has reference more, in this connection, to the appearance of macroscopic elements as distinguished from those which are purely microscopic. (i). Cheesy Masses. Frequently one finds in the sputum small cheese-like particles which vary in size from that of a pin-point to that of a pea, the large majority being about the limit of ordinary vision. These cheesy masses are fragments of nec- rotic tissue and appear in the larger form in cases of abscess or gangrene of the lung, while in tuberculosis they are always small unless the cavity, from which the material is derived, is markedly necrotic. The color of these masses varies from a yellow to a black. Those fragments which come from an abscess are of yellow color due to the presence of much pus, the darker ones contain decompo- sition products of hemoglobin, while many of them are tinged a deep black with coal pigment. If the sputum be squeezed between two glass plates these cheesy particles or fragments can sometimes be more distinctly seen. They are pre- ent in largest numbers in the so-called "nummular" sputum from a tubercu- lous cavity. This nummular sputum derives its name from the fact that such material floats upon the surface in a coin-like mass, and then sinks to the bottom. 1 Arch, of int. Med., vol. 8, 1911, p. 163. 2 Med. Record, vol. 80, 1911, p. 870. 3 Arch, of Diag., vol. 5, 1912, p. 220. 4 Jour. A. M. A., vol. 59, 1912, p. 1537. 6 See Scott, Jour. A. M. A., vol. 60, 1913, p. 440; also Acs-Nagy, Wien. klin. Wochs Bd. 25, 1912, S. 1904. 6 See Eiselt, Zeitsch. f. klin. Med., Bd. 75, 1912, S. 71. THE SPUTUM. 7 (2). Dittrich's Plugs. These masses are similar to the small caseous particles above mentioned and are frequently expectorated by perfectly normal individuals. The true plugs are distinct casts of the bronchi or bronchioles and vary in size from pin-point to that of a bean, the majority being about the size of a mustard seed. The smaller ones are opaque and yellowish-white in color, while the larger ones have a distinct gray tinge. They are usually expectorated free from pus or mucus, so that they frequently give rise to anxiety, especially in those of a hypochondriac tendency. These plugs have a distinctly disagreeable odor, which is more evident if they are crushed on the glass plates. Microscopic examination of these masses shows large clumps of bacteria, fatty acid crystals, free fat globules, and cellular detritus. Occasionally a few leucocytes are found, but these are rare, while pigment granules, either of hematogenous or extraneous origin, are sometimes observed. While such plugs are especially numerous in cases of putrid bronchitis and bronchiectasis, they are frequently found in the crypts of the normal tonsil as well as in cases of follicular tonsillitis or of ozena. Fig. i.-Curschmann's spirals {Tyson after Curschmann). I, Natural size II and III, enlarged; a, a, central thread. (3). Curschmann's Spirals. These structures are found in the sputum in practically every case of true bronchial asthma and have been reported in acute bronchitis, croupous pneumonia, chronic pulmonary tuberculosis, and occasionally in chronic bronchitis. They are not present in every paroxysm of asthma, but are more frequently found just at the end of the paroxysm and are absent when the spu- tum becomes mucopurulent. They seem to be derived through true exuda- tion from the bronchioles, as Curschmann says, a bronchiolitis exudativa. These structures are recognizable, to a certain extent, by the naked eye, but for their absolute identification a microscopic examination is essential. They are composed of a spirally-twisted network of very delicate fibrils, in the meshes 8 DIAGNOSTIC METHODS. of which are numerous epithelial cells and eosinophile leucocytes. Along with these cellular bodies one frequently finds large numbers of diamond-shaped crystals, known as the Charcot-Leyden crystals, which will be discussed later. This spirally-twisted mass (the mantle) seems to be wound around a central light thread. While this is the structure of the complete spiral, we frequently find variations pointing apparently to two distinct forms. The first is the spirally-twisted strand of mucus with the enclosures above mentioned. The second is the tight spiral mass of mucus wound around a central fibre. This central fibre is very refractive and is homogeneous in structure, varying in size from 1/2 to 18 microns in diameter. The length of these spirals is from 1 to 2 cm. and their breadth about 1 mm. (4). Fibrinous Casts. By the term fibrinous cast we refer more directly to true bronchial casts, which are composed of fibrin. These are observed in pneumonia, in which case they are derived from the smaller bronchioles and are brownish or reddish in color and contain many red and white blood-cells. These smaller casts vary from 1/2 to 3 cm. in length. In the chronic fibrinous bronchitis we find the so- called arborescent casts which are usually whitish in color and contain many epithelial cells. These casts vary in size from 1 to 15 cm. in length by several mm. in thickness. We may have an acute form of fibrinous bronchitis accom- panying various febrile conditions, so that similar casts may appear in almost any of the infectious diseases. These larger casts are fairly .firm, usually have a lumen, and branch dichotomously five to ten times. Microscopic examina- tion shows them to consist of large numbers of longitudinal fibres containing blood and epithelial cells in their meshes. They may be stained with the Wei- gert fibrin stain in a very beautiful way. Such staining methods show that not all of the material in such casts is fibrin, so that it may be necessary to rename these structures as simple bronchial casts rather than as fibrinous forms. Besides these fibrinous casts, one occasionally may find distinct bronchial casts composed of the mycelium of various fungi. Thus mycelial threads of the aspergillus have been reported by Osler, Devillers and Renon. (q). Concretions. This name is applied to anything, expectorated with the sputum, which has the appearance or consistency of a stone. These concretions are formed in dilated portions of the bronchi or in cavities by the calcification of the stag- nated contents. Although concretions consisting of cartilaginous or osseous material are frequently found postmortem, yet there seems to be little mention in the literature of any such formations being expectorated. (d). Bronchioliths. These concretions are formed by the deposition of calcium salts in the stagnated contents of the bronchus or of a cavity. They may be derived from the smaller or larger bronchi, but rarely form arborescent shapes, being usually THE SPUTUM. 9 irregular and varying in size from a pin-head to that of a walnut. They are usually single, but may be multiple. They vary in consistency from that of chalk to a stone, and may be expectorated in small numbers over long periods of time. (6). Pneumoliths. These lung stones are in the majority of cases of tuberculous origin. They usually arise by the calcification of caseous areas which later ulcerate into a bronchus and are expectorated unless too large. They may, also, arise by the calcification of a pulmonary cavity or of a bronchial lymph-gland. These lung stones consist of the carbonates, phosphates, and sulphates of calcium and magnesium, in some one or more salts predominating, while in others still different combinations may exist. These pneumoliths either have a chalky or a calcareous consistency and vary in size from that of a pin-head to that of a tennis-ball. These lung stones are usually expectorated en masse or in the form of smaller portions of a larger stone. In some cases these smaller stones may reach the number of 500 (Portal). (6). Echinococcus Membranes. Rarely one may find in the sputum fragments of the walls of echinococcus cysts or their contents. These may come from a perforating cyst of the liver, kidney, or lung. The presence of the laminated membrane and of the parasitic scolices and hooks makes it possible to arrive at an absolute diagnosis of the origin of such material. The membrane is thick, tough, and of a porcelain- like color and may show a laminated or fibrillated structure. The parasite is discussed in a later section. (7). Foreign Bodies. Examination of the sputum may reveal the presence of material which has lodged in the air-passages and been retained for long periods of time. Such bodies are coins, fish-bones, and cherry-stones. Heyfelder reports a case of the expectoration of a wooden cigar-holder 111 / 2 years after its disappearance. IV. Microscopic Examination. The microscopic examination of the sputum is almost the only one to which it is subjected at the present day. This is to be regretted, as much may be learned from a careful macroscopic examination. However, a microscopic examination reveals evidence which points to an absolute diagnosis more fre- quently than does the macroscopic examination. Before making the micro- scopic examination, it is wise to place the sputum in a flat-bottomed dish (Petri dish) which has half of its base blackened, so that the more suspicious particles may be selected for microscopic investigation. Some experience in this work is necessary, as one is frequently called upon to recognize material which is purely extraneous and has absolutely nothing to do with the sputum. Such material practically always comes from the buccal cavity and consists of frag- 10 DIAGNOSTIC METHODS. ments of various food-stuffs, such as bread, fruit pulp, meat fibers, vegetable tissue, and portions of tobacco leaf. Naturally, such material should not mis- lead one, but it very frequently does. The fragments of meat tissue contain elastic tissue fibers and may lead one to state that such material is present in the sputum, thus giving expression to the possibility of a diagnosis of incipient tuberculosis. A portion of the sputum selected for examination is taken up with a plati- num loop and spread in a thin layer upon a glass slide. It is then dried by Fig. 2.-Objects found in the sputum (Landois). i, Detritus and dust-particles; 2, pig- mented alveolar epithelium; 3, fatty degenerated and partially pigmented alveolar epithelium; 4, alveolar epithelium showing myelin-degeneration; 5, free myelin forms; 6, 7, desquamated ciliated epithelium, partly changed and deprived of its cilia; 8, squamous epithelium from the mouth; 9, leucocytes; 10, elastic fibers; 11, fibrinous cast of a small bronchus; 12, lep- tothrix buccalis, together with cocci, bacilli, and spirochete; a, fatty acid crystals and free fatty granules; b, hematoidin; c, Charcot's crystals; d, cholesterin. passing the slide several times through the flame, care being taken not to burn the specimen. The smear is allowed to cool and is then stained either with Loffler's methylene blue for general purposes or with special stains for the various specific organisms. (o). Pus-cells (Leucocytes). There is practically no specimen of sputum which does not contain leuco- cytes in larger or in smaller numbers. The true pus-cell is the polymorpho- nuclear neutrophile and appears in the sputum frequently filled with fat globules or pigment granules. In cases of asthma, the eosinophile cells are very abun- dant, while basophile cells may occasionally obtain. Although these cells THE SPUTUM. 11 are so frequent in asthma, a diagnosis should not rest upon such evidence alone, as there seems to be a form of bronchitis, possibly of the tubercular variety, which has been named "eosinophilic bronchitis" from the large number of such cells observed. In pulmonary tuberculosis, which is not associated with a mixed infection, it is very common to find the small mononuclear leucocyte (lym- phocyte) in place of the polynuclear type. This finding is of such frequent occurrence that the writer is often led to search many slides for tubercle bacilli in case the organisms have not been found in the earlier specimens examined. Basophilic cells are not infrequently observed in the sputum. Pottenger1 reports the presence of macro-phagocytes in a case of chronic hemop- tysis. The cytology of the sputum shows little of diagnostic value. The thin smears of sputum may be treated like blood-smears and stained with the same stains when one desires to study the cellular types present. (b). Red Blood-cells. These cells are frequently found in the sputum and may have much signifi- cance. They may occasionally arise from contamination with nasal or buccal discharges, but the true bronchial or pulmonary sputum usually shows them only in cases of hemorrhage or exudation. The rusty sputum of pneumonia contains large numbers of such cells and the hemorrhage from a tubercular cavity may be very extensive. These cells are occasionally well-preserved, but at times they are difficultly recognizable. They may be very much distorted in shape, so that their color and staining properties must be relied upon for differentiation. Various types of epithelial cells are found in the sputum. Pavement epithelium may come from the mouth, pharynx, and upper larynx. Cylindri- cal epithelium may be derived from the nose or the bronchi. These cylindrical cells may be ciliated, but rarely does one find these ciliated forms except in cases of asthma and bronchitis. Alveolar epithelial cells are present in normal sputa as they are constantly desquamated from all of the epithelial surfaces. These cells assume a large variety of forms and frequently show various types of de- generation. They are very numerous in bronchitis and in general inflammatory conditions of the lungs, but may occur in almost any condition, associated with irritation along the respiratory tract. These cells contain large numbers of granules which are probably referable to coal pigment. Such cells give the sputum a grayish or green color. An abundance of such cells in the sputum was designated in earlier times as "phthisis melanotica." Occasionally these alveolar cells are filled with fat globules. In other cases one finds the so-called myelin globules, which are irregular in shape, often showing concentric lines, with very little retractility, and of a dull greenish or blue appearance. The cell may be invisible and only the myelin appear as a large irregular mass. Much doubt exists as to the origin of these globules, but it is more probable that they represent simply the fatty products of degenerated protoplasm. (c). Epithelial Cells. 1 Jour. A. M. A., vol. 59, 1912, p. 1794. 12 DIAGNOSTIC METHODS. These cells are especially frequent in the normal morning sputum, in acute or chronic influenza, and in the so-called desquamative catarrhal pneumonia, when they appear as small lumps resembling boiled sago. Occasionally free myelin globules are found in the sputum. These globules stain poorly with the aniline dyes, are stained yellow with iodin, but do not stain black with osmic acid or red with Sudan-Ill. Frequently these alveolar cells contain pigment granules, derivatives of hemoglobin. This material is in the form of amorphous granules of a brown color and seems to be identical with hemosiderin, but may become iron-free, when it more closely resembles hematoidin. Cells containing such granules are especially numerous in chronic heart disease and are styled, therefore, ^heart- disease cells." They occur, however, in any condition in which red blood-cells escape into the alveoli and are found, therefore, in pneumonia, infarction of the lung, and hemorrhagic pulmonary tuberculosis. (d). Elastic Tissue. The presence of elastic tissue in the sputum is indicative of destruction of the lung tissue and in many cases is found in the sputum before one can detect tubercle bacilli. It has, therefore, some importance in the early diagnosis of tuberculosis. When this elastic tissue is grouped in masses it is usually visible to the naked eye. However, one more frequently relies upon microscopic ex- amination for its detection. The method of Clark, as applied to the detection of elastic tissue, is usually the best one to follow. The sputum is placed upon a glass plate about 14 inches square and pressed out by a smaller one, about 6 inches square, into a thin layer. The plates are then placed upon a dark back- ground and examined with a hand lens. Instead of these glass plates one may use Petri dishes. The elastic tissue fibers appear either in the form of distinct strands some- times grouped in an elongated network, in that of the alveolar type, in which the fibers preserve the outline of the alveoli and are long and branching, or in that from the arteries in which we may have a distinct sheet-like arrangement. These fibers are characterized by their undulating outline, their curling ends, their sharp edges and uniform diameter, their frequent branching, and their intense refractility. These characteristics are brought out both with the low and with the high power of the microscope. The fibrous tissue fibers are dif- ferentiated from elastic fibers by the fact that the former are present in bundles of fine wavy lines without the coarse black refractive appearance of elastic tissue. Chains of bacteria, especially the leptothrix forms, frequently interlace in such a way as to simulate the alveolar structure of elastic tissue. These chains differ, however, from elastic tissue in their refractility, in the absence of the wavy outline, and in their denser arrangement. The elastic tissue fibers derived from food substances have the same characteristics as the lung elastic fiber, with the exception that they are not arranged in the alveolar form and may at times be coarser and more irregular in outline than the pulmonary elastic THE SPUTUM. 13 tissue. Vegetable cells and fibers, as well as fatty acid crystals, which may occur in the sputum, should not mislead one into assuming the presence of elastic tissue. Should one wish to stain elastic tissue, he may use the orcein stain of Unna- Tanzer. This stain consists of one gram of orcein dissolved in a mixture of 80 c.c. of 95 per cent, alcohol and 35 c.c. of distilled water, 40 drops of strong hydrochloric acid being added after solution is complete. In using this stain the elastic fibers are treated with a few c.c. of the dye and then warmed for five minutes, after which the preparation is decolorized with acid alcohol. The elastic tissue fibers will be stained a brownish violet by this process. If the elastic tissue is very small in amount we usually resort to chemical means for the isolation of this material. Ten c.c. of sputum are mixed with an equal volume of 10 per cent, sodium hydrate solution and the mixture is boiled until it becomes homogeneous. Four volumes of water are then added, the entire mass well mixed, and either allowed to stand or to settle by centrifugation. In this way the constituents of the sputum are destroyed with the exception of the elastic tissue which has, however, become swollen and paler and does not have its characteristic appearance. About 90 per cent, of cases showing elastic tissue in the sputum are of tuberculous origin, according to Dettweiler and Setzer. As the healing process begins and proceeds this elastic tissue gradually diminishes in amount, so that a constant presence or an increase in the amount indicates a progressive condi- tion. It is seen in abscess of the lungs, in bronchiectasis, in pulmonary infarct, and occasionally in pneumonia. It is found in cases of gangrene of the lungs, although there are many statements that it is digested by the trypsin-like fer- ment so common in such conditions. Osler states that he has never seen a case of gangrene of the lung in which elastic tissue fibre could not be found. Crystals are never found in the freshly formed sputum, being indicative of stagnation of the material within the body or of decomposition after being expectorated. (e). Crystals. These crystals occur most frequently in the sputum of gangrene, putrid bronchitis and of chronic tuberculosis. They occur as distinct needles, either singly or in groups, may be short and relatively thick with pointed ends, or they may be long and very closely resemble elastic fiber. Pressure upon the cover- glass will produce varicosities in these crystals which do not appear in the case of elastic tissue. They are soluble in alkalies and in ether and melt into fat globules if the slide be warmed. (1). Fatty Acid Crystals. Crystals of cholesterin are found in the sputum of chronic lung abscesses, empyema and chronic tuberculosis. They are not as frequent as are the fatty (2). Cholesterin. 14 DIAGNOSTIC METHODS. acid crystals, but are usually found associated with them when they are present. These crystals have a distinct rhomboid form with notched angles. (3). Hematoidin. Crystals of hematoidin occur very rarely in sputum, and then only when extravasation has taken place into the alveoli. They are found very rarely after direct hemorrhage unless the extravasated blood remains for some time in the alveoli. These crystals occur especially in abscess of the lungs, empyema, or perforating liver abscess. They are rhomboid or needle-shaped crystals, ruby-red in color, and may show small curved filaments projecting from the angles of the larger forms. (4). Leucin and Tyrosin. These substances are formed only by the decomposition of protein material and are found, therefore, in the putrid sputum of an empyema, or from a per- forating liver abscess, and in the very early discharges of true lung abscess. The tyrosin is found in the form of long refractive needles, frequently arranged in bundles, while the leucin appears as distinctly spherical masses with con- centric striations and radiating lines. (5). Magnesium-ammonium Phosphate. These crystals appear in the sputum under the same condition as do the preceding tyrosin and leucin crystals. They are usually the typical coffin-lid crystals so common in decomposed urine, but may at times assume an irregular structure. These crystals occur in the sputum in conditions associated with decom- position and appear either in the typical octahedral crystal with a cross con- necting the corners or as the more atypical dumb-bell shaped crystal. (6). Calcium Oxalate. These crystals are apparently derived from the eosinophile cells as they are more frequently present only in conditions in which the eosinophiles are very numerous, hence the term "leucocyte crystals." They increase in the sputum either from stagnation within the system or after being expectorated. They are associated in asthmatic attacks with the spirals of Curschmann, being frequently included in the meshes of the spiral. These crystals form straight, pointed, colorless, hexagonal, double pyra- mids, resembling a very much elongated diamond. They have sharp elon- gated points with clear-cut edges, are very brittle, are colorless, show little retrac- tility, and vary greatly in size. They may occur singly or in groups, forming either clusters or distinct Greek-cross types. They have been supposed to be identical with the spermin crystals of Bbttcher, but the hexagonal type of the crystal as well as the facts that they do not show marked double refraction by polarized light and have but a single optical axis should serve to differentiate them (7). Charcot-Leyden Crystals. THE SPUTUM. 15 from the spermin crystals. They are colored yellow with Florence's reagent and may be stained with the polychrome and other blood dyes. (See Semen.) (/). Bacteria. The bacteria found in the sputum are very numerous and under normal conditions are purely saprophytic. We may even at times find many truly pathogenic organisms in the sputum which are of no clinical significance, although one may be led into making a diagnosis without sufficient clinical evidence. These saprophytes may occur in the fresh sputum or develop therein after the specimen has stood for some time. In either case their chief effect is to bring about gradual decomposition of the sputum. The various chromogenic bacteria are of particular interest as their development along the respiratory tract may so change the color of the sputum that the examiner may be led astray. Among the ordinary saprophytes found in the sputum we find representa- tives of the streptothrix1 and the leptothrix groups. Flexner and Warthin and Olney have reported the presence of a streptothrix, the streptothrix ep- pingeri, in the sputum of cases showing the clinical symptoms of pulmonary tuberculosis. These organisms are about four times as thick as the tubercle bacillus; when stained they are resistant to decolorization by acids, but are slowly decolorized by strong alcohol. Stained specimens are easily made with the use of methylene-blue dyes or of Gram's method. The leptothrix group is particularly abundant in the mouth and is found in large numbers in the lungs in cases of pulmonary gangrene. Yeast fungi occur in the sputum at times, but rarely in the fresh specimens. They are oval or elliptical cells and are very refractive, sometimes resembling very closely fat droplets. Their appearance may vary from that of a simple oval cell without distinct limiting membrane to those with definite membrane and vacuoles. These cells are especially characterized by their tendency to throw out projections or buds at various points of their periphery. They vary in size from i to 40 microns in diameter. While these yeast fungi are usually extraneous, cases are reported (Busse) in which pathogenic yeasts have been found in anomalous pulmonary conditions. These organisms stain with the ordinary aniline dyes and appear in some instances to be acid-fast (resisting decolorization by acid). Various types of molds are found in the sputum. Some of these appear to be distinctly pathogenic, while the maj'ority are merely saprophytic. These molds are found in the true sputum only in cases associated with destructive processes of the lungs. It has been doubted that they could cause primary in- fections, but there are at present many reports of cases showing that some of them at least may be distinctly answerable for primary infections. Among these pathogenic molds we find certain types of the Mucor, of (1). Saprophytes. 1 See Claypole, Jour, of Exp. Med., vol. 17, 1912, p. 99. 16 DIAGNOSTIC METHODS. the 130 varieties of which six are known to be distinctly pathogenic. Besides this type we find, as the most important pathogenic mold of the sputum the aspergillus fumigatus, 16 cases of pulmonary affection (pneumonomycosis aspergillina) having been traced by Sticker to this fungus. Other types of the Fig. 3.-Aspergillus fumigatus. aspergillus are the flavus, niger, and the sub-fuscus. The penicillium glaucum and the Oidium albicans are occasionally found in the sputa. Among the bacteria wh'ch may be purely saprophytic we may find almost any of the pus-forming organisms. More frequently, however, when these or- ganisms are present in excessive numbers a contamination or a direct pathogenic Fig. 4.-Micrococcus catarrhalis. (From Emery's " Clinical Bacteriology. ") influence should be suspected. A form known as the micrococcus tetragenus occurs both as a pathogenic and a harmless organism. It consists, as its name implies, of four cocci arranged in a square within a mucous capsule. It stains with the ordinary dyes and is Gram-positive. This organism is found, in its THE SPUTUM. 17 pathogenic state, in bronchitis, tubercular cavities, and hemorrhagic infarctions. The harmless form differs from the pathogenic type in the fact that it cannot be cultivated. The sarcinae are rarely found in the sputum. These organisms are somewhat smaller than those occurring in the stomach and are probably purely saprophytic in the sputum. They are found, however, in cases of putrid bron- Fig. 5.-Budding forms of blastomycetes found in sputum. (From photograph by W. A. Pusey.) chitis, especially when this occurs in emphysematous lungs, in gangrene, tuber- culosis, and pneumonia. The micrococcus catarrhalis is found frequently in the sputum as a sapro- phyte, but may become pathogenic especially in some epidemics of la grippe.1 It is larger than the ordinary staphylococcus, while its arrangement in lateral pairs and its Gram-negative properties obscure its differentiation from the gonococcus and meningococcus unless cultural methods be employed. 1 Hastings and Niles. Jour, of Exp. Med., vol. 13, 1911, p. 638. 18 DIAGNOSTIC METHODS. In cases of general systemic blastomycosis, Eisendrath and Ormsby have found the blastomycetes in the sputum. They recommend the examination of the unstained specimens after the addition of 10 per cent, sodium hydrate solu- tion. In such preparations examined with a high-power dry lens, the typical re- fractile blastomycetes are observed. (See Parasitic Diseases.) (2). Pathogenic Types. (a). Tubercle Bacillus. This organism is the most important pathogenic type found in the sputum. Its detection is usually easy and should be attempted in all suspicious cases, as an early diagnosis may frequently save the life of the patient. In the days before the organism was recognizable, physicians based their diagnosis of con- sumption upon the macroscopic appearance of the sputum. While such ex- aminations frequently lead to a presumptive diagnosis of tuberculosis, nothing can settle the question except the microscopic examination of the sputum. This statement needs some modification in several ways. In the first place, specimens of tubercular sputum may not show the presence of the bacilli so that several examinations of sputum, collected at different periods, must be made. In the second place, the sputum may be examined in the very early stage of the disease and no tubercle bacilli be found, but in such cases the presence of elastic tissue fiber would be very significant of tubercular changes. In these days we have, fortunately, recourse to other diagnostic measures in case tubercle bacilli cannot be found in the sputum. I have reference here to the use of tuberculin, either introduced in the form of an injection as done by Koch, dropped into the eye as advocated by Calmette, as advised by Pirquet applied after the manner of vaccinatipn, or used as an inunction as suggested by Moro. This test is usually certain and can be relied upon to settle the diagnosis. In examining the sputum microscopically the fine cheesy particles pre- viously mentioned are selected and smears of such material made upon glass slides. It is always advisable to make at least five such smears to insure definite results. It has frequently been found that many suspicious-looking sputa show no tubercle bacilli when subjected to the ordinary methods of examination. In order, therefore, to increase the possibility of positive results, the sputum should be rendered homogeneous and fluid. The best method of accomplishing this is, in the writer's opinion, Loeffler's1 modification of the antiformin process of Uhlenhuth. The technic is as follows: 5, 10, or more c.c. of the sputum are placed in a flask and mixed with an equal quantity of a 50 per cent, solution of antiformin (a 10 per cent, solution of sodium hypo- chlorite containing 5 to 10 per cent, of sodium hydrate) and boiled for a period not exceeding 15 minutes. Solution occurs associated with considerable foam- ing and browning of the mixture. For every 10 c.c. of this solution are now added 1.5 c.c. of a mixture of 1 part of chloroform and 9 parts of alcohol. After thoroughly shaking to produce a fine emulsion, portions of the fluid are placed in sedimenting tubes, the tubes are corked, and centrifuged for fifteen 1 Deutsch, med. Wchs., Bd. 36, 1910, S. 1987. THE SPUTUM. 19 minutes. The heavier elements collect in a film just above the chloroform, which film holds the tubercle bacilli owing to the marked affinity of chloro- form for the fatty and waxy material in these organisms.1 The supernatant liquid is poured off and the film above mentioned is removed and placed upon a glass slide, the excess of fluid being taken up with filter-paper. As a fixa- tive a drop of egg-albumin, preserved with 1/2 per cent, carbolic acid, is added and a thin spread made by means of a second slide. This smear is allowed to dry and is then stained by one of following methods. This enrichment process of Loeffler furnishes preparations which often show a remarkable in- crease in numbers of tubercle bacilli as compared with those found by the usual smear methods. A further advantage of this method is that practically all organisms, with the exception of those of the acid-fast type to which the tubercle bacillus belongs, are destroyed. If the mixture be allowed to stand for a few hours instead of being boiled, one may obtain material for pure-culture or for inoculation purposes, as the secondary invaders are eliminated. Naturally, in the study of the mixed infections in tuberculosis one should examine prepara- tions made in the usual way. Instead of following this method of Loeffler, one may use Spengler's process of digestion with soda and pancreatin or, preferably, Ellermann and Erlandsen's technic of double digestion with 0.6 per cent. Na2CO3 solution at 370 C. for 24 hours followed by boiling with 0.25 per cent. NaOH for a few minutes. These methods are, however, inferior to that given above. Staining Characteristics. The methods of staining the tubercle bacillus depend upon the property, possessed by this organism, of taking up the aniline dyes with greht difficulty, but, when once stained, of becoming just as resistant against decolorization. Ziehl-Neelsen Method. The smears are made upon glass slides and are fixed by passing several times through a flame. The smear is covered with carbol-fuchsin solution2 (a mixture of 90 parts of 5 per cent, carbolic acid water and 10 parts of a concen- trated alcoholic solution of fuchsin), and is then heated over a flame for one to three minutes in such a way that the staining solution steams, but does not boil. If the staining solution is heated too strongly the smear decolorizes less readily so that it is very good practice never to boil the staining solution. Where several slides are to be examined, the writer has found the copper plate of Ehrlich very useful. The slides are laid upon the plate, are covered with the stain, and allowed to heat for ten minutes. The usual technic is, however, to heat a single slide at one time, more of the stain being added as the first evaporates. Some workers find that the immersion of the smear in cold carbol-fuchsin for 24 hours gives somewhat clearer pictures, but the time is too long for the ordi- nary laboratory diagnosis. Having thus stained the smear with the carbol- fuchsin solution, it is then decolorized. The tubercle bacillus is not only acid- 1 See Panzer, Zeitsch. f. physiol. Chem., Bd. 78, 1912, S. 414. 2 See Verhoeff, Jour. A. M. A., vol. 58, 1912, p. 1355. 20 DIAGNOSTIC METHODS. fast, but also alcohol-fast, so that we use decolorizing agents containing both acid and alcohol. There are many acid-fast organisms known, such as the bacillus leprae, the smegma bacillus, the timothy bacillus, the butter bacillus and many saprophytic bacilli found in water, soil, and manure. Few or none of these organisms are absolutely both alcohol and acid-fast, so that the use of the combined decolorizer will usually differentiate the tubercle bacillus. Many decolorizing agents have been advised, but the writer finds the use of a 10 per cent, solution of sulphuric acid in 95 per cent, alcohol very reliable. This decolorizing agent does not burn the specimen nor does it prevent the morpho- logical characteristics from appearing in a clear-cut way. Some workers ad- vise the use of 2 per cent, hydrochloric acid in 80 per cent, alcohol., while others use 25 per cent, nitric acid, followed by alcohol. The technic of decolori- zation is as follows: Wash the smear, which has been stained with carbol- fuchsin, in water and flood the specimen with the decolorizing solution until only the faintest pink color is seen in the smear. It frequently happens that the thicker portions of the smear resist this decolorizing so that it may be neces- sary either to make a new smear or to examine merely the portions which have been decolorized. After decolorization is complete the specimen is washed in water and counter-stained with Loffler's methylene blue (saturated alcoholic solution of methylene blue 30 c.c., 100 c.c. of a 1 to 10,000 aqueous potassium hydrate solution) for a few seconds, after which the specimen is washed with water, dried between filter-paper, and examined with the oil immersion lens. In such preparations the tubercle bacilli are seen as bright red rods, some- what bent, sometimes much curved and occasionally showing distinct branching forms. In a few preparations one may find the curves of the bacillus so marked that a very close resemblance to the spirillar forms obtains. These organisms occasionally show a distinct beading, giving the appearance of bright red cocci. The size of these organisms varies from 1 1/2 to 3 1/2 microns and about 2/10 microns in width. They may be single or arranged in clumps, sometimes in the form of distinct crosses, sometimes parallel, and very frequently forming acute angles by the joining of two bacilli. Gabbet's Method. This method is much more simple than the preceding, but is not as reliable. By this method the decolorization and counter-staining are carried out in one operation. The smears are prepared as previously described and stained with the carbol-fuchsin solution. The excess of the staining solution is drained off without washing and is replaced by Gabbet's methylene blue solution (methy- lene blue 2 grams, sulphuric acid 25 c.c., water 75 c.c.). This solution is allowed to act for one to three minutes and is then washed off with water and the specimen dried and examined. The tubercle bacilli will appear as bright red rods as previously described, while the other organisms as well as the various cellular types will be stained blue. This method is not as reliable as is the former, owing to the fact that the alcohol-fast bacilli resist decolorization and may confuse one in making the PLATE I. Tubercle Bacilli in Sputum. Ziehl-Neelsen Method. THE SPUTUM. 21 diagnosis. Moreover, the use of the strong acid may cause decolorization of some of the tubercle bacilli and will, therefore, give rise to wrong ideas. These methods are the ones usually followed i« routine laboratory work for the de- tection of the tubercle bacillus. The combined decolorization with alcohol and acid being the only reliable method, one should use the Gabbet's method only for obtaining somewhat clearer pictures. Pappenheim's Method. The technic of this method is as follows: The preliminary staining is carried out with carbol-fuchsin solution as previously outlined. The speci- men is then drained and covered with the decolorizing solution, which is made by dissolving one gram of rosolic acid in 100 c.c. of absolute alcohol, saturating the mixture with methylene blue and adding 20 parts of glycerin. This solution is drained off slowly and the process repeated several times. The slide is then washed in water, dried between blotting-paper and examined with the immersion lens. The tubercle bacilli are stained red and the other organisms blue. Much's Method. As is well known, certain specimens of undoubted tubercular sputum as well as tissues frequently do not show tubercle bacilli when any of the above methods of staining are used, owing to the facts that neither is every acid- alcohol-fast bacillus the tubercle bacillus nor is every tubercle bacillus abso- lutely acid-fast. It is to Spengler1 and, more especially, to Much2 that we are indebted for a method which will demonstrate these non-acid-fast types of tubercle bacilli, which are distinctly granular, frequently appearing as mere granules rather than as true bacilli. These granules, under certain unknown conditions, change into true bacillary types and vice versa. Much's method is a modified Gram method, the tubercle bacilli being Gram-positive.3 Prepare smears as described above. Cover this smear with a carbol-methyl violet solution (10 c.c. of a saturated alcoholic solution of Gruebler's methyl violet B. N. mixed with 90 c.c. of 2 per cent, aqueous carbolic acid solution) and heat to bciling several times. Wash off stain with water and cover smear with Lugol's solution( iodine 1 gram, potassium iodide 2 grams, water 300 c.c.) for 5 minutes. Wash with water and treat with 5 per cent, nitric acid for one minute and follow this with 3 per cent, hydrochloric acid for 10 seconds. Without washing place the slide in a mixture of equal parts of acetone and absolute alcohol until the smear is colorless. Wash with distilled water and counter-stain with 1 per cent, aqueous solution of safranin for afew seconds. Wash in water, dry thoroughly, and examine with the immersion lens. The tubercle bacilli and the granular forms appear bluish while the other organisms are red. Value of Examinations. Brown4 in a recent article has summed up the value of the sputum examina- tion for tubercle bacilli. He gives as his reasons for believing that one should 1 Deutsch, med. Wchs., Bd. 31, 1003, S. 1228 and 1333; Ibid., Bd. 33, 1007, S. 337. 2 Beitr. z. Klin. d. Tuberk., Bd. 8, 1907, S. 85 and 357. 3 See Benians, Jour, of Path, and Bact., vol. 17, 1912, p. 199. 4Montreal Med. Jour., vol. 30, 1901, p. 769; Jour. A. M. A., vol. 40, 1903, p. 514. 22 DIAGNOSTIC METHODS. be guarded in forming an opinion of the prognosis of certain cases the follow- ing points: (i) Many of the tubercle bacilli may not be stained at all. (2) Old foci may give off very few and young foci no bacilli at all. (3) By the occlu- sion of a bronchus the contents of a focus may be shut off entirely for a time and thus the expelled sputum may contain a large number of tubercle bacilli. (4) The organisms may be present one day and not again for months. (5) The organisms may be abundant in one part of a specimen and none be found in others. (6) Some patients with fatal tuberculosis (caseous pneumonia or acute miliary tuberculosis) may have no bacilli in the sputum, while in other cases the organisms are present even before physical signs obtain. (7) In severe cases with bronchitis the secretion of the bronchus will dilute the sputum and give the appearance of a reduction in the number of organisms. While the number of bacilli in the sputum may thus vary, it is usually in direct ratio to the severity of the disease, although for the reasons above men- tioned too much reliance should not be placed upon the number of organisms found. Brown recommends the use of a somewhat modified Gaffky's table in judging of the prognosis in any particular case. The cases are classified as follows, being designated by the Roman numerals: I. Only one to four bacilli in whole preparation. II. Only one on an average in many fields. III. Only one on an average in each field. IV. Two to three on an average in each field. V. Four to six on an average in each field. VI. Seven to twelve on an average in each field. VII. Thirteen to twenty-five on an average in each field. VIII. About fifty on an average in each field. IX. About 100 on an average in each field. There has been some attempt to base a prognosis upon the form and group- ing of the tubercle bacilli, the short rods indicating a rapid growth while the longer form shows a slower development. The continued expectoration of large numbers of bacilli would indicate a cavity, while the sudden increase in numbers associated with an increase in the cellular elements would point to lung disintegration. A steady decrease over a long period of time would indi- cate improvement, but it must be remembered that occlusion of the bronchus may shut off large numbers of these organisms from the sputum. It should be stated as a working rule that the finding of a single or a very few organisms in the sputum should be looked upon with suspicion, but that an absolute diag- nosis should be made only after repeated examination has shown the presence of the tubercle bacilli. The worker wrill find in the study of preparations stained as above that artefacts are very common so that a hurried examination is never warranted. The sputum in tubercular cases rarely shows tubercle bacilli in pure cul- ture. One usually finds large numbers of streptococci, staphylococci, micro- THE SPUTUM. 23 cocci catarrhalis, and frequently influenza bacilli and pneumococci. Pus cells may be few or many, while the large number of leucocytes are more frequently of the mononuclear type. Blood-cells may or may not be present, while elastic tissue fiber is very frequent, appearing in many cases before tubercle bacilli can be demonstrated. It should be remembered that mixed infections, which are the rule in tubercular conditions, may be so marked that few if any tubercle bacilli appear in the sputum. This statement is especially true when the sputum has been allowed to stand for some time before examination. The writer has frequently found sputa showing fairly large numbers of tubercle ba- cilli becoming practically negative for these organisms if the sputum be allowed to stand exposed to the air for 48 hours. The explanation is that the other organisms so far outgrow the tubercle bacillus that they prevent any further development of this latter organism and bring about such degeneration that the staining qualities of the tubercle bacillus are markedly affected. This fact has been taken advantage of in the clinical use of injections of pus organisms into tubercular joints. (b). Lepra Bacillus. The bacillus leprae, first described by Hansen, is a small slender bacillus from 4 to 6 microns in length and surrounded by a slimy envelope. These bacilli behave toward staining reagents very much like the tubercle bacillus, but are less resistant toward acid and alcohol than is the tubercle bacillus, so that a differentiation is possible provided decolorization is rather severe. The stained bacilli often show clear spots or appear as if made up of stained granules. These organisms may be found in many cases of leprosy in the sputum or nasal secretion, so that in doubtful cases a differentiation is necessary. While these organisms stain much more easily than do the tubercle bacilli and are more easily decolorized, it may be necessary to resort to inoculation ex- periments to make the absolute differentiation.1 This bacillus may be found normally in the saliva, coating of the tongue, the tartar of the teeth, and in the crypts of the tonsils. Pathologically, it may be found in cases of simple bronchitis, in the sputum in cases of gangrene of the lungs, and in the suppurative discharges from the ears. While these organisms are much more commonly confused with the tubercle bacillus when the urine is examined than when the sputum is investigated, yet they must be borne in mind in every sputum examination. It may be necessary to resort to inoculation experiments to decide the question, but ordinarily the use of the differentiating stain of Pappenheim (previously described) as well as the following method of Bunge and Trantenroth may be used. After fixation of the smear, the fat is removed by soaking the specimen in absolute alcohol. The preparation is now covered with a 5 per cent, solution of chromic acid for 15 minutes, after which it is (c). Smegma Bacillus. 1 See Duval, Jour. Exper. Med., vol. 13, 1911, p. 365; also Duval and Wellman, Jour. A. M. A., vol. 58, 1912, p. 1427; Clegg cultivates this bacillus in symbiosis with amoebas and cholera vibrios (Philip. Jour, of Sci., Sec. B., vol. 4, 1909, p. 403). 24 DIAGNOSTIC METHODS. washed with water. The smear is stained with carbol-fuchsin, decolorized with 16 per cent, sulphuric acid for three minutes, and is then counter-stained for five minutes in a concentrated alcoholic solution of methylene blue. This method is said to give the tubercle bacillus as distinct red, while the smegma bacillus is blue. While both of these methods of differentiation are usually applicable, yet one occasionally finds the smegma bacillus resisting the action of 16 per cent, sulphuric acid for 30 minutes and occasionally of strong alcohol for 12 hours. It is, therefore, necessary in such cases to resort to the court of last appeal, the guinea-pig test. (J). The Timothy Bacillus. This organism is present in the mouth reaching it through the medium of butter and milk, which may contain it in large numbers. These bacilli resist the decolorizing action of both alcohol and acid to almost the same extent as do the tubercle bacilli, but they usually appear as somewhat longer and thicker rods. Strangely enough, this organism produces a lesion in guinea-pigs which re- sembles very closely that of true tuberculosis, so that the inoculation test will not always be conclusive unless other animals are inoculated with material from the first one, in which case no lesions develop in the later animals. Fortunately, the cultural peculiarities of this organism are markedly different from those of the tubercle bacillus, as the former develops readily on the ordinary culture media. Fig. 6.-Diplococcus pneumoniae. (Williams.) (e). The Pneumococcus (Diplococcus pneumoniae'). This organism, discovered by Frankel and elaborated by Weichselbaum, is generally recognized as the etiologic factor in cases of acute croupous pneu- monia, although other organisms not infrequently give rise to this condition. It is found in large numbers in the sputum and other exudates, appearing as a small slightly elongated conical or lance-shaped coccus, which shows a THE SPUTUM. 25 marked tendency to occur in pairs (diplococci) with the broader ends in appo- sition. Occasionally it is arranged in short chains resembling streptococci. In exudates and in the blood this organism is usually enveloped by a well- defined hyaline zone or capsule. In clinical work it is rarely necessary to stain this capsule to identify the organism but if this be desirable it is best done by the method of Rosenow.1 Cultural peculiarities and methods of differentiating this organism from other encapsulated cocci must be found elsewhere. It stains well with the ordinary dyes and is Gram-positive. Friedlander's Bacillus (bacillus mucosus capsul- atus) is occasionally found in some cases of lobar and lobular pneumonia and occasionally may be considered the etiologic factor in such conditions, although it is usually a secondary invader. These bacilli grow readily on artificial media, are encapsulated and stain easily with the ordinary dyes but are Gram-negative. Fig. 7.-Friedlander's bacillus (above): pneu- mococcus (below). (Greene.) (f). The Influenza Bacillus. This bacillus, known as Pfeiffer's bacillus, is found in the bronchial sputum, especially in the pulmonary type of this disease. The most characteristic sputum is greenish-yellow in color with lumps of pus in nummular form. The organisms are found in such sputum as small, short bacilli measuring 2/10 to 3/10 micron in breadth by 5/10 micron in length. They usually occur singly, but may form chains. In the stained specimens these organisms show distinct polar stainings, appear- ing frequently as diplococci. They are stained with dilute carbol-fuchsin solution, faintly with the ordinary methylene blue solution, or are identi- fied by their Gram-negative character- istics. The best counter-stain used in the Gram method is either Bismarck brown or safranin, the organisms ap- pearing both intra- and extracellular. [Fig. 8.-Bacillus influenzae in sputum. (Abbot t) (g). The Bacillus Pertussis. This organism, discovered by Bordet and Gengou, and elaborated by Klimenko, has been frequently found in the sputum in cases of whooping- cough. It resembles very closely the influenza bacillus, appearing as short, plump, ovoid bacilli, with rounded ends and lying singly or in small groups between the pus and epithelial cells. It stains feebly with the usual dyes and is Gram-negative. This organism is rarely intracellular and may thus be dis- tinguished from the influenza bacillus. 1 Jour, of Inf. Dis., vol. 9, 1911, p. 1. 26 DIAGNOSTIC METHODS (A). Typhoid Bacillus. This organism has been found in the sputum in typhoid fever cases show- ing a coexistent bronchitis or pneumonia. The sputum is usually hemorrhagic in character and shows the bacilli as short, thick rods, staining with the ordinary dyes and negative to Gram's stain. For the absolute identification of this or- ganism cultural and agglutination tests are essential. (i). Staphylococcus and Streptococcus Pyogenes.. These organisms are found in practically every sputum examined and can be identified only by the use of cultural methods. They stain well with any of the aniline dyes and are Gram-positive. Their presence in the sputum has little pathologic significance. This bacillus of bubonic plague was discovered by Kitasato and Yersin in 1894. It is a short, thick bacillus, measuring from 0.8 to 2 microns in length and from 0.4 to 0.8 micron in thickness. A capsule may be usually made out and the stained organism frequently resembles a diplococcus, owing to the intense polar staining with intermediate faint staining. These organisms are Gram- negative. The bacillus pestis is found in the sputum of persons suffering from the pneumonic type of this disease and should be recognized owing to the markedly infective character of the material.1 It may be necessary absolutely to identify the organism by inoculation and cultural experiments. (;). The Bacillus Pestis. The sputum of cases of pulmonary anthrax may contain large numbers of these bacilli. These organisms are from 5 to 10 microns in length and from 1 to 1 1/2 in breadth. They are frequently grouped in long segmented threads, the segments varying in length, but usually being two or three times as long as broad. Occasionally these bacilli may be single, but are usually multiple. They form oval spores in the middle of the short segments. The organism stains with the ordinary dyes and is also Gram-positive. For absolute identifica- tion cultural and inoculation experiments, the latter into white mice, may be necessary, but the morphological characteristics of this organism will usually identify it. (k). The Bacillus Anthracis. (T). The Bacillus Mallei. This organism of glanders is found in the sputum in the pulmonary form of this disease. Morphologically, there is nothing characteristic in the ap- pearance of this organism beyond the fact of the presence of faintly staining areas in the protoplasm of the rather long bacilli. These organisms stain by Gram's method as well as with the ordinary aniline dyes. For a final diagno- sis inoculation into a guinea-pig should be made. 1 See Strong and Teague, Philip. Jour, of Trop. Med., vol. 7, 1912, p. 187. PLATE II. Streptococcus Pyogenes. (Methylene Blue Stain.) THE SPUTUM. 27 (m). Actinomyces Hominis (Ray Fungus). This fungus, which gives rise to the condition known as lumpy jaw in cattle, occasionally infects man, causing pulmonary conditions designated strep- tothricosis. The mucopurulent sputum in such conditions contains elastic tissue and small sulphur-yellow granules which are visible to the naked eye and are the characteristic findings of such cases. Macroscopically these granules are yellowish, grayish, greenish, or brownish in color, and are sometimes abundant and sometimes scarce. They are very friable, and when gently crushed beneath the cover-glass and examined microscopically appear to have broken up into hyaline rounded masses at the margins of which, on close inspection, fine radial Fig. 9.-Actinomyces. (Williams.) striations or filaments or hyaline club-shaped bodies, all closely set together, may be seen. The club-shaped bodies are variable in size and are composed of a hyaline refringent substance. In the granules obtained from the lesions in man the club-shaped formations are much less frequently observed than those obtained from the lesions in cattle (Mallory and Wright). If cover-glass preparations be made and stained with Gram's method, one will usually find isolated and matted filaments, many of which may be seen to branch, in addition to longer and shorter fragments of filaments and fine detritus of the same. If clubs are present in the granules they may be found scattered throughout the preparations. In this connection see Bridge.1 (g). Animal Parasites. (a). Amebae. Artault has described a unicellular ameboid body which resembles very closely a leucocyte when stained, but, while motile, differs in refractility and staining quality. This he calls the ameba pulmonalis. In cases of perforating liver abscess the true amebae coli may be found in the sputum, and, according to Flexner, in cases of abscess of the jaw communicating with the mouth. These amebae will be discussed fully in the chapter on Feces, to which the reader is 1 Jour. A. M. A., vol. 57, 1911, p. 1501. 28 DIAGNOSTIC METHODS. referred. It should be noted here that these organisms may not be numerous so that many slides may have to be examined. Nothing should be called an ameba unless it shows true ameboid movement. (b). Flagellates. Flagellated organisms, such as the trichomonas pulmonalis and the cercomonads, are found in the sputum associated with Dittrich's plugs incases of gangrene, putrid bronchitis, and tubercular cavity formation. The tricho- monas is probably identical with the trichomonas vaginalis or intestinalis which will be discussed later. Fig. to.-Paragonimus westermanii; (ventral view); ioXi. A, oral sucker; B, ceca; D, acetabulum; E, genital pore; F, uterus; G, ovary; H, testicles; I, vitelline glands; K, excretory canal; L, excretory pore. (Tyson after Braun.) (c). Cestodes. Not infrequently the lung is the seat of infection with the taenia echinococcus. In such cases various foreign bodies, such as fragments of membranes, scolices, booklets, and cysts, may be found in the sputum. Such formations may, also, be found in cases of liver abscess perforating into the lung. Any one of the above formations is characteristic of this condition. The parasite will be dis- cussed fully in the section on Feces, to which the reader is referred. The sputum in such cases is usually purulent or mucopurulent and may be tinged with blood. This sputum may be expectorated over a long period THE SPUTUM. 29 of time and may even contain tubercle bacilli from a coexistent tuberculosis. A distinguishing point between the sputum of pulmonary echinococcus disease and that of a perforating liver abscess is that the sputum in the latter is usually bile stained. (d}. Trematodes, The most common of this class of animal parasites is the ordinary "lung fluke," which has been called also distoma pulmonale, distoma Westermanii, distoma Ringeri, and Paragonimus Westermanii. The eggs of this parasite are much more frequently found in the sputum than are the parasites themselves, so that the diagnosis will rest with the finding of these ova. These eggs measure from 80 to 100 microns in length and 40 to 60 microns in width. They are brownish in color, oval in shape, have a smooth thin shell and a lid near one end which is quite characteristic The parasite is from 8 to 10 mm. long, 4 to 6 mm. wide, and is very markedly rounded anteriorly, being nearly as thick as broad. The sputum in such cases is usually small in amount, is very tenacious, and is reddish or rusty due to admixture of blood with the mucus Frequently no blood is found, in which case the sputum will still be of a yellowish or brown color due to the eggs themselves. The sputum, also, contains many spirals, which resemble very closely the Curschmann spiral, and, also, the Charcot-Leyden crystals. The eggs of another species of distoma, the distoma hematobium, have been found in the sputum by Manson. Gage1 has recently reported the finding of the larvae of strongyloides intestinalis in the sputum. Fig. ii.-Ovum of paragonimus wester- manii, from sputum: 1000 X 1. (Tyson after Braun.) V. The Sputa in Disease. (i). Pulmonary Tuberculosis. It has been truly said by Brown that pulmonary tuberculosis has no charac- teristic form of sputum The amount voided may vary from the very slight type of fibroid tuberculosis to the very abundant sputum of cavity formation. It is to be said that the amount of pus will usually depend upon the extent of the secondary infection, although caseous degeneration may lead to the expectora- tion^of large amounts of material resembling pus. In the early cases of pulmonary tuberculosis we may find a small amount of sputum which is expectorated only in the morning. This may be very tenacious and resemble very much the sago-like sputum previously mentioned. Sooner or later, depending upon the extension of the disease, there will appear small caseous particles which are very suggestive. As ulcerative processes proceed, the sputum becomes more profuse, yellowish or greenish in color, and 1 Arch, of Int. Med., vol. 7, 1911, p. 561. 30 DIAGNOSTIC METHODS. muco-purulent in character. In any stage of this ulcerative tuberculous con- dition we may find blood in amounts ranging from a few blood-cells to a sputum loaded with blood from a hemorrhagic focus. Likewise we will find elastic tissue in more or less amount and tubercle bacilli varying from a few to many in each field. The color of the tubercular sputum may range through all the shades of the spectrum, fhe greenish shade being associated with a most marked decomposition. As stated previously, the most suspicious looking sputa fre- quently contain no tubercle bacilli, so that frequent search must be made for these organisms which are the only diagnostic point of the sputum in this condition. (2). Croupous Pneumonia. The early sputum of acute lobar pneumonia is usually yellowish-red in color and very tenacious in consistency. In some cases the sputum is mucoid and abundant for a few days, but soon takes on the characteristic reddish color from the presence of unchanged red blood-cells. Its consistency is so great that the receptacle may be inverted without allowing any material to run out. The characteristic rusty sputum, which is found when the exudation into the alveoli is taking place, is homogeneous, glairy, very tenacious, and deep red in color. This rusty sputum, while characteristic of pneumonia, is sometimes replaced by one ranging in color from a yellow to a green. These colors are due to dif- ferent oxidation products of hemoglobin, and are, perhaps, more frequently, observed in the stage of resolution when the sputum becomes less tenacious and more abundant. The greenish sputa in pneumonic conditions have some im- portance. This coloration may be due to a coincident jaundice or may arise from delayed resolution, especially when the exudate has been particularly hemorrhagic. It is, moreover, sometimes an indication that a true tubercular condition has intervened and, hence, that the prognosis must be guarded. The so-called prune-juice sputum usually indicates a severe type of the disease, while at times it may signify merely a beginning resolution.1 Fibrinous coagula are found, according to Osler, in every case in which search is made. These may vary from very small bronchial casts to very large branching types. Curschmann's spirals as well as the Charcot-Leyden crystals are frequently observed. The characteristic organism of this condition, the diplococcus lanceolatus of Frankel, is usually found, but has only incidental im- portance, as it is so frequently present in the sputum of normal individuals. (3). Bronchopneumonia. The sputum of this disease is rarely characteristic. It partakes of both the type of a bronchitic and a pneumonic sputum. It may, therefore, contain much mucus and pus, may be viscid, may be streaked with blood, but is rarely 1 Pacini (Interstate Med. Jour., vol. 19, 1912, p. 536) has recently advanced the following reaction as characteristic of early pneumonic sputum. Mix 1 volume of sputum with 10 vol- umes of distilled water, agitate for 5 minutes and filter. To a test-tube containing 10 c.c. of distilled water add 5 drops of a 1 per cent, aqueous methyl-violet solution. Add to this latter, 10 drops of the sputum filtrate. A positive reaction is shown by the appearance of a distinct red color. \ THE SPUTUM. 31 so distinctly rusty as in the croupous type of this disease. As the disease is so limited in extent it is more or less rare to find an abundant sputum or to observe fibrinous coagula. Microscopical examination shows various organisms, but nothing diagnostic. (4). Acute Bronchitis. The sputum in this condition is very scanty in the early stages, is usually very tenacious and is expelled with difficulty. This early bronchitic sputum is known as "sputum crudum" and consists of practically pure mucin, containing within its meshes a few leucocytes, red cells, bronchial epithelial cells, and a few myelin drops. After a few days the sputum is increased in amount, becomes less viscid, and assumes the type of a distinct muco-purulent sputum. This sputum, called the sputum coctum, contains numerous pus-cells, is yellow or yellowish- green in color, shows the presence of large numbers of red cells, as a rule, and an increase of the polynuclear leucocytes over the mononuclear form. These mononuclear forms are more characteristic of the sputum of true tubercular conditions. Fat may be found, either in isolated drops or in large masses. As improvement in the condition occurs, the sputum becomes more abundant, and more distinctly purulent, and then gradually diminishes until it ceases. The sputum of acute bronchitis may give much information as to the course of the disease, as the transition from the viscid mucoid sputum through the abun- dant purulent stage to the final cessation is quite characteristic. (5). Chronic Bronchitis. (a). Simple Chronic Bronchitis. In most of these cases the sputum is either very little in amount or is much more abundant than in the acute forms. Such cases of simple chronic bron- chitis are usually those following the acute type of the disease in which we find the expectoration, for long periods of time, of a tenacious, viscid, and scanty sputum. Later it may become more abundant and muco-purulent, and may have a dark color and a distinctly foul odor. In the type of chronic bronchitis associated with cardiac disease we find large amounts of blood which may be fresh or changed, giving the typical prune-juice appearance. In such sputum we frequently find large numbers of the so-called "heart-disease cell" which have been previously described. (6). Putrid Bronchitis- This condition is brought about by dilatation of the bronchial tubes fol- lowing a chronic bronchitis. The sputum lies stagnant in these dilated bronchi so that it decomposes to a great extent. The sputum in such cases is very abundant, is of an ash-gray or brown color, is markedly purulent, and has a very disagreeable odor. On standing it separates into the three layers which have been previously discussed. In such conditions no elastic tissue fiber is found, so that we have here a differentiation from gangrenous or tuberculous pul- 32 DIAGNOSTIC METHODS. monary conditions. The sputum in this condition is very similar to that found in bronchiectasis, which is usually associated with decomposition of the sputum. Whether a diagnosis is possible between a straight putrid bronchitis and bron- chiectasis is doubtful, if one relies merely upon the sputum. The sputum in bronchiectasis occurs usually in the morning and is then very profuse. It shows, however, the characteristics of the sputum of putrid bronchitis, but is more commonly associated with the presence of pus, while 50 per cent, of cases show more or less profuse hemorrhage. (c). Fibrinous Bronchitis. The chief characteristic of this condition is the expectoration of more or less perfect bronchial casts. These may be single or may be distinctly branch- ing, showing the arrangement of the entire bronchial tree. This condition occurs quite frequently associated with many febrile diseases, but in the dis- cussion at this point we have reference to the idiopathic type of the disease. The sputum in this latter class of diseases is mucoid and very abundant in the earlier stages. After a few days there is expectorated, following a severe coughing spell, a bronchial cast. This expectoration is usually tinged with blood. Such casts may be expectorated over long periods of time and their form may vary as previously described. (6). Bronchial asthma. The sputum in bronchial asthma is, perhaps, more characteristic than that of any other pulmonary condition. During the paroxysm of asthma there may be no sputum, or it may be scanty, consisting of the glairy mucoid plugs known as the pearls of Laennec. The sputum contains many eosinophile cells and many alveolar epithelial cells with myelin degeneration. The mucoid sputum usually contains large numbers of the spirals of Curschmann along with the Charcot-Leyden crystals. In some cases of asthma one finds small cylindrical casts of bronchi. Some of these branch while the majority are straight and may taper at one end into the central fiber of a true spiral. 7). Influenza. The sputum of the pulmonary type of this condition shows in the early stages as a very scanty tenacious expectoration. Later it increases in amount, becomes muco-purulent and often blood-streaked, and is greenish-yellow in color. This sputum contains large numbers of Pfeiffer's bacilli, which have been previously discussed. (8). Gangrene of the Lung. The sputum in this condition is very profuse, is greenish-brown in color, is very offensive in odor, and is extremely fluid in character. It contains shreds of elastic tissue which serve to distinguish it from the sputum of putrid bronchi- tis or bronchiectasis. This sputum separates, as do other forms of sputum which have undergone stagnation and decomposition, into three distinct layers. THE SPUTUM. 33 Microscopic examination shows fragments of necrotic tissue varying from very minute particles to those several cm. in length. Very few epithelial cells or leucocytes are found, but red blood-cells are more or less frequent. The bacterial content is usually very high, but nothing characteristic is found among these organisms. (9). Abscess of the Lung. The most characteristic feature of true abscess of the lung or of liver abscess which has perforated into the lung is the sudden appearance of a large amount of pure pus containing fragments of lung tissue This material usually has the normal odor of pus, but may become offensive, although never as mark- edly so as in gangrene or putrid bronchitis. The sputum of the perforating liver abscess is usually distinguished from that of the true liver abscess by the so-called "anchovy-sauce" appearance. The color may vary, due to the presence of various types of bile pigment. Microscopically, bilirubin crystals may be found. (10). Perforating Empyema. The sputum of such conditions is composed almost entirely of pus and is thin and liquid. It contains many hematoidin crystals, but very little elastic tissue fiber or other tissue fragments. The odor is usually described as that of old cheese in the beginning, but soon becomes offensive owing to decomposition. (11). Pneumonoconioses. The sputum in these various conditions will depend upon the pigment with which the lung has been infiltrated. The expectoration is usually muco- purulent, very profuse, and is laden with coal-dust (anthracosis), iron-dust (siderosis), with stone-dust, chalk-dust or plaster of Paris (chalicosis), and with starch granules (amylosis). BIBLIOGRAPHY. i. Andre. La grippe ou influenza. Paris, 1908. 2. Aufrecht. Die Lungenentziindungen. Wien, 1901. 3. Cornet. Die Tuberkulose. Wien, 1901. 4. De Jong. Etude histo-chimique et cytologique des crachats. Paris, 1907. 5. Hoffmann. Die Krankheiten des Bronchien. Wien, 1900. 6. Sticker. Lungenblutungen. Wien, 1901. CHAPTER II. ORAL, NASAL, AURAL, AND CONJUNCTIVAL SECRETIONS. I. Oral Secretions. (1). General Considerations. The oral secretion is a mixture derived from the various buccal glands, the submaxillary, sublingual, parotid, and mucous glands. To this secretion has been given the name saliva. It is a colorless, odorless, and tasteless fluid, which appears somewhat stringy and frothy, separating on standing into two layers, the upper one of which is clear and the lower one cloudy. The function of this secretion is to moisten the mouth and throat and, also, to aid in swallow- ing the food as well as partially to digest the starchy food through the action of a specific ferment (ptyalin) which it contains. The normal daily amount of saliva secreted is usually about 1,500 c.c., this quantity varying under the influence of many factors, both physiologic and pathologic. The specific gravity ranges between 1,002 and 1,009 giving a total solid content of 3 to 12 grams. Its reaction is alkaline, corresponding to 0.006 to 0.048 per cent, of sodium hydrate. While the reaction of the saliva is normally always alkaline, we occasionally find an acid reaction, especially in children and in the early morning hours, due to the production of lactic acid by the bacteria which are always present in the mouth. Likewise we find an acid reaction especially in conditions associated with acidosis, such states very frequently leading to dental caries and to many other irritative conditions of the mouth. The recent work of Talbot along this line is especially interesting. The chemical composition of the saliva does not have any great clinical significance with the exception of the presence of the sulphocyanates, the nitrites, and the characteristic ferment ptyalin. These substances seem to have some importance both from a diagnostic and symptomatic standpoint, so that a few remarks may be timely. The presence of potassium sulphocyanate (KCNS) is more or less characteristic of normal saliva and may be detected as follows: Collect a few c.c. of saliva before meals and allow this to filter. Add a few drops of hydrochloric acid and then a drop or two of ferric chlorid solution, when a distinct red color will be observed, whose depth will depend upon the amount of sulphocyanate present. It has been stated that heat should be ap- plied in this test, but the writer has never found it necessary as the characteristic reaction almost invariably appears in the cold. This color disappears on the addition of mercuric chlorid solution, which fact may serve to differentiate it from the similar one given by the saliva of opium habitues and due to meconic acid. Very little pathologic significance has been attached to variations in 34 PLATE III. Leptothrix and Spirocheta Buccalis. (Unstained Specimen.) ORAL, NASAL, AURAL, AND CONJUNCTIVAL SECRETIONS. 35 the amount of the sulphocyanate in the saliva, but it is interesting to note that in many cases of diabetes as well as in cases of severe stomatitis this sub- stance is frequently absent. The nitrites may be detected by the more delicate tests used in water analy- sis, as their amount is usually not sufficient for the ordinary qualitative tests. A very good test is the use of the Griess-Ilosvay reagent (1/2 gram of sulphan- ilic acid is dissolved in 150 c.c. of dilute acetic acid and treated with 1/10 gram of naphthylamin dissolved in 20 c.c. of boiling water. On standing, a blue sediment forms which is separated and dissolved in 150 c.c. of dilute acetic acid). On treating 10 c.c. of saliva with a few drops of this reagent and heating, a red color will develop in the presence of nitrites. The most important constituent of saliva is the ptyalin which has a defi- nite hydrolytic action upon starch, converting this polysaccharid into maltose through the intermediate stages of erythrodextrin and achrobdextrin. This action may be readily seen by treating a little starch paste with a few c.c. of filtered saliva and placing the vessel in the incubator for 10 to 15 minutes. At the end of this time iodin solution is added when a distinct red color or an entire absence of color will be noticed. It is to be remembered here that starch, treated with iodin, is colored blue, so that a change of the starch is evident by the color reaction. Recent work by Litmanowicz1 has shown that the dias- tatic power of saliva is unaffected by physiologic or pathologic variations in general body functions.2 (2). Microscopic Examination. On allowing saliva to stand it separates into two distinct layers, the upper one clear and containing the liquid portion, while the lower is cloudy and con- tains the morphological elements. In the microscopic examination of this lower layer we observe many epithelial cells in the form of large, irregular, squamous cells which are derived from the mucous membrane of the mouth and tongue. The number of these cells present depends, of course, upon the erosion to which the mouth has been subjected by various irritants either of the food or of disease. The characteristic cells of the saliva are the salivary corpuscles, which resemble the leucocytes, but are larger and more granular. Occasion- ally red blood-cells are seen, but these have no direct significance other than to denote ulcerative or markedly irritative conditions somewhere in the naso- pharynx. Beside these constituents of the saliva, we find, microscopically, many micro-organisms of the mold, yeast, and bacterial types. The bacteria are always present in the mouth as they are taken in with the air, food, and drink. Few of these have any direct significance, although the Spirochaeta buccalis and microdentium should be borne in mind, especially when an examination is being made for the Spirochaeta pallida.3 The former is differentiated from the latter by the fact that its ends lie upon a line drawn longitudinally through 1 Zentralbl. f. d. ges. Phys. u. Path, des Stoffw., Bd. 4, 1909, S. 81. 2 See Hirata, Biochem. Zeitsch., Bd. 47, 1912, S. 167. 3 See Thibaudeau, Jour. A. M. A., vol. 59, 1912, p. 446. 36 DIAGNOSTIC METHODS. the center of its spirals, while such a line drawn through the pallida lies above and below its ends. Moreover, it should be remembered that the smegma bacillus is an occasional habitant of the mouth and throat and may occur in specimens of sputum, giving rise to the assumption of the presence of tubercle bacilli unless proper means of identification are used. Simon has pointed out an interesting fact that the majority of the micro-organisms which are constantly present in the mouth cannot be cultivated on artificial media, while the temporary invaders easily develop. Many pathogenic bacteria have been found in the mouth of the healthy subject. This is interesting clinically as showing the constant danger to which we are all subject, in case our resistance becomes low- ered. The writer recalls that the most virulent culture of pneumococci obtained from 200 throats, both diseased and normal, was from his own at the time when he was in perfect condition and showed no symptoms thereafter. Beside the pneumonia organism, streptococci and diphtheria bacilli are frequently found in the mouths of perfectly healthy individuals. Molds and yeast fungi are rarely found in the saliva during health, but they are frequently present in pathological conditions. (3). Pathologic Changes. The normal daily secretion of the saliva is, as stated above, about 1,500 c.c. The composition of the secretions of the various glands, which contribute to the mixed secretion, differs rather widely, the one from the other. We may, therefore, have changes, not only in amount of saliva, but, also, in the quality, depending on the diseased condition of one or more of these glands. The quantity of saliva is diminished in inflammation of the salivary glands, such as in parotitis, in all febrile diseases, in diabetes, and in nephritis. The secretion is also diminished by the therapeutic use of preparations of belladonna and of opium. It is increased by certain poisons, such as pilocarpin and mercury, by excessive irritation with acids and alkalies, and, also, by irritations arising fromcarious teeth. Occasional cases have been reported of a greatly increased amount of saliva through some obscure nervous reflex, while such a condition is not unusual in pregnancy. An increased flow of saliva is known as salivation or ptyalism. In determining whether or not salivation really exists, obser- vation will frequently show increased amounts of saliva at all times. In some cases, however, it is necessary to measure the amount and, also, to make later chemical examinations of the saliva. The best way of obtaining saliva free from contamination is to wash the mouth thoroughly with a solution of sodium bicarbonate, brush the teeth thoroughly with the same solution, and then rinse out the mouth with cold water. On now touching the inner surface of the teeth or the edge of the tongue with a glass rod that has been dipped into dilute acid, saliva will be seen pouring into the mouth from many points. This saliva is then collected in clean receptacles and the quantity measured. Variations in the reaction of the saliva are not uncommon in pathologic conditions. In various intestinal diseases with which we may have an associ- ORAL, NASAL, AURAL, AND CONJUNCTIVAL SECRETIONS. 37 ated stomatitis, an acid reaction is frequently noted. Also in fevers, diabetes, starvation, and other conditions giving rise to acidosis (overloading of the system with acid products) the reaction of the saliva is always acid. Strauss and Cohn believe that the saliva is practically always alkaline, even under pathologi- cal conditions. Coating of the Tongue. A coating of the tongue is practically always abnormal, as the normal appearance is a bright reddish color with no visible deposits. A change in the normal appearance of the tongue has so long been indicative, in the minds of the profession, of disturbed conditions not only in the mouth, but in the stomach and the bowels, that one should always take into consideration any such change. In severe infectious fevers a brownish coating with a furred appearance is practically always seen. This consists of remnants of food and of incrusted blood, along with large numbers of micro-organisms and dark desquamated epithelial cells. The white coating contains no blood and is more indicative of simple gastro-intestinal disturbance than is the brown coating. The so-called "tartar" which forms upon the teeth seems to consist of deposited calcium carbonate and contains many actively motile spirochete as well as large segmented leptothrices, along with leucocytes and epithelial cells. Pharyngomycosis Leptothrica. In many pathological conditions of the throat, such as tonsillitis, diphtheria, and thrush, we frequently find the tonsillar and other buccal structures covered with a coating which is, in many cases, a distinct membrane containing the pathogenic organisms in large numbers. Many perfectly normal subjects complain of the formation, in the tonsillar crypts, of plugs of material which are easily removed by pressure. These are frequently found in patients subject to tonsillitis, but, also, in those showing no pathological conditions of the tonsils and are closely related to Dittrich's plugs, which have been discussed. In the pyoid masses of pharyngomycosis leptothrica, one finds large numbers of lymphocytes, epithelial cells and long segmented fungi, the lepto- thrices buccalis, which are colored bluish-red by a solution of iodopotassic iodid. In such conditions the polynuclear neutrophiles are present in only small numbers. In some cases patches of these fungi extend over quite an area of the tonsils so that the appearance may be one of the formation of a diph- theritic membrane, although microscopic examination will at once clear up the diagnosis. Diphtheria. One of the most important examinations of the oral cavities consists in the detection of the diphtheria bacillus (Klebs-Lb filer bacillus), as an early diagnosis of this disease frequently enables the physician to institute antitoxin treatment. Such an examination should never be omitted in any case of suspected sore throat, especially where any membranous patches are present. 38 DIAGNOSTIC METHODS. By means of a stout platinum loop or a swab of cotton a piece of membrane or a portion of the exudate is scraped from the throat. This material is then spread over the surface of Lbffler's blood serum and is allowed to incubate at 370 C. for six to eight hours. This period of incubation is of some importance as it has been definitely shown that at the end of six to eight hours the diphtheria organism is the only one which will attract much attention, while if left for a longer time, other organisms, especially the streptococcus and staphylococcus, will so far outgrow the diphtheria bacillus that this latter may be unrecognizable unless the incubation be carried 36 hours, when the diphtheria bacillus then assumes the ascendency. From this culture, cover-glass preparations are then made and stained for one to five minutes in Loffler's alkaline methylene blue solution. They are then rinsed in water, dried, and examined with the im- mersion lens. Frequently attempts are made to diagnose diphtheria by examinations of smears from the fresh exudate. Such cultureless smears rarely show the characteristic appearances of the diphtheria bacilli, so that the writer would advise the use of a preceding culture in all cases. The character- istic cultural peculiarities upon different media must be looked for in works on bacteriology. Neisser's Stain. This stain is supposed to differentiate the diphtheria bacillus from all others The smear is stained for five minutes with a methylene blue solution (methylene blue, Grubler, i gram, 20 c.c. of 96 per cent, alcohol, glacial acetic acid 50 c.c., and water 950 c.c.). The stain should be filtered before use. The specimen is heated gently during the staining process and the dye renewed as the stain evaporates. Wash in water and stain for two minutes with an aqueous solution of Bismarck brown or, better, a dilute solution of safranin. The polar bodies will be stained a deep blue, while the body of the bacillus will take a light brown or red color.1 Microscopically, the stained organism appears as a slightly curved rod, but especially characteristic are the bizarre forms, such as rods with alternate staining and nonstaining portions, rods with distinct deeply staining polar bodies, club-shaped or " narrow-waisted' rods, many of which lie together in distinctly parallel lines. Diphtheria bacilli may be found in the throat for weeks after all symptoms have disappeared so that it is wise to enforce isolation of the patient until a negative examination for these organisms is obtained. Occasionally in examination of smears from the throat, true diphtheria bacilli may be confounded with pseudo-diphtheria bacilli, and in examination of other specimens, such as those taken from the eye, the bacillus xerosis may be confusing. These different organisms are best differentiated by the study of their action in fermenting or not fermenting certain sugars. According to Knapp, the pseudo-bacilli will ferment none of the sugars, the diphtheria bacilli will ferment dextrose, mannite, maltose, and dextrin, but not saccharose, while 1 See Raskin, Deutsche med. Wchs., Bd. 37, 1911, S. 2384. PLATE IV. Diphtheria Bacilli Showing Polar Staining. (Neisser Method, Counter Stained with Safranin.) ORAL, NASAL, AURAL, AND CONJUNCTIVAL SECRETIONS. 39 the xerosis bacillus ferments dextrose, mannite, maltose, and saccharose (cane sugar), while it does not ferment dextrin. The details of the methods for these tests must be found in bacteriological works. Vincent's Angina (Ulceromembranous Angina and Stomatitis). In this condition, smears taken from the throat, as well as the free saliva will be found to contain many organisms of two especial types, the first, spirilla, and the second, long fusiform bacilli. Usually both of these types are found together, but occasionally the spirilla are absent. The spirilla usually measure from 36 to 40 microns in length and 1/2 micron in breadth, while the bacilli are 6 to 12 microns in length and are somewhat thicker in the center than at the end. These organisms may be readily stained with Lbffler's methylene blue, gentian violet, or dilute carbol-fuchsin, but they decolorize with Gram's method. They have so far shown negative cultivation and inoculation results. They are regarded by some as representing definite stages in the development of a trypanosome. Fig. 12.-Vincent's spirillum and bacillus. (Coplin.') Streptococcic Sore Throat. During the last year several milk-borne epidemics of sore throat with severe constitutional symptoms have occurred, especially in Boston, Chicago and Balti- more, of which the causative organism is a peculiar streptococcus (the strep- tococcus epidemicus). In smears from the throat and tonsillar exudates this highly virulent organism occurs in short chains, the spherical cocci appearing in twos in the chain. They are strongly Gram-positive and are surrounded by a definite capsule. They produce a relatively narrow zone of hemolysis on blood agar but little or no greenish color.1 Gonorrheal Stomatitis. 1 See Davis and Rosenow, Jour. A. M. A., vol. 58, 1912, p. 773; Hamburger, Ibid., p. 1109; Davis, Ibid., pp. 1283 and 1852; Capps and Miller, Ibid., p. 1848; Heinemann, Ibid., vol. 59, 1912, p. 716; Leutscher, Ibid., p. 869; Frost, Pub. Health Reports, vol. 27, 1912, p. 1889; Stokes and Hachtel, Ibid., p. 1923; Winslow, Jour. Inf. Dis., vol. 10, 1912, p. 73; Hamburger, Bull. Johns Hopk. Hosp., vol. 24, 1913, p. 1. In this condition the usual changes of infection are observed along with 40 DIAGNOSTIC METHODS. the appearance of the gonococci in the smears. Boston reports several cases of supposed gonorrheal stomatitis in which cultural methods showed the absence of this organism, although the smears showed the presence of intra- cellular Gram-negative diplococci. Such reports are not surprising in view of the fact that so many saprophytic diplococci are found which may or may not stain by Gram's method. Thrush. This is a condition most commonly seen in children, but may occur in adults, especially in those with tubercular tendencies. The saliva in this con- dition is usually acid and somewhat increased in amount. Microscopic ex- amination of the membrane shows many epithelial cells, leucocytes, and much Fig. 13.-Oidium albicans. (Kolle and Wassermann.) granular detritus with a network of branching band-like formations, showing distinct segments. The contents of the segments are clear and usually contain two highly refractive granules, one at each pole. This organism is known as the Oidium albicans. It stains well with the ordinary aqueous methylene blue solution. II. Nasal Secretion. This secretion does not present many points for study and seems to be of pathologic significance only in infectious conditions. Normally, the nasal secre- tion is comparatively scanty, clear, tenacious, odorless, salty in taste, and alka- line in reaction. It is largely composed of mucus, showing squamous and cili- ated epithelium in abundance, occasionally leucocytes, large numbers of bacteria and Charcot-Leyden and triple phosphate crystals. The bacterial content of the nasal secretion is made up of both pathogenic and non-pathogenic organisms, the tubercle bacilli having been frequently obtained from a nor- mal mucous membrane. ORAL, NASAL, AURAL, AND CONJUNCTIVAL SECRETIONS. 41 Pathologic Changes. In most acute infections, as well as in the so-called acute colds, the nasal secretion is at first diminished in amount, but soon becomes very profuse. This secretion shows the same appearance as does the normal fluid, but as ulcer- ation ensues may be heavily loaded with pus-cells and bacteria. The chronic suppurative process in the nose may affect any or all of the accessory sinuses, so that we may have very severe conditions arising from simple ulceration. Frequently the ulcerative and membranous conditions spoken of above may ex- tend from the mouth to the nose, so that distinct diphtheritic membranes are frequently found in the nasal cavities. Hay Fever. In this condition the nasal secretion is found to be increased to a large extent at certain times of the day and much diminished at others, depending upon the paroxysms of the disease. Nothing of pathological importance has been found, however, in the examination of the nasal secretion in this condition, no specific organism having been identified. Meningitis. In some cases of meningitis the cerebrospinal fluid passes into the nasal cavity as a result of caries of the bones of the skull. This fluid may be dis- tinguished by the fact that it contains practically no albumin, but does show the presence of a reducing substance which may or may not be sugar. This fluid may also contain the diplococcus intracellularis meningitidis of Weichselbaum. While this organism is not found in all cases of epidemic meningitis, yet it is found in many, so that the nasal secretion may be of some diagnostic importance from this standpoint. The above diplococcus is similar, both in morphological and staining characteristics, to the gonococcus, and presents many difficulties in differentiation (see cerebro-spinal fluid). In the course of glanders, leprosy, plague, pneumonia, typhoid fever, influenza, and many other infectious diseases, the characteristic organisms of these conditions may be found in the nasal secretion. Recently Goldberger and Anderson1 have demonstrated the presence of the unknown virus of measles in the mixed buccal and nasal secretions.2 Occasionally concretions are found in the nose, but these rarely reach a large size and do not have very great pathologic significance. They are largely composed of vegetable fibers taken in by inhalation and cemented by mucus which is hardened by the deposition of lime salts. In the condition known as ozena the nasal secretion is found to contain many large diplococci (Lbwenberg). Klemperer and Schierer believe that the organism of ozena is probably Friedlander's bacillus, as it is very plentiful in the nasal secretions of this disease. 1 Jour. A. M. A., vol. 57,1911, pp. 476 and 971; Amer. Jour, of Dis. of Child., vol. 4, 1912, p. 20. 2 See also Lucas and Prizer, Jour, of Med. Res., vol. 26, 1912, p. 181. 42 DIAGNOSTIC METHODS. III. The Aural Secretion. Normally no secretions appear in the external ear, with the exception of that of cerumen, while the secretion of the middle ear and of the internal ear is normally inaccessible to examination. We find, therefore, that the chief im- portance which is attached to the clinical examination of the aural secretions, is entirely a pathological one. In catarrhal and inflammatory conditions of the external auditory canal, one finds naturally very large numbers of organisms, with which the disease may or may not be associated. In the chronic inflam- matory processes of the middle ear, the more important organisms found are the pneumococcus, streptococcus pyogenes, staphylococcus pyogenes, bacillus pyocyaneus, the bacillus of Friedlander, the bacillus coli communis, the diplo- coccus intracellularis, the typhoid bacillus, and especially the diphtheria bacil- lus. As disease of the middle ear is so commonly associated with disease of the naso-pharynx, it is possible to find in the discharge from the ear any organism which is causing trouble either in the nose or in the throat. Hamilton has shown the almost constant presence of the pseudo-diphtheria bacillus in the discharge of the running ears following scarlet fever. It is not an uncommon thing to find certain inflammatory processes of para- sitic origin in the external auditory canal. This condition is known as otomyco- sis and is frequently caused by the aspergillus niger. Besides this organism, many other fungi belonging to this group have been found in the external ear. Among these we find the aspergillus flavus and fumigatus, the aspergillus nidu- mus, the mucor septatus, the eurotium malignum, and the penicillium minimum. These parasites are very readily detected by removing a small portion of the mycotic mass and spreading it thinly on a slide. Add a small drop of water, apply a cover-glass, and examine under a high-power dry lens. For the details of structure of such organisms, bacteriological works should be consulted. Besides these fungi we occasionally find in the external auditory canal larvae of various insects. These larvae may develop later into the full-grown insect and may be removed from the ear by the movements of the animal. In other cases such larvae have incited inflammatory processes which are occasionally troublesome, as in the case reported by Richardson. IV. The Conjunctival Secretions. Under normal conditions the secretion of the conjunctiva and of the lacri- mal gland concerns us only very little. It is in the course of an inflammatory process in which these secretions may be greatly increased and greatly changed by the inflammation that any clinical importance attaches to them. In in- flammatory conditions of the conjunctiva we find certain organisms which require identification in order that proper treatment may be instituted and the proper prognosis given. It is to be recalled that the pseudo-diphtheria bacillus is practically always found in smears made from the conjunctival secretion, yet it is rarely, if ever, pathogenic in this situation. PLATE V. Morax-Axenfeld Diplobacillus. (Gram's Stain.) Courtesy of Dr. Brown Pusey. Koch-Weeks Bacillus. (Gram's Stain.) Courtesy of Dr. Brown Pusey. ORAL, NASAL, AURAL, AND CONJUNCTIVAL SECRETIONS 43 Pathologic Changes. In cases of conjunctivitis which are traceable to infection with the diph- theria organism, we frequently find the formation of an extensive membrane which consists of epithelial cells, leucocytes, and large numbers of streptococci along with the diphtheria bacilli. Clinically, a membrane formation on the conjunctiva may arise from infection of this tissue with organisms other than the diphtheria bacillus; hence, it is wise in all suspicious cases to submit a portion of the membrane to both direct microscopic and cultural examinations. The appearance of the organism has been previously described. Diphtheritic Conjunctivitis. Infectious Conjunctivitis. In acute infectious conjunctivitis, various organisms have been found, the most common ones being the Koch-Weeks bacillus, the pneumococcus, and the gonococcus. Occasionally one finds the staphylococcus, streptococcus, colon bacillus, influenza bacillus, Morax-Axenfeld bacillus, the diphtheria bacillus, and other organisms. In some regions the Koch-Weeks bacillus is frequently found as the etiologic factor in acute infectious conjunctivitis, while in others it is rarely, if ever, observed. This is an organism of the influenza group, a small, thin, Gram- negative bacillus, which is, so far as known, pathogenic only for the human conjunctiva (see cut). It grows best on media containing a slight amount of human blood, especially in symbiosis with the xerosis bacillus. This latter organism is differentiated from the Klebs-Lbffler bacillus only by the appli- cation of the fermentation tests spoken of under Examination for Diphtheria Bacilli in the Throat. The most common bacterial cause of chronic conjunctivitis is the bacillus of Morax-Axenfeld, usually seen as a diplobacillus, several groups of which may at times be arranged in chains (see Plate V). It is a Gram-negative organism and grows well on Lbffler's blood-agar, which it digests, forming on the sur- face, at the beginning of its growth, very characteristic small pits. This form of conjunctivitis is much more common than is generally sup- posed, so that the identification of the organism is of great importance. In the development of this type of conjunctivitis, large or small amounts of pus are invariably present between the folds of the conjunctiva. A portion of this pus may be collected by means of a cotton swab or a platinum loop and smeared thinly over a slide. This smear is then fixed in the flame and is stained for the gonococcus by both the methylene blue and Gram stains. The technic of this latter staining process, as well as the characteristic appearance of the organism, will be discussed in a later section Gonorrheal Conjunctivitis. Trachoma. Lately, a great deal of attention has been given to some bodies, which are found in trachoma and which are, possibly, the long-sought cause of this infec- 44 DIAGNOSTIC METHODS. tious disease of the conjunctiva. These organisms are known as the Prowazek- Greeff bodies and are shown in the accompanying cut. They are best stained by the Giemsa stain (see Blood), smears being made in the usual manner. The present status of these bodies is that they are almost always found in the acute stages of this disease and are, occasionally, observed in other conditions which do not present the clinical features of trachoma. Whether or not they are the true etiologic factor must still be considered an undecided question.1 Vernal Conjunctivitis. An interesting point in the diagnosis of the above condition, as well as of the conjunctivitis of hay fever, is the fact, first observed by Herbert2 and confirmed by many, that in these two types of conjunctivitis eosinophile poly- morphonuclear leucocytes are found in abundance, whereas in the ordinary forms of conjunctivitis such cells do not obtain. Pusey3 has, therefore, been lead to believe that these facts point strongly to a similar etiologic factor, namely pollen, in the vernal conjunctivitis and in the conjunctivitis of hay fever. BIBLIOGRAPHY. 1. Axenfeld. Die Bakteriologie in der Augenheilkunde. Jena, 1907-The Bacteriology of the eye. New York, 1908. 2. Davis. Les Mikrobes de la Bouche. Paris, 1890. 3. Kohl. Die Heferpilze. Leipzig, 1908. 4. Litchfield. Diphtheria in Practice. London, 1908. 5. Mallory and Wright. Pathological Technic. Philadelphia, 1912. 6. Miller. Die Mikro-organismen der Mundhohle. Leipzig, 1889. 7. Nuttall and Smith. Bacteriology of Diphtheria. Cambridge, 1908. 8. Sticker. Die Bedeutung des Mundspeichels in physiologischen und pathologischen Zustanden. Berlin, 1899. 1 See report from the Laboratories of the New York Health Department found in Proc. New York Path. Soc., vol. 12, 1912, p. 17. 2 Brit. Med. Jour., vol. 2, 1903, p. 73 3 Jour. A. M. A., vol. 57, 1911, p. 1207. PLATE VI. Trachoma Bodies of Prowazek-Greeff. (Giemsa Stain.) Courtesy of Dr. Brown Pusey. CHAPTER III. GASTRIC CONTENTS. I. General Considerations. The gastric juice is the product of the secretory activity of the glands of the stomach. Different series of glands contribute separate elements to the secretion, so that we find much variation, under pathologic conditions, in the composition of this fluid. The stomach should be regarded as a dilated and specialized portion of the general digestive tube, its walls consisting of the following four coats: mucous, submucous, muscular, and fibrous. From the standpoint of secretory activity the internal or mucous coat is the most important. This mucous membrane is covered throughout its entire length by a single layer of simple columnar epithelium. It follows the various folds or ruga dipping down in places to line the orifices and ducts of the tubular glands which are of such importance in the digestive activity of the stomach. The gastric glands are of two kinds, the peptic or fundus glands, situated in the middle and cardiac thirds of the stomach, and the pyloric glands, found in the pyloric third of the stomach. Peptic Glands. These glands are slightly wavy simple tubular depressions, in which a duct, a neck, and a fundus are recognizable. In exceptional cases the fundus is divided, while in nearly all it is tortuous or spiral its extremity being often sharply bent at right angles to the general axis of the tube (Piersol). In these peptic glands are found two types of cell. The first, known as the central, chief or adelomorphous cells, bound the lumen of the gland and form the bulk of the glandular epithelium. These cells are either polyhedral or columnar in form and each contains a spherical nucleus situated within the granular pro- toplasm. These cells do not stain readily with aniline dyes. The chief func- tion of these central cells of the peptic glands is to secrete the rennet and lipase which are present in the gastric juice. The second type of cell in the peptic gland is known as the parietal, acid or oxyntic cell and is situated in the periphery of the gland immediately below the basement membrane. These cells are more oval or angular in form, are larger than the chief cells, are more finely granular in structure and stain deeply with the aniline dyes They are directly concerned with the secretion of hydrochloric acid. The Pyloric Glands. These glands are characterized by their relatively long wide ducts into which the several divisions of the body open; the tubular compartments are 46 46 DIAGNOSTIC METHODS. wavy and tortuous and frequently end in slightly expanded extremities. The duct is lined by tall columnar epithelium, the cells becoming lower and broader as they approach the neck and toward the fundus. The cells contain finely granular protoplasm and do not secrete mucus but a thin albuminous liquid. Parietal or acid cells do not occur in the pyloric gland, being confined to the true peptic gland (Piersol). It will thus be seen that the active portions of the gastric juice are secreted by the fundus glands, the pyloric glands contributing nothing except a small amount of the ferments and liquid portion, the mucus being largely derived from the goblet cells which line the entire stomach and the wider portion of the glandular ducts. It would lead me too far astray to discuss the formation of the ferments in the cells, but it is well in passing to state that these ferments do not exist in the cells as such, but rather in the form of zymogens or pro- zymogens which become active only in the presence of the free hydrochloric acid. The free hydrochloric acid of the gastric juice is formed in the parietal cells of the peptic gland.1 The mechanism of this formation is not absolutely estab- lished, but it seems probable that this free acid arises from the chlorids taken up from the blood by these cells. Just what is the active agent in causing the con- version of the chlorids into free acid seems to be in doubt, but it may be either the continuous action of carbonic acid or, as Maly assumes, the interaction of the sodium phosphate (Na2HPO4) with the chlorids of the cell. It is also probable that the osmotic influences may be very great in the production of this free hydrochloric acid as Koeppe advocates. This acid is present at all times in the normal stomach, being found even in cases of extreme starvation. The recent work of Pawlow2 has shown that various factors influence the quantity and quality of the normal gastric juice. He asserts that the "appe- tite is the first and mightiest exciter of the secretory nerves of the stomach, a factor which embodies in itself a something capable of impelling the empty stomach of the dog in the sham feeding experiment to secrete large quantities of the strongest juice. A good appetite in eating is equivalent from the outset to a vigorous secretion of the strongest juice; where there is no appetite this juice is also absent." Moreover, under natural conditions, the stimulation of food is a very important factor. The administration of a diet causes a secre- tion of gastric juice which is directly proportionate, both in amount and activity, to the diet taken. We find, according to Chigin,3 that the greatest digestive power is shown by the juice excreted after the administration of bread, although the total acidity is greatest following an intake of meat. If we compare equiva- lent weights of food material we find that flesh requires the most gastric juice and milk the least; but taking equivalents of nitrogen, bread needs the most and flesh the least. In this connection it is well to remember that the gastric secre- tion varies from hour to hour. Thus the most active juice occurs with flesh 1 See Harvey and Bensley, Biol. Bull., vol. 23, 1912, p. 225. 2 The Work of the Digestive Glands. London, 1902. 2 Loc. cit. Pawlow. GASTRIC contents. 47 in the first hour, with bread in the second and third hour, and with milk in the fifth to the sixth hour. The point of all this is that the rate and time of secretion of the gastric juice is always characteristic for each diet. Moreover, it has been found that the hydrochloric acid first secreted com- bines at once with the proteins of the various food stuffs, so that we may find no free hydrochloric acid in the gastric contents, although the secretion may be normal and may show a very high degree of total acidity. Usually, however, we find the presence of free hydrochloric acid in amounts ranging from 0.2 to 0.3 per cent. II. Methods of Obtaining the Gastric Contents. Unless the patient is one who can easily eject the contents of the stomach by vomiting, it is necessary to resort to the introduction of the so-called stomach- tube for the removal of the contents. This stomach-tube consists of a long, soft rubber tube about 75 cm. in length, having a lumen 6 to 7 mm. in diameter and provided with either two oval lateral openings or with three, one being at the end of the tube. Before introduction of the stomach-tube, it should be moistened with warm water and should be thoroughly cleaned. It is frequently a wise precaution, in general work where all classes of patients are to be examined, to have separate tubes which may be used by patients affected with tubercular or syphilitic con- ditions. While this may seem unnecessary, it is not impossible to bring about an infection in a patient free from these conditions. In those patients who require frequent washing-out of the stomach or frequent examination of the stomach contents, it is wise to order separate tubes for each. Fig. 14.-Stomach tube. Introduction of the Tube. The patient must be in a sitting posture, a towel or a rubber sheet being placed about his neck to prevent soiling of the clothes with the saliva or material which is occasionally brought up during the passage of the tube. False teeth should be removed and anything interfering with the passage of the tube should be avoided. In patients who are hypersensitive, a io per cent, solution of cocain is applied to the pharynx. The head of the patient is now bent slightly forward, never backward as some advise, and the mouth slightly opened, care being taken never to use a depressor on the tongue. The rubber tube, held as one would a pen, is passed gently backward over the tongue until its tip strikes the posterior wall of the pharynx, when it turns downward and may be readily introduced into the stomach, by slight forcing. As the tube reaches the esopha- 48 DIAGNOSTIC METHODS. gus, many patients complain of a sense of suffocation, which is not real but apparent. The tube interferes in no way with the normal respiratory move- ments and hence the patient should be cautioned to breathe normally and not forget to breathe. If the patient will swallow normally, the passage of the tube is greatly facilitated. It occasionally happens that highly nervous patients have great difficulty in swallowing this tube, so that it may be necessary to defer the withdrawal of the contents to a second or even a third period. It is never wise to excite a patient by forcing matters at any stage of the investigation. If any sign of cyanosis or marked pallor is evident the tube should be immedi- ately withdrawn and a second attempt made at some later time. When the tube has reached the floor of the stomach, which is in normal cases about 40 cm. from the incisor teeth, a distinct resistance to further passage of the tube will be noticed. This point should be carefully observed as the forcing of the tube beyond this point may produce rupture of the stomach wall or may cause the tube to "buckle." In this latter condition it will be impossible to withdraw the stomach contents. Many of the tubes used for gastric examination have a mark indicating the normal length of tube from the incisor teeth to the stomach wall, so that one has a definite idea when he has introduced the tube to the right point. In some cases the gastric juice will commence to flow from the tube as soon as it is properly introduced, but in the majority of cases some help is necessary to start the siphonage. Frequently all that is needed is to ask the patient to bear down with his abdominal muscles or to cough a little. In other cases aspiration is necessary. This may be done by the mouth, but this method does not seem advisable. Better prac- tice is to employ an ordinary Politzer bag or a Boas bulb for starting the fluid in the tube. This is very readily done by compressing the bulb and applying it, while compressed, to the end of the tube in such a way that the suction will be sufficient to draw the material into the tube. Once started, the material flows quite readily, but it may be necessary to use aspiration several times as the tube may become clogged with tenacious mucus or particles of food stuff. The material as it flows from the tube is collected in appropriate vessels and set aside for future work. If it is desired to wash out the stomach, either to obtain the total gastric contents or for the purpose of mere lavage, a funnel is attached to the external end of the stomach-tube and about 500 c.c. of water are allowed to flow through the tube into the stomach. In this operation the funnel is held either on a level with the patient's mouth or a very little bit above. By depressing and Fig. 15 -Turek's aspiration apparatus. GASTRIC CONTENTS. 49 inverting the funnel over a suitable vessel, before all the water has left it, return flow will soon set in and the stomach will be practically emptied by siphonage. In some cases it becomes necessary to add more water, but in no case should any be added after the patient complains of a feeling of distress. In collecting the stomach contents one should avoid as far as possible any admixture with the saliva which is more freely excreted at this time than normally. This is best done by wrapping a cloth about the tube so that the material may be absorbed as it runs along the side of the tube. After one has obtained the gastric contents, the tube is compressed with the fingers and is rapidly withdrawn, care being taken to keep up the compression so as to hold in the tube material which has not already passed into the receiving vessel. This residual material is added to the portion first received. In cases in which water has been introduced to wash out the stomach after the gastric contents have been obtained, one should be careful to note the amount of fluid poured into the stomach so that he may be able to judge of the amount again received. In this way only may he arrive at approximate results regarding the total acidity of the contents withdrawn. Not every case with which the practitioner meets is amenable to such manipulation. We find as especial contraindications to the use of the stomach- tube, uncompensated valvular lesions of the heart, arteriosclerosis, aneurysm, advanced pulmonary tuberculosis, marked emphysema, acute febrile diseases, severe hemorrhage, especially from ulcer or carcinoma, and excessively devel- oped nervous antipathy. Test Meals. As the secretion of the gastric juice is so dependent upon administration of food, it has become the custom to use certain combinations of food principles, which will excite gastric activity and enable us to obtain a juice which will give us more or less definite ideas of its composition in the condition investi- gated. It must be remembered that marked idiosyncracy toward certain foods exists, so that we may not use in all cases the same sort of a diet for ex- citing the gastric juice. The results obtained in pathologic conditions are compared with those obtained from normal individuals under the influence of the same diet. In this way we are able to say, with some degree of certainty, that a suspected case shows normal or abnormal gastric relations. These diets, the so-called test meals, are always given to the fasting stomach and are removed after a suitable time by the use of the stomach-tube. The time best suited for the administration of these meals is in the morning, as the stomach has had occasion during the night to empty itself of most of its contents. Ewald Test Meal. This meal, which is, perhaps, the most frequently employed in general work, consists of a roll or piece of bread or toast without butter and two cups of water or tea1 without milk or sugar. In approximate figures this will represent 1 Kober, Lyle and Marshall (Jour, of Biol. Chem., vol. 8, 1910, p. 95) have shown that it is almost impossible to detect blood in the presence of tannic and gallic acids, as found in tea. It is wise, therefore, not to use tea in the test meal when blood is suspected. 50 DIAGNOSTIC METHODS. 35 grams of wheat bread and 400 c.c. of water or tea.1 The bread should be well masticated so that the later withdrawal of the contents may not be inter- fered with by the plugging up of the openings in the tube. The contents are removed one hour later and consist normally of 30 to 50 c.c., depending both upon the skill of the operator and upon the condition of the stomach. Hyper- motility of the stomach will diminish the quantity of contents received, while a hypomotility will increase the quantity. Boas Test Meal. This meal consists of a dish of oatmeal prepared by concentrating to 500 c.c. a liter of water to which a tablespoonful of oatmeal is added. This meal was advised to prevent the introduction into the stomach of lactic acid which is a normal constituent of bread. While this small amount of lactic acid introduced in the Ewald meal has little significance, yet in doubtful cases it is well to avoid it. The contents of the stomach are withdrawn one hour later when the amount may be very small. If the stomach shows normal digestive powers most of the material will be then passed into the intestine, while an appreciable amount of material would indicate either a dilatation of the stomach or pyloric obstruction. Riegel Test Meal. This test meal has the advantage of permitting the patient to use a diet which is more normal than either of the ones previously mentioned. This diet is more important in America, where we are not accustomed to the con- tinental breakfasts, than it is in Germany where the Ewald and Boas meals are more usual. The Riegel meal is given in the middle of the day at a time when the patient is accustomed to such a meal. It consists of about 400 c.c. of soup, 200 grams of beef-steak, and either two slices of white bread or 150 grams of mashed potato along with one glass of water. This meal is withdrawn at the end of three to four hours. It has the advantage of allowing us to judge of the length of time which the food remains in the stomach under normal conditions and, also, to form an opinion of the rate and amount of digestion which has taken place. This meal incites a more nearly normal gastric juice than does the Ewald or Boas meal, but it is such that clogging of the stomach tube by particles of undigested food frequently occurs. Fischer Test Meal. This meal, introduced by an American physician, has the advantage of more nearly approaching an American breakfast than the others. It consists of the bread and tea of the Ewald meal along with a quarter of a pound of finely chopped lean beef broiled and seasoned. The contents are removed at the end of three hours. Fischer has shown by comparing results after his 1 Roberts (Jour. A. M. A., vol. 58, 1912, p. 753) adds to this breakfast 30 grams of lactose. He extracts the meal as usual and, then, washes out the stomach with 200 c.c. of water. The total gastric juice excreted and the acidity when excreted are calculated by the method of Matthieu (p. 53). GASTRIC CONTENTS. 51 meal with those of the Ewald breakfast that those with his are much more constant and somewhat higher than with the latter. Salzer Test Meal. This is in reality a double meal and is given as follows: For breakfast the patient receives 30 grams of lean cold roast meat, finely chopped, 250 c.c. of milk, 60 grams of rice, and one soft-boiled egg. Four hours thereafter a second meal is given, consisting of 35 to 70 grams of stale wheat bread and 400 c.c. of water. The contents are then removed one hour after this second meal. Under normal conditions of digestion and motility the stomach contents should show no remnants of the first meal. Sahli Test Meal. This meal was introduced to enable the worker to examine quantitatively the material withdrawn. The inconstant composition of the ordinary test meals makes it rather difficult to judge of the digestive power of the stomach. Sahli has introduced, therefore, a soup prepared as follows: Twenty-five grams of flour and 15 grams of butter are placed in a pan and browned over a fire. Three-hundred and fifty c.c. of water are then added and the whole boiled for five minutes (the loss in volume being replaced by fresh water) after which it is seasoned with a little salt. In this soup the fat is in the form of a very fine emulsion and the taste is so pleasant that a more nearly normal stimu- lus to gastric secretion is offered. The patient is now allowed to take 300 c.c. of this soup, while the remaining 50 c.c. are retained for a determination of the fat content. The contents are withdrawn, one hour after the meal, from the stomach which must have been thoroughly washed out prior to the administra- tion of the meal. We then determine the absolute amount of fat remaining in the stomach after the test digestion and compare this amount with that introduced. As we cannot be sure that the entire stomach contents have been withdrawn, we must know the residual amount of gastric juice. For this purpose one resorts to the method of Matthieu, which will be discussed later (p. 53). The amount of fat both in the original soup and in the withdrawn stomach contents is then determined and the total gastric juice calculated. This method of fat determination will be given in detail under Milk, to which the reader is referred. Calculation of Results. "The following calculations are possible from a consideration of the residue from the acidity of the gastric filtrate, and from the difference between the amount of fat found in the ingested flour soup and that found in the expressed contents. " By the addition of the value X, found in the calculation of the residue (p. 53), t0 ^e amount of contents expressed after one hour, there is obtained the volume of the contents which were actually present in the stomach at the end of that period. This we designate as To. From the absolute fat-content of 52 DIAGNOSTIC METHODS. To, there can be determined how much of the volume can be ascribed to the ingested flour soup. The amount of fat remaining in the stomach serves, there- fore, as a measure for the amount of soup remaining. This is designated as Su. Representing this mathematically, we have the proportion To: Su:: F: f, in which F represents the percentage fat-content of the soup and f that of the expressed contents. To-Su will give, of course, the volume of gastric juice in the expressed contents. If the acid-content of To has been determined, it is possible from these data to proceed further and to calculate what acidity was possessed by the pure gastric juice as it was excreted. Thus, if 75 c.c. of pure gastric juice are present in the stomach contents, whose volume amounts to 150 c.c. with 2 per cent, acidity, then the acid-content of the pure gastric juice is evidently 4 per cent. The determination of this acidity may be ex- pressed by the proportion, To: A:: Ma: a, in which To represents the amount of expressed contents including the residue, Ma the amount of secretion con- tained in the expressed contents (To-Su), A equals the acidity of the pure secretion in per cent, and a the percentage acidity of the expressed contents" (Sahli). It is important to remember in selecting a test meal for any given case that the tastes of no two persons are alike and that no two persons will react identically toward the stimulation of the same test meal. We should in every case endeavor to pay more attention to the administration of meals similar to those to which the patient is accustomed and, also, to give them at the time when such meals are ordinarily taken. The results can only be comparative and have, in this sense, some value. Too much rigidity in administration of such meals will lead frequently to mistakes in diagnosis, so that one should learn to vary his test meals rather than to rely upon a single one in all cases. Another point to be borne in mind is that the meals should be removed at the time of optimum secretion, which may not in all cases be at the end of one hour with the Ewald meal. For this reason a single examination of the stomach con- tents should not be implicitly relied upon in making a diagnosis. III. Macroscopic Examination. The gastric juice is a clear, colorless, easily filtered, levorotatory fluid having a distinctly acid reaction, an acid taste, and a characteristic odor. Its specific gravity, when the stomach is empty, ranges between 1004 and 1006.5; after the ingestion of food from 1010 to 1020 and more than 1020 when the production of acid is diminished (Landois). Its cryoscopic point is -0.38 degree to -0.444 (Roth and Strauss). Amount. The figures for the total amount of gastric juice secreted in 24 hours are variable. Beaumont gives 180 grams per diem while Bidder and Schmidt give a figure corresponding to about one-tenth of the body weight. The amount of fluid obtained one hour after an Ewald meal, is from 20 to 50 c.c., although larger amounts ranging from 200 to 500 c.c. indicate either diminished motilitv GASTRIC CONTENTS. 53 or hypersecretion, on the one hand, or dilatation associated with pyloric stenosis, on the other. It is to be remembered that the quantity of juice secreted is in- fluenced by the appetite and by the amount and character of the food taken, as well as by the age and sex of the patient and the time of day at which the food is taken. The largest amounts of gastric juice are found in cases of hyper- secretion when it is not uncommon to find a liter or more of gastric juice in the non-digesting stomach. In order to determine the total amount of gastric juice secreted, one cannot rely upon the quantity removed as there is always a slight residuum. The method of Matthieu and Remond is commonly used for such determinations. This gives results very nearly exact, at least for clinical purposes. With this method the gastric contents are removed, as nearly as possible, after an Ewald meal. A definite quantity of water, usually 300 c.c., is then poured into the stomach through the tube and is thoroughly mixed by moving the funnel up and down and by pressure upon the stomach. As much as possible of this added fluid and the remaining gastric juice is removed and collected in a sepa- rate vessel. The acidity of the undiluted as well as of the diluted stomach con- tents is then determined by titration. From the difference in these two values conclusions may be drawn as to the degree of dilution and to the residual amount of stomach contents which was not expressed. The amount expressed plus the residual amount equals the total gastric contents. The following is the method of calculation according to Matthieu: Let a = acidity of the undiluted gastric contents. Let b = acidity of the diluted gastric contents. Let x = amount of the test meal remaining in the stomach after expres- sion. Let 300 c.c. = the amount of water introduced into the stomach for dilu- tion. Then a : b : : x + 300 : x ax = b (x + 300) 300 b a - b An absolutely accurate result, in the study of gastric activity, can be ob- tained only when the total quantity of gastric juice is known. It is, therefore, necessary in stating, for instance, the acidity of a stomach contents to calculate the total available acidity rather than the mere degree of acidity. It is self- evident that a stomach contents expressed in the ordinary way, which shows an acidity of 400, may have this acidity in a total quantity of 50 c.c., while an acidity of 400, with a total quantity of 200 c.c., would represent actually four times as much hydrochloric acid available. It would seem, therefore, to the writer that the method of representing acidity in terms of degrees without any refer- ence to the amount of gastric contents is absolutely irrational. 54 DIAGNOSTIC METHODS. Color. Gastric juice is normally a practically colorless liquid, although at times it may be somewhat opaque and, therefore, much whiter in color. Variations in this colorless fluid are observed after test meals due to admixture of various food products, so that we may have distinctly brownish colorations due to the tea or particles of toasted bread, while in the test meals consisting of meat the color may be more of a reddish tone. Pathologically, we may find a distinct red color due to the presence of blood. This bright red color comes from the presence of fresh blood from a hemor- rhagic gastric ulcer or may be derived from abrasions of other portions of the alimentary tract. If the blood has been thoroughly mixed with the stomach contents for some time it may appear in the form of a brownish-black deposit, the so-called coffee-ground material. The blood in these cases is in the form of hematin and must be tested for as later outlined. This coffee-ground appearance is particularly evident in cases of gastric carcinoma. The color of the gastric contents may be either a yellow or a green, due to the presence of bilirubin in the former case and biliverdin in the latter. This biliary pigment should be detected by the tests outlined under Urine. The presence of bile in the gastric contents is indicative of duodenal occlusion. In cases of intestinal occlusion below the duodenum we occasionally find fecal matter in the gastric contents. This is characterized by the brownish- black coloration and by its intense odor. Odor. The normal gastric juice is practically odorless or very slightly sour. Ad- mixtures of material coming from the intestines cause a very intense odor, while the material rising from abscesses along the alimentary tract above the stomach will frequently give rise to a very offensive odor. In the vomitus ob- tained under various pathological conditions the odor may be very characteristic. Thus, in uremia we may find a distinct odor of ammonia, in alcohol intoxication a distinct alcoholic odor is evident, while in cases of stagnation of gastric con- tents an intensely strong odor is observed. In cases of dilatation we frequently find the organic acids so much increased in amount that distinct odors are noticeable. Consistency. The normal stomach contents are usually watery in character, but may vary due to admixture with extraneous material. After test meals or following a vomiting spell we may find portions of unchanged protein or carbohydrate material. The amount of bread taken with an Ewald meal should be so far digested in one hour as to form a puree-like mass which settles out on standing. Various food residues are, of course, present in the vomitus so that the consist- ency and appearance may give us much valuable information regarding the digestive process. In cases of mucous catarrh or in those showing either a diminution or an increase in the amount of hydrochloric acid we may find after GASTRIC CONTENTS. 55 a test meal the presence of large amounts of tough, slimy, mucoid material, which may be so abundant as practically to make it impossible to filter the con- tents. The consistency of such material may be almost that of a paste, or may be simply that of a thick syrup which on pouring from the vessel onto the filter will form distinct mucoid threads. The presence of an increased amount of mucus is of some diagnostic importance and should, therefore, be looked for under all circumstances. Gastric Contents from Fasting Stomach. The stomach is practically never empty, always containing a certain amount of acid fluid. Boas considers anything between io and 100 c.c. as a nor- mal amount of material for the fasting stomach. Anything above this amount would mean either motor insufficiency or hypersecretion. One may differen- tiate these two conditions by washing out the stomach at night, when the mate- rial withdrawn in the morning will be extremely scanty if the condition is one of motor insufficiency. Riegel regards any material in the fasting stomach as pathological. This fluid from the fasting stomach is thin, has a specific gravity of 1004 to 1005, contains some free hydrochloric acid, no lactic acid, and no bacteria. It is very commonly bile-stained, may be alkaline from the presence of pan- creatic juice and may contain large amounts of mucus. As such material is always found in the fasting stomach it is well to make it a rule to wash out the stomach the night before giving a test meal. Vomitus. In those cases which are associated with frequent vomiting we may obtain much valuable information from the examination of the ejected material. It is in these cases that one finds, frequently, much difficulty in passing the stomach-tube. The amount of material vomited will depend, of course, upon the motility of the stomach. In cases of dilatation or of stenosis we frequently find two or three quarts of material, while in conditions associated with hyper- motility we may have simply a scanty highly mucoid vomitus. The presence of food particles will give much information as to the digestive power of the stomach. If undigested meat fibers are found in the vomitus, ejected three hours after eating, one may assume more or less disturbance of protein diges- tion. If particles of unchanged bread are found, three hours after taking, the disturbance in protein digestion is probably more marked than in the case of the meat fibers. If an individual vomits bits of food more than seven hours after a meal, some impairment of motility must exist, according to Sahli, for after that interval even a hearty meal should have completely left the stomach. The vomiting of an acid liquid containing no food particles is quite character- istic of hypersecretion of gastric juice. The degree of acidity of the vomitus as well as the amount of hydrochloric acid present very frequently enables us to judge of the activity of the juice. 56 DIAGNOSTIC METHODS. These figures will not be as reliable as are the ones obtained after a test meal, but may serve in cases in which the stomach-tube cannot be passed. Frequently one finds a vomitus which is quite foamy and smells strongly of the volatile fatty acids. In such conditions we may assume a diminution in the amount of hydrochloric acid, which normally prevents the occurrence of any such decomposition, or we may ascribe this condition to simple stagna- tion of the gastric contents. Such contents will show microscopically the presence of large numbers of sarcinae ventriculi, yeast fungi, and various bacteria. The blood in the vomitus varies from a slight streaking of the material to a fluid which shows intimate mixing with the gastric contents. In cases of re- cent hemorrhage, which is particularly common in ulcer of the stomach, an abundant admixture of fresh arterial blood or of dark coagulated blood is observed. Brown or black coffee-ground-like material is particularly suggestive of carcinoma, although the same condition may result from erosion of the gastric mucous membrane when associated with hyperacidity or hypersecretion. An admixture of bile, producing a yellowish or greenish discoloration, may occur with any type of vomiting, but more especially from an empty stom- ach and in that associated with duodenal obstruction. A biliary vomiting is frequently observed in peritonitis and may be due to the fact that there is no counterpressure from the gastric contents to prevent regurgitation from the duodenum. This green vomitus may not always be due to the presence of bile, but may come from contamination with various chlorophyll-containing organisms. Almost all types of vomitus contain mucus. In some cases we find abun- dant tough, slimy masses which seem to be indicative of mucous catarrh of the stomach or of a diminution in the amount of hydrochloric acid. Fecal vomiting is a sign either of complete motor insufficiency of the intes- tine as found at times in peritonits, or indicates intestinal obstruction, either in the lower part of the small intestine or in the large bowel. The brownish- black color of this vomitus and the distinct odor render it very characteristic. Asiatic cholera and cholera nostras are associated with a vomitus which is abundant, alkaline in reaction, contains white flakes of mucus and epithelial cells, and large numbers of bacteria, both Koch's spirillum and the Finkler- Prior spirillum and various other unidentified types. This vomitus of cholera is known as the 11 rice-water" vomitus. The time at which vomiting occurs is frequently of great importance from the diagnostic standpoint. If it be at the height of digestion and during in- tense pain the condition is probably one of ulcer. If during or shortly after eating we may have either gastritis, a neurosis, or cancer. If it is frequent in the morning before breakfast and seems to be independent of eating the con- dition is probably one of dilatation. While these statements are not infallible, yet they are applicable in the majority of cases. GASTRIC CONTENTS. 57 Gastric Contents After Test Meals. The amount of material obtained after a test meal has some diagnostic importance. As previously stated, one obtains after an Ewald or Boas meal from 20 to 50 c.c., of contents, but these figures may vary to as high as 500 c.c. Hypersecretion or motor insufficiency are the chief causes of such increased amounts, the former being more probable if a large amount of free hydrochloric acid is present along with the excessive amount of fluid. A larger proportion of solid undigested material is observed in cases of pure motor insufficiency, but we frequently have a combination of both conditions. Absolute proof of the diminished motility is found in the presence of more than a trace of food in the stomach seven to eight hours after a meal. The general appearance of the material obtained after a test meal will be practically those previously discussed. IV. Microscopic Examination. The microscopical examination of the gastric contents is usually made on material withdrawn from the stomach after test meals, but the vomitus is oc- casionally examined. The gastric juice is practically never free from remnants of food, such as meat threads or starch granules, although nothing has been taken for many hours. Moreover, small masses of mucus, which occasionally assume a snail-like spiral form, and saliva which is recognized by the presence of large flat epithelial cells and the so-called salivary corpuscles are quite frequent. A few bacilli and yeast cells are almost always observed. As such elements as the ones above mentioned are present in all gastric juice, we must not attach undue importance to the presence of small amounts of such material in the contents obtained after a test meal. After the Ewald meal one rarely finds anything beyond the presence of numerous starch granules and more or less mucoid material, along with bacteria of the various types which flourish particularly in the buccal and gastric cavities. In cases associated with diminished motility of the stomach we may find remains of food which has been introduced many hours previously. In such specimens we observe numerous fat globules or fatty acid crystals, many vege- table fibers and cells and a few red blood-cells which have come from slight abrasion of the mucous membrane of the pharynx by the stomach-tube. These red cells usually are much altered in appearance by the hydrochloric acid and do not show their ordinary hemoglobin color, but take on a more brownish tint, which is due to the presence of hematin. Boas-Oppler Bacillus. This organism is found quite commonly in patients suffering with carci- noma of the stomach, and is almost always absent in nonmalignant disease. It is found more frequently in the gastric contents at a time when lactic acid is present in large amounts, so that in the incipient stages of carcinoma these 58 DIAGNOSTIC METHODS. organisms may be absent. These bacilli are very long (3 to 10 microns), 1 micron broad, and are frequently joined end to end forming very long chains. They are readily stained with the usual aniline dyes and by Gram's method and, on treatment with iodin, take on a brown color which distinguishes them from the large mouth bacillus (leptothrix buccalis), which stains blue with iodin. This organism is not absolutely pathognomonic of carcinoma of the stomach, but is found in 75 to 85 per cent, of all cases, being rarely present in dilatation or benign stenosis of the pylorus. Sarcinae. Fig. 16.-Boas-Oppler bacilli. (Hemmeter.) Occasionally in normal gastric juice and especially in cases of dilatation with marked fermentation one finds the so-called sarcince ventriculi which are cocci arranged in squares or tetrahedra which resemble, very much, cotton bales. These organisms have no pathologic significance, but are indicative of stagnation of gastric contents. Along with these sarcinas one may find large numbers of yeast cells. Protozoa. These unicellular parasites have been occasionally found in the gastric contents. Flagellates, amebas, and monads seem to be more frequent than the other types of protozoa.1 They seem to be more commonly found in cases of carcinoma of the stomach; quoting from Simon, "from the available data there can be no question that the presence of protozoa in the stomach contents is sug- gestive of nonobstructive carcinoma." Fragments of Tissue. Frequently small shreds of mucous membrane are found in the expressed gastric contents. One finds these in cases of chronic gastritis, ulcer, hyper- chlorhydria, and especially in cancer. These tissue fragments should be studied carefully under the microscope, as not infrequently a diagnosis of cancer is pos- sible from such examination. 1 See Smithies, Amer. Jour, of Med. Sci., vol. 144, 1912, p. 82. GASTRIC CONTENTS. 59 Crystals. Various types of crystal are occasionally noted in the gastric contents, among which may be mentioned bile acids, cholesterin, fatty acids, leucin, tyrosin, and calcium oxalate. If the reaction of the juice is alkaline triple phosphate crystals may appear. V. Chemical Examination. The chemical examination of the gastric juice is the most important of all laboratory methods in the diagnosis of various pathologic gastric conditions. As previously stated, the acidity of gastric juice is referable to the presence of free and combined acids. The free acidity is traceable largely to hydrochloric acid, although organic acids, such as lactic, acetic, and butyric, may increase the free acidity under abnormal conditions. Besides this free acidity, we have hydrochloric acid which is bound chemically to the protein substances and does not react with tests for free acidity. There are also present in the gastric juice acid salts, especially the sodium dihydrogen phosphate (NaH2PO4). Besides these factors which have to do with the reaction of the gastric juice, we find certain ferments which act only in the presence of the free hy- drochloric acid.1 The first of these, pepsin, has the power of acting upon al- bumin in an acid medium and converting it, through various stages, into lower splitting products of albumin. This peptic digestion will be discussed in detail later. A second ferment, known as rennin, lab, or chymosin, has the power of curdling milk by coagulating the casein. A third ferment, lipase, acts upon fat, especially when this is present in a finely divided form. This lipolytic action is not ordinarily great, but should nevertheless be remembered. The experiments of Sahli show that this action is negligible during the period covered by the administration of his test meal. Although many statements have been made to the contrary, the gastric juice, through the agency of the hydrogen ions of its free hydrochloric acid, acts upon certain polysaccharides, especially cane sugar, hydrolyzing them into the simpler monosaccharides. Careful determinations indicate that the speed of inversion is about the same as that of an equal strength of hydrochloric acid, so that we do not need to assume any ferment action. Besides these substances gastric juice contains a small amount of albumin, carbohydrates, and various inorganic salts. None of these constituents have any importance from a clinical standpoint and will be disregarded. (1). Total Acidity. As previously stated, the total acidity of the gastric juice is referable to the presence of free and combined hydrochloric acid, organic acids, and acid salts. This factor may be readily determined by titrating 10 c.c. of gastric juice with tenth-normal sodium hydrate, using phenol-phthalein as an indicator. This indicator is colorless in the presence of acid and becomes red at the point of neutralization, being used as a 1 per cent, alcoholic solution. On adding a 1 See Ewald, Deutsches Arch. f. klin. Med., Bd. 106, 1912, S. 498. 60 DIAGNOSTIC METHODS. few drops of this solution to the filtered gastric contents, a white cloud will be observed due to the precipitation of the reagent by the water of the gastric juice. The titration is carried to the point at which the addition of sodium hydrate produces a definite pink color which remains permanent and does not deepen on the addition of further alkali. If sodium chlorid be added to the point of saturation of the gastric contents, the end point becomes somewhat sharper owing to the diminished dissociation which the disodium hydrogen phosphate undergoes into sodium dihydrogen phosphate in the presence of increased sodium ions. This precaution is rarely taken, however, as the clinical result is never so accurately determined as is the scientific factor. The total acidity varies between rather wide limits. Normally it ranges from 75 to ioo°, being made up of approximately 5o°of free hydrochloric acid, 250 of combined hydrochloric acid, and 250 of organic acids and acid salts. The chief variation under normal conditions is an increase in the combined hydrochloric acid and a decrease in the organic acids and acid salts. In pathologic conditions we may find the total acidity high with very little free hydrochloric acid, or we may find the total acidity low, with a normal amount of hydrochloric acid present. Sahli has stated the variations in the acid factors of the stomach contents as follows: If the total acidity is high and the hydrochloric acid is normal, the high acidity can be due only to a deficient motility and absorption and hence we find an increase in the organic acids. Such a gastric juice may show lactic acid, but will more probably give the tests for the other organic acids. A low total acidity with an excess of hydrochloric acid shows that the motility and absorptive powers of the stomach are good. If the total acidity be mod- erate and free acid small in amount, a poor motility may be assumed. Gen- erally speaking, when much lactic acid is present we find low HC1 and only combined HC1; that is, diminished secretion and diminished motility. No lactic acid is found when the HC1 is normal or increased. (2). Free Hydrochloric Acid. A number of tests have been devised for the detection of free hydrochloric acid in the stomach and its differentiation from lactic and acetic acids. It should be remembered that the tests outlined below are not specific tests for hydrochloric acid, but are common to all mineral acids. Many of these tests react also with the organic acids, providing they are present in sufficient con- centration. The efficiency of any acid is due merely to the ionic decomposition which it suffers when in solution; in other words, is due to the presence of free hydrogen ions. Hydrochloric acid appears to be more efficient in the digestive processes than do the organic acids, owing simply to its greater degree of disso- ciation. The tests commonly employed in clinical work are based upon the reaction which certain coloring matters undergo when treated with free hydro- chloric acid. These tests can be, therefore, only approximate and must be used with discretion in scientific work, although in clinical work they are near GASTRIC CONTENTS. 61 enough for all purposes. Were we able to express completely the stomach con- tents and thus to obtain material which would give us absolute data, we would then require better clinical methods. What should be measured in testing stomach contents for free acidity is the number of hydrogen ions which the gastric juice contains, since this is the important factor in the efficiency of the juice. This may be determined directly by measuring the speed of any cata- lytic reaction due to hydrogen ions, the most convenient one being the rate of inversion of cane sugar. Several indirect methods have been devised for deter- mining the presence of free hydrochloric acid as distinguished from free organic acids, combined hydrochloric acid, or acid phosphates. It is wise to have some quick method by which one may determine the presence of a free acid in the gastric juice. This may be done with litmus which will, however, not show the presence of free hydrochloric acid to the ex- clusion of other acids or acid salts. To determine whether the acid reaction is due to free acid it is customary to employ the Congo-red paper. This con- sists simply of filter-paper which has been saturated with an alcoholic solution of Congo-red and dried. On treating gastric juice with such paper we obtain a blue color in the presence of free acid. This paper reacts with a blue color to any free acid so that one should never assume the presence of free hydro- chloric acid when he obtains a blue coloration. Although many writers state that a blue coloration is never given by free organic acids, the writer has seen too many cases in which distinct action was referable either to free lactic or free acetic acids to agree with this statement. The tests which are applicable to the detection of free mineral acids in general may be used as indicative of free hydrochloric acid, as this is the only mineral acid which one would ordinarily find in the gastric contents. Qualitative Tests. (a). Tbpfer's Test. The test is based upon the coloration which a 0.5 per cent, alcoholic solu- tion of dimethyl-amido-azobenzol takes when treated with gastric juice contain- ing free hydrochloric acid. A few c.c. of filtered gastric juice are placed in a dish and one or two drops of the above solution added. In the presence of free mineral acids a carmin red color is obtained. This reagent is a very deli- cate one and does not react to organic acids unless they are present in amounts exceeding 0.5 per cent. The coloration with free hydrochloric acid varies in intensity with the amount of acid present, and may range from a deep orange to an intense carmin. According to Simon, lactic acid does not give the typical red color with this reagent, especially if albumoses are present, unless it be in a concentration of at least 1 per cent. This reagent will detect the presence of 0.02 parts of hydrochloric acid per thousand. (&). Gunzburg's Test. The reagent employed in this test consists of 2 grams of phloroglucin and 1 gram of vanillin dissolved in 30 c.c. of absolute alcohol. This yellowish solu- 62 DIAGNOSTIC METHODS. tion should be kept in dark bottles, as it gradually changes to a dark red and then to brown when exposed to the light. Boas claims that the reagent becomes more delicate and stable if one dissolves the phloroglucin and vanillin in 100 c.c. of 80 per cent, alcohol. Two or three drops of this solution are added to an equal amount of the gastric juice contained in a porcelain dish and the mixture evaporated over a a water-bath. In the presence of free mineral acid a rose-red color is developed, varying in intensity with the amount of acid present. This mixture must not be boiled or heated too rapidly as the resulting color will then be brown or brownish-red and may mislead one into believing that no free hydrochloric acid is present. This test does not react to organic acids or to acid salts, nor is it interfered with by the presence of products of food digestion. It may, therefore, be used with the unfiltered gastric juice. This test reacts with the rose-red color in the presence of 0.05 parts of HC1 per thousand.1 This reagent consists of 5 grams of resorcin and 3 grams of cane sugar dissolved in 100 grams of 95 per cent, alcohol. It has the same delicacy as Gunzburg's test and is more stable. The test is applied in the same way as the preceding, taking particular care to use a low flame in evaporating, and gives a rose-red or vermilion color in the presence of mineral acids. This color gradually fades on cooling and is not given by organic acids or acid salts. (c). Boas' Test. The reagent for this test is a saturated alcoholic solution of tropeolin 00. This test is applied in the same way as the preceding and gives a lilac-blue color in the presence of free acid. This test is not as delicate as the preceding, re- acting only in the presence of 0.3 parts of free hydrochloric acid per thousand, and has the objection that it strikes the blue color much more easily with the organic acids. (d). Tropeolin Test. Other tests have been advocated for the qualitative detection of free hydrochloric acid, but they are not as delicate as the above and have nothing to justify their existence. Of the tests given, the Gunzburg test seems to be the most reliable, although the Tbpfer's test is clinically sufficient and has the advantage of being much less expensive and more convenient than is the for- mer reagent. The quantitative estimation of free hydrochloric acid is of great importance in the study of all pathologic conditions of the stomach. Any determination made with our present methods must have reference to the fact that the acidity of the gastric juice is due almost entirely to the free hydrochloric acid. While this is not absolutely true, yet the organic acids are rarely present in sufficient amounts to react with the indicators in conditions in which hydrochloric acid Quantitative Methods. 1 See Christiansen, Biochem. Zeitsch., Bd. 46, 1912, S. 24, 50 and 71. GASTRIC CONTENTS. 63 is normal or increased in amount. However, in the conditions associated with diminished amount of hydrochloric acid one must be on his guard in the inter- pretation either of the qualitative or quantitative tests for free hydrochloric acid. The writer has seen several cases in which all of the indicators, with the excep- tion of Gunzburg's reagent, showed positive results for hydrochloric acid, the cases being those in which no free hydrochloric acid was actually present. Bear- ing this in mind we may determine the acidity, referable to free hydrochloric acid, by the titration of a known amount of the filtered gastric juice with tenth- normal sodium hydrate solution, using as indicators the solutions mentioned under the head of qualitative tests.1 As the Gunzburg reagent is the most delicate and reliable of all the tests for free hydrochloric acid, it is wise to use this reagent as an indicator. The test as usually followed by the writer is to add 20 to 30 drops of the Gunzburg reagent to 10 c.c. of the gastric juice. On adding the sodium hydrate solution no color change will be visible, as the reaction takes place only when the solu- tion is warmed. This warming cannot be done directly as the evaporation would necessarily have to proceed to the point at which loss of hydrochloric acid might occur. Following the recommendation of Sahli, the rod with which the solutions are stirred is warmed before being used. A distinct red color will be evident along the sides of the rod as the neutralization point is reached. The older method of Fleiner, consisting in the removal of a drop or two of the gastric juice after the sodium hydrate solution had been added and the evaporation of this mixture in a small dish, is too time-consuming and introduces the error of loss of substance at each trial. The writer would recommend, therefore, either the use of the warm glass rod or a combination of the Congo-red paper as an approximate guide to the point of neutralization and a determination of the final result with the Gunzburg method. Mintz Method. Tdpfer's Method. This method is the simplest, is the most generally used, and at the same time is one of the most delicate of all the clinical quantitative methods for free hydrochloric acid. It consists in the use of dimethyl-amido-azobenzol as an indicator, the titration of the filtered gastric juice being done with tenth- normal sodium hydrate solution. Although this reagent does give, under some conditions, a red color in the absence of free hydrochloric acid when the organic acids are largely increased, yet such conditions are so rarely found in clinical work that the result of test-tube experiments cannot be applied to clinical cases. The experiments of Simon show that lactic acid must be present to the ex- tent of 1 per cent, before any cherry-red color is obtained, providing albu- moses are present. As these latter substances are always found in the gastric juice after intake of protein material, one can readily see that lactic acid need not be considered. We find, however, conditions associated with 1 See Knapp, New York Med. Jour , vol. 97, 1913, p. 437. 64 DIAGNOSTIC METHODS. fermentative processes in the stomach in which acetic and butyric acids are present in fairly large amounts. These acids will give a red color with the indi- cator, but should not mislead as their strong odor in such concentrations permits of easy recognition. In the titration the sodium hydrate solution is added from a buret to the filtered gastric juice to which one or two drops of indicator are added for every io c.c. of juice. In the presence of free hydrochloric acid this indicator strikes a distinct cherry-red tone and thus enables the worker to decide at once as to the presence or absence of the acid. No evaporation is necessary, hence the test has the advantage of simplicity and does not occasion any loss of substance. As the sodium hydrate solution is added the reddish tint of the mixture changes to a distinct yellow. The titration must be carried to the point at which every trace of red disappears and the color becomes a pure yellow. This reaction requires considerable experience, hence the writer would advise the student to make his titration with a known solution of hydro- chloric acid so that he may become familiar with the end point. Much confusion exists in the literature regarding the use of filtered or of unfiltered juice for these titration tests. It is the writer's custom to use the filtered contents, as an exact measurement of the quantity of j uice taken can rarely be made otherwise, owing to the presence of food particles which will raise the meniscus in the measuring glass to the desired point without giving in reality an actual measurement of the amount of true juice desired.1 As the hydrochloric acid is in solution what should be tested is a known quantity of the solution and not a mixture of the solution with the food particles. Another source of error in using the unfiltered gastric juice is the possibility of the combination of the sodium hydrate used in titration with the undigested protein material, thus giving a somewhat higher result than is really present. Using the Gunz- burg reagent, the results are lower than with the Topfer reagent, as the removal of the substance in the former test introduces a distinct error. Amount of Free Hydrochloric Acid. The researches of Ewald, Szabo, and Boas show that free hydrochloric acid is present in the stomach to the extent of two to three parts per thousand (0.2 to 0.3 per cent.). This amount of acid is present only at the height of digestion as the hydrochloric acid first secreted is bound to the protein material in the form of combined hydrochloric acid, so that if a stomach contents be examined at the height of digestion and no free hydrochloric acid be found, one may assume that a disturbance of this function of the stomach is present. The time at which this physiologic excess will appear depends upon the kind of food upon which the gastric juice has acted. After an Ewald breakfast this excess should be present in from 45 to 60 minutes, while after a Riegel meal in from two to three hours. These points should be borne in mind as the re- moval of such meals before these periods will always give a deficit in the amount of free hydrochloric acid. 1 See, however, Christiansen, Biochem. Zeitsch., Bd. 46, 1912, S. 82. GASTRIC CONTENTS. 65 Certain clinical conditions lead to an excretion of gastric juice which is normal, increased, or diminished in amount of free hydrochloric acid. We must, therefore, have a method of interpretation of the free HC1 acidity of the stomach. It is customary to report the results of titration of a given specimen of gastric juice in one of two ways: (i). We may represent the acidity referable to free hydrochloric acid by the number of c.c. of tenth-normal sodium hydrate necessary to neutralize 100 c.c. of filtered gastric juice, using dimethyl-amido-azobenzol as an indicator. This is called degree or percentage of acidity. Thus, if 5 c.c. of tenth-normal sodium hydrate were used to neutralize 10 c.c. of filtered gastric juice, the de- gree or percentage of acidity would be, obviously, 50. (2).. We may report the actual amount of free hydrochloric acid present. This is the more scientific way, as we have a much better means of comparison with the normal standards. One c.c. of tenth-normal sodium hydrate neutral- izes 0.00365 gram of free HC1. If, now, we multiply this factor by the number of c.c. necessary to neutralize 100 c.c. of filtered gastric juice, we obtain a figure representing the absolute amount of free HC1 in the gastric content. Thus, if the gastric juice showed an acidity of 50 degrees, we would have 0.1825 (50 x 0.00365) gram of HC1. The usual text-book statement is that the free hydrochloric acid is normally about 40 degrees, but the writer is accustomed to consider an acidity of 50 to 550 as much more nearly normal than one of 40. Euchlorhydria. This is a condition in which the amount of free hydrochloric acid is between o.i and 0.2 per cent. In speaking of percentage one must not confuse the two types of percentage reckoning. Thus, io per cent, or io° of HC1 rep- resents only 0.0365 gram or true per cent, of hydrochloric acid. The lower figure of 0.1 per cent, is given for euchlorhydria owing to the fact that normal variations may permit of this low point, although the higher figure of 0.2 per cent, is the more usual. When this euchlorhydria exists in the presence of clini- cal symptoms pointing to gastric disturbance, we usually have to do with a neurosis. Gastritis may be absolutely ruled out, a carcinoma excluded except when the new growth has taken place upon an old ulcer and an ulcer practically always ruled out, although this latter may show an euchlorhydria. This con- dition may be associated with a certain amount of atony along with more or less marked dilatation. Hypochlorhydria. This is a condition associated with excretion of the gastric juice showing o.i or a lower per cent, of hydrochloric acid. It is found especially in subacute or chronic gastritis, in incipient carcinoma, in fevers, severe anemias, many mental diseases, passive congestion due to valvular heart lesions, many cases of chronic nephritis, and dilatation of the stomach. Some rare cases of ulcer of the stomach show a low degree of free hydrochloric acid, but this is not usual. 66 DIAGNOSTIC METHODS. Anachlorhydria. This is a condition characterized by the excretion of the gastric juice show- ng the complete absence of free hydrochloric acid. This condition has been supposed to be pathognomonic of gastric carcinoma. There are, however, many cases of cancer which show either a hypo- or a hyperchlorhydria, and there are, also, many other conditions which show an anachlorhydria. Thus we find this condition in a large majority of cases of advanced chronic gastritis, in the severe anemias, especially of the pernicious type, in neurasthenia and hysteria, in many severe febrile diseases, and in atrophic gastritis. Hyperchlorhydria. This conditions exists when we have the excretion of a gastric juice showing more than 0.2 per cent, hydrochloric acid. This figure may run anywhere from 0.2 to 0.9 per cent. It is a very common occurrence in nervous individuals, in ulcer of the stomach, some cases of chlorosis, in some chronic cachexias, in the early stages of chronic gastritis, in carcinoma which is grafted on to an old ulcer, in continuous hypersecretion, in chronic passive congestion of the stomach, and in cases of migraine. It will thus be seen that the hydrochloric acid of the gastric juice varies under clinical conditions to quite an extent. While an increased amount of hydrochloric acid is usually present in ulcer of the stomach, we must not neces- sarily make our diagnosis on this point alone. Likewise in carcinoma of the stomach we should not exclude this condition if the examination of the stomach contents does not show a lessened amount of hydrochloric acid or even a total absence, as many cases of carcinoma may show all varieties of chlorhydria. It is to be said that here as in practically all other laboratory examinations a single determination is not conclusive. A series of examinations should be made so that one may have comparative figures. In this way only may one be sure of his ground. (3). Combined Hydrochloric Acid. As the hydrochloric acid first secreted by the gastric glands combines with the protein of the food material, it is necessary to have some method by which we may determine just how much of this material has been formed in the stomach. The physiologically active hydrochloric acid consists of both the free and combined acid so that we may have only a slight amount of the free acid, but a relatively large amount of the combined. Not infrequently we find cases which show no free hydrochloric acid, but quite a percentage of the com- bined acid, indicating that a certain amount of acid has been secreted by the gastric juice. This combined hydrochloric acid has, therefore, a certain clinical importance and should be investigated in every case. No direct methods are known for its determination so that we must resort to indirect methods. Method of Martins and Liittke. This method, as modified by Reissner, is much more applicable to scientific work than to clinical investigations, as it is too complicated and time-consuming GASTRIC CONTENTS. 67 for the ordinary practitioner. It is based upon the facts that the free hydro- chloric acid as well as the acid combined with protein material escape upon in- cineration of the gastric juice, while the inorganic chlorin in combination with inorganic bases remains in the ash. If the total amount of chlorin present in the filtered gastric juice be determined (a), and then the amount of chlorin in the ash (b) investigated, subtraction of the latter (b) from the former portion (a) will give the amount of chlorin referable to free and combined hydrochloric acid except a small loss referable to volatilization of ammonium chlorid. If the gastric juice be neutralized with sodium hydrate before it is incinerated and the chlorin in this ash determined, the amount of this chlorin (a') subtracted from (a) represents the ammonium chlorid volatilized. Hence a'-b equals the free and combined HC1. By now determining the amount of free hydro- chloric acid according to Tbpfer's method we may at once calculate the combined hydrochloric acid by subtracting this result from the amount of free and com- bined hydrochloric acid previously obtained. The method of determining the chlorin will be fully discussed in the section on urine to which the reader is referred. Method of Topfer. This method embraces three separate determinations. In the first place the total acidity of the gastric juice is determined by titration of xo c.c. of filtered gastric juice with tenth-normal sodium hydrate, using phenolphthalein as an indicator. This result is termed a. This indicator has the advantage of re- acting toward anything of an acid nature, and will give us, therefore, the dif- ferent factors which go to make up the total acidity of the gastric juice, namely, the free and combined acids, the organic acid, and the acid salts. Having determined this factor (a), a second portion of io c.c. of gastric juice is titrated with tenth-normal sodium hydrate solution using a i per cent, aqueous solution of alizarin (alizarin monosulphonate of sodium) as an indi- cator. Two or three drops of this indicator are added to io c.c. of filtered gastric juice when the mixture becomes distinctly yellow. The titration is carried to the point of production of a pure violet color which does not deepen on the further addition of alkali. As this reaction demands the recognition of a change from yellow through a faint violet to a deep violet color the worker must have considerable practice before he is able accurately to determine the end point. The result is termed &. No trace of red should be present, a pure violet color being the true reacting point. This color may be observed by treating a few drops of alizarin solution with a i per cent, solution of sodium carbonate. Alizarin reacts with free acid, both mineral and organic, and with acid salts, but not with organically bound HC1. If, therefore, we subtract the figure obtained when alizarin is used as an indicator (&) from that obtained with phenolphthalein (a) the result will be the combined hydrochloric acid (a - b) If now we add this combined hydrochloric acid to the free hydrochloric acid, which has been obtained by titration of the gastric juice using dimethyl- 68 DIAGNOSTIC METHODS. amido-azo-benzol as an indicator, (c), we obtain the total physiologically active hydrochloric acid {c+^a-b}). The difference between the total acidity and this factor gives us the amount of organic acid and acid salts present, (a - (c + (a - b) ). If but a small amount of gastric juice be available for chemical ex- amination, recourse may be had to a modification suggested by Einhorn. This is a double titration of the same portion of juice. A few c.c. (5) of filtered gastric juice are treated with a few drops of dimethyl-amido-azobenzol and the solution titrated for free HC1 with sodium hydrate. When the point of neutrali- zation of the free acid is reached a few drops of phenolphthalein solution are added and the titration continued to the point of neutralization of total acidity. These indicators do not interfere at all with one another as their reacting points are usually widely different. The writer has found this method reliable and convenient. Hydrochloric Acid Deficit. In those cases in which the gastric contents shows no free hydrochloric acid, it is customary to determine the HC1 deficit. By this is meant the amount of hydrochloric acid which must be added to the gastric contents before it shows a reaction for free acid. This amount will naturally depend on the amount of combined HC1 already present, the amount of protein in the gastric contents and the amount of alkali secreted. Sahli suggests the expression "saturation deficit" for this figure. Ten c.c. of unfiltered gastric juice are titrated with tenth-normal hydrochloric acid, using dimethyl-amido-azobenzol as an indi- cator and titrating to the point of production of the red color. The result is expressed in terms of degrees as under the representation of the free hydro- chloric acid. This factor enables one to follow the course of the disease, showing how little hydrochloric acid is excreted for combination with the proteins of the food. (4). Organic Acids. The organic acids, outside of the lactic acid, have very little clinical signifi- cance. The food practically always contains a certain amount of fatty acids which appear in the stomach contents and contribute to the total acidity. In the normal digestion of the carbohydrates, lactic acid is practically always formed, so that excess of this acid would indicate excessive fermentative proc- esses in the stomach, due to a combination of diminished amount of hydrochloric acid along with a lessened motility of the stomach. Other acids, such as buty- ric and acetic acids are formed in this same process of carbohydrate fermenta- tion, so that the organic acids may represent a large portion of the total acidity. Besides this, bacterial decomposition, in the absence of hydrochloric acid, plays a role in the production of these fatty acids. The fat-splitting ferment, lipase, may produce these organic acids in fairly large amounts. Total Organic Acid. It is sometimes of importance to know just how much organic acid is pres- ent in the stomach contents. This may be done directly by the Hehner-Maly GASTRIC CONTENTS. 69 method, which is based upon the fact that, if a mixture of organic and inor- ganic acids be neutralized and then incinerated, the organic acids will be con- verted into carbonates while the inorganic acids remain as neutral salts. If the alkalinity of these carbonates be then determined and this factor subtracted from the total acidity we obtain directly the organic acids. This is possible owing to the fact that the degree of alkalinity of the carbonates is equal, in terms of tenth-normal solutions, to the acidity referable to the organic acids. The technic is as follows: The total acidity of io c.c. of gastric juice is deter- mined by titration with tenth-normal sodium hydrate solution as previously described. The neutralized solution is evaporated to dryness in a platinum dish and is then incinerated. The ash is dissolved in distilled water and the alkaline solution titrated with tenth-normal oxalic acid solution. As i c.c. of the tenth-normal oxalic acid solution is equivalent to i c.c. of tenth-normal sodium hydrate solution, we subtract the factor obtained in the latter titration from that of the former and obtain directly the degree of acidity due to mineral acids. The number of c.c. of tenth-normal oxalic acid used represents directly the total organic acids present. In this method the acid salts are included in the factor referable to mineral acids, so that we may subtract from this factor the degree of acidity, attributable to free hydrochloric acid, and obtain the amount of acid salts present. In some cases fatty acids are present which are not soluble in water and consequently are not neutralized by the addition of the sodium hydrate solution. These acids may be extracted from the neutralized solution with ether and may then be neutralized and added to the neutral aqueous solution. The mixture is now evaporated as before and incinerated. This estimation of the higher fatty acids requires the use of the unfiltered gastric contents. However, such acids do not play a large clinical role and may be ordinarily omitted. (a). Lactic Acid. The ordinary foods such as milk, bread, and meat contain a certain amount of lactic acid, so that any test for the presence of lactic acid can be of value only when the meal contains very little of such foods or when the portion taken in with the food has disappeared from the stomach. After the Boas meal there is always less lactic acid than after the Riegel meal, so that the former is much preferable when a special test is to be made for the presence of lactic acid. Boas has shown that under physiological conditions no appreciable amount of lactic acid is formed during the process of digestion. At the height of digestion practically no lactic acid is demonstrable in the stomach contents. This may be due to its absorption, on the one hand, or, on the other, to the fact that the hydrochloric acid interferes with the delicacy of the reactions. Pathologically, lactic acid is found in any condition associ- ated with stagnation of the gastric contents as a result of motor insufficiency, provided the amount of hydrochloric acid is below the normal amount. As this condition of affairs is found most frequently in cases of carcinoma of the 70 DIAGNOSTIC METHODS. stomach, an excess of lactic acid is very strongly suggestive of malignancy, although it must be remembered that such an excess may appear in cases of benign stenosis and gastric insufficiency. If the stomach be washed out the evening before giving a test meal, preferably the Boas meal, and lactic acid be found in appreciable amounts, carcinoma is the probable diagnosis. This finding of increased lactic acid and diminished hydrochloric acid is not always observed in every case of carcinoma of the stomach. In some cases periods of increased production of hydrochloric acid alternate with increased formation of lactic acid, and in some cases, especially those in which the carcinoma has developed upon the base of an old ulcer, no lactic acid may be present, but hydrochloric acid may be found in large amounts. Uffelmann's Test. This test is, perhaps, more commonly employed for the detection of lactic acid than is any other, but the writer prefers the Kelling test. Uffelmann's reagent consists of 20 c.c. of 1 per cent, carbolic acid solution, to which are added one drop of dilute ferric chlorid solution and sufficient water to form a transparent amethyst-blue solution. This solution is not permanent and must, therefore, be made fresh before each test. If a few drops of the filtered gastric juice be added to 5 c.c. of this reagent, the solution will be decolorized in the presence of lactic acid, taking on a beautiful canary-yellow or greenish-yellow tint. The mere decolorization of this solution is not sufficient for a positive test. A pure lemon yellow or canary color must be present before one may assume the presence of lactic acid. Even when this color appears one must eliminate such factors as the acid sodium phosphate, cane sugar, glucose, alcohol, and various organic acids, such as tartaric, citric, or oxalic, before he can say that lactic acid is present. A considerable excess of hydrochloric acid in the gastric juice may prevent the appearance of this color and likewise a yellowish tint of the stomach contents may obscure the result. Under such conditions it is necessary to extract the gastric contents with ether, which takes up the lactic acid, leaving the interfering substances behind. The ethereal solution is then evaporated, the residue taken up with distilled water, and the test applied to this solution. Kelling's Test. This test is in reality a modification of the previous one and consists in the addition of a few drops of filtered gastric juice to a very dilute solution of ferric chlorid. As used in the writer's laboratory, the method is as follows: To a test-tube full of water are added one or at most two drops of a io per cent, solution of ferric chlorid. The mixture is thoroughly shaken and divided into two portions, one of which serves as a control. On now adding a few drops of filtered gastric juice to one of these portions a distinct canary-yellow color will appear in the presence of lactic acid. The color of the two solutions should be compared so that any change in the one, to which gastric juice was GASTRIC CONTENTS. 71 added, may be observed. This test has the same objections as the Uffel- mann test, so that it is always wise to extract the gastric juice with ether before applying either one of these tests. Strauss' Method. This method is, perhaps, the very best clinical method at our disposal, as it shows lactic acid when present in pathological amounts. It does not, however, give a quantitative result, nor does one seem necessary in the ordinary clinical work. Into a special separatory funnel (see cut) are introduced 5 c.c. of the gastric juice. The funnel is then filled to the 25 c.c. mark with alcohol-free ether and well shaken. The ethereal layer will take up the lactic acid from the gastric contents. After the fluids have settled the gastric juice and ether are allowed to run out to the mark 5 by opening the stop-cock, after which distilled water is added to make up the 25 c.c. volume. Two drops of 10 per cent, ferric chlorid solution are then added with a medicine dropper and the mixture well shaken. The water will now extract the lactic acid from the ether. The aqueous layer is colored an intense greenish-yellow if more than 0.1 per cent, of lactic acid is present, while smaller amounts will show a slight greenish tinge. This test may be negative if the lactic acid present is completely combined with the proteins of the gastric juice. In such cases hydrochloric acid may be added to liberate this lactic acid before shaking out with ether. Other qualitative tests as well as several quantitative tests have been given for lactic acid. Quantitative deter- minations of lactic acid do not seem to be of any great clinical importance, as any marked reaction for this sub- stance is indicative of a pathologic condition whose extent may bear no relation whatever to the amount of lactic acid present. A general idea of the amount of lactic acid may be obtained by evaporating 10 c.c. of gastric juice, acidulated with a few drops of sulphuric acid, to the consistency of a syrup and then extracting this residue several times with acid-free and alcohol-free ether. The ether may be removed by evaporation and the residue taken up with water. This watery solution may now be titrated with tenth-normal sodium hydrate, each c.c. of alkali used representing 0.009 gram of lactic acid. The method of Boas, while very exact, is much too complicated and time- consuming for clinical work so that the writer will refer to other books for a description of this test. The principle of the method is based upon the fact that when lactic acid is heated with a strong oxidizing agent it is decomposed into acetic aldehyd and formic acid. If now the aldehyd be distilled off and Fig. 17.-Strauss' separatory funnel. (Hemmeter.} 72 DIAGNOSTIC METHODS. transformed into iodoform by the addition of alkaline iodin solution, this iodoform may be quantitatively determined. (&). Butyric Acid. This acid does not occur in the gastric contents, under physiological con- ditions, unless much milk or carbohydrate food has been introduced. Fliigge has shown that butyric acid may be derived from lactic acid and conse- quently may be present under the same conditions in which lactic acid is found. As butyric acid may be introduced from without and may have been formed in the mouth, one should be careful in drawing conclusions as to the clinical significance of butyric acid. If present in any large amount, butyric acid may be usually recognized by its distinct odor which is that of rancid butter. This test may not be sufficient for the recognition of butyric acid so that it is advisable to shake out the gas- tric juice with ether, evaporate, and take up with water as described under Lactic Acid. If a small pinch of powdered calcium chlorid be added to this watery solution and the mixture warmed, butyric acid will separate from the fluid in the form of small fat drops which float on the surface and have a characteris- tic odor of rancid butter. If a portion of the dried ethereal extract of the gastric juice be treated with a few drops of concentrated sulphuric acid and a little alcohol, the odor of ethyl- butyrate is perceptible on slight warming. This odor is the peculiar one of pineapples and is very easily recognized. This test is known as the pine- apple test. (c). Acetic Acid. It is not an infrequent occurrence to find acetic acid in rather large amounts in pathological conditions, although it is rare physiologically except when in- troduced with the food. The pathologic acetic acid is formed by the bacterial decomposition of the alcohol which is produced by the action of yeast upon carbohydrates. As yeast fungi are so frequently present in cases of dilatation of the stomach, associated with stagnation of its contents, acetic acid may be found under such conditions and may constitute a large portion of the total acidity. In testing for acetic acid the aqueous extract of the ethereal residue of the gastric juice is carefully neutralized with sodium carbonate solution. If a few drops of ferric chlorid solution be added to this neutralized solution a deep red color will appear if acetic acid be present. If this solution be boiled a reddish precipitate of basic ferric acetate is formed. This neutral- ization of the aqueous solution is an essential point in this test, as the presence of free acid will prevent the appearance of any precipitate and the presence of free alkali will cause the formation of ferric hydroxid which will mislead, as the coloration is very much the same. The writer has seen several cases in which acetic acid was mistaken for lactic acid when the Kelling test was applied. On adding gastric juice containing a large amount of acetic acid to the dilute GASTRIC CONTENTS. 73 ferric chlorid solution a change in color is observed, but in no case do you get the distinct canary-yellow color which can be possibly referable to acetic acid. (5). Gastric Ferments. The enzyme pepsin is the most important of the ferments occurring in the gastric juice. As previously stated this ferment is excreted in the form of the zymogen (pepsinogen or propepsin) by the chief cells of the fundus glands. It becomes active, that is, converted into pepsin, by the free hydrochloric acid of the gastric juice. This ferment acts only in acid media and is destroyed by the presence of minute traces of alkali. Its action is continuous, a small portion being capable of digesting large amounts of albumin, providing the products of this digestion are gradually removed. Should the products of ferment activity remain in the stomach an undue length of time, this fer- ment will cease to be active owing to the accumulation of the products of its own activity. Pepsin acts in the presence of many other acids, but the concen- tration of these acids must be higher than in the case of hydrochloric acid. Thus, a 0.2 to a 0.4 per cent, hydrochloric acid gives the best results with pepsin, while a 1 to a 1.or8 per cent, lactic acid is necessary to bring about good results. Very little data exists as to the amount of pepsin or of its zymogen so that we are forced to draw our conclusion regarding a normal or abnormal amount of this ferment from the rate at which known amounts of albuminous material are digested. Pepsin acts only upon protein substances, giving rise to a series of decomposition products which will be discussed later. Normally 25 c.c. of gastric juice will dissolve (digest) 0.05 to 0.06 gram of serum albumin in one hour, the same amount of coagulated egg-albumin in three hours, and a similar amount of fibrin in one and one-half hours. A diminution in the amount of pepsin must be referable to a direct disease of the secreting gland, as general abnormalities do not affect this function as much as they do the production of free hydrochloric acid. Pepsin is usually present when the free hydrochloric acid is either increased or diminished, but in cases of carcinoma, atrophic gastritis, and in occasional cases of pernicious anemia we may find no pepsin and no hydrochloric acid. Such a condition is known as achylia gastrica and occurs sometimes as a direct pathologic con- dition without a known etiology. It frequently happens that pepsin is present when no hydrochloric acid is found. In such conditions no digestion will take place in the stomach unless hydrochloric acid be added or the organic acids be present in very large amounts. (a). Pepsin. Qualitative Tests. The digestive power of the filtered gastric contents will depend, of course, upon the amount of pepsin and the amount of free acid present. Artificial digestion experiments are at present the only methods by which we may test the amount of pepsin. The substances used in these digestion experiments 74 DIAGNOSTIC METHODS. are egg-albumin and fibrin. The fibrin may be prepared by beating freshly drawn ox-blood with a glass rod until the coagula are distinctly formed. These stringy masses are washed thoroughly in water to remove the coloring matter of the blood, are then cut into small pieces of uniform size and are kept in al- cohol for a few days. These hardened masses are placed for one to two days in a neutral concentrated solution of carmin. They are then washed in water, thoroughly pressed, and are preserved in glycerin to which a little carmin has been added. Before being used they should be washed in water to remove the glycerin and free coloring matter. The egg-albumin is prepared for these experiments by boiling an egg until the albumin is distinctly coagulated. This material is then cut into cylinders of about 5 mm. in diameter with a cork borer and are then sectioned into disks 1 mm. thick. These disks may be preserved in glycerin. In testing for the presence of pepsin 25 c.c. of gastric juice, which must contain free hydrochloric acid, are placed in a flask which contains a few pieces of fibrin or a disk of coagulated egg albumin. This flask is then placed in the incubator from 37 to 400 and allowed to remain until the protein is completely dissolved. If pepsin is present the fibrin will show signs of digestion by swelling up in from 15 to 30 minutes, the egg-albumin in from one-half to one hour. Within an hour and a half the fibrin should be practically dissolved, while the egg-albumin will require about three hours. If no hydrochloric acid is present in the gastric juice a few drops of 10 per cent, hydrochloric acid are added to 25 c.c. of gastric juice and the test per- formed in the same manner. A positive result will indicate the presence of the zymogen, pepsinogen. Certain laws have been discovered regarding the action of ferments in general and these are applicable to pepsin. Schutz has found that the relative quantities of pepsin in digesting mixtures containing the same quantity of hydrochloric acid are proportional to the squares of the quantities of albumin digested in the same time, or, in other words, the activity of a ferment varies as the square root of its amount. Nirenstein and Schiff1 have found that this law applies only for the less concentrated pepsin solutions. If the quantity of pepsin in the digesting fluid is so large that more than 3.6 mm. of albumin (see Mett's Test) are digested in 24 hours, the above law does not give reli- able values for the quantity of pepsin. The gastric juice under such circum- stances must be diluted before this law applies. Sahli rightly calls attention to an important reason for diluting the gastric juice before testing for pepsin, namely, the constant presence of the products of peptic digestion which inhibit further peptic activity. The gastric juices which contain a diminished amount of hydrochloric acid are the richest in these inhibiting substances and should be carefully studied. The presence of these substances gives rise to conditions Quantitative Examination. 1 Arch. f. Verdauungskr., Bd. 8, 1903, S. 559. GASTRIC CONTENTS. 75 which make it impossible to arrive at accurate conclusions if the pepsin value is calculated from the pure gastric juice. Hammerschlag's Method. Ten c.c. of a 1 per cent, filtered solution of egg-albumin in 0.4 per cent. HC1 are poured into two test-tubes. As fresh egg-albumin contains about 13 per cent, of dry protein, it should be diluted about 13 times to make a 1 per cent, solution. To one of the test-tubes 5 c.c. of gastric juice are added, to the other 5 c.c. of distilled water, both being placed in the incubator at body temperature for one hour. At the end of this time the albumin in each tube is estimated by Esbach's method (see Urine). The difference between the pre- cipitate of albumin in the two tubes is equal to the amount of albumin which has been digested and forms, therefore, a measure of the peptic activity of the gastric juice, the square root of the amount of pepsin being proportionate to the quantity of albumin dissolved. This test is open to the objection that the albumin of the gastric juice, as well as the albumoses are precipitated by the reagent. In one hour not all of the egg-albumin will be digested, normally only about 90 per cent. Mett's Method. This method1 is, perhaps, the one most frequently used and the one which gives much more accurate results than the others advised.2 The whites of sev- eral eggs are mixed, in order to avoid accidental variations in the egg-albumin, and are filtered. The gas should be removed from this material by the use of a suction pump as far as possible. A number of glass capillary tubes, each from 10 to 30 cm. in length and 1 to 2 mm. in diameter, are then filled by suction with this albumin. They are then laid in the bottom of a vessel which is placed for five minutes in boiling water in order to coagulate the albumin. The tubes are taken out, wiped carefully, and the ends sealed with paraffin or sealing wax. It occasionally happens that these tubes contain air-bubbles, which will, however, disappear in a few days. These longer tubes may be kept in stock for a considerable period of time. In performing the test for pepsin by the use of these tubes of coagulated albumin, the longer tubes are cut into lengths of about 2 cm. and are placed in a small dish or watch-glass with 5 c.c. of the gastric juice, which must necessarily be acid in reaction. These dishes con- taining the gastric juice and filled glass tubes are then placed in the incubator for 10 to 24 hours. At the end of this time the length of the digested column at each end of each tube is measured and the average length of the column of albumin digested estimated. The square of this digestion length is the measure of the relative amount of pepsin in the gastric juice. The unit upon which one may base comparative results of the relative amount of pepsin is that quantity of pepsin by which 1 mm. of albumin in a Mett's tube will be digested in 24 hours with an acidity of 0.18 per cent, free HC1. The length 1 J. A. D., Petersburg, 1889. 2 See Christiansen, Biochem. Zeitsch., Bd. 46, 1912, S. 257; also, Waldschmidt, Pf. Arch., Bd. 143, 1912, S. 189. 76 DIAGNOSTIC METHODS. which theoretically pure pepsin would give is 4 mm. the units represented by the pepsin of the gastric juice being anywhere from o to 256. Nirenstein and Schiff, for the reasons previously -mentioned, advise the dilution of a gastric juice before applying this test, believing that a dilution of 16 will give more nearly exact quantitative results. Their method is as follows: One c.c. of the filtered gastric juice is diluted with 16 c.c.of 0.18 per cent. HCL The procedure is then the same as in Mett's method, the results being multiplied, of course, by 16 in order to obtain the actual num- ber of units in the gastric juice. In some cases it has been found that a dilution of 1 to 32 gives better results, but this is rather unusual. The results of these workers show that striking differences exist in individual gastric secretions, figures ranging between o and 256 units being obtained. This points to the fact that the pepsin concentration is independent of the amount of acid in the gastric juice. It may be necessary, therefore, to pay more attention in the future to the determination of the pepsin values than in the past. The quantitative estimation of pepsinogen may be carried out by either one of the methods pre- viously outlined for pepsin by rendering the gastric juice acid with hydro- chloric acid up to 1 to 2 parts per thousand. This method1 is based upon the digestion of an acid casein solution in 0.2 per cent, hydrochloric acid by the gastric juice. The acid casein solution is prepared by dissolving 100 grams of finely powdered dry casein in 1,900 grams of a solution of hydrochloric acid containing 5.04 grams HC1. A measured quantity of the gastric juice is added to a definite amount of the acid casein solution and placed in the incubator at body temperature. After the end of one hour the digestion mixture is poured into 100 c.c. of 20 per cent, sodium sulphate solution, in which the non-digested casein is completely precipitated. This is collected on a weighed filter and washed with distilled water until no trace of sulphate reaction is evident. It is then dried with alcohol and ether and weighed. The difference in weight between this undigested residue and that contained in the original amount taken gives the amount of casein digested. This method, as modified by Volhard, consists in the titration of the acid- ity of the filtrate from the solution to which the sodium sulphate has been added. The total acidity is higher, the more the casein is in the uncoagulated form and the increase in the acidity will vary as the square root of the amount of pepsin. The writer has had no experience with the ricin method of Jacoby-Solms and must, therefore, neglect a discussion of it. Method of Thomas and Weber. The normal gastric juice contains a second ferment, chymosin, which has the function of coagulating milk independently of the presence of acid. The zymogen of this ferment becomes active, however, only in the presence of (b). Chymosin (Rennin). 1 Central, f. Stoffw. u. Verdauungskr., Bd. 2, 1901, S. 365. GASTRIC CONTENTS. 77 acids; that is, the zymogen is converted into rennin by acids. In this process of coagulating milk, insoluble casein is formed from the caseinogen of the milk by the combined action of the rennin ferment and calcium salts, while the curdling of milk is due to the precipitation of unchanged caseinogen by acids. It is evident, therefore, that this process resembles very closely that of the co- agulation of the blood, as calcium salts are absolutely essential for its success. Leo's Method. Three to five drops of gastric juice are added to 5 or 10 c.c. of fresh uncooked neutral or amphoteric milk and the mixture placed in the incubator for 15 to 20 minutes. If rennin is present in normal amounts coagulation will be observed. In this process one may not be sure as to whether the curdling is due to the action of rennin or to that of the acid. Rennin action occurs typically only when no change in the reaction of the milk has taken place. Riegel's Method. Three to 5 c.c. of neutralized gastric juice are added to 5 to 10 c.c. of fresh milk. This mixture is placed in the incubator and left for 15 minutes, when distinct coagulation will occur in the presence of rennin. If the milk be boiled previous to treatment the result is not so typical. Quantitative methods for the determination of rennin are at present uncertain and are even of doubtful utility. So little is known about the varia- tions in the rennin ferment of the gastric juice that an exact determination would add little to the clinical history. According to Glassner, pepsin and rennin are both diminished in cases of tumor of the fundus, while pepsin is diminished and rennin normal in tumors of the pylorus. (c). Lipase. While it is undoubted that lipase occurs in the gastric secretion, its action is very slight. Normally, gastric digestion is not much concerned with the splitting of fat into lower products, but, as Volhard has shown, this action does occur. In testing for the presence of lipase in gastric contents it is necessary that the examination be made after the stomach is thoroughly washed out following the administration of a test meal free from fat. This ferment may be detected qualitatively by adding a small piece of fresh neutral butter to the gastric juice and placing the mixture in the incubator for one hour. At the end of this time a distinct odor of butyric acid will be observed. Quantitative methods for the determination of this ferment do not seem advisable, although Volhard as well as Stade have used such a method for scientific purposes. It may be said that owing to this lipolytic action statements have arisen that the Sahli test meal gives erroneous results. Careful work by Seiler shows that the amount of fat decomposed under the conditions of the test meal is so slight that it may be neglected. Volhard found that after two hours from 30 to 36 per cent, of the fat was split up into fatty acids, which aid in dissolving the bile and in forming an emulsion with the neutral fat in the intestinal canal. 78 DIAGNOSTIC METHODS. The Products of Protein Digestion. It is generally stated that the action of the pepsin and hydrochloric acid of the gastric juice upon protein material passes through the following stages: The albumin is first converted into acid-albumin (syntonin), then into albu- moses, of which there are four (prot-albumose, hetero-albumose, dys-albumose, and deutero-albumose), and ultimately into peptone. It is well established that pepsin itself will not carry the hydrolysis beyond the stage of peptone or somewhat lower polypeptids. The finding of amino-acids and hexon bases, by the author as well as by others, in the contents of a dilated stomach must be interpreted as meaning a conversion of the pepsin products into lower cleavage units through the action of some other ferment, whether derived from the intestinal canal or excreted into the stomach from a malignant growth. Pepsin has, therefore, a function distinct from that of trypsin, attacking different linkings within the protein molecule. It is practically never necessary to test the stomach contents for such material in clinical work, so that the writer will refer to text-books on physio- logic chemistry for a description of the products of gastric protein digestion. The Products of Carbohydrate Digestion. The pure gastric juice, owing to its acid content, inverts sugars to a certain extent. The amount of this inversion depends on the number of free hydro- gen ions arising from the acid, and is only very slight under normal conditions. As previously described, the saliva converts starch into soluble substances through the stage of soluble starch, erythrodextrin, achroddextrin and finally maltose. The action of the ptyalin of the saliva is inhibited by the free acid of the gastric juice, but the action of this salivary ferment is so rapid that from 50 to 75 per cent, of the starch is converted into a soluble form. While these products are not ordinarily tested for in the gastric contents, the fact of their for- mation and presence must be remembered. An excess of starchy material in the food will lead to an increased amount of such decomposition products in the stomach when the acid of the gastric juice is not present in sufficient amount to inhibit the action of the ptyalin of the saliva. Blood. Blood is not a normal constituent of the gastric contents, but is found fre- quently in conditions associated with erosion, in ulcer, and in carcinoma. The appearance of the blood in cases of ulcer is usually that of fresh bright red blood, which may, however, be changed to a brownish substance due to the action of the excess of acid commonly present in this condition. In carcinoma the blood is more intimately mixed with the stomach contents and appears in the form of brownish-black clumps, constituting the so-called "coffee-ground" material. The tests for the presence of blood will be given in detail under Feces, to which the reader is referred. Gases. The stomach usually contains a certain amount of gas which may have been swallowed, may have passed into the stomach from the duodenum, or GASTRIC CONTENTS. 79 may have been produced in the stomach by processes of fermentation. The examination for these gases is not of great clinical importance, but a general idea of the different kinds of gases present seems essential. During the proc- esses of normal digestion, nitrogen, oxygen, and carbon dioxid may occur from the protein digestion, while hydrogen, marsh-gas, and olefiant gas may arise from the carbohydrate hydrolysis. In abnormal processes of digestion we may find ammonia and hydrogen sulphid arising from decomposing protein material. The work of Boas seems to indicate that the hydrogen sulphid is more commonly present in cases of benign gastric dilatation and is rare in car- cinoma. This hydrogen sulphid does not seem to be produced either in the presence of free hydrochloric or of lactic acid. The presence of hydrogen sul- phid can, however, not be considered as a specific substance in the stomach contents, as Dauber has shown that almost every stomach contains bacteria which may produce this gas from sulphur-containing bodies. In cases of dila- tation of the stomach, providing the motility be sufficiently diminished, we find fermentation with resulting gas production even though hydrochloric acid be present. Such a condition never occurs if the motility is normal, utterly regardless of the amount of hydrochloric acid, as in cases of diminished hydro- chloric acid lactic acid will usually prevent such a process. One may show the presence of the gases in the stomach contents by filling a fermentation tube with the well-mixed unfiltered gastric contents and placing it in the incubator for some time. If there is no gas within 24 hours it may be wise to wait at least 48 hours to permit of the proper diffusion of the gas. If gas is formed its nature may be determined by the ordinary chemical tests. This test has some value in determining the degree of stagnation of the stomach contents, but it must be remembered that a small amount of gas is contained in the normal stomach. Other substances, such as acetone, have been found in the gastric contents in pathologic conditions, but tests for these substances are rarely of importance. In conditions associated with the presence of acetone in the stomach contents this substance is usually detected in the breath, so that the examination of the stomach contents is unnecessary. Functions of the Stomach and its Contents. The stomach is to be regarded as a specialized portion of the alimentary tube in which the first stages of digestion of protein material take place. This occurs under the combined action of the hydrochloric acid and pepsin, the resulting products being gradually passed into the duodenum through the py- lorus. An increased acidity of the gastric juice may be associated with a distinct spasm of the pylorus so that food material cannot pass into the duodenum. On the other hand, a lessened degree of acidity is associated with hypermotility, the contents passing rapidly into the intestine, where it is acted upon by the pan- creatic ferments. Beside the function of digestion, principally of the protein foods, the stomach serves as a reservoir to hold the food material, allowing it to 80 diagnostic methods. pass only in small portions into the bowel at any one time. Owing to the pres- ence of hydrochloric acid, the gastric juice is antiseptic, rendering inert many but not all types of bacteria. Moreover, the hydrochloric acid activates the zymogens and thus permits of action upon all types of food material. The work of Pawlow has shown that the acid of the gastric juice is one of the most powerful stimulators of pancreatic secretion. The mechanism of correlation between the stomach and bowel is more easily understood if this point be borne in mind. The acid chyme coming from the stomach is poured out only in small portions at a time so that the pancreatic juice secreted may act upon the smaller portions as they are passed into the bowel. VI. Motility of the Stomach. It is probably true that disturbances in the motility of the stomach are in reality of more importance than are those in the secretory activities.1 Under normal conditions of motility the food material passes into the intestine and is digested there, although no previous gastric digestion has taken place. If the motility be much impaired, stagnation of food with resulting dilatation of the stomach will occur, which will give rise to more or less serious disturbance. The motor disturbances are of three types, (i) vomiting, (2) hypermotility, and (3) motor insufficiency. The most important of these is the latter, as the former has little influence upon actual digestion in the stomach, although the patient may suffer for want of adequate nutrition; while in hypermotility the gastric disturbance will not be much noticed, owing to the fact that the food is rushed into the duodenum where it is digested. The consequences of motor insufficiency may be either disorders of secretion, decomposition, or both. Simple pathologic conditions which hinder the emptying of the stomach, such as ulcers, cicatrices, spasm of the pylorus, and simple atony, are usually asso- ciated with hyperacidity, while malignant conditions usually show a dimin- ished secretion. It is quite rare to find a case of motor insufficiency without disorder of secretion, rarer at all events than disorders of secretion without motor disturbances (Schmidt). Motor insufficiency is quite commonly followed by decomposition of the gastric contents and may even be considered the chief cause of such decomposition. Under these conditions of disturbed secre- tory and digestive activity associated with motor insufficiency we find the absorp- tive power of the stomach very much affected. Hypermotility is seen in many cases of hyperacidity, but it must be remembered that primary hyperacidity may cause spasm of the pylorus and hence bring on a distinct motor insufficiency. We cannot, therefore, judge of the motility of the stomach from the degree of acidity of the gastric contents. An enlarged stomach is not necessarily associated with motor in- sufficiency. Cases of megalogastria are more or less frequent in which the motility is practically normal. When a dilatation is associated with motor insufficiency it is clinically styled an ectasia or ectasis, being known as atonic 1 See Carlson, Amer. Jour, of Physiol., vol. 31, 1912, p. 151. GASTRIC CONTENTS. 81 gastric ectasis if the condition is due to weakness of the muscle, while it is styled hypertonic gastric ectasis if due to pyloric stenosis. Normally, no food should be found in the stomach within seven to eight hours after taking, no matter how large the meal. We have, therefore, a method by which we may judge of the motive power of the stomach. Leube's Method. Leube administers a Riegel test meal and washes out the stomach with a liter of water six hours after. If only very slight traces of food are found in the washings the motor power is regarded as normal. Boas' Method. Boas administers a simple evening meal consisting of meat, bread and butter, and tea, washing out the stomach the following morning. If any food material is found the motor insufficiency is considerable. If the stomach be washed out previous to the administration of the evening meal no food should be found in the stomach in the morning. Method of Ewald and Sievers. This test is based upon the observation that salol is decomposed into car- bolic and salicylic acids only in an alkaline medium. As the salicylic acid is eliminated in the urine in the form of salicyluric acid, it is possible to determine the rate of passage of salol from the stomach to the small intestine. It seems necessary to state that the assumptions on which this test are based are partially wrong. In the first place salol is split into its constituents by gastric juice within 15 minutes, although the degree of dissociation is slight. Moreover, a certain amount of absorption of salicylic acid takes place from the stomach so that a reaction may be obtained in the urine within 15 minutes in cases in which no hypermotility exists. One gram of salol is given to the patient immediately after a meal. The urine is then collected every 15 minutes for two hours and tested by the addi- tion of a small amount of ferric chlorid solution, which will give a violet color in the presence of salicyluric acid. Under normal conditions, according to Ewald, a positive reaction occurs in from 45 to 75 minutes. A further delay above 75 minutes is indicative of motor insufficiency, the degree of insufficiency bearing some relation to the time of appearance of this reaction. Should no result be obtained after 24 hours a stenosis of the pylorus is highly probable. As the writer has so frequently found a reaction for salicyluric acid in the urine within 15 minutes, which is due not to the action of the hydrochloric acid or the ferments of the gastric juice, but to the moisture, temperature, and bacteria, he is accustomed to use the time at which a reaction for this substance disappears from the urine rather than the time at which it makes its first appearance, as the basis of judgment regarding the motility of the stomach. Normally, no re- action for salicyluric acid should occur in the urine after 24 hours although 82 DIAGNOSTIC METHODS. Huber states that it may take 26 to 27 and hence limits his time to these latter figures. It might be wise to follow the suggestion of Sahli and determine both the time of appearance and disappearance. Sahli has called attention to the fact that we are not justified in assuming a pyloric stenosis in case food material is found in the stomach even several days after being taken. He adds that the emptying of the stomach is regu- lated by the intestine rather than by the stomach itself, since nutritive substances reaching the intestine effect a reflex closure of the pylorus (von Mering's reflex) until the intestine has completed its work. The motor activity of the stomach should, therefore, be examined under conditions in which this reflex does not occur. This he determines by estimating the length of time re- quired by the stomach to empty itself of a half-liter of water, the stomach being thoroughly washed out previously. Winternitz Test. Winternitz has recommended the use of iodipin instead of salol for testing the motility of the stomach. This substance is not affected by the gastric contents, but is acted upon in the intestine by the pancreatic secretion and bile in such a way that iodin is set free. This may be tested for in the saliva by adding to it a little starch paste, when a distinct blue color will be observed within 15 to 45 minutes. VII. Absorptive Power of the Stomach. The absorptive power of the stomach is not of great importance clinically, as the greatest part of absorption occurs from the intestinal tract. However, tests for such power have a certain associated value and are, therefore, usually made. For this purpose Penzoldt and Faber advance a method depending upon the principle that under physiologic conditions potassium iodid is rapidly absorbed by the gastric mucous membrane and is immediately eliminated in the saliva. A capsule containing two to three grains of potassium iodid is given to a patient shortly before a meal. The saliva is then tested as follows for the presence of potassium iodid at intervals of two to three minutes. The saliva is slightly acidified with nitric acid and treated with a few drops of starch paste when the characteristic blue color of iodid of starch will be formed by the action of the iodin liberated from the potassium iodid by the nitric acid. Under physiological conditions the first trace of iodin will appear in the saliva within ten minutes of its administration upon an empty stomach. Under pathological conditions a delayed reaction may be observed in almost all dis- eases of the stomach, especially in dilatation and in carcinoma. The test will naturally be delayed in case the stomach is filled with food. This test has little value, as it may appear or not in all types of gastric disease. Von Mering1 has found that potassium iodid is not absorbed at all from the stomach even within 'Klin. Jahrb., vol. 7, 1899. GASTRIC CONTENTS. 83 t"wo or three hours, so that the iodin appearing in the saliva may be due to absorption from the intestine. If his results are confirmed, this test becomes useless as one indicating the absorptive power of the stomach, but may then be invaluable as an indicator of the motor power of the stomach. VIII. Indirect Examination of the Stomach Contents. As not all cases of disease of the stomach permit of examination by removal of the contents through the stomach-tube, methods have been advanced to permit of indirect determination as to the activity of the stomach contents. These methods do not permit of accurate determination of the acidity or of the ferments of the juice, but do give much information regarding the normal digestive powers and motility of the stomach. Gunzburg's Method. A tablet of 0.2 gram of potassium iodid is placed in a piece of the thinnest possible strongly vulcanized rubber tubing measuring about 2.5 cm. in length. The ends of the tubing are folded and the package tied with three threads of fibrin which have been hardened in alcohol. The package is now tested by placing it in warm water for several hours and examining the water for potas- sium iodid. The patient swallows one of these packages three-quarters of an hour after an Ewald meal, the saliva being tested for potassium iodid at intervals of 15 minutes. In the presence of free hydrochloric acid in normal amounts the threads of fibrin are dissolved and the potassium iodid is absorbed, giving a reaction in the saliva in from one to one and three-quarters hours. In cases of hypochlorhydria the reaction is delayed, a delay of six hours indicating a practical absence of free hydrochloric acid. This test very frequently gives reliable results, but the threads of fibrin soon become brittle and break on swallowing the package so that a reaction for potassium iodid under these conditions would have no value. Sahli's Desmoid Reaction. Sahli1 has recently introduced the 11 Desmoid bag" for use in estimating the functional activity of the stomach. These bags are made of the ordinary rubber-dam used by dentists and contain a pill of 0.05 gram of methylene blue and 0.1 gram of iodoform. The bag is tied, in a manner especially outlined by Sahli, with cat-gut which has been allowed to dry but has been untreated chemically. This gut, according to Sahli, is digested only by the gastric juice and not by the pancreatic juices. This pill is administered to the patient immediately following the noon meal and the urine and saliva tested at intervals of one hour, beginning three hours after administration of the pill. The digestion of the gut by the gastric juice liberates the pill and Fig. 18.-Sahli's Desmoid bag. 1 Correspbl. f. Schweiz. Aertz., 1905. 84 DIAGNOSTIC METHODS. permits of the absorption of both the methylene blue and the iodoform. The methylene blue will appear in the urine coloring it green within six hours, while the iodin will be found in the saliva within two hours. Should the color of the urine not be distinctly green, this tint may be more clearly brought out by adding a few drops of acetic acid and boiling. Variations from the periods indicated above denote a hyperacidity or a hypoacidity of the gastric juice according as the time of appearance of the reactions is lessened or increased. As the gut is digested only by the gastric juice a nonappearance of either reaction would indicate an anachlorhydria. The writer has used these desmoid bags in a large number of cases and has found them fairly reliable, giving results which have, in many cases, been confirmed by chemical analysis. As these bags are not obtainable in the market, he has been forced to make them himself and has found that the technic of Sahli must be followed very closely, especially as regards the tying of the gut. Other methods, such as those of Dunham, Turek, and Einhorn, have been advocated, but possess no advantages over those outlined. The writer will refer, therefore, to other works for their description. IX. The Gastric Juice in Disease. (1). Hyperchlorhydria. By hyperchlorhydria is meant the secretion of an abnormally acid gastric juice whose acidity is due to an excess of free hydrochloric acid. This secretion is much more marked during digestion, being less frequent on an empty stomach. Under these conditions we usually find an increased total acidity along with the increase of free hydrochloric acid. A condition which is characterized by a high total acidity with a very high amount of organic acid would not, of course, be considered in this connection. A hyperacidity or hyperchlorhydria exists when we have more than 0.2 per cent. (6o°) of free hydrochloric acid. This condition may be due to pathologic changes in the mucosa or to direct nervous influences. Cases of pure hyperchlorhydria are occasionally very stubborn and may be associated with almost any variety of abnormal gastric function. While we find hypermotility of the stomach in many cases of hyper- chlorhydria, we very frequently note a diminished motility due to spasm of the pylorus. This condition brings about a stagnation of the stomach contents and a consequent increase in the fermentative processes. The acidity in such cases may run as high as 2000 or over and the digestive powers of the gastric juice, as regards proteip substances, may be much increased, the carbohydrate digestion being correspondingly diminished. These facts point to the reason for the administration of an increased protein diet in such cases, the protein combining with the hydrochloric acid and thus taking a portion of the excess from the field of action. While this condition is not a distinct entity, yet we find many cases which come under the heading of idiopathic hyperchlorhydria and which are not GASTRIC CONTENTS. 85 associated with other pathologic conditions. Some of these cases are purely functional and clear up promptly under proper treatment, while others are of nervous origin and are remedied only when the etiologic factor is eliminated. In this latter type of hyperchlorhydria the degree of acidity varies with the nervous symptoms, giving rise to the term " heterochylia." (2). Hypersecretion (Gastrosuccorrhea). By hypersecretion is meant an excessive secretion of gastric juice which is out of proportion to the physiologic stimulus. This hypersecretion occurs even when no stimulus is present, is always pathological, and, according to Riegel, always produces pathological results. A hyper- or continuous secretion may be determined by finding a fairly large amount of gastric juice in the fast- ing stomach under conditions which rule out stenosis and stagnation. The stomach is washed out before the patient retires, the contents being withdrawn the following morning. If a quantity of highly acid fluid is obtained, a hyper- secretion is proven. The quantity taken from the fasting stomach should never be more than 100 c.c., while Strauss regards 40 c.c. as an indication of hypersecretion. This secretion, to be called a hypersecretion, must contain no food remnants, no sarcinse nor yeast cells, but should be distinctly acid. This condition is probably a functional neurosis, being constant or inter- mittent and a part of a general neurosis, a secretory neurosis, or the result of organic nervous disease, such as the gastric crises of tabes dorsalis. Reich- mann has reported cases of the periodic or intermittent type, during the inter- vals between the attack the digestion of the patient being normal. Such cases are known as Reichmann's disease. The chronic cases are of long duration and have a gradual onset. The patient complains of much discomfort, feeling of weight or depression in the stomach, pain during digestion, vomiting, especially at night, and a gastric contents with a large amount of free hydrochloric acid. Dilatation of the stomach sooner or later comes on as the result of spasm of the pylorus induced by the hyperacidity. In these dilated stomachs we find, of course, products of fermentation and many yeast cells and sarcinae. (3). Achylia Gastrica. This condition may arise either from a functional disturbance of the mucosa or a true atrophy of the mucosa. This latter state, known as atrophic gastritis, may be the end stage of a chronic gastritis or the result of carcinoma. When this condition is not due to direct gastric disturbance, it is more fre- quently seen in connection with pernicious anemia, in which the general nutri- tion is very much below par. The local condition may not be suspected, as the hypermotility, so common in achylia gastrica, may prevent attention being drawn to the stomach. For a diagnosis of achylia, the test meal of Ewald gives very good results. Examination of the gastric contents shows that the food is little changed, the 86 DIAGNOSTIC METHODS. total acidity very low (1 to 6°), no free hydrochloric acid, gastric ferments much diminished or entirely absent, and lactic acid only in small amounts. The motility of the stomach is usually little impaired, so that the retention of food is unusual. (4). Acute Gastritis. The stomach contents of acute gastritis shows a diminished total acidity, little or no free hydrochloric acid, organic acids relatively increased, much mucus and undigested food. This condition is usually brought on by direct irritation and is generally easily remedied by total abstinence. The material for chemical examination is usually obtained in these cases from the vomitus, as the passing of the stomach-tube is very rarely tolerated. (5). Chronic Gastritis. All grades of this condition may exist up to complete atrophy of the mucosa. Examination of the stomach contents shows practically no digestion of the food material, much mucus intimately mixed with the food particles, the secretion usually diminished, free hydrochloric acid diminished or absent, ferments much reduced, protein digestion small, starch digestion little affected, micro- scopic examination showing the presence of many epithelial cells and leuco- cytes. There are some cases of chronic gastritis in which a hyperacidity of the juice is evident, but these are rare, the usual findings being one of diminished acidity. The motility of the stomach in these cases is sometimes normal, some- times increased, or may be diminished. One of the most characteristic findings in this condition is the presence of a large amount of mucus containing either leucocytes or their nuclei and epithelial cells from the walls of the stomach. If there is little acid present in the gastric contents the mucus may swell up and appear greater in volume. As a general rule, it may be said that the mucus and the hydrochloric acid vary inversely as their amounts. (6). Nervous Dyspepsia. This condition of nervous dyspepsia is part of a general neurosis and may show no characteristic findings in the stomach contents. The degree of acidity may range from a normal to either a hyper- or a hypoacidity, while the amount of ferments present will not usually vary. The findings in nervous dyspepsia are not at all constant, varying at different examinations. We have, therefore, more or less distinct methods of differentiation between this con- dition and chronic gastritis. In the first place, the acidity of chronic gastritis remains constant for several examinations while that of nervous dyspepsia is variable. The ferments are diminished in cases of gastritis, while they are normal in nervous dyspepsia. Much mucus is found in gastritis while little or none is present in dyspepsia. The cases of nervous dyspepsia are partially associated with distinct errors in eating, the American frequently bolting his food in such a way as to make it difficultly digestible. The influence of nervous conditions over gastric function has been very well expressed by Emerson GASTRIC CONTENTS. 87 when he says, "a neurasthenic will often worry his subliminal gastric sen- sations into the sphere of consciousness." (7). Ulcer of the Stomach. The diagnosis of ulcer of the stomach depends, to a large extent, upon the clinical symptoms of the disease rather than upon the examination of the stomach contents. While the symptoms of this disease, increasing dyspepsia, pain, vomiting, blood in the vomitus, and hyperacidity of the vomitus, are well- known, the stomach-tube should rarely ever be used in obtaining the contents of the stomach, owing to the danger of perforation in such cases. The vomitus of such cases is usually ejected from one to three hours after a meal and contains well-digested food. Blood may or may not be present and may be either fresh red blood or dark in color from the formation of hematin. The total acidity of the gastric contents is usually increased, hydrochloric acid constituting a large part of this total acidity. A single examination of the gastric contents will rarely determine anything about an ulcer, so that repeated examinations of the vomitus must be made to obtain a general idea of the acidity. In these cases blood is usually present in the feces and may be de- tected as outlined later. When an ulcer is complicated by a beginning carcinoma, we may find all types of variation in the acidity of the stomach contents. (8). Carcinoma of the Stomach. In no other condition of the stomach is an absolutely certain early diagnosis to be so much desired as in carcinoma of the stomach. The chemical features of the gastric juice in this condition may be very suggestive or may be negative. The clinical history of the case along with the age of the patient are probably of more importance in making a diagnosis of carcinoma than are variations in the chemical composition of the juice. However, such changes may usually be found and are, therefore, considered as at least of presumptive evidence. The local symptoms of carcinoma of the stomach are sometimes as variable as are the changes in the gastric juice, so that every possible point in diagnosis should be taken advantage of, if for no other purpose than to exclude this condition. Perhaps the most important sign of carcinoma is the absence of free hydro- chloric acid. Although this condition is present in about 85 per cent, of cases, it cannot always be traceable to carcinoma, as it may occur in atrophic gastritis and advanced chronic gastritis. This lack of free hydrochloric acid is due to the union of this acid with some body which in itself does not show an alkaline reaction. Von de Velden suggests that the secretion from the cancer is the active agent in neutralizing the hydrochloric acid. Moreover, the products of protein digestion might have some power in neutralizing the acid, Emerson having shown that hexone bases are present as a result of the action upon the protein of a ferment derived from the tumor itself. Certainly, in cases of carcinoma the total nitrogen of the stomach contents is much increased, so that 88 DIAGNOSTIC METHODS. such bodies are probably a very great factor in the diminution of the hydro- chloric acid. This reduction in hydrochloric acid is also influenced by changes in the mucosa to such an extent that the active secretion is diminished. The failure of free hydrochloric acid is usually a very early symptom, but it must be remembered that hydrochloric acid may be present in normal amounts or even in increased amounts when the carcinoma is small and occupies the pyloric region or when this growth develops on the base of an old ulcer. The acidity may vary markedly from day to day, sometimes showing no free hydro- chloric acid and sometimes a considerable amount. This variation is of con- siderable practical importance. It is to be said, moreover, that absence of free hydrocholoric acid may be noted in cases of pernicious anemia, and carcinoma of gall-bladder, pancreas, or duodenum. In cases of carcinoma of the esoph- agus the disappearance of hydrochloric acid from the gastric contents seems necessarily to be the result of the secretion of the tumor neutralizing the gastric juice. Along with this diminution in free hydrochloric acid the total acidity is also diminished. The presence of an increased amount of lactic acid is a very valuable sign in cancer. About 90 per cent, of patients show the presence of lactic acid when there is no free hydrochloric acid, but when there is a large amount of combined HC1 pointing to a free secretion of this acid. Lactic acid may not be present in cases of carcinoma, especially those in which the growth is upon the base of an old ulcer, or it may be present in conditions other than carcinoma, such as chronic gastritis, associated with atrophy of the mucosa and dilatation of the stomach, especially when a benign stenosis of the pylorus exists. It must be stated, however, that we usually find an increased secretion of hydrochloric acid in cases of simple benign pyloric stenosis, so that lactic acid cannot be formed in the presence of this increased hydrochloric acid. According to Riegel, the chief cause of the lactic acid formation is the combination of motor insufficiency with a hypoacidity associated with a diminished secretion, both of acids and ferments. This diminution in the amount of ferments is in no way specific for cancer, as it is in reality due to the chronic gastritis which is set up by the tumor. As lactic acid cannot be formed in the presence of an appre- ciable amount of free HC1, we do not always find after a test meal a large excess of this acid, especially in those cases in which considerable combined HC1 is present. It seems advisable, therefore, in testing for the presence of lactic acid, to examine the contents of the fasting stomach in the morning after it has been well washed out the preceding evening. Whether wre are to assume that lactic acid is formed by the action of the organisms in the stomach upon food material, whether it be a normal product of digestion, or whether it is a product of the activity of a ferment derived from the tumor, must be left for future investigation. The vomited material or the material obtained by washing out the stomach shows very little digestion of the protein elements, while the carbohydrates are well hydrolyzed. The amount of material in the stomach will be large or small GASTRIC CONTENTS. 89 depending upon the degree of stenosis, so that it is nothing unusual to obtain several pints of material containing undigested protein residue. Microscopic examination of this material may show cellular masses washed off from the tumor or may show these fragments embedded in masses of blood. Blood is a usual finding in cases of carcinoma and may be detected by the methods dis- cussed under Feces. Sarcinae and yeasts are rare, the former occurring more frequently in cases of marked dilatation. The Boas-Oppler bacilli are more or less constant findings in carcinoma, occurring in about 80 per cent, of the cases and only very rarely in any other condition. The characteristics of these bacilli have been previously discussed. This test is based upon the fact that albumin is secreted from the car- cinoma itself and passes into the gastric contents. The patient is placed upon an absolutely protein-free diet for 24 hours and the stomach carefully washed out at the end of this time with 400 c.c. of physiological salt solution. A few hours thereafter the contents of the stomach are removed and the remaining material washed out with 400 c.c. of physiological salt solution. The total nitrogen and the albumin are then estimated, the former by the Kjeldahl method and the latter by the Esbach method (see Urine). The nitrogen ranges from 10 to 70 mg. per 100 c.c. in cases of carcinoma, while in other conditions it varies from o to 16 mg. per 100 c.c. The Esbach reaction gives a distinctly appreciable precipitate for albumin, anything over 0.5 parts per 1000 being considered indicative of carcinoma. While this test is not infallible, yet the writer has found it present in so many cases of carcinoma that he is inclined to make a very strong presumptive diagnosis on the basis of this test. Salomon's Test. These investigators1, working in Muller's clinic, have taken advantage of the well-known fact that normal peptic digestion does not proceed to the formation of amino-acids but stops at the peptone or peptid stage. Further, it has been shown by Emerson, Fischer and others that digestion of protein goes somewhat further in carcinomatous stomachs than in normal ones, owing to the fact that carcinomatous tissue, as well as various body tissues, contains pro- teolytic ferments capable of producing more complete hydrolysis than does pepsin (see Lewin2). Neubauer and Fischer, following the investigations of Erdmann and Winternitz,3 Glaessner,4 and Volhard,5 use this hydrolysis of peptids by such proteolytic ferments for diagnostic purposes. Their method is as follows: The contents of the stomach are withdrawn one-half to three- fourth hour after an Ewald meal; 10 c.c. of the filtered material are mixed with Neubauer and Fischer's Test. 1 Deutsch. Arch. f. klin. Med., Bd. 97, 1909, S. 499. 2 Ergebn. d. inn. Med., Bd. 2, 1908, S. 212. 3 Miinch. med. Wchs., Bd. 50, 1903, S. 983. 4 Berl. klin. Wchs., Bd. 40, 1903, S. 599. 5 Miinch. med. Wchs., Bd., 50, 1903, S. 2130. 90 DIAGNOSTIC METHODS. a little glycyl-tryp tophan1 and a little toluol is added to prevent bacterial action. The mixture is placed in the incubator at 370 C. for 24 hours. At the end of this period, withdraw 2 or 3 c.c. of the mixture from beneath the toluol layer, place this in a test-tube, add a few drops of 3 per cent, acetic acid and then allow the fumes of bromine to pass into the tube from an open bottle of bromine. This part of the technic must be carried out very carefully as an excess of bromine will obscure and even destroy the reaction. After shaking the tube a positive reaction is shown by the appearance of a rose-red or intense red color. If no red color appears, carefully add more bromine vapor and shake. This addition may be continued until an excess of bromine is evidenced by a light yellow color of the mixture. These workers have shown that errors may arise in this test due to the presence of (1) tryptophan in the gastric contents, (2) peptid-splitting bacteria, (3) trypsin or pancreatic juice, and (4) blood. To do away with these sources of error they advise the rejection of gastric contents showing any of these substances. Filtering the contents removes the peptid-splitting bacteria to some extent while the addition of toluol prevents their development. Tryp- tophan, as such, may be tested for as usual, while the presence of pancreatic juice may be assumed if bile be present. Blood is shown by tests outlined under Feces. Warfield2 has recently shown another fallacy of this test, which may render the test of little clinical value. He has found that the saliva of many patients is capable of decomposing glycyl-tryptophan into its constituent elements. He concludes, therefore, that swallowed alkaline saliva, when mixed with neutral or faintly acid (not over 0.05 per cent. HC1) gastric contents, may introduce sufficient error to account for the discordant results obtained. Koelker3 has found a di- and tripeptid-splitting ferment in the saliva. Lyle and Kober4 believe that a negative result with this test is of far greater value than is a positive one. Weinstein5 believes that the presence of tryptophan in the gastric con- tents, as withdrawn, is more positive as a test for carcinoma than is the glycyl- tryptophan test, providing intestinal fluid has not regurgitated into the stomach. He has confirmed the findings of Warfield and asserts that this latter fallacy can not be applied to his "tryptophan test" as the salivary proteolytic action is not such as to produce trytophan from protein in non-cancerous gastric contents. He, therefore, suggests the following modification. A portion of the gastric contents is withdrawn 4 or 5 hours after a full dinner and the filtered contents tested directly for tryptophan after acidifying the filtrate with acetic acid. If no reaction appears, incubate for 24 hours and repeat the 1 Glycyl-tryptophan, as prepared for the test, is a clear solution. It is furnished by Kalle & Co., Biebrich a. Rhein, under the name of "Fermentdiagnostikum." It comes in small bottles which are used directly for the test by simply adding the io c.c. of filtered gas- tric contents and incubating for 24 hours. 2 Bull. Johns Hopkins Hosp., vol. 22, 1911, p. 150. 3 Zeitsch. f. physiol. Chem., Bd. 76, 1911, p. 27. 4 New York. Med. Jour., vol. 91, 1910, p. 1151. 5 Jour. A. M. A., vol. 55, 1910, p. 1085; Ibid. vol. 57, 1911, p. 1420. GASTRIC CONTENTS, 91 test. This latter test appeals to the writer much more than the original, as it is not open to the same fallacies and is more simple. See Taylor and Hall, Jour, of Path, and Bact., vol. 17, 1912, p. 121; Sanford and Rosenbloom, Arch, of Int. Med., vol. 9, 1912, p. 445; Smithies, Ibid., vol. 10, 1912, p. 357 and 521; Jacque and Woodyatt, Ibid., p. 560; Hamburger, Jour. A. M. A., vol. 59, 1912, p. 847. BIBLIOGRAPHY. i. Boas. Die Erkrankungen des Magens. Berlin, 1906. 2. Billings. Diseases of the Digestive System. New York, 1906. 3. Debove, Achard, et Castaigne. Manuel des Maladies du Tube Digestif. Paris, 1907. 4. Ewald. Die Erkrankungen des Magens. Berlin, 1900. 5. Harley and Goodbody. The Chemical Investigation of Gastric and Intestinal Diseases. London, 1906. 6. Hemmeter. Diseases of the Stomach. Philadelphia, 1906. 7. Loeper. Manuel des Maladies du Tube Digestif. Paris, 1907. 8. Lyon. Diagnostic et Traitement des Maladies de L'Estomac. Paris, 1909. 9. Reed. Diseases of the Stomach and Intestines. New York, 1907. 10. Riegel. Die Erkrankungen des Magens. Wien, 1903. 11. Roeser. La Chimie Alimentaire. Paris, 1906. 12. Sigaud. Traite clinique de la digestion. Paris, 1907. CHAPTER IV. THE FECES. I. General Considerations. The feces are composed of substances of different origin, which may be divided as follows: (i) Food remnants, either undigestible constituents or digestible but unabsorbed elements; (2) secretions of the alimentary tract; (3) decomposition products and bacteria; (4) formed and unformed elements derived from the intestinal wall, and (5) foreign bodies, such as hair, wood fiber, parasites, parasitic ova, and enteroliths. It is not possible to draw a sharp line between a truly physiologic and pathologic composition of the feces. In each individual case this division will depend upon a number of factors, among which may be mentioned composition of the food, method of taking the food, individual functional capacity of the intestines, frequency of the bowel movements, and general systemic conditions. The condition of the food remnants will give much information regarding the functional capacity of the bowel, although examinations of this factor are at present made to a very slight extent. Under normal conditions, about one-third of the dry substance of the feces arises from the bacteria; this amount increasing under certain patho- logic conditions which will be later discussed. As products of bacterial activity we observe the formation of hydrobilirubin from bilirubin as well as the reduction of certain medicaments and the rare formation of certain diamins, which will be treated later. This bacterial activity will vary much under pathologic conditions and may have much to do with the symptomatology of the case investigated. The ordinary decomposition products are derived from the carbohydrates and proteins of the food; from the former are produced, by fermentation, volatile fatty acids, lactic acid, succinic acid, alcohol, carbon- dioxid, hydrogen, and methane, while from the latter are formed by putre- faction indol, skatol, phenol, ammonia, and hydrogen sulphid. The fats are decomposed only to a very slight extent. The normal products of digestion in the intestinal tract as well as the factors bringing about these changes will be discussed later. Normal Feces. For proper comparisons between the feces of various individuals the food must be the same. Starvation feces, meat feces, and milk feces are typical types, but are not normal in the strict sense. For comparison, however, an arbitrary norm must be established in order to judge of slight variations which have to do with special differences in utilization of food and which are not 92 THE FECES. 93 observable by the eye except under certain conditions (as, for instance, fatty stools). Through the work of Praussnitz and of Schmidt and Strasburger a new field for study of normal and abnormal feces has been opened up. A normal feces should be one which consists almost entirely of remnants of the digestive juices and intestinal secretion arising from a purely digestible, properly prepared, and assimilable food. It contains approximately 8.6 per cent, of nitrogen, 16 per cent, of ether extract, and 15 per cent, of ash calculated on the dry basis. Any appreciable variation from this composition would indicate a diminution of the functional activity of the bowel. In the investi- gation of pathological cases this composition is scarcely to be assumed, as the Fig. 19.-Normal Feces. (Landois.) a, Muscle fibers; b, tendon; c, epithelial cells; d, leucocytes, e-i, various forms of plant- cells, among which are large numbers of bacteria; between h and b are yeast-cells k, ammonium-magnesium phosphate. diet is more or less restricted, the appetite is capricious, the intestinal activity is variable, and the feces are, therefore, different from those of the healthy. It must not be understood from this that a feces showing the above composition is necessarily normal in other respects. One may not rely entirely on the chemical composition, but must largely consider the macroscopic and micro- scopic findings. In order that we may have a sound basis upon which to judge of intestinal activity it seems wise to have some sort of test diet which may be given to suspected cases. Diet of Schmidt and Strasburger. This diet1 is so selected that it can be used by the healthy as well as by those with intestinal trouble; its amount is sufficient to satisfy the maximal calorie requirement of the individual while at rest; it contains the three chief groups of food material in definite relation to each other; is as free as possible from remnant-leaving food, and can be easily obtained and prepared. The starchy food present is of the amount and kind which have been shown most favorable for the prevention of excessive fermentation in the bowel. The daily diet is as follows: One and five-tenths liters of milk, 100 grams of 'Die Faeces des Menschen, Berlin, 1910. 94 DIAGNOSTIC METHODS. zwieback, two eggs, 50 grams of butter, 125 grams of beef (raw weight), 190 grams cooked potato, and a gruel of 80 grams of oatmeal. This is distributed through the day as may best suit the patient. This diet contains 102 grams of protein, in grams of fat, and 191 grams of carbohydrate, yielding 2,234 raw calories. In cases which show a diarrhea, due' to the milk, one may sub- stitute instead of 1/2 liter of milk the same amount of cocoa made from 20 grams of cocoa powder, 10 grams of sugar, 100 grams of milk, and 400 grams of water. Small variations in the amount of milk, sugar, butter, and even of eggs may be permitted, but the outline, as regards meat, zwieback, potato, and gruel should be rigidly adhered to. This diet should be administered for three days or longer if necessary to obtain a stool which comes from it. In order to judge of the first appearance of the stool from this diet, the patient should be given a capsule containing 0.3 gram of powdered carmin, both preceding and following the diet. Instead of carmin, one may use cork, charcoal, or silicic acid. Folin's Diet. This diet1 is especially serviceable in case one wishes to follow the metabo- lism in any special case. Its easy application and its fairly constant values for nitrogen, phosphorus, chlorin, and sulphur make it invaluable. The standard diet, which is given to the patient daily for several days is as follows: Whole milk, 500 c.c. Cream (18 to 22 per cent, fat), 300 c.c. Eggs (whole), 45° gm- Horlick's malted milk, 200 gm. Sugar, 20 gm. Sodium chlorid, 6 gm. Water, 2100 c.c. This diet contains approximately 119 gm. of protein, 148 gm. of fat, and 225 gm. of carbohydrate, yielding 2787 raw calories. The intake of one day is nitrogen 18.9 gm., 5.9 gm. of P2O5, 3.8 gm. of S03, and 6.2 gm. of Cl. It is not to be compared with that of Schmidt and Strasburger for estimating the intestinal activity, but is more reliable if metabolism relations are to be studied. The normal motility of the feces is from 6 to 20 hours, while on a milk diet it may vary from 36 to 48 hours. This factor may be of importance in judging of intestinal obstruction. The isolation of a stool under a special diet is of great importance, as the success of an investigation will depend upon the obtaining of a feces arising from a specified diet. Obtaining Intestinal Juice. Boas has shown that it is often possible, by the use of the stomach tube, to obtain sufficient intestinal secretion to permit of examination. His technic 1 Jour, of Physiol., vol 13, 1905, P. 45. THE FECES. 95 is, however, rather difficult, is trying to the patient and is often unsatisfactory. Boldyreff, Volhard, Faubel and, more recently, Koziczkowsky1 have shown that the intestinal juice may easily be obtained by the following method. The patient is allowed to take 150 c.c. of pure olive oil when the stomach is empty, perferably in the morning. At the end of three-fourth hour, during which time the patient must remain quiet in a sitting posture, the gastric contents are removed by means of a stomach tube. The material, obtained by means of this oil test-breakfast, readily separates into two layers, an upper of oil and a lower of the mixed gastric and intestinal secretions. This may then be tested for the various ferments in the usual ways. Trypsin is best detected by the method of Gross.2 Either the fluid obtained as above or a portion of the feces, rubbed up with three times its amount of a 1 to 1,000 Na2CO3 solution and filtered, may be used for the test. 100 c.c. of a o. 5 per thousand solution of casein in 1 to 1,000 Na2CO3 solution are treated with 10 c.c. of fluid mentioned above and the mixture placed in the incubator for eight to twelve hours. At the end of this period the addition of a few drops of a dilute (1 per cent) acetic acid should produce no precipitate in case digestion is complete, that is if trypsin is present. Functions of the Intestinal Juices. The pancreatic juice as excreted into the intestine is an alkaline fluid con- taining three ferments, trypsin which hydrolyzes protein, amylopsin which acts upon the carbohydrates, and lipase (steapsin) which aids in the digestion of fat. The trypsin appears to be excreted in the form of a zymogen which is activated by a second ferment, enterokinase, derived from the intestinal mucosa, while the lipase and amylopsin are active when secreted. It has been shown by Pawlow that the passage of free hydrochloric acid into the duodenum is a direct stimulant to the excretion of pancreatic juice. Bayliss and Starling believe, however, that the stimulus to the pancreatic secretion is not the free acid but is a ferment, secretin, which is formed by the action of the hydrochloric acid upon the intestinal mucosa. While the trypsin acts upon protein bodies, splitting them through various stages into the ultimate products, amino-acids and hexone bases, there is a second ferment, erepsin, discovered in the intestinal mucosa by Cohnheim, which acts upon the intermediate splitting products of pro- tein, such as the albumoses and pepton, carrying this conversion to the same lengths as does trypsin.3 It is interesting to note that we find in the intestinal juice of the infant a ferment, lactase, which hydrolyzes lactose into the simpler saccharids. The presence of the bile, which reaches the duodenum through the common duct, is necessary for the proper digestion and absorption of the fatty sub- stances of the food. Variations in this constituent may reflexly cause disorders in the secretion of the stomach, one finding very frequently a hyperacidity 1 Zeitsch. f. klin. Med., Bd. 68, 1909, S. 261. 2 Deutsch, med. Wchs., Bd. 35, 1909, S. 706. 3 See Schlecht and Wittmund, Deut. Arch. f. klin. Med., Bd. 106,1912, S. 517; Friedmann, Med. Record, vol. 81, 1912, p. 355; Abderhalden, Zeitsch. f. physiol. Chern., Bd. 78, 1912, S. 344- 96 DIAGNOSTIC METHODS. associated with obstructive jaundice. Whether the bile maybe assumed to have a disinfecting power must be left for the future. Were this influence actually present, it is hard to understand, why, in cases of long-standing exclusion of bile, we do not find catarrhal conditions as we do in cases of marked putrefactive processes in the bowel. Estimation of Intestinal Digestion. A careful chemical examination of the feces, coupled with a macroscopic and microscopic investigation, will give much information regarding the degree of intestinal digestion and absorption. However, such methods are time- consuming and not easily performed by the general practitioner. For this reason Sahli has introduced a method, similar to his stomach method, of investigating such activity. He employs glutoid capsules, which are made from gelatin hardened with formaldehyd. These capsules either do not dissolve in the gastric juice at all or only after considerable time, although they are quickly soluble in the intestinal juice. In these capsules is placed material which will not diffuse through the capsule wall and whose absorption may be studied from an examination of the saliva or urine. Sahli uses either iodoform or salol. In the former case 0.15 gram of iodoform is placed in a glutoid capsule, and given with an Ewald test meal. Under the best conditions (normal gastric motility, normal intestinal digestion, and normal intestinal absorption) the iodin reaction may be expected to appear, according to Sahli, in the saliva within four to six hours; that is, within one and one-fourth hours after the capsule has been dissolved by the pancreatic juice. Instead of iodoform salol may be used, being given in the amount of one-half gram along with an Ewald test meal. The reaction for salicyluric acid may be obtained in the urine within one and one-half hours after the capsule has been taken. Sahli gives the following results obtained by the use of this capsule: When the stomach contains neither free hydrochloric acid nor pepsin, the reaction is not delayed so long as gastric motility is good. In cases of diarrhea due only to an increased peristalsis without any marked disturbance of intestinal diges- tion, the reaction is either normal or even somewhat hastened. In other types of diarrhea characterized by an involvement of the intestinal chemistry or in- testinal absorption, the reaction is either absent or distinctly delayed or the capsules may be found undigested in the feces. This method also aids in differentiating an icterus due to occlusion of the ductus choledochus at its point of entrance into the intestine, in which case the digestion of the capsule may be interfered with, from one where the obstruction to the bile flow is higher up near the liver. This test may be of some presumptive evidence in the diagnosis of pancreatic carcinoma, although all cases of pancreatic carcinoma do not necessarily occlude the duct, in which case a positive result would obtain. Galli reports a case of carcinoma of the pancreas in which the glutoid capsule was dissolved, although no pancreatic juice was present in the bowel. THE FECES. 97 II. Macroscopic Examination. (1). Method. The macroscopic examination of the feces embraces not only the study of the physical characteristics, but also a recognition of various normal and abnormal substances. It is convenient, when examining the feces, to employ some form of a washing apparatus to separate the coarser from the finer particles. Boas has introduced a special sieve for such work which, however, does not have much advantage over the ordinary flour-sifter which Einhorn advises. Strauss' method of washing the feces by a current coming from below seems to be much the best of any of these types of apparatus. In the absence of any other equipment, an ordinary house-hold sieve of various-sized mesh will answer the purpose. Small amounts of mucus or connective tissue frag- ments may be recognized by rubbing up a portion of the stool in a mortar with a little water, when these substances will float upon the surface. In some cases it may even be advisable to place the feces in tall glass jars in which the stool is mixed with water and allowed to arrange itself in layers. If the stool is of ordinary consistency, it should be spread out in a more or less thin layer so that the larger particles may be easily recognized. If it be, however, a watery stool the contents are thoroughly mixed and examined as such. (2). Amount. The amount of feces excreted will de Fig. 20.-Boas' stool-sieve. (Hemmeter.) pend upon (i) the quantity and quality of the food; (2) the remnants of intestinal juices and debris; (3) the condition of the digestive organs, and (4) the bacteria. These factors are all important, each being dependent to a certain extent upon the other. The average moist weight may be considered as varying from 100 to 250 grams with a dry weight of 20 to 40 grams. This amount may, however, reach as high as 20 kilograms as in a case reported by Lynch. Number of Stools. The number of stools which may be passed in 24 hours is subject to very wide variation, even under physiological conditions, but is usually constant in the same individual. At least one stool a day should be considered normal. 98 DIAGNOSTIC METHODS. although many persons are accustomed to have only one movement of the bowels in 48 hours and others in longer periods. We must, therefore, consider each individual case before judging as to abnormality in the number of stools. Lynch1 has reported a case of a patient having only one stool each 100 days, while Geib and Jones2 discuss a case in which there was no stool during an entire year the patient at the end of that time voiding 32 liters of feces. A diarrhea' is said to exist when the stools are frequent and fluid. The frequency may vary from 2 to 50 in 24 hours, although a single liquid stool in 24 hours may constitute a diarrhea in a patient unaccustomed to having a movement every day. A normal stool is never fluid, so that the character of the stool is of more importance than the number, as individual peculiarities must be considered. A diarrhea may be due to increased peristalsis, to in- creased intestinal secretion, a diminished gastric secretion, or a decreased absorption from the bowel. The most extreme grades of diarrhea are observed in Asiatic cholera, dysentery and the summer diarrhea of infants, although a marked diarrhea does occur in enteritis, peritonitis, intestinal tuberculosis, and uremia. Some infectious diseases are more frequently associated with diarrhea than are others, but little of diagnostic importance is obtained from this symptom. By constipation is meant the infrequent and irregular movement of the bowels, associated with symptoms which are relieved by administration of laxatives. The habits of the individual must be taken into consideration before judging as to a real or apparent constipation. This condition is physio- logically a result of a sedentary life or of a diet lacking in elements which will stimulate intestinal peristalsis.3 Pathologically, we find constipation in cases of dilated stomach, occlusion of the bowel, atony of the bowel wTall, in con- ditions causing increased cerebral pressure, and in obstruction by pressure from without the bowel. It must be stated here that we frequently have an apparent diarrhea merely as a symptom of constipation high up in the bowel. Such being the case, it is better practice to administer a laxative in such conditions than it is to give an astringent. (3). Consistency and Form. The consistency and form of the normal stool vary considerably depending upon the nature of the food ingested. The stool is much softer with a purely vegetable diet, of which about 85 per cent, is water, than with an animal diet, of which only 65 per cent, is water. One differentiates, as regards consistency, a well-formed, a mushy, and a fluid stool, between which types there are many gradations, some stools being partly formed and partly fluid. Many factors may influence the amount of water present in the stool, such as (1) a lessened absorption of water from the intestinal canal, (2) intake of a large amount of water, and (3) an increased secretion from the intestinal glands. The consistency of the normal stool varies from day to day and can be 1 Thesis, Buenos Aires, 1896. 2 Jour AM.A.,Vol. 38, 1902, P. 1304 3 See Cannon, Jour. A. M. A., vol. 59, 1912, p. 1. THE FECES. 99 constant only when the patient is placed upon a standard diet. It may be abnormally too fluid or too solid, in the latter case being frequently voided in the form of very hard masses, known as "scybala." As a general rule, it may be stated that the greater the absorption of water from the intestine the more firm will the feces be and, in consequence, the more frequently will these scybala form. Stools are frequently observed of which the consistency is normal but of which the size of the cylinder is quite small. Such small "lead-pencil" sized stools have been supposed to be indicative of stricture of the lower bowel, but this condition is not necessarily present. Besides the water-content of the feces, the amount of fat, mucus, and vegetable residue has much to do with the consistency of the stool. One may differentiate the fat from water in the stool by placing a small portion of the feces upon a slide and pressing a cover-slip down upon it. If the cover-glass remains when the pressure is removed increased fat may be assumed, while if the softness of the feces be due to increased water the cover-glass will spring away from the feces. In some cases, especially those associated with achylia gastrica, the stools- are very frothy, indicating a marked bacterial decomposition. Such stools, should not be confused with those of the ordinary diarrhea or with the char- acteristic "rice-water" stools of Asiatic cholera, in which particles of mucus are readily detected. Such stools are never found associated with large fat contents. (4). Odor. The peculiar odor of the normal stool is referable to the presence of indol and skatol, Which arise from the putrefaction of protein material in the large intestine. Along with these we may find other substances, such as hydrogen sulphid, methane, and phosphine. The odor of the stool is much more marked following a meat diet than that after a vegetable diet; it is very slight on a milk diet and is practically lost in the fasting condition. If the processes in the intestine are of such a nature that fermentation of the carbohydrates exceeds the putrefaction of the proteins, the stool shows a distinct sour odor traceable to the presence of butyric or acetic acid. The odor in cases of acute and chronic diarrhea is frequently very slight, while that of the loose, watery discharges of cholera is peculiar and sperm-like, referable to the presence of cadaverin. In the diarrhea of children a distinct putrid odor may be present, although this is not necessarily the case. The so-called acholic stools have in themselves, according to Schmidt, very little odor, showing this property only when complicated by decomposition processes arising from the lack of bile. In cases of severe dysentery and carcinoma of the large intestine an intensely dis- agreeable odor is observed which differentiates these conditions from those associated with other types of decomposition. (5). Color. The color of the stool varies from a light brown to even a black, depending upon the kind of food, the residue of the intestinal secretions, the presence of 100 DIAGNOSTIC METHODS. pathologic products from the intestinal wall, and the administration of thera- peutic agents. The dark color of the normal stool is due to the presence of hydrobilirubin, which is formed from the secreted bilirubin by the reducing processes in the intestines. Bilirubin is found normally in the stool of a nursing child, being converted either into biliverdin or hydrobilirubin under abnormal conditions. This change from a light yellow infantile stool to a greenish one has much importance to the pediatrician. A well-formed stool is always darker in color than is the thin stool, which is equivalent to saying that the stool is darker the longer it remains in the intestine. The color of a stool under a meat diet is dark brown, with a vegetable diet a lighter brown, and following a milk diet a light yellow. This color which is traceable to the diet is shown best in those cases in which bile is excluded from the bowel. The dark color of the meat feces is probably traceable to the conversion of the blood- coloring matter into hematin and not to the formation of sulphid of iron which is so commonly stated. Food products may color the feces a characteristic shade. Thus, coffee may give a dark brown color, cocoa a brownish-red color, red wine a smoky black-brown color which has a shade of green. Chlorophyll- containing plants, such as spinach and lettuce, give rise to greenish shades. Occasionally the reduction process in the intestines may go so far as to convert the bilirubin into leucohydrobilirubin instead of into hydrobilirubin. This hydrobilirubin is identical with the urobilin found in the urine. Such stools may be practically colorless when voided, but will be converted into the dark brown normal-colored stool on contact with the air. Besides these pig- ments, biliprasin has been isolated from the feces by Fleischer, while Muller obtained cholecyanin. The variations in color of the stool following administration of therapeutic agents is frequently characteristic. Thus, after the use of calomel one ob- serves a distinct green coloration due to the conversion of the bilirubin into biliverdin. Bismuth preparations color the stool a distinct black, due to the formation of the oxid or sulphid of bismuth. Rhubarb, senna, santonin, and gamboge cause a distinctly yellow coloration which will change to a reddish tone in the presence of alkali. Iron compounds produce shades ranging from dark brown to black, while methylene blue gives rise to the production of a blue-green color and sandal-wood a reddish-violet color. Cases showing the presence of large amounts of mucus or of pus in the feces are characterized by a gray-white or yellowish-gray coloration of the feces. Such cases are seen in membranous colitis and rupture of an abscess, especially of the appendiceal variety, into the intestinal tract. In cases of syphilitic or carcinomatous ulceration of the colon or rectum this characteristic color tone may be more or less influenced by the presence of blood. Stools showing the presence of a large amount of fat are clay colored. That this coloration is due to the excess of fat, rather than to the absence of Pathologic Variations in Color. THE FECES. 101 bile, may be shown by extracting the feces with alcohol and ether, in which extraction the bile will be taken up along with the fat and will color the solvents. These acholic stools as they have been called occur both in cases associated with biliary obstruction and in those showing no obstruction. It would be better practice, therefore, to style these stools colorless, instead of acholic stools. Strumpell was able to obtain stools of a light brown color by feeding patients a diet containing*small amounts of fat, thus proving that the increased fat was more important than the diminution of bile, although this latter does account for some cases. This clay-colored stool may also be found in diarrhea, while in Asiatic cholera and dysentery the stools may be absolutely colorless. They have also been found in cases of leukemia, carcinoma of the stomach or intestine, tuberculous enteritis, and chronic tuberculous peritonitis. The cause of this lack of color may be the same unknown cause that produces the formation of leucohydrobilirubin spoken of above. In some cases we find a distinct golden-yellow or even a green color of the feces. This is due to the presence of unaltered bile, on the one hand, and bili- verdin, on the other. Biliverdin is usually found in abnormal decomposition processes in the intestine of the infant, while unaltered bile may appear in cases of increased intestinal peristalsis, in which the contents of the bowel are rushed onward before absorption can take place. This is probably the partial ex- planation of the green stool following the administration of calomel. Normally, bilirubin is not found in the intestine below the ascending colon. Such being the case, it may be possible to judge of the point of irritation :n a diarrheal attack by the fact that the higher up in the bowel the point of disturbance the more of this pigment will be found in the feces. Bilirubin indicates, therefore, an enteritis especially of the small, but also of the large intestine. This bile pigment may be found most frequently on cellulose material, mucus, muscle fibers, and fat. It is readily detected by rubbing up 2 or 3 c.c. of the feces with a concentrated aqueous solution of pure mercuric chlorid. This mixture is allowed to stand 24 hours and is examined microscopically thereafter. The fragments to which the pigment is attached will stain red if due to hydro- bilirubin, while those stained with bilirubin will show a green color. Naturally, in this examination chlorophyll-containing fragments must be excluded. The most favorable material for such examination is the mucus. A green color of the stools may also be observed due to infection with the bacillus pyocyaneus. Blood may give rise either to a distinct red color, a brownish-red shade, or even a black tint. If the blood be adherent to the scybalous masses or to the well-formed feces, it is usually derived from the rectum or anus and indicates hemorrhoids; if it be evenly distributed with the food material and is changed from a bright red to a brownish color, it indicates a hemorrhage in the stomach or high up in the small intestine, especially if the stool be solid; while an evenly mixed bloody feces of fluid character will usually point to the colon as the seat of the trouble. As a general rule, it may be said that the darker the color the more remote from the rectum will be the hemorrhage. Tarry black blood is 102 DIAGNOSTIC METHODS. seldom of low origin, usually indicating trouble in the stomach or duodenum, while fluid scarlet blood usually arises from the colon or rectum, although in some cases of typhoid fever the blood may be a bright red, although the hemor- rhage may be fairly well up in the bowel. In cases of intussusception the blood may appear mixed with serum, but with no fecal matter. In deciding as to the importance of blood in the feces one must naturally exclude that arising from food or from hemorrhages above the gastro-intestinal tract. Thus blood coming from the mouth, nose, throat, or lungs may pass into the stomach and out with the feces, making a mistake in diagnosis very possible. Further, blood arising from vaginal discharges, which may be mixed with the feces at the time of defecation, should be excluded. The detection of blood in the feces is more or less simple and direct in the fresh state, but when intimately mixed with the feces its recognition is a matter of some difficulty. In many cases of hidden or occult bleeding blood is never detected microscopically or macroscopically. It is practically useless to search for blood-cells in the feces, as rarely are perfect cells found, unless the blood is present in very large amount, many specimens showing no cells. The chemical tests outlined in the chapter on Blood serve very well for its detection with certain modifications. Guaiac Test (Van Deen's Test). A small portion of the stool is rubbed up with water and one-third of its volume of glacial acetic acid added. This mixture is well shaken in a test- tube and a few cubic centimeters of ether added. After thoroughly shaking this mixture, it is allowed to settle, when the ether, in the presence of blood, will have assumed a brownish color. In case the ethereal extract is not clear, a few drops of alcohol may be added. On adding to this ethereal extract a mixture consisting of equal parts of fresh tincture of guaiac and ozonized turpentine, a blue ring will form at the point of contact or a blue coloration will be seen throughout the mixture if the tube be shaken. This test is much more reliable in its negative phase than in its positive phase. The writer has frequently found positive tests for blood arising from the employment of tubes previously used with copper solutions or with nitric acid, so that he would advise the worker invariably to use either new or thor- oughly clean test-tubes when testing for the presence of blood. Moreover, this test is given by other substances, which are enumerated in the section on Blood and may react positively in the absence of blood. If the patient eats water- melon, potatoes or rice, has been taking iron, or the feces contain much pus, a distinct reaction may be present. In this test as well as in all of the other tests outlined, the presence of hematin arising from the meat of the diet must be excluded. This can be done only by placing the patient upon an absolutely meat-free diet for several days preceding the examination. The Schaer-Klunge Test. This test is very similar to the preceding, but is much more delicate, being THE FECES. 103 positive after the ingestion of only three grams of blood. It is even more important when using this test to exclude all hemoglobin and chlorophyll- containing foods for some days preceding the examination. The stool is rubbed up with water and treated as in the preceding test with acetic acid and ether. To this ethereal extract is then added a mixture of 1 c.c. of ozonized turpentine and 1/2 c.c. of fresh 3 per cent, alcoholic aloin solution. This may be prepared by dissolving what aloin will lie on the point of a spatula in 1/3 of a test-tube of 75 per cent, alcohol. At the line of contact a distinct red ring will be observed in the presence of blood in from three to five minutes. Careful work with this test has shown that fat interferes to some extent with its delicacy. It is, there- fore, customary to treat the feces with an equal volume of ether and to shake thoroughly to remove all the fat present. The ethereal solution is then poured off and the remaining fecal material mixed with one-third its volume of glacial acetic acid and 10 c.c. of ether, being then thoroughly shaken and set aside. A portion of this brownish ethereal extract, which contains the hematin formed by the action of the glacial acetic acid upon the hemoglobin of the blood, is poured into a thoroughly clean test-tube and treated as above described with the turpentine and aloin solution. If the tube be shaken after the red contact ring has formed the whole mixture will assume a cherry-red color. If the tube is allowed to stand for a few minutes the aloin solution may sink to the bottom, forming a distinct red layer beneath that of the ether and turpentine. A reaction to be positive should appear within 10 minutes as the aloin itself will gradually turn red under the conditions of the experiment if left for a much longer period. This test, as previously stated, is much more delicate than the guaiac test, especially when the feces are first extracted with ether. Charcoal instead of carmin should be used to mark the feces. The writer has found this test very reliable and very easy of application. This test has the advantage of excluding practically every other factor which might influence the tests previously given. A portion of the feces is extracted with ether to remove the fat and is then separated from the ethereal solution. This fat-free feces is then rubbed up with water and treated with glacial acetic acid and ether. The ether, as in the preceding test, takes up the hematin which is now detected by the spectroscope, showing the characteristic spectrum, namely, an intense narrow band in the red between C and D and a somewhat more definitely marked group of three broader bands, (i) in the yellow, (2) at the boundary between yellow and green, and (3) at the boundary between the green and blue, this last band being difficultly recognizable. In order to avoid confusion with the spectrum of methemoglobin or of chlorophyll, one may convert the hematin into reduced hematin (hemochro- mogen) by the addition of alcoholic potassium hydrate, water, and ammonium sulphid solution. The spectrum of this latter substance is characterized by the two bands in the green. Weber's Test. 104 DIAGNOSTIC METHODS. This test is probably the most sensitive and should be more frequently employed. It is very simple of application, is very reliable and has practically no fallacy, especially when the hematin is converted into the hemochromogen. This test is, however, not so delicate as the aloin test, so that small amounts of blood, as found in occasional cases of occult bleeding, may escape detection. Weber originally used the guaiac test with a modification of extracting the fat from the feces before testing with guaiac and turpentine. Schumm states that if the feces be thoroughly extracted with alcohol and ether most of the fat and urobilin will be removed and that under these circumstances Weber's test, either with the guaiac tincture and turpentine or with the spectroscope becomes much more valuable. Adler's Test. O. and R. Adler have introduced the use of benzidine as a test for the presence of blood. The stool is extracted with a mixture of alcohol and ether for the reasons above mentioned. It is then treated with glacial acetic acid and with ether as described in the other tests. This acid ethereal extract, which contains the hematin, is then treated with 2 c.c. of a saturated alcoholic benzidine solution and 2 c.c. of hydrogen peroxid (3 per cent.). In the presence of blood an intense green color appears. This test is almost too delicate for clinical work, as it shows the presence of 1 part of blood in 100,000 parts of water. However, it has a great value, as a negative test with this reagent will absolutely rule out of consideration the presence of blood. It is especially necessary in applying this test that absolutely every trace of hemo- globin be removed from the diet, so that any blood appearing in the feces may have some diagnostic importance.1 (6). Mucus. From the diagnostic standpoint the recognition of much mucus in the feces is of the greatest importance. Any amount of visible mucus should be considered pathological, although it is to be remembered that mucus may be increased physiologically as the result of hypersecretion, in which case it will appear as a slimy coating of the scybalous masses or as small adherent particles. Boas regards the mucus found after strong cathartics as normal, but this is to be questioned as the irritation may be sufficient to set up a mild hypersecretion. The mucus expelled with the meconium and, according to Lynch, even that passed by infants up to the second week of life should be considered normal. The fecal mucus is a true mucin, being precipitated by acetic acid, but dissolving in io per cent. HC1. The larger portions of mucus may be easily recognized with the naked eye, but the smaller bits are more clearly brought out by rubbing up the feces with water and holding the material, in a thin layer, toward the light. If the feces be well formed, the mucus may be separated from the exterior of the cylinder, as it is never found in the interior of a firm feces. In mushy stools the mucus 1 Floersheim has recently shown that prunes must be avoided in the test meals given when the stool is to be tested for blood bv the benzidine test. THE FECES. 105 is intimately mixed with the fecal material, usually in the form of smaller particles, the exterior type of mucus being of much larger flakes. Nothnagel has reporteda jelly-like consistency of a mushy stool in a case of jejunal diarrhea, in which the mucus was not derived from the intestinal wall, but came, probably from the bile. It is important that one be able to recognize mucus, as mistakes have been made in the presence of swollen vegetable tissue, fruit pulp, echino- coccus membranes, and even of parasites. The ordinary form in which mucus appears in the feces is in clumps, flocculi, or shreds with irregular margins. These pieces may vary in size from those just visible to those several inches in length. In some cases strips, tubes, ribbons, or macaroni-like pieces are observed, which are especially frequent in enteritis membranacea or mucous colitis. The amount of mucus passed may vary from a few flakes to an enormous mass, Bories having seen 120 grams in one movement. Occasionally one may see, especially in the stools of infants, masses resembling cooked sago granules or "frog spawn," which Kitagawa has identified as mucinous material. The consistency of fecal mucus varies from that of a jelly-like mass to one having the density of thin leather. The larger the piece the firmer it appears, although exceptions do occur. Pure mucus is usually glassy or jelly-like, certain inclusions changing its consistency. Cellular inclusions in the mucus change it to a paper-like mass, while some specimens appear tenacious due to absorption of protein material or to a diminution of its water-content. These inclusions, as well as those of fat or bacteria, will diminish the usual transpar- ency of the mucus. On remaining long in the bowel, mucus may take on the normal brown color of the feces due to hydrobilirubin, a dark orange shade from bilirubin, a greenish hue from biliverdin, or a red to reddish-brown tinge from blood pigments. The usual state of the mucus, however, is colorless. It may be stated generally that the most of the macroscopically recognizable mucus of the feces arises from the large intestine. Owing to its easy digesti- bility the mucus from the mouth, stomach, and upper intestines passes out only under conditions of great motility. The secretion of mucus in the small bowel is much less than in the large intestine. Of all recognizable forms of fecal mucus only the smallest particles arise from the small bowel, and then only when they are found in a fluid feces. The higher the secreting point the smaller will the particles be, as a rule. These upper intestinal flecks contain much detritus of digestion and half-digested cells or free nuclei and crystals in cellular form, frequently a few bilirubin granules being observed. The mucus from the large intestine, especially from the sigmoid, is large in amount, is jelly-like in appearance, and has ordinarily no inclusions. In cases of mucous colitis the mucus may be passed in the forms of the large strips or bowel-casts, appearing as pure transparent material grayish-white or bloody in color and having no inclusions. No fecal matter may be present. Such a condition is a pure secretion neurosis and becomes a true inflammation only when cell-inclusions are observed. It is to be said that mucus, even with a 106 DIAGNOSTIC METHODS. small inclusion of pus or blood, does not necessarily point to ulceration. On the other hand, one must not judge from the absence of mucus that no catarrh exists, as one frequently finds variations in such excretions. Microscopic mucus is much rarer than the macroscopic type, although both may be present at the same time. Microscopically, mucus appears as a structureless mass, characterized by irregular lines running through it and by a difficultly recognizable margin, showing a more or less transparent ground substance in which may or may not be found epithelial cells, pus-cells, blood- cells, bacteria, protozoa, food remnants, and crystals. If the mucus comes from the higher sections of the bowel the food remnants will predominate, while in that from the lower bowel the cellular elements are in excess. Mucus is stained with difficulty. For a successful stain the reaction must be neutral and very little admixture with foreign material is permissible. Thionin colors mucin a specific violet, while the other tissue elements are stained blue. Methylene blue and methyl violet stain it but slightly, while other aniline dyes color only the enclosed cells. lodin may give a diffuse yellow. (7). Pus. Macroscopically visible masses of pus of a gray-white color are occasionally found in the stools, but these may hardly be distinguished from mucus particles without the microscope. If in large amounts, attention should be directed to the perforation of an abscess into the bowel. Such pus is more or less intimately mixed with the fecal material. The passage of pus from the small intestine and even from the cecum is associated with such a marked decomposition that it can be recognized only with the greatest difficulty. The nuclei may still persist, but these resemble very closely those of the food cells. A few isolated leucocytes are usually present in the feces, as a result of diapedesis through the mucous membrane. A pure pus excretion is never seen in cases of uncomplicated catarrh, but in ulcerative processes of the large intestine and in many affections of the small bowel (dysentery, ulcerative colitis, syphilis, carcinoma, tuberculosis, and typhoid) pus-cells may be found in small masses. Casein flocculi should be differentiated from pus masses by microscopic examination, which will show fat droplets mixed with the albumin. (8) Food Remnants. The appearance of macroscopic amounts of food remnants in the feces is known as lientery and is dependent upon the nature and amount of the food, its method of preparation, the degree of its mastication, and upon the condition of the digestive organs. Much less undigested material remains on a meat diet than after a vegetable ration. In the former case we may find small bits of bone, cartilage, tendon, hair, feathers, skin, fish scales, and connective tissue; while in the latter we may observe cellulose-containing cells, such as those of cereal, cotyledonous and THE FECES. 107 leguminous vegetables, skins of fruits, nuts, and vegetables, grape seeds, cherry stones, etc. Cooking has a great influence upon the digestibility of any food. Boiling seems to be the best method, although much of the nutritive material is taken up by the cooking water. The method of roasting is such as to produce a more nutritive but a less digestible substance, unless it is carried only to the point of slight coagulation of the protein. The outside layers of roasted meats seem to be practically indigestible, according to the work of van Ledden-Hul- sebosch. Smoking seems to be the least desirable form of preparing meats. Vegetables become more digestible through the cooking process, owing to the bursting of the indigestible cellulose membrane. Those vegetables, such as lettuce, cucumbers, onions, turnips, cabbage, and radishes, which are eaten raw, appear in the feces absolutely unchanged. While great individual differences exist in the power of utilizing digestible or undigestible food, yet we must assume that any appreciable residue, especially following the Schmidt-Strasburger diet, is pathological. The digestive insuffi- ciency may begin in the mouth as a result of too little mastication. The im- portance of thorough mastication has been especially emphasized by Fletcher. Lack of digestive power of the stomach will have but little influence on intestinal digestion, providing the pancreatic secretion is sufficiently active, with the exception that raw or smoked connective tissue will be undigested and will appear in the feces. Lientery is much less frequent in motor disturbances of the stomach than in those of the intestines. Perhaps the greatest amount of undigested residue is seen in cases of perforation of an ulcer or carcinoma of the stomach into the intestine, a direct communication being established between the stomach and colon. Naturally, in cases of increased intestinal motility, one will find more food residue in the feces than under normal conditions. This motor insufficiency is especially answerable for the appearance of undigested starchy remnants, as, even under the most severe conditions, starch digestion is not interfered with to any extent in the bowel, providing mastication and preparation of food have been sufficient and the succus entericus has not been so diminished by catarrhal processes as to permit of fermentation. The direct irritation from the hard particles of cellulose may cause an increased peristalsis and, hence, directly lead to lientery. Marked disturbance of protein and fat digestion, evidenced by an intense lientery, will occur if the pancreatic secretion be insufficient. This is even more marked if gastric disturbance coexists. A lack of bile will cause an exclusive disturbance of fat digestion, which is seldom manifested by the appear- ance of macroscopic particles of fat, but is characterized by the typical clay color of the stool. The final factor which influences lientery is the insufficiency of the absorptive power of the intestine. This increases the food residue by retarding digestion, according to the law that an accumulation of the products of ferment 108 DIAGNOSTIC METHODS. activity will prevent further action of the ferment, as well as by holding within the bowel the products already digested. Only in exceptional cases can one judge of the specific factor at the bottom of a lientery, a boundary line between normal and abnormal being drawn with difficulty. It is, however, important that one be able to recognize food rem- nants both macroscopically and microscopically. The writer recalls a case in which the residue of orange and banana pulp was mistaken for a new parasite. This recognition of the nature of a fecal residue is especially important in the examination of the infantile stool. Normally, the nursing child shows no food residue in the feces, but frequently white flecks or clumps are observed, which may consist of casein but perhaps more frequently of fat or of soaps. The proper recognition of such particles will require both chemical and microscopic examination, so that one should not jump to the conclusion that all such sub- stances are casein, as is too frequently the case. The following discussion of the appearances of the various food residues will include both the macroscopical and microscopical examinations, as these are inseparable in practical work. (?4). Protein Residues. The appearance of muscle fibers in increased amount in the stool is known as azotorrhea.1 The muscle fibers appear as isolated pieces of different size and shape. The smallest pieces have a circular or oval contour, the medium-sized particles are jagged, while the larger masses have parallel sides and angular surfaces. These fibers are colored yellow or yellowish-brown by the hydro- bilirubin, but may be tinged with bilirubin or by foreign pigments. The color will depend both upon the amount of pigment and upon the time the fibers have lain in the bowel. The smaller pieces may be entirely homogeneous, although these as well as the larger bits show both transverse and longitudinal striations. If the fibers are well digested, the longitudinal striations may be the only ones showing, and even these may disappear. No nucleus is visible unless the pancreatic secretion is entirely lacking. No specific micro-chemical reactions are known for these fibers. They may be tested with any of the color reactions for protein material. Connective tissue and elastic tissue fibers are occasionally associated with muscle fiber and may be recognized by their appearance. (a). Muscle Fibers. (&). Casein. This is found especially in the infant stool and is always pathological. These masses, known as curds, are more or less round clumps which vary in size from that of a pin-head to that of a hazel-nut. They are either pale or golden-yellow, the larger masses being a pure white inside. These curds always have a distinct yellow tone exteriorly, but the interior is pure white. If these par- ticles are pressed between a slide and cover-glass, they spread out like white 1 Also called creatorrhoea. See Wertheimer, Zeitsch. f. klin. Med., Bd. 76, 1912, S. 57; Tileston, Arch, of Int. Med., vol. 9, 1912, p. 525; also, Fronzig, Zeitsch. f. klin. Med., Bd. 77, i9r3, s- 4°. THE FECES. 109 cheese and show absolutely no structure microscopically. These masses show the protein reactions. Leiner's Test. A small amount of fecal matter containing these curds is spread on a slide and dried in the air. It is then fixed by heat and stained with a mixture of equal parts of a 0.75 per cent, solution of acid fuchsin and methyl green in 50 per cent, alcohol (dilute mixture 10 times with water). At the end of 15 minutes the slides are placed in distilled water and left for one hour. Casein and para- casein will take a pale blue or violet color, while similar substances will show a greenish tone. Microscopically, one may differentiate from these true casein particles (curds), certain products which are more or less normal in the stools of the child. These masses are smaller than the curds, being not usually over pin- head size, are more yellow in color, and appear under the microscope as clumps of fatty acid crystals or of fat droplets and bacteria bound together by mucus.1 (B). Fat Residues. Fatty substances are present in all stools, if there be any in the diet, either in the form of neutral fat, free fatty acids, or soap, especially of calcium and magnesium. If present in macroscopically recognizable amounts, the con- dition is known as steatorrhea. Neutral fat may be present in the form of white colorless clumps of different size and irregular in shape, some being globular while others are distinctly angular. The more usual form is the refractive, opaque, more or less yellow globule. Occasionally the fat may appear as a melted oil which hardens over the surface of a formed stool or gives the appearance of vaselin to a semi- solid stool. Neutral fat may be recognized by its ready solubility in ether and its black color on treatment with osmic acid and deep red color when acted upon by Sudan III or Scharlach R. Fatty acids appear either as scales or as crystals of varying form. The scales may not be distinguishable from those of pure fat except by their easy solubility in cold alcohol. The crystals are thin, delicate, curved needles, which run to a distinct point and are grouped in thick masses. Other types of fatty acid needles may occur, such as the small lancet-shaped plate or those closely resembling the soap crystal. These crystals are colorless and are soluble in cold alcohol, showing no stain with the above reagents for free fat. The fatty acid scales are, however, stained by these dyes. The soaps likewise appear in the feces as scales or as crystals. The scales are less refractive, more firm, usually more angular, and may be colorless or yellow-brown from hydrobilirubin or yellow from bilirubin. The crystals appear most frequently as uncolored needles, which are shorter, plumper, not so pointed as the fatty acid crystals, and are arranged in clusters. Schmidt has described a peculiar form of crystal, which he styles the "cracknel" form, a round type with a sunken center and a raised border. These crystals are ^ee Courtney, Amer. Jour, of Dis. of Child., vol. 3, 1912, p. 1. 110 DIAGNOSTIC METHODS. insoluble in ether, do not melt on warming, and are uncolored with stains. Treated with acids they form fatty acid crystals. Any condition which interferes with the proper absorption of fat by the mucosa or lymphatics will lead to steatorrhea. This condition may be observed, physiologically, after the ingestion of large amounts of fat, as in the oil treatment of gall-stones, but . we are little concerned with such findings. It is seen, pathologically, in cases of atrophy of the mucosa, in amyloid disease of the intestines, tuberculous ulceration, tubercular peritonitis, tabes mesenterica, and even in catarrhal enteritis. The peculiar, glistening, gray-white, pasty, acholic stool of steatorrhea is seen more frequently, however, in cases of biliary obstruction and of pancreatic disease. The fat in these cases of biliary stasis is in the usual form, that is three-fourths of the ether extract of the bile-free feces is present as fatty acids and soaps and one-fourth as neutral fat. In cases of acholia 50 to 80 per cent, of the fat will be unabsorbed instead of the normal 5 to 10 per cent.1 In the fatty stools of pancreatic disease, the fat is largely present as fatty acid, no soap being found. The association of steatorrhea with the absence of decomposition products, few bacteria, and the presence of maltose in the feces is much more indicative of pancreatic disease than is steatorrhea alone (Le Nobel). fC). Carbohydrate Residues. Starch may be present, in the normal feces, enclosed within plant cells which have resisted digestion, but well-preserved starch granules are never found normally. A few partially digested or complete colorless granules may indicate no abnormality, but many will point to a hyperacidity or to a disturb- ance in the small intestine, especially to an insufficiency of the succus entericus leading to the so-called "fermentative dyspepsia" (Schmidt). Starch may be detected by the addition of Lugol's iodin solution to a small portion of the feces spread upon a slide. A blue coloration will indicate starch, while a red tone will show the presence of erythrodextrin. The varieties of cellulose-containing substances found in the feces are as numerous as are the types of such food material. These have been discussed previously and cannot be here elaborated. Certain microchemical tests may prove valuable in the identification of such cellulose material. Cellulose treated with sulphuric acid and then with iodin solution gives a blue color due to the conversion of cellulose into amyloid. If a solution of zinc chlorid be allowed to act upon a suspected mass of cells, which have been previously treated with Gram's iodin solution, a purple coloration will be produced. If a neutral or weakly alkaline solution of Congo- red be added to cellulose, it will be stained a distinct red. Cellulose is soluble only in an ammoniacal solution of cupric oxid, known as Schweitzer's reagent. (o). Biliary Constituents. As a rule, the unchanged biliary acids, glycocholic, taurocholic, and 1 See von Hoesslin and Kashiwado, Deutsches Arch. f. klin. Med., Bd. 105, 1912, S. 576. THE FECES. 111 cholalic acids, are absorbed from the bowel so that they do not appear in appreciable amounts in the feces. Schmidt believes that cholalic acid is present in slight amounts in all feces and that all the bile acids may be increased in pathologic conditions. Bile pigments, on the other hand, are always present, chiefly in the form of hydrobilirubin, although bilirubin may occur. In pathologic conditions we may find, beside these two pigments, biliverdin, bilifuscin, bilicyanin, and bilihumin. The tests for these pigments as well as for the free bile acids and their salts will be considered in detail in the section on Urine, to which the reader is referred. In cases of cholelithiasis gall-stones of varying size may be found in the feces, although not in every case. It is, therefore, of the greatest importance that the feces be carefully examined in suspected cases for the presence of these concretions. This is especially the case whenever a severe, colicky, abdominal pain of doubtful origin exists. The feces should be well mixed with water and passed through a fine sieve which will retain any of the suspected particles. A single examination of the feces is not sufficient in such cases unless the stones be found. In cases of negative findings, the stool should be searched for at least two weeks following a suspected attack. According to Naunyn,only the very firm stones will leave the bowel, the softer ones breaking up into small bits in the bowel. These gall-stones vary from the size of a pin-head to that of a pigeon's egg and are found as small crumbling masses or as hard stones which either have a jagged surface or a smooth surface with characteristic facets, indicating the presence of many such stones. These stones consist for the most part of cholesterin mixed with the biliary pigment, or of compounds of calcium with the biliary pigments. These calcareous stones may be combinations of the various bile pigments and are always the hard facetted type, while the choles- terin stones are softer and may be colorless or tinged with bile pigment. The nucleus of these stones is usually a mass of organic detritus, in some cases being made up of clumps of bacteria, such as the typhoid bacillus and bacillus coli communis. A gall-stone may be usually recognized from its fractured surface, but frequently it becomes necessary to submit it to chemical examination for purposes of identification. From the clinical standpoint it is a matter of more or less indifference as to the composition of a gall-stone, so that the writer will refer to works on physiological chemistry for the methods of detection of the various constituents. In this examination for gall-stones the worker must not be deceived by the presence of extraneous substances, such as seeds, cherry-stones, fats and soaps of high melting point and of masses of impacted feces. In the treatment of gall-stones by the use of olive oil many of these soft fatty translucent masses appear in the feces which are in no way associated with gall-stones, although one may be deceived in supposing them to be true stones. 112 DIAGNOSTIC METHODS. (10). Intestinal Sand and Concretions. By intestinal sand we have reference to the small granules or masses of inorganic salts which appear in the feces. These masses are very small, may be spherical or irregular, are usually hard, usually have a reddish-brown or green color and consist of inorganic compounds mixed with organic detritus, especially with fat and bacteria. While most of this material is made up of ingested sub- stances, we may have in cases of neurasthenia and especially in those associated with mucous colitis the excretion of large numbers of these particles as the result of a true secretory neurosis. Myer and Cook1 have recently shown that the banana may be a source of this intestinal sand. The massing together of this intestinal sand may form intestinal concretions or enteroliths. As a rule, however, these intestinal concretions have as a nucleus some foreign body around which calcium and magnesium salts of phosphoric and carbonic acid or ammonium magnesium phosphate have been deposited. These enteroliths are usually hard, heavy, and round, being colored, as a rule, brownish. A second form of intestinal calculus, known as coproliths, are irregular in shape, usually softer than the enteroliths, and consist of inorganic material, mixed with inspissated feces. Neither of these types of intestinal concretion have any great pathologic significance, although the possibility must be granted that some of them might either form or lodge in the appendix and thus be accountable for an acute appendicitis. (11). Tissue Fragments. The examination of feces for tissue fragments is more or less unsatisfactory, owing to the fact that these fragments are difficultly recognizable after being partially digested by the juices of the intestinal canal. It not infrequently happens, however, especially in carcinoma of the lower bowel, that such frag- ments may be obtained and a diagnosis made possible by microscopic exami- nation. A diagnosis of malignancy would better not be made unless these tissue fragments show typical cellular arrangement, or at least, typical arrange- ment of the nuclei. The writer has seen a diagnosis of carcinoma made on the basis of a strip of mucus with cell enclosures and would, therefore, W'arn the worker to be on his guard against the possibility of such an absurd mistake. III. Microscopic Examination. This examination includes the search for many substances which have been described under the macroscopic examination, so that these need not be here considered. There are, however, some elements, both morphological and crystalline, which have not been treated. For any success whatever in the microscopic examination of the feces, great care must be taken in the selection of the material for examination. This is especially true when searching for parasites or their ova, as well as for the differentiation between various food particles and secretions from the intestinal 1 Amer. Jour. Med. Sci., vol. 137, 1909, p. 383 THE FECES. 113 wall. If the stool be soft and mushy it should be thoroughly mixed by stirring and suspicious partic'es looked for macroscopically. Fresh specimens as well as stained preparations are then made and examined both with the low and high power. In examining feces for bacteria it is always well to mix the feces with water and centrifuge for a few minutes. The bacteria will remain sus- pended in this process and the coarser fecal material will be deposited. The fluid is then poured off and mixed with equal parts of alcohol, the mixture being again centrifuged when the organisms will be found in the sediment. This procedure is the one adopted by Strasburger in estimating the proportion of bacteria in the dry stool. Various microchemical reactions are carried out with isolated fecal material and frequently are of great importance. In this procedure a few drops of reagent are allowed to flow under a cover-glass which covers the specimen. The staining material may be drawn under the cover-glass by placing a piece of ordinary filter-paper on the opposite side of the cover-slip. If the examination for amebae is to be made it is essential that the stool be kept warm, as the characteristic differentiating point of an ameba is its active motility. This can be done by placing the feces in a bottle which is surrounded by warm water until the examination is made. It is wise, also, to make the examination upon a warm stage, although this is not absolutely essential unless extended search has to be made for the parasite. There are no specific stains for the feces, the ordinary Lbffler's methylene blue serving very well for a general staining of bacteria, while characteristic stains must be used when searching for special organisms. Morphological Elements. Epithelial cells are present in every specimen of feces. These may be squamous in form and are usually found in the mucoid particles of the stool. They are especially present in rectal carcinoma and in ulcerative conditions of the lower bowel. The cylindrical type of epithelium is much the commonest form and is found, also, in association with the mucus. Many well preserved cells may be present if their source is in the lower bowel, but, as a rule, these cells show all types of degeneration. They occur in catarrhal inflammations of the intestinal mucosa and are rarely associated with pus-cells unless ulceration has taken place. If the irritation is in the small bowel these cells are always more or less digested and contain bilirubin particles in cellular arrangement, while if from the lower bowel the cell is usually intact. The presence of red blood-cells is determined more by the chemical ex- amination than by the microscopical. It may occasionally happen that these cells may be seen, but a good result will depend upon a fortunate selection of the material for examination. A few scattered leucocytes are practically always found in the feces,1 and assume pathologic importance only when present in large amounts as pus-cells, whose occurrence has been previously discussed. 1 Albu and Werzberg (Zeitsch. f. klin. Med., Bd. 74, 1912, S. 394) call attention to the frequent presence, in the feces of amebic dysentery, of eosinophile leucocytes and myelocytes. 114 DIAGNOSTIC METHODS. Crystals. Besides the crystals of free fatty acids and soaps previously mentioned we find large numbers of other crystals in the feces. Among these we observe neutral phosphates of calcium and of magnesium which appear as wedge- shaped crystals occasionally forming rosettes in the former case while in the latter the crystals are more in the form of rhombic plates. The ammonium magnesium phosphate crystals are practically always present in the feces, appearing either as the typical coffin-lid crystal or as irregular fern-like masses. These triple phosphate crystals were at one time supposed to be characteristic of typhoid stools, but like so many other things their importance has been exaggerated. Various other crystals of calcium compounds are observed, such as calcium carbonate, calcium sulphate, and calcium oxalate, along with calcium salts of unknown fatty acids, which have been described by Nothnagel as irregular, oval, or circular masses, either fissured or showing concentric striations, and being always bile-stained. The lactate of calcium is seen in the form of radiating needles arranged in sheet-like masses in the stools of children on a milk diet. Cholesterin crystals are found as thin, transparent, rhombic plates with notched corners. These crystals do not always appear in the typical shape, so that they should be tested by the addition of concentrated sulphuric acid when the cholesterin crystals will change from a yellow to a blood-red, violet, green, and finally blue color. The Charcot-Leyden crystal appears in typical form in the feces as a colorless, diamond, double pyramid-shaped crystal. The presence of these crystals is practically characteristic of helminthiasis, although they may indicate the presence of any parasite from the least harmful to the most pernicious. Hematoidin crystals occur as reddish-yellow rhombic plates or as groups of needles or amorphous masses in stools showing the presence of blood chemically. These have no especial significance and are not found as frequently as are the other blood pigments. After the use of bismuth preparations we find the oxid of bismuth appear- ing in the form of black irregular rhombic crystals with notched edges. Char- coal appears in the form of irregular black masses which are larger and not so rhombic in form as are the bismuth crystals. The examination of the feces for the various bacteria as well as the para- sites and parasitic ova will be discussed in a special section. Various sub- stances appearing in the feces have been mistaken for parasites or their ova and these also will be discussed in the later sections. IV. Chemical Examination. The chemical examination of the feces would naturally embrace both the qualitative and quantitative estimation of the products of digestion and decomposition as well as the estimation of the undigested portion of the food along with the products derived from the gastrointestinal canal itself. Clini- THE FECES. 115 cally, such work is rarely carried out and has questionable value from a diagnostic standpoint. However, in following the metabolism in any special case it is essential that the absorbed and unabsorbed portion of the food be known. This applies more particularly to the proximate food principles, protein, fat, and carbohydrate, although in some cases the estimation of the inorganic intake and excretion is of first importance. The writer cannot attempt in the scope of this work to go into great details regarding the chemical examination of the feces, but must limit himself to a few selected topics. Reaction. The normal reaction of the feces does not vary much from a neutral one, although it has a tendency to be slightly alkaline, owing to the presence of the alkaline secretions of the intestinal tract. The alkalinity of these secretions is diminished both by the combination of the alkali salt with the digestion products and by the absorption of some of this alkaline material into the blood. This reaction may at times be acid owing to the formation of lactic and butyric acids in the fermentative processes, while an increased alkalinity may be observed in cases of markedly increased ammoniacal decomposition. A pure meat diet gives, as a rule, an alkaline feces, while a pure carbohydrate or fat diet will have a tendency to form an acid feces. The reaction of the feces under pathologic conditions is of little importance, although in typhoid fever the reaction is somewhat more strongly alkaline than in almost any other condition. So much depends upon the diet as well as upon the condition of the digestive organs that no conclusions at present may be drawn from the reaction of the feces. Total Solids. The normal amount of feces passed in 24 hours ranges from 100 to 250 grams, of which about 75 per cent, is water and 25 total solids. Although the determination of the total solids of the feces has little of importance in itself, it is essential that one should know the dry weight of the feces in order that he may properly calculate the amount of the various substances in the dry feces. In determining the total solids in the feces, the specimen, preferably the 24-hour specimen or each movement separately, is placed in an evaporating dish, covered with a small amount of alcohol, and heated over a water-bath with frequent stirring. Small amounts of alcohol should be added as evaporation proceeds in order to hasten the drying process. It is, moreover, necessary to add a small amount (10 to 15 c.c.) of dilute sulphuric acid and to mix it thoroughly with the feces. This combines with the free ammonia, forming a non-volatile salt, and thus prevents loss of nitrogen. If the stool be rich in fats, it is wise to add a weighed amount of dry washed sand to make the mass more porous and thus permit of quicker drying. When the water has been driven off on the water-bath, the specimen may be placed in the drying oven at 1050 and left for several hours, after which it is placed in the desiccator and dried to constant weight. Knowing the weight of original substance taken and the weight of the 116 DIAGNOSTIC METHODS. dry substance, it is very easy to determine the percentage of total solids in the feces. An increased total solids of the feces is observed in most cases of con- stipation, while in diarrhea the solid residue is much diminished. Increased separation of fluid into the bowel and diminished absorption from the bowel are the factors regulating the fluid content. Nothing can be determined, how- ever, by the examination of the amount of total solids, as regards the pathologic conditions accountable for an increase or a decrease. Total Nitrogen. The determination of the total nitrogen must always be made in estimating the metabolism in any given case. In so doing it is likewise essential that one should know the absolute nitrogen intake, as without such a factor nothing can be learned from the excretion either in the urine or in the feces. The daily excretion of nitrogen in a fasting condition varies from i to 4 grams, on a mixed diet this may run from 3 to 7 grams, while on a vegetable diet the nitrogen of the feces may be as high as 10 grams, a pure meat diet yielding between 2 and 6 grams. In pathologic conditions this amount is practically always in- creased, due both to lack of digestive power and to decrease in the absorptive function of the intestines. Outside of metabolic experiments the absolute amount of nitrogen of the feces has no great diagnostic importance beyond show- ing some perversion of intestinal activity.1 In insufficiency of the pancreatic secretion, we are apt to find the greatest loss of nitrogen by way of the feces. The method for determination of total nitrogen will be given in detail in the section on Urine to which the reader is referred. The constituents which go to make up this total nitrogen are the various undigested protein bodies, such as albumin, globulin, nucleoprotein, nucleo- albumin, and gelatin, along with partially or completely digested products of protein origin, bacteria, and secretions from the intestines. These latter bodies embrace albumoses, peptone, the various amino-acids, and the hexone bases along with a certain amount of ammonium salts. The chemical properties of these substances, as well as the methods of their chemical detection and estimation must be looked for in works on physiological chemistry. The bacteria form a large percentage of the total nitrogen. It is rare, even in metabolic work, that the nitrogen partition of the feces is determined. Much more importance attaches, with our present knowledge, to such division of the nitrogenous material of the urine, which will be discussed by the writer at a later point. Likewise the products of abnormal decomposition of protein material, taking place in the bowel with the formation of such products as indol, skatol, and phenol, are rarely searched for in the feces.2 Our clinical knowledge of the importance of these substances is confined largely to their detection and esti- mation in the urine so that they must be passed for the time being. In the study of the metabolism of various cases of cystinuria, which is associated with 1 See McCrudden and Fales, Jour, of Exp. Med., vol. 17,1912, pp. 20 and 24; also Koplik and Crohn, Amer. Jour. Dis. of Children, vol. 5, 1913, p. 36. 2 See Moewes, Zeitsch. f. exp. Path, und Ther., Bd. 11, 1912, S. 555. THE FECES. 117 abnormal protein disintegration, certain diamines, such as cadaverin and putrescin, have been isolated from the feces by Udransky and Baumann, but these cannot be discussed at this time. Fat. The chemical estimation of the amount of fat is of importance only in metabolic work, as in the more direct clinical examinations thi? substance is detected by macroscopic and microscopic methods. It is, however, essential that the loss of fat by way of the feces should be known before a proper metabolic balance can be struck. The dried feces is treated with a small amount of i per cent, acid- alcohol and evaporated to dryness in order to convert any soaps which may be present into the fatty acids. The dry residue thus treated is then placed in a Soxhlet apparatus and extracted with ether for at least 72 hours. The ether, which has taken up the free fat and the fatty acids, is then evaporated and the residue of fats weighed. From this weight the percentage of fat may be deter- mined, and, knowing the original dry weight of the total 24-hour feces, the total fat lost by way of the feces may be readily calculated. The amount of fat in the feces will depend much upon the amount and quality of the diet. In fasting conditions the amount is usually about 1 gram. On diets poor in fat the amount in the feces may exceed that of the diet, thus indicating a loss of fat from the body. It seems to be a general rule that the higher the melting point of the fat of the diet the greater will be the loss in the feces. Thus on a diet containing only butter as a fat about 4 per cent, of the intake will be lost, while with pork fat the loss may be as high as 13 per cent. It is impossible to say, a priori, just what the fat content of a normal feces would be. With the ordinary diet of our country it would range from 2 to 7 per cent, of the intake, which should be somewhere about 100 grams of fat per day, representing a loss of from 2 to 7 grams daily. Carbohydrates. As previously stated, carbohydrates may appear in the feces either in the form of starch or of cellulose. Only in exceptional conditions do we find any of the monosaccharids present. Disaccharids, such as lactose and maltose, may be occasionally found, but only when there is a combination of insufficiency, both of a salivary and pancreatic ferment. The presence of any appreciable amount of starch granules must be considered pathological, while the amount of cellulose will depend upon the amount in the food as well as upon the preparation of the food and its mastication. Nothing is to be learned, from the clinical standpoint, by the determination of the absolute amount of carbohydrate in the feces. In the study of the utilization of food substances by the system and in general metabolic work, it is, however, necessary to know just how much of the carbohydrate intake is absorbed. This may be determined indirectly by subtracting from the weight of the dry feces the sum of the protein, fat, and ash. This result will represent a 118 DIAGNOSTIC METHODS. much higher figure for carbohydrates than the one obtained by direct deter- mination. In the direct determination much difference exists between the soluble forms and the insoluble cellulose. In the latter case we have to do with a useless form of carbohydrate and should, therefore, direct our attention rather toward the estimation of the amount of undigested or partially digested starch. We should determine the amount of cellulose and subtract this factor from the total carbohydrate obtained in the previous subtraction. This cor- rected figure will represent more nearly the true amount of carbohydrate than will the former, although for scientific purposes a direct determination is essential. The starch may be estimated by taking a weighed amount of the dry feces and treating with 50 c.c. of 10 per cent. HC1. This is then boiled for one-half hour in order to convert the starch and digested portions of starch into the monosaccharids. This acid solution is then filtered and washed with sufficient water to make the total approximately the same as that of the original solution. The filtrate is then neutralized with sodium hydrate and made up to exactly 100 c.c. in a'volumetric flask. The sugar in this solution is then determined by the methods outlined under Urine. This method, although not absolutely accurate will yield results sufficiently correct for ordinary purposes. More accurate results will be given by the Volhard-Pfluger method to be discussed later. The amount of glucose, as determined by either of these methods, will give, if multiplied by 0.94, the amount of starch in the feces taken. The percentage may then be readily calculated and the total quantity of the 24-hour excretion of starch determined. The fermentation method of Schmidt was advanced to permit of rough estimation of the presence of pathological amounts of starch, which, under ordinary circumstances, should have been digested. The principle of the method is the estimation of the amount of gas produced by the action of the intestinal bacteria upon the sugar which is formed from the starch by the intestinal amylolytic ferments. Prior to this determination the patient is placed upon the test diet outlined on page 93. The test is made as follows: Approximately 5 grams of the moist stool are placed in the vessel a which is then filled with water and the contents thoroughly mixed (see cut). The stopper is then placed in the fermentation flask in such a way that no air-bubbles are left. The tube b is filled with tap-water and closed with a small stopper without the inclusion of any air-bubbles. The tube c is now placed in position as shown in cut and the whole apparatus put in an incubator for 24 hours. The tube c has a small opening in the top so that water may be readily forced from the tube b into c by pressure of the gas produced. According to Schmidt 0.1 gram of starch Fig. 21. Schmidt's fermen- tation apparatus. THE FECES. 119 will cause the tube b to show about one-half its volume of gas. Normally, a positive result is said to occur when the tube is one-fourth to one-third filled with gas. The material in the fermentation flask a should be tested with litmus-paper after the test is complete to obtain the reaction of the mixture. If the gas formation is, as it should be, due to carbohydrate fermentation there will be a slight increase in the acidity of the mixture, while if it be due to protein putrefaction the reaction will show a slightly increased alkalinity. The various decomposition products of carbohydrate digestion and fermentation are rarely of importance in fecal examinations. These products consist of volatile fatty acids, lactic acid, saccharic acid, alcohol and aldehyde, and the conjugated glycuronic acids. Tests for these various substances will be found in various parts of this -work or in works on physiological chemistry to which the reader is referred. From the clinical standpoint the gases produced by the processes of fermentation and putrefaction in the bowel have little value. The same is to be said of the quantitative estimation of the inorganic constituents of the feces. These latter may be determined by methods outlined in quantitative analysis and may be passed by for the present. The various ferments found in the intestinal canal may be detected in the feces, but such examinations are at present of little clinical value. Methods of isolating these ferments and of determining their activity belong more to physiological chemistry than to clinical diagnosis and will, therefore, be neglected in this discussion.1 V. Bacteriology of the Feces. The bacteria of the intestinal canal are of many types, many of which are purely saprophytic, while others may or may not be pathogenic. The number of these bateria is usually enormous, Sucksdorff finding on an average 53,124,000,000 in the 24-hour specimen of normal feces. One may determine the amount of these bacteria by the method of Strasburger, who uses the following technic. Two c.c. of feces are rubbed up in a porcelain mortar with 30 c.c. of 1/2 per cent, hydrochloric acid. This mixture is then placed in centrifugal tubes and whirled for one minute. The bacteria will remain in suspension in the liquid which is poured off from the sediment. The sediment is again rubbed up with a little hydrochloric acid and again centrifuged, the liquid being added to the first portion. This procedure may be repeated until the fluid no longer becomes turbid on centrifuging. This acid solution, holding in suspension the bacteria, is then mixed with an equal portion of ordinary alcohol and placed in a beaker which is allowed to remain on a constant water- bath at 400 for 24 hours. At the end of this period a portion of the fluid may be evaporated and more alcohol added. This mixture is then placed in the centrifuge tubes and whirled for several minutes. The bacteria are now deposited and the supernatant fluid is poured off and mixed so that it may be again centrifuged. The bacterial sediment is washed with alcohol and ether 1 See Crohn, Amer. Jour, of Med. Sci., vol. 145, 1913, p. 393. 120 DIAGNOSTIC METHODS. in the centrifuge tube and is then placed in a weighed dish. This is dried at ioo° and again weighed. In order to determine the amount of bacteria as compared with the total feces, the dry weight of 2 c.c. of the fresh feces is determined as previously outlined. Knowing these factors, the percentage of bacteria in the dry feces may be easily calculated. Strasburger found that this was approximately one-third the weight of the dry stool and represented about 8 grams per day. As this dry feces is used for the determination of total nitrogen, we must bear in mind that the bacteria of the feces will represent about one-half of the total nitrogen of the feces. This fact is especially important in metabolic work. The method of counting the bacteria in the feces by cultural and plate methods has little clinical value and will be passed in this discussion. It is necessary, however, in attempting to isolate definite organisms from the feces, that cultural methods be adopted, and these will be briefly mentioned under the discussion of the pathogenic organisms. It is to be said that the bacterial flora of the intestine is so varied that very careful and long-continued work is necessary to isolate a special organism. From the stand-point of preventive medicine it should always be remembered that the feces contain large numbers of typhoid bacilli and cholera spirilla, so that measures should be taken to treat properly such ejecta as soon as voided. While the majority of the bacteria of the intestines are saprophytic, being introduced with the food or drinking- water, yet we find the bacillus coli communis as a normal habitant of the intestine. This organism is usually harmless, but may, under certain con- ditions, become distinctly pathogenic, many obscure cases of gall-bladder infection, for instance, being traceable to this organism. The presence of non-pathogenic organisms seems to be an essential for proper performance of intestinal function. This idea, originally advanced by Pasteur, has been denied by Schottelius, Nuttall, and Thierfelder, although their experiments extended only over periods of 17 days. These intestinal bacteria not only aid digestion, but also prevent a certain amount of abnormal decomposition, owing to the fact that they inhibit the development of foreign types to a large extent, especially under normal conditions. These normal bacteria may under certain conditions, however, become pathogenic, but only when the intestinal wall loses its continuity. In case the normal bacteria of the intestine become too numerous, the products formed by their activity upon protein material may be absorbed and bring about certain toxic effects.1 It is interesting in this connection to find that the bacteria are diminished in amount in chronic constipation, according to Strasburger. The toxic symptoms must, therefore, be referable to the absorption of other than bacterial products of decomposition, or the products have increased toxicity. For the best conditions to exist the symbiotic relations of the intestinal bacteria should be such that there are neither too few nor too many bacteria. Just how this is to be brought about is the problem of the clinician. 1 See Zuntz, Die Naturwissenschaften, Bd. i, 1913, S. 7. THE FECES. 121 The normal adult stool shows a preponderance of Gram-negative organisms (colon bacilli). Pathologically an increase in this type may be noted (cholera spirilla) or Gram-positive organisms may prevail. (a). The Cholera Spirillum. This organism, known as the comma bacillus, is about 2 microns long and 1/2 micron thick. It is very actively motile and has a single delicate flagellum at one end. It stains easily with the ordinary bacterial stains and is decolorized by Gram's method. The cultural peculiarities of this spirillum may be learned from any text-book on bacteriology. This organism is usually recognizable in the stools of Asiatic cholera, which are the characteristic "rice- ivater" stools. The blood serum of patients affected with cholera will show Fig. 22.-Cholera spirilla. (Pitfield.') very characteristic agglutination of cultures of these organisms. The blood is usually used in a dilution of i to 15, the reaction being observed in from 5 to 20 minutes. Closely related to this comma bacillus of Koch is the bacillus of Finkler- Prior. This latter organism may be distinguished from the spirillum of Asiatic cholera by its morphology, the organism being larger and thicker than the comma bacillus, and by the appearance of the stab cultures on gelatin. The cholera spirillum forms a typical funnel-shaped depression, while the bacillus of Finkler-Prior shows a stocking-like depression. The Finkler-Prior bacillus is found in cases of cholera nostras. It may be necessary for the absolute identification of these organisms to apply bacteriolytic tests with serum of animals immunized against a specific type of organism.1 (&). Typhoid Bacillus. This organism, discovered by Eberth, is so similar in morphology to numerous other organisms, especially to the bacillus coli communis, that simple staining methods do not suffice for its detection. The characteristic stool of typhoid fever is a copious watery stool, having a strong odor and an alkaline 1 See Public Health Reports, vol. 27, 1912, p. 371. 122 DIAGNOSTIC METHODS. reaction. This stool is known as the 11 pea-soup" stool, and may be tinged with blood and contain many pus-cells. The typhoid bacilli are medium-sized organisms with rounded ends, generally short, but sometimes long or thread-like and frequently showing faintly-stained sharply-defined areas in their protoplasm. They are actively motile and have both polar and lateral flagella. This organism stains with the ordinary dyes and is decolorized by Gram's method. Several methods have been advanced to permit of isolation of the typhoid bacillus from the feces. The writer selects, therefore, those that have proven the most satisfactory in his work. Fig. 23.-Bacillus typhosus, stained to show flagella (Oertel after Frankel and Pfeiffer.) Method of Drigalski and Conradi.1 Three pounds of minced beef are mixed with 2 liters of water and allowed to stand overnight. The beef is then pressed and the juice boiled for one hour and filtered. To the filtrate are added 20 grams of Witte's peptone, 20 grams of nutrose, and 10 grams of sodium chlorid. Boil this mixture one hour and filter. To the filtrate 60 grams of agar are added and the mixture boiled for three hours, one of which should be in the autoclave. Slightly alkalinize the mixture to litmus-paper, filter, and boil for one-half hour. To this hot agar solution, which should be now about 6o°, add the following litmus- lactose solution. Two hundred and sixty c.c. of litmus solution are boiled for 10 minutes, after which 30 grams of chemically pure lactose are added and the mixture boiled for 15 minutes longer. This litmus-lactose solution is added while boiling to the hot agar solution, the mixture being well shaken and again faintly alkalinized to litmus. Four c.c. of a hot sterile 10 per cent, solution of sodium carbonate and 20 c.c. of a freshly prepared 0.1 per cent, solution 'Zeitsch. f. Hyg. u inf. Krankh., Bd. 39, 1902, S. 283. THE FECES. 123 of crystal-violet B. (Hochst) in warm sterile distilled water are then mixed in. This medium may be poured directly into plates or kept in flasks. It soon hardens to a firm mass and does not become dry readily. The principle upon which the use of this medium depends is that in the presence of both lactose and protein the colon bacillus will first attack the milk-sugar, while the typhoid bacillus will act upon the protein. In the presence of litmus the colonies of colon bacilli become distinctly red, while those of the typhoid bacillus are blue. The crystal-violet inhibits the growth of many of the other organisms, especially of the acid-producing type. If the stool be fluid, as is usually the case, one series of two plates is inocu- lated with the undiluted stool, another with a stool diluted with 10 volumes of sterile normal salt solution, while other dilutions, such as 1 to 100 and 1 to 1000 may also be made. If the stool is solid it is rubbed up into a homogene- ous mass with sterile salt solution and the various dilutions made as above. In making the inoculations from the stools, the material is rubbed over the surface of the medium, the plates being left open to allow the surface to dry. The dry plates are then placed in the incubator at 370 and examined at the end of 24 hours. Any contamination of the media will be killed by the crystal-violet. It frequently happens that certain strains of the paratyphoid bacillus develop blue colonies in this medium. It is, therefore, necessary for absolute differentiation that the agglutination test described under blood be carried out. In this way the typhoid bacillus may be absolutely identified, especially in the presence of its well-known morphological characteristics. This method1 permits of rapid diagnosis and is especially of importance in the examination of stools of suspected "carriers." The culture medium is a modification of that of Endo. Plain nutrient sugar-free agar is prepared as follows: 1000 c.c. of cold tap water, 15 grams of powdered agar, 10 grams of Witte's peptone, and 3 grams of Liebig's extract of beef are mixed and cooked in the double boiler for one hour. Just alkalinize to litmus by careful addition of sodium hydrate solution, cook for 15 minutes and filter through absorbent cotton. This filtrate is then stored in flasks containing known amounts (100 c.c.) and is sterilized in the autoclave. When it is desired to use this media, add 1 gram of chemically pure lactose to each 100 c.c. and heat in the steri- lizer until the media is melted and the lactose thoroughly distributed. Now prepare a mixture of 1 c.c. of a filtered 10 per cent, alcoholic solution of fuchsin and 10 c.c. of a fresh 10 per cent, aqueous solution of sodium sulphite. Add 1 c.c. of this sterilized and decolorized fuchsin mixture to each 100 c.c. of the liquid lactose media and mix thoroughly. Plates are poured and allowed to harden (with the covers removed) in the incubator for 30 minutes, when they are ready for inoculation. A small portion of feces is emulsified in 10 c.c. of sugar-free2 bouillon Method of Kendall and Day. 1 Jour, of Med. Res., vol. 25, 1911, p. 95. 2 This sugar-free broth is specially prepared to remove the muscle-sugar.. Meat extract is inoculated with colon bacilli and incubated for 16 hours. Boil and strain through cloth. The filtrate is then made up as ordinary bouillon. 124 DIAGNOSTIC METHODS. and incubated for 1 hour at 370 C. This suspension is rubbed gently but firmly over the surface of the agar plates by means of sterile hooked glass rods and in- cubated for 18 hours at 370 C. The translucent, colorless, "dewdrop" colonies are removed to small tubes, which contain 1 c.c. of sugar-free bouillon and have been kept at 370 C. prior to use.1 Incubate for two hours and make agglutin- ation tests, using a known serum of high agglutinating power (see blood). (c). The Bacillus of Dysentery. Bacillary dysentery is distinctly different from the amebic type of dysentery. The bacillus dysenteriae, or Shiga's bacillus, is now generally recognized as the specific organism of this type of dysentery. Flexner found a similar bacillus in the dysentery of the Philippines, while Kruse has found practically the same type in Germany. In the United States Flexner and Harris find an organism which answers the description of the ordinary Shiga bacillus. The difference between the different organisms described is very slight, so that one may generally consider them as varieties of the same species. The Shiga bacillus is a short rod with rounded ends, very much resembling in morphology the typhoid bacillus. It does not seem to have very active motility as far as progression is concerned, although it does show a high degree of molecular motility. It stains with the usual dyes and is decolorized by Gram's method. The only sure method of identification of the various types of the dysentery bacillus seems to be the agglutination test. The fecal material is best obtained by curettage of the rectum, as these bacilli seem to be present in the mucus and are thus more easily con- centrated. A bouillon culture is then made from these particles and agar slants are made as soon as possible. At the same time agar plates are pre- pared and placed in the incubator for 24 hours. The Shiga bacillus is not a rapid grower so that practically no colonies of this organism will develop within 24 hours. The colonies which do develop at the end of 24 hours are marked on the plates and these are again incubated for 24 hours. From the colonies which develop at the end of 48 hours, tubes of glucose agar and litmus mannite agar are inoculated. If any of the tubes show fermentation at the end of 24 hours they are placed aside as not indicative of the Shiga bacillus. From these tubes showing no fermentation, litmus milk, litmus mannite, and bouillon are inoculated. The Shiga bacillus will first render the milk slightly acid, later changing it to alkaline, the litmus mannite remains unchanged with the true Shiga bacillus. (d). The Tubercle Bacillus. The examination of the stools for tubercle bacilli is not always satis- factory. The enormous number of bacteria of the feces may prevent a recog- nition of the tubercle bacillus even though it is present. In the feces we find certain organisms which are acid-fast, such as the timothy bacillus, which is verv closelv related in mornholosv and staining characteristics to the tubercle 1 Or the colonies may be transferred to Russell's medium (litmus-Iactose-glucose agar) as recommended by Lumsden (Pub. Health Reports, vol. 27, 1912, p. 789). THE FECES. 125 bacillus. We should be on our guard, therefore, lest we make a wrong diagnosis from the occasional presence of acid-alcohol-fast organisms. If these organisms are present on repeated examination and there are clinical symptoms pointing to such a trouble of the bowel, one may give a presumptive diagnosis of tuberculosis, remembering that the tubercle bacilli may have come from swallowed tubercular sputum. In selecting the material for examination, pick out the particles of mucus, especially those which are blood-stained or purulent. One may be much more certain of definite results if the antiformin method be used upon the feces, which has been rubbed up with distilled water (see p. 18). The staining methods are as usual. VI. Parasitology of the Feces. In the examination of the feces for parasites one should obtain the feces as fresh as possible. This is especially the case where the examination for protozoa such as the ameba is to be made. The feces should be kept in a warm vessel prior to the examination, as the motility of these unicellular organisms is shown only with great difficulty after they have become cold. The other types of protozoa are less sensitive to changes in temperature and show their active motility, providing the feces be examined soon after voiding. The formed feces are not as suitable for examination for protozoa as are the more fluid stools, the types present in the former case more frequently being the resting forms and not the true actively motile organisms. It seems to be generally accepted that these organisms are the more easily found the more fluid, more mucoid, and more alkaline the feces. The particles for examination should be preferably the masses of mucus which can be found by careful search in the liquid stool. The organisms are best examined in their fresh condition, as the staining agents usually require fixation of the specimen and consequent death of the parasite.1 The following discussion of the intestinal parasites is taken directly from Tyson's modification of Braun's work. (i). Protozoa. The protozoa are unicellular animal organisms. These, although living occasionally symbiotically, are more usually found as isolated single organisms. A few of these organisms are sufficiently large to be detected by the naked eye, but the majority are minute and require the finer microscopic detection. They consist essentially of a mass of protoplasm (cytoplasm or sarcode), with differentiation for functional purposes (organelles) of a variable character, constancy, and prominence. In the ameba, for instance, the sarcode may be separable into an internal distinctly granular portion known as the endosarc, and a peripheral clearer portion known as the ectosarc; a cell-membrane in some instances is a well-marked feature, while in others it is absent; and in some of the free-living protozoa special external coverings of chitinous, siliceous, or chalky composition enclose the protozoon. Of the various parts 1 Ross (Further Researches into induced cell-reproduction and cancer, Philadelphia, 1911 and 1912) has advocated the method of embedding the suspected feces in agar-jelly which may be impregnated with the stain. See also, Smithies, Arch, of Int. Med., vol. 9, 1912, p. 736. 126 DIAGNOSTIC METHODS. the nucleus is, after the cytoplasm, the most constant, varying much in appear- ance, shape, size and number in the individual form (single nucleus of variable size and shape; double or dimorphic nucleus, a macronucleus of vegetal character, a micronucleus with creative function; polymorphous nucleus, multiple nuclear granules more or less widely distributed in the cytoplasm). Not uncommon examples of specialization are met in the contractile vacu- oles, in pigment spots, in mouth-like ingestion foci and their pits on the surface of many forms with relatively firm cell membrane, in the anus-like excretory points of the same form, or the peripheral motor organelles, in the sucking tubes of the suctoria, and in the hook-like fixation apparatus of the gregarines. Motile protozoa move in a variety of ways. The naked rhizopods move by a peculiar rolling due to currents in the internal substance of the cell or by the protrusion of the cell-substance as extensions or pseudopoda, these move- ments being always accompanied by change in the cellular shape of the animal. Ciliates and flagellates move through the activity of the special cuticular appendages known as ciliae and flagella. (a). Rhizopoda (Sarcodina). Amoebina. The ameba was first discovered in the large intestine by Lambl, although we are indebted to Lbsch for the first accurate description of this organism. He did not believe it to be the cause of dysentery, but regarded it as a secondary invader. Since that time much work has been done, a controversy having arisen regarding its specificity. In a great many cases of dysentery this organism is not found, but in its stead bacteria, especially the Shiga bacillus, are present in large numbers. This bacillary dysentery is something entirely different from the amebic type of this disease. It is true that in certain cases of dysentery as well as in normal individuals, amebae are found which are differentiated with great difficulty from the true amoeba coli. There seem to be two distinct types: one pathogenic, to which Councilman and Lafleur give the name amoeba dysen- teriae, which Lbsch styles amoeba coli, and which Schaudinn designates enta- moeba histolytica; a second type, which is non-pathogenic, has been styled by the first writers amoeba coli Lbsch and by Schaudinn entamoeba coli. Viereck1 has described a third form, which is pathogenic and is known as the entamoeba tetragena. In this form four distinct amebas develop within the cyst. Craig2 has reported this type as a cause of dysentery in the Philippines.3 One may sum up the points relative to the pathogenicity of the entamoeba histolytica as follows : (i) It appears in a form of dysentery which is anatom- ically characterized by peculiar ulcerations which are markedly different from the diphtheritic inflammatory processes of the bacillary dysentery. (2) The more recent the case the more numerous are these parasites; (3) they are depos- (a) Amoeba coli (entamoeba histolytica) 1 Beiheft z. Arch. f. Schiffs, -u. Tropenhyg., Bd. i, 1907, S. 1. 2 Arch, of nt. Med., vol. 7, 1911, p. 362; Amer. Jour, of Med. Sci., vol. 145, 1912, p. 83; Jour, of Med. Res., vol. 26, 1912, p. 1. 3 See Darling, Arch, of Int. Med., vol. n, 1913, p. 1; also, Whitmore, Arch, of Int. Med., vol. 9, 1912, p. 515. THE FECES. 127 ited in the dysenteric ulcers and tend to pass into the deeper tissue appearing as true tissue parasites; (4) they frequently deposit themselves in the liver caus- ing abscesses which contain these organisms in large numbers and practically no other infectious material; (5) by injection of amebae-containing feces into the large intestine of animals typical amebic dysentery may be caused. As these or- ganisms have thus far resisted successful attempts at pure cultures, experiments cannot, of course, be made with the single organism. It is to be borne in mind that symbiosis with bacteria is apparently necessary for the development of amcebse. The entamoeba histolytica is an actively motile, roundish, pear-shaped, oval, or irregular unicellular organism having an endosarc which is typically Fig. 24.-Amoeba coli. {Hemmeter.} granular and may contain leucocytes, red blood-cells, bacteria, particles of food, or pigments which the parasite has ingested, and shows the clear hyalin ectosarc, which is, perhaps, best seen in the pseudopoda. The pseudo- poda are the typical motile portions of the parasites, these projections being thrown out from any point of the periphery, the protoplasm seeming to flow into them and drawing the animal after it. The parasite may not move, but will change its external appearance by throwing out these pseudopoda in various directions. The nucleus is a homogeneous, little refractile, chromatin-poor, spherical mass, about 6 microns in diameter (the diameter of the organism itself ranging between 10 and 50 microns, the average being 35). This nucleus is not always clearly visible, appearing much more frequently in the animal killed by corrosive sublimate. In the granular endosarc one frequently sees several vacuoles which may or may not pulsate, the general opinion being that pulsation is absent, although change in shape is frequent.1 1 If one wishes to stain such preparations, the method of Darling (Jour. A. M. A., vol. 59, 1912, p. 292, and Science, vol. 37, 1913, p. 58) is the best. Stain with Wright's stain in the usual way and follow this with Giemsa's stain until the film has a purple cast. Then plunge this into 60 per cent, alcohol to which 10 to 20 drops of aqua ammoniae have been added. Dif- ferentiate in this way until the film has a violet color. Dry and examine. 128 DIAGNOSTIC METHODS. The stools of this amebic dysentery are thin and watery, show an alkaline reaction, and have a peculiar lime-like odor. Much mucus, blood, and occa- sionally many pus-cells are found for which reason the mucus should be selected for examination in case these organisms are suspected. Frequently one may obtain better specimens for examination by examining mucus which is obtained with a rectal tube. Q3). Entamoeba coli. (amoeba coli Lbsch). This parasite varies in size between 10 and 15 microns. The hyaline protoplasm of the pseudopoda is not distinctly differentiated from the ecto- plasm. It is opaque, gray in color, and its nucleus is sharply defined, being characterized by its richness in chromatin. The movements of this organism are not as rapid as those of the entamoeba histolytica, which does not show as active phagocytic power. According to Craig, about 65 per cent, of normal persons show these nonpathogenic entamoeba coli in the feces, especially after a dose of Epsom salts. Fig. 25.-Coccidium hominis, from intestine of rabbit: 1, A degenerate epithelial cell containing two coccidia; 2, free coccidium from intestinal contents; 3, coccidium with four spores and residual substance; 4, an isolated spore; 5, spore showing the two falciform bodies-X 1140. (Tyson after Raillietf (&). Sporozoa. This organism appears in the feces as an oval or spherical parasite about 22 microns long and showing a thin periphery. A large number of nuclei are usually observed. The infection with these organisms seems to arise from rabbits in whose intestines these parasites develop in large numbers. Coccidium Hominis (coccidium perforans; cystospermium hominis). THE FECES 129 (c). Flagellates. This organism was first studied by Marchand and Zunker and later elaborated by Grassi, Roos, and Janowski. It is probably identical with the one known as trichomonas vaginalis which may live in the vagina, the urethra, large and small intestine, the stomach, and may be found in the sputum. Various forms have been described as being found in the intestine, but they are in all probability the same organism; among these we find protoryxomyces coprinarius, monocercomonas hominis, cimano- monas hominis, trichomonas hominis, cercomonas coli hominis, and cercomonas seu Bodo urinarius. This is a colorless protozoon of a pyriform or spindle shape/ rounded in front and bearing three flagella which are apt to be merged at the base and easily lost, the posterior end pointed but not bearing a flagellum. It is from 20 to 25 microns in length and 8 to 12 broad. Along the body, starting from the base of the flagella, runs an undulating membrane in a somewhat spiral manner to the posterior end. It has a finely granular cytoplasm and at its anterior end a vesicular nucleus, behind which one or more non-pulsating vacuoles may be seen. At times this organism may be observed to assume an ameboid form, the movements of the flagella having then ceased and projections resembling pseudopoda being observed. (a). Trichomonas intestinalis. Fig. 26.- Trichomonas i n t e s t inalis. (Tyson.) Cercomonas hominis. This organism was first studied by Davine and has been found by many other workers. It is known under the names of cercomonas intestinalis, mono- cercomonas hominis, cimcenomonas hominis. The adult organism is a small, colorless, pyriform parasite, with round anterior end provided with one long flagellum and a pointed posterior end. It is 8 to io microns long and has no undulatory membrane, as has the trichomonas intestinalis. (v). Megastoma entericum. This organism was first found by Lambl in the feces of children. It is known under several names among which are Lamblia in- testinalis, hexamitus duodenalis, dimorphous muris, and megastoma intestinale. This parasite is a colorless pear-shaped protozoon with a rounded anterior end and a pointed posterior end bearing a pair of flagella. The anterior end has one side concave with a raised border or lip, one pair of flagella, arising at the anterior border of this disk-like concavity, and two pairs together from its posterior margin. The cytoplasm is finely granular and the dumb-bell shaped nucleus is anteriorly about the middle level of the concavity. Vacuoles are absent and solid inclusions are never observed. The length of these organisms is from Fig. 27.-Cercomonas hominis; A, larger and B, smaller varieties. (Tyson.') 130 DIAGNOSTIC METHODS. 15 to 16.5 microns, while the width is from 10 to 12.5 microns. The number of these organisms found in the feces may be very large. The surest points for their diagnosis seem to be the concavity and the dumb-bell shaped nucleus. The source of infection for man is the drinking of infected water. (d). Infusoria. Balantidium coli (paramoe- cium coli). This parasite is colorless, ovoid in shape, 70 to 100 microns long and 50 to 70 microns broad, having a finely granular cytoplasm containing frag- ments taken from the intestinal ma- terial, and a clear ectoplasm showing numerous longitudinal striations, 'j It is covered completely with actively motile cilia which are more dense about the funnel-shaped mouth which extends about one-fourth the length of the body. The nucleus is kidney- shaped and is usually accompanied by one or more accessory nuclei, while two or more contractile vacuoles are seen which pulsate to a slight extent. This parasite has been found in connection with various types of diarrheal affection and also in persons entirely free from intestinal symptoms. It may be present in the stools in very large numbers, being found especially in the colon, but in severe cases also in the jejunum. The hog is the most common host of this parasite, which at times shows marked patho- genicity (see Bowman1). Fig. 28.-Megastoma entericum, showing disc-surface and lateral views in larger figures, and three epithelial cells with attached examples to the right. (Tyson.) (2). Entozoa (Enthelmintha). (A). Platodes (Flat Worms). (a). Cestodes (Tape-worms) The cestodes are naked, flat worms of elongated ribbon shape, endoparasitic, at least in their adult stage, and in many instances in all stages, without a digestive canal, and always more or less distinctly divided into segments. The entire parasite, or strobile, may be looked upon as a colony of indi- viduals united in ribbon fashion from their mode of origin, for convenience in their development and functional performance; the various segments being derived by a process of constriction from the originally acquired parasite, which is spoken of as the head, nurse, or scolex of the strobile. A characteristic of the cestodes is the differentiation of two developmental stages: the first, cysticercus stage, in which the connective tissue or parenchymatous organs are invaded, and, Fig. 29.-Balantidium coli: a, Nucleus; b, vacuoles; c, cytostome, with pit and peristome; d, ingested material. (Tyson after Leuckart.') 1 Jour. A. M. A., vol. 57, 1911, p. 1814. THE FECES. 131 secondly, the development of the sexually mature animal in the intestine. The scolex obtains entrance as a larva into the intestinal tract of the host, becomes attached by a special fixation apparatus to the mucous membrane, and there develops into the adult parasite, forming the anterior extremity of the strobile in the developed worm. The head is usually a very small and inconspicuous object of globular, pyriform or club shape with a short posterior extension, spoken of as the neck. In the middle of the frontal face of the head there is often a small prominence, known as the rostellum, about which may be arranged in one or more rows, as one of the means of fixation of the parasite, small booklets as a crown. As more constant means of fixation the head is provided with two or four suckers, rounded or linear depressions with more or less definite lips. Back of the head by a process of constriction from the neck, the segments, also known as links or proglottids, arise, the newest form always being placed between the neck and the next older link. Thus the older segments are always separated more and more from the head by each newly formed proglottid, each as it grows older and recedes further from the neck developing in size; the length of the strobile being thus dependent upon the two factors, growth of the individual length and the new formation of segments. These new segments as they are first formed are usually very short and pro- portionately broader, but as they increase in size with age they generally enlarge especially in their long diameter, and come to be more or less square. The number of these links may vary from three or four to several thousands, the length of some worms being 100 or more feet. The structure of each link and hence of the whole strobile includes an interior or matrix of an indeterminate connective reticular material, from which the various organs appear to develop and in which they are imbedded; over which are to be recognized exteriorly a delicate cuticle and beneath the latter, two layers of so-called muscle, the outer layer longitudinal and the inner transverse and circular. Aside from the common parts the various links may be looked upon as individuals. There is no digestive canal, all nutrition being obtained by the parasite from absorption of dissolved material from the fluids in the infested intestine. The only really highly organized parts are the generative organs, each link containing both male and female organs. The terminal links are the ones containing the most ova, while the links nearest the head are usually but partially developed. In their development the ova fill the canal of the oviduct more and more often causing the appearance of side pockets of more or less branching character. The terminal ripe links either actually containing the ova or, after discharge of more or less of the original number, are apt to become separated from the strobile and be carried with the fecal matter from the intestine. Either with or without intermediate development the embryo is in some way, by water or in solids, carried into the alimentary canal of a second host. Arrived in this situation, both by its own activity and by passive convection by blood and lymph streams, the embryo penetrates the intestinal wall and becomes deposited in one or other situation as a larva, bladder-worm, 132 DIAGNOSTIC METHODS. or cysticercus. This larval or cysticercus form is surrounded by a delicate connective tissue outer wall, derived from the host by a process of reactive in- flammation, within which lies the true bladder-worm. This is essentially the head of the future parasite. This cysticercus, after remaining a variable period in the tissues, is devoured with the flesh of its host by a third (definitive) host. (a). Taeniidae. This parasite, which is derived from infection through the cysticercus cellulosae of pork, has been also named taenia cucurbitina, taenia dentata, cystotcenia solium, and pork tape-worm. The average length of the strobile is 2 to 3 meters, occasionally reaching twice this measurement. The head is somewhat spherical or slightly tetragonal from the four rather prominent cup-like suckers with thick lips. The head varies from 1/2 to 1 mm. in diameter, while the suckers range from 1/4 to 1/2 mm. in diameter. It is provided with a short, thick, rostellum bearing a double crown of booklets, usually 28 in number. The neck is thin, about 3 cm. in length and is unsegmented. The proglottids number between 800 and 900, the fully grown and ripe segments measuring from 9 to 12 mm long and 5 to 6 mm. broad. The uterus consists of a large, median, longi- tudinal trunk with from 7 to 10 coarsely dendritic branches on each side. The ova are round or oval, the shell very thin but surrounded by an embryonic layer which is thick and shows distinct radiating lines. These eggs are usually of a brownish color and may show on their interior the booklets of the embryo. This parasite in its adult stage is practically limited to the small intestine of man, while the larval form has been found in swine, monkeys, dogs, etc. It has been shown that careful cooking and prolonged and thorough salting and drying of the meat will destroy the vitality of the cysticercus cellulosae. It is, therefore, plain that any meat should be more or less thoroughly cooked before being eaten. (1). Taenia solium. Fig. 30.-Head and neck, and ovum X 300, of tasnia solium. Embryophore surrounded by vitellus. (Tyson after Gould.} (2). Taenia saginata. This parasite is the beef tape-worm and is also known as the tania medio- canellata, tcenia inermis, and tcenia dentata. The adult worm varies from 3 to 8 meters in length, has a head from 1 to 2 mm. in diameter, tetragonal in shape without booklets or rostellum, with four cup-shaped suckers each 0.8 mm. in diameter and placed at the corner of the frontal face. The ripe seg- ments are from 18 to 20 mm. long and 5 to 7 mm. broad. The uterus shows a distinct median longitudinal trunk with 20 to 35 lateral single or dichotomously THE FECES. 133 branching and slender diverticula. The eggs are spherical with a thin shell surrounded by a thick radially striated embryonic shell. These eggs are from 30 to 40 microns long by 20 to 30 microns wide. Infection with this parasite is relatively rare in the United States. It is found almost exclusively in children, the infection occurring through dogs and cats, the larval form of the parasites being found in the body lice and fleas.1 This organism has other synonyms, among them being taenia canina, taenia moniliformis, taenia elliptica, dipylidium caninum, and dipylidium cucumerinum. The parasite is from 15 to 35 cm. in length. The head is small, rhomboidal, with a clavate rostellum surrounded by three or four crowns of booklets (48. to 60 in number), suckers rather large with radially-marked borders, and neck very short. The segments are from 80 to 150 in number, the older ones being 8 to 11 mm. in length and one to three mm. in breadth and often showing a reddish- brown color. The links frequently swell out in the middle so that the parasite has an appearance not unlike a chain of beads. A single uterus is common to the two oviducts, consisting of a network of tubes in which the ova lie in groups, filling small saccules, each containing 10 or 15 ova and surrounded by a reddish material which gives the color to the worm. The ova are spherical, 43 to 50 microns in diameter, and have a double wall. Within the wall one observes an embryo armed with booklets. (3). Taenia cucumerina. Fig. 31.-Head and neck of taenia saginata: A, retracted; B, extended. (Tyson after Gould,.') Fig. 32. Taenia cu- cumerina. (Tyson after Leuckart.) This worm is known as the "dwarf tape-worm" and is perhaps best known in Italy and Southern Europe, although it has been found in many cases in the eastern and southern portions of the United States. It has been called the tomia cegyptica, hymenolepis nana, hymenolepis murina, and diplacanthus nana. The infection with the ova is probably through the use of unfiltered water tainted with human or murine dejecta. It is most frequently seen in children, and inhabits the ileum, usually from the middle toward the ileocecal valve. The parasite is from io to 15 mm. in length and from 0.5 to 0.7 mm. broad, is provided with a subglobular head measuring 0.2 to 0.3 mm. in transverse diameter, shows four large rounded suckers and a large rostellum retractile into an infundibulum. The rostellum is surrounded by a single row of characteristic booklets, 24 to 30 in number and 14 to 18 microns in (4). Taenia nana. 1 See Lins, Wien. klin. Wochs., Bd. 24, 1911, S. 1595. 134 DIAGNOSTIC METHODS. length. The neck is rather long and slender, being followed by about 150 small proglottids, which are broader than long (0.4 to 0.9 mm. broad by 0.14 to 0.3 long). The ova are characteristic. They are round or oval in shape, 32 to 36 by 42 to 56 microns in size, and have two distinct membranes. At each pole Fig. 33.-Taenia nana: X 10. (Tyson after Gould ) of the inner membrane is seen a small protuberance from which springs a number of clear refractile threads, which are distributed in a waving fashion through the substance intermediate to the outer and inner walls. (5). Taenia diminuta. Synonyms-Hymenolepts dimtnula; hymenolepts jlavopunctata; tama leptocephala; tania flavopunctata; txnia minima; and tania varerina. The parasite is io to 60 mm. long, head small, globular, with four globose suckers situated close to apex, rostellum small, pyriform, and devoid of booklets. The segments number 800 to 1300 and are broader than long. Gravid uterus nearly fills the segments, showing as trans- verse line when not ripe. The ova are round or slightly oval, yellowish in color, double-walled, inner wall showing slight protuberances at poles, a layer of albuminous material being seen between the walls. (6). Taenia echinococcus. This parasite in its adult stage is met with in the upper part of the small intestine of dogs, wolves, and jackals. The larval form, known as the hydatid cyst, is found in man, although more frequently in the ox, hog, horse, dog, cat, rabbit, etc. This disease is rare in America, and is acquired through association with the dog more frequently than in other ways. The most common seat of hydatid disease is in the liver. The parasite is 2 to 5 mm. long, has a small subglobular head measuring 0.3 mm. in transverse diameter and bearing a rostellum with a double row of very characteristic booklets (28 to 50 in number) and four prominent cup- Fig. 34.-Head and neck of taenia diminuta. (Tyson after Braun.) Fig. 35.-Ovum of taenia diminuta. (Tyson after Braun.) THE FECES. 135 shaped suckers. The neck is short and rather thick; proglottids three or four in number, the last of which is usually longer than the rest of the worm put together. Uterus consists of a thick longitudinal median trunk with a few short lateral branches. Ova spheroidal with thin radially striated shells and containing a granular hexacanthus embryo. Length of ova 30 to 36 microns, transverse diameter 25 to 30 microns. Fig. 36.-Tsenia echinococcus: a, Adult; b, head from echinococcus cyst. On left a detached booklet, as seen in fluid from cyst. {Tyson after Coplin and Bevan ) Fig. 37.-Hydatid cyst, showing daughter cysts. In the lower part of field is a whitish mass containing parts of the walls of ruptured daughter cysts. The thick wall of the mother cyst is well shown. From liver of man, X j. {Coplin ) (&). Bothriocephaloidea Synonyms.-Dibothriocephalus latus; tania lata; dibothrium latum; bothriocephalus latissimus; fish tape-worm. This parasite is most commonly met in the human intestine, but may be found in dogs and cats. It is most common in central Europe and in the maritime countries of Europe, British Islands, and Japan. The examples found in America occur in foreigners, as a rule. The ova, which are usually in large numbers in the feces, require for their further development immersion in water. The liberated embryo is then taken up by fresh-water fish and con- veyed to man. Strobile 2 to 10 meters (20 in a few cases) in length, marked in ripe segments by brownish central rosette (uterus with ova). Head elongated, almond shaped, 2 to 5 mm. long and 0.7 mm. transversely, with two lateral grooves or bothridia as suckers. Neck variable according to degree of contraction. The (1). Bothriocephalus Latus. 136 DIAGNOSTIC METHODS. segments number 3000 to 4000 and begin about 50 cm. from the head. The anterior links are poorly defined, in their growth increasing slowly in length but markedly in breadth. The ripe links measure 2 to 4 mm. in length and 10 to 12 in width, with opaque brownish rosettes in the middle line. Uterus formed of a number of plicated tubes in the form of a rosette The ova are brownish in color, ellipsoidal in shape, 68 to 71 microns in length and 44 to 45 in transverse diameter, have a thin shell, and a lid which may be opened or closed. The contents of the ova are coarsely granular or mulberry-like. Infection with this worm is not always single, as high as 100 worms having been reported in the same individual. Many cases of infection with this parasite are associated with a high-grade anemia, which is dis- tinguishable from pernicious anemia only by the effects of removal of the oarasite. (2). Dibothriocephalus cordatus. This is a tape-worm of the same genus as the above and is parasitic in seals, being transmitted from them to man. It varies in length from 80 to 115 cm. Proglottids about 600, 7 to 8 mm. broad. The head is heart-shaped, 2 mm. long and broad. The ova are similar to those of the latus, but are a little larger in size. Fig. 39.-Diboth- riocephalus cor- datus: adult. (Tyson after Leuckartf Fig. 38.-Bothrio- cephalus latus. (Tyson after Leuckart.) Synonyms.-Diplogonoporus grandis; Krabbea grandis. » Strobile measures up to io meters in length. Proglottids short and broad, head, neck, and number of segments unknown. Uterus rosette-shaped with several loops on each side. Ova brownish, operculated, oval, 63 microns in length and 48 to 50 in width. Intermediate host unknown, probably fish. (3). Bothriocephalus sp. Ijima et Kurimoto. (/?). Trematodes (Fluke-worms). The various forms of distoma, which belong to this class, are more prop- erly hepatic parasites, although they and their ova may at times appear in the intestines and feces. A detailed discussion of these parasites will be taken up in the chapter on Parasites. THE FECES. 137 (B). Nematodes (Round Worms). The nematode worms are unsegmented, elongate, circular or nearly so in their transverse section, cylindrical or more or less delicately fusiform and tapering toward each end. They are with but few exceptions parasitic and include many important examples, which are parasitic in man. Intermediate hosts, so essential for the intermediate development of the flukes and tape- worms, are practically absent in the nematodes. (a). Ascaridae. (1). Ascaris lumbricoides. This is the common round worm or maw- worm seen so frequently in children. The number in a single host is usually small, but may be very large. Its habitat is the small intestine, but the eggs may occur in the vomitus as well as in the feces. The male worm is whitish to reddish-yellow in color; 15 to 17 cm. long, 3 103.5 mm. thick; elongate, fusiform; cuticle finely ringed; oral orifice terminal with three lips (one dorsal and the other two meeting in the median ventral line), each with fine denticulations on margins; at base of superior lip two papillae, one only at base of other two lips; posterior end terminating conically, curved ventrally, with two slightly curved, short, equal spicules projecting from the subventral cloaca; 70 to 75 papillae on the ventral face of posterior end, of which seven pairs are postanal. The female parasite is 20 to 25 cm. long, 5 to 5.5 mm. thick; anterior end and general appear- ance as in male; posterior end tapering, ending in a conical, pointed, straight tail, vulva at level of first third of body length (in a slightly depressed annular band); anus subterminal. The ova are elliptical in shape, 50 to 75 microns long and 40 to 58 microns broad; shell thick and transparent; stained yellowish with fecal material when in feces, but colorless in uterus; protoplasm unsegmented and coarsely granular; covered with a mammilated albuminous envelope, which may be lost. Fig. 40.-Ascaris lumbri coides: to left, male in lateral aspect; to right, female, ventral aspect, natural size. (Tyson after Raillietf 138 DIAGNOSTIC METHODS (2). Ascarismystax. Synonyms.-Ascaris canis; ascaris lumbricus canis; ascaris teres; ascaris caniculce; ascaris cati; ascaris tricuspidata; ascaris felis; ascaris werneri; ascaris marginata; ascaris alata; and fusaria mystax. The male parasite is whitish or slightly brownish; 40 to 60 mm. long, 1 mm. thick; anterior end usually curved, with lateral cuticular expansions making the end look somewhat arrow-like; mouth terminal with three nearly equal lips with denticulate margins; at base of superior lip two papillae, on inferior lip one ordinary and two minute papillae; posterior end curled and with A, Male; B, female; C, anterior extremity, enlarged and shown from dorsum to exhibit the lateral wing-like cuticular expansions; D, same showing in profile. (Tyson after Railliet.) Fig. 41.-Ascaris mystax, lateral cuticular alar expansions, and on each side of cloacal aperture 26 papillae, of which five are postanal. The female worm is 120 to 180 mm. long; anterior end and general appear- ance as in male; posterior end straight, terminating obtusely; vulva at anterior fourth of body length; anus subterminal. The ova are almost spherical; 68 to 72 microns in diameter; shell thin with thin albuminous envelope showing an alveolated surface.1 (3). Oxyuris vermicularis. Synonyms.-Ascaris vermicularis; fusaria vermicularis; ascaris grcccorum; pin-worm; thread-worm; seat-worm. The male is whitish in color; 3 to 5 mm. long, 0.3 to 0.4 mm. thick; cuticle transversely striated and at head end showing a vesicular swelling along the dorsal and ventral median lines; lateral lines distinct; mouth termi- nal, with three retractile lips; esophagus with distinct bulb; posterior end conical, curved ventrally, with six pairs of papillae and slight cuticular expan- sion on each side; one spicule hooked at free end. 1 See Beisele, Munch, med. Wochs., Bd. 58, 1911, S. 2391. THE FECES. 139 The female is io mm. long, 0.6 mm. thick; anterior end and general appearance as in male; posterior end straight, extended to a long mucronate tail; vulva at anterior third of body length. The ova are oval, flattened on one side with a characteristic asymmetry, 50 microns long and 16 to 20 broad; shell thin; colorless. This parasite inhabits the rectum and colon, but it may travel even into the stomach. The ova are rarely found in the feces, except in the mucus or about the anus. (bf Angiostomidae. Synonyms.-Anguillula intestinalis el stercoralis; leptodera intestinalis et stercoralis; pseudorhabditis stercoralis; rhabdonema strongyloides; rhabdonema intestinalis. This organism is found in two different forms, the first dioic and free, the second parasitic, as parthogenetic females. The parasitic form lives in the upper intestinal tract of man; is 2.5 mm. long, cylindrical, with pointed tail end, cuticle smooth; mouth simple with four lips; long, slender, cylin- drical esophagus reaching one-fourth of the length of the worm; anus close to tail; vulva at posterior third, containing yellowish-green oval ova, 50 to 58 microns long and 30 to 34 microns broad. The larvae develop in the intestine and are passed in the fecal material. These larvae are at first from 200 to 240 microns in length, but increase to two or three times this length. The larvae differ essentially from the parent in having a rhabditiform esophagus. In the discharged feces at about 300 C. these develop with one moulting of the cuticle to a free-living generation with separate sexes. In this free sexual generation the worms are smooth, cylindrical, and tapering, with pointed tail- ends; the mouth is the same as in the parasitic form; esophagus rhabditiform with its anterior portion long and with the pos- terior pyriform and containing a Y-shaped chitinous armature; anus at base of tail; male with tail curved and two spicules, body length 0.7 mm.; female 1 mm. long, with straight pointed tail, vulva a little back of the middle; ova few, yellowish, ellipsoid, thin-shelled, 70 by 45 microns in size, sometimes hatching in the uterus. The larvae of this generation look much as their Strongyloides intestinalis. Fig. 42.-Oxyuris vermi- cularis: to left, female; to right, male (considerably enlarged); A, anus; O, mouth; v, vulva. {Tyson after Braun.) 140 DIAGNOSTIC METHODS. free parents, are at first 0.22 mm. in length,but grow to 0.55 mm., then moult and assume a filariform or strongyloid character like that of the parasitic grand- parent. These gain access to the intestine of a new host in an unknown manner or shortly die. These worms may be found throughout the upper gastrointestinal tract, especially in the duodenum and upper part of the jejunum. The time elapsing Fig. 43.-Strongyloides intestinalis; on the left, a gravid female from human intestine (natural size 2.5 m.m). In the middle, a rhabditiform larva from fresh fecal matter, X 120; to the right, a filariform larva from culture, X 120. (Tyson after Braun.) between infections with the filariform larvae of the sexual generation and the appearance of the rhabditiform embryos of the parasitic type in the stools is between two and three weeks. The parthogenetic female types are usually called the strongyloides intestinalis, while the free sexual form is styled strongy- loides stercoralis. This form is found widely distributed in Indo-China, the East Indies, Africa, Europe, and North and South America. The mode of transmission to THE FECES. 141 the second host is probably through the means of unfiltered water or of unclean, uncooked vegetables. (c). Trichotrachelidae. Synonyms.-A scans tnchtura; tnchocephalus tnchturus; tnchocephalus hominis; tnchocephalus dispar; tnchocephalus mastigodes; whip-worm. The male worm is 35 to 45 mm. long; whitish; anterior three-fifth slender and thread-like; posterior two-fifth thicker, cylindrical, terminally rounded and curled; anus terminal; single spicule in a tubular sheath containing small spinules. The female parasite is 35 to 50 mm. long; shape as in male for front and body; posterior extremity straight, bluntly pointed terminally; vulva at beginning of thick posterior portion of body. 1 - The ova are very characteristic, brown, oval, thick- walled, with a colorless shining button-like protuberance at each pole. The eggs are 50 to 54 microns long and 23 microns wide, with an unsegmented yolk. Occasionally these eggs show variations in shades of brown, some being very much lighter than others. (1). Trichiuris trichiura. Fig. 44.-Tri- chiuris trichiura, natural size: A. Male; B, female. (Tyson.) (2). Trichinella spiralis (trichina spiralis). The male worm is 1.4 to 1.6 mm. long and 0.04 mm. thick; cylindrical in shape; anterior end tapering, posterior end gradually and slightly thickening and terminating in a bifid extremity with two lateral somewhat conical tail appendages; cloacal aperture between these, which form a sort of bursa; back of cloacal aperture two pairs of papillae. The female worm is 3 to 4 mm. long; anterior end as in the male; posterior end nearly of same thickness to tail, which is rounded; anus terminal; vulva at anterior fifth of body; viviparous. The larvae when born are 90 to 100 microns in length, obtuse anteriorly, posteriorly prolonged to a pointed tail; when encysted as "muscle trichinae" the larvae measure about 1 mm. long and 0.04 mm. in thickness, tapering anteriorly, more thick and obtuse posteriorly with complete organization as in the adult and showing the characters of the different sexes. This parasite or its larvae are rarely if ever found in the stools. In its adult sexual stage it infests the intestinal tract of man and a number of mammalians, gives origin to a large number of larval worms, after which the adults die. The larvae pass through the muscular wall of the intestine and are carried by the blood-current into various muscles of the host. Here they pass an indefinite encysted stage, a capsule forming around them and becoming calcified. Herrick and Janeway,1 Packard,2 and Cross3 found the embryo in the blood 1 Arch, of Int. Med. vol. 3, 1909, p. 263. 2 Jour. A. M. A., vol. 54, 1910, p. 1297. 3 Arch, of Int. Med., vol. 6, 1910, p. 301. 142 DIAGNOSTIC METHODS. the latter in blood from an ear puncture. The blood was laked with 3 per cent, acetic acid, was centrifuged, and the sediment examined as outlined by Staubli. The chief means of infection of man with the trichinella spiralis is through the eating of insufficiently cooked pork, especially ham. Fig. 45.-Trichinella spiralis: a, Gravid female "intestinal trichiura"; E, embryos; G, vulva; Ov, ovary; b, adult male, "intestinal trichiura"; T, testicles; C, young arva; d, larva in musculature; e, encapsulated larva in muscle. (Tyson after Braun ) THE FECES 143 (d). Strongylidae. (1). Uncinaria duodenalis. Synonyms.-Anchylostoma duodenale; strongylus quadridentatus; dochmius anchylostomum; sclerostoma duodenale; strongylus duodenalis; dochmius duo- denalis, and European hook-worm. The male parasite is whitish or blotched posteriorly with brownish when the intestine contains blood; 8 to 10 mm. long; cuticle finely striated transversely; tapering to a blunt point anteriorly and with head curved upon the dorsum so as to give a slightly hooked anterior end; on each side of the median line on the ventral side of oral border two hook-like chitinous teeth and on dorsal border on each side of the median line one less curved chitinous tooth; with a dorsal conical tooth extending along back of oral cavity from the base of the cavity; in the oral cavity about the esophageal opening a delicate armature consisting Fig. 47.-Anterior end, showing mouth parts, of uncinaria duodenalis, dorsal view. (Tyson.) Fig. 46.-Tail, with expanded bursa, of male uncinaria duodenalis. (Tyson.) of two dorsal and two ventral lancet-like pieces; posteriorly the body ends in an abruptly pointed tail in a copulatory bursal expansion of the cuticle, this having one dorsal and two lateral lobes; in the folds of the bursa one dorsal subdivided muscular ray, each division ending tridigitally, and on each side symmetrically placed an undivided dorsolateral, a divided lateral, undivided lateroventral, subdivided ventral and undivided small subventral muscular rays; cloacal aperture superterminal; two equal spicules. The female worm has the general appearance of the male and is shaped anteriorly like it; 12 to 18 mm. long; posteriorly tapering to a finely pointed tail; anus subterminal; vulva about the posterior third of the body length; two uterine and ovarian tubes. The ova are colorless, elliptical, thin-shelled with unsegmented or early segmenting material, 50 to 60 microns long and 30 microns broad. This worm has a wide distribution both in tropical and subtropical countries. Its habitat is in the duodenum, jejunum, and upper part of the human ileum. Infection with this organism is known as uncinariasis or anchylostomiasis. A particularly severe type of anemia is set up by this para- 144 DIAGNOSTIC METHODS. site both through the influence of loss of blood and the elaboration of toxic hemolytic material by the parasite. The eggs of this organism are frequently found in the stools and should be carefully searched for in every case of severe anemia. The mode of infection by these organisms may be direct ingestion of dirty water or unclean vegetables, but it is probable that the most frequent method of infection is through the skin, these larvae attaching themselves to the feet of people walking in infested sand or water. The larvae penetrate the skin and make their way through the blood- and lymph-currents to the lungs, whence they penetrate to the air-passages and are supposed to be carried upward toward the mouth by the bronchial mucus and are then swallowed. (2). Uncinaria americana1 (American Hook-worm). The male parasite differs from the former organisms discussed in being of smaller size (6 to 9 mm. long and more slender than the duodenalis), in the smaller size and more conical shape of the head, in having no booklets on the Fig. 49-Anterior end of uncinaria americana, showing mouth parts (dorsal view). (Tyson.) Fig. 48.-Tail, with expanded bursa, of male uncinaria ameri- cana. (Tyson.) oral rim, but instead on each side a large ventral and smaller dorsal chitinous lip, extending from the the rim toward the median line; in a greater prominence and projection into the oral cavity of the dorsal conical tooth; in the smaller size of the copulatory bursa, its dorsal lobe being subdivided and the ventral margin being extended so as to form an indefinite ventral lobe; and showing the dorsal muscular ray of the bursa divided, each division ending in a bipartite tip. The female worm differs from the duodenalis in being shorter and more slender (8 to 15 mm. long), with similar differences of the anterior end as above outlined for the male; vulva just in front of the middle of body length instead of at the posterior curve as in the uncinaria duodenalis. The ova of this parasite are somewhat larger (68 to 70 microns long and 38 to 40 microns broad) than those of the uncinaria duodenalis, but are other- wise similar. * Stiles has found that these parasites2 are the common cause of the frequent "anemia of the South." Smith believes that uncinariasis exists in every case 1 Also called Necator Americanus. See Stitt, Jour. A. M. A., vol. 59, 1912, p. 1706; also, Stiles, Pub. Health Reports, vol. 28, 1913, p. 7. 2 See Glover, Jour. A. M. A., vol. 58, 1912, p. 1837, for a discussion of uncinariasis in oriental immigrants. THE FECES. 145 Fig.50.-Parasitic bodies, ova, and larvae met in human feces; color approximate only (Tyson.) 1. Larval strongvloides intestinalis. 2. Ovum of fasciola hepatica. 3. Ovum of taenia nana. 4. Ovum of uncinaria duodenalis. 5. Ovum of uncinaria americana. 6. Ovum of taenia saginata. 7. Ovum of taenia solium. 8. Ovum of opisthorchis sinensis. 9. Ovum of opisthorchis felineus. 10. Ovum of cotylogonimus heterophyes 11. Ovum of taenia cucumerina. 12. Ovum of ascaris lumbricoides. 13. Ovum of dicrocoelium lanceatum. 14. Ovum of bothriocephalus latus. 15. Ovum of trichiuris trichiura. 16. Ovum of oxyuris vermicularis. 146 DIAGNOSTIC METHODS. in which "ground-itch" has occurred within eight years and that the disease is rarely ever present in those who have not had this condition during that period. This would point to the general transmission of infection through the skin. The uncinaria duodenalis has long been known as the cause of the so-called Egyptian chlorosis, tunnel-workers' anemia, brick-layers' anemia, and other conditions necessitating work in low-lying watery places. Pseudoparasites. It not infrequently happens that extraneous substances are found in the feces which very closely simulate a parasite or its ova in appearance. These substances are, for the most part, food residues and should be carefully differen- tiated by applying tests for cellulose, which will show in all vegetable cells. Stiebel has described, under the name of diacanthos polycephalus, a frag- ment of the woody portion of a bunch of raisins. Sultzer describes a mulberry seed as a vesicular worm under the name of ditrachyceros rudis. Bastiani considers the larynx of a bird which he found in the fecal material as being a biped worm, under the name of sagitula hominis. Scopoli regards a frag- ment of the trachea of a bird as an entozoon form under the name of physis intestinalis. A pupil of Moquin-Tandon describes a strip of lettuce under the name of slriatula, regarding it as a worm intermediate between the ascaris and the oxyuris (Guiart and Grimbert). Perhaps the most frequent pseudoparasite of the feces is the pulp of an orange, it shows in the feces in the form of large oblong masses terminated by two slender extremities, one of which ends in a sort of parenchyma. These vesicular masses are the large cells which secrete the orange juice and which are generally found intact in the fecal material. These have been frequently mistaken for hydatids or for parasitic ova. Certain spores, such as those of the truffle or of lycopodium, have not infrequently been mistaken for para- sitic eggs. Moreover, one may find in the fecal material the pollen of the coniferous plants which very closely simulates parasitic ova. The spines which form the down on certain fruits, such as raspberries, strawberries, peaches, and quinces, so closely resemble parasites that careful study is in some cases essential. It is, therefore, wise in all cases before pronouncing a finding as one of a parasite to be perfectly sure of your ground. BIBLIOGRAPHY. i. Braun. Die tierischen Parasiten des Menschen Wurzburg, 1908. 2. Briancon. De L'Ankylostomiase. Lyon, 1904. 3 Hemmeter. Diseases of the Intestines. Philadelphia, 1902. 4. Lynch. Coprologia. Buenos Ayres, 1896. 5. Musgrave and Clegg. Amebas: Their cultivation and etiologic significance. Manila, 1904. 6. Nothnagel. Beitrage zur Physiologic und Pathologic des Darmes. Wien, 1895. Krankheiten des Darmes. Wien, 1900. 7. Von Oefele. Technik der chemische Untersuchung des menschlichen Kothes. Leipzig, 1908. 8. Schmidt und Strasburger. Die Faeces des Menschen. Berlin, 1910. 9. Van Ledden Hulsebosch. Makro- und mikroskopische Diagnostik der mensch- lichen Exkremente. Berlin, 1899. PLATE VII. "Vegetable Cells found in Feces. (After Schmidt and Strasburger.) CHAPTER V. PARASITES. I. General Considerations In the previous portions of the work, the writer has introduced those parasites which are more particularly related to the parts under discussion. There are, however, a large number of organisms which do not fall naturally within the scope of any of the chapters outlined for this book. These will, therefore, be discussed in general without much regard to distinct classifications, as many of the subdivisions have been treated previously. This discussion is taken, for the most part, from Tyson.1 II. Trematodes (Fluke-worms). Flukes are naked and unsegmented flat worms, usually of the shape of a leaf or of the tongue (occasionally pyramidal or elongated and more or less cylindrical), provided with incomplete digestive canal (without anus), possess- ing one or more suckers and occasionally booklets; with but few exceptions hermaphroditic and, as a rule, presenting a complicated series of metamorphoses in their development. In structure it is customary to speak of the surface upon which the genital pore opens as the ventral; this surface commonly shows also the orifices of the mouth and one or more suckers. On the dorsal surface in many occurs the opening of a small canal, spoken of as Laurer's canal, of unknown function. The surface of the body is covered by a fairly thick and firm cuticle, often provided over variable areas with small spines or tubercles. Beneath the cuticle over its internal surface is spread the superficial muscular layer (not showing the structure of muscle of higher animals, however), with longitudinal, circular, and diagonal fibers, while within this along the borders is met the parenchymatous muscle. The general internal tissue of the body, spoken of as the parenchyma, is a fine reticular connective tissue which closely surrounds the various organs. The suckers of trematodes vary in number and arrangement on the anterior and posterior extremities, the ventral surface and its borders, and in a few cases also on the dorsal surface. Usually the oral opening is surrounded by such a sucker and in addition, in the forms likely to be met with in man, on the ventral surface some distance posterior to the oral sucker is a second, known as the ventral sucker or acetabulum, in the median line. Not infrequently in the lining of these suckers, on their lips, or on the cuticle close to the lips, chitinous booklets are to be found. 1 See Ward, Ill. Med. Jour., vol. 22, 1912, p. 417. 147 148 DIAGNOSTIC METHODS. The alimentary system consists of a mouth, opening m the oral sucker and situated terminally or on the ventral surface of the anterior end of the worm. This cavity continues into a dilated tube with thick walls, the pharynx, this extending posteriorly by a short, straight, and usually narrow esophagus, which divides in the anterior portion of the body into the two intestinal tubes or ceca. At the posterior extremity of the body is a small orifice, the excretory pore, which serves as outlet for a series of more or less complex canals for the convection of the fluid waste, the arrangement representing a low nephridial apparatus, while the mouth serves as an anus. The reproductive system is highly developed, showing numerous minor variations in the different genera and species. Adult flukes are parasitic upon a wide range of animal life, including the higher animals, fish, amphibia, reptiles, and birds, living as ectoparasites and endoparasites. The varieties affecting man are compara- tively few. The most common parts of the human body to be infested are the intes- tines, gall-ducts, respiratory tubes, and blood-vessels. (a) Fasciolidae. (1). Fasciola hepatica. Synonyms.-Distomum hepahcum; dis- tomum cavia; fasciola humana, cladocalium hepaticum; common liver-fluke. A comparatively large fluke, measuring 20 to 50 mm. long and 8 to 13 mm. wide, of leaf-shape, with anterior extremity pro- longed into a small cone; greatest width of body about the anterior third of length; light brown color; cuticle provided with alternating transverse rows of spines, ex- tending on ventral surface to the posterior level of testes, but not as far posteriorly on the dorsal surface; the oral sucker at the anterior end of cephalic cone, inclining to ventral surface, 1 mm. in diameter; ventral sucker near anterior end behind cephalic cone, 1.6 mm. in diameter; well developed larynx and short esophagus; intestinal branches extending nearly to the posterior extremity of the worm, ap- proaching the median line posteriorly with few median and numerous lateral branches; excretory pore at posterior extremity, with well developed Fig. 51.-Showing the sexual glands of fasciola hepatica; 5X1. O, oral sucker; D, intestinal ceca; Do, vitelline glands; Dr. ovary; Ov. uterine canal; T, testicles; Sq, "shell gland"; V, trans- verse vitelline duct; Gp, genital pore; 5, ventral sucker. (Tyson after Braun.} PARASITES. 149 system of excretory tubes; genital pore in the median line anterior to the ventral sucker; two large highly branched testes, mostly posterior to the transverse vitelline duct; ovary single, branched, lying in front of testes and to one side of the median line; the uterus coiled into a rosette, showing as a brown spot just back of the ventral sucker on the ventral surface; vitelline glands numerous, ranging along each lateral border from the level of the ventral sucker to the posterior extremity of the worm; vitelline ducts run- ning transversely at about the end of the anterior third of the body. The ova are yellowish-brown, oval, operculated, and measure 130 to 145 microns in length and 70 to 90 microns in width. This fluke is a common one in most mammalians and has a wide geographical distribution over the world, being found not infrequently in America. Its usual habitat is the gall-ducts, but it has been seen in the gall-bladder, intestines, in the portal and other venous channels and in subcutaneous cysts. The ova appear in the feces and are the chief means of diagnosis of this condi- tion. The embryo which develops from the ovum completes its developmental cycle in the body of a snail, the Limnsea truncatula. From the body of the snail the cercariae escape in the water and become attached to grass or other aquatic material, which is taken in by the animal. Infection in man is quite rare, about 32 cases having been reported in the literature. Synonyms.-Distomum buski; distomum crassum. (2). Fasciolopsis buski. Fig. 52.-Fasciolopsisbuski: a, ora sucker; b, acetabulum: c, cirrus pouch; d, vitelline glands; e, " shell gland "; /and g, posterior and anterior testicles; h, ovary; i, cecum; k, uterus. {Tyson after Braun.') The length of this organism is variable, ranging between 24 and 70 mm.; breadth 5.5 to 14 mm.; lance-shaped; narrowing more rapidly anteriorly than poste- riorly, maximal width about the middle of the length; no cephalic cones; brownish in color; cuticle without spines; oral sucker small and placed on ventral surface of anterior extremity; ventral sucker two or three times as large as the oral, placed near anterior end and showing a saccular distention extend- ing posteriorly; very short esophagus, back of the fairly developed pharynx; the two ceca without branches; genital pore at the anterior quarter of the 150 DIAGNOSTIC METHODS. acetabulum; cirrus pouch large; testicles branched; in posterior part of body one back of the other; uterus in anterior half of body, tortuously coiled; ovary at middle of length of body, to the right of median line; Laurer's canal present; vitelline follicles numerous along lateral margin from the level of the ventral sucker to the posterior extremity; the transverse vitelline ducts at the equator of body. The ova are brownish, ovoid, operculated, and measure 125 microns in length and 75 in width. Very little is known of the intermediate stage and of the host of Busk's intes- tinal fluke. This has been found in the small intestine of man and probably arises from eating infected food. (3). Opisthorchis Felineus. Synonyms.-Distomum conus; distomumlanceolatum (Siebold); distomum sibiricum; distomum tenuicolle; the European cat-fiuke. Variable in size according to the state of contrac- tion, 8 to ii mm. long and 1.5 to 2 mm. broad; yellow- ish-red and nearly transparent; flat and lanceolate; anterior end constricted and attenuated into a cone; posterior end more obtuse; cuticle without spines; oral sucker toward the ventral surface at anterior extremity; ventral sucker at base of cone about one-fourth of body length; pharynx and esophagus of equal length; ceca comparatively straight and unbranched, reaching nearly to the posterior extremity and often seen filled with blood; excretory pore terminal; its tubular vesicle winding in the median line between the testes and branching in front of the anterior testes; testes in the posterior part of body, the anterior four-lobed, the other five-lobed; cirrus and pouch absent; genital pore in the median line in front of the ventral sucker; slightly lobate ovary in the median line anterior to the testes; recepta- culum seminis prominent; uterus anterior to the ovary and testes coiled in the middle third of the body; vitelline follicles occupy about the middle third of the body, beginning anteriorly at the level of the ventral sucker. The ova are oval in shape, operculated, and measure 30 microns in length and 11 in width. This worm has been found in the gall-ducts of man and other animals, especially in Russia, Siberia, Hungary, and Japan. Fig. 53.-Opisthor- chis felineus; from liver of cat. 10 X 1. (Tyson after Braun.) (4). Opisthorchis Sinensis. Synonyms.-Distomum sinense; distomum spathulatum; distomum hepatis endemicum seu pcrniciosum; distomum hepatis innocuum; distomum japonicum; the Japanese or Chinese liver-fiuke. This parasite measures io to 20 mm. in length and 2 to 5 mm. in breadth; PARASITES. 151 long and lanceolate; reddish and nearly transparent when fresh; cuticle with- out spines; oral larger than ventral sucker, on the ventral face of anterior extremity; ventral sucker about one-fourth of body length posterior to former; pharynx and esophagus small and a bifurcation of the latter close to oral sucker; ceca unbranched, reaching close to the posterior extremity; excretory system as in the previous species; in the posterior fourth of body the two testes, one in front of the other, with from four to six dendritic branches; no cirrus or pouch; genital pore in the median line just in front of the acetabulum; ovary tri- lobed and placed just anterior to a large gourd- shaped receptaculum seminis both in the median line and anterior to testes; uterus well-developed and coiled in the middle area of the body between the ovary and ventral sucker; vitelline follicles in the marginal fields along the middle third of the body. The ova are brown, oval, operculated, and measure 27 to 30 microns in length and 15 to 17 in width. This fluke is comparatively common in Japan and eastern Asia and has been reported in America. It infests the gall-ducts and has been found both in the pancreatic duct and intestine. Little is known of the intermediate hosts. Many other types of these distomata are met with in the liver of various animals and are occa- sionally found in man. Among these we find the distomum lanceolatum, distomum heterophyes, distomum conjunctum, and amphistomum hominis (gastrodiscus hominis.) Fig. 54.-Opisthorchis sinen- sis: ventral surface, stretched; a, oral sucker; b, ceca; c, geni- tal pore; d, acetabulum; e, uterus; f, vitelline glands; g, ovary; h, receptaculum sem- inis; I, Laurer's canal; i, test- icles; k, excretory canal; m, ex- c retory pore. (Tyson after Braun.) III. Nematodes (Round Worms) Eustrongylus Gigas Synonyms.-Ascaris canis et martis; ascaris visceralis et renalis; strongylus gigas; strongylus renalis; eustrongylus visceralis. The male worm is red in color; 14 to 40 cm. in length and 4 to 6 mm. in thickness; slightly tapering anteriorly; mouth terminal; with a hexagonal orifice surrounded by six lips bearing papillae; cuticle thin and transparent, finely striated transversely; about 150 papillae along the longitudinal lines laterally; caudal extremity with an oval plate-like expansion serving as a bursa, its margin bearing small papillae and slightly indented dorsally and ventrally; single sexual spicule. 152 DIAGNOSTIC METHODS. The female parasite shows the general appearance and head end as in the male; 20 to 100 cm. in length and 5 to 12 mm. in thickness; caudal extrem- ity obtuse, straight, with anus subterminal; vulva 50 to 70 mm. posterior to mouth; single ovarian and uterine tube plicated from near the anterior end along the intestine nearly to the anus, then returning to the vulva near the anterior end. The ova are brown, ellipsoid, with thick shell marked by external cribri- form depressions, 64 to 68 microns in length and 40 to 44 microns in breadth. Fig. 55 -Eustrongylus gigas: female, natural size, in kidney of dog. (Tyson after Raillietf This worm, which is more common in the dog, has been reported a num- ber of times in man. It is the largest of the nematode worms and has its habitat in the pelvis of the kidney. Little is known of its life history. IV. Parasites of the Skin. Arthropoda. These are bilaterally symmetrical segmented animals whose segments do not correspond, but vary in structure, and which primitively bear upon each segment a pair of jointed appendages. The segments are often more or less fused, thus forming special body-regions which may themselves be more or less fused together as well. The covering of these animals is a comparatively thick and strong cuticle which remains pliable between the segments of the body and of the jointed appendages, but which commonly becomes hard and PARASITES. 153 shell-like from chitinous or calcareous material directly over the different body- segments and internodes of the jointed appendages. This arrangement re- quires that in the growth of the individual the firm external covering should from time to time be shed, such changes taking place periodically and being known as moults. While each segment in the primitive animal is provided with a pair of jointed appendages, these in the individual species are often lost from this or that part of the body, or may remain rudimentary and inconspicu- ous or may take on special features of structure from the assumption of special Fig. 56.-Acarus scabiei: A, female, dorsal view; B, portion of human epidermis, show ing burrows with contained ova and young acarians. {Gould,) function which causes their wide departure from the original and common type used for locomotion. The arthropods commonly reproduce by ovulation, the development of the embryo to the adult often showing more or less compli- cated metamorphoses. The true parasitic forms of the arthropoda thus far met in man are limited to the Arachnoids and Insects. (4). Arachnoidea. (a). Sarcoptes or acarus scabiei (the Itch Parasite). This parasite is oval in shape, is provided with horns and bristles, is barely visible to the naked eye, the male being from 0.2 to 0.3 mm. in length by 0.145 to 0.19 mm. in breadth; the female is somewhat larger, showing a length of 0.33 to 0.45 mm. and a breadth of 0.25 to 0.35 mm. The female lies at the end of a burrow in the epidermis, in situations where the skin is most delicate, as between the fingers, at the elbows, under the knees, and in the groin. In this burrow which varies from a few milli- meters to a centimeter in length the female deposits her eggs, after which she dies. The eggs hatch in from four to eight days, and in about 14 days the larvae are sufficiently matured to make their own burrows. The disease is communicated either by the clothing or by personal contact. To demonstrate the parasite, the burrow is opened with a needle and the female pressed out on a slide, which is then covered and examined. 154 DIAGNOSTIC METHODS. (b). Demodex folliculorum. This parasite is very small, varying in length from 0.3 to 0.4 mm. It is somewhat cylindrical, tapering to an obtuse point at the posterior end. This parasite has its habitat in the sebaceous follicles, especially of the face and nose. (c). Leptus autumnalis (Harvest-bug). This is a minute red parasite, from 0.3 to 0.5 mm. long, which has three pairs of legs, with rows of bristles upon its back and belly. It prevails in summer on grass and plants and attaches itself to the skin of man by its booklets. (B). Insecta. (a). Hemiptera. (i). Pediculus capitis (Head-louse). The male is from i to 1.5 mm. long, the female 1.8 to 2 mm. long. The color of the parasite varies somewhat with the race of its host. In the Caucasian it is gray with a dark border, in the Negro and China- men it is much darker in color. The eggs are 0.6 mm. in length and are attached to the hairs, forming the so-called "nits." These nits are whitish oval masses which are easily visible. This parasite, while usually found upon the hair of the head, may be found in other portions of the body.1 The symptoms may be severe or very slight. Fig 57. Demodex folli- culorum: from dog, enlarged. (Tyson after Braun.') Fig. 58.-Leptusautumnalis: enlarged. (Tyson after Braun.) (2). Pediculus vestimenti (Body-louse). This parasite is considerably larger than the former, being from 2 to 5 mm. long and whitish-gray in color, the back part of the body being wider than the thorax. The antennae are longer than are those of the head-louse. The eggs are from 0.7 to 0.9 mm. in length, about 70 being laid by each female. This parasite is found upon the clothing in which it deposits its eggs, especially about the neck, back and abdomen. (3). Pediculus pubis (phthirius inguinalis or Crab-louse). This parasite is smaller than the head louse, grayish-yellow or grayish- white in color, the male being from 0.8 to 1 mm. in length, the female about 1 Goldberger and Anderson (Pub. Health Rep., vol. 27, 1912, p. 297) show that typhus fever may be transmitted by this parasite as, also, by the body louse. See, also, Nicolle, Comte and Conseil, C. R. des Acad. Sci., Tm. 149, 1909, p. 486; Ricketts and Wilder, Jour. A. M. A., vol. 54, 1910, p. 1304; Goldberger and Anderson, Pub. Health Rep., vol. 25, 1910, p. 177; Ibid., vol. 27, 1912, p. 835; and Jour. A. M. A., vol. 59, 1912, p. 514. PARASITES. 155 1.12 mm. in length. The eggs are pear-shaped, from 0.8 to 0.9 mm. in length and from 0.4 to 0.5 mm. in breadth. This parasite infests the parts of the body covered by the shorter hairs, such as the pubis, axilla, eye-brows, and chest. Fig. 60.-Pediculus vesti- menti: x 10. {Tyson after Braun.) Fig. 59.-Pediculus capitis: X 15. {Tyson after Braun.) (4). Cimex lectularius (acanthia lectularia or Bed-bug) While, strictly speaking, the bed-bug is not a parasite of man, yet as its habitat is the bed, bedding, and walls of the sleeping-apartment of man, it may be considered as indirectly parasitic. It usually emerges at night from Fig. 61.-Pediculus pubis. {Tyson after Braun.) its lodging for the purpose of securing its nourishment in the blood of its victims. This parasite is reddish-brown in color, oval in shape, from 4 to 5 mm. in length and 3 mm. in breadth. These insects, if crushed between slides or as more usual between the hand and a part of the victim's body have a 156 DIAGNOSTIC METHODS. characteristic odor very much resembling kerosene. The blood is drawn from the victim by means of a long proboscis. The eggs are approximately 1.12 mm. in length and require about n months for their development to the sexually ripe insect. These eggs are retained in the crevices of the bed, floors, furniture, wall-paper, and other parts of the dwelling so that the complete removal of these eggs and parasites is a matter of some difficulty.1 That these insects have more or less importance from the standpoint of transmission of disease from one person to another must be remembered. Individuals vary in their susceptibility to the bite of the bed-bud, some being indifferent to it while others are maikedly affected by it. Fig. 62.-Pulex irritans: X 14. (Tyson after Braun.) (&). Diptera. (1). Pulex irritans (Common Flea) The male is from 2 to 2.5 mm. in length, the female as much as 4 mm. It is a red or brownish-red insect, having a laterally compressed body, an oral haustellum., serrated soft mandibles, a tongue sheathed in an inferior labium, and a pair of labial four-jointed palpi. Each of the triple segments of the thorax bears a pair of five-jointed double-clawed legs. The female deposits her eggs, not on the human being, fortunately, but in the fissures, crevices, or holes of garments or furniture which may be accessible. (2). Pulex penetrans (Sand-flea or Jigger). This parasite is a minute, brownish-red, egg-shaped insect which penetrates the skin of man. The female is the infecting insect and produces painful irritation and even suppuration. Vegetable Parasites. This organism is the cause of the disease known as favus or tinea favosa. This fungus invades the root sheaths, the bulbs, and the shafts of the hair (1). Achorion Schdnleinii. 1 See Rucker, Pub. Health Reports, vol. 27, 1912, p. 1854. PARASITES. 157 filaments of the scalp, but it also occurs upon the "non-hairy" portions of the skin and upon the nails. The spores gain access to the deeper layers of the skin and develop around the hair-shaft, forming a characteristic yellowish cup-shaped crust which has a peculiar mouse-like odor. In searching for this parasite, a favus crust is softened by the addition of a few drops of water or dilute sodium hydrate solution and placed upon a slide and examined with the high-power dry lens. The hairs may also be examined in the same manner or may be stained by methods outlined in the discussion on Tinea tricophytina. Fig. 63.-Pulex penetrans: young female, enlarged. (Tyson after Braunf The mycelial threads appear as narrow, flattened, ramifying, short or elongated, linear cells or tubes, which may be simple and empty, or be divided more or less regularly by transverse partition walls transforming the longer and simple into shorter and compound cells. The latter often contain in their cavities sporules clinging to either side, in which case the mycelial threads are termed sporophores. The conidia are encapsulated or are strung together like the beads of a necklace, and appear as round, oval, angular, or very irreg- ularly contoured bodies. These mycelial threads branch at right angles, the spores measure from 3 to 10 microns in diameter (Hyde). This organism is the cause of tinea circinata (herpes tonsurans, ring- worm of the body), and of tinea sycosis (hyphogenous sycosis, tinea barbae, ring-worm of the beard, barber's itch). The trichophyton is composed of spores which vary greatly in size, but which, as a rule, are somewhat larger than those of the type next to be discussed. They are frequently cuboidal, oval, or irregularly rounded, but their chief characteristic lies in their arrangement in lines or chains, extending up and (2). Trichophyton megalosporon endothrix. 158 DIAGNOSTIC METHODS. down the hair shaft. The mycelium is found without, but never within the hairs (Hyde). These fungi may be stained by the method of Morris and Calhoun. The hair is first washed in ether to remove all fatty debris; it is then put for one or two minutes in Gram's iodin solution and is stained after drying for from one to five minutes in gentian-violet. It is again dried and treated for a minute or two with the iodin solution and for an equal length of time in aniline oil containing pure iodin, after which it is cleared with aniline oil, washed in xylol, and mounted in Canada balsam. Fig. 64.-Achorion schonleinii, X 500 diameters. (Van Harlingen.) This parasite appears under the microscope chiefly in the form of a large number of round spores, irregularly grouped or massed about the follicular portions of the hair. Mycelial threads, large and branching, are often seen within the hair. The sheath of spores surrounding the hair is often continued upward for 1/16 to 1/8 of an inch above its exit from the follicle and may be recognized as a whitish or grayish coating of the hair. These mycelial threads are all within the hair proper, thus differing from those of the tricho- phyton which are never within the hair; after repeatedly dividing and subdivid- ing they terminate on the outer surface of the shaft in fine filaments, at the extremities of which are the spores. This parasite is the cause of the disease tinea tonsurans, or ring-worm of the scalp. (3). Microsporon audouini (Trichophyton Microsporon) PARASITES. 159 Fig. 65 -Normal hair X 900 Fig. .66.-Hair showing trichophyton endo-ectothryx X 900. 160 DIAGNOSTIC METHODS. (4). Microsporon furfur. This parasite is readily recognized by the microscopic examination of the scales scraped from the skin. Innumerable clustered spores, highly refractive and resembling in their circular and oval contours droplets of oil, are quite characteristic. The mycelial threads are not usually branched, but lie in a close network, among which sporophores are distinguishable, with conidia and terminal elements emerging at one extremity of the spore case. Both elements of this organism are more readily stained by the aniline dyes than are those of the trichophyton or favus. This organism is the cause of the con- dition known as tinea versicolor. Fig 67.-Hair showing microsporon audouini X 900, (5). Microsporonminutissimum This organism is the etiologic factor of erythrasma. It is characterized by the extreme delicacy and fineness of its threads and very minute spores. The threads are either simple, cylindrical bodies of variable size or they may exhibit partition septa, may divide dichotomously, and may terminate in hooked or knobbed expansions. The largest transverse diameter is 0.6 microns, in length'the mycelium presents the greatest variations. (6). Blastomycetes. These organisms may be found in the cutaneous eruptions of the skin in blastomycosis and may be described as follows, according to Montgomery and Ormsby. In unstained preparation the organisms appear as round or oval bodies with a double-contoured highly refractive capsule. Within the capsule, in many instances, granules or spore-like bodies can be distinguished. The PARASITES. 161 addition of a 1 to 10 per cent, solution of potassium hydrate to the specimen under examination facilitates the recognition of these bodies. In stained sections the double contoured, homogeneous capsule is usually separated from a finely or coarsely granular protoplasm by a clear space of varying width. Vacuoles of different sizes are found in some organisms. In both pus and tissue, organisms in pairs or in various stages of budding"are commonly seen. The Fig. 68.-Mycelial threads of blastomyces from old agar culture. (From photograph by W. A. Pusey.) * ' parasite, as a rule, varies in size from 7 to 20 microns, though slightly smaller and much larger forms occur in some cases. The organisms are readily obtained in pure culture from unbroken ab- scesses, from miliary abscesses in the borders of the cutaneous lesions, and from the miliary nodules and abscesses in the deep-seated organs. The peculiarities of the cultures of blastomycetes must be looked for in other works. Micro- 162 DIAGNOSTIC METHODS. scopically, the organism obtained in culture appears at first as a fine, branching mycelium with a few small spore-like bodies. Later a large, segmented, often pod-like mycelium appears, together with large, round, or oval bodies with bud-like projections. (7). Sporothrix Schenckii. This organism is the cause of sporotrichosis and was first recognized by Schenck, Hektoen and Perkins. Since its discovery a large number of cases have been identified in all parts of the world, especially in the middle- western portions of the United States. (See Hamburger, Jour. A. M. A., vol. 59, 1912, p. 1590; Sutton, Ibid., vol. 60, 1913, p. 115; Chipman, Jour, of Cut. Dis., vol. 30, 1912, p. 339; also, Adams, Journal-Lancet, vol. 32, 1912, p. 395.) The lesions in this condition resemble, at times, those of tuberculosis, while, occasionally, they appear more like those of syphilis. The chief characteristics are the sharply defined, painless, cutaneous or subcutaneous abscesses which follow the lymphatics and do not yield to the ordinary surgical procedures. It is a striking fact that attempts to obtain this sporothrix, either in sections of the affected tissue or in smears from the broken-down nodules, have usually failed. Cultural methods are essential. Some of the light yellow pus is taken, by aseptic methods, from the suspected lesion, is transferred either to glucose or maltose agar, and is kept at room temperature. After a few days small points appear from which circular growths with marked striations develop in 10 to 12 days. Later these may show a deep black pigmentation. Smears made from these cultures show a typical sporothrix, the mycelium being coarse (2 jj. broad) and forming a dense, branching, septate network. Spores (4 to 5 // long and 2 to 3 // broad) develop by budding either from lateral or terminal filaments or from the sides of the thread. The organism stains readily with the ordinary dyes and is, also, Gram-positive. BIBLIOGRAPHY. i. Carpenter. The Microscope and its Revelations. Philadelphia, 1901. 2. Daniels. Laboratory Studies in Tropical Medicine. Philadelphia, 1903. 3. Emery. Bacteriology and Hematology. Philadelphia, 1905. 4. Hyde AND Montgomery. Diseases of the Skin. Philadelphia, 1902. 5. Jackson. Tropical Medicine. Philadelphia, 1907. 6. Oertel. Medical Microscopy. Philadelphia, 1902. 7. Pusey. Diseases of the Skin. New York, 1907. CHAPTER VI. THE URINE. I. General Considerations. The examination of the urine is one of the most important features of clinical diagnosis. So constant are the physical and chemical properties of the normal urine that any marked abnormality is easily detected. The relation between the kidneys and the blood is so close that the kidneys soon excrete any abnormal substances which have found their way into the blood-current. F or this reason we find in the urine the abnormal products of perverted metabolism of the system or of special organs. It is true that the urinary findings may be, in any special case, secondary to those of the blood or of clinical examination, but in such conditions we may detect substances in the urine, which put us on our guard against making a specific diagnosis or point out the way to a correct differentiation. More or less marked changes in the character of the urine will occur whenever a pathologic condition exists anywhere in the system. These changes may not always be sufficient to be of direct diagnostic value, but in many cases may settle a differential diagnosis. In the urine we find excreted the products arising from the metabolism of the various proximate principles, both of the tissues and the food. Knowing the intake of such material, we are able from our examination of the urine to judge of the manner in which the system is handling the material brought to it. In recent years the study of the metabolism in various conditions has been so extended that determinations which a few years ago were unusual are now matters of almost daily routine. An examination of the urine will frequently reveal the presence of irregular digestive and absorptive powers of the intestines, through the appearance of certain abnormal products of protein decomposition. Moreover, the study of the nitrogen partition of the urine is taking on increasing importance from day to day, so that the estimation of the factors determining this division should be possible by any one attempting to follow the metabolic activity of the system in any specified condition. When the oxidative powers of the system are lessened, we find the urine showing abnormal products as an indication of such deficiency. These products are more or less characteristic and may be determined with a great degree of exactitude. Thanks to the work upon metabolism in diabetes, for instance, we now know that the glycosuria is clinically not of as great importance as is the presence of many other abnormal products associated with the sugar; in other words, a glycosuria must not be considered as identical with diabetes. 163 164 DIAGNOSTIC METHODS. The system has a definite disintoxicating power toward certain noxious substances, whether introduced from without or formed within. While the blood is of special importance in such processes as far as bacterial products are concerned, an examination of the urine will frequently reveal much information regarding the metabolic toxins or the medicinal poisons. The estimation of the conjugated glycuronic and sulphuric acids of the urine will throw much light on the degree of this activity. In this connection it may be mentioned that indican, a product of bacterial decomposition of protein in the intestinal canal, may be taken as a direct indicator of the degree of such decomposition, but that we must not assume that ail of the conjugated acids have such an origin. Besides these general indirect points of interest, an examination of the urine will often reveal a direct anatomical lesion of the kidneys. Time was when we regarded the mere presence of albumin in the urine as indicative of a kidney lesion, but we know that a thorough clinical examination is necessary before a diagnosis is possible. Albumin may or may not mean kidney trouble and may even be purely physiologic. Too much stress can hardly be laid upon the necessity of closely associating the urinary findings in any condition with the clinical symptoms of the case. The writer will have much to say later regarding the various abnormalities of the urine, but he wishes to impress at this point the fact that no finding, no matter how abnormal it may seem, should be con- sidered absolutely pathognomonic, without taking into consideration the clinical manifestations of the case. An albuminuria, glycosuria, cylindruria, or pyuria may mean one thing at one time and another at a second period, so that the worker is cautioned against jumping at conclusions. The urine will yield much information, providing the worker knows how to interpret his findings. A laboratory worker must be cautious in his attitude and report his findings without any attempt at interpretation, unless he is aware of the clinical history of the case in point. Frequently, however, a simple urinary examination may give much unexpected information and put the physician in a much better position to make an obscure diagnosis. The writer must refer to works on physiology for the various theories which have arisen from time to time regarding the mechanism of secretion of the urine. Suffice it to say at this point that the urine is excreted through the activities of the kidneys, the water and salts being secreted by the glomeruli, while the majority of the excretory products are eliminated by the vital or select- ive activity of the epithelium of the renal tubules. It is evident, therefore, that the physical and chemical characteristics of the urine will depend both upon the blood-pressure1 within the capillaries and the rate of flow through these vessels as well as upon the condition of the secreting epithelium. Collection and Preservation of the Urine. In all urinary examinations in which quantitative relations are to be studied it is necessary that a portion of the total 24-hour specimen be examined. This 1 See Lawrence (Amer. Jour, of Med. Sci., vol. 144, 1912, p. 330) who shows that the amount of blood in a unit of time is of much more importance than the pressure. THE URINE. 165 should be thoroughly mixed and carefully measured. The composition of the different voidings is so variable that no definite idea regarding the elimina- tion can be gained from a single specimen. If a mere qualitative examination is to be made, a single specimen may be studied. In chronic nephritis, for instance, the morning urine may show points of interest as compared with that voided in the evening after a day's activity. In diabetes it may be desired to study the effect of a carbohydrate meal upon the sugar excretion. This may best be done by an examination of the urine passed three or four hours after such a meal. The microscopic examination is best made as soon as possible after voiding, but i.t is to be remembered that much variation may be noted in the sediment of the different voidings. In making the 24-hour collection, the patient is instructed to empty his bladder at a specified time, preferably at 7 a. m. This portion is thrown away and all urine passed from that time until the bladder is emptied at 7 a. m. the next day is saved. If one desires to separate the day and night urine, the voidings from 7 a. m. to 7 p. m. may be kept in one container and those from 7 p. m. to 7 a. m. in a second vessel properly labeled. A thoroughly clean bottle of one-half to one gallon in capacity should be used as the container. This should be well corked after each addition of urine and kept in a cool place. As urine undergoes decomposition more or less readily, depending upon bacterial activity, some preservative should be added to prevent such processes. The writer is accustomed to advise the use of a slight excess of chloroform. This may be removed by heating the urine and will not then interfere with the later reactions. If not removed, its presence will lead to a pseudocarbohydrate reaction. Three or four drops of formalin may be added for each pint of urine. This is an efficient agent, but it will lead to reactions simulating those for sugar and even albumin, and, moreover, will introduce a crystalline compound of formalin and urea into the sediment as well as markedly interfering with the bile and indican tests. Thymol may be added,, but this may give a reaction similar to those for bile pigments and albumin. Cam- phor, chloral, and boracic acid have been used, but do not possess any virtues over the other preservatives mentioned. In any case, the worker should be on his guard in reporting abnormal findings without convincing himself that the reaction is not due to an added preservative. II. Physical Properties (i). Quantity. The amount of urine passed within 24 hours depends upon several factors and varies both for individuals and for different races of people. It is self- evident that under normal conditions, the amount of urine will vary with the quality and quantity of the substances to be excreted, the condition of the renal parenchyma, the pressure and rate of flow of the blood-current, the vasomotor disturbances, the stage of digestion, the loss of fluid in the perspiration as 166 DIAGNOSTIC METHODS. influenced by the surrounding temperature, amount of exercise, and extent of fluid intake; upon the weight of the subject, the sex, and age. As a rule, the quantity of urine excreted varies between 1200 and 1500 c.c. (40 to 50 ounces), reaching a maximum two or three hours after a large fluid intake. Women excrete somewhat less than men, while children void relatively more than do adults, although the actual amount is less. In the adult we find the amount of urine is almost directly proportional to his weight, a normal large individual excreting nearer 1,800 than 1,500 c.c., the amount being about 1 c.c. per kilo and hour, while with a child the excretion is about 4 c.c. per kilo and hour. The physiologic limits of the urinary excretion are about 750 and 3000 c.c. In cases showing as high an output as 3,000 c.c., one is justified, perhaps, in assuming the presence of some pathologic condition. The kidneys are not easily deranged by the excess work put upon them in excreting large quantities of urine, so that we may find a secretion of many liters per day continuing for an extended period without endangering the normality of the excreting organ. Normally, the amount of urine excreted during the day by far exceeds that voided during the night, while the afternoon urine is usually more than that of the morning. We find, however, that in edematous conditions either of hepatic, cardiac, or renal origin, the night urine usually exceeds that of the day. This condition is known as nycturia. As it is, perhaps, more frequently asso- ciated with cardiac insufficiency, it may have a diagnostic importance in these cases. Polyuria. By this is meant an excretion of an increased amount of urine. Just what amount of urine is to be considered as indicative of polyuria will depend much upon the habits of the patient as regards daily intake of fluid. As a rule, anything above 2,500 c.c. is at least suggestive of this condition. One should, however, not be content with the examination of a single day's specimen in judging of a polyuria, but should demand that the increase extend over a period of several consecutive days. Just what factors are to be held accountable for the polyuria, is not always easy to decide in every case. An increased intake of fluid together with an increased general blood-pressure will cause both an increased local renal pressure and an increased blood-flow through the kidney. While such a polyuria rarely exists in cases of ordinary chronic or active renal hyperemia, we find under the influence of drugs that a very decided increase in the urinary output may occur. The most important of such drugs, known as diuretics, are caffeine and digitalis. A polyuria is observed in the convalescent stages of acute nephritis, in both chronic parenchymatous and interstitial nephritis, and in amyloid degenera- tion of the kidney. This excretion, especially in the chronic interstitial type, THE URINE. 167 may be one in which the total solids are normal or reduced, and is then known as hydruria. Diabetes mellitus is more frequently, perhaps, than other conditions associated with a polyuria. The quantity eliminated is dependent both upon the increased intake as a result of the polydipsia as well as upon the dehydrating powers of the sugar. A certain relationship exists between the amount of fluid and the sugar, the polyuria being usually diminished by measures which decrease the amount of sugar excreted. This polyuria is not necessarily continuous and may alternate with periods showing a normal or subnormal amount of urine. In cases of diabetes insipidus we find the daily excretion of as much as 50 liters or more of urine. According to Meyer, this polyuria is due to the attempt on the part of the kidneys to secrete sufficient water to hold the solids in solution. He believes that a distinct insufficiency of the kidneys to secrete a urine of normal concentration exists, so that more water must be excreted to take care of the normal salts.1 In cases associated with abnormal accumulations of fluid, such as pleuritis, ascites, and general edema, a polyuria will exist at the time of absorption of the ex- udates, owing to the presence of such large amounts of fluid in the blood-vessels. The so-called "epicritic polyuria" is frequently observed during convales- cence from acute febrile attacks. This is probably indicative of the elimination of toxic products which have accumulated in the system during the progress of the disease. It is supposed to be of favorable import when occurring in a febrile condition, but it is to be recalled that this polyuria may be followed by a later oliguria which is of grave significance. As a rule, however, it may be said that as the case improves the urine is increased in amount. Polyuria may be observed in many nervous conditions, both functional and organic. The cause is probably some disturbance of the vasomotor appa- ratus as a result, perhaps, of irritation of the floor of the fourth ventricle, cerebellum, or cord. Hysteria, neurasthenia, epilepsy, and chorea are fre- quently associated with a polyuria. A paroxysmal polyuria in the course of a suspected nervous disease is more indicative of a functional derangement, while a continuous polyuria is more frequently associated with true organic disease. Oliguria. This is a condition characterized by the excretion of a diminished amount of urine, 800 c.c. being given as the lower normal point of the urinary output. Here, again, the absolute figure must depend upon the patient and upon his customary excretion. A single examination is not sufficient to decide whether or not an oliguria exists. This condition is found, perhaps, most frequently in cases of broken compensation of the heart, where the blood-pressure is markedly diminished. It is present whether the cardiac incompetency be primary or secondary to hepatic, renal, or pulmonary lesions. 1 See Lewis and Matthews (Trans. Chic. Path. Soc., vol. 9, 1913, p. 16) who show the close relation of hypersecretion of the pars intermedia of the hypophysis to diabetes insipidus. 168 DIAGNOSTIC METHODS. Oliguria is noted in practically all acute febrile disorders, especially in typhoid fever. This is due, probably, to a combination of cardiac weakness with the increased loss of water by the skin and lungs. Moreover, we may have, in such states, a retention of fluid along with a direct contraction of the renal vessels. Acute nephritis as well as chronic parenchymatous nephritis are associated with a more or less extensive oliguria. This condition is probably referable to diminished functional activity of the glandular elements as well as to increased resistance within the tubules. A bilateral diffuse lesion is always necessary to cause much oliguria, as the sound kidney, if the trouble be unilateral, will take on vicarious activity. The more acute the condition, the greater the degree of oliguria. Oliguria may also occur following the administration of an anesthetic, in connection with eclampsia, hysteria, or epilepsy, after the loss of large quantities of fluid by hemorrhage, diarrhea, or vomiting, in cases of portal obstruction as seen in acute yellow atrophy or hepatic cirrhosis, or in cases in which pressure is exerted upon the vascular system, especially the vena cava, by tumors. Anuria. This oliguria may, in almost any case, proceed to complete anuria, which may or may not be of vital significance. Cases of anuria do occur without any preceding oliguria, as shown in acute nephritis and in some cases of hysteria. Anuria, per se, cannot be held responsible, however, for the uremic symptoms so frequently associated with it, as it may persist for many days, 19 in a case of Adams, without any uremic signs. Anuria may be due to obstructive, reflex, renal, and prerenal causes. We may have an occlusion of the urinary passages1 on one side and a reflex closure on the other. Tumors, prostatic hypertrophy, and toxic and nervous bladder disturbances may lead to a great degree of oliguria amounting, almost, to anuria. The so-called prerenal causes of anuria include scarlet fever, which may lead to a severe nephritis, phosphorous poisoning, action of ether and chloroform, collapse, ureteral and urethral calculus, and cholera. (2). Appearance. Freshly voided urine should be clear and transparent. Only the faintest trace of any turbidity should be normally present, except soon after a meal rich in vegetable food, when a distinct turbidity may be noticed due to the precipitation of the phosphates in the alkaline urine. When allowed to stand for a short time, a light cloud is noted which gradually settles to the bottom of the container in the form of the so-called "nubecula." This contains a few small granular cells and a few epithelial cells and is composed largely of mucus. 1 See MacNider (Jour, of Med. Res., vol. 26, 1912, p. 79) for a discussion of the relation of swelling of the tubular epithelium to the urinary output. THE URINE. 169 On standing for a somewhat longer period, as for instance over night, at the ordinary temperature, distinct crystals of uric acid may separate and appear in the sediment. If the temperature of the urine is allowed to fall to a considerable extent during this period, a somewhat more marked turbidity will be produced owing to the precipitation of the acid urates. This sediment is particularly noticeable if a highly acid urine becomes very cold. If kept for a longer period at room temperature, or a shorter period during the warmer months, a diffuse cloudiness will appear, due to the precipitation of the phosphates, owing to the lessened acidity or abnormal alkalinity of the urine. This alkalinity is due to the decomposition of the urea into ammonium carbonate. The crystals in this alkaline urine will be triple phosphates, calcium phosphates, ammonium urate, and calcium carbonate. Even before the urine becomes alkaline, a diffuse cloudiness may be present, due to the development of numerous saprophytic bacteria. This bacterial cloud is removed only with the greatest difficulty, as filtration of the urine has practically no effect upon it. Frequently, the addition of lead acetate to the urine will produce a voluminous precipitate, which may carry down the bacteria and permit of their filtration. This procedure is, however, not to be recommended, as other substances, if present in small amounts, may be carried down and thus escape detection. If the urine is cloudy when freshly voided, the turbidity may be the result of the precipitation of phosphates through the alkalinity or it may indicate the presence of an organized sediment, such as casts, epithelial cells, blood, and pus. The normal urine shows but very little viscosity, differing little from ordinary water in this respect. In certain conditions we find a marked degree of viscidity, which becomes especially apparent on attempting to filter the urine. In cases of chronic cystitis the excretion of a large amount of mucus may make the urine ropy and gelatinous. This increased viscosity may also be seen in cases of pyuria associated with decomposition. (3). Color. The color of the urine varies normally between various shades of yellow, the depth of color depending upon the concentration or specific gravity of the specimen. While the color is usually much paler in the urines of low specific gravity and very dark in those of high density, we find in diabetes mellitus a very pale urine with a high specific gravity. In cases of anemia the urine is always paler than normal, but in pernicious anemia the urine is highly colored owing to the marked destruction of the erythrocytes. As a rule, it may be said that an acid urine is more highly colored than an alkaline one, although many exceptions to this rule occur. There seems to be some difference between the urines passed at different periods of the day; thus the urine of the day is usually a distinct amber, while that of the night may take on a greenish tinge. To what this color in the latter case is due is at present unsettled. Several color scales have been introduced, such as those 170 DIAGNOSTIC METHODS. of Neubauer and Vogel and of Radde, but these are not sufficiently extensive to take in pathologic variations where they would be most important. As a rule, it is sufficient to divide the colors of the urine into those of the spectrum, making allowance for light, medium, and dark shades of each color. The pigments causing the normal and abnormal colorations of the urine will be discussed in detail in a later section. At this point the writer would say that normally these pigments are urochrome, uroerythrin, and urobilin, while the various conjugated glycuronic and sulphuric acids, blood pigments, biliary pigments, melanin, etc., are found in pathologic conditions. Pathologic Colorations. Deviating from the rule that the higher the specific gravity the more intense the color, diabetes mellitus shows an extremely light color with a high specific gravity. Owing to the lack of pigment, in chlorosis we find a very pale urine. Chronic interstitial nephritis and amyloid degeneration of the kidneys are associated with an extremely pale urine. In febrile conditions the coloration may range from an orange-red to a distinctly red tint, owing to the increase in the amount of urobilin. This deep color is especially noticeable in cases of severe pneumonia.1 This reddish urine may also be due to an increase in the amount of uroerythrin, which is responsible for the deep color of the urate sediment so frequent in the concen- trated urines of febrile conditions, circulatory disturbance of the liver, and in cases associated with profuse perspiration. In cases of jaundice the urine may vary from a dark yellow or green to a brown or black, depending upon the concentration of the urine, the amount of bile pigments, and upon certain chemical activity which occurs in such urine. Not only will the color of the urine be deeper, but the foam which appears on shaking the specimen will take on a distinct yellowish-brown tint. If this biliary urine be allowed to stand in the cold for some time, crystals of bilirubin may separate out and be seen in the sediment. Urine which contains blood may have a violet shimmer, may appear smoky, blood red, brownish-black, or even deep black in color. These vari- ations depend both upon the amount and kind of pigment present. Hemo- globin gives a more reddish tint to the urine, while methemoglobin produces a brownish shade. Such urine is always cloudy, owing to the admixture of corpuscles and other organic material. The blood found in the urine may arise from any point in the genitourinary tract or may be of systemic origin. In the latter case conditions which give rise to hemolysis will cause the appear- ance of hemoglobin in the urine. The condition of chyluria is characterized by the presence of large numbers of highly refractile globules of fat along with many morphological constituents. This gives rise to the appearance of a milky urine and is especially characteristic of infection with the filaria. It is not infrequent to find, in certain cases of hysteria, a specimen of milky urine, owing to the fact that the patient has 1 See Hildebrandt, Zeitsch. f. klin. Med., Bd. 73, 1911, S. 189. THE URINE. 171 added milk to the urine before sending it to be examined. The presence of a large quantity of pus will also give the urine a milky appearance. The urine of patients suffering with melanotic tumors may be perfectly clear when freshly voided, but becomes black or dark brown on exposure to the air. This reaction is due to the transformation of the pigment melanogen to melanin, and may be hastened by the addition of oxidizing agents to the urine. This darkening of the urine extends characteristically from above downward. The condition known as alkaptonuria, which is characterized by the excretion of homogentisic and uroleucic acids, gives rise to the passage of a urine which is brownish-black in color and may be syrupy in consistency. This color is not always evident in the fresh specimen, but appears soon after being voided. In cases of peritonitis, suppuration anywhere in the system, gangrene, and marked intestinal putrefaction, the urine is frequently dark colored owing to the passage of certain aromatic products of decomposition, either indican or various derivaties of phenol. The coloration in these cases may vary from a dark brown or greenish-black to a distinct blue. These urines differ from those containing melanin"^n the fact that ferric chlorid does not blacken the urine as it does in the presence of melanin. This urine may contain a distinct amount of indigo, although the substance present is usually a different oxidation product. If indigo be present, a bluish-black scum will frequently rise to the surface of the specimen. Medicinal Coloration. After the use of carbolic acid either internally or externally, guaiacol, creosote, resorcin, naphthalin, salol, and various tar preparations the urine may vary from a dark brown to a black color. This coloration is due to the excretion of hydroquinon and of pyrocatechin, and may be' evident only on allowing the urine to stand for some time. The urine containing pyrocatechin may reduce alkaline copper solutions, but will not affect such bismuth prep- arations. While a dark brown or black coloration of the urine may be found in cases of hemorrhage, melanosis, malaria, alkaptonuria, ochronosis, and chronic tuberculosis, one should be on his guard, as the medication of the case may be responsible for such coloration. Methylene blue will color the urine a greenish to deep blue shade, which may last for several days. Usually within an hour after this drug is taken the urine may show a faint tinge of green which may be more clearly brought out by acidifying with acetic acid and warming. The use of the hypnotics, trional, sulphonal, and tetronal, frequently gives rise to the voiding of a urine which has a deep red-wine color, due to the presence of hematoporphyrin. Pyramidon produces a urine of rose-red color, the pigment of which is soluble in ether, chloroform, and amyl alcohol. Anti- pyrin and purgatin both produce distinctly red urines. Chrysarobin, senna, rhubarb, cascara, and santonin produce a golden-yellow urine which 172 DIAGNOSTIC METHODS. becomes red in the presence of alkali. This coloration is due to the excretion of chrysophanic acid. According to Gorup-Besanez, the pigment of beets, huckleberries, blackberries, etc., may under certain conditions be excreted in the urine and color it the corresponding shades. (4). Odor. The normal urine usually has a distinct aromatic odor which very much resembles that of beef broth. This odor is due to the presence of certain volatile acids and is more marked in urines of high concentration. If the urine undergoes decomposition either within the bladder or on standing, a so-called "urinous odor" appears which is due to the decomposition of protein material. This odor is very markedly ammoniacal. Should such an odor appear in the freshly voided specimen, it is evidence of marked cystitis. Abnormal decomposition of the urine, as evidenced by change? in the odor, may be found in conditions associated with decomposition of pus and may be due to the presence of hydrogen sulphid along with the ammonia. This condition may be observed in cases of perforation of an abscess into the urinary tract, in which case the urine may have a distinctly fecal odor if the intestine be involved, while in carcinoma of the bladder this repulsive odor of the urine may also be noticed. A distinct fruity odor is often present in cases of diabetes mellitus, in many febrile conditions, and in some stomach and intestinal troubles, which may be directly traceable to the presence of acetone, although Folin believes it is due to some unknown substance other than acetone. Certain medicaments, such as oil of turpentine, give rise to a distinct odor of violets in the urine. Menthol causes an odor of peppermint, while cubebs, copaiba, sandal-wood oil, tolu, and saffron produce a peculiar spicy odor. Valerian and asafetida are excreted as such in the urine and produce their characteristic odor. Certain foods, such as meat, bouillon, and coffee, produce a slight odor of the urine, while asparagus gives a peculiar characteristic odor due to the presence of methyl mercaptan. (5). Reaction. The normal urine has an acid reaction. According to the views recently held, this acidity was directly due to the presence of acid salts, especially to sodium dihydrogen phosphate (NaH2PO4), and not to the presence of any free acid. As certain procedures had shown that not all of the phosphates were in the form of the diacid type, but that some of them were present as the monacid phosphates, the variations in the acidity were supposed to represent differences, both absolute and relative, in the amounts of the diacid phosphates present. The recent work of Folin has shown that the phosphates in the clear urine are all of the monobasic (diacid) type. His figures indicate that the acidity of normal clear urines is ordinarily greater than the acidity of all the phosphates THE URINE. 173 present and that the excess must be due to free organic acids. For this reason the methods of Freund and of Lieblein for the determination of the acidity of the urine must be given up. To quote from Folin:1 "The current attractive, and in a measure plausible, belief that the acidity of urine is regulated by variations in the relative proportion of the two forms of 'acid phosphates' is, therefore, erroneous. If urine does at no time contain comparatively strong acids in the free form, the reason is in part the variability of the ammonia formation and in part the presence of salts of organic acids. In a mixture of salts containing an excess of acids it is the weakest which will remain uncombined and the strongest organic acids will, therefore, exist as salts; but if the total amount of acidity becomes abnormally great, the quality (the strength) of the free acids may change." From the standpoint of physical chemistry the acidity of the urine, as of all other acid solutions, should represent the absolute number of dissociated hydrogen ions in a definite quantity of the urine. We are, therefore, face to face with the same problem confronting us in the examination of the alkalinity of the blood. In the case of the urine the question of indicator to be used in the titration test is a matter of great moment, as no two indicators will give the same degree of acidity. The one naturally to be selected would be that which will react to every possible substance of an acid nature. If we use the methods of physical chemistry we find, according to Hbber,2 that the urine is only about 30 times as acid as is distilled water and only about one ten-thousandth as acid as the titration figures would indicate. Such being the case, we must either entirely revise our figures for the acidity of the urine or employ methods which can be more easily carried out by the general worker than can those of physical chemistry.3 Folin4 has, therefore, introduced a method which uses direct titration of the urine and employs phenolphthalein as an indicator. This indicator reacts to all bodies of an acid nature, but cannot overcome certain difficulties which are in the way of direct titration. These obstacles are (1) the occur- rence of calcium in the urine in the presence of the monobasic phosphates, and (2) the presence of ammonium salts. He has found that the addition of potas- sium oxalate to the urine will do away with these difficulties by holding in solution both the di- and tri-calcium phosphates and by preventing the dissocia- tion of the ammonium compounds. Folin's Method. Total Acidity. Twenty-five c.c. of urine are treated with 15 to 20 grams of powdered potassium oxalate and one or two drops of a 1 per cent, alcoholic solution of phenol-phthalein. The mixture is shaken rapidly for one or two minutes and titrated at once with a tenth-normal sodium hydrate solution until a 'Amer. Jour, of Physiol., vol. 13, 1905, p. 45. 2Beitr. zur chem. Physiol, u Path., Bd. 3,1903, S. 525. 3 See Henderson and Palmer, Jour, of Biol. Chem., vol. 13, 1913, p. 393; Ibid, vol. 14, 1913, p. 81. 4 Loc. cit. 174 DIAGNOSTIC METHODS. faint, distinct, permanent pink color is obtained. It is advisable to shake the flask during the titration so as to prolong the effects of the potassium oxalate. The acidity is expressed in terms of the amount of tenth-normal sodium hydrate solution necessary for neutralization of the 24-hour amount of urine. This is expressed as T, which is, on an average, 617. Determine the amount of total phosphates present by titration with uranium nitrate solution as described later. Seven and one one-hundredth mg. of P2O5 have an acidimetric value equal to i c.c. of tenth-normal acid. The total acidimetric value of the phosphates of the 24-hour urine may be easily determined with the help of this factor, by converting the amount of phosphates into terms of N/10 acid. From the total acidity (T) subtract the acidimetric value of the phosphates (P). The remainder is the acidity due to uncombined organic acids, and the difference, that obtained from calculating all the phosphoric acid as di- acid phosphate, is the free mineral acidity. For all ordinary studies of the acidity of the urine the direct titration of the total acidity and of the phosphates gives the necessary information. The excess of the total acidity above that calculated from the phosphates gives the total free acids present. If the acidity calculated from the total phosphates is greater than the titrated acidity, then there are practically no free organic acids present, and the titrated acidity represents the amount of phosphates present in the diacid form (Folin). While the acidity of the urine is best determined and expressed as outlined above, it seems wise to the writer to retain the same style of expression for the acidity as used in stomach analysis. With this nomenclature one would state the acidity of the urine in degrees; that is, the amount of tenth-normal sodium hydrate necessary to neutralize 100 c.c. Under normal conditions this will vary from 35 to 450. It may be increased by a diet rich in meat, while it is decreased by a vegetable diet. There are many acids produced in the oxida- tion of protein, among which we find sulphuric, phosphoric, uric, and the oxyaromatic acids. Ordinarily, these play an indirect part in the acidity of the urine, although this phosphoric acid may exist in part as the dihydrogen phosphate and in consequence increase the acidity of the urine. The regulation of the metabolism is such that an increase of the acids produced in the system or taken into it from without is neutralized by an increased formation of am- monia, the salts appearing in the urine as the ammonium salts which do not, of course, increase the acidity of this fluid. This is the basis upon which one estimates the amount of ammonia in following a condition of acidosis.1 The reaction of the urine varies at different times of the day. The acidity appears to be highest in the morning before breakfast and is diminished after a meal, due to the secretion of hydrochloric acid into the stomach. The reaction Free Mineral and Organic Acidity. 1 Fischer (Nephritis: An experimental and critical study of its nature, cause and the principles of its relief, New York, 1912) believes that the abnormal production or accumulation of acid in the cells of the kidney and the action of this acid on the colloidal structures of the kidney are accountable for all the changes which characterize nephritis. THE URINE. 175 of the urine may even be alkaline for a period of two or four hours after each meal, in which case the urine will be turbid from the precipitation of phosphates. This reaction of the urine following meals is known as the "alkaline tide" of the urine. Between meals the acidity of the urine will gradually increase until the next meal is taken. The reaction of the urine is modified to a great extent by the use of drugs.1 Thus, alkalies, such as carbonate and bicarbonate of sodium, will render the urine alkaline if taken between meals, while if taken just preceding a meal they will be neutralized by the gastric juice. All organic acids of the fatty series are oxidized in the system to carbonic acid and combine with bases forming basic salts which render the urine alkaline or less acid, providing these acids are not taken above the point of tolerance, as the writer has shown that large doses of such acids as citric acid will increase the acidity of the urine. The mineral and aromatic organic acids will, however, practically always increase the acidity. In many pathologic conditions we find the reaction of the urine variable. Abnormal gastric activity may be either associated with an increase or a decrease in the acidity of the urine, depending upon a condition of hypo- or hyperacidity of the gastric juice. The rapid absorption of a transudate or exudate will lead to the excretion of an alkaline urine from the presence of an increased amount of alkaline salts. An alkaline urine is not infrequently seen after intestinal hemorrhage, in certain cases of pneumonia, typhoid fever, chronic nephritis, and in cases in which exudates from the urinary tracts have become mixed with the urine. In certain cases of nervous diseases and in some cases of anemia we may also find an alkaline urine. The urine in all of the above cases will show, if tested by litmus-paper, an alkaline reaction in which the bluing of the red litmus-paper is permanent. This condition is known as fixed alkalinity and is quite distinct from the following type. In cases of decomposition of the urine within the urinary tract, through the influence of bacteria, the urea is decomposed into ammonium carbamate and carbonate. The alkaline reaction of the urine in such cases will be shown by a blue color of the red litmus-paper either held above it or placed in it, the blue color disappearing when the paper is dried. This condition is known as volatile alkalinity. If the urine shows this volatile alkalinity on being voided, the finding is significant of trouble somewhere along the urinary tract, especially within the bladder. Benedict has recently introduced the term "acid unit" into the study of the urinary output. One c.c. of urine with an acidity of i degree is equivalent to 100 acid units. In other words, one may determine the acid units in the 24-hour specimen by multiplying the number of cubic centimeters of the urine by the degrees of acidity. This is normally about 40,000. (6). Specific Gravity. 1 See de Jager (Biochem. Zeitsch., Bd. 38, 1912, S. 294), who shows that magnesium sulphate increases the acidity, while sodium sulphate diminishes it. The specific gravity of the normal urine ranges between 1,015 and 1,025, 176 DIAGNOSTIC METHODS. with an average of 1020. This specific gravity will depend, of course, upon the amount of fluid intake, the quantity of the 24-hour specimen of urine, the degree of tissue activity, and the condition of the secreting organs. The intake of a large volume of water may reduce the specific gravity of the urine to a very low figure and, correspondingly, a small intake may lead to a urine of high gravity. We find, therefore, that perfectly normal urines may show specific gravities ranging from 1010 to 1030, with pathologic variations from 1,002 as high as 1,060 or more.1 In general routine work it is essential that the specific gravity of the 24-hour specimen be determined. Except in unusual cases, absolutely noth- ing of diagnostic value may be learned by the determination of the specific gravity of a single voiding of urine. The variations at different times of the day, under the influence of food, digestion, activity of the skin and lungs, and exercise, may be so great that apparent pathologic figures may be obtained from a single specimen. Such variations are overcome for the most part in the 24-hour specimen unless pathologic conditions are present to keep up such variations. In some cases, especially in chronic diffuse nephritis, the morning specimen of urine almost invariably has a lower specific gravity than that of the other periods of the day. For this reason one may determine the specific gravity of a single specimen of such urines. It is of especial importance that the total amount of urine in the 24-hour specimen be taken into consideration in judging of the value of a specific gravity. Thus in chronic interstitial nephritis we may find a large volume of urine with a low specific gravity, while in diabetes mellitus an even higher volume of urine may be present, showing a very high specific gravity. Technic. The most accurate method of determining the specific gravity is, of course, the use of the pycnometer. The principle of this method is the determina- tion of the weight of a definite volume of urine as compared with that of the same volume of distilled water under the same conditions of temperature and atmospheric pressure. This method will be discussed in the section on Blood, to which the reader is referred. The clinical method of estimating this factor is a distinctly areometric one. The principle of this method is that a body immersed in a fluid will displace an amount of fluid equivalent to the loss in its own weight. By the use of instruments known as hydrometers or in the case of the urine as urinometers, this displacement is measured by immersing the hydrometer in the fluid and observing the point to which this instrument sinks. The stem of the hydrometer is graduated in divisions from 1,000 to 1,060 by differ- ences of i°, the 1,000 point being that to which the instrument sinks when im- mersed in distilled water at the temperature to which the instrument is cali- brated. Any variation in the density of the solution in which this hydrometer is immersed will be evident by the depth to which it sinks, the more concen- trated the solution the less will the instrument sink. 1 See Frey, Deut. Arch. f. klin. Med., Bd. 106, 1912, S. 347. THE URINE. 177 The vessel in which the urine is poured should be cylindrical in shape, with parallel sides and wide base and sufficiently tall to permit of the complete sinking of the hydrometer. The forms of this cylinder with fluted sides are perhaps more desirable than the plain cylinder, as the bulb does not tend to stick to the sides of the vessel so readily. The vessel is filled about four-fifths full of urine, any foam being removed by the use of filter-paper. The hydro- meter is placed in the urine with a twisting motion and allowed to come to rest. The depth to which the stem is immersed is then read off by observing the mark which coincides with the lower meniscus of the urine as seen from below. The worker should never attempt to read the specific gravity from above as a slight meniscus interferes with the accuracy of his reading. The worker must be absolutely sure that the urinometer neither rests upon the bottom of the cylinder nor touches the sides, but should see that it floats perfectly free in the urine. The temperature at which the reading is taken is a matter of some moment as some of these instruments are graduated at 150 C. The ordinary model, as made by Squibb, is graduated at 250 C., which is, perhaps, more nearly the working temperature of the room. A variation of 30 in temperature between that of the room and that at which the instrument is calibrated, will give a difference of i° of specific gravity; that is, a difference in the fourth place of the specific gravity. In ordinary clinical work corrections for variations in temperature are usually unnecessary, as variations of two or three points in the fourth place of the specific gravity are of absolutely no importance, as such changes might be attributed to chemical variations on standing, even though the most accurate methods of esti- mating the specific gravity were used. If the quantity of urine be very small, it may be diluted with distilled water, so that the measuring cylinder may contain enough material to permit of a density estimation. The specific gravity of this diluted urine is then determined as above and the last two figures of the specific gravity are multi- plied by the degree of dilution. Fig. 69 - Urin- ometer and cylin- der. {Hawk.) Rough Estimate of Total Solids As the degree of specific gravity is directly proportional to the amount of solids contained in the urine, one may roughly judge of the total solids by a simple calculation as follows: If the last two figures of the specific gravity be multiplied by 2.33 (Haeser's coefficient), the result will be the approximate number of grams of total solids in every 1,000 c.c. of urine. Know- 178 DIAGNOSTIC METHODS. ing the quantity of urine passed in the 24 hours, a simple calculation will yield the total 24-hour excretion of solids. Long uses the coefficient 2.6. Instead of the above figure one may multiply the last two figures of the specific gravity by 1.1 (Haines' coefficient) and obtain the number of grains of solids in each fluidounce of the urine. This figure when multiplied by the total number of ounces of the 24-hour specimen ■will yield the excretion of solids in grains. This latter method has some advantage for the older practitioner who has not accustomed himself to the use of the metric system. It is to be said that neither one of the above methods can give anything but approximate results and in pathologic urines are absolutely unreliable. Highly albuminous urines invariably show a reduced specific gravity, while a high sugar content is associated with an increase in the density. In either one of these cases a calculation of the total solids by the above method will yield inexact figures. If the worker desires to know the exact amount of total solids in the 24-hour specimen, and this is sometimes advisable, recourse should be had to more exact methods of determination which will be discussed later. The specific gravity of a specimen of urine varies, of course, with the amount of total solids. Normally, these range from 60 to 70 grams with a 24-hour excretion of 1,500 c.c. of urine. The urea usually constitutes about one-half of the total solids. As the normal percentage of urea is approxi- mately two with a specific gravity of 1,020, the writer has been struck with the usual close relationship of the percentage of the urea to the specific gravity. In watching this point in over 2,000 examinations of urine within the last year, the writer has observed that the percentage of urea will practically parallel the last two figures of the specific gravity; in other words a specific gravity of 1,015, for instance, will normally be associated with a urea content of 1.5 per cent. This statement is true only in those cases which contain neither albumin nor sugar. As a rule, it may be said that the specific gravity of the urine is inversely proportional to the amount of fluid eliminated. It will, therefore, be evident that conditions leading to an oliguria will produce a high specific gravity, while those causing polyuria will give a low specific gravity. This state- ment must be modified when considering certain pathologic conditions, as we may find a diminished amount of urine with low specific gravity as in chronic nephritis in which the salts are diminished, although the organic albu- minous bodies are much increased; while in diabetes mellitus we have an abun- dant urine of high specific gravity. The specific gravity of the urine is of more or less importance in judging of the activity of the kidneys. In acute nephritis we find a urine of high specific gravity, while in the chronic types of renal disease the specific gravity is low owing to the diminution of the salts. It is to be noted in this connection that in the case of so-called "functional" albuminuria, the specific gravity of the urine is above the normal figures. A marked reduction in the specific THE URINE. 179 gravity of any case of nephritis is of dangerous import. The writer will have much to say regarding the functional activity of the kidneys in a later section. (7). Optical Activity. According to Haas,1 the normal urine is slightly levorotatory, ranging from 0.01 to 0.18. This optical activity is due to traces of the conjugated glycuronic acids which will be discussed later. An increase in this levorotatory power is observed due to the presence of increased amounts of glycuronic acid, /3-oxybutyric acid, albumin (in amounts over one-half part per thousand), and levulose. Dextrorotatory urines depend upon the presence of glucose, maltose, and lactose, while the presence of pentose usually gives rise to an optically inactive urine or one at least showing only a slight degree of dextrorotation. This optical activity under the influence of pathologic products will be discussed in the section on Carbohydrates, to which the reader is referred. III. Chemical Properties. (A). Normal Composition. The tables given in most text-books showing the chemical composition of the urine cannot be regarded as absolutely indicative of the excretion as shown in the every-day specimens of urine. The composition of the urine is absolutely dependent upon the diet under normal conditions, so that a table to be exact must embrace the findings under a specified diet.2 Perhaps the most frequently quoted table is that of Parkes, which may be found in almost any text-book dealing with the urine. This table gives the various figures for the different substances excreted, but does not take into considera- tion the amount of the various types of foods used in the diet leading to this excretion. The writer, therefore, feels that it is wise to omit such a table at present, as we have no method of comparison of the excretion with the intake. The recent work of Folin upon the urine of persons both under a mixed diet and a nitrogen-free diet gives an "approximately complete" determination of the urinary constituents under absolutely fixed dietary conditions. The figures for such diets will be given under the head of each individual substance discussed. The daily urine of a healthy adult will vary between 1,200 and 1,500 c.c. in amount and will contain from 60 to 70 grams of total solids, of which the inorganic constituents form from 25 to 30 grams and the organic between 35 and 40 grams. The inorganic constituents consist of the phosphates of sodium, potassium, calcium, and magnesium, the chlorids and sulphates of the alkali metals, various types of ammonium salts, traces of nitrates, cal- cium carbonate (especially under vegetable diet), and traces of iron compounds. While these inorganic substances have not hitherto been credited with much 1Centralbl. f. d. Med. Wissensch., 1876. 2 See Long, Jour. A. M. A., vol. 58, 1912, p. 757; also, Long and Gephart, Jour. Amer. Chem. Soc., vol. 34, 1912, p. 1229. 180 DIAGNOSTIC METHODS. importance, to-day we are realizing more and more that much is to be learned by a careful study of the inorganic excretion.1 The organic substances are of especial importance both in metabolic work and in the diagnosis of pathologic conditions. While we do not by any means know everything concerning the variations in excretion of these organic products, yet we do know much which is helpful in our direct diagnostic work as well as in our study of the progress of the disease. Among the organic substances which are more or less normal (although not always in large amounts) in the urine we find the lower and higher fatty acids, oxalic acid, acetone, glycerophosphoric acid, a trace of glucose, lactose (especially in nursing mothers), carbamic acid, urea, oxaluric acid, allantoin (especially a few days after birth), creatinin, uric acid, purin bases, thiosulphuric acid, tauro- carbamic acid, cystin, chondroitin-sulphuric acid, inosite, hippuric acid, benzoic acid, phenaceturic acid, p-oxyphenyl-acetic acid, hydro-p-cumaric acid, skatol-carbonic acid, conjugated sulphuric and glycuronic acids, oxy- proteic acid, pigments, organic iron compounds, traces of protein, and ferments. Under pathologic conditions we may find lactic acid, large amounts of acetone, aceto-acetic acid, /3-oxybutyric acid, fats, large amounts of glucose, levorotatory carbohydrates, r-arabinose, lecithin, cystin, putrescin, cadav- erin, ptomaines, oxymandelic acid, leucin, tyrosin, homogentisic acid, uro- leucic acid, cholesterin, cholic acid, glycocholic acid, tauro-cholic acid, various derivates of phenol, hematin, hematoporphyrin, methemoglobin, other blood pigments, bile pigments, melanin, and protein material. (a). Total Solids and Total Ash. The estimation cf the total solids of the urine is a matter of considerable difficulty, owing to the fact that evaporation of the urine leaves a syrupy residue which is dried only with much trouble and constant loss of ammonia, formed through the action of diacid sodium phosphate upon urea in concen- trated solution. One may come very close to accurate results by placing a weighed amount of dry clean sand in a weighed platinum dish and adding io to 15 c.c. of urine. This is then evaporated upon the water-bath and later in the drying oven at 1050 C. The dish is then placed in the desiccator and allowed to remain until the weight becomes constant. Knowing the weight of the dish and sand, the amount of the urine added, and the weight of the dish after the urine is evaporated the total solids may be easily calculated. In case a large amcunt of residue is desired for quantitative work, recourse may be had to the method of Slagle,2 of adding 5 c.c. of concentrated H2SO4 to each liter of urine showing a specific gravity up to 1,020 and evaporating this to dryness. More H,SO4 must be used with the more ccncentrated urines. The estimation of the ash is as follows: Fifty c.c. of urine are evaporated to dryness over the water-bath in a weighed platinum or porcelain dish. The dish is then heated, while covered, over the free flame until gases cease 1 See Baumann and Howard (Arch, of Int. Med., vol. 9, 1912, p. 665) for a discussion of the inorganic metabolism in scurvy. 2 Jour, of Biol. Chern., vol. 8, 1910, p. 77. THE URINE. 181 to be evolved, especial care being taken not to permit sputtering of the con- tents. In some cases it is possible completely to incinerate the urine by long- continued heat. However, a more usual procedure is to treat the carbonized residue with distilled water, thoroughly stir the mixture and filter through a filter-paper whose ash is known. The contents of the dish should be washed onto the filter several times and the material upon the filter also washed with boiling water. The filter-paper and its contents are now placed in the dish and completely incinerated. After this procedure the filtrate and washings of the original carbonized material, which contain most of the inorganic con- stituents, are placed in the dish and evaporated at xoo° C. to dryness, and then incinerated over the free flame. The dish is now placed in a desiccator and dried to constant weight. Knowing the weight of the dish with its con- tents and the original weight of the dish a simple calculation will give the amount of ash in the 50 c.c. of urine taken. As originally stated the inorganic constituents will range between 25 and 30 grams under normal conditions. (&). Inorganic Constituents. (1). Chlorids. The chlorids are one of the most important groups of inorganic solids in the urine. They are derived entirely from the food and, in consequence, the amount of excretion will depend upon the intake. The chlorids actually forming constituent parts of the food exist in combination with potassium and calcium, while those which are added as seasoning to the food are practically always in the form of sodium chlorid. As the amount contained in the food is trivial in comparison to that added, we are accustomed to regard practically all of the intake as sodium chlorid. This is a constituent of the serum of the blood and of other tissues, while the potassium salt is in more direct relation with the cellular elements. Under normal conditions from io to 15 grams of sodium chlorid are eliminated in 24 hours. The administration of a diet rich in salts will in- crease this amount, while a salt-poor diet will diminish the amount up to a certain point. If the diet be a starvation one, or an absolutely salt-free one, the chlorids will disappear almost entirely from the urine. The regulation of the metabolic activity of the system is such that a certain amount of salt must be retained in order to preserve the osmotic equilibrium. For this reason we find that withdrawal of salt from the diet does not lead to any appreciable diminution of the normal chlorid content either of the blood or tissues. An increase in the elimination of salt is practically always followed by a retention of salt unless a sufficient supply is furnished by the food. If food containing sodium chlorid be given after a period of salt-free diet, a por- tion of this salt will be retained. Conversely, we find, if the body has for patho- logic reasons retained sodium chlorid, that an increased elimination will follow. This metabolic activity is intimately associated with the general protein metabolism of the body. Any increase in the amount of circulating protein as compared with the living protoplasm will be followed by an increased 182 DIAGNOSTIC METHODS. elimination of the chlorids, which have been previously retained by the living or active protein material of the protoplasm. This fact is shown by the rela- tion between the elimination of the chlorids and the total nitrogen. With an ordinary diet this ratio is as one to one, but in disease it may be much disturbed owing to chlorid retention through renal insufficiency. The blood and tissues of patients with nephritis show a higher chlorid content than those of normal individuals. We should, therefore, expect, if the law of increased chlorid excretion being dependent upon increased circulating albumin were to hold, that the chlorids would be increased along with the albumin in nephritis. But we find that the kidney under these circumstances is unable to excrete the increased amount of salts circulating in the blood. A further method of withdrawing sodium chlorid from the body con- sists in the administration of large quantities of alkaline carbonates or of compounds of the alkalies with vegetable acids. As soon as these are given the body becomes poor, not only in acid substances and in HC1, but also at the same time in sodium and potassium. Practically speaking, the body becomes directly impoverished in NaCl. The body may also lose chlorin when vomiting is frequent, when absorption is diminished, when the stomach is regularly washed out, and when diarrhea is marked. This loss may be especially noted in cases of hyperacidity of the gastric juice associated with vomiting. The chlorin in these conditions is withdrawn in the form of free acid, and in consequence the alkalinity of the tissues may be increased, although not to any great degree nor for any great length of time. Sodium chlorid is, therefore, seen to be of more than passing importance in the general meta- bolic activity of the system. Physiologic Variations. The amount of sodium chlorid excreted will depend directly upon the amount ingested. We may find as high as 30 grams of salt in the 24-hour specimen or it may be as low as 2 grams, salt-free diet reducing the elimination to a mere trace. The elimination may be increased by active exercise, by increasing the water intake and hence the water output, and by the intake of a large amount of vegetable food. Much more chlorid is apparently excreted during the day than during the night. Pathologic Variations. A marked diminution of the chlorids, which may in some cases be al- most complete, has been supposed to be pathognomonic of pneumonia. This, however, has been shown to be fallacious, as the same condition occurs in most acute febrile states, with a possible exception of intermittent fever. In a doubtful fever a large diminution in the amount of urinary chlorids might be strongly presumptive of pneumonia, but would be conclusive only in the presence of distinct clinical signs of this disease. While the retention of these chlorids in the exudate of pneumonia may partially explain the diminution in the urine, it cannot explain the fact that the chlorids of the THE URINE. 183 food are also retained. The explanation is more likely to be found, in the writer's opinion, in an existing renal insufficiency. This same retention of chlorids will be found also in any condition in which there is a transudate or exudate of any considerable bulk, so that this factor must play a certain role. As crisis approaches in pneumonia, the chlorids of the urine will in- crease in favorable cases, while in those of bad prognosis no such increase will, as a rule, be observed. Van der Bergh believes the explanation of the dimin- ished urinary chlorids in pneumonia to be an attempt on the part of the blood to maintain its osmotic pressure, the chlorids remaining fixed in the tissues owing to the increase of the products of metabolism in the plasma. The chlorids are diminished in all acute and chronic renal diseases associated with albuminuria. The work of Widal upon the influence of chlorids upon the progress of a renal disease has brought out the facts that not only do we have such a chlorid retention, but that the presence of chlorids in the food will increase both the albuminuria and the edema of these con- ditions. While these facts are incontrovertible, we must take into considera- tion, as Richter has shown, the amount of water intake as well. This is such an important field to the clinician that the writer would refer to other works giving the details of the " dechloridization" treatment and its effects in reducing the symptoms of a nephritis. A severe diarrhea will also diminish the amount of chlorids in the urine, as the chlorids of the food are carried off by the bowel too quickly to permit of absorption. In cases of carcinoma of the stomach, in dilatation either from hypersecretion or stricture of the pylorus, and in some cases of ulcer of the stomach a diminution or even total absence of chlorids in the urine may be observed. In most chronic diseases, in anemic conditions, in rickets, and in marked nervous diseases, such as melancholia or mania, the amount of chlorids may be greatly reduced. If the output of chlorids be very low in a chronic disease, the prognosis becomes grave unless the diet can explain the diminution. A diminution is observed in most febrile diseases, especially in the exan- themata, while in typhoid fever the reduction is not so marked. This slight diminution in typhoid fever may serve as a distinguishing point in the diagnosis of meningitis from typhoid fever, in the former case the diminution being much more marked than in the latter. In acute yellow atrophy of the liver the chlorids are diminished, while in cirrhosis of the liver they are somewhat increased.1 The Chlorids are increased in all conditions which have previously shown a retention, according to the law which has been previously discussed. We find thus an increase in the period of convalescence from acute febrile diseases, especially pneumonia. Strangely enough, the chlorids are found markedly increased in diabetes insipidus,2 which is associated also with the excretion 1 Burnham (Jour. A. M. A., vol. 58, 1912, p. 851) reports a case of alcoholic cirrhosis in which there was a constant decrease in the excretion of the urinary chlorids. 2 See Forschbach and Weber, Zeitsch. f. klin. Med., Bd. 73, 1911, S. 221; also, Meyer, Ibid., Bd. 74, 1912, S. 352. 184 DIAGNOSTIC METHODS. of a large amount of urine. In epilepsy an increase may be observed follow- ing the attack. The chlorids are increased in the urine after the use of chloroform, whether administered internally or as an anesthetic. Some of the diuretics, •especially potassium acetate, produce an increase in the urinary chlorids. In metabolic work it is frequently of advantage to study the effects of an ash-free diet upon the pathologic condition. Taylor1 has recently intro- duced such a diet, consisting of the whites of 18 eggs, 120 grams of olive oil and 200 grams of crystallized sugar. Little work has been done under the influence of such a diet so that no definite conclusions can be drawn at present. The work of Goodall and Joslin2 with this diet confirms the earlier views that it is practically impossible to diminish the chlorin of the body by more than 10 to 14 per cent, and that the loss of water is proportionate to this. Estimation of the Chlorids. For rough clinical purposes the amount of chlorids in the urine may be estimated as follows: A few c.c. of clear, filtered urine, from which albumin if present is removed by heating with acetic acid, are placed in a test-tube and acidified with 10 drops of chemically pure nitric acid. This mixture is then treated with a few drops of 10 per cent, silver nitrate solution. If the chlorids are present in normal amount a distinct, curdy white precipitate will settle out. If the chlorids be increased a heavy precipitate will be observed, while if they be diminished only a cloud without any flakes will be seen. Quantitative Determination. The best method for such determination is, in the writer's opinion, the Arnold modification of Volhard's method. The principle of the test is the precipitation of the chlorids in a definite amount of urine by a standard solution of silver nitrate in the presence of an excess of free nitric acid. If the precipitate of silver chlorid be filtered from the solution, the excess of silver nitrate may be determined in the filtrate by titration with a standard solution of potassium sulphocyanate, using a strong solution of iron-ammo- nium-alum as an indicator. The urine should be as fresh as possible and should contain no nitrites. Albumin, unless present in very large amount, need not be removed. It is wise, however, in case the urine shows a high albumin content, to acidify the urine with acetic acid, boil, and filter off the precipitated albumin. In doing this one should take a definite volume of urine, precipitate as above, and wash the precipitate thoroughly with water in order to dissolve any chlorids which may have been retained by the al- bumin. The filtrate is made up to a definite volume, which represents the amount of urine originally taken. Thus, 20 c.c. of urine are treated as above and washed with sufficient water to make 50 c.c. In the test as outlined later, 'Univ, of Cal. Pub., Pathology, vol. i, 1904, p. 71. 2 Arch, of Int. Med., vol. 1, 1908, p. 615. THE URINE. 185 in which 10 c.c. of urine are used, 25 c.c. of this filtered albumin-free urine will represent 10 c.c. of original urine. Solutions Necessary. (i). A solution of silver nitrate of such a strength that i c.c. is equiva- lent to o.oi gram of NaCl or 0.00606 gram of Cl. In making this solution 29.055 grams of pure anhydrous crystallized silver nitrate are dissolved in 1 liter of distilled water. The chemically pure AgNO3 as found on the market is perfectly reliable and needs only an accurate chemical balance for weighing the exact amount. It is essential that this solution should contain exactly the amount specified, as the accuracy of the method depends upon the correct strength of the volumetric solutions. (2). A solution of potassium sulphocyanate of such a strength that 20 c.c. will correspond exactly to 10 c.c. of the silver solution or, in other words, so that 2 c.c. of the cyanate solution are necessary to precipitate the silver from each c.c. of the silver nitrate solution. Other workers use solutions of potassium sulphocyanate of somewhat different strength, but the simplicity of the calculations necessary to determine the chlorids of the urine is much increased by such a relation between the two volumetric solutions. As po- tassium sulphocyanate is very hygroscopic, it is impossible accurately to weigh the exact amount necessary to make this solution. We, therefore, dissolve a slight excess (9 grams) of potassium sulphocyanate in approximately 1 liter of water. In order to make this solution correspond exactly to the silver solution it is necessary to find out how much water must be added to make 20 c.c. of this neutralize 10 c.c. of the silver solution. The technic is as follows: ten c.c. of the known solution of silver nitrate are measured from a buret and diluted with 50 or 60 c.c. of distilled water. Five c.c. of chemically pure nitric acid (specific gravity 1.2) and 5 c.c. of a strong solu- tion of iron-ammonium-alum are added and thoroughly mixed. This mixture is then titrated with the potassium sulphocyanate solution whose strength is to be determined. The principle of this titration is that the KCNS first combines with the AgNO3, forming a white precipitate of silver sulphocyanate. At the exact point at which this combination is complete the potassium sulpho- cyanate will combine with the iron of the indicator forming sulphocyanate of iron which is distinctly red in color. The titration is, therefore, carried to the point at which a permanent faintly reddish-brown color appears on shaking the mixture. The number of c.c. of the sulphocyanate solution neces- sary to produce this end point is then read off from the buret and we are ready for our correction. As the sulphocyanate solution was intentionally made too strong, the titration should yield fewer than 20 c.c. of this solution. Supposing 18.5 c.c. of sulphocyanate solution were used we must obviously add to every 18.5 c.c. of the remaining sulphocyanate solution 1.5 c.c. of water according to the equation C = -in which C represents the number of c.c. of water which must be added to the remaining solution; N the total 186 DIAGNOSTIC METHODS. number of c.c. remaining after titration; n the number of c.c. consumed in one titration, and d the difference between the number of cubic centimeters theoretically required and that actually used in one titration. The calculation would, therefore, run as follows: 79.58. We must, therefore, add to the remaining 981.5 c.c. of potassium sulphocyanate solution 79.58 c.c. of water to make the sulpho- cyanate solution of such a strength that 20 c.c. will exactly precipitate the silver from 10 c.c. of the AgNO3 solution. (3). A cold saturated solution of iron-ammonium-alum. This must be absolutely chlorin free. (4). Chemically pure nitric acid, chlorin-free and having practically no trace of nitrous acid. Specific gravity 1.2. Technic. Ten c.c. of urine or 25 c.c. of the diluted urine from which the albumin has been removed are accurately measured with a pipet and placed in a 100 c.c. volumetric flask. Five c.c. of nitric acid, 50 c.c. of water, and 20 c.c. of the standard silver solution are then added and the mixture thoroughly shaken. After this mixture has stood for about 10 minutes distilled water is added up to the graduating mark of the flask, after which the whole is thor- oughly mixed and the precipitated silver chlorid allowed to settle. This mixture is then filtered through a perfectly dry filter into a thoroughly dry 50 c.c. volumetric flask. This 50 c.c. of filtrate will represent, therefore, only 5 c.c. of urine, but the calculation made later will compensate for this. This 50 c.c. is then poured into a beaker of about 250 c.c. capacity and the volumetric flask is thoroughly washed out with water, the washings being added to the solution in the beaker. Five c.c. of the alum solution are then added and the mixture titrated with the potassium sulphocyanate solution to the appearance of the first permanent reddish tinge of the solution. The number of c.c. of sulphocyanate solution, necessary to neutralize the excess of silver remaining after the chlorid of silver has been filtered off, is then read off from the buret. Calculation. As 20 c.c. of the sulphocyanate solution are equivalent to io c.c. of the silver solution, it is evident that the number of c.c. of silver solution not used in the precipitation of the chlorids corresponds to the number of c.c. of sulpho- cyanate solution necessary to neutralize the 50 c.c. of the filtrate. We, there- fore, subtract the number of c.c. of sulphocyanate solution used from 20 (the number of c.c. of silver solution added) and obtain directly the number of c.c. of silver solution necessary to precipitate the chlorids in 10 c.c. of urine. As each c.c. of silver solution represents 0.01 gram of NaCl or 0.00606 gram of Cl, multiply these factors by the number of c.c. used, the percentage of chlorids in the urine being obtained by multiplying the amount of chlorids THE URINE. 187 in 10 c.c. by 10, and the total amount by simply multiplying this figure by the number of hundreds of c.c. in the total 24-hour specimen of urine. If the urine is very highly colored it is advisable to add a few drops of a concentrated solution of potassium permanganate before the titration. This will usually decolorize the urine so that the end point will be much more distinct. Purdy's Centrifugal Method. This method, while having nothing in common with the accuracy of the preceding one, is very convenient and has the advantage of yielding quick results which are clinically available. Ten c.c. of clear, filtered, albumin-free urine are placed in a centrifuge tube which is graduated to 15 c.c. One c.c. of strong nitric acid and 4 c.c. of a 5 per cent, solution of silver nitrate are then added. The tube is shaken by inversion and the mixture allowed to stand for a few minutes, after which it is placed in the centrifuge and whirled for three minutes at the rate of 1,200 revolutions per minute. The bulk percentage of silver chlorid is then read off, from which the percentage by weight both of sodium chlorid and of chlorin, equivalent to the precipitated silver chlorid, may be calculated. One per cent, by bulk represents 0.13 per cent, by weight of NaCl and 0.08 per cent, of Cl. As previously stated, the amount of chlorin in the urine depends upon the amount ingested, ranging normally between 10 and 15 grams. By the use of Folin's standard diet, which contains 6.2 grams of Cl, the excretion is found to be 6.1 grams of Cl in 24 hours. On the ash-free diet of Taylor the excretion at the end of 12 days of such a diet was 0.17 gram of Cl in the 24 hours. (2). Phosphates. The phosphates occurring in the urine are the sodium, potassium, cal- cium, and magnesium salts of the tribasic orthophosphoric acid (H3PO4). As previously stated, in the discussion on the reaction of the urine, normal, clear, acid urine contains no dibasic monacid phosphates, but all of the phos- phates under these conditions are of the monobasic diacid type. If the urine becomes less acid or amphoteric in reaction we find, however, in addition to the above, the disodium monohydrogen phosphate, the monocalcium phos- phate, and the monomagnesium phosphate; while if the urine be alkaline we may find the neutral phosphates in the ascendency. It must be remembered, therefore, that the normal acidity of the urine is not strictly regulated by variations in the relative proportions of the monosodium dihydrogen phosphate and of the disodium monohydrogen phosphate as usually stated. Besides these mineral phosphates, phosphoric acid is found in the urine in combination with glycerin as glycero-phosphoric acid, which is derived largely from the hydrolytic cleavage of lecithin compounds. According to Mandel and Oertel, Mathison,1 and Yoshimoto,2 the output of organically-bound phosphorus is 1 Biochem. Jour., vol. 4, 1909, p. 274. 2 Zeitsch. f. physiol. Chem., Bd. 64, 1910, S. 464. 188 DIAGNOSTIC METHODS. not increased by a phosphorus-rich diet, while Kondo,1 shows that, although absolutely increased in amount, its relation to total P2O5 is diminished. The larger portion of the urinary phosphoric acid is derived from the food, the smaller portion coming from the metabolism of the tissue protein, especially the nucleins. This endogenous phosphoric acid may be of special importance as variations will be found depending upon the degree of destruc- tion of the lecithin and nuclein compounds. It is to be remembered here that not all of the phosphoric acid ingested is excreted, as between a fourth and a third of the total quantity may remain in the feces in combination with calcium. In studying the effects of increased ingestion of phosphates, the feces must, therefore, be examined quite as closely as the urine. Physiologic Excretion. The amount of phosphoric acid excreted in the 24 hours is always ex- pressed in terms of P2O5. The normal P2O3 excretion of the adult varies from 1 to 5 grams with an average of about 3.5 grams. The figures of Folin, based upon a diet containing 5.9 grams of P2O5, show this excretion to average 3.87 grams in 24 hours, while patients on an ash-free diet eliminate approxi- mately 0.75 gram. In this excretion the phosphates of sodium and potas- sium usually exceed those of calcium and magnesium, the former being ex- creted in the amounts of 2 to 4 grams in the 24 hours, the latter from 1 to 1.5 grams. Little data exists regarding pathologic variations in the relation of these two types of phosphates so that no conclusion may at present be drawn. The excretion of P2O5 will vary with the food, especially with the amount of calcium and magnesium of the food. These bases combine in the intestine with the phosphoric radical forming phosphates which are difficultly soluble.2 This fact is taken advantage of by Croftan in the administration of calcium salts to precipitate the phosphates and thus diminish their activity in conditions attributable to uric acid. The phosphates are increased on an animal diet and diminished on a vegetable diet as Ziilzer has shown. During starvation an increase of the phosphates may be observed, as an indication of decomposition of the tissues. Administration of phosphates at this time will usually lead to a retention to counterbalance the previous loss. This same fact was observed in the discussion of the chlorids and may be stated as a general law, that an increased excretion is followed by a retention and a retention by an increased elimination. It must be stated, however, that an insufficient supply of phosphoric acid is not compensated for by such a great retention of phosphates as of the chlor- ids. The organism eliminates even more phosphoric acid in starvation than in cases of deprivation of salt, as the decomposing protein sets free the salts bound up with it. Many attempts have been made to determine where and in what form phosphoric acid is retained in the body and where and from what sources the body draws upon it for excretion. It is a difficult matter 1 Biochem. Zeitsch., Bd. 28, 1910, S. 200. 2 See Wurtz, Biochem. Zeitsch., Bd. 46, 1912, S. 103. THE URINE. 189 to determine what amount of the phosphoric acid retained reaches the bones, what portion is devoted to the soft tissues, and how much of it remains organi- cally combined in the body (Magnus-Levy). It has been found that the relation between the excretion of phosphoric acid and nitrogen is normally about one to seven, the same relations which exist between the amount of nitrogen and phosphoric acid in the human muscular tissue. It is, therefore, plausible to assume that a retention of both nitrogen and phosphoric acid will lead to a deposition of increased flesh. In starvation we find that this relation is markedly disturbed, the phosphoric acid being both relatively and absolutely increased. Such being the case the loss of P2O5 must be largely sustained by the bones, which are relatively poor in nitrogen. For a full discussion of this subject the writer would refer to von Noorden's work on Metabo- lism and Practical Medicine,1 which gives great detail regarding all phases of metabolism. The phosphates are increased during hard muscular exercise, while mental exercise seems to lead to a diminished excretion of the alkaline phosphates and an increased output of the earthy phosphates. The ingestion of large quantities of water is frequently associated with an increased elimination of the phosphates, although this is later followed by a slight retention. It not infrequently happens that a freshly voided urine shows a marked turbidity and even precipitation due to the deposit of earthy phosphates. This has been supposed to be due to an increased output of the phosphates, but it is now known to be nothing but the natural consequence of a change of urinary reaction from acid to alkaline. This condition which has been called 11 phosphaturia" would, therefore, much more appropriately be styled il alkalinuria." This subject will be discussed in a later section to which the reader is referred. A diminished elimination may be observed in cases of acute febrile disease, especially at the height of pneumonia. The degree of diminution is usually proportionate to the severity of the disease and usually lessens as convalescence comes on. According to Gouraud, the earthy phosphates are considerably reduced in pneumonia, while in tuberculous conditions the phosphates are increased, an interesting point in differential diagnosis. This retention in pneumonia as well as in the other acute febrile diseases is pos- sibly due to the renal insufficiency which may be very great in such conditions. This diminished phosphatic excretion may not always obtain in the acute febrile conditions, in some cases a sudden increased output being observed. In typhoid fever Robin believes an increased elimination during the febrile rise to be an unfavorable sign, while an increase during defervescence indi- cates a favorable prognosis. The phosphates appear to be diminished in most chronic diseases. In all renal diseases, whether acute or chronic, a diminished excretion is present Pathologic Variations. 1 Chicago, 1908, Keener. 190 DIAGNOSTIC METHODS. due to the renal insufficiency. This diminished phosphoric acid excretion is regarded by Purdy1 as a factor almost as constant as is the excretion of al- bumin. In gout the phosphoric acid excretion runs parallel to that of uric acid, decreasing immediately preceding the acute attack and rising as the attack subsides. In cases of pregnancy a diminished excretion is observed which is attributable to the withdrawal of phosphoric acid from the maternal organism for the purpose of the fetal bone formation. In certain bone dis- eases, such as osteomalacia, a diminished excretion is usually observed, although at times an actual increase is seen. The earthy phosphates, especially, are diminished in these latter conditions while the alkaline phosphates may be increased. In cases of myositis ossificans the excretion of inorganic phosphates does not seem to be much affected as one might expect from the new bone formation. In cases of hystero-epilepsy the phosphates are diminished, the diminu- tion usually being proportionate to the intensity of the attack, while in true epilepsy the phosphates appear to be more or less markedly increased. It is in just the nervous diseases that one would expect to find much variation in the phosphatic excretion, but very few data are found bearing on this sub- ject. Folin and Shaffer find that in the periods of nervous excitement the relative amount of phosphoric acid is diminished, but that the absolute amount is little changed. In Addison's disease, hepatic cirrhosis, acute yellow atrophy, and chronic lead-poisoning we may find an extensive decrease of phosphates in the urine. In certain cases which show most of the symptoms of diabetes mellitus without any sugar output, a phosphatic increase is observed in the urine. This condition has been called "phosphatic diabetes" and may be associated with the excretion of as high as io grams of P2O5 within 24 hours. In true diabetes mellitus the phosphates may be increased at one time and diminished at another, as there seems to be an inverse ratio between the excretion of sugar and that of the phosphates. The phosphates seem to be increased in cases of pseudoleukemia, leuke- mia, hemorrhagic purpura, in cases of acute or chronic inflammatory processes of the genito-urinary tract, and in cyclic vomiting of children. As previously stated, the output of urinary nitrogen bears a relation of about seven to one to that of the phosphate excretion. This relation has been termed the "relative value" of phosphoric acid and represents the amount of P2O5 corresponding to 100 grams of N. Normally this ranges between 15 and 20. Estimation of Phosphates. Ten c.c. of urine are rendered alkaline with ammonia. The earthy phos- phates are precipitated in the form of a flocculent precipitate and may be roughly estimated by the volume of the precipitate. 'Practical Urinalysis, Phila., 1900, p. 56, THE URINE. 191 If this alkalinized urine be filtered and the filtrate acidified with acetic acid, the addition of a few drops of ferric chlorid or of uranium nitrate solu- tion will precipitate the alkaline phosphates. These methods are purely qualitative and can have no clinical value beyond giving a general idea of the relative amounts of the earthy and alkaline phosphates. Instead of ferric chlorid or uranium solution, magnesium mixture may be used for this purpose. Quantitative Determination. The best method for the estimation of the urinary phosphates is that of titration with uranium nitrate or acetate solution. The principle of this method is. that phosphoric acid compounds in acetic acid solution give, on treatment with uranium nitrate, a yellowish-white flocculent insoluble pre- cipitate of uranium phosphate (UO2HPO4). As a means of recognizing the point at which an excess of uranium solution is present in the titrated fluid, one may use either a solution of ferrocyanid of potassium which gives a dis- tinct browish color at the end point, or, preferably, a few drops of tincture of cochineal, which gives a grass-green color and has the advantage that it can be added directly to the titrated fluid, which is not the case with the ferro- cyanid of potassium. (i). A solution of uranium nitrate or acetate of such a strength that 20 c.c. shall correspond to 0.1 gram of P2OS. It is a matter of absolute indifference whether the acetate or the nitrate be used, but the writer prefers the nitrate as this is more easily obtained in the pure state. In making up a solution of uranium nitrate of the above strength, one may not rely implicitly on the weighing, as the uranium nitrate may contain impurities or excess water and thus vitiate the results. It is, therefore, necessary to have a standard phosphate solution against which the uranium solution may be titrated. The usual solution recommended by various writers is one of disodium monohydrogen phosphate. This salt varies in its degree of hydration and its solutions do not keep well. Moreover, it is absolutely necessary when this salt be used that a definite amount of it be taken and converted into sodium pyrophosphate, after which a corresponding dilution of the solution must be made to make it of such a titer that every 50 c.c. shall be equivalent to 0.1 gram of P2OS. In view of these facts, the writer is accustomed to follow the suggestion of Giles1 and use chemically pure dihydrogen monopotassium phosphate. This salt crystallizes well without any water of crystallization and does not alter on exposure to the air. This solution is to be made such a strength that 50 c.c. corresponds to 0.1 gram of P20s, in other words a liter must contain 2 grams of P2OS. In order to find out just how much of this salt must be dissolved in a liter of water we must have recourse to a simple calculation. The formula of dihydro- Necessary Solutions. 1 Sutton's Volumetric Analysis, Philadelphia, 1904, p. 294. 192 DIAGNOSTIC METHODS. gen monopotassium phosphate is KH2PO4, its molecular weight being 136. Two molecules of this salt are necessary to yield one molecule of P2OS accord- ing to the equation 2KH2PO4=P2Os + K2O + 2H2O. If, therefore, one liter of the solution must contain 2 grams of P2O5, the amount of KH2PO4 which must be dissolved in a liter is easily calculated from the following proportion : 272 : 142 :: x : 2. x = 3-83- We, therefore, dissolve 3.83 grams of dihydrogen monopotassium phosphate in 1 liter of water and obtain directly a solution which contains 2 grams of P2OS or one in which every 50 c.c. is equivalent to 0.1 gram of P2OS. It is Fig. 70.-Volumetric flasks. perhaps, needless to add that this solution should be made in an accurately standardized volumetric flask, and at the temperature at which the flask is calibrated. Having thus obtained our standard phosphate solution we are now in a position to make up our standard uranium nitrate solution, the titer of which must be such that 20 c.c. corresponds to 0.1 gram of P2OS, or, in other words, one liter of which must be equivalent to 5 grams of P2OS. The formula of uranium nitrate is UO2(NO3)26H2O, its molecular weight being 502.6. Ura- nium nitrate combines with dihydrogen potassium phosphate according to the following equation : UO2 (NO3)2 + kh2po4 =uo2hpo4+ hno3 + kno3. THE URINE. 193 As seen above, two molecules of the dihydrogen phosphate are necessary to yield one molecule of P2O5. We will, therefore, have, when uranium nitrate acts upon the dihydrogen phosphate, only the equivalent of 1/2 molecule of P2OS; that is, 71 parts. As the uranium solution must contain 5 grams of P2OS to the liter we may then calculate how much uranium nitrate is necessary to form the equivalent of such a solution by the following proportion: 502.6 : 71 " x •• 5- x = 35-39- Were we absolutely certain of the purity and state of hydration of our uranium nitrate, all that would be necessary would be to weigh out this exact amount. As this is not the case, we weigh out a slight excess (35.75 grams) and dissolve in one liter of distilled water. We are now ready to determine the strength of the uranium solution as follows: Fifty c.c. of the dihydrogen monopotassium phosphate solution are placed in a beaker and treated with a few drops of tincture of cochineal and 5 c.c. of acetic acid mixture (see below, solution 2). Some workers prefer the ad- dition of potassium ferrocyanid as an indicator, but this does not give as distinct a contrast at the end point, and if tests are made by adding a drop of the mixture to the ferrocyanid solution on a white plate, loss of substance must occur. This mixture is then heated and titrated, as soon as the boiling point is reached, with the uranium solution until a trace of a distinct green color becomes permanent on stirring the mixture. Duplicate determinations are then made, the results of which should agree exactly with the original. The number of c.c. of uranium solution used is then read off and we are prepared for the calculation of the amount of water which must be added to stand- ardize the solution. As 20 c.c. of this uranium solution should correspond exactly to 50 c.c. of the standard phosphate solution, we may, for the sake of example, use the same figures given for obtaining the dilution in the case of the sulphocyanate solution discussed under the heading of Chlorids. Thus if 18.5 c.c. of ura- nium solution were used we must add, according to the previous explanation, 79.58 c.c. of distilled water to the remaining 981.5 c.c. of uranium solution in order to make every 20 c.c. equivalent to 0.1 gram of P2O5. (2). An acetic acid mixture prepared by dissolving 100 grams of sodium acetate and 30 grams of glacial acetic acid in sufficient water to make 1000 c.c. This solution must be added in the determination of the urinary phosphates in order to overcome the influence of the nitric acid liberated in the reaction and to convert any monacid phosphates into the diacid type. (3). An indicator, preferably tincture of cochineal prepared by digesting the ground cochineal bugs in 25 per cent, alcohol and filtering. This indicator has the advantage that it may be added directly to the solution to be titrated, while potassium ferrocyanid must be used by the plate method of adding a few drops of the solution to the indicator after each addition of uranium solution. 194 DIAGNOSTIC METHODS. Technic. Fifty c.c. of clear filtered urine are placed in an Erlenmeyer flask and treated with 5 c.c. of the acetic acid mixture, for the purpose of transforming any monacid phosphates into the diacid form and of neutralizing the nitric acid formed during the titration. A few drops (5 to 10) of tincture of cochineal are added, the mixture heated to the boiling-point, and then titrated as de- scribed above. It is wise invariably to run duplicate determinations. After each addition of the uranium nitrate the precipitate is allowed to settle so that one may see more clearly the first trace of any green coloration or precipitate. The calculation is as follows: Supposing 10 c.c. of the uranium solution were used, the corresponding amount of P2O5 in the 50 c.c. of urine examined would then be found from the equation: 20 : 0.1 : : 10 : x. x=o.o5. The percentage of P2O5 would, therefore, be 0.1 (2 x 0.05). If the total 24-hour urine were 1500 c.c., the total P2O5 excretion would be obviously 1.5 grams. Total Phosphoric Acid. The above determination gives the total P2O5 excretion as far as the inorganic phosphates are concerned, but it does not take into consideration the organically-bound phosphorus. This is usually very small in amount, but it is sometimes of value to know whether any variations exist. This may be determined by incinerating 50 c.c. of urine and determin- ing the total phosphoric acid in the ash. This, however, is not as easily done as might be imagined, owing to the fact that the urinary residue is difficultly incinerated. The writer prefers, therefore, the method of Asboth-Folin which will be discussed in the section on Sulphates. The mass fused by this method is dissolved in hot water, the solution filtered and washed with hot water. This solution is then slightly acidified with HC1, treated with the acetic acid mixture, and titrated as in the previous determination. The difference between the figures obtained here and those obtained by direct titration of the urine will yield the amount of organically-bound phorphorus in terms of P2O5. The normal output of this organic phosphorus varies from 0.01 to o. 1 gram in 24 hours, forming about 2 or 3 per cent, of the total P2O5. Phosphorus-rich diets increase the absolute amount but diminish the relative percentage. Purdy's Centrifugal Method. This method cannot be relied upon for accurate results in metabolic work, but may be of some service from the clinical standpoint. Ten c.c. of clear filtered urine are placed in a centrifuge tube graduated to 15 c.c. Two c.c. of 50 per cent, acetic acid, and 3 c.c. of 5 per cent, uranium nitrate solu- tion are then added and thoroughly mixed with the urine by inversion of the tube. The tube is then placed in the centrifuge and operated at a speed of THE URINE. 195 1,200 revolutions for three minutes. According to Purdy, 1 per cent, by bulk of uranium phosphate equals 0.04 gram of P2O5 in each 100 c.c. of urine. Each succeeding percentage by bulk increases by the figure 0.01. Thus a bulk percentage of five of uranium phosphate would equal 0.04 plus 0.04, or 0.08 gram of P2O5 in each 100 c.c. These figures are much at vari- ance with those of Ogden, who states that "he has found that each 1/10 of a c.c. of precipitate calculated as P2O5 is equivalent to 0.0225 Per cent- by weight." Owing to these differences, the writer would suggest that the bulk percentage be stated as such rather than as parts by weight of P2O5. (3). Sulphur Compounds. The sulphur is present in the urine in three forms: (i) preformed or neutral sulphates; (2) ethereal or conjugated sulphates, sulphuric acid in combination with aromatic compounds, and (3) neutral, unoxidized, or organic sulphur. The total output of sulphur depends essentially upon the protein metabolism, both of that of the tissues and of the food. It is to be remembered that the sulphur elimination is much less accurate than that of the nitrogen as an indi- cation of the degree of protein metabolism, owing to the fact that different protein substances vary in their sulphur-content. The daily excretion of sulphur, in terms of SO3, varies from 1 to 3.5 grams, when the subject is upon a mixed diet. Ordinarily, the ethereal sulphates form about one-tenth of the total output. The neutral sulphur does not vary under normal conditions as far as its absolute amount is concerned, but we note, on changing the diet to one which is relatively free in protein material, that the relative amount of the neutral sulphur is markedly increased. Thus, Folin finds on a diet containing 18.9 grams of nitrogen and 3.8 grams of SO3 the daily excretion of total SO3 is 3.31 grams, of which the inorganic SO3 is 2.92 (87.8 per cent, of total), the ethereal SO3 is 0.22 (6.8 per cent), and the neutral SO3 0.17 (5.1 per cent.). On a nitrogen-free diet, consisting of cream and arrowroot, a total SO3 excretion of 1.04 grams is noted, of which 0.63 gram (60.6 per cent.) is traceable to the inorganic SO3, 0.12 (11.5 per cent.) to the ethereal SO3, and 0.29 (27.9 per cent.) to the neutral SO3. We are, therefore, con- fronted with the following fact, "the distribution of the sulphur in urine among the three chief normal representatives, inorganic sulphates, ethereal sulphates, and 'neutral sulphur,' depends on the absolute amount of sulphur present." On the ash-free diet of Taylor we find, according to Goodall and Joslin, a total SO3 excretion of 0.96 gram, of which the inorganic SO3 forms 0.71 gram (74 per cent.), the ethereal sulphates 0.05 gram (5.2 per cent.), and the neutral sulphur 0.2 gram (20.8 per cent.). It is doubtless true that practically all of the urinary sulphur is derived from protein metabolism, a definite relation being usually established between the nitrogenous and sulphur output. Normally, N : SO3 :: 5 : 1, apparently regardless of whether the patient is on a nitrogen-rich or a nitrogen-free diet; the absolute amounts of each, however, differ markedly, depending upon the 196 DIAGNOSTIC METHODS. diet. Folin's figures show that as the total urinary sulphur is reduced, the percentage represented by the inorganic sulphates sinks from about 90 to 60 per cent. This fact has been expressed in the above quotation from Folin. The reduction in the inorganic sulphates must be made up by a relative increase in the other forms of sulphur. The ethereal sulphates have for a long time been held to be an accurate index of the degree of absorption of the products of intestinal protein decomposition. There can be no question but that increased intestinal decomposition is associated with increased out- put of the conjugated sulphuric acids, especially the indoxyl and skatoxyl sulphuric acids. These ethereal sulphates are diminished on a milk diet or on the cream and arrowroot diet of Folin or the vegetarian diet adopted by many of Chittenden's subjects. This is true especially as regards the urinary indican which is absolutely negative in such cases, while the total amount of ethereal sulphate is diminished only about 50 per cent. Such being the case we must have some other than intestinal origin for the large relative increase of the ethereal sulphates under a nitrogen-free diet. I quote from Folin r1 "(1) The urinary indican is not to any extent a product of the general protein metabolism, is therefore probably, as is generally sup- posed, a product of intestinal putrefaction, and may consequently be assumed to indicate approximately the degree of putrefaction in the intestinal tract. (2) The ethereal sulphates can only in part be due to intestinal putrefaction, and neither their absolute nor their relative amount can be accepted as an index of the extent to which the putrefaction is taking place in the intestines. (3) The ethereal sulphates, on the contrary, represent a form of sulphur metabolism which becomes more prominent when the food contains little or no protein." Here the sulphuric acid is conjugated with aromatic bodies formed from decomposition of tissue protein. The sulphates, as a whole, must be increased in any condition associ- ated with increased protein catabolism. Thus we find in febrile conditions an increased output of sulphur corresponding to the intensity of the process, this increase being followed by a diminution as convalescence comes on. An increased elimination of sulphates has been observed in leukemia, diabetes mellitus and insipidus, progressive muscular atrophy, and following the use of such drugs as morphin, potassium bromid, sodium salicylate and acet- anilid. From the clinical standpoint the elimination of sulphates has little practical value, the variation in the amounts of ethereal sulphates and neutral sulphur being the chief factors of value. While, as shown above, the ethereal sulphates are subject to great varia- tion, the indoxyl-potassium sulphate (indican) varying according to the degree Qf intestinal decomposition, we find some points of clinical interest in their Pathologic Variations, 1 Loc. cit. THE URINE. 197 study. As the putrefactive processes normally occur below the ileo-cecal valve, any condition increasing such decomposition with a consequent in- crease of urinary indican would indicate trouble in the lower bowel, more frequently of the chronic type. They are increased in cases showing abnormal intestinal absorption, as, for instance, in typhoid fever, intestinal tuberculosis, peritonitis, and chronic intestinal catarrh. Obstructive jaundice is usually associated with increase in the output of indican as the bile seems to have a great influence upon putrefactive processes in the intestine. They appear to be increased in cholera, while in ordinary diarrhea they are diminished in absolute amount, but may be relatively increased. In cases of gastric hypoacidity associated with bacterial decomposition in the stomach the urine may show a marked increase of indican. In acute nephritis a very intense indican reaction may be obtained, while in the chronic form the amount of indican is usually diminished. In cases of pus-formation almost'anywhere within the system an increased elimination of ethereal sulphates may be ob- served due to the absorption of the products of decomposition of the pus. This point is of some importance in differentiating purulent from non-purulent affections of various organs. The writer has seen a persistent intense indi- canuria, associated with middle-ear infection, which cleared up completely after thorough drainage. The neutral sulphur does not have a very definite relation to the amount of sulphur of the intake or to the amount formed in the decomposition of tissue protein. While it is said to vary in amounts representing from 12 to 15 per cent, of the total SO3 of the urine, this must be true only when the absolute amount of excreted SO3 is taken into consideration. By this is meant that although the absolute amount of neutral sulphur does not normally vary to any great extent, its percentage relation to the total sulphur varies with the amount of sulphur. Thus Folin finds on an intake of 3.8 grams of SO3 an output of 0.17 gram of neutral SO3, while on a nitrogen-free diet the out- put is 0.2. The nature of the neutral sulphur of the urine is somewhat uncertain, as we seem to have only two well-established bodies, namely, the sulpho- cyanates and hydrogen sulphid. The sulphocyanates are derived largely from the absorption of material from the saliva and represent approximately one-third of the total neutral sulphur. The hydrogen sulphid may be regarded as a decomposition product. Besides these we find cystein, which is an in- termediary product of normal protein metabolism, and tauro-carbamic acid, which is derived from the biliary material. In cases of jaundice we may find as high as 60 per cent, of the sulphur in the neutral form, which may be due to the absorption of this and other biliary material. Traces of chondroitin- sulphuric acid, oxyproteic acid, alloxyproteic acid, and uroferric acid con- tribute to the neutral sulphur-content of the urine. Neutral Sulphur 198 DIAGNOSTIC METHODS. The greatest increase of neutral sulphur is probably associated with the presence of cystin which is not normally present in the urine. This is undoubtedly derived from abnormal protein decomposition. The sulphur in the neutral form may reach as high as 30 or 40 per cent, of the total, due to the presence of cystin. The writer will refer to a later section for a discus- sion of the subject of cystinuria. The variations in the neutral sulphur of the urine must be regarded as indicative of abnormal metabolic processes which are not associated with variations in the other types of urinary sulphur. As Folin has shown, "the neutral sulphur is not at all due to processes identical or similar to those which give rise to indican. The neutral sulphur represents products which in the main are independent of the total amount of sulphur eliminated or of pro- tein catabolized." The method followed by the writer in determining the total urinary sulphur in terms of SO3 is the Folin1 modification of the Asboth-Modrakowsky method. Twenty-five c.c. of urine (50 c.c. if very dilute) are measured into a large nickel crucible of about 200 c.c. capacity and treated with 3 grams of so- dium peroxid.2 The mixture is evaporated to a syrupy consistency and is then carefully heated to dryness. The latter part of this procedure requires about 15 minutes. The crucible is removed from the flame and allowed to cool. Moisten the residue with 1 or 2 c.c. of water, add about 7 grams of sodium peroxid and heat the mixture to complete fusion for about 10 minutes. Allow the fused mass to cool, add about 100 c.c. of water and heat to boiling for at least one-half hour to dissolve the alkali and decompose the peroxid. This mixture is transferred to an Erlenmeyer flask (about 400 c.c. capacity) by means of hot water and is diluted to about 250 c.c. Concentrated HC1 is slowly added to the almost boiling solution until the oxid of nickel (formed in the process) just dissolves. This will require about 18 c.c. of acid for 8 grams of Na2O2. After boiling for a few minutes, the solution should be perfectly clear. If it is not clear too much water or too little peroxid were used in the last fusion. Allow the mixture to cool and filter from the insoluble residue, if there is any. To this clear acid solution add 5 c.c. of very dilute (1 to 4) alcohol and boil for a few minutes to remove the last traces of chlorine, which are formed on acidi- fying the solution. Now add, drop by drop, 10 c.c. of a 10 per cent, barium chlorid solution and allow the precipitate of BaSO4 to settle out for two days in the cold. At the end of this time the precipitate is filtered off, using either ash- free filters or, preferably, weighed porcelain Gooch crucibles with properly prepared asbestos mats. Wash the precipitate with about 250 c.c. of cold water, dry and ignite for 15 minutes. In this heating of the Gooch crucible, the flame must touch neither its perforated bottom nor its sides. If the crucible Estimation of Total Sulphur (A) 1 Jour, of Biol. Chem., vol. i, 1906, p. 131; Jour. Amer. Chem. Soc., vol. 31, 1908, p. 284. 2 Benedict, Denis, and Schmidt use, as the oxidizing agent, a solution of copper nitrate and potassium chlorate. THE URINE. 199 be placed upright upon the cover of a platinum crucible, which rests upon a platinum triangle, the flame may be applied directly to the bottom of the platinum. After the heating, the crucible is placed in the desiccator, is allowed to cool and is weighed. Knowing the weight of the crucible and mat before and after this process, a simple calculation yields the amount of BaSO4 obtained from the urine taken. If the filter-paper be used, the precipitate is washed as previously mentioned. Transfer the filter-paper and precipitate directly to a weighed platinum or porcelain crucible, add 3 or 4 c.c. of alcohol and ignite, to dry and partially burn the filter-paper. Heat the residue until complete incineration occurs and the ash becomes colorless. Cool the crucible in a desiccator and weigh. If the amount of BaSO4 obtained by either of these methods be multiplied by 0.3429, the result is the amount of SO3 in the urine taken. A simple calculation yields the percentage and total amount excreted. Determination of Total Sulphates (B). Folin's Methods. Under this head we determine the neutral (preformed or inorganic) and ethereal sulphates. Folin1 has advanced two methods, the first of which is preferable. i. Precipitation in the Cold.-25 c.c. of urine and 20 c.c. of dilute hydro- chloric acid (1 to 4), or 50 c.c. of urine and 4 c.c. of concentrated HC1, are gently boiled in an Erlenmeyer flask (capacity about 200 c.c.) for 20 to 30 minutes (never less than 20). To prevent loss of material it is wise to keep the flask covered with a small watch-glass during the boiling. The flask is cooled for two or three minutes in running water and the contents are then diluted with cold water to about 150 c.c. To this solution are then added 10 c.c. of a 5 per cent, barium chlorid solution, care being taken not to shake or stir the contents during the addition. At the end of an hour or more, the mixture is shaken up and filtered. The rest of the determination is as outlined above. 2. Precipitation in the Heat.-The boiling of the urine with hydrochloric acid is conducted exactly as in the preceding method. At the end of 20 to 30 minutes, the boiling urine is diluted to about 150 c.c. with hot water. The mixture is heated once more to the boiling point, is removed from the flame, and immediately precipitated by the addition of 5 c.c. of 10 per cent, barium chlorid solution, which must be added drop by drop. The mixture is allowed to stand for 2 hours in order to cool. The remainder of the process is as pre- viously given. Determination of Inorganic (Neutral) Sulphates (C). About 100 c.c. of water (not less), io c.c. of dilute hydrochloric acid (i to 4), and 25 c.c. of urine (50 c.c. if dilute, when a correspondingly smaller amount of water is taken) are measured into an Erlenmeyer flask of about 200 c.c. capacity. 10 c.c. of a 5 per cent, solution of barium chlorid are added, using the drop method, the urine solution not being shaken or disturbed during 'Loe. cit. 200 DIAGNOSTIC METHODS. this addition. At the end of an hour or more, the mixture is shaken and filtered. The remaining steps are given above. The figure obtained here represents the amount of SO3 referable to the preformed sulphates. The ethereal sulphates (D) need not be determined separately as the difference between the total sulphates (B) and the inorganic sulphates (C) will represent the amount of these ethereal sulphates in the specimen examined. If the SO3 referable to the total sulphates (B) be subtracted from the SO3 of the total sulphur (A) the remainder represents the neutral sulphur in terms of SO3. Purdy's Centrifugal Method. Ten c.c. of clear urine are placed in a centrifuge tube and to it are added 5 c.c. of barium chlorid mixture, consisting of four parts of barium chlorid, one part of concentrated hydrochloric acid, and 16 parts of distilled water. The tube is inverted to insure mixing of the reagent and urine and is allowed to stand for a few minutes. It is then placed in a centrifuge and whirled three minutes at the rate of 1,200 revolutions per minute. Each percentage of BaSO4 by bulk represents approximately 0.25 per cent, of SO3 by weight. This result can, of course, not be as accurate as the preceding, but has some clinical advantage. (4). Carbonates. A freshly voided specimen of urine may contain small quantities of car- bonates and bicarbonates and some free carbonic acid. The amount of free carbonic acid varies with the degree of acidity of the urine and the amount of carbonate-forming material in the food. It has been found that vegetable foods are almost always productive of an alkaline urine, owing to the fact that the organic acids of the vegetables may be converted into carbonates and excreted as such. The carbonate which most frequently forms as a sedi- ment in the urine is calcium carbonate which will be treated in a later section. If the urine be acidified and a stream of air passed through this acidified urine into a vessel containing a solution of barium hydrate, the carbon-dioxid liberated by the acids will unite with the barium forming barium carbonate which may be filtered, dried, and weighed. Such quantitative determinations are rarely made and have practically no clinical value. (5). Sodium and Potassium. These metals exist in the urine in the form of the oxids Na2O and K2O, the former being present in amounts of 4 to 7.5 grams, the latter varying from 2 to 4 grams. The normal relation between the excretion of these prod- ucts is as 5 to 3. The sodium is largely derived from the addition of sodium chlorid to the diet, while the potassium is a constituent of most vegetable foods. We find, therefore, that both of these substances depend largely upon the diet in normal cases, while in pathologic conditions the potassium may be excreted in larger amounts.1 Thus we find in fever the potassium salts predominate 1 See Blumenfeldt, Inaug. Dissert., Berlin, 1912. THE URINE. 201 over the sodium compounds up to the time of crisis, after which the sodium salts again assume their normal proportions. Increased exercise as well as increased decomposition of protein from pathologic causes will tend to increase the potassium of the urine. From the clinical standpoint the quantitative determination of the sodium and potassium of the urine has little significance, so that the writer will refer to text-books of physiological chemistry for such procedures. (6). Calcium and Magnesium. Both of these alkali-earth metals are excreted in the urine largely in the form of the phosphates. Calculated as the oxids, the calcium excretion (CaO) varies from o.i to 0.3 gram and the magnesium (MgO) from .0.15 to 0.4 gram in 24 hours, the relation between the calcium and magnesium excretion being about 1 to 1.5. The chief source of these compounds is the food, but it must be remem- bered that only a small portion of these substances is excreted in the urine. Calcium and, to a less extent, magnesium form compounds with phosphoric acid in the bowel and are excreted without being absorbed. Even if these compounds be injected subcutaneously the excretion is largely into the intes- tine. We should, therefore, study much more closely the calcium metabolism by examination of both the feces and urine. Little is known regarding the output of either of these metals, but what work has been done is more closely connected with the calcium excretion than that of magnesium. During starvation calcium oxid is increased both relatively and absolutely, its probable source being the bones. It is increased to some extent by exercise and is diminished after the administration of alkalies. In chronic diseases we note an increase due, probably, to inanition. In tuber- cular conditions the calcium output is occasionally found to be diminished, while some cases of a marked increase have been noted. In diabetes mellitus, as in other conditions associated with an acidosis, the output of CaO may be greatly increased, thus running parallel to the ammonia output. In a case of myositis ossificans studied by the writer there seemed to be no great variation in the calcium output, while the magnesium content of the urine was markedly reduced. Austin has reported similar findings. McCrudden and Fales report (Arch, of Int. Med., vol. 9, 1912, p. 273) a loss of calcium and a retention of magnesium in non-puerperal osteomalacia. The same authors (Jour, of Exp. Med., vol. 15, 1912, p. 450) find the calcium excre- tion in the urine, in cases of intestinal infantilism, to be almost negligible. Peabody (Jour, of Exp. Med., vol. 17, 1913, p. 71) shows a retention of cal- cium and a normal excretion of magnesium in pneumonia. Quantitative Determination. Calcium. McCrudden1 has recently introduced a method for this determination which, in my opinion, yields the best results. If the urine be alkaline it must 1 Jour, of Biol. Chem., vol. 7, 1910, p. 83; Ibid., vol. 10, 1911, p. 187. 202 DIAGNOSTIC METHODS. be made neutral or slightly acid with HC1 and be filtered. If turbid or faintly acid to litmus, add 10 drops of concentrated HC1. Two hundred c.c. of the clear acid urine are treated with 10 c.c. of a 2.5 per cent, oxalic acid solution and 8 c.c. of 20 per cent, sodium acetate solution. Allow the mixture to stand over night at room temperature or shake vigorously for 10 minutes. Filter off the precipitate of calcium oxalate and wash it Cl-free with 0.5 per cent, ammonium oxalate solution. Save the filtrate for the determination of mag- nesium. The precipitate of calcium oxalate is then dried on the filter-paper at ioo° C. and is placed in a weighed platinum crucible and burned over the Bunsen flame until the filter-paper is completely ashed. The blast-lamp is then applied and the crucible and contents blasted for 15 or 20 minutes. The crucible is dried in the desiccator and weighed, the increase in weight repre- senting the calcium oxid (CaO) in the 200 c.c. of urine. A simple calculation will yield the amount of CaO in the 24-hour specimen. Magnesium. The filtrate and washings obtained in the above determination are treated with one-third the volume of 25 per cent, ammonium hydrate. A precipitate of magnesium ammonium phosphate (NH4MgPO4) occurs, which is allowed to settle for several hours, is collected on an ash-free filter, is thoroughly washed with water containing one-third its volume of ammonia, and is dried in the oven at ioo° C. The filter-paper and precipitate are placed in a weighed platinum crucible and burned until the filter-paper is completely destroyed, after which the crucible and contents are heated over the blast-lamp for 15 minutes. This heating converts the magnesium ammonium phosphate (NH4MgPO4) into Magnesium pyrophosphate (Mg2P2O7); one part of mag- nesium pyrophosphate represents 0.36243 part of MgO. All that is neces- sary, therefore, is to multiply the amount of magnesium pyrophosphate by the above factor to obtain the amount of MgO in 200 c.c. of urine. (7). Iron. Iron is practically always present in the urine in organic combination. The amount actually present is very small, being given by Marnier as varying between 3 and 11 mg. to the liter, while Neumann and Mayer find the normal output to average 0.983 mg. The urinary iron has little clinical significance. It is increased in fever, in malaria, diabetes mellitus, and pernicious anemia, the amounts in some cases running as high as 20 mg. during the 24 hours. Any method for the estimation of the urinary iron must be very delicate and very accurate. The urine must be completely incinerated, the method of Neumann1 employing a mixture of concentrated sulphuric and nitric acid being by far the best. For a discussion of the technic the writer will refer to other works.2 ^rch. f. Anat. u. Physiol., Physiol. Abth., 1900, S. 159. 2 See Goodman (Jour, of Biol. Chem., vol. 12, 1912, p. 37) for the urinary iron in pneu- monia. THE URINE. 203 (c). Organic Constituents. (1). Nitrogenous Bodies From the metabolic standpoint the estimation of the total nitrogen of the urine is one of the most important features of its chemical examination. The excretion of total nitrogen varies with the amount of nitrogen of the food and with the degree of tissue metabolism. Normally, the system so adapts itself to the nitrogenous intake that the excretion of nitrogen, in the urine, feces, perspiration, etc., is equal to the intake. In other words, a normal person is in nitrogenous equilibrium. This subject of nitrogenous equilibrium is of such great importance and the factors which influence it so varied that the writer feels that it will be unwise to discuss it briefly for fear of not pre- senting it clearly to the student's mind. As a full discussion of this subject would be too extensive for the scope of this work the writer must be content with reference to the admirable discussion by Magnus-Levy1 in von Noorden's Hand-book of Pathology of Metabolism. The total nitrogen of the urine may be taken as a direct index of the protein metabolism. Upon a starvation diet, or one from which the nitrog- enous factors have been eliminated, we find a gradual reduction in the amount of urinary nitrogen. From about the fourth day of starvation the excretion becomes practically constant and continues until severe tissue-decomposition, as a sign of impending death, occurs. If, at this time, a diet rich in nitrogen be given, a certain amount of the intake will be retained, but not all. If this increased nitrogenous diet be continued a certain portion will be retained each day until the system again assumes a condition of nitrogenous equilibrium, the output equaling the intake. It is to be remembered in this connection that the carbohydrates and fats of the diet both have a certain direct influence in diminishing the protein metabolism; in other words, these substances act as protein-sparers. The normal amount of total nitrogen of the urine, with the subject upon a mixed diet, varies between io and 16 grams per day. The work of Chit- tenden2 has proven that this is much too high for individuals who desire to get the most out of their system with the least possible work. In other words "physiological economy" is much better subserved by a diet yielding from 5 to 6 grams of total nitrogen. The subjects of his experiments showed normal activities, both mental and physical, and at the end of the experiments felt much better than before them and had gained in weight. Folin3 in his xyork has shown that the normal excretion of six subjects, each observed for a period of five days, was 16 grams of total nitrogen on a diet of 119 grams of protein yielding 18.9 grams of nitrogen. On a nitrogen-free diet the excretion averaged 3.6 grams. His work has led him to the following statement which has greatly changed our ideas of nitrogenous metabolism. He says: "It (d). Total Nitrogen. 1 Chicago, 1908. 2 Physiological Economy in Nutrition, New York, 1905. 3 Loc. cit. 204 DIAGNOSTIC METHODS. may, therefore, be positively stated, as a principle in the chemistry of metabolism, that the distribution of the nitrogen in urine among urea and the other nitrog- enous constituents depends on the absolute amount of total nitrogen present." The subject of the distribution of the various nitrogenous products of the urine has been much changed by this work of Folin and of Chittenden. This distribution or, as it is better called, the "nitrogen partition of the urine," varies according to the diet. The following table shows the excretion of the various nitrogenous constituents and their percentage relations to the total nitrogen under a mixed diet and under one which is nitrogen free. Excretion in grams. Percentage of total N. Mixed N-Free Mixed N-Free Diet. Diet. Diet. Diet. Nitrogen, 16. oo 3.60 100.00 100. 0 Urea N, 13.90 2. 20 86.87 61. 7 Ammonia N, 0. 70 0. 42 4-37 u-3 Creatinin N, 0. 58 0. 60 3-63 17. 2 Uric Acid N, 0.12 0. 09 o- 75 2- 5 Undetermined N, 0. 70 0. 29 4-37 7-3 It will thus be seen that the total nitrogen excretion of the urine is made up of several factors. The principal points to be gained from the above table are that the urea, on a nitrogen-free diet, is markedly reduced. This reduction must naturally be made up by increase in the other factors. We find the ammonia and especially creatinin markedly increased from the per- centage standpoint, although both are absolutely diminished. We have, therefore, a distinctly endogenous nitrogen metabolism as well as an exogenous one. The urea content should, consequently, be considered as of direct importance in estimating the degree of protein tissue metabolism, although not as usually taught. This urea output cannot at the present time be con- sidered as representing from 85 to 90 per cent, of the total endogenous nitrogen, but should be regarded more properly as representing between 60 and 65 per- cent. It is true that the urea as found under mixed diets gives this higher figure, but at least 20 per cent, of this must be placed against useless activity on the part of the system. In other words, the intake of sufficient nitrogen to yield a urea excretion amounting to 85 per cent, of the total N must be considered unnecessary. This is a somewhat enlarged expression of the fact that most people eat much more than is utilized by the system. It is seen, therefore, that in the general run of urine examinations the urea output does not represent to us the extent of the tissue metabolism, as ordinarily we have not sufficiently controlled the diet. The above table presents a remarkable percentage increase in the amount of creatinin. The figures of Folin show that the absolute quantity of creatinin eliminated, whether upon a nitrogen-rich or on a nitrogen-free diet, is remarkably constant for the same individual. Although this is influenced to a slight extent by the diet, it is so slight that it can be disregarded (see Creatinin). THE URINE. 205 Physiologic Variations. An increase in the total nitrogen is observed after a heavy protein meal. As previously stated, in starvation the nitrogen of the urine becomes constant after about the fourth day. This nitrogen excretion observed in starvation is, however, less than the minimum amount that must be given in the form of protein in order to maintain a nitrogenous equilibrium. As a rule, the amount of protein taken in is much in excess of the requirement of the system, so that the amount excreted probably represents protein which has never become a part of the system. For this reason the fallacy of considering urea as a direct representative of the protein metabolism of the tissues becomes evident. The intake of protein is, in reality, readjusted to suit the actual needs of the body, so that the urea can represent only a portion, about 60 per cent., of the total metabolic activity of the tissue protein. A physiological increase in the excretion of nitrogen is observed in the infant for four or five days after birth. The nitrogen excretion is increased when the intake of water has been greater than normal. This fact should be borne in mind in metabolic experiments in which the intake of water should be quite as much regulated and as well known as the intake of other substances. Physiologically, a diminished output of nitrogen is observed on a low nitrogen diet and also on a diet rich in carbohydrates and fat, as these latter substances provide the greater part of the necessary energy. The system must then utilize its own protein. Increased exercise is supposed to give a slight increase in the nitrogen output owing to increased muscular activity, but it is to be said that the loss of water through the increased perspiration may be such a factor in diminishing the urinary nitrogen that no increase may be observed. Certain medicaments as quinin and opium will usually diminish the output of nitrogen. Pathologic Variations. Perhaps the most marked increase in the urinary nitrogen is observed in the acute febrile infections. This increase is not due to the temperature -per se nor is there any parallelism between the urinary nitrogen and the degree of temperature. Whether or not this increase in nitrogen can be directly traced co the effects of the toxins produced by the organisms causing the d'sease is still unsettled. That this cannot be the only element is proven by the fact that Krehl and Matthes have shown that more protein is destroyed in the so-called aseptic fever than is the case in the normal organism under similar conditions, dietetic and physical. In some febrile conditions we find that the elimination of nitrogen may be reduced during the febrile period, while about the time of crisis a very marked output of urinary nitrogen may be observed. This is the well-known epicritical elimination of nitrogen. In the fever associated with acute nephritis the urinary nitrogen is not increased, but is rather diminished owing to the renal insufficiency as well as to the edema which occurs. A toxogenic decomposition of protein is found in cases of Increased Excretion. 206 DIAGNOSTIC METHODS. carcinoma, pernicious anemia, chronic tuberculosis, leukemia, scurvy, and especially in exophthalmic goiter. In cases of acute yellow atrophy and phosphorous poisoning the total nitrogen may be increased, but the percentage of urea will be very much diminished. In cases of diabetes mellitus the nitrogen excretion is usually much in- creased, due more to the effect of the increased nitrogenous diet than to in- creased endogenous protein metabolism. Likewise, in diabetes insipidus a large increase in urinary nitrogen may be observed. An increase is oc- casionally observed in cases of nephritis owing to the large albumin content of the urine. During the progress of absorption of an exudate a very high ex- cretion may be observed, the resolution of a pneumonic exudate, for instance, being easily followed by the variations in the urinary nitrogen. Diminished Excretion. A diminution in the nitrogen excretion is usually observed in convalescence from acute and chronic conditions. This is probably due to the attempt on the part of the system to make up for the losses incurred during the active progress of the disease. It may be diminished in conditions in which the absorptive power of the intestine is much reduced. If the oxidative powers of the system are very much reduced as the result of chronic conditions, the urinary nitrogen will usually be much diminished. In cases of nephritis, both acute and chronic, a large diminution in the urinary nitrogen is observed. This is due primarily to the renal insufficiency and to the associated dropsy. In such conditions a marked increase may be observed in the fecal nitrogen, especially when marked diarrhea is a complicating factor. These periods of retention of nitrogen in nephritis may alternate with periods of increased elimination, so that examinations at different periods may show greatly con- flicting results. When the water output of the urine is largely diminished, as a result of transudation, of exudation, or of increased perspiration, the total nitrogen may be reduced. Estimation of Total Nitrogen (Kjeldahl). The principle of this method is as follows:1 The nitrogenous constituents of the urine are oxidized by various oxidizing agents into ammonia. This ammonia is converted into ammonium sulphate by the sulphuric acid which is added at the same time as the oxidizing agents. After the preliminary decomposition and oxidation of the organic nitrogen into ammonium sulphate, free ammonia may be liberated by the action of strong sodium hydrate and distilled into a standard acid solution. Knowing the strength of the acid solution, one may then titrate the remaining acid with a standard alkali solution and determine how much ammonia has combined with the acid. One c.c. of tenth-normal sulphuric acid, used by the ammonia liberated in the dis- tillation, represents 0.001401 gram of nitrogen. 1 See Folin and Farmer (Jour, of Biol. Chern., vol. n, 1912, p. 493) for a micro-chemical method based on this Kjeldahl process. The ammonia formed is then determined by nessler- ization. THE URINE. 207 Technic. Five c.c. of urine are accurately measured, either with a pipet or buret, into a Kjeldahl flask of Jena glass of 800 c.c. capacity. Ten c.c. of concen- trated sulphuric acid and approximately 1 gram of copper sulphate are added, the flask placed in a hood and heated over a low flame until white fumes of sulphuric acid are given off (Gunning's modification). Five grams of potassium sulphate are then added and the mixture heated with an increased flame to boiling for one-half to three-fourths of an hour. The solution should have lost every trace of a yellowish color and should have become by this time a clear bluish-green. The worker should be cautioned regarding the fumes given off in this process and conduct his work only in a hood with a good draft. It is frequently necessary to wash down the carbon from the sides of the vessel by shaking the fluid in such a way that the carbonized material Fig. 71.-Kjeldahl's nitrogen apparatus. is carried down to the bottom of the flask. One should be cautious lest he lose some of the liquid in this manipulation, which would not only throw cut his determination but might result in a very severe burn should any of the material fall upon him. The mixture is allowed to cool completely before the further steps of the determination can be taken. Most workers advise at this juncture the transference of the material from the first flask into a second distilling flask. The writer has convinced himself that this procedure is not only unnecessary but is even unwise, as the transference may result in slight loss of material. He is, therefore, accustomed to use the same flask both for the oxidation and distillation. After the mixture has cooled the neck of the flask is thoroughly washed 208 DIAGNOSTIC METHODS. with a stream of distilled water so that every trace of material may be carried from the neck into the body of the flask. Sufficient additional water is added to bring the total up to approximately 250 c.c. A little talcum powder, a few pieces of pumice stone, or a few pieces of granulated zinc may then be added to prevent bumping of the contents when sodium sulphate separates out later in the process. Fifty c.c. of 40 per cent, sodium hydrate are then added for every 10 c.c. of sulphuric acid used in the original oxidation. Care should be taken in adding this strong alkali that none of it touches the upper portion of the neck of the flask. The alkalinized mixture is then shaken and connected with a Fresenius bulb which is attached to a Liebig condenser as shown in the accompanying cut. The outlet tube passes into an Erlen- meyer flask which should contain 50 c.c. of tenth-normal sulphuric acid. In cases with abnormally high nitrogen values it may be necessary to use a larger quantity than 50 c.c. of standard acid, but the writer has found only two instances in over 1,500 determinations in which an increased amount was necessary. The connection of the distilling flask to the bulb and con- denser should be done rapidly to avoid any possible loss of ammonia. The distilling flask is heated slowly at first and the heat increased only after boiling has become regular. If heated too quickly, spurting of the liquid may occur and traces of alkali be carried into the bulb and thence over into the standard acid. The distillation should continue until about 150 c.c. have been distilled over, which will take from 20 to 30 minutes. The writer has observed that bumping of the mixture rarely occurs, under the conditions outlined above, before the ammonia is completely driven over. This can, however, not be taken as an absolute sign that every trace of ammonia has been distilled off. In order to see whether such is the case, one must test the outlet tube with a piece of moist red litmus-paper which will turn blue in the presence of traces of ammonia. If all ammonia has not been given off, the distillation must be continued until such is the case. If no more ammonia is being evolved, the distilling flask is disconnected from the bulb so that no suction may draw the standard acid into the condenser. The connecting bulb is removed and the tube of the condenser washed with a spray of distilled water so that any material adhering may be washed into the standard acid. The outlet tube is disconnected and washed both inter- nally and externally into the standard acid solution. The standard acid solution is then titrated with tenth-normal sodium hydrate solution, using cochineal, methyl-orange, alizarin-red, or rosolic acid as an indicator. The writer prefers the use of the latter. As each c.c. of tenth-normal alkali is equivalent to each c.c. of tenth-normal acid, we sub- tract, from the original number of c.c. of acid (50), the number of c.c. of stand- ard alkali used to neutralize the remaining standard acid. The difference gives us the number of c.c. of standard acid neutralized by the ammonia given off in the distillation. As each c.c. of tenth-normal acid is equivalent to 0.001401 gram of nitrogen, we multiply this factor by the number of c.c. of acid neutralized by the ammonia and obtain the amount of total nitrogen THE URINE. 209 in 5 c.c. of urine. This result multiplied by 20 yields, of course, the percentage of total nitrogen, which may be changed into the actual total amount of nitrogen by multiplying it by the number of hundreds of c.c. in the 24-hour specimen. It goes, without saying, that the reagents used in this determination must be ammonia-free or, at least, that their ammonia content be known. Urea (NH2)2CO. A discussion of the various factors which have to do with urea excretion cannot be taken up at this time. The recent work in the laboratory of Hof- meister1 as well as that of Chittenden and of Folin has shown us conclusively that it is no longer possible to consider the rate of urea formation as a direct measure of protein metabolism. It is without question true, as Leathes2 has said, that the nitrogen, or a great part of it, may be removed from the protein, converted into urea, and expelled with the urine before the oxidation of the rest of the protein molecule has been started upon; and the fact that we can trace in the urine excreted in a given time all or the greater part of the nitrogen of the protein taken at a meal, tells us nothing whatever about the fate of that part of the protein which contains, it may be, as much as 80 or 90 per cent, of the total energy of the protein. Further, urea is not a measure of the true protein catabolism, because a great part of it is formed from nitro- gen that has never been beyond the liver; and it is not the measure of the protein energy because it is largely derived from protein by reactions which leave the energy value of the molecules from which it originates but little altered. The importance of the denitrifying and desamidization reactions of the tissues must be much more considered in the future than they have been in the past. As usually stated in text-books, urea constitutes from 80 to 85 per cent, of the total nitrogen output. It has been customary to figure directly the amount of tissue protein which must have been decomposed in order to yield this amount of urea. These figures can hardly be taken as conclusive of such decomposition. Folin has shown that a definite exogenous as well as an endogenous protein metabolism occurs. With a patient upon a nitrogen- free diet, the urea constitutes only about 60 per cent, of the total nitrogen. This would represent the true endogenous urea formation. A diet which requires the patient to eliminate much higher percentages of urea is, there- fore, causing increased systemic activity, but is not increasing the direct tissue decomposition, as this excess never becomes a part of the system. One of the most important laws of protein- metabolism is that the amount of nitrogen in the body is not increased by, or not in proportion to, an increase in the nitrogen intake. This field is too extensive to warrant a discussion by the writer, so that he will refer to the work of Magnus-Levy3 in von Noorden's 1 Lang. Beitrage zur chem. Phys, und Path., Bd. 5, 1904, S. 340. 2Problems in Animal Metabolism, Phila., 1906. 3Loc. cit. 210 DIAGNOSTIC METHODS. Hand-book of Pathology of Metabolism and to Leathes1 lectures on "Problems in Animal Metabolism." The amount of urea excreted on an average diet varies from 15 to 40 grams. Folin finds this excretion to be, on a diet of 119 grams of protein yielding 18.9 grams of nitrogen, 29.8 grams. On a nitrogen-free diet the amount of urea is 2.2 grams. This excretion will, of course, vary, depending upon the diet. Von Jaksch states that the excretion of urea bears a definite relation to the total nitrogen excretion, so that for clinical purposes direct urea determinations may well be dispensed with, as the correct urea-content of the urine may be found by multiplying the simple nitrogen of the urine by the factor 2. This statement should not be regarded seriously by the practitioner as the rule, as Folin shows, would require that 93.3 per cent, of the total nitrogen in the urine be in the form of urea. Physiologically the variations in urea will follow those of the total nitrogen so that the previous discussion applies to normal variations in the urea output. Pathologic Variations. The pathologic increase in the amount of urea is observed under the same conditions as those mentioned under an increase of total nitrogen. Thus in febrile conditions, in diabetes mellitus and insipidus, after the resorption of an exudate, in maliganant conditions, and in exophthalmic goiter, the urea may be markedly increased. In conditions associated with destruction of hepatic parenchyma, or with a diminished rate of blood-flow through the liver, the urea excretion may be very considerably diminished. Thus we find in acute yellow atrophy, carcinoma, cirrhosis, and phosphorus poisoning that the normal urea of the urine is replaced by other nitrogenous constituents. The normal function of the liver in converting ammonium compounds and amino acids into urea is so markedly interfered with that the urea may completely disappear from the urine in such cases. In acute nephritis there may or may not be a diminution in the excretion of urea, depending upon the extent of the renal insufficiency. In the chronic types of nephritis we find that the urea excretion fluctuates to a great extent, periods of increase varying with those of decrease. In the early stages, even though large amounts of albumin and casts be present, the urea may be normal, while in the later stages it is often greatly diminished. A diminished excretion of urea is observed in melancholia and in the advanced stages of general paresis, while in epilepsy and hysteria an increase or a decrease may be observed. In some cases of diabetes mellitus Hirsch- feld has shown that the urea output may be diminished as the result of delayed absorption from the intestine. In these cases of diabetes the ammonia of the urine may be markedly increased owing to its combination with acid bodies and consequent withdrawal from hepatic activity. 1 Loc. cit. THE URINE. 211 While urea is a very important substance both clinically and chemically, it is very rarely tested by qualitative methods in medical work. The writer feels, therefore, that a description of the properties of this substance would best be learned by consulting works on physiologic chemistry. Determination of Urea. The methods for the determination of urea are numerous. Many of them are inaccurate although giving results clinically of importance. In the selection of a method for the determination of urea one must be governed entirely by the importance of the urea determination in any specific case. It should be said in advance that a urea estimation is absolutely useless unless taken in conjunction with the total nitrogen. The general practitioner usually insists on knowing both the percentage and total amount of urea excreted with utter disregard both for the nitrogen of the intake and of the output. If the urea is of any value at all it should be determined with these points in view. The prevailing idea is that a percentage output of two is approxi- mately normal and he therefore bases his conclusion upon an increase or decrease with this as a standard. Not infrequently single voidings of urine are examined for the urea output. Such examinations are worse than useless and may even be harmful. The methods for the determination of urea are distinctly separable into those useful for purely clinical purposes and those for the more exact metabolic work. The general practitioner, believing as he dose in the importance of urea as an indicator of systemic activity and excretion, must have a rapid and easy method for estimation of urea. It is to be said, however, that the more exact methods would better be applied if reliable conclusions are to be drawn. Knop-Hiifner Method. The principle of this method is the decomposition of urea by means of sodium hypobromite and the measurement of the nitrogen evolved. Sodium hypobromite acts upon urea according to the following equation: CO(NH2)2 + 3NaOBr=3NaBr+2N + CO24-2H2O. The carbon-dioxide evolved in this reaction is absorbed by the excess of alkali used, so that all that is necessary is to measure the amount of nitrogen evolved. This can be done by direct measurement or by collection in tubes which are so calibrated that each c.c. of nitrogen represents a certain percentage of urea. Various forms of apparatus, which are termed ureometers, have been advised for the estimation of urea by this method. The form introduced by Knop and Hufner is probably the most accurate, but is too complicated for general clinical purposes. As this method has absolutely no claim to accuracy it is useless in scientific investigations of the nitrogen partition of the urine. For clinical purposes, however, it serves as a rough approximation of the urea output and is, for this reason, largely used by the practitioner. 212 DIAGNOSTIC METHODS. This method, applied as in the following discussion, is given merely because of the fact that it is almost the only method of estimating urea which can be carried out by the general practitioner. Doremus Ureometer. This instrument is seen in the accompanying cut. The graduations of the tube are such that the number of mg. of urea in the i c.c. of urine used in the test are directly read off instead of the number of c.c. of nitrogen formed in the reaction. As o.oi gram of urea in one c.c. represents i gram per 100 cc., this tube will furnish directly the percentage values of urea. The tube is filled with a solution of sodium hypobromite made by adding i c.c. of bromin to 40 c.c. of 20 per cent, cold sodium hydrate solution. This hypobromite solution decomposes after standing for a few days so that it is never wise to attempt to keep such a solution for any length of time. In the writer's laboratory a stock solution of 20 per cent, sodium hydrate is prepared and the bromin added to it only as occasion requires for preparing fresh solutions. In this way one will always have the material at hand and need have no fear of his stock solution decom- posing. After filling the tube with this hypo- bromite solution, 1 c.c. of urine is added by means of the curved pipet accompanying the instrument. In injecting the urine into the solution, the curved end of the pipet should be passed well under the curve of the bulb, the tube tilted slightly forward, and the urine forced into the hypobromite solution with a slow steady pressure. An evolution of gas will be observed at once and will cease in a short time. The carbon dioxid given off by the reaction of the urea upon the sodium hypobromite is absorbed by the excess of alkali and the nitrogen collects in the upper portion of the tube. As soon as the evolution of gas has ceased (5 to 10 minutes),the amount of urea is read off directly from thecalibrations of the tube as previously described. In this determination the urine should, theoretically, be free from both albumin and sugar. Practically, however, these substances are never removed as the method yields comparative rather than accurate results. Indeed, a very delicate reaction for albumin is given by this test. One may be practically certain of the presence of albumin, if the urine, on being injected into the solution, forms rather characteristic heavy bubbles and if a dense froth collects in the upper portion of the tube. This froth settles slowly so that the percent- age reading requires some time. The forms of this apparatus which substitute for the glass foot a wooden base are much to be preferred as they are not so Fig. 72.-Doremus ureometer. THE URINE. 213 easily broken. The modification of this instrument, as introduced by Hinds, is seen in the accompanying cut. In this form the urine is allowed to run in from the smaller graduated tube by opening the stop-cock. This modification is an ad- vantage, but does not yield any more accurate results than does the preceding. Folin's1 Method. The principle of this method is as follows: At a temperature of 1600 C. crys- tallized magnesium chlorid (MgCl26H2O) boils in its water of crystallization. If urea be present it is decomposed by this boiling solution into ammonia and carbon dioxid. If the conversion be carried out in acid solution, the ammonia formed will combine with the acid and may then be liberated by alkalinizing the mixture. The ammonia is distilled into a standard acid solution and may then be determined as given under the total nitrogen. In this process the pre- formed ammonia as also the trace present in the magnesium chlorid will also be deter- mined so that a separate estimation of these factors must be made and subtracted from the total amount. Technic Five c.c. of urine are measured into an Erlenmeyer flask of about 200 c.c. capacity, 5 c.c. of concentrated hydrochloric acid, 20 grams of crystallized magnesium chlorid, a piece of paraffin about the size of a hazelnut ana two or tnree arops 01 a 1 per cent, aqueous solution of alizarin-red are added. An especially con- structed safety-tube (see cut) is then inserted and the mixture boiled until the drops flowing back from the safety-tube produce a very perceptible bump or hissing sound on coming in contact with the solution (10 to 15 minutes). The temperature2 is then somewhat reduced and the heating con- tinued for one hour. It is important in this process that the reaction must not remain alkaline, and, therefore, as soon as the material in the flask turns red a very few drops of the acid distillate in the safety-tube are shaken back into the flask.3 At the end of the hour the contents of the flask are washed into a liter Kjeldahl flask with about 700 c.c. of water. Twenty c.c. of 10 per cent, sodium hydrate are then added and the ammonia distilled, as de- Fig. 73.-Doremus-Hinds ureometer. (Hawk.) ^eitsch. f. physiol. Chem., Bd. 32, 1901, S. 504; Ibid., Bd. 37, 1903, S. 548. 2 Folin advises (Handb. d. Biochem. Arbeitsmeth. Bd. 5, 1911, S. 286) the use of small glass bulbs containing solid HgCll as an indicator of the proper temperature of this reaction. This substance melts at 1530 C., which is the optimum temperature to insure complete de- composition of the urea. 3 See Folin, Jour, of Biol. Chem., vol. 11, 1912, p. 507. 214 DIAGNOSTIC METHODS. scribed under total nitrogen, into a standard tenth-normal solution of sulphuric acid. This distillation should be continued until the contents of the liter flask are nearly dry or till the distillate shows no trace of ammonia with litmus- paper. This will require about one hour. The distillate is then boiled to drive off the carbonic acid, is then cooled, and titrated with tenth-normal sodium hydrate to determine the amount of acid which combined with the ammonia formed. Alizarin-red or rosolic acid are used as indicators. One c.c. of tenth-normal sulphuric acid is equivalent to 0.001704 gram of ammonia (NH3), or 0.001401 gram of nitrogen. If the nitrogen value be multiplied by 2.143 the re- sult will be the amount of urea in the 5 c.c. of urine taken. From the total c.c. of n/10 sulphuric acid neutralized must be subtracted the n/10 sulphuric acid values for the pre- formed ammonia as well as for the ammonia which may be present as an impurity in the 20 grams of magnesium chlorid used. It has been found by Schoorl that when carbohydrates and urea are heated together they form very stable condensation products (ureids). For this reason this method of Folin does not give accurate results with saccharin urine. A combination of this method with that of Mbrner, which will be described later, will give absolutely accurate results. In the determination of the ammonia values of the preformed ammonia and of the magnesium chlorid, the later methods of Folin must be used. Morner-Sjbqvist Method. This method1 is an extremely accurate one, but no more so than that of Folin, except in saccharin urines. If albumin be present it must, however, be removed by heat and acetic acid and the original volume of the urine re- stored. If the urine contains large amounts of hippuric acid this method may not give accurate results, as Salaskin and Zaleski havu shown. Fig. 74.-Folin's urea apparatus. (Hawk.) Technic. Five c.c. of urine are placed in a flask with 5 c.c. of baryta mixture con- sisting of a saturated barium chlorid solution containing 5 per cent, of barium hydrate. One hundred c.c. of a mixture of two parts of 97 per cent, alcohol and one part of anhydrous ether are then added and the mixture allowed to stand in a closed flask overnight. It is then filtered and the residue v*ashed with fresh alcohol and ether mixture and the combined filtrates evaporated ^kand. Arch. f. Physiol., Bd. 2, 1891, p. 438; Ibid., Bd. 14, 1903, S. 247. THE URINE. 215 at a low temperature (6o° C). Urea will be practically the only nitrogenous body left in solution, with the exception of traces of ammonia. When the evaporated filtrate has been reduced to about 25 c.c. in volume, a few c.c. of water and a small amount of calcined magnesium oxid are then added, the mixture stirred, and heated to drive off the ammonia; or the residue may be treated by Folin's method as previously described. This heating is con- tinued until the vapor shows no alkalinity when tested with moistened litmus paper, a result usually obtained when about 10 to 15 c.c. of the mixture remain. The fluid and the residue are then washed into a Kjeldahl flask and treated with concentrated sulphuric acid, copper sulphate, and potassium sulphate as in the determination of total nitrogen. One part of nitrogen is equivalent to 2.143 grams of urea. If much hippuric acid be present, Braunstein1 advises oxidation of the urea by heating the evaporated residue with 10 grams of solid phosphoric acid (in an air-bath) to 140 to 1450 C. for five hours. Alkalinize and distill as previously described. Benedict's2 newer method also yields accurate results. (c). Ammonia (NH3). This substance, although chemically belonging in the class of inorganic compounds, is so closely related to the nitrogen metabolism that it is more properly discussed under the heading of Nitrogenous Bodies. Ammonia is one of the most important products of protein metabolism. It is constantly present in small amounts in normal urine averaging about 0.85 gram of NH3 in 24 hours, representing from 4 to 5 per cent, of the total nitrogen. It is present in combination with various acids and may represent largely a portion of the nitrogen which has not been transformed into urea, but has been used to combine with acid substances formed in the protein metabolism of the body. Any increase in the production of acid in the system or any increased intake of noncarbonate-forming acids -will lead to an increased excretion of ammonium salts. This is an important factor in the metabolism of conditions associated with an acidosis. The total output of ammonia will vary under normal conditions with the diet or, in other words, with the intake of total nitrogen. While the increase of the total nitrogen of the urine on increased nitrogen intake is largely in the form of urea, yet a small increase in the absolute amount of ammonia must occur. Likewise we observe a diminished intake of nitrogen reducing the absolute value of ammonia, but largely increasing its relative value. Thus Folin finds with a total excretion of 16 grams of nitrogen, an ammonia output of 0.85 gram (4.3 per cent.); while on a nitrogen-free diet a total nitrogen output of 3.6 grams was observed with an ammonia elimination of 0.51 gram (11.3 per cent.). We therefore conclude with Folin as follows: "With pro- nounced diminution in the protein metabolism (as shown by the total nitrogen in thj urine), there is usually, but not always, and therefore not necessarily, 1 Zeitsch. f. physiol. Chern., Bd. 31, 1901, S. 381. 2 Jour, of Biol. Chem., vol. 8, 1910, p. 405. 216 DIAGNOSTIC METHODS. a decrease in the absolute quantity of ammonia eliminated. A pronounced reduction of the total nitrogen is, however, always accompanied by a relative increase in the ammonia nitrogen, provided that the food is not such as to yield an alkaline ash." Although the ammonium salts of many organic acids are converted into urea in the system we find the ammonium salts of the sulphuric and phosphoric acids formed in the decomposition of protein material are excreted as such. Moreover, we observe that an increased consumption of fat, either taken in as food or derived from the tissue, is associated with a combination of am- monia with the fatty acids. This provision of metabolism, by which the system is protected against the deleterious effects of increased acidity by neutralization of acid compounds with ammonia, is of the greatest importance, as the fixed alkalies of the tissues are thereby maintained in their usual con- centration unless the pathologic processes be extreme. Pathologic Variations. An increased output of ammonia is observed in cases of diminished oxida- tive powers of the system, in febrile diseases, in hepatic disturbances such as carcinoma and acute yellow atrophy, in uremia, in acid intoxication, in dyspnea from any cause, in the toxic vomiting of pregnancy, in delayed chloro- form poisoning, and especially in diabetes mellitus. In this latter condition the degree of acidosis may be conveniently followed by watching the ammonia output. A reduction in the amount of ammonia is observed in many cases of nephritis and in some cases of carcinoma of the stomach, although there is at the same time a diminished excretion of hydrochloric acid in the gastric contents. Edsall reports a reduction in cases of periodic insanity preceding the attack, while a rise is observed as the attack proceeds. Administration of large doses of the fixed alkalies will usually diminish the ammonia output.1 Quantitative Determination of Ammonia. Like the methods given under Urea, many have been advanced for the determination of the urinary ammonia. One of these, though inaccurate, has been so long used and even to-day is so relied upon in many quarters that the writer includes it with the understanding that he advocates only the ac- curate methods. If the ammonia is worth determining, definite results should be sought and hence the most accurate methods are the ones to be used. They are no more complicated, not as time-consuming, and give more reliable results. Method of Schlbsing. This method is the one most commonly used, but is open to the objection that it does not yield accurate results and is time-consuming, but has the advantage of simplicity. It is argued by many that the element of time is of no importance, as clinically one would not wait for an ammonia determination 'See Janney, Zeitsch. f. physiol. Chem., Bd. 76, 1912, S. 99. THE URINE. 217 before instituting vigorous treatment. On the other hand, in metabolic work it is of a great advantage to get the work out of the way as quickly as is possible and consistent with accurate results. Twenty-five c.c. of urine are placed in the vessel B (preferably a Petri dish) (see cut). Above this is placed a glass triangle upon which rests a dish (C) containing 20 c.c. of tenth-normal sulphuric acid. Twenty c.c. of milk of lime are then poured into the dish containing the urine and the whole covered with a'bell-jar, the borders of which have been well greased to make an air- tight union when the jar is placed upon the glass plate. This apparatus Technic. Fig. 75.-Schlosing's ammonia apparatus. is then allowed to stand at room temperature from four to five days, during which time the ammonia, liberated by the action of the milk of lime upon the ammonium salts of the urine, will be taken up by the sulphuric acid in the vessel C. At the end of this time the bell-jar is removed, the acid titrated with tenth-normal sodium hydrate, and the number of c.c. of remaining acid determined. One c.c. of tenth-normal sulphuric acid neutralized by the evolved ammonia represents 0.001704 gram of ammonia. This figure is multiplied by 4 to obtain the percentage ammonia value. If any moisture is present on the inside of the bell-jar it should be washed into the sulphuric acid before titration. This method, as previously stated, does not give accurate results owing to the fact that ammonia may be split off from urea and thus give figures which are somewhat high. It has been found that if the apparatus be kept at 38° C., the time necessary for this reaction may be reduced to 48 hours. If we add to the urine instead of the milk of lime, 0.5 gram of sodium carbonate 218 DIAGNOSTIC METHODS. and about 10 grams of sodium chlorid, no ammonia will be split off from the urea and no decomposition of the urine will occur (Schaffer). Folin's Method. The ammonia in this method1 is set free by the addition of a weak alkali (sodium carbonate), is then removed from the urine at ordinary room tem- perature by means of a strong air-current, is collected in tenth-normal sulphuric acid and then titrated. Technic. Twenty-five c.c. of urine are poured into an aerometer cylinder (30 to 40 cm. high), 15 grams of potassium oxalate,2 1 gram of dry Na2 CO3 and some crude petroleum (to prevent foaming) are added. The upper end of the cylinder is then closed by means of a doubly perforated rubber stopper, through which pass two glass tubes, only one of which is long enough to reach below the Fig. 76.-Folin's ammonia apparatus. (Hawk.) surface of the liquid.3 The shorter tube (about 10 cm.) in length is connected with a calcium chlorid tube filled with cotton, which in turn is connected with a glass tube extending to the bottom of a wide-mouthed bottle (capacity about 500 c.c.) which contains 20 c.c. of tenth-normal sulphuric acid, 200 c.c. of water and a few drops of an indicator (alizarin-red). The complete ab- sorption of the ammonia by the sulphuric acid is most easily insured by the use of a simple absorption tube which compels a very intimate contact of the air coming from the cylinder with the acid and water in the absorption bottle. This absorption bulb consists of a glass tube, measuring about 8 mm., in diameter, one extremity of which has been blown into a small bulb. By means of a heated platinum wire, 10 or 12 holes, each about 1 mm. in diameter, are made in this bulb. 'Zeitsch. f. physiol. Chem., Bd. 37, 1902, S. 161; Ibid., Bd. 39, 1903, S. 477. 2 Jour, of Biol. Chem., Bd. 8, 1910, p. 497. 3 SeeFolin and Macallum (Jour, of Biol. Chem., vol. 11,1912, p. 523) for a micro-chemical method. THE URINE. 219 The absorption bottle is then attached by means of a glass tube and rubber connection to a filter-pump which will draw 700 liters of air per hour. The air passing through the alkaline urine will draw the ammonia into the standard acid in from one and one-half to two hours. In order to exclude any error due to the presence of ammonia in the aspirated air, a similar absorp- tion apparatus is attached to the distal side of the evolution flask. The amount is then determined by titration of the standard acid with tenth-normal sodium hydrate, using alizarin- red as an indicator and titrating to the first red point instead of the violet color. Steel, Gies and others advise the use of NaOH instead of Na2CO3 in this method on the ground that the latter will not liberate NH3 from ammonium magnesium phosphate. This would not seem to militate against the method as the triple phosphate is rarely, if ever, present in the urine in appreciable amounts except when the urine is decomposed. Under such circumstances the ammonia factor would be useless from the metabolic standpoint. Formalin Method. This method, originated by Ronchese1 and Malfatti,2 de- pends on the fact that a solution of an ammonium salt, treated with formaldehyd, decomposes with the formation of hexa- methylen-tetramin, the acid combined with the ammonia being liberated. This can then be determined by titration. Dilute io c.c. of urine with 50 c.c. of water, add 2 or 3 drops of a 1 per cent, alcoholic solution of phenolphthalein and neutralize with N/10 NaOH. Five c.c. of formalin, previously neutralized with N/10 NaOH, are added to the neutralized urine and the mix- ture is again titrated with N/10 NaOH to the appearance of a faint, permanent pink color, the amount of alkali used being noted. It is evident that 1 c.c. of this N/10 NaOH is equivalent to 1 c.c. of N/10 NH3 or, in other words, repre- sents 0.001704 gram of NH3. Multiply the number of c.c. of N/10 NaOH used by this factor and obtain the NH3 in 10 c.c. of urine. It has been found that formalin combines, also, with the NH2 group of the amino acids, thus leading to results higher than those of the Folin method. If the figure obtained by the Folin method be subtracted from that of the formalin method, the result is the NH3 referable to amino acids. Fig. 77.- Folin's absorp- t io n bulb. (Hawk.) (d). Uric Acid, CSH4N4O3. In times past uric acid has been credited with much more clinical im- portance than is to-day ascribed to it. It is not, as many believe, a product of protein decomposition, as such, but can be derived only by the splitting of the nucleo-proteins. These nuclein bodies are compounds of protein with nucleic 1 Jour. Pharm. et China., Tm. 25, 1907, p. 611; Bull. Soc. China, de France, Tna. 1, 1907, p. 900. 2 Zeitsch. f. anal. Chem., Bd. 47, 1908, S. 273. 220 DIAGNOSTIC METHODS. acid, the latter constituent splitting up into thymic acid and derivatives of "purin," among which we find uric acid, xanthin, hypoxanthin, etc. The chemistry of these nuclein compounds is more or less complicated, so that the writer cannot take the space to present them in detail. Uric acid, as well as the purin bases to be discussed later, are derivatives of Fischer's hypothetical purin nucleus, with the formula C5H4N4 and the following graphic structure. The figures in brackets represent the number of the atom or group of atoms: Each of the purin derivatives is formed by the replacement of one or more of the hydrogen atoms in this nucleus by various atoms or groups of atoms. Thus we find that uric acid is a derivative in which 3 atoms of oxygen have been substituted for the hydrogens in positions 2, 6, and 8. The hydrogen atoms are not replaced, but shift in the direction of the double bonds, passing to the nitrogen atoms 1, 3, and 9. The structure of uric acid with the formula C5H4N4O3 is Its chemical name is, therefore, 2, 6, 8, tri-oxy purin. Uric acid is derived from the nucleins of the food, as well as from the nucleins of the tissues. We can conceive, consequently, that we may have a uric acid excretion which may be, in part, referable to each of these factors. This is the basis of the theory of endogenous and of exogenous purin metabolism, the former representing the uric acid formation from the tissue nucleins, the lat- ter that from the food directly. It is self-evident that the exogenous purin me- tabolism, being directly dependent upon the diet, will vary more than will the endogenous form. This latter type is dependent not only upon the metabolism of the various cellular elements of the body, but also upon the degree of direct synthesis of uric acid in the system, as well as upon the extent of conversion of uric acid into urea. This latter statement must not be inter- preted as meaning that uric acid is simply a stage in the conversion of all protein into urea. This idea, which held sway for so long, has fortunately THE URINE. 221 been abandoned. However, we do know, from the work of Frerichs and Woh- ler that the system does transform a certain amount of uric acid into urea. Burian and Schur point out that the uric acid eliminated by man on a purin-free diet (endogenous uric acid metabolism) is for each individual a constant quantity and entirely independent of the total amount of nitrogen eliminated. This fact has so firmly fixed itself in the minds of the profes- sion that the important results of Folin have been largely unnoticed. This latter worker finds that "when the total amount of protein metabolism is greatly reduced, the absolute quantity of uric acid is diminished, but not nearly in proportion to the diminution in the total nitrogen, and the per cent, of the uric acid nitrogen in terms of the total nitrogen is, therefore, much in- creased." It would seem, therefore, that this point is still very debatable. Quoting again from Folin "if the endogenous uric acid is to be considered as derived from the cell nucleins exclusively, it would, indeed, seem highly plaus- ible that the quantity should tend to remain constant, even with very great variations in diet. Rigid proof that the endogenous uric acid elimination is for each individual a constant quantity would be strong evidence in favor of such a theory. Burian and Schur support the view that the endogenous uric acid is derived from the cell nucleins, but they contend that in man about one-half of the uric acid so derived is destroyed inside the organism and that only the other half is eliminated. With the introduction of this important modification of the nuclein theory, there is no longer any reason why the uric acid elimination should not be a decidedly variable factor which might well be susceptible to change under the influence of many different changes in the conditions, among others, changes in diet." It will be seen, therefore, that one must be on his guard in drawing conclusions from the amount of uric acid in the urine, as to the degree of nuclein decomposition within the system. Physiologic Variations. The output of uric acid varies, depending upon the diet, from 0.2 to 2 grams in 24 hours. On a diet of 119 grams of protein with a total urinary nitrogen output of 16 grams the uric acid eliminated was 0.37 gram (0.8 per cent, of total N); while on a nitrogen-free diet with a urinary nitrogen of 3.6 grams the output of uric acid was 0.09 gram (2.5 per cent, of total N). It is increased physiologically by increase in the nucleins of the diet, sweet-breads, liver, kidneys, and brain yielding very large amounts of uric acid. A meat diet will lead to a larger excretion of uric acid than will a vegetable diet, the maximum output being observed about five hours after a hearty protein meal. According to Horbaczewski, this increase is dependent upon the leucolysis which occurs at the time of the disappearance of the digestive leucocytosis. The amount of exercise will also influence the output in the urine; while an intake of a large amount of water will increase the normal uric acid value. A certain relation between the amount of urea and of uric acid excreted seems to obtain. As a rule, it may be said that the nitrogen of the urea is to the nitro- 222 DIAGNOSTIC METHODS. gen of the uric acid as 50 or 60 to 1. It is probable that variations in this rela- tion have at present much less clinical value than in times past. It does not seem to the writer consistent to assume a special "uric acid diathesis" if this normal relation be disturbed in the sense that the uric acid is increased, as our knowledge of the factors controlling the excretion of both urea and uric acid is still more or less hazy. Pathologic Variations. A pathologic increase of uric acid is observed whenever we have increased protein catabolism. Thus, in fever, the output of uric acid runs parallel to that of urea. In cases associated with marked leucocytosis a large increase in the uric acid output may be noted, which is referable to the constant leucolysis. This increase is especially marked in leukemia, an output of more than 12 grams in 24 hours having been observed by Magnus Levy. In cases of pneumonia, associated with a high leucocyte count, an increase in the uric acid output is seen, being especially marked after the crisis, but may even precede the crisis. Gout has so long been associated, in the minds of the profession, with increased uric acid formation that such a relationship is generally accepted. It is true that during the acute attack the blood may contain an increased amount of uric acid, but never in such a large amount as in cases of leukemia, for instance. The mere excess of uric acid in the blood can, therefore, not be the determining factor. As our knowledge of the true etiology of gout is so ob- scure the writer will not attempt a discussion of uric acid in such relations. According to Futcher the uric acid is below the normal standard preceding the acute attack, rises to much increased values during the attack, and again falls below the normal limit after the subsidence of the acute symptoms. In acute articular rheumatism an increased elimination is noted during the febrile period, a decrease being observed as convalescence approaches. In diabetes mellitus an increase or a decrease may be observed, the uric acid excretion varying inversely as that of the sugar, giving rise to the term diabetes allernans. In cases showing much degeneration of hepatic tissue, as in cirrhosis and acute yellow atrophy, uric acid may be largely increased. A diminished excretion of uric acid is usually observed in the ordinary forms of anemia while in pernicious anemia an increase may be noted. Chronic interstitial nephritis, chronic lead-poisoning, purpura hemorrhagica, and some cases of epilepsy, are associated with a diminished output of uric acid. Large doses of quinin or of opium may diminish the uric acid, while salicylic acid, chinic acid, colchicin, urotropin, piperazin, nucleinic acid, etc., may increase the amount. / The chemical properties of uric acid may be found in works on physio- logic chemistry. Qualitative tests are frequently desirable for the detection of the nature of certain deposits, as, for instance, renal infarcts. These will be discussed in a later section in association with the description of the various crystalline forms which uric acid may assume in a urinary deposit. THE URINE. 223 Quantitative Determination. Folin's Method. This modification1 of Hopkins' method depends upon the precipitation of uric acid, as ammonium urate, by addition of ammonium sulphate. The urate is decomposed with sulphuric acid and the liberated uric acid is determined by titration with a standard solution of potassium permanganate. Two hundred c.c. of urine are treated with 50 c.c. of a reagent consisting of 500 grams of ammonium sulphate, 5 grams of uranium acetate, and 60 c.c. of 10 per cent, acetic acid, dissolved in 650 c.c. of water. The mixture is al- lowed to stand without stirring for about one-half hour. The precipitate of uranium phosphate has then settled and the clear supernatant liquid is re- moved by siphonage or by decantation. One hundred and twenty-five c.c. of this clear fluid, representing 100 c.c. of urine, are measured into a beaker, 5 c.c. of strong ammonia are added, and the mixture set aside until the following day. The precipitate of ammonium urate, which is produced by alkalinizing the solu- tion saturated with ammonium sulphate, is then filtered off, and washed with 10 per cent, ammonium sulphate solution until the filtrate is practically free from chlorids. The filter is removed from the funnel, opened, and the precipitate rinsed back into the beaker, enough water to make about 100 c.c. being used. The pre- cipitate is now dissolved by adding 15 c.c. of concentrated sulphuric acid. The solution is then titrated with a twentieth-normal solution of potassium perman- ganate, until the first pink coloration is observed extending through the entire liquid after the addition of two drops of the permanganate solution. Each c.c. of potassium permanganate used corresponds to 3.75 mg. of uric acid. Owing to the solubility of the ammonium urate, a correction of 3 mg. per 100 c.c. of urine must be made. The corrected result gives the percentage of uric acid.2 A method of making a twentieth-normal potassium permanganate solution may be found in any work in general chemistry. The student is to be cautioned that a twentieth-normal solution of potassium permanganate, as used in this connection, has reference to one of such a concentration that one liter would contain 0.05 gram of available oxgyen for oxidizing purposes. This solution is obtained by dissolving 1.581 grams of pure KMnO4 in one liter of water. As this weighing may not be sufficient, owing to slight impurities in the permanganate, it is wise to titrate this solution against a known twentieth- normal solution of oxalic acid. If the solutions correspond, 1 c.c. of each should be the equivalents. The titration of the oxalic solution is made by meas- uring out 10 c.c. of oxalic acid solution, diluting to approximately 100 c.c. with distilled water, and adding 15 c.c. of concentrated sulphuric acid. The per- manganate solution is then added drop by drop until a faint permanent red color is observed throughout the liquid, which does not disappear on stirring but persists for at least 3c seconds. If these solutions do not correspond the per- Technic. 1Zeitsch. f. physiol. Chem., Bd. 32, 1901, S. 552. 2 Skorczewski und Sohn (Wien. klin. Wochs., Bd. 24, 1911, S. 1700) show that the urine of patients taking atophan may reduce the permanganate and, hence, introduce an error. 224 DIAGNOSTIC METHODS. manganate solution must be diluted according to the formula previously given under the determination of the chlorids in the urine. Salkowski-Ludwig Method. This method is not as convenient as that of Folin, but is inserted as it serves as a combined method of determining uric acid and the purin bases at the same time. The principle is as follows: Uric acid is precipitated, in the form of a double urate of silver and magnesium, by an am-moniacal silver solution in the presence of magnesium salts. The silver is removed by hydrogen sulphid and the uric acid precipitated by hydrochloric acid, after which it may be es- timated by direct weighing or by determination of the nitrogen. Two hundred and forty c.c. of urine are treated with 60 c.c. of magnesium mixture, which is made up as follows: One hundred grams of crystallized magnesium chlorid and 200 grams of ammonium chlorid are dissolved in about 500 c.c. of water, ammonia is added until the mixture smells strongly of this substance and the whole is then made up to one liter. The above mixture is well stirred and immediately filtered through a dry filter-paper into a beaker. Two hundred and fifty c.c. of filtrate (representing 200 c.c. of urine) are then measured off and treated with 10 to 15 c.c. of an ammoniacal silver nitrate solution (one liter of which contains 26 grams of AgNO3 and enough ammonia to dissolve the precipitate formed). The precipitate should be of a flocculent gelatinous nature and of a yellowish color. If the precipitate be white, too much silver chlorid is present and more ammonia must be added. The clear solution above the precipitate should contain an excess of silver chlorid which may be shown by adding a little nitric acid to a few drops of the clear super- natant liquid. This mixture is then filtered, any particles adhering to the beaker being transferred to the filter by means of water and a rubber-tipped glass rod. The residue on the filter is washed with distilled water until the wash fluid shows no trace of silver or of chlorids. The funnel is now placed in the neck of a liter flask, the tip of the filter-paper is perforated with a glass rod, and the precipitate washed into the flask and thoroughly mixed with the water. This solution is made faintly acid by the addition of two or three drops of hydrochloric acid. Three to four c.c. of 10 per cent, copper sulphate solution are then added and the mixture boiled, after which hydrogen sulphid is passed through the hot mixture to precipitate the silver salts. This gas should be passed until the solution is saturated with it, after which the solution is boiled and filtered. The precipitate is washed with hot water and the filtrate, which must be clear and colorless, is evaporated to a small bulk (10 to 15 c.c.). Ten to fifteen drops of hydrochloric acid are then added, the mixture stirred and allowed to stand for a few hours, preferably over night. This addition of hydrochloric acid precipitates the uric acid and leaves the purin bases in so- lution. The crystals of uric acid are filtered off on a small weighed filter, are washed with water slightly acidified with HC1 to such an extent that the total wash-water and filtrate should not be more than 50 to 60 c.c. The precipitate Technic. THE URINE. 225 is washed with alcohol, carbon disulphid, and ether, and is then dried and weighed. The difference between the original weight of the filter-paper and that of the paper and precipitate is the amount of uric acid in the 200 c.c. of urine. Owing to the slight solubility of uric acid in acidulated water, a correction of 0.00048 gram must be added for every 10 c.c. of the filtrate and wash-water. Instead of weighing the uric acid, the filter-paper and contents may be placed in a Kjeldahl flask and a nitrogen determination made as previously described. The nitrogen value multiplied by 3 will give the weight of uric acid in 200 c.c. of urine. The filtrate and wash-water from the uric acid precipitation contains the purin bases. This filtrate is alkalinized with ammonia and again precipitated with the ammoniacal silver solution. This precipitate is collected on a small filter, washed with water, dried and carefully incinerated. The ash is dis- solved in nitric acid and the silver chlorid estimated by titration with potassium sulphocyanate as described under the determination of chlorids. One c.c. of the potassium sulphocyanate solution is equivalent to 0.00734 gram of silver. One part of silver, in the form of the silver compounds of the purin bases, represents 0.277 gram of nitrogen, or 0.7381 gram of the purin bases. It is evident, therefore, that 1 c.c. of the potassium sulphocyanate solution will represent 0.002 gram of nitrogen and 0.00542 gram of purin bases accord- ing to the following proportion: i : 0.00734 :: 0.277 : x x = 0.002 By multiplying the number of c.c. of potassium sulphocyanate solution used to precipitate the chlorids by 0.00542 we obtain the number of grams of purin bases in the 200 c.c. of urine originally used. It is wise in determining the amount of purin bases to start with a larger quantity of urine, as, for instance, 600 to 700 c.c., as the amount present in 200 c.c. would be very small. This method is apt to give slightly high values for uric acid, as the purin bases may not be entirely soluble in the acid solution used to separate them from the uric acid. It requires much more equipment than is usually at the disposal of the practitioner so that it can hardly be recommended for routine use in general work. Method of Rudisch and Kleeberg. This method1 is quite as accurate as is the preceding and has the advantage that it can be carried out in from 20 to 30 minutes. The principle is as follows: The total purin bodies are precipitated by an excess of silver nitrate, and the excess of silver determined volumetrically by potassium iodid, using a mixture of nitrous and sulphuric acid with starch solution as an indicator. As the silver compounds of the purin bases are soluble in strong ammonia solution, it is possible to make a determination of the uric acid and then estimate the purin bases by subtraction of this value from the total purin compounds. Technic. One hundred and ten c.c. of urine are treated with 55 c.c. of fiftieth-normal AgNO3 solution and diluted with strong ammonia to 220 c.c. (The fiftieth- 1Amer. Jour, of Med. Sci., vol. 128, 1904, p. 899. 226 DIAGNOSTIC METHODS. normal solution of silver nitrate is made by dissolving 3.3932 grams of AgNOs, which has been heated for 10 minutes with a small amount of water to 1200 C., in 500 c.c. of water, adding 75 c.c. of strong ammonia and 10 grams of ammo- nium chlorid and making the whole up to one liter.) The addition of the ammonia dissolves the purin bases, leaving the uric acid in the precipitate. The original mixture is now filtered in such a way that two 100 c.c. portions are obtained, each portion representing 50 c.c. of urine. With the first of these portions an approximate estimation of the uric acid is made, while with the second the accurate one is carried out. To one of these portions, fiftieth-normal potassium iodid solution is added from a buret, a few drops being removed from time to time and added to a solution of nitrous-sulphuric acid mixed with a little starch paste. (The fiftieth-normal potassium iodid solution should contain 3.32 grams of KI in one liter. The nitrous-sulphuric acid mixture is made by mixing 25 c.c. of H2SO4 w'ith 75 c.c. of H2O and 1 c.c. of fuming HNO3.) The addition of the drops of the solution to the indicator is for the purpose of determining the point at which an excess of potassium iodid occurs. This will be shown by the appearance of a distinctly blue contact ring. When the solution shows such a reaction, on adding a few drops of the mixture to the indicator, the number of c.c. of KI is read off, and we are then prepared for the more accurate control. It is wise to keep the indicator cold by immersion in ice-water, as otherwise a violent reaction may occur. The second 100 c.c. portion is then carefully titrated with the fiftieth- normal KI solution by running directly about 1 c.c. less of this solution into the urine than was required in the first titration. After this point is reached the KI solution should be added drop by drop and a test made after the addi- tion of each five drops. In this way an accurate end-point may be reached. As each 100 c.c. portion represents 50 c.c. of urine it will contain 25 c.c. of fiftieth-normal AgNO8 solution. Subtract from this 25 c.c. the number of c.c. of fiftieth-normal KI used and the number of c.c. of silver nitrate which com- bined with the uric acid of the urine is obtained. One c.c. of fiftieth-normal AgNO3 solution represents 0.00336 gram of uric acid. By multiplying this factor by the number of c.c. of silver nitrate used, we obtain the amount of uric acid in 50 c.c. of urine. The second part of this method consists in the determination of the total purins. One hundred and ten c.c. of urine are treated with 55 c.c. of fiftieth- normal AgNO3 solution and diluted with water to 220 c.c. The remainder of the process is exactly as outlined above. The number of c.c. of fiftieth-normal silver nitrate used will represent the values for the total purins. If the number of c.c. of silver nitrate solution used in the previous deter- mination of uric acid be subtracted from the number used in the determination of total purin, the result will be the number of c.c. referable to the purin bases. One c.c. of fiftieth-normal AgNO3 solution represents 0.00152 gram of purin bases calculated as xanthin. By multiplying this factor by the number of c.c. of AgNO3 used we obtain the amount of xanthin in 50 c.c. of urine. THE URINE. 227 1 he writer has used this method frequently and has tound its results very similar to those of the Salkowski-Ludwig method, although the amount of uric acid is somewhat less in this method than in the latter one. It is very simple, is quick and accurate and may, therefore, be recommended for general work. A somewhat similar method is advocated by Bartley1. The writer finds this latter method fairly reliable and simple. Ruhemann's Method. This method is a very convenient clinical one, although its results are by no means as accurate as those of the preceding methods.2 What the general practitioner desires, as a rule, is to know whether the uric acid is increased or diminished and does not care as to the absolute value. Such results, giving the total purins, may be obtained for clinical purposes by this method. It consists in the use of a specially graduated tube, the uricometer, in which are placed the reagents and the urine to be tested (see cut). The calibrations of the tube are such as to represent directly the amount of uric acid in parts per 1000. The principle of the method is the decolorization of an iodin solution by the uric acid of the urine, and the measurement of the amount of urine which must be added to a definite amount of iodin solution to effect this decolorization. Carbon disulphid is placed in the tube up to the mark S, in such a way that the lower meniscus of this reagent rests upon the mark. A solution of iodin in potassium iodid is then added so that the upper portion of the meniscus coincides with the mark J. This iodin solution has the following composition. Technic. lodin, 0.50 gram Potassium iodid, 1.25 grams Absolute alcohol, 7.50 grams Glycerin, 5.00 grams Distilled water, q.s., 100.00 grams The urine is added slowly by means of a pipet until the lowest calibration is reached. The glass stopper is inserted and the contents of the tube mixed by repeated inversion for about 15 seconds. The carbon disulphid absorbs the iodin, taking on a distinct purple coloration. If this amount of urine does not completely decolorize the iodin, shown by the porcelain-like color of the carbon disulphid solution, more urine is added and the tube again inverted for 15 seconds. This process is continued until repeated Fig. 78.- Ruhemann's uricometer. 1 Medical Chemistry, Philadelphia, 1904. 2 See Bradley and Bunta, Jour. A. M. A., vol. 60, 1913, p. 44. 228 DIAGNOSTIC METHODS. shaking of the tube causes the carbon disulphid to assume a pale pink color. The reaction is practically ended at this point, as by a little more shaking of the contents the indicator will assume the characteristic porcelain-white appear- ance. This process requires from 6 to 15 minutes. The amount of uric acid is then read off directly from the tube in parts per liter. Should the urine contain less uric acid than can be read off from the calibrations, a second test is made adding the iodin solution to the mark midway between S and J, the amount indicated on the tube being of course divided by 2. Conversely, should the urine contain more uric acid than is represented by the lower calibration, one adds the iodin solution to the point above J and multiplies his reading by 1.5; or adds the iodin solution to the second mark above J and multiplies the reading by 2. With this method the urine must be acid in reaction. If the urine contains a sediment of the urates, it should be thoroughly shaken before being added, so that the urates may be in suspension. Any free uric acid which may have sepa- rated in the sediment is not determined in this method. Strongly colored urines have no influence upon the decolorization. The presence of sugar does not interfere with the results, but if albumin be present in large amounts it should be removed by acidifying with dilute acetic acid, boiling, and filtering. The writer has used this method very frequently and finds it very useful for rough estimation of the uric acid outputs when the absolute amount is not of much clinical moment. The purinometer, introduced by Hall,1 seems to be a much more reliable and useful instrument, as it employs the reagents required by the Salkowski- Ludwig method.2 (e). Purin Bases. These bodies have been called purin bases, alloxur bases, xanthin bases, and nuclein bases. They are of extreme importance from the standpoint of physiologic chemistry, but are, clinically, of less value than is uric acid. The following bodies have been isolated from the urine in various conditions: adenin, guanin, epiguanin, carnin, episarkin, xanthin, hypoxanthin, heter- oxanthin, paraxanthin, and methylxanthin. Certain methylated xanthin compounds are found in tea and coffee and are, therefore, introduced into the system as caffein, theobromin, and theophyllin, being excreted either as xanthin or hypoxanthin. Xanthin (CSH4N4O2) is 2, 6, dioxypurin. These nuclein bases of the urine arise either from the nuclein of the food or from the increased nuclein metabolism of the system. Very little is known at the present time regarding the absolute variations in the excretion of these purin bases in the urine. Salkowski finds an excretion ranging between 0.0286 1 The Purin Bodies, Phila., 1904. 2 Folin and Macallum (Jour, of Biol. Chem., vol. 13, 1912, p. 363) have devised a new method for uric acid based upon the action of phosphotungstic acid in alkaline sloution upon uric acid. The depth of color of the blue solution formed is then compared with that of a standard solution. Their "uric acid reagent" is prepared by boiling 100 grams of sodium tungstate with 80 c.c. of 85 per cent, phosphoric acid and 750 c.c. of water for a couple of hours and then diluting to 1 liter. See, also, Folin and Denis, Jour, of Biol. Chem., vol. 14, 1913, P- 95* THE URINE. 229 and 0.0561 gram (calculated as xanthin), while Camerer regards an average output as 0.087 gram in 24 hours. It is interesting to note that a vegetable diet appears to increase this output more than a meat diet, 0.044 gram being excreted, according to Camerer, on a meat diet and o.m gram on a vegetable regime. This finding is the reverse of that for uric acid. As a general rule, it may be said that the output of purin bases is from 8 to 10 per cent, of that of uric acid, varying from 16 to 60 mg. per diem. It is evident that foods containing these substances should be absolutely interdicted in conditions which may be traceable to disturbances in the nuclein metabolism. The pathologic variations in the excretion of these bodies are little under- stood. An increase in the amount of uric acid is usually associated with an increase of the xanthin bases, but at times no such relations obtain, a de- crease in these bodies being observed. In leukemia an excretion of 0.321 gram has been reported by Magnus-Levy1. In certain cases of tuberculosis, nephritis, epilepsy, migraine, and pneumonia the output may be much increased. Edsall finds the urinary output increased as a result of X-ray treatment. Xanthin is occasionally found as a constituent of the urinary sediment and may form calculi. The tests for the recognition of such excretions will be discussed in a later section. The quantitative estimation of these bodies may be made by the tests pre- viously outlined under Uric Acid. There can be little doubt that careful study of the urine with reference to these purin bases, using exact methods of analysis, will reveal much valuable information.2 The older ideas regarding the excretion of this nitrogenous body have suffered a severe shock from the work principally of Folin, Shaffer, Hoogen- huyze and Verploegh, Mellanby, and Klercker. Acid urines were supposed to contain creatinin and little or no creatin; while alkaline urines were thought to show creatin instead of creatinin. It has been proven, however, that "normal fresh urine, whether acid or alkaline, contains creatinin, and if the normal subject has not taken creatin in his food during the preceding days his urine will not contain creatin, whatever its reaction. There is no normal excretion of endogenous creatin,3 as this, when ingested, is largely retained in the body unless the food contains a large amount of protein." Thus Folin has shown that the excretion of creatinin on a diet yielding a urinary nitrogen value of 16 grams, was 1.55 grams or 3.6 per cent, of total nitrogen; while on a nitrogen- free diet with a urinary nitrogen of 3.6 grams, the creatinin output was 0.6 gram or 17.2 per cent, of total N. He showed, further, that while the actual amount excreted varied with different individuals, yet for the same person the output was practically constant, under the same conditions of health and mus- cular activity. He believes, therefore, that creatinin is by far the most re- (/). Creatinin (C4H7N3O). 2 See Hefter, Deutsch. Arch. f. klin. Med., Bd. 109, 1913, S. 322. 3 This statement does not apply to growing children who show a normal excretion of a relatively large amount of creatin. See Mendel and Rose, Jour, of Biol. Chern., vol. 10, 1911, p. 265; also Folin and Denis, Ibid., vol. n, 1912, p. 253. 1 Loc. cit. 230 DIAGNOSTIC METHODS. liable index as to the amount of a certain kind of protein metabolism occurring daily in any given individual. He bases his conclusions upon the facts that "the absolute quantity of creatinin eliminated in the urine on a meat-free diet is a constant quantity, differing for different individuals, but wholly independent of quantitative changes in the total amount of nitrogen eliminated." Shaffer1 believes that creatinin is not an index of the total endogenous protein catabolism, as patients in whom the endogenous catabo- lism is much increased may excrete very little creatinin. Folin finds that the chief factor determining the amount of creatinin eliminated on any special diet is the weight of the patient. Fat or corpulent persons yield less creatinin per unit of body-weight than do lean ones. It is, therefore, necessary in meta- bolic work to consider not only the body-weight but, also, the excess of fat in increasing the weight. According to Shaffer, the normal excretion of creatinin varies between 7 and u mg. of creatinin-nitrogen per kilo of body-weight. In pathologic subjects it is low, varying from the normal to 2 mg. per kilo of body-weight in 24 hours. He calls the creatinin-nitrogen excretion per kilo of weight the "creatinin coefficient." He thinks that creatinin is an index of some special process of normal metabolism taking place largely, if not wholly, in the muscles. Upon the intensity of this process appears to depend the muscular efficiency of the individual. In acute febrile conditions, in which an increased destruc- tion of muscle tissue occurs, an increase is seen in the creatinin output during the active febrile period, while in the period of convalescence a diminished excretion will be observed. This excretion in fever does not run parallel to the muscular efficiency of the individual (Shaffer). Simon has shown that a dimin- ished excretion occurs in anemia, marasmus, myositis ossificans, chlorosis, phthisis, chronic parenchymatous nephritis, progressive muscular atrophy, and pseudohypertrophic paralysis. According to Shaffer, creatin may be excreted by subjects of acute fevers, in the acute stages of exophthalmic goiter, in other conditions in which there is a rapid loss of muscle protein, and by women during the postpartum reso- lution of the uterus. Krause2 has reported the constant presence of creatin in diabetes mellitus referable, probably, to a deprivation of carbohydrates. Although administration of carbohydrate usually diminishes the creatin excre- tion, Folin's theory would indicate that an increased excretion of creatin may follow a carbohydrate-rich diet, if the state of nutrition of the body is high, as a greater fraction of the creatin of the food would be eliminated. See McCrudden, Jour, of Exp. Med., vol. 15, 1912^.457; also Sedgwick, Amer. Jour, of Dis. of Child., vol. 3, 1912, p. 209, who finds an increased excretion of creatin in cases of recurrent vomiting, while Ellis3 presents such a finding in a case of acromegaly. For an excellent survey of this field see Mendel4 as well as Ditman and Welker.5 1 Amer. Jour, of Physiol., vol. 23, 1908, p. 1. 2 Quart. Jour. Exp. Physiol., vol. 3, 1910, p. 289. 3 Jour. A. M. A., vol. 56, 1911, p. 1870. 4 Science, vol. 29, 1909, p. 584. N. Y. Med. Jour., vol. 89, 1909, pp. 1000, 1046 and 1091. THE URINE. 231 Creatinin has the formula, C4H7N3O, with the graphic structure Qualitative Tests for Creatinin. The chemical reactions which serve for the detection of creatinin depend upon the formation of different colored compounds. It forms a distinct crys- talline compound with zinc chlorid, which may be used in the quantitative estimation by Salkowski's method. For this process see works on physio- logic chemistry. A point to be remembered regarding creatinin is that it re- duces copper solutions and may be mistaken for sugar unless care be exercised. Weyl's Test. To the urine to be tested are added a few drops of a very dilute aqueous so- lution of freshly dissolved sodium nitroprussid and a few drops of dilute sodium hydrate solution. In the presence of creatinin a ruby-red color appears which changes, after a short time, to an intense yellow. If this solution be heated with a little glacial acetic acid the color will change to green and finally blue. Acetone gives a similar reaction, but on the addition of acetic acid changes to a purplish-red instead of green. If the uri ne be heated previous to the application of this test, the acetone may be driven off. This test for creatinin is sensitive to 6 parts in 10,goo. Jaffe's Test. To the urine to be tested are added a few drops of a saturated solution of picric acid and a few drops of dilute sodium hydrate solution. If creatinin be present a red color appears immediately, which increases in intensity and remains permanent for a long time. If glacial acetic acid be added the color becomes yellow. Acetone gives a reddish-yellow color of less intensity than that produced by creatinin. Glucose, if present, may give a red color if the mixture be warmed. This test indicates one part of creatinin in 5,000. Quantitative Determination. Folin's Method. The principle upon which this determination1 is based is the comparison of the color produced by Jaffe's reaction with that of a standard solution of potassium bichromate. A high-grade colorimeter is necessary for this compari- son. Folin recommends the use of the Duboscq instrument, while the writer finds one made by Sargent & Co., of Chicago, very satisfactory. This latter has the advantage of being much less expensive. The reagents necessary are (i) a half-normal potassium bichromate solu- tion containing 24.55 grams per liter, (2) a saturated picric acid solution con- taining about 12 grams per liter, and (3) a 10 per cent, solution of sodium hydrate. 'Amer. Jour, of Physiol., vol. 13, 1905, p. 45. 232 DIAGNOSTIC METHODS. Technic. Ten c.c. of urine are measured into a 500 c.c. volumetric flask, 15 c.c. of the picric acid solution and 5 c.c. of the sodium hydrate solution are then added, and the mixture allowed to stand for five or six minutes. This interval is used to pour a little of the standard bichromate solution into each of the two cylinders of the colorimeter. The depth of the solution in one of the cylinders is then accurately adjusted to the 8 mm. mark. With the solution in the other cylin- der a few preliminary colorimetric readings are made simply for the sake of insuring greater accuracy in the subsequent readings of the unknown solution. The two bichromate solutions must, of course, be equal in color, and in taking :heir readings no two should differ nore than 0.1 mm. or 0.2 mm. from 'he true value (8 mm.), leaving out of .onsideration the very first reading made, which is sometimes less ac- curate. Four or more readings should be made in each case, and an average taken of all but the first. After a while one becomes sure of the true point, and can take the average of the first two readings. At the end of five minutes the con- tents in the 500 c.c. flask are diluted up to the 500 c.c. mark. The bichromate solution is thoroughly rinsed out of one of the cylinders by means of the unknown solution and several colorimetric readings are then made at once. The calculation of the results is very simple. It is based on the experimentally determined fact that 10 mg. of perfectly pure creatinin give, under the conditions of the de- termination, 500 c.c. of a solution, 8.1 mm. of which have exactly the same c olorimetric value as 8 mm. of a half- normal bichromate solution. If, for example, it is found that it takes 9.5 mm. of the unknown urine- picrate solution to equal the 8 mm. of the bichromate, then the 10 c.c. of urine contain Fig. 79.-Sargent's colorimeter. (Courtesy of E. H Sargent & Co.) 8.i io x - = 8.4 +mg. of creatimn 9-5 If the io c.c. of urine used in the test are found to contain more than 15 mg. or less than 5 mg. of creatinin, the determination should be repeated with THE URINE. 233 a correspondingly different amount of urine, because outside of these limits the determination is much less accurate. The color of the urine does not materially affect the results on account of the great dilution. Sugar and albumin do not interfere with the determination, while acetone and diacetic acid do unless one allows a few minutes before reading.1 The technic2 for the estimation of creatin is as follows: 10 c.c. of urine are heated in the autoclave at a temperature of 1170 to 1200 C. for 30 minutes with 20 c.c. of 3 per cent, phosphoric acid to convert the creatin into creatinin. Cool the mixture and neutralize. Apply the technic above outlined when one obtains both the preformed creatinin and that derived from creatin. Subtract the preformed (as determined by the direct Folin method) from the total crea- tinin and obtain the creatinin derived from creatin. Multiply this latter figure by 1.16 to determine the actual amount of creatin as such. If one uses HC1, as in the Benedict-Myers method, in the presence of sugar, dark brown compounds are formed which interfere with the later colorimetric readings. The above method of Rose is equally accurate in diabetic and non- diabetic urines. By the undetermined nitrogen of the urine is meant the nitrogen remaining after that attributable to urea, uric acid, xanthin bases, ammonia, and creatinin has been subtracted from the total nitrogenous output. This factor is made up of many substances present in variable amount and determined with more or less difficulty. In this nitrogenous fraction we find various mon-amino and di-amino acids, hippuric acid, oxy-proteic and alloxy-proteic acids, and allan- toin. Physiologic variations in these separate factors are not at present well understood and seem to depend upon the age of the individual, the diet, and the condition of the intestines and liver. As the variations in the other nitrogenous constituents have been shown to depend to a great extent upon the nature of the diet, so we would expect the values for the undetermined nitrogen to show similar fluctuations. Folin finds on a diet yielding 16 grams of total nitrogen in the urine an undeter- mined nitrogen output of 0.6 gram or 3.75 per cent, of total nitrogen; while on a nitrogen-free diet with a urinary nitrogen value of 3.6 grams the undeter- mined nitrogen is 0.27 gram or 7.3 per cent. He is led, therefore, to make the following generalization: "The absolute quantity of undetermined nitrogen decreases under the influence of the starch and cream diet, but in per cent, of the total nitrogen there is always an increase." The separation of the various factors included in this undetermined nitrogen has clinically little value at the present time, as our knowledge of the importance of any or all of these constituents is extremely limited. (g). Undetermined Nitrogen. (1). Amino Acids. Theoretically, we should find in the urine, whenever hepatic metabolism is disturbed, both mono- and di-amino acids, as these substances are normal 'See Greenwald, Jour, of Biol. Chem., vol. 14, 1913, p. 87. 2 Rose, Jour, of Biol. Chem., vol. 12, 1912, p. 73. 234 DIAGNOSTIC METHODS. products of protein hydrolysis in the intestine and are directly converted into urea in the normal liver. It has long been known that acute yellow atrophy and phosphorus poisoning were associated with a diminished urea output and the presence of the two mono-amino acids, leucin and tyrosin, in the urine. Recent work1 with improved methods has shown that the urine contains these and other amino acids in any condition in which hepatic activity is impaired, so that our old-time diagnostic point of acute yellow atrophy must fall by the wayside. Although these substances are found in much larger quantities under pathologic conditions, traces of them are present in normal urine, especi- ally when the nitrogen intake has been large. Thus von Noorden regards glycocoll (amino-acetic acid) as a normal urinary constituent, its excretion averaging i gram daily. These acids are also found in cases of gout, pneumonia, especially during the absorption of the exudate, in diabetes, and in leukemia. A large number of these mon-amino acids have been isolated from the direct products of protein hydrolysis, but the number found in the urine has not been as large owing to the uncertain methods of examination. In this con- nection it is to be stated that the proportions of the various mon-amino acids will vary, depending upon the kind and type of protein subjected to hydrolysis or digestion. The introduction by Fischer2 of the esterification method has added much to our knowledge of these bodies. The writer cannot take space to give this method in detail and will, therefore, refer to works on physiologic chemistry for this phase of the subject, as well as for the characteristic crys- talline compounds formed by treatment of these mon-amino acids with /3-naphthalin-sulpho-chlorid. By the di-amino acids we mean the bodies lysin,3 arginin and histidin, which are collectively known as the hexone bases. These bodies, together with the mon-amino acids, constitute the end-products of the normal tryptic as well as the abnormal proteolytic digestion. An approximate method of determining these bodies has been given under the heading of the "Formalin Method" as applied to ammonia in urine (p. 219). This is not absolutely accurate but is clinically of value for comparative purposes. This acid is a normal constituent of urine and varies between 0.1 and 1 gram in 24 hours. It is derived to a large extent from foods containing benzoic acid, such as prunes, cranberries, green-gages, bilberries and many other fruits, and is also formed from the metabolism of tissue and food protein. This substance is directly synthesized in the system by the combination of benzoic acid with glycocoll, but the exact place of synthesis is uncertain. A pathologic increase in the excretion of hippuric acid has been observed in acute febrile diseases, marked intestinal putrefaction, hepatic disturbances, and in diabetes mellitus; while in cases of acute diffuse and chronic parenchymatous nephritis (2). Hippuric Acid (C9H9NO3). 1 See Signorelli, Biochem. Zeitsch., Bd. 39, 1912, S. 36; Ibid., Bd. 47, 1912, S. 482; also Galambos and Tausz, Zeitsch. f. klin. Med., Bd. 77, 1913, S. 14. 2 Untersuchungen uber Aminosauren. Berlin, 1906. 3 See Ackermann and Kutscher (Zeitsch. f. Biol., Bd. 57, 1911, S. 355) who report the presence of lysin in a patient with cystinuria. THE URINE. 235 as well as in amyloid kidneys hippuric acid is practically absent from the urine. For methods see larger works.1 (3). Oxyproteic and Alloxyproteic Acids. These acids have been isolated from the urine and seem to be constant constituents, derived from protein catabolism. Both of them contain sulphur and have been credited with forming a large percentage of the neutral sulphur output as well as contributing 2 to 3 percent, of the total N excretion.2 Bondzyn- ski believes the oxyproteic acid accountable for Ehrlich's diazo-reaction (page 322). Salkowski3 has shown that a certain amount (3 to 4 per cent.) of the urinary nitrogen is precipitable by absolute alcohol and is non-dialysable. This is the so-called "Colloidal N" and is largely referable to oxyproteic acid and N-containing carbohydrates. As this factor is increased in many cases of carcinoma up to 9 or 10 per cent, of total N, he believes this may be of value in diagnosis. However, Mancini,4 Marriott and Wolf,5 Caforio,6 and Kojo7 have demonstrated that this increase does not always obtain in cases of carcin- oma and is found in pregnancy, some hepatic disorders and in pulmonary tuber- culosis. Salomon and Saxl8 have outlined a method for the determination of the sulphur factor of this oxyproteic acid as related to total N and believe that an increase in this factor is indicative of carcinoma. Both of these methods are, as yet, too complicated for clinical work and yield uncertain results. (See Neuberg.9) (4). Allantoin (C4H6N4O3) This substance is a product of oxidation of uric acid. It is somewhat variable in amount, usually being found only in traces, but after a large intake of meat especially rich in nuclein the amount may be quite perceptible. It is found in fairly large quantities in the urine of the new-born, in the amniotic fluid, in ascitic fluid, in traces in the urine of pregnancy, and in leukemic blood. Pathologic variations have been little studied. The best method for the determination of this substance is that of Wiechow- ski.10 A point to be remembered in connection with allantoin is that it may be present in the urine in sufficient quantities to reduce copper solutions and must, therefore, not be mistaken for sugar. 1 Dakin, Jour, of Biol. Chern., vol. 7, 1910, p. 103; Steenbock, Ibid., vol. 11, 1912, p. 201; Folin and Flanders, Ibid., p. 257; Hryntschak, Biochem. Zeitsch., Bd. 43, 1912, S. 315. 2 See Browinski and Dombrowski, Zeitsch. f. physiol. Chem., Bd. 77, 1912, S. 92; Doman- sky, Monatsschr. f. Psych, u. Neurol., Bd. 31, 1912, S. 53; Erben, Prager med. Wochs., Bd. 37, 1912, S. 427. 3 Berl. klin. Woch., Bd. 42, 1905, S. 1581 and 1618; Ibid., Bd. 47, 1910, S. 533, 1746 and 2297. 4 Deutsch. Arch. f. klin. Med., Bd. 103, 1911, S. 288. 5 Amer. Jour, of Med. Sci., vol. 133, 1907, p. 404. 6 Berl. klin. Woch., Bd. 48, 1911, S. 1843. 7 Zeitsch. f. physiol. Chem., Bd. 73, 1911, S. 416. 8 Wien, klin. Woch., Bd., 24, 1911, S. 449; Riforma Med., Tm. 27, 1911, p. 421. 9 Der Harn. Berlin, 1911, p. 788; also Semenow, Folia Urol., Bd. 7, 1912, S. 215. 10 Beitr. z. chem. Physiol, u. Path., Bd. 11, 1908, S. 129; Biochem. Zeitsch., Bd. 19, 1909, S. 378; Ibid., Bd. 25, 1910, S. 431. 236 DIAGNOSTIC METHODS. (2). Fatty Acids. Traces of volatile fatty acids are present in all normal urines. The most important of these are formic, acetic, propionic, and butyric acids. They are doubtless formed in the intestinal tract by bacterial action upon the carbo- hydrates of the food, and may, therefore, be to some extent an index of the degree of carbohydrate fermentation. Their normal amount does not much exceed o.oi gram in 24 hours, although Blumenthal1 gives the figures under an ordinary diet as being equivalent to from 50 to 80 c.c. of tenth-normal sul- phuric acid (0.25 to 0.39 gram). These acids are increased in febrile conditions, the amount running parallel to the rise in temperature. This increase in the fatty acids of the urine is known as lipaciduria. According to Rosenfeld, this increase in febrile states is observed only in those cases in which absorption of decomposing al- buminous material occurs, as in all suppurative processes within the system. In the convalescent stage of pneumonia these acids may be excreted in increased amount, while a diminished output is usually observed preceding the crisis. In hyperacidity of the gastric contents the fatty acids of the urine are increased, while in hypoacidity they are diminished. In contradistinction to the increase of fatty acids in cases of fever associated with suppurative conditions, we find a diminution in the amount of fatty acid in scarlet-fever, erysipelas, measles, and diphtheria. In cases of acute rheumatism formic acid is said to be excreted in large amounts. In cyclic vomiting lactic acid2 may be increased. A simple method of determining the fatty acids of the urine is to acidify from 250 to 500 c.c. of urine with 50 to 75 c.c. of dilute sulphuric acid and distill. The distillate is then titrated with tenth-normal sodium hydrate solution using phenol-phthalein as an indicator. The results are expressed in terms of the corresponding number of c.c. of tenth-normal acid. (3). Oxalic Acid. The amount of oxalic acid eliminated in the urine in 24 hours varies from 10 to 20 milligrams. A portion of this excretion is undoubtedly derived from the diet, but some of it is produced in the metabolism of the tissues, especially of those containing nuclein substances. Although the carbohydrates do not take part in the production of oxalic acid normally, we find, in conditions asso- ciated with fermentation of the carbohydrates in the stomach, a large increase in the output of oxalic acid. As oxalic acid is more or less readily formed by oxidation of uric acid, passing through the intermediate stage of oxaluric acid, we can readily see that increase of nuclein metabolism may increase the output of this substance. Strangely enough, it has been found that the intake either of pure nuclein or of nuclein-containing foods was not associated with a corresponding increase in the urinary oxalic acid, so that we must assume that a large portion of this intake escapes in the feces in the form of calcium 1 Pathologic des Harnes, Berlin, 1903. 2 For methods of determination see von Furth and Charnass, Biochem. Zeitsch., Bd. 26, 1910, S. 199; also, Mondschein, Ibid., Bd. 42, 1912, S. 91 and 105. THE URESTE. 237 oxalate. It has, moreover, been found that the administration of oxalates by the mouth is not associated with increased excretion either in the urine or feces, so that we must assume a decomposition of the oxalic acid into carbon dioxid and water somewhere in the system. This conversion appears to take place in the intestinal canal under the influence of bacterial action. Crystals of cal- cium oxalate are frequently found in the urine in cases showing a marked in- crease in the output of ethereal sulphates, so that we must either assume an increased formation from the carbohydrates of the food, an increased intake and absorption of oxalates of the food, or an increase in the nuclein metabolism as a result of absorption of the toxic products of intestinal putrefaction. Among the foods which are known to contain relatively large amounts of oxalic acid we find spinach, rhubarb, tomatoes, carrots, celery, string-beans, green peas, potato, figs, plums, strawberries, pepper, cocoa, tea, and coffee. Those which contain very little are meat, milk, eggs, butter, cereals, rice, asparagus, cucumbers, mushrooms, lettuce, cauliflower, cabbage, pears, peaches, grapes, and melons. The increased elimination of oxalic acid may or may not be associated with a deposition of crystals oi calcium oxalate. An increased elimination is observed in cases associated with irregular activity of the gastro-intestinal tract. These cases usually show many types of nervous disorder, especially neurasthe- nia, and are characterized by the large deposit of oxalate crystals in the urine. To this condition has been given the name oxaluria, which can hardly be digni- fied as a definite pathologic entity. No inference can actually be drawn regard- ing the degree of elimination of oxalic acid from the appearance of a deposit in the urine, as it has been shown by Fiirbringer that the urine may contain a large amount of oxalic acid without a sediment of calcium oxalate crystals being formed. Such cases, however, should be watched more or less closely as sepa- ration of calcium oxalate may occur within the pelvis of the kidney and lead to the formation of a calculus. In cases of jaundice a marked oxaluria may be observed which is directly referable to the associated cholemia. In occasional cases of diabetes mellitus the elimination of oxalic acid may be much increased. Quantitative Determination. Baldwin's Method.1 Five hundred c.c. of a mixed 24-hour specimen of urine are treated with 150 c.c. of 95 per cent, alcohol and the mixture set aside for 48 hours to allow the calcium oxalate to precipitate. It is then filtered, each particle of the pre- cipitate being transferred to the filter by means of hot water and a rubber- tipped glass rod. This precipitate is then washed with hot water and later with 1 per cent, acetic acid. The precipitate is then washed from the fil- ter-paper by a stream of dilute hydrochloric acid from a wash-bottle until every trace of the precipitate is removed from the filter and dissolved. The filter is then washed with hot water until the washings are no longer acid in reaction. The hydrochloric acid solution and the washings are evaporated 1 Jour, of Expr. Med., vol. 5, 1903, p. 27. 238 DIAGNOSTIC METHODS. to about 20 c.c., a little calcium chlorid solution is added, the solution is neutralized with ammonia, is then rendered slightly acid with acetic acid, and 95 per cent, alcohol added in an amount equal to one-half the volume of the liquid. The mixture is then set aside for 48 hours, after which the precipitate of calcium oxalate is collected on an ash-free filter, washed with cold water and dilute acetic acid until free from chlorids, and the filter with its contents is incinerated in a weighed platinum crucible. This latter process is first carried out over a Bunsen burner and later over a blast-lamp. The crucible is then dried in a desiccator and weighed. The difference in weight represents the amount of calcium oxid obtained from 500 c.c. of urine. Each gram of this oxid represents 1.6 grams of oxalic acid. Other organic substances are found in traces in the urine, both normal and abnormal, but do not have any clinical significance. Among these we find lactic acid, succinic acid, traces of carbohydrates, and certain little-understood organic acids. These may be found discussed in works on physiologic chemistry. Several ferments1 have been demonstrated in the normal and pathologic urine, but do not seem to have any great clinical importance at the present time. (4). Ferments. This ferment is present in practically every specimen of urine. It has been found by Grober, Gehrig, Griitzner, Mathes, Stadelmann, and others. It seems to be absent or very much diminished in cases of typhoid fever, gastric carcinoma, and hypoacidity.2 In cases of pneumonia, Lenobel and Kun and Lochbihler have observed the presence of a ferment, which seemed to be pepsin in increased amounts. Scola reports a diminution of the normal pepsin con- tent in severe diseases of the nervous system. Pepsin. Lipase. This ferment is present normally only in minute traces. It is found, however, in cases of hemorrhagic pancreatitis, jaundice, and in diabetes melli- tus. It may be detected by the method of Kastle and Loevenhart which is as follows: In each of three flasks are placed 5 c.c. of urine. One of the flasks is boiled to destroy the ferment which may be present. To a second flask are added a few drops of phenol-phthalein solution and the acidity determined by titration with tenth-normal sodium hydrate solution. The amount of alkali necessary to neutralize the 5 c.c. of urine is then added to each of the other flasks. To these flasks are then added 0.25 c.c. of ethyl-butyrate and 0.1 c.c. of toluol, the flasks being then placed in the incubator at 370 C. for 24 hours. To each of these flasks there is then added 1/2 c.c. more tenth-normal hydro- chloric acid than the amount of tenth-normal alkali previously added. The mixture is then shaken out with 50 c.c. of ether and 25 c.c. of alcohol to remove 1 See Lindemann, Zeitsch. f. klin. Med., Bd. 75, 1912, S. 58. 2 See von Kozawa, Int. Beit. z. Path. u. Ther. d. Ernahr., Bd. 4, 1912, S. 44; Fuld and Hirayama, Zeitsch. f. exp. Path. u. Ther., Bd. 10, 1912, S. 248; also, Tachau, Zeitsch. f. klin. Med., Bd. 76, 1912, S. 167. THE URINE. 239 the butyric acid. This is then titrated with tenth-normal sodium hydrate, each c.c. of tenth-normal sodium hydrate representing 0.0088 gram of butyric acid. (5). Mucin-like Substances. (a). Mucin. True mucin is present in traces in practically all urine. It is found both as an insoluble portion which forms the nubecula and as a soluble portion which is much smaller in amount than the insoluble form, is precipitated by acetic acid, but is easily soluble in a slight excess of the acid. This form of protein is derived from the urinary passages and has practically no pathologic importance, although it may be much increased in catarrhal conditions of the urinary tract in which it may appear as a gelatinous ropy material, or in rare cases in the form of casts of the ureter or urethra from i to io cm. long and 3 to 4 mm. thick. These cases are rare and have been reported by von Jaksch under the name of "ureteritis membranacea" and by Frank under the name of "pyelitis productiva." Mucin is precipitated by the ordinary reagents for albumin (to be dis- cussed later), but is soluble in an excess of these reagents, so that it need not be mistaken for albumin. It is frequently confused with the"nucleo-albumins," but may be distinguished, chemically, by the fact that it contains no phosphorus and gives on heating with acids a substance which reduces copper solutions. The amount of true mucin in the urine is clinically of little importance, so that the writer will refer elsewhere for quantitative methods. (&). Nucleo-albumin. The large majority of specimens of urine contain a substance (other than true mucin) which is precipitated on the addition of cold acetic acid. Dilute acetic acid does not redissolve the precipitate, so that true mucin is excluded, as the latter is usually dissolved even by dilutions of acetic acid which do not preipcitate the other bodies. The reaction with acetic acid is somewhat inten- sified if the urine be diluted. Practically every normal urine contains traces of such substances, which will give a precipitate with acetic acid, especially if the salts be removed by dialysis. In all probability this body, found in normal urines, is either euglobu- lin or a mixture of this protein with fibrinogen. It has been called "nucleo- albumin," but this is in all probability a misnomer. It is probably true that real nucleo-albumin is never a normal constituent of the urine. In this con- nection it must be stated that much confusion exists regarding the nature of true nucleo-albumin. This substance has been considered to be identical with nucleo-protein, but with absolutely no basis of chemical facts. Nucleo-protein (true nuclein) is a combination of protein with the prosthetic group, nucleinic acid, which splits up into phosphoric acid and purin bodies; while nucleo-al- bumin (pseudo-nuclein) is a combination of protein with paranucleinic acid, which is rich in phosphorus, but does not yield purin bases on hydrolysis. 240 DIAGNOSTIC METHODS. Care should, therefore, be taken in calling a substance by a specific name unless it can be shown chemically to be such a body. Mbrner1 believes that most of the.so-called "nucleo-albumin" is a com- pound of true serum-albumin with an albumin-precipitating body formed on addition of acetic acid. He showed that there were three such precipitating bodies present in the urine; chondroitin-sulphuric acid was practically always present, nucleinic acid occasionally present, and tauro-cholic acid, which may be normally present in traces, but in certain pathologic conditions is much in- creased. He believes that these precipitating bodies are normally present in excess and, therefore, any increase of a precipitate on addition of the acetic acid would mean an increased excretion of albumin. The more these precipi- tating bodies predominate the more the precipitate resembles "nucleo-albumin." We see, therefore, that the true nature of this body or group of bodies is still unsettled. It is wise, in the writer's opinion, not to attempt a differentiation at present, but to regard these bodies as normal constituents, which at times may be excreted in abnormal amount. From the pathologic standpoint what has been called true "nucleo- albumin" appears in conditions destroying the integrity of the epithelium of the uriniferous tubules or of the bladder as well as in conditions associated with the excretion of pus in the urine. Thus we would expect to find such a body in acute nephritis, whether the result of bacterial or of exogenous toxins, in the acute febrile diseases, in renal hyperemia, in leukemia, in acute yellow atrophy, and in obstructive jaundice, in which case this body is derived doubtless from the bile. In cases of nephritis this body, precipitable by acetic acid in dilute solution, may precede and follow the true albuminuria. In amyloid kidney this body seems to be the chief type of protein present. In orthostatic albumin- uria this substance may be the only protein present and may persist after the others have cleared up. In cases in which the urine contains a large number of epithelial cells, casts, and pus-cells, Matsumoto finds a substance precipi- table by acetic acid, but only in very small amounts. This finding would seem to indicate that nucleo-albumin, if such ever occurs, is at least not an indica- tion of cellular origin from increased epithelial desquamation. This body may be found in general catarrhal conditions of the urinary tract, as in cystitis or pyelitis, but in such cases we are more apt to obtain true mucin. In the above discussion the writer has not attempted to differentiate these bodies, as the reports in the literature have little importance beyond the fact that a body precipitable by acetic acid was obtained. Each worker has named this body as he understood it and in many cases has had no definite basis for such conclusion. The writer feels, therefore, that this body, whatever it may be found to be, cannot at present be considered of any pathologic importance as it occurs in such widely varying conditions and has been chemically so little understood. To remove it from the urine, add a solution of lead acetate and filter; the precipitated phosphates and chlorids carry down this protein. 'Skand. Arch. f. Physiol., Bd. 6, 1895, p. 332. THE URINE. 241 (6). Pigments and Chromogens. (a). Urochrome. This pigment is the chief coloring matter of normal urine, imparting a yellow, orange, or a brownish color to the urine, depending upon its concentra- tion. It is closely related to urobilin, especially the so-called normal urobilin of MacMunn, as this latter body may be readily converted into urochrome by evaporation of its aqueous ethereal solution. This body has not been isolated, at least in the pure state, so that little is known of its chemistry. It is, in all probability, a mixture of one or more pigments, contains about 4 per cent, of nitrogen, and is free from iron. It is readily soluble in water and alcohol; sparingly soluble in acetic ether, amyl alcohol, and acetone; insoluble in ether, chloroform, and benzol. Much difference of opinion exists regarding the spectrum of this pigment, some stating that there is no absorption spectrum, while others either describe a faint, narrow absorption band between F and G, or a broad absorption band at F, which extends more to the left than to the right of this line. As this substance has so little clinical importance, the writer must refer to works on physiologic chemistry for its detection. (&). Uroerythrin. This pigment is a constituent of a large majority of normal urines, although it is probable that it indicates a slight deviation from the normal. It has been called rosacic acid by Prout and purpurin by Golding-Bird To this pigment is due the salmon or brick-red color which the urinary sediments take in highly concentrated febrile urines. Normally, it may not be present in sufficient amount to color the urine, but under pathologic conditions it may impart a deep orange tint to this fluid. Little is known regarding the chemistry of this pig- ment, but it seems to be iron-free. It is soluble in amyl alcohol, slightly soluble in acetic ether and absolute alcohol and very difficultly soluble in water. For the chemical and physical properties of this pigment the writer must refer elsewhere. This pigment appears to be increased on a meat diet, after severe exercise, profuse perspiration, or by irregular digestion. Pathologically, it is observed especially in cases of hepatic insufficiency, in chronic cardiac and pulmonary disease, in acute articular rheumatism, in malarial fever, and in general acute febrile diseases. In typhoid fever one does not find this pigment as frequently as in most other acute febrile conditions. (c). Urobilin. This substance appears in the urine not as a free pigment, but in the form of the chromogen urobilinogen, which is decomposed into urobilin through the influence of the light. It is claimed that various types of urobilin are found, as for instance the normal urobilin of MacMunn and the pathologic urobilin of Jaffe. Whether these are really different bodies is at present an unsettled question. Urobilin appears to be identical with the stercobilin of the feces and is not the same as the normal fecal hydrobilirubin. Much discussion has centered around the origin of this pigment. Passing through the stages of the 242 DIAGNOSTIC METHODS. hepatogenous, hematogenous, nephrogenous and histogenetic urobilinuria, the general concensus of opinion seems at present to be that most of the urinary urobilin is of enterogenous origin.1 According to this theory, the pathogenesis of urobilinuria may be presented as follows: "The liver-cell, both in normal and abnormal conditions, forms only bilirubin from the blood pigment. Providing there is no marked obstruction to the passage of bile into the intestine, the bilirubin is acted upon by bacteria which reduce it so completely to urobilin that only traces of bilirubin appear in the feces. A part of the urobilin is absorbed and is excreted in the urine, while traces appear in the bile and in pathologic transudates. When bacte- rial action is excluded, as in the new-born, no urobilin is found in the urine. Further, when bile is not present in the intestine, as in cases of absolute oc- clusion of the ductus choledochus, urobilinuria does not occur. It is sparingly excreted when the production of biliary pigment is diminished, as in hunger, while the amount is small or at most normal in cases of incomplete exclusion of bile from the intestine. On the other hand, the amount excreted may reach abnormal limits if a preceding obstruction, accompanied by stasis, has been overcome, and bile flows freely into the intestine. Likewise the quantity may be abnormally large if the production of biliary pigment from the red blood- corpuscles increases as a result of infection and intoxication, or of hepatic lesions, such as cirrhosis and cyanotic induration. In these cases the bile is tenacious and the condition may give rise to jaundice although it is seldom that the stasis is so great that the bile is completely shut off from the intestine. Indeed, in most cases, owing to the excretion of excessive pigments in the bile (pleiochromia), a more than normal amount of pigment passes into the intestine, and as a result of this there arises a marked urobilinuria with a mild degree of biliary stasis. In many cases the stagnation of bile is great enough to cause a passage of biliary pigment from the blood into the urine, leading to a marked urobilinuria, a mild degree of bilirubinuria, and a yellow-tinting of the tissues. In other cases the absorption of bile is so slight that only yellowing of the tissues results, yet the concentration of pigment in the blood does not suffice to permit of its excretion by the kidneys, so that marked urobilinuria and yellowing of the tis- ues without bilirubinuria ensues" (Weintraud). Urobilin is found in febrile conditions, chronic passive congestion, lead- poisoning, cases in which extravasation of blood into the tissues occurs, in any condition associated with marked hemolysis, hepatic cirrhosis, and in the cases of jaundice outlined above by Weintraud. It has been noticed in increased amounts in Addison's disease, extrauterine pregnancy, hemophilia, and in secondary syphilis. The presence of an increased amount of urobilin usually causes a dark- yellow color of the urine, the foam in such cases, being colored more perhaps due to the presence of other pigments than to urobilin itself. Urobilin is soluble in ethyl alcohol, amyl alcohol, and chloroform, and 1 See Wilbur and Addis, Jour. A. M. A., vol. 59, 1912, p. 929; also, Fromholdt and Ner- sessoff, Zeitsch. f. exp. Path. u. Ther., Bd. 11, 1912, S. 400. THE URINE. 243 slightly soluble in ether, acetic ether, and in water. If an acid solution of urobilin be examined with a spectroscope, it shows a broad absorption band to the right of E, the left-border of which reaches nearly to b, while the right border encloses F. If the solution be alkaline, the spectrum shows a less broad absorption band between E and F enclosing b. This solution is best made as follows: Ten to 20 c.c. of the urine are acidulated with a few drops of hydro- chloric acid and shaken out with from 6 to 10 c.c. of amyl alcohol, which then shows the characteristic spectrum of acid urobilin. If to a small portion of this amyl alcohol solution be added a few drops of 1 per cent, solution of zinc chlorid, which has been strongly alkalinized with ammonia, a beautiful green fluores- cence appears. With Ehrlich's dimethyl-amidobenzaldehyde reaction (p. 325) this pigment gives a distinct red color. In the decomposition of protein occurring in the intestinal canal indol (CgH7N) and skatol (C9H9N) are found among the products of this bacterial cleavage. These substances are absorbed and oxidized in the blood to indoxyl (CgH7NO) and skatoxyl (C9H9NO). These bodies are then conjugated with sulphuric acid forming indoxyl and skatoxyl sulphuric acids, after which they are excreted in the form of the potassium salt, indoxyl potassium sulphate (CgH6NO-SO2-OK) and skatoxyl potassium sulphate (C9H8NO-SO2--OK). To the former of these is given the name indican. This chromogen is present in much larger amounts than is the skatoxyl compound so that our discussion will include principally the former. The absolute amount of indican occurring in the urine depends upon the amount of decomposition occurring in the intestine. This subject has been treated of previously under the head of ethereal sulphates to which the reader is referred. The largest amounts are naturally observed in health following a meat diet, while the urine may be practically free from indican on a vegetable or a milk diet. The pathologic variations may be summed up as follows. An increased elimination of indican is observed in all diseases which are associated with an increased intestinal decomposition.2 This decomposition usually occurs in the large intestine, but may at times take place in the small bowel, in which cases the degree of indicanuria may be greater owing to the increased absorptive power of the small bowel. Although many writers state that indicanuria is not seen in cases of simple constipation, the writer must take exceptions to these statements as some of his most intense and persistent cases of indicanuria have been found in uncomplicated constipation. Secondly, an increased excretion of indican is observed in cases showing diminished peristalsis, as, for instance, in ileus and peritonitis. These cases frequently show an intense reaction.3 If the obstruction leading to ileus be in the large (4/). Indican.1 1 See Baar, Die Indicanurie, Wien 1912. 2 See Morgan, Amer. Jour, of Med. Sci., vol. 144, 1912, p. 827; also, Benedict, Arch, of Diag., vol. 5, 1912, p. 159. 3 Although there appears to be no definite relation of indicanuria and albuminuria, the writer has rarely failed to observe large numbers of casts in urines showing marked indican reactions. 244 DIAGNOSTIC METHODS. bowel, indican is either absent or appears much later than is the case if the small bowel be involved. Lastly, in any condition associated with protein decomposition anywhere in the system, as, for instance, in empyema, putrid bronchitis, abscess formation, etc., indican may be much increased. The color of the urine is usually normal when voided, although a large amount of indican may be present. In some cases oxidation of this chromogen has occurred within the system and the urine appears greenish or even blue when voided. If the urine be allowed to stand until decomposition occurs, a reddish or bluish metallic-like scum may be observed, due to the conversion of the indican into indigo-blue. Most of the tests for the presence of indican in the urine depend upon the oxidation of the indican, according to the following equation (this body having been previously decomposed by concentrated HC1 into indoxyl and sulphuric acid). 2C8H7NO + 20 = C16H10N2O2 + 2H2O. Tests for Indican. Jaffe's Test. A few c.c. of urine are treated with an equal volume of concentrated hydrochloric acid, two or three drops of a strong solution of calcium hypochlorite (bleaching powder or "chlorid of lime") are added and the contents mixed. Two c.c. of chloroform are then added and the tube inverted several times. In this process the indican is oxidized to indigo-blue which is taken up by the chloroform. The depth of the blue coloration of the chloroform will serve as an approximate estimate of the amount of indican present. In absolutely normal urine no blue coloration or, at most, a faint bluish tinge is observed. Care must be taken in this test to avoid an excess of the hypochlorite, as this will convert the indigo-blue into isatin which is distinctly yellow, according to the following equation: C16H10N2O2 + O2 = 2C8H5NO2. Obermayer's Test. A few c.c.of urine are mixed with an equal volume of Obermayer's reagent (a 0.2 per cent, solution of ferric chlorid in concentrated hydrochloric acid) and the solutions mixed by repeatedly inverting the tube. A few c.c. of chloro- form are then added and the tube inverted as before. The presence of any appreciable amount of indican is indicated by a dark brown to black coloration of the urine and the later absorption of this color by the chloroform, which becomes a more or less deep shade of blue. If the urine be very dark in color or if bile pigment be present, the pigments may be removed by adding a solution of lead subacetate and filtering. If potassium iodid be present this test yields a red color which will disappear on addition of sodium thiosulphate solution. Salicyluric acid gives a violet coloration. If formalin be used as a preservative, indican is incapable of detection by any of the tests given. Instead of using the Obermayer reagent as such, the urine may be mixed with the hydrochloric acid and one or two drops of a io per cent, solution of THE URINE. 245 ferric chlorid may then be added to the mixture. This test is a much more reliable one than is the Jaffe test, as the latter is very prone to carry the oxi- dation to the isatin stage rather than to the indigo-blue phase. Even under the best conditions a certain amount of isatin will be formed so that quantitative determinations are a matter of some difficulty. A certain amount of skatoxyl-potassium sulphate is formed along with the indican. In some cases for reasons not well understood, this pigment or one closely allied to it appears in the urine in excess of the indican, so that the above test may give distinct red colorations of the choloroform instead of the usual blue. Just exactly what pigment causes these red variations is doubtful. Some of this pigment is always present along with the indican and may be extracted either with hot water or with a mixture of alcohol, ether and water. To this red pigment has been given the name indigo-red, skatoxyl-red, urorubin, urorhodin, and several others. If the urine be heated on applying Jaffe's test, a dark red coloration is observed, while if Obermayer's test be used the coloration will be of a reddish- violet. Skatoxyl-potassium Sulphate. A few c.c. of urine are boiled and? concentrated nitric acid added drop by drop during the boiling. The urin© takes a deep red color and the foam appears bluish-red. If the nitric -acid be much in excess the urine will assume a yellow- ish-red color and the fo'am a distinctly yellow tint. If sodium hydrate or ammonia be now added drop by drop, a bluish-red precipitate is observed which is soluble in an excess of the alkali with a brownish-red coloration. This test is due to the presence of indigo-red. Its clinical significance is the same as that of indican. This pigment may be obtained directly from the urine by neutral- izing it with sodium hydrate and shaking out with ether, when the ether will take a distinct red color. Rosenbach's Test. Quantitative Determination. Wang's Method. The principle of this method1 is the decomposition of indican and its oxidation to indigo-blue. This compound is then transformed into indigo- sulphuric acid, which is directly determined by titration with potassium per- manganate solution. A preliminary determination of the relative amount of indican is made with Obermayer's reagent. If a strong reaction is obtained, from 25 to 100 c.c. of urine are used, while if the reaction be slight 200 to 500 c.c. are necessary. The urine should be acidified with acetic acid, unless its reaction be already acid. Fifty c.c. of urine, or a larger amount if the conditions mentioned above obtain, are treated with 5 c.c. of a 20 per cent, solution of lead acetate, or Technic. ^eitschr. f. physiol. Chem., Bd. 25, p. 406. 246 DIAGNOSTIC METHODS. one-tenth the volume of urine in case a larger amount of urine be taken. The urine is then filtered and a large and accurately measured portion of the filtrate is treated, in a separatory funnel, with an equal volume of Obermayer's reagent. This mixture is then shaken out with chloroform, using 30 c.c. of this menstruum and shaking for one minute. At least four such chloroform extractions should be made, more if the chloroform still extracts indigo-blue from the mixture. The chloroform extract is placed in a small flask and the chloroform distilled. The residue in the flask is dried for a few minutes on the water-bath to remove the last traces of the chloroform, and is washed either with hot water or with a mixture of equal parts of alcohol, ether, and water. These solvents remove the red coloring matter, leaving the indigo-blue undissolved. The extract is filtered through a small filter and the indigo-blue completely transferred to the filter, after which it is thoroughly washed with hot water. This indigo-blue is dissolved on the filter with boiling chloroform, the filtrate being allowed to run into the original flask. The chloroform is again distilled and the residue dried on the water-bath. This purified indigo blue is dissolved in 10 c.c. of con- centrated sulphuric acid and the solution diluted to 100 c.c. with water. This solution of indigo-sulphuric acid is then titrated with standard potas- sium permanganate solution of such a strength that 1 c.c. will represent approxi- mately 0.0062 gram of indigo-blue. This solution of potassium permanganate contains about 3 grams of potassium permanganate to the liter. Its titer is determined before use by titrating against a solution of pure indigo-blue in sulphuric acid. In making this titration the concentrated solution is not used, but a dilute one made by diluting 5 c.c. of the stronger permanganate solution to 200 c.c. (each c.c. of which will represent 0.000155 gram of indigo-blue). In this titration the blue color of the indigo-sulphuric acid does not change to any extent on the addition of the first drops of the permanganate solution, but gradually turns greenish and then becomes yellowish or entirely colorless. The amount of indigo-blue in the urine used is readily ascertained by multiply- ing the number of c.c. of permanganate solution used by the ainount of indigo- blue represented by the titer of the solution (in the writer's laboratory 1 c.c. of the diluted solution equals 0.000155 gram of indigo-blue). It has been found by Ellinger that only about 87 per cent, of the theoretical yield is obtained by this method, owing, probably, to the simultaneous for- mation of isatin from the indigo-blue. The average daily excretion of indican, as revealed by this test, ranges between 5 and 20 mg. Folin's Method. This method has not, as yet, been extended so as to give absolutely quan- titative results. Its principle is the comparison of the color given when urine is treated with Obermayer's reagent with that of Fehling's solution as a stand- ard. To this standard has been given the arbitrary value of 100. Exactly one one-hundredth of the 24-hour specimen of urine is taken for each determination and treated with an equal volume of Obermayer's re- agent. The indigo-blue is then extracted with 5 c.c. of chloroform until all of THE URINE. 247 the pigment has been dissolved. With the chloroform solutions are then made colorimetric comparisons with Fehling's solution, using either the Duboscq or Sargent instrument. Folin finds the average indican excretion, on this basis, from a diet of 119 grams of protein to be 77, while on a nitrogen-free diet no indican is found. This method while not as yet giving absolute figures is very convenient and is useful for the comparative work of each physician. (e). Uroroseinogen. This chromogen, shown by Herter1 to be indol-acetic acid, is converted by oxidation into the pigment urorosein, which is soluble in ethyl and amyl alcohol but insoluble in chloroform, ether and benzol. If the urine stands for some time bacterial action produces this pigment, although an alkaline reaction may destroy the coloration. Its amyl-alcoholic solution shows a sharp narrow absorption band between D and E. Although this pigment appears in normal urine only in traces, it may be markedly increased by a strict vegetable diet. Pathologically, it appears in cases of severe gastric and intestinal disturbances, tuberculosis, pernicious anemia, nephritis, severe chlorosis, diabetes mellitus, carcinoma, osteomalacia and typhoid fever. It has, therefore, little differential value although it is, often, as significant as indican of excessive protein decom- position in the bowel. Test for Urorosein (indol-acetic acidf. If 5 c.c. of urine be mixed with 5 c.c. of chemically pure HC1 and a few drops of a 1 per cent, solution of sodium nitrite a characteristic rose-red coloration develops, which is due to oxidizing rather than nitrifying action. This pigment also gives a red coloration with Ehrlich's dimethyl-amidobenzaldehyd reaction. (B). Abnormal Composition. (1). Proteins. There has been much discussion as to whether a true physiologic pro- teinuria occurs. Without going into a discussion of the subject the writer wishes to state his belief that a true physiologic proteinuria occurs, but that such may never be detected by the usual clinical methods of examination. For our purposes, therefore, the presence of protein of one type or another in amounts amenable to detection by clinical methods must be considered pathological. (a). Serum Albumin. From the pathologic standpoint, serum albumin is the most important protein found in the urine. The amount excreted in 24 hours is variable and does not necessarily have any relation to the severity of the kidney lesion should one really exist. From 5 to 10 grams of this protein per diem may be regarded as a moderate pathologic excretion, while a lesser would be of little significance and a greater would be regarded as excessive. As large amounts as 40 grams have been observed, but such findings are exceptional. Mbrner considers an excretion of albumin varying from 25 to 75 mg. per liter as a normal output, which amount would, however, escape the usual clinical tests. 1 Jour, of biol. Chem., vol. 4, 1908, pp. 107, 239 and 253. 248 DIAGNOSTIC METHODS. As some of the more delicate tests for albumin, as for instance Spiegler's reagent, will show traces of albumin in practically every specimen of urine examined, we should limit our conception of albuminuria to those cases which react with the ordinary tests rather than with the most delicate. According to Hofmeister, the standard upon which one bases a judgment as to the presence of a pathologic albuminuria is the formation of a distinct albumin ring within three minutes after the urine and nitric acid are in contact in Heller's test (see below). Moreover, the term albuminuria should be limited to those cases in which there is some disturbance of the renal epithelium, especially of the glomeruli. This does not exclude those cases of purely functional albuminuria in which no distinct lesion of the kidney exists, as these cases are all associated with some abnormality of the excreting organ, although it is not extensive enough to constitute a definite pathologic entity. It not infrequently happens that albumin, as found in the urine, is derived from some portion of the urinary tract below the kidney, as when inflammatory exudates, blood, lymph, spermatic or prostatic fluids, pus and other extraneous material are mixed with the urine after it is excreted by the kidney. To such cases is given the name false or accidental albuminuria, in contradistinction to the true albuminuria, in which the albumin is present when excreted by this organ. Before reporting a finding as one of true albuminuria, it is absolutely imperative that extraneous albumin be excluded. Not infrequently do we find in perfectly healthy individuals a urine which is normal in every way with the exception of the presence of albumin, which is easily detected by our ordinary methods. This type of albuminuria is observed following severe muscular exercise beyond the point to which the subject is accustomed. Thus, in raw recruits under the severe forced marches during the early days of their service we find this type very prevalent. Leube states that 59 per cent, of such soldiers show a temporary albuminuria, which disappears after the subject becomes accustomed to the increased exercise. It is a common occurrence to find such a functional albuminuria in football players, bicycle riders, crew men, and general athletes after every period of increased exertion. This question may be summed up by the statement that the excretion of albumin is simply dependent, under these conditions, upon the limit of endurance of the subject, practically everyone being able to produce an albuminuria if he overdoes in any way. Moreover, cold baths, excessive mental labor, and severe emotion may lead to quite an extensive albuminuria, especially in an individual with somewhat lessened resistance. According to Rem-Picci1 albuminuria is a constant finding after cold baths, different sub- jects reacting differently to the same stimuli. This albuminuria never lasts over 24 hours and may be associated with the appearance of casts and blood, as may also the type following increased exercise. The colder the bath and the longer the immersion, the more rapid the appearance of albumin. He gives Functional Albuminuria. *11 Policlinic©., Tm. 8, 1901, p. 389. THE URINE. 249 as the limit of temperature necessary to produce this condition 12 to 130 C* with an immersion of not longer than three minutes. Another type of functional albuminuria is that following the intake of a heavy protein meal. This is known as "alimentary albuminuria." It has been supposed to be due to filtration from the blood, the foreign protein being absorbed and excreted without intervening hydrolytic cleavage. Uhlenhuth,1 Inouye,2 and Croftan,3 working with the precipitin test, have found egg albu- min in the blood and urine in these cases. Ascoli4 later demonstrated both serum and egg albumin by this test. Wilson5 using the fixation of complement test, found the main element to be of human origin. Recently Wells,6 applying the anaphylaxis reaction,7 has shown that sensitized animals do not react to egg albumin but show marked reactions to human protein. Owing to this disagreement of the anaphylaxis and precipitin tests, it would seem probable that the excreted albumin is derived from the blood rather than directly from the food, little if any unaltered food protein ever reaching the urine.8 The albuminuria observed in the new-born for the first few days of life, aside from the influence of feeding upon foreign proteins, is probably a further example of a true functional albuminuria. Ribbert explains this by the des- quamation of epithelium of the glomerular capsule. Likewise, the albuminuria observed in pregnancy is usually distinctly functional. About 50 per cent, of pregnant women show this albuminuria, little difference being observed between primiparse and multipane. The kidney undoubtedly shows some functional dis- turbance in its attempt to eliminate the toxic products absorbed from the fetus. According to Senator9 such albuminurias as outlined above should be considered functional when slight in degree, transitory in character, occurring after unusual strain, either physical or mental, the subjects showing a further negative history after the removal of the direct stimulus. Whether this type is to be considered physiologic rests entirely upon the conception of the normal or abnormal nature of the stimuli leading to the excretion of albumin. It would seem to the writer that the term "physiologic albuminuria" would better be given up or at least used with great caution, as it is difficult to believe that the severe strain put upon the system and kidneys, in particular, can be dis- tinctly physiologic.10 He is inclined to reserve this term for the excretion of the minimal amounts of albumin, which cannot be detected by the simpler clinical 1 Deutsch, med. Wochs., Bd. 26, 1900, S. 734. 2 Deutsch. Arch. f. klin. Med., Bd. 75, 1903, S. 378. 3 New York Med. Jour., vol. 89, 1909, p. 474. 4 Miinch. med. Wochs., Bd. 50, 1903, S. 201 and 1761; Zeitsch. f. physiol. Chem., Bd. 29, 1903, S. 283. 5 Jour. Path, and Bact., vol. 13, 1909, p. 484. 6 Jour. A. M. A., vol. 53, 1909, p. 863. 7 See Schittenhelm and Weichardt, Zeitsch. f. exp. Path. u. Ther., Bd. 11, 1912, S. 69; Pearce, Jour, of Exp. Med., vol. 16, 1912, p. 349; also, Schloss, Amer. Jour. Dis. of Child., vol. 3, 1912, p. 341. 3 See Van Alstyne and Grant, Jour, of Med. Res., vol. 25, 1911, p. 399. 9 Erkrankungen der Nieren, Wien, 1902. 10 See Wolf, Arch, of Diag., vol. 5, 1912, p. 260. 250 DIAGNOSTIC METHODS. tests, and to regard the types of albuminuria, as discussed above, as "func- tional" (Pavy), or "constitutional" (Martius). A second group of cases showing albuminuria is observed in which the excretion of albumin may persist for a period varying from days to months, may disappear for a variable space and then return for another indefinite length of time. To this type is given the name "intermittent albuminuria." This is purely functional, no lesion of the kidney being manifest. The cases usually show a history of an acute infection or of an antecedent nephritis as the result of such infection. The most frequent cause, however, is an uncompensated heart lesion which may or may not be associated with a direct renal lesion. Not infrequently we find a.n hereditary intermittent albuminuria in those with a distinct neurotic family history. The functional albuminuria may, at times, follow a definite course, disap- pearing and reappearing with such regularity that it has been styled "cyclic albuminuria." In this form the albumin usually disappears from the urine at night and when the patient is flat on his back, but reappears during the day or when the subject is erect. The terms "orthostatic, orthotic, or postural albuminuria" would, therefore, seem to be more appropriate than the former appellation. This type is dependent, according to Erlanger and Hooker, upon a lowering of the pulse pressure which constantly occurs when the individual changes from the recumbent to the erect position. Jehle and, more re- cently, Nothmann believe that this type is due to a lordosis, and style it, therefore, lordotic albuminuria.1 Certain pathologic cases, as beginning nephritis, may show a cyclic albuminuria which may extend well into the period of recovery. In true cyclic cases the negative physical findings would lead us to class the condition among the functional albuminurias. If a case of this type becomes persistent and casts are frequently found, sub- sequent cardiovascular changes will usually appear, changing the case into one of true nephritis. The albumin in such cases usually appears in the urine after rising and reaches a maximum from noon to 3 or 4 o'clock, then gradually declines, disappearing from 8 to 10 p. m. If the subject changes his habits of life, the cycle of albuminuria will also change. Along with the variations in the excretion of albumin, the other urinary constituents fluctuate in the same manner, the sequence being, according to Teissier, increase in pigments, al- bumin, uric acid, and urea. A peculiarity of this type of albuminuria is that it may be even diminished by exercise and is less, therefore, after a hard day's work, differing from the albuminuria in nephritis or in the cardiac types. While there has been much discussion as to whether such cases are not truly pathological, most of them at any rate must have a certain insufficiency of the renal epithelium. This may be due to changes in the circulation following change in posture, but it is rather surprising that increased exercise would not 1 See Hamburger, Wien. klin. Wochs., Bd. 25, 1912, S. 262; also Frankel, Deutsch, med. Wochs., Bd. 38, 1912, S. 1985, who finds the urinary acidity increased in these cases and states that, possibly, the albuminuria is due to abnormal production of acid as Fischer's theory assumes. See, also, Gomolitsky, Zeitsch. f. klin. Med., Bd. 77, 1913, S. 96. THE URINE. 251 work more strongly in this way. Krehl regards these conditions as relatively harmless as they do not usually show any subsequent history of nephritis; Broadbent does not believe such cases ever develop actual renal disease, while Senator insists that most of them are cases of nephritis, either at the onset or during a latent attack of the disease. The patients showing this type of albuminuria are weak anemic individ- uals about the age of puberty, subject to fainting spells, with a heart showing intermittent attacks of dilatation and palpitation, and with a probable con- genital weakness of the kidney (37.5 per cent, having a movable kidney). The adults are neurasthenics with distinct vasomotor paralysis. A type of albuminuria has been observed in some patients with enlarged spleen, in which albumin is present when the patient is flat on his back but is absent when he is erect. This has been called 11 hypostatic albuminuria." Its pathogenesis is uncertain, but it can have only an indirect relation to the enlarged spleen, as many cases of splenic tumor do not show this albuminuria. A still further type of functional albuminuria is known as 11 albuminuria oj adolescence." This occurs between the ages of 14 and 16, and then disappears. It is different from the cyclic type, although this latter occurs in young people. The children are usually anemic, have a neurotic family history, have an unstable vasomotor system and, possibly, a congenital weakness of the kidney. In this class should be included the albuminuria shown by masturbating children or after sexual excess at this period. The kidney does not keep pace with the physical growth and activity of the system, so that the association with the unstable vasomotor system may account for the albuminuria (Emerson). According to Sutherland, movable kidney may be accountable for some of these cases, as he finds it present in one-third of his cases. Febrile Albuminuria. During the course of any acute fever, an albuminuria may be observed which is not associated with distinct changes in the renal parenchyma and dis- appears with the fall of temperature. The amount of albumin excreted may be small or great, depending upon the severity of the toxic action of the bacterial products. In ordinary cases there is practically no inflammatory condition present in the kidney, the albuminuria being due to an ischemia and a later hyperemia. In some of the infectious fevers the influence of the toxins is so great that a true nephritis originates, as especially noted in scarlet fever and diphtheria. Any febrile albuminuria may pass into a true nephritis, so that the case must be closely watched for the appearance of symptoms indicating such a complication. In some cases, an increase in the albuminuria is observed during convalescence, when only traces were previously noted. This is known as 11 colliquative albuminuria. " A transitory albuminuria may be observed following injuries to the kidneys or even after bimanual palpation of this organ. The albumin and casts may persist for a variable period with no other signs of renal involvement. In Traumatic Albuminuria. 252 DIAGNOSTIC METHODS. cases of movable kidney, especially during Dietl's crises, an albuminuria may be noted due to obstruction in the renal circulation. In cases of ureteral stenosis, an albuminuria may be observed as the result of the impeded outflow of urine. This same type of albuminuria may be seen after blocking of the ureter by a calculus, pressure from a tumor, or twisting of the ureter.1 Hematogenous Albuminuria. By this type of albuminuria we have in mind one in which the albumin is excreted as a result of some alteration in the quality and quantity of the normal protein of the blood. On the other hand a distinctly hematogenous albuminuria may be the result of the excretion of an abnormal protein. This type is observed in purpura, scurvy, pernicious anemia, chronic lead or mercury poisoning, syphilis, leukemia, jaundice, cachexia, after the inhalation of anesthetics, and in diabetes. This type is directly referable to the influence of various toxic agents upon the kidneys. The changes in the kidney may be either of a degenerative order, leading to a distinct nephritis, or may be purely circulatory. Among the substances causing such an albuminuria we find ether, chloroform, mustard, cantharides, mercury, lead, arsenic and antimony compounds, oil of turpentine, potassium nitrate and chlorate, phosphorus, carbolic acid, salicylic acid, tar compounds (aniline derivatives), and petroleum. Toxic Albuminuria. 'Neurotic Albuminuria. A slight transitory albuminuria may be observed in epilepsy (in which condition it may not always be found but is present invariably when marked cyanosis is seen during the attack), in apoplexy, tetanus, progressive paralysis, exophthalmic goiter, mania, delirium tremens, migraine, brain tumor, injuries to the head especially affecting the floor of the fourth ventricle, neurasthenia, and various psychoses. Not infrequently we find neurotic patients showing an albuminuria as the direct result of perverted metabolism and not as the consequence of pathologic changes in the nervous system. Albuminuria with Definite Renal Lesions. In acute nephritis an intense albuminuria is a constant and important symptom. The more acute the case the larger will be the amount of albumin, the elimination being generally proportionate to the severity of the disease, although some acute cases may show no albuminuria (Herringham2). The percentage of albumin varies inversely as the amount of urine, as a rule, so that it is much better to excrete a larger amount of urine with a low percentage of albumin than a diminished amount of urine with an increased percentage of albumin. The absolute quantity of albumin excreted varies from 0.2 to 1 per cent. It may reach as high as 5 per cent, or higher, in one case of Senator being 8 per cent., but this is rare. The total excretion in 24 hours is rarely over 1 See Evans, Wynne and Whipple (Bull, of Johns Hopk. Hosp., vol. 23, 1912, p. 311) who have shown that a reflex albuminuria results from irritation of the urinary bladder. 2 Trans. Clin. Soc. of London, vol. 34, p. 901. THE URINE. 253 25 grams. Nephritis of syphilitic origin appears to be associated with the largest outputs of albumin. In cases of active renal congestion from exposure to cold or through the action of drugs, or in chronic passive congestion due to cardiac, pulmonary, or hepatic lesions an albuminuria may be observed without any trace of an active renal lesion. The albumin, in these cases, is small in amount and runs parallel to the quantity of urine, thus differing from the excretion in true nephritis. In chronic parenchymatous nephritis the elimination may be relatively large, exceeding, in some cases, that of the acute form.' In the chronic inter- stitial type the albuminuria is very slight, rarely amounting to more than 5 grams. In this type of nephritis the albumin may be absent at various exami- nations, so that frequent investigations of the urine must be made. In amyloid kidney, the urine closely resembles that of the interstitial type of nephritis, a total absence of albumin being, however, less frequently observed. The serum globulin in this type of kidney disease is relatively more increased than in any other type of renal disorder, so that the albumin-globulin quotient may be of some importance in diagnosis. This protein is soluble in distilled water and is coagulated by heat, if the solution be acid, at a temperature varying between 56 and 8i° C. The tem- perature at which any protein coagulates on heating will depend upon the amount of salts present. This protein is precipitated by absolute alcohol and by salts of the heavy metals, as well as by the ordinary alkaloidal precipitants. It is levo-gyrate, its degree being represented by the following formula (a)D= -62.6°. It is precipitated by concentrated mineral acids, but is dissolved by somewhat large excess. With acetic acid the precipitate first formed is readily soluble in a slight excess of the acid. With concentrated alkali serum albumin forms an alkali-albuminate which is less soluble in water than is albumin, but which is soluble in an excess of the alkali. This fact accounts for the spon- taneous precipitation of albumin in a concentrated urine which is alkaline in reaction. Numerous tests have been given for the detection of albumin in the urine. The writer cannot attempt to describe all of these, but must select, therefore, those which he has found most useful. The qualitative tests for proteins in general may be found in any work on physiologic chemistry. Before any test for albumin may be made, the urine must be absolutely clear. It is advisable always to use fresh specimens, but if these are not at hand methods must be adopted to clear up the urine. In the majority of cases fitration through several folds of filter-paper will usually accomplish this. If this does not succeed, as it practically never does if the urine be cloudy from the presence of bacteria, recourse must be had to precipitating agents which will carry down the suspension of bacteria. Such agents are powdered mag- nesium oxid or carbonate, silicic acid, or saw-dust. These substances are Tests for Serum Albumin. ^ee von Hosslin, Deutsch. Arch. f. klin. Med., Bd. 105, 1912, S. 147. 254 DIAGNOSTIC METHODS. thoroughly mixed with the urine and the mixture then filtered through double folds of filter-paper or plugs of asbestos fiber. The addition of lead acetate or any of the salts of the heavy metals is inadvisable, as the precipitates formed will include a large part of the albumin, the other precipitants not affecting the albumin directly. Occasionally the urine may be cleared by centrifugation. It has been found that the tests for albumin are rendered more distinct if the urine be somewhat diluted. Hallauer1 has shown that the excess of urea and phosphates in a concentrated urine interfere to some extent with the delicacy of the reactions. As a rule, a 24-hour specimen is examined or a speci- men of the urine voided in the morning and that voided at night. The varia- tions of the voidings of the different periods of the day are occasionally quite marked, the morning specimen frequently showing no albumin while the evening specimen may show quite appreciable amounts. Heat Test. This test is based upon the principle that serum albumin is coagulated by heat especially in the presence of acid. One may use either acetic acid or nitric acid, but the conditions of this addition are different in each case. If acetic acid be added one must be careful lest he add an excess, as the albumin precipitate is soluble in a very slight excess of acetic acid; with nitric acid the condition is the reverse, care being taken not to add too little else the albumin will not be precipitated by the acid. A few drops of dilute acetic acid are all that is required while with nitric acid between one-twentieth and one-tenth of the volume of the urine must be added (one to two drops of 25 per cent, nitric acid per c.c. of urine). Technic. A test-tube is filled about three-quarters full of the clear neutral or faintly acid urine and heated by directing the flame upon the upper portion of the tube, the lower portion being held in the hand. If the fluid remains clear and the reaction is acid, no albumin is present. If a cloud is noticed it may be rendered more distinct by holding the tube against a black back-ground when the upper portion will appear more turbid than the lower. This cloud may be due to albumin or calcium phosphate, rarely to calcium carbonate. To determine which is the cause, acidulate with a few drops of 5 per cent, acetic acid. If the urine becomes clear, the precipitate first noticed is calcium phosphate; if it remains turbid and even increases in intensity of turbidity the precipitate is albumin; while if due to carbonates an effervescence will be observed. It is wise to boil the urine after the addition of each drop of acid, so that the danger of getting an excess of acid may be more easily avoided. It is to be remembered that, if the protein be very slight in amount, and especially if the urine be originally alkaline, the protein will remain in solution owing to the formation of acid albumin. If not enough acid is added, the precipitate of phosphates may not dissolve, while if too much be added the albumin will dissolve. For these reasons it is better to add the acid after boiling. In some cases the fresh 1 Munch med. Wochensch., Bd. 50, 1903, p. 1539. THE URINE. 255 urine is already too acid to permit of coagulation, so that alkali may be added to diminish the acidity. The presence of "nucleo-albumin" may lead to a wrong interpretation in this test. This substance is precipitated in the cold by acetic acid and may thus be differentiated. The resinous acids, which are excreted in the urine after the intake of such drugs as copaiba, cubebs, and benzoin, are not so apt to interfere with this test unless a large excess of acetic acid be used, which is never admissible. If nitric acid be used in place of acetic acid the urine is boiled as above and concentrated nitric acid added to a strongly acid reaction. The nitric acid should never be added before boiling the urine nor should the urine be boiled after the nitric acid is added, as traces of albumin will be dissolved by the hot nitric acid. A flocculent precipitate is indicative of albumin. The phosphates and carbonates do not confuse in this reaction as they are readily dissolved. The "nucleo-albumin" is also eliminated by this test as it is readily soluble in the excess of acid. In this test the urine should be set aside and allowed to cool after the boiling is complete, as albumoses, if present, will separate out on cooling as a distinct white flocculent precipitate. A precipitate of uric acid may also form on cooling, but this is more granular and is usually colored, while the albumin precipitate is white unless an admixture of blood be present. In case the urine be poor in salts, the tests are improved by the addition of a saturated solution of sodium chlorid. The salts hinder to a great extent the solution of the albumin by the acids. In such cases, therefore, it is wise to acidify strongly the urine with acetic acid and then add one-sixth its volume of a saturated solution of sodium chlorid as recommended by Purdy. The urine is now boiled as above when a precipitate on heating will indicate albumin. The nucleo-albumin reaction is slight, the albumoses appear only on cooling, while the resinous acids may be precipitated but are soluble in alcohol, while the albumin is rendered more compact by this reagent. Heller's Nitric Acid Test. This test is, perhaps, more frequently employed than any other of the tests for albumin in the urine. It has a very wide field of usefulness, although it is not as delicate as some of those to be mentioned. A few c.c. of concentrated nitric acid are placed in a test-tube and the urine to be tested is allowed to run slowly down the side of the tube in such a way as to form a distinct layer of urine above the acid. Some workers advise the addition of the acid after the urine, but if this is done it is much better practice to allow the nitric acid to flow from a pipet introduced to the bottom of the tube. The writer is always accustomed to allow the urine to flow upon the nitric acid from a long pipet so that the urine does not perceptibly mix with the acid, the tube being held at an angle of 45 degrees. Albumin, if present, is precipitated at the zone of contact in the form of a white opaque cloud or ring. This precipitate is acid-albumin which is insoluble in a slight excess of acid. 256 DIAGNOSTIC METHODS. A red or reddish-violet transparent ring is always obtained with normal urine owing to the reaction of the urinary pigments with the nitric acid. If the urine contains abnormal coloring matters this colored ring may assume various tints. Thus if bile be present a play of colors from red to yellow through blue or green takes place, the green being the characteristic coloration; if indican be in excess this colored ring may be bluish or even black; while pigments due to drugs will give colors ranging from deep red to violet This colored ring is usually below the white opaque ring due to albumin and tends to extend down into the acid instead of up into the urine. If much nitrous acid be present in the nitric acid effervescence may be observed to such an extent that the ring of albumin may be lost. This albumin ring is usually sharply defined and separated both from the urine and acid as a white opaque ring, whose breadth will depend upon the amount of albumin in the urine. If the tube be allowed to stand for some time the ring will lose its distinct outline, a more or less diffuse cloudiness rising throughout the urine. Albumin is, however, not the only substance to be precipitated by nitric acid under the conditions of this test. Thus we find globulin, albumoses, and Fig. 80 -Conical test-glass. Fig. 8i.-Horisma- scope. resins precipitated exactly at the line of contact of urine and acids. If the urine be heated the albumoses will dissolve while the albumin becomes more compact. If this precipitate be due to resins the precipitate will dissolve in alcohol or ether while the albumin will remain unchanged. It is sometimes wise to shake out the urine with ether before applying this test. The globulin ring may be differentiated from that of albumin only by separating these proteins by the method to be discussed under Serum Globulin. Globulin is usually associated with albumin, is practically never found by itself, and has practically the same clinical significance, so that for clinical purposes the differentiation of these two bodies is unnecessary. Weinberger1 has recently shown that the addition 1 Jour. A. M. A., vol. 52, 1909, p. 1310. THE URINE. 257 of thymol, as a preservative of the urine, leads to the formation of a grayish- white ring just at the junction of the nitric acid and urine when this test is applied. "Below the ring there is a greenish zone extending somewhat into the acid, above it a reddish somewhat smaller zone." If this substance is suspected, the urine should be extracted by agitation with an equal volume of petrolic ether. A somewhat similar reaction has been reported by Kenney 1 in cases in which several drops of formalin were added as a preservative to a small amount of urine. Besides these rings at the zone of contact, a further white or yellowish ring may be observed at this point. This ring is found in urines which are especially rich in urea and appears as a distinctly crystalline ring due to the formation of urea nitrate. If the urine be previously diluted this ring does not appear. If an excess of uric acid be present in the urine, we observe, on allowing the tube to stand for a few minutes a distinct white ring in the urine about i to 2 cm. above the point of contact of the acid and urine. If the mucin-like bodies previously discussed are present in slight excess, a diffuse cloud appears throughout the urine if the fluids have been slightly shaken or a distinct ring i to 2 cm. above the albumin ring may be observed if the urine is carefully added to the acid. This ring is seen in practically all urines, is never at the point of contact, and does not appear to have any clinical significance. Instead of performing this test in a test-tube as described above, it may be done in a conical wine-glass, as recommended by Simon and Ogden, or may be employed as recommended by Boston. This latter worker uses a flat- pointed pipet into which is drawn from 1 to 2 inches of the urine to be tested. The exterior of the tube is then wiped perfectly dry and the pipet, with its upper end closed with the finger, is introduced to the bottom of a bottle con- taining pure nitric acid. By lessening the pressure of the finger the acid gradually flows up into the pipet forming a distinct line of contact between it and the urine. The same points mentioned above obtain with this test. This modification is very simple and may be recommended for general use. A further method of performing this test is the use of the horismascope (see cut). The urine is placed in the larger tube C and the nitric acid allowed to flow through the capillary tube A so that a distinct line of contact is observed. The use of this instrument frequently brings out much more clearly the albumin ring than do the other modifications. This test is frequently employed in combination with heat. Two methods are available, either one of which may be used, although the results are some- what different in the two cases. Some workers advise heating the nitric acid previous to the addition of the urine. This does not seem wise to the writer, as traces of the acid albumin are undoubtedly dissolved by the hot acid and may thus escape detection. The writer is accustomed to heat only the upper portion of the tube which contains the somewhat diluted urine. In this way the urates 'New York Med. Jour., vol. 80, 1904, p. 403. 258 DIAGNOSTIC METHODS. and albumoses are thrown out of the field of action while there is much less danger of traces of albumin being dissolved. If the various points included by this test are remembered and each interfering substance removed by proper differentiation, this test is practically the most useful and reliable one for the detection of albumin in the urine. Potassium Ferrocyanid Test. A few c.c. of urine are strongly acidified with acetic acid and a few drops of a 10 per cent, solution of potassium ferrocyanid are added drop by drop. In the presence of albumin a faint turbidity or a flocculent precipitate will be observed, depending upon the amount of albumin present. The slow addition of the ferrocyanid is necessary, as an excess of this reagent will dissolve the precipitate first formed. In this test albumoses and nucleo-albumin are both precipitated. The former is dissolved on heating the mixture, while the latter is detected by the precipitate on the addition of the acetic acid unless the acid be added in excess. The urates do not interfere if the urine is diluted previous to making the test. This test is recommended by many as a much more delicate one for albumin than those previously described. It reacts with a smaller amount of albumin, but it does not, in the writer's opinion, give as much general information about the urine as does Heller's test. Sulpho-salicylic Acid Test. This substance may be used either in the form of a 20 per cent, solution or in the solid state. If the solution be used it is added to the acidified urine in such a way that a distinct line of contact is formed. Albumin will be shown by a distinct white ring at the point of contact. It is preferable, however, to add to the urine a small fragment of this substance, when in the presence of albumin a turbidity or a white flocculent precipitate will be observed, depending upon the amount of albumin. The albumoses are also precipitated by this reagent, but dissolve on heating. Neither uric acid nor resins react with this substance, while the mucin-like substances are not appreciably affected. This test is, perhaps, the most convenient one for the use of the general practitioner, as the substance may be readily carried in the medicine case and added to the urine at the bedside. Spiegler's Test. As the original reagent of Spiegler was found to be of little value in many cases, Jolies1 has modified it with the following composition: io grams of mer- curic chlorid, 20 grams of citric acid, and 20 grams of sodium chlorid dis- solved in 500 c.c. of water. The urine is acidified with acetic acid to precipitate the nucleo-albumin if present. This substance is filtered off and the filtrate superimposed by means of a pipet upon a few c.c. of the above reagent. In the presence of albumin a distinct white ring appears at the zone of contact. This reagent precipitates 1 Zeitsch. f. physiol. Chem., Bd. 81, 1912, S. 205. THE URINE. 259 the albumoses which are soluble on heating. Nucleo-albumin, if present, should be removed before applying the test. In case iodids are present, mercuric iodid will be precipitated but may be removed by alcohol. This test is the most delicate test for albumin. It shows one part of albu- min in 350,000 parts of urine. This is almost too delicate for clinical work as it will show albumin in practically every specimen of urine. The writer has found this test extremely valuable in cases which showed only a faint reaction with Heller's test. If such an urine be treated with Spiegler's reagent a much more distinct albumin ring will be observed so that all doubt is thus cleared up. Many other tests have been advocated, but the writer does not feel that they have any advantages over those outlined above. As far as delicacy of reaction is concerned Spiegler's reagent is the most delicate, the sulphosalicylic acid, heat and acid, ferrocyanid, and Heller's test following in the order named. It will thus be seen that Heller's test is the least delicate of any of the ones spoken of above, but the writer is accustomed to use it in general work, relying upon Spiegler's reagent to settle mooted points of delicacy of reaction. As a general working rule it should be said that no one of the less delicate tests should be relied upon without being confirmed by some one of the others. Quantitative Determination of Albumin. Scherer's Method. Fifty c.c. of urine are placed in a beaker and heated upon a water-bath. Two or three drops of dilute acetic acid are then added and the mixture boiled. A flocculent precipitate of albumin should separate out; if not, a drop or two more of acid is added until such a precipitate is obtained. The solution is then filtered through an ash-free filter which has been previously dried and weighed. The filtrate should be tested by Spiegler's reagent to see if any albumin has been dissolved. If this shows no albumin, the precipitate may then be dried on the filter-paper after being washed with water, alcohol, and ether. If the filtrate shows albumin, another test with a fresh 50 c.c. of urine must be made. Much time is saved if a larger quantity of urine be originally taken, treated as above, and small portions filtered off and tested for albumin. The addition of a small amount of a saturated solution of sodium chlorid will facilitate the precipitation of the albumin. The precipitate on the filter-paper is dried and weighed, the difference in weight between the original dried paper and that with albumin representing the amount of albumin in the 50 c.c. of urine used. The total amount is determined by a simple calculation. Esbach's Method. This test is carried out in a standard graduated glass tube, known as an albuminometer (see cut). This tube is filled with acidified urine to the point U and the Esbach reagent added to the mark R. This reagent consists of a solution of io grams of picric acid and 20 grams of citric acid in one liter of distilled water. The tube is now closed with the rubber stopper and inverted several times in order to mix thoroughly the contents. It is then allowed to stand in a test-tube rack for 24 hours, after which the amount of albumin is- 260 DIAGNOSTIC METHODS. read off, the graduations on the tube representing the number of grams of albumin per liter of urine. This test has many fallacies and can give only a very approximate determina- tion of albumin. The precipitate does not settle evenly. If the albumin reads more than 4 grams per liter, the urine must be diluted to a specific gravity between 1006 and 1008 and must be kept at constant acidity. If the urine con- tains less than 1/2 gram of albumin per liter the test is absolutely useless. .Further the room temperature must be kept constant at about 150 C. or errors as high as 100 per cent, may occur. More- over, albumoses, uric acid, creatinin, resinous acids, etc., are precipi- tated by this reagent. This test should, therefore, rarely be used, not even for comparative purposes. This method1 is a modification of the above, the reagent oeing a solution of phosphotungstic acid in acidulated alcohol with the following composition: Tsuchiya's Method. Phosphotungstic acid, i. 5 grams Concentrated HC1, 5.0 c.c. 95 per cent, alcohol, q. s. ad., 100.0 c.c. This reagent is used with the Esbach tube as in the above test or in the Purdy centrifuge tube discussed later, giving far more accurate results than the Esbach test. Normal urine gives only a faint unreadable precipitate, while small amounts of albumin as well as large ones are completely precipitated by it. This albu- minous precipitate settles regularly and quickly, foaming or floating of the precipitate being rarely seen. Variations in temperature have little influence on this precipitation. It is to be remembered, however, that other urinary albuminous substances will be also pre- cipitated. Mattice2 has shown that this method yields fairly accu- rate results for comparative purposes, the figures agreeing closely with those of the gravimetric methods. He has also demonstrated that the modification of this test, as introduced by Goodman and Stern,3 is nether accurate nor satisfactory, largely because the point at which albumin is shown by the first turbidity with this reagent (and hence the amount of albumin reacting) varies markedly with the dilution of the urine. The writer advises, therefore, that this method of Tsuchiya be used for reliable comparative clinical results, the gravimetric or Kjeldahl methods being applied for scientific purposes. In case the latter method be used, multiply the nitrogen values of the precipitated and washed albumin by 6.3 to obtain the albumin values. Fig. 82.- Esbach's al- buminometer, 1 Zentralbl. f. inn. Med., Bd. 29, 1908, S. 105. 2 Arch. f. Int. Med., vol. 5, 1910, p. 313. 3 Jour. A. M. A., vol. 51, 1908, p. 2055. THE URINE. 261 It is perhaps unnecessary to state at this point that in speaking of per cent, of albumin, one should have reference only to the number of grams of albumin by weight in 100 c.c. of urine. It is not infrequent to hear of case histories in which albumin has been reported as 50 to 75 per cent. This can mean only per cent, by volume and harks back to the time when the amount of precipitate formed on boiling the urine was taken as the quantitative criterion. The urine very rarely contains more than 5 per cent, of albumin, although Salkowski has reported a case in which 8 per cent, was observed, the albumin separating out as a white amorphous precipitate on standing. This fact should be remembered, as the appearance of such a precipitate in the untreated urine might be very misleading. Purdy's Centrifugal Method. To io c.c. of the urine placed in a centrifuge tube, 3 c.c. of a 10 per cent, solution of potassium ferrocyanid and 2 c.c. of 50 per cent, acetic acid are added. The reagents and urine are then mixed by placing the thumb over the end of the tube and inverting it several times, after which the tube is allowed to stand for 10 minutes. It is then placed in a centrifuge the radius of which, with its tubes extended, must be 6 3/4 inches. The tubes are revolved for exactly three minutes at a uniform speed of 1,500 revolutions per minute. The amount of albumin is then read off in bulk percentage, each division of the tube representing 1 per cent., as only 10 c.c. of urine are used and the divisions represent tenths of a c.c. One per cent, by bulk represents 0.021 per cent, by weight of albumin. This method is very satisfactory, although not absolutely accurate. It is difficult to keep a centrifuge running uniformly at the above rate, unless a speed indicator be watched during the entire period. The method is more exact and more expeditious than the Esbach method and is to be recommended, therefore, for clinical estimations of the amount of albumin. It is advisable in many of the general quantitative tests applied to the urine that the albumin should be removed if it is present in more than traces. Usually this may be done by acidifying with acetic acid and boiling until the precipitate is flocculent. The filtrate in such cases will usually be clear and contain no albumin. As this test does not eliminate the albumoses and further as the boiling acid may hydrolyze a small amount of the albumin into albumoses, Hofmeister recommends the following method. Ten c.c. of a 40 per cent, solution of sodium acetate and the same amount of 10 per cent, ferric chlorid are added to the urine when it will be colored a bright red. The urine is neutralized or rendered very faintly acid and is then boiled. The albumin separates out along with the basic ferric acetate and is filtered off. This method is not applicable if glucose is present. Removal of Albumin. (&). Serum Globulin. Serum globulin is associated in the blood with serum albumin. This 262 DIAGNOSTIC METHODS. protein is not a single body, but is probably a mixture of two forms known as euglobulin and pseudo-globulin. These two fractions differ in their precipita- tion and solubility constants and must be looked for in works on physiologic chemistry. The former of these probably occurs in most urines, constituting a large part of what has been called "nucleo-albumin" (see above). Serum globulin, in the specific sense, is usually found in the urine in every case in which serum albumin is observed. Cases are reported in which each one of these protein bodies has appeared separately, but this is not the normal finding. Its excretion as compared with that of serum albumin varies from io to 75 per cent, of the total protein. The relation of the albumin to the globulin of the blood is as 1.5 is to 1. This relation, known as the "albumin quotient," is by no means the same in the urine. In cases of amyloid degeneration of the kidney the globulin may be very much increased beyond that observed in other chronic affections of the kidney, the albumin quotient being usually lower than one. Senator considers this an important point in the diagnosis of this condition. In the various types of nephritis we find the globulin being greater the more acute the condition, so that in acute diffuse nephritis the albumin quotient may be very low, while in chronic parenchymat- ous nephritis the quotient ranges from two and five-tenths to five and five-tenths. As the nephritis improves the relative amount of globulin diminishes, but increases with each acute exacerbation. Globulin is insoluble in water, but soluble in dilute solutions of sodium chlorid, dilute acids, or alkalies, unless these be exceedingly dilute and their action not prolonged. If, therefore, urine containing globulins be highly diluted with water the globulin will be precipitated in the form of a distinct cloud. The precipitation constants with various inorganic salts must be learned from text-books on physiologic chemistry. The urine is rendered alkaline by the addition of a few drops of ammonium hydrate and the precipitated phosphates filtered off. To the filtrate is then added an equal volume of a saturated solution of ammonium sulphate, the mixture is allowed to stand one hour and is then filtered. The albumoses and nucleo-albumin may also be precipitated in this way. Ammonium urate does not usually separate out in the hour. The precipitate on the filter is then washed with a half-saturated solution of ammonium sulphate until the filtrate is albumin-free. A distinct precipitate is usually evidence of the presence of globulin, as the albumin is not precipitated until the urine is completely satu- rated with the ammonium sulphate. In order to eliminate the other factors the precipitate is dissolved in water and heated on a water-bath to coagulate the proteins. The solution is filtered, the precipitate washed with water and heated on a water-bath with a i per cent, solution of sodium carbonate. It is then filtered and neutralized with acetic acid. If a precipitate occurs it is globulin, as the albumoses and nucleo-albumin would not be precipitated by such treatment. Paton advises the use of a contact method in detecting globulin. The Qualitative Test for Globulin. THE URINE. 263 phosphates are removed as mentioned above and the filtered urine allowed to run down the side of a test-tube containing a few c.c. of a saturated solution of sodium sulphate. A white ring will indicate the presence of globulin. The phosphates are removed as above and 100 c.c. of the clear filtered urine are treated with an equal volume of a saturated solution of ammonium sulphate or directly saturated with magnesium sulphate. The precipitated globulin is collected on a dried and weighed filter and washed with a half- saturated solution of ammonium sulphate in the former instance or with a saturated solution of magnesium sulphate in the latter case. The final washings should show no trace of a reaction for albumin. The funnel with the filter- paper and its contents are then dried at no0 C. The ammonium sulphate is washed from the precipitate with hot water and the precipitate is then dried with alcohol, ether, and finally at no0 C., until the weight becomes constant. The difference between the weight of the filter-paper and the filter-paper plus the globulin gives the amount of globulin in 100 c.c. of urine. Quantitative Determination. These are intermediate products of the digestion of protein by ferments, by acids, or by bacteria. In the normal digestion we find the products passing through the following stages: protein, acid albumin, primary proteoses (of which there are two, namely, protalbumose and heteroalbumose), second- ary proteoses (the only well-established representative being deuteroalbumose), peptone, amino-acids and hexone bases. The proteoses and peptones are very soluble diffusible bodies which are not coagulated by heating. The primary proteoses are precipitated by half-saturation with ammonium sulphate, the secondary proteoses are precipitated only after complete saturation with this salt, while the peptones are not precipitable in either one of these ways. The primary proteoses are precipitated by nitric acid, thus differing from the secondary types which are not so precipitated. In the urine we find repre- sentatives of both types of proteoses, while more or less doubt exists as to whether true peptone has ever been isolated from the urine. (c). Proteoses. Primary Proteoses. This body was first discovered by Bence-Jones1 and was regarded as a heteroalbumose. Recent work by Magnus-Levy2 has shown that it is in all probability a true albumin, as its digestion products include protalbumose which could hardly be derived from true heteroalbumose. As the exact chemical nature of this body has not been definitely settled, it will be discussed under the above heading, although it probably does not properly belong there. This body differs from all other types of protein material which occur in the urine in its property of precipitating when heated to as low a temperature as Bence-Jones' Protein. 'Med. and Chir. Trans., vol. 33, 1850; Phil. Trans. Royal Soc., vol. 1, 1848, p. 55. 2Zeitsch. f. physiol. Chern., Bd. 30, 1900, S. 200. 264 DIAGNOSTIC METHODS. 40° C. and of practically completely dissolving on boiling, to appear again on cooling. A second characteristic of this body is the readiness with which it dissolves in dilute ammonia after it has been precipitated with alcohol. The excretion of this body in the urine has been called "heteroalbumosuria," "myelopathic albumosuria of Bradshaw," "Kahler's disease" and "Bence- Jones' albumosuria." The amount of this protein body excreted is somewhat variable. In Bence-Jones' original case an output of 6.7 per cent, or a total amount of 70 grams in the 24 hours was observed, while Coriat1 reports a case in which none was found in the urine, although 4 per cent, was present in the pleuritic effusion. Between these limits we find the majority of cases showing an output usually not over 1 per cent. The literature contains about 35 cases showing the excre- tion of this body. The output of this body appears to be constant during.the day and not affected in any way by the diet. Little is known regarding the direct origin of this body.2 It undoubtedly has some association with the bone- marrow, but just what is not clear. "We may imagine, however, that through the agency of the cells of the abnormal tissue, that is their products of metabo- lism, the normal transformation of the ingested albumin into tissue-albumin is impeded, resulting in the production of the substance in question, which is then eliminated as foreign matter" (Simon). This body is excreted in cases associated writh the occurrence of multiple myelomata of the bones, especially when these affect the thoracic skeleton. In only one other case, namely, one of lymphatic leukemia, has this body been reported, so that a urinary finding may be regarded as practically pathogno- monic of multiple myelomata. Boggs and Guthrie (Amer. Jour, of Med. Sci., vol. 144, 1912, p. 803, and Bull, of Johns Hopk. Hosp., vol. 23, 1912, p. 353) report the unique finding of this protein in a case associated with metastatic carcinoma. In some cases of this disease the urine does not show the Bence- Jones body, so that a negative finding does not necessarily preclude the condi- tion. Ellinger has shown that this disease may take its course without the occurrence of local bone symptoms but may be associated with a marked anemia. It is, therefore, wise in cases of obscure anemia to test the urine repeatedly for this body. The specific reaction for this protein is observed on heating the acidified urine very slowly. At a temperature varying from 500 to 6o° a slight cloud changing to a marked turbidity and then into a dense cloud will be observed. This may be so intense that the urine appears distinctly milky. This turbidity may change into a heavy sticky precipitate or coagulum as the temperature approaches the boiling-point. When the boiling-point is reached the pre- cipitate entirely or partially dissolves, especially if the boiling be continued from Tests for Bence-Jones' Body. 1 Amer. Jour. Med. Sci., vol. 126, 1903, p. 631. 2 Rosenbloom (Arch, of Int. Med., vol. 9, 1912, pp. 236 and 255) advances the possibility of osseo-albumoid as a precursor of this protein; see, also, Cathcart and Henderson, Jour, of Path, and Bact., vol. 17, 1912, p. 238. THE URINE. 265 one to three minutes. The precipitate may not absolutely all dissolve on boil- ing, as variations in the acidity of the urine and the amount of mineral salts present may affect this process. If the tube be allowed to stand after being boiled the precipitate returns as the fluid cools. None of the other protein bodies give this sequence of precipitation, dissolving, and reprecipitation. Hugounenq suggests the name "thermolytic albtinvimiria" for the excretion of this body in the urine. Upon the addition of concentrated nitric acid, drop by drop, a temporary turbidity develops which disappears on shaking, but persists if more acid be added. If the mixture be heated the precipitate will dissolve and reappear on cooling. This same reaction may be observed on applying any of the tests for serum-albumin outlined above. This protein is precipitated from its solution by the addition of two volumes of a saturated solution of sodium chlorid to urine which has been previously acidified with acetic acid. The addition of two volumes of saturated solution of ammonium sulphate likewise causes its complete precipitation. It may be then washed with alcohol and ether and dried over sulphuric acid. Boston1 has proposed the following test for this body. Fifteen to twenty c.c. of filtered urine are placed in a test-tube and mixed with an equal volume of a saturated solution of sodium chlorid, the tube being shaken to insure a thorough mixing of the fluids. Two or three c.c. of a 30 per cent, solution of sodium hydrate are added and the mixture vigorously shaken. The upper one-fourth of the mixture is then gradually heated to the boiling-point and a solution of 10 per cent, lead acetate added drop by drop, the heating being continued after each addition. When the drop of lead solution comes in contact with the liquid a copious pearly or creamy cloud appears at the surface, becoming less dense as the boiling-point is neared; and when ebullition is prolonged for from one-half to one minute the upper portion of the liquid shows slight browning, which deepens to a dull black color. Standing intensifies the reaction, and if this be prolonged for several hours the black precipitate falls through the clear stratum of liquid, collecting in the bottom of the tube as a coarsely granular pigment. This reaction is based upon the fact that this body contains a large pro- portion of loosely-bound sulphur. Lindemann2 finds that this body really contains no more such sulphur than does serum albumin, while Wood could obtain no more blackening than with other proteins. This test would seem to have little value as it is in no way distinctive for the Bence-Jones protein. The quantitative determination of this body may be made by precipitation with two volumes of saturated solution of ammonium sulphate and the washing and drying of this precipitate as previously mentioned. The Tsuchiya method is useful as an approximate estimation in the absence of albumin. Secondary Proteoses. 1 Clinical Diagnosis, Philadelphia, 1905. 2 Deutsch. Archiv. f. klin. Med., Bd. 81, 1904, S. 114. Deuteroalbumose is probably the body which has been found in the urine 266 DIAGNOSTIC METHODS. in cases in which peptone was reported. This secondary proteose differs in its reactions from the primary proteoses and the Bence-Jones protein. It is precipitated only on complete saturation of the urine with ammonium sulphate. In performing this test the urine should be made albumin-free, preferably by the Hofmeister method previously discussed. Nucleo-albumin may be precipi- tated by basic lead acetate. Urine containing deuteroalbumose does not become cloudy on boiling; does not regularly give Heller's test, but does react with the ferrocyanid test when the neutral salts are present in fairly large quantities, and reacts in the cold with sulpho-salicylic acid and with Spiegler's reagent, but the precipitate dissolves on heating to reappear again on cooling. This reaction with the latter reagents in the cold may be distinguished from that of albumin by boiling the mixture and filtering while hot. The albumin remains on the filter while the albumose is present in the filtrate, appearing as a distinct precipitate as the solution cools. Tests for Albumoses. To a few c.c. of urine, from which albumin and nucleo-albumin have been removed as outlined above, add one-fifth its volume of concentrated acetic acid. A 10 per cent, solution of phosphotungstic acid is then added, when the urine remains clear on standing-if albumoses are absent; while a milky turbidity is observed in 5 to 10 minutes if these substances are present. If this precipitate be filtered off (warming will facilitate the clumping of the precipitate), washed with distilled water and then dissolved on the filter with a very dilute solution of sodium hydrate, the solution will have a distinctly blue color. This solution is warmed until it becomes clear, more sodium hydrate being added if necessary. It is then cooled and the biuret test applied by adding a few drops of strong sodium hydrate and a few drops of dilute (2 per cent.) copper sulphate solution. On warming this mixture a beautiful-red color will be observed. Fittipaldi1 advises that 10 c.c. of urine be treated with 60 c.c. of absolute alcohol and allowed to stand until the next day. The alcohol is carefully poured off, the precipitate is dissolved in the least possible amount of 30 per cent. NaOH and this alkaline solution is treated with a few drops of a freshly prepared ammoniacal nickel solution.2 In the presence of albumose or peptone an orange-red coloration appears. This test seems reliable and does not react with albumin. Bang's Method. Ten c.c. of urine are heated in a test-tube with 8 grams of finely powdered ammonium sulphate until the salt has been dissolved. The mixture is boiled for a few seconds and is then centrifugalized for one-half to one minute. The supernatant fluid is then poured off and the precipitate extracted with alcohol to remove urobilin. After pouring off the alcohol the residue is dis- solved in a little water, the solution is boiled to remove albumin and filtered. The filtrate is shaken out with chloroform to remove any traces of urobilin 1 Deutsch, med. Wochs., Bd. 37, 1911, S. 1890. Gazz. d. osp., Tm. 32, 1911, p. 515. 2 A 5 per cent, solution of nickel sulphate mixed with an equal volume of ammonia water. THE URINE. 267 which may have escaped previously. The watery solution is then poured off from the chloroform and tested as above for the biuret reaction. Clinical Significance. Deuteroalbumose may occur in the urine either alone or associated with albumin. It is observed in a great variety of conditions, so that distinct types of albumosuria may be noted. Large accumulations of pus anywhere in the system lead to the excretion of deuteroalbumose as a result of the breaking down of the pus cells and the later absorption of the hydrolyzed material. This form, known as pyogenic albumo- suria, is observed in pneumonia during the stage of resolution, in gangrenous processes anywhere in the system, in empyema, bronchiectasis, abscess for- mation, and in epidemic cerebrospinal meningitis. In this latter condition the differential diagnosis from a tubercular meningitis often rests on the appear- ance of albumose in the urine. A hepatogenous form of albumosuria occurs in any condition associated with marked disturbance of hepatic function, as for instance in acute yellow atrophy, phosphorus poisoning, cirrhosis, carcinoma, and catarrhal jaundice. Little is known of the origin of the albumose in this condition. An enterogenous type is observed in cases of gastric or intestinal ulcer, whether the latter be due to typhoid fever or dysentery, while intestinal tuber- culosis is less frequently associated with the appearance of albumose in the urine. In these cases the breaking down of the tissues may be responsible for increased absorption both of the products of hydrolysis of the tissues and of the food. An albumosuria of hematogenous origin has been observed in cases of scurvy, leukemia, purpura, dermatitis, poisoning with hemolytic agents, preg- nancy, especially after the death of the fetus, and in various psychoses, as well as in carcinoma affecting any part of the system. The albumosuria in these cases is probably referable to the increased lysis of the cells under the influence of the exogenous or endogenous toxins. A febrile type is observed in practically all fevers, more especially the infectious types, such as measles, scarlet fever, diphtheria, acute articular rheu- matism, smallpox, and mumps. This is referable both to the influence of the toxins of the disease in producing increased protein disintegration as well as to the associated septic conditions. A large number of other conditions are associated with the appearance of albumose in the urine. Such a state is due to the breaking down either of tissue or of an exudate, and may, therefore, appear in almost any type of disease. In some cases albumose may appear in the urine, following the ingestion of a large amount of albumose. This is the digestive or alimentary albumosuria and appears to be indicative of an ulcerative condition somewhere along the intestinal tract. As previously stated, albumose may be associated in the urine with albumin, constituting the mixed albuminuria of Senator. In these cases the albumosuria 268 DIAGNOSTIC METHODS. may precede the albuminuria, may alternate with it, or continue after it has disappeared. This condition is particularly prominent in cases of nephritis, especially of the syphilitic type, and should be watched with care. In any case of albumosuria it is necessary to exclude contaminations with foreign material, especially with spermatic or prostatic secretions. (d). Peptone. True peptone rarely if ever appears in the urine. Peptones are the last hydrolytic products of protein which give the biuret reaction. This body is not precipitated on saturation with ammonium sulphate as are the other types of protein material. Ito reports the finding of true peptones in the urine in cases of croupous pneumonia, pulmonary tuberculosis, ulcer of the stomach and in women during the puerperal period. In these cases deuteroalbumose was also present, so that there is a possibility that the peptone isolated by Ito was derived from the urinary albumose. Many reports of peptonuria are found in the literature, but the substance dealt with in practically all of these cases was probably some type of albumose and not true peptone. As this substance has no clinical value, the writer must refer to works on physiologic chemistry for a discussion of its properties and tests. (e). Hemoglobin. This body is the normal coloring matter of the blood and is to be regarded from the chemical standpoint as a chromoprotein. In the normal metabolism disintegration of red blood-corpuscles is constantly occurring, but this is not suf- ficient to lead to a hemoglobinemia and a resulting hemoglobinuria. These two conditions must go hand in hand, the latter being impossible without the former. When the destruction of the red cell becomes so extensive that the accu- mulation of the blood pigment in the blood-current is so great that the liver is unable to convert it into bilirubin, hemoglobinemia and hemoglobinuria must result. While the distinct limit of destruction of cells necessary to produce this condition is not definitely settled, it may in general be said to occur when approximately one-sixtieth of the hemoglobin of the corpuscles is set free. The protein really excreted in the urine is not hemoglobin, but methemoglobin, so that the term methemoglobinuria would be better used, although a direct hemoglobinemia does obtain. From what has been said above it is evident that the excretion of this protein will inevitably occur after the use of the so-called hemolytic poisons. Among these we find ether, chloroform, snake-venom, arseniuretted hydro- gen, phosphorus, hydrogen sulphid, toluylendiamin, mushrooms, anilin, lacto- phenin, bile salts, chlorates, pyrogallic acid, naphthol, carbolic acid, carbon monoxid, and tuberculin. Hemoglobin will appear in the urine in cases of poisoning with the above substances only when the hemoglobinemia is extensive. In mild cases, the liver will be called upon to form increased biliary pigment and the urine will, therefore, contain bile pigments instead of blood pigments. Likewise, we find a hemoglobinuria following transfusion of the blood of ani- THE URINE. 269 mals into man, after severe burns, exposure to cold, in the course of any of the specific infectious diseases, and in malaria1 and syphilis. The so-called " black- water fever" is more probably a malarial hematuria than a hemoglobinuria. The use of quinin in malaria is said in some instances to lead to a hemoglo- binuria, but the writer has never been fortunate enough to see such a case. A paroxysmal type of hemoglobinuria has been occasionally reported in the literature. This occurs in typical paroxysmal forms after exposure to cold or exertion, and is often preceded by a typical "infectious" onset, such as chill, fever, and malaise, along with pain in the lumbar region. The hemo- globin may be excreted for several days and then disappear with no untoward symptoms. This condition is very rare and its cause uncertain.2 An epidemic hemoglobinuria occurs at times in the new-born and is associated with a distinct hemoglobinemia, jaundice, and cyanosis. Some unknown toxic agent is at the bottom of this condition. The urine, in cases of hemoglobinuria, may be clear, but is generally turbid, and varies in color from a bright red to almost a black. The turbidity gives the appearance of a peculiar smoky or hazy urine. The urine must be ex- amined when freshly voided, as blood-corpuscles soon disintegrate in the urine, giving it the same appearance as noted in hemoglobinuria. The clinical significance of hematuria and hemoglobinuria are much different and should not be confounded. If the urine be centrifuged the supernatant fluid will be a clear blood-colored liquid and the sediment will show none or very few red cells. The tests for the presence of blood pigments must be applied in order to differentiate this protein from the other types. Naturally, any specimen of urine containing hemoglobin will react to the albumin tests previously given so that it may be very difficult to determine whether a true albuminuria is coex- istent. As a rule, in such conditions there is an associated nephritis so that all the findings of this latter condition may obtain. The chemical tests indicate the presence of hemoglobin or of any of its derivatives and do not differentiate a hemoglobinuria from a hematuria. Microscopic examination for the presence of red blood-cells is the only possible way of clearing up such a diagnosis. A spectros'copic examination of the urine will differentiate the types of blood pigments (see Blood). Tests for Hemoglobin and Derivatives. Heller's Test. A few c.c. of urine are strongly alkalinized with sodium hydrate and heated. Either at once or on standing a brownish-red precipitate of the phosphates and the carbonates of the alkaline-earths is formed, the color being due to the hematin carried down by the phosphates if blood is present. If the urine contains a large amount of foreign pigments, this red coloration may not be easily noted. In this case filter off the precipitate and dissolve it in Arriola (Interstate Med. Jour., vol. 19, 1912, p. 74) claims that the black urinary pig- ment of the urine is pathognomonic of malaria. 2 According to Cooke (Amer. Jour, of Med. Sci., vol. 144, 1912, p. 203) syphilis is the most important, if not the only, etiologic factor in paroxysmal hemoglobinuria. 270 DIAGNOSTIC METHODS. acetic acid, when the solution becomes red if blood pigment is present, the color gradually fading upon exposure to air. If this test be controlled by the spectroscopic tests for hematin in alkaline solution it becomes quite reliable and very delicate. It indicates one part of oxyhemoglobin in 4,000 of urine. Ten c.c. of urine are treated with i c.c. of ammonium sulphid solution and i c.c. of pyridin. If blood be present, the urine will assume a more or less intense orange color, which may be more evident on looking through the test-tube lengthwise. In this case the hemoglobin has been converted into hemochromogen, which may be recognized by the spectroscopic test. This test is more delicate than the previous, showing one part of blood to 8,000 of urine. Instead of the above tests, which arc more directly applied in urine, the guaiac and aloin tests as discussed under Feces may be applied. These tests have a special importance from the negative standpoint, as a positive test does not necessarily prove the presence of blood. They are more delicate than is the spectroscopic, but the latter is more reliable. The spectrum of the various blood pigments will be discussed later, so that the writer need only refer to the section on Blood for this. If the urine contains fresh blood the spectrum is that of oxyhemoglobin, while in cases of hemoglo- binuria or of hematuria of renal origin the spectrum is that of methemoglobin. The urine to be tested spectroscopically should be slightly acid and perfectly clear. If a large amount of blood pigment be present, the spectrum will be much clearer if the urine be diluted. This dilution should not be carried too far, otherwise the absorption lines will not appear. In testing for methemo- globin the spectrum of neutral as well as of alkaline methemoglobin should be looked for. Donogany's Test. (/). Fibrin. The occurrence of this protein in the urine is very rare. As fibrin is derived from fibrinogen through the action of the fibrin ferment, the presence of the former body presupposes that the latter two substances have been present somewhere along the genitourinary tract. This substance is an elastic, grayish, stringy material insoluble in water and alcohol. Chemically, fibrin belongs to the group of globulins; it is soluble with difficulty in dilute saline solutions, is coagulated by heat, and precipitated either by great dilution with water or by saturation with magnesium sulphate. This protein may occur in the urine either in the coagulated form or in solution. It is found in any condition in which large amounts of blood are present in the urine, whether the blood comes from the kidneys or points below. It may coagulate immediately after voiding or may occur as preformed clots which are formed either in severe inflammations of the pelvis of the kidney, of the ureter, bladder, or urethra. It occurs also in cases of chyluria and rarely in direct nephritis. THE URINE. 271 In some cases the fibrin is in solution, especially in urines containing no blood. This fibrin separates out in the form of a coagulum on standing or may change the urine into a distinctly gelatinous mass. This so-called "spon- taneously coagulable urine" is seen more frequently in cases of chyluria, but may be observed in rare cases of nephritis. Test for Fibrin. The clotted material is filtered off from the urine, is thoroughly washed with water and boiled in a i per cent, solution of sodium carbonate. On cooling this solution may be tested as outlined under Serum Albumin. Other protein bodies, such as histon and nucleo-histon, have been reported in the urine by several workers. It is also possible that protamin has been observed. As these substances are little understood, have no clinical value at present, and require more or less elaborate methods for their absolute identification, the writer must refer to other works for a discussion of them. (2). Carbohydrates. Normally the urine contains traces (0.01 to 0.03 per cent.) of carbohydrates which are incapable of detection by the ordinary clinical tests. Besides these true carbohydrates the urine contains other substances which react, especially toward copper solutions, as do the monosaccharides. These latter reducing bodies are uric acid, creatinin, conjugated glycuronic acids, and various pig- ments, either normal ones excreted in unusual amounts or abnormal ones excreted in usual amounts. The total output of the reducing bodies of the normal urine varies between 2 and 3 grams in 24 hours, while the normal true carbohydrates of the urine vary from 0.2 to 1 gram per diem. (a). Glucose (d-Glucose) (CH20H-(CH0H)4-CHO). The normal blood contains about one part per thousand of glucose. Whether this sugar is in the free state or in combination with other molecules, as for instance as the so-called jecorin of Drechsel, is at present an unsettled question. According to Claude Bernard, sugar will appear in the urine when- ever more than three parts per thousand are present in the circulating blood. This figure is, in view of the recent work of von Noorden, Stern, and Liefmann, much too high, as the average finding in their cases was 0.85 part per thousand. The excretion of sugar in the urine is known as glycosuria and presupposes an excess of sugar in the blood (hyperglycemia). In only one form of glyco- suria do we find absence of this hyperglycemia;that is, in conditions in which the kidneys become less impervious than normally to the sugar circulating in the blood. This type of glycosuria is most frequently observed in cases of poisoning with phloridzin, and has lead to the assumption of the clinical entity "renal diabetes mellitus." The pathology of such a condition is little understood so that we may for the present disregard this type and limit our discussion to the glycosuria which inevitably follows a hyperglycemia. Glycosuria. The normal metabolism is such that any excess of carbohydrate food is 272 DIAGNOSTIC METHODS. converted, up to a certain point, into glycogen and stored up in the liver. Should this ingestion of carbohydrates exceed the functional power of the liver to convert it into glycogen, the excess will pass through the hepatic fdter into the circulating blood, thus causing directly a hyperglycemia. Unless increased muscular activity is sufficient to utilize this excess, the kidneys will excrete sugar until the normal relations again obtain. This is the purely alimentary type of glycosuria and ceases as soon as the intake is diminished. The type of food ingested has much to do with the extent of the glycosuria. Under normal conditions it matters relatively little how much carbohydrate is ingested in the form of starch, as the products of hydrolysis are gradually absorbed and do not lead to overactivity of the liver with a resulting hyperglycemia and glycosuria. On the other hand, a certain limit, different for each individual, is observed in the amount of sugar which may be ingested without causing a glycosuria. For this reason Naunyn has regarded alimentary glycosuria as of two distinct types: (i) that following the ingestion of starch, which he styles glycosuria ex amylo, and (2) that following the ingestion of an excess of sugar, glycosuria e saccharo. The amount of starchy or of saccharine food which a person may ingest without a glycosuria is known as the assimilation limit or degree of tolerance for such food. This factor varies for each individual under normal conditions and under pathologic influences is dependent upon the state of the intestines, liver, pan- creas, muscles, and kidney. A normal person may stand an intake of from 200 to 300 grams of glucose without excreting more than traces in the urine, but in the majority of persons this figure would, perhaps, be found to be more nearly 150 grams than 300. It has been found that the administration of as small amounts as 50 grams of galactose and lactose was followed by an excretion of these sugars in the urine, while maltose, dextrose, levulose, and saccharose required much larger intakes. The excreted sugar, following the increased ingestion, is in most cases similar to that taken in, although Moritz has shown that some of the polysaccharides may be partially hydrolyzed into their mono- saccharide components. It has been found by Worm-Muller that a large intake of cane-sugar is not followed by a maximum excretion of this sugar, even though the assimilation limit has been greatly exceeded. Thus he observes, after an intake of 50 grams of cane-sugar, an excretion of 0.1 gram, while after an intake of 150 grams the excretion was only 0.85 gram. "Alimentary glycosuria occurs in a healthy person only by saturating the organism with soluble carbohydrates. Therefore it is absent after admin- istering starch, as in this case no more sugar will be absorbed than can be metabolized in the body. It is also scanty, or quite absent, if sugar solutions be given on a full instead of on an empty stomach. Naunyn has observed, regarding alimentary glycosuria, that there is excreted in the urine only that sugar which, according to Ginsberg, reaches the general circulation through the thoracic duct, thus avoiding the liver. Still other external influences may come into play, such as altered capacity of the tissues, especially those of the importantly concerned liver. The result of this may be that excessive doses THE URINE. 273 of glucose are stored up as glycogen or fat in a given time. For saccharosuria and lactosuria the relationships are somewhat different. Here it is very obvious that these double sugars, if given in excessive quantities, are not com- pletely split up in the intestine or during their passage through the intestinal wall, but enter the general circulation as such. The organism, like most of the yeasts, cannot decompose these sugars to any extent, so that they leave the body with the molecules unaffected" (Magnus-Levy). In testing for the pathologic type of alimentary glycosuria, it is customary to give 100 grams of cane-sugar or glucose either in the morning on an empty stomach or two hours after a very light breakfast. The urine of healthy persons should remain free from sugar, while a pathologically lowered limit of tolerance will be observed by a more or less extensive glycosuria beginning in about one hour, reaching a maximum within two to four hours, and lasting about 8 hours. If this glycosuria follows the administration of starchy foods, the condition is probably a pure diabetic one, while a glycosuria fol- lowing 100 grams of grape-sugar does not necessarily indicate a diabetes. It is probable that cases showing a somewhat low assimilation limit for sugar belong to the type of mild diabetes, although many class them in the general category of "hepatic insufficiency." That the liver is incapable in such con- ditions of polymerizing the sugar into glycogen cannot be disputed, but the question at issue is, as von Noorden shows, upon what does the insufficiency of the liver depend ? I quote from the article in his Handbuch der Pathologic des Stoffwechsels as follows: "The cause cannot be overfilling of the gly- cogen reservoir, by which we explain the alimentary glycosuria of the healthy. In most cases the subjects are in ill health and their previous diet anything but excessive, so that there is little reason to suppose that their glycogen repository was already filled to overflowing. There is no more reason for supposing that they have a diminished power of utilizing sugar. In none of the affections in question are the processes of oxidation and of energy production decreased; rather is there a great increase in oxidation and especially in the combustion of carbohydrate in certain of the diseases, such as Graves' dis- ease and high fever, which predispose to alimentary glycosuria. Either the liver-cells are unable to polymerize all the sugar reaching them or else the gly- cogen formed is too quickly converted into sugar, through some increase in the diastatic process. The latter would imply that the automatic regulation between sugar combustion and sugar-formation is no longer so evenly balanced as in health. Physiologically the diastatic process is dependent only on the sugar requirements of the body; here it would also be controlled by the sugar- supply." The idea of hepatic insufficiency as a distinct clinical entity responsible for the alimentary glycosuria can hardly hold. One should not be satisfied with a mere statement that such a condition exists, but should attempt to ex- plain why it does obtain. In many conditions of the liver in which an un- doubted insufficiency is present, no glycosuria can be caused by administration 274 DIAGNOSTIC METHODS. of a fairly large amount of glucose. On the other hand, administration of levu- lose in these conditions produces a distinct alimentary levulosuria. To the classes of cases which give this levulosuria should be, according to Strauss, applied the term "hepatic insufficiency" rather than to those showing an ali- mentary glycosuria. (See p. 293.) It would lead me too far afield to discuss the various factors which influence the appearance of sugar in the urine.1 In many cases a restriction of the diet within the assimilation limit will rid the urine of sugar, while in others a constant production of sugar within the system occurs. This endogenous production of sugar may again be of several types. Thus we find in nervous conditions, especially in affections in the region of the fourth ventricle, an ex- cretion of sugar which continues only so long as the glycogen of the body is not used up. If no carbohydrates be taken in the food, glycosuria will soon dis- appear. This condition is due to interference with the normal nervous con- trol of the glycogenic function of the liver. This type of glycosuria, to which the name "neurohepatogenous glycosuria" has been applied, is found in a variety of conditions, such as progressive paralysis, multiple sclerosis, cerebral tumors, peripheral neuritis, traumatic neuroses, mania, melancholia, and hyste- ria. In this type we find the administration of carbohydrates in the food being followed by a glycosuria, because the diastatic processes are continually urged to increased activity, to such an extent that the hepatic artery becomes loaded with sugar. In other words, we have, in this type, an increased rate of sac- charification rather than a diminished formation of glycogen. In other cases a continuous glycosuria is observed even after complete exclusion of carbohydrates from the diet. In this type there is a constant formation of carbohydrates from the protein and fat of the food. The liver appears to be capable of storing up glycogen in the usual amount, but the system in general is practically unable to utilize the sugar brought to it. In consequence of this the liver-cells are repeatedly called upon to convert the glycogen reserve into glucose, and as a result the blood becomes laden with sugar. Even under these circumstances the hyperglycemia would not neces- sarily cause a glycosuria, providing the normal ferments of the blood arising from the internal secretion of the pancreas were present. This type of glyco- suria is in reality typical diabetes mellitus and is so dependent upon such a large number of factors that the writer must refer to other works for its discussion. Recently Pfliiger has shown that a very close relationship exists between the nervous influences of the duodenum and pancreas. His work apparently in- dicates that what is generally known as pancreatic diabetes is dependent to some extent upon disturbance of the proper correlation between the duo- denum and pancreas. As so little is definitely known regarding the glycolytic ferment which the pancreas furnishes to the blood it would seem at present useless to speculate upon the probable influence of the duodenum in producing a typical diabetes mellitus. A peculiarity in the true diabetic glycosuria is that other types of sugar, especially levulose, appear to be well tolerated by the 1 See Woodyatt, Jour, of Biol. Chern., vol. 14, 1913, p. 38. THE URINE. 275 diabetic individual without leading to a glycosuria. Besides, the administra- tion of the primary hydrolytic products of glucose, as for instance gluconic and saccharic acids, apparently diminishes an already existing glycosuria. This would seem to indicate that the system is primarily unable to bring about the initial cleavage of the glucose molecule. As it has been shown that many diabetic cases are associated with sclerosis of the islands of Langerhans in the pancreas, this organ is usually regarded as the principal seat of the patho- logic changes in true diabetes mellitus. On the other hand, many cases which are clinically indistinguishable from those of pancreatic diabetes show no lesion in this organ postmortem nor any characteristic lesion in other organs. We are, therefore, confronted with the possibility that this disease is a representa- tive of a truly chemical type of pathologic perversion of metabolism rather than as one which yields gross or microscopic evidences of pathologic changes. Occasionally, cases are seen in which a diplomellituria exists. This is the contemporaneous or alternate occurrence of diabetic and non-diabetic glycosuria in the same individual (Stern1). The clearing up of one type usually reveals the other. A transient glycosuria may obtain in conditions in which the oxygen supply is reduced, as in suffocation, poisoning with carbon- monoxid, curare and amyl nitrite, or after administration of such drugs as strychnine, cocaine, caffeine and adrenalin.2 Further, a transient glycosuria is not infrequently seen following the use of alcohol, especially in the form of beer or champagne. A post-anesthetic glycosuria is observed after the use of chloroform and ether, probably as a result of the action of these agents in stimulating the transformation of glycogen into dextrose (Hawk3). Qualitative Tests for Glucose. The qualitative tests for sugar depend for the most part upon the chemical structure of its molecule. The hexoses belong either to the class of aldehyds or ketones, and as such will reduce metallic oxids to lower forms. The CHO and CO groups of the aldoses and ketoses, respectively, are the reacting points in all of the reduction tests, such as those with copper and bismuth solutions as well as in the tests showing the formation of the characteristic osazones. Moreover, these carbohydrates show the peculiarity of fermenting, in the presence of yeast, into alcohol, carbonic acid, and other products. This fermentation test, especially with the saccharomyces cerevisiae, is given only by the sugars having three or a multiple of three carbon atoms in the molecule. Fischer's work has shown that only those sugars may be fermented by a specific ferment in which the ferment and sugar stand to one another in such a relation that a chemical union is possible between them or as he expresses it, only when the ferment fits into the sugar molecule like a key in 1 Arch, of Diag., vol. 3, 1910, p. 236. 2 See Emerson (Jour. A. M. A., vol. 59, 1912, p. 2245) for a discussion of glycosuria in the insane. Also, Herbert, Biochem. Zeitsch., Bd. 48, 1913, S. 120. 3 Arch, of Int. Med., vol. 8, 1911, p. 39; see, also, King, Moyle and Haupt, Jour, of Exp. Med., vol. 16, 1912, p. 178. 276 DIAGNOSTIC METHODS. a lock. This explains why only certain types of the hexoses will ferment in the presence of yeast. Before any qualitative or quantitative test may be made for the pres- ence of sugar in the urine, albumin must be removed, especially if present in more than traces. This may be done by precipitation with lead acetate and filtering or boiling with dilute acetic acid and filtering. To a few c.c. of urine in a test-tube are added one-third its volume of a io per cent, solution of sodium hydrate and then, drop by drop, a io per cent, solution of copper sulphate. This copper sulphate should be added with con- stant shaking until a slight excess of the precipitated cupric hydrate (Cu(OH)2) remains undissolved and is visible, on shaking the tube, as a distinctly greenish- blue flocculent precipitate. The upper layer of the urine is then warmed, when a yellow or red precipitate appears in the heated urine if sugar be present. As a rule, it is unnecessary and even unwise to boil the solution, as otherwise substances other than sugar may produce the reaction. It is true that the re- action is not as sensitive unless the solution be heated to boiling, but other sub- stances do not so readily interfere with the reaction at a lower temperature. The yellow or red precipitate will gradually form throughout the mixture, from above downward, and will finally settle out, leaving a colorless or yellow fluid above. This reaction is due to the reduction of the cupric hydrate, which is formed by the action of the sodium hydrate upon the copper sulphate, into cuprous hydrate, which becomes dehydrated, on heating, into cuprous oxid. If no sugar be present in the urine and other reducing substances are not exces- sive in amount, the cupric hydrate will settle out on warming as a black pre- cipitate of cupric oxid. The equations showing these points are as follows: Trommer's Test. CuSO4+2NaOH = Na2SO4 + Cu(OH)2 Cu(OH)2 + heat= CuO4-H2O 2CuO+glucose (CHO- (CHOH)4 - CH2OH) = Cu2O + gluconic acid (COOH-(CHOH)4-CH2OH). Whether a yellow or a red precipitate forms will depend upon the alka- linity of the solution, the stronger the alkalinity the more pronounced is the red color due to cuprous oxid, while in less strong alkaline solutions the yellow color of cuprous hydrate will predominate. Certain substances normally present in the urine as well as some which may be added to it have the property of holding in solution the cupric hydrate first formed. This property is shown by the deep blue color which the solution assumes. This color is very intense in the presence of sugar, but it is unwise to assume the presence of sugar from this fact alone. Among the substances which dissolve cupric hydrate and which may be present in the urine in varying amounts we find ammonium compounds, albumin, uric acid, creatinin, allantoin, mucin, glucose, lactose, maltose, pyrocatechin, hydroquinon, alkapton acids, bile pigments, and gly- curonic acid. On warming the solution, which may not contain sufficient THE URINE. 277 amounts of glucose to give a typical reaction, a slight reduction of the copper solution will occur leading to the formation of a dirty yellow solution. If these bodies be present in excess a distinct precipitate may occur, and as a result marked confusion may arise regarding the presence of sugar.1 It has been shown that uric acid and creatinin do not readily reduce at as low a temperature as does sugar, so that slight warming is much better than boiling. The pres- ence of these bodies very frequently leads to a change in color from the bright blue to the greenish-yellow which may be due to the presence of sugar in small amounts, but should never be regarded as indicative of a pathological glycosuria. It should be a working rule, therefore, that mere decolorization of the fluid should not be regarded as due to sugar. Glucose reduces so much more markedly than the other bodies mentioned that a distinct granular pre- cipitate either of cuprous oxid or hydrate forms and settles out, leaving a supernatant fluid partially or completely decolorized. Normal urine or urine containing an excess of uric acid, creatinin, or ammonium compounds practically never produces an immediate precipitate unless the solution be boiled for some time. This is due to the fact that these substances hold in solution the amount of cuprous oxid which is formed by their reducing action. If such solutions be allowed to stand for a time a reddish-yellow precipitate may occur, but this should not confuse as the typical sugar reaction occurs im- mediately on warming unless too much copper solution be added. In case only traces of sugar be present, we may obtain little or no positive reaction because the sugar holds in solution the traces of cuprous oxid formed just as do the other substances above mentioned. As a rule, however, sugar in patho- logic amounts shows a reducing action over and beyond its dissolving action, so that a red or yellow precipitate must settle out. The limit of this test for sugar is about 0.2 per cent., in which case the reduction will occur as in normal urine without the separation of the characteristic copper precipitate. Even in this case the yellow color is somewhat more intense and is clearer than the dirty yellow color produced in the presence of excessive uric acid or creatinin. If the urine contain an excess of the conjugated glycuronates, the test may be quite as distinctive as that for sugar. To differentiate such reactions, one should resort either to the phenylhydrazin or the fermentation tests to be described later. Benedict's Test. This qualitative test2 is, in the writer's opinion, the best modification of Trommer's test. The reducing action of glucose in alkaline solution is dimin- ished by strong alkalies such as the hydrates. This property may prevent such well-known solutions as Fehling's and Haines' (discussed below) from demon- strating small quantities of sugar. Further, the urine contains many compounds which interfere with the detection of glucose by the strongly alkaline solutions, 1 See Schulz, Zeitsch. f. physiol. Chem., Bd. 77, 1912, S. 121; also, Salkowski, Ibid., Bd. 79, 1912, S. 164. 2 Jour. A. M. A., vol. 57, 1911, p. 1193. 278 DIAGNOSTIC METHODS. the loss of delicacy in the other tests being probably due to the fact that normal reduction of the solution is inhibited for a period long enough to allow the strong alkali to decompose the reducing substance. However, in solutions of the alkali carbonates the reduction develops more slowly and is not destroyed by this weak alkali. The addition of sodium citrate, instead of the Rochelle salt as used by Fehling, makes the solution very stable, although little more than does the glycerin of Haines' solution. For the reasons given, this solution of Benedict is more sensitive (about ten times) to urinary sugar than are the other copper solutions. Moreover, it is not appreciably reduced by creatinin and uric acid, nor by such preservatives as chloroform, chloral or formaldehyd, which introduce errors into the other tests. It is, however, promptly reduced by the alkapton acids and the conjugated glycuronic acids. The formula of the reagent is as follows: Copper sulphate (C. P. crystallized), 17.3 grams Sodium or potassium citrate, 173.0 grams Sodium carbonate (crystallized),1 200.0 grams Distilled water, q. s. ad., 1000.0 c.c. Dissolve the citrate and carbonate (with aid of heat) in about 700 c.c. of water and filter if necessary. Dissolve the CuSO4 in about 100 c.c. of water and pour into the alkaline solution. Cool and make up to 1 liter. The technic is as follows: 5 c.c. of the reagent are placed in a test-tube and not more than 8 or 10 drops of urine are added. The mixture is then heated to vigorous boiling for one to two minutes and allowed to cool spontaneously. In the presence of glucose, the entire body of the solution will be filled with a precipitate which may be red, yellow or greenish in tinge. If the quantity of glucose be low (under 0.3 per cent.) the precipitate forms only on cooling. If no sugar is present, the solution either remains perfectly clear or shows a faint turbidity, which is blue in color and consists of precipitated urates. This test may be especially recommended for all qualitative work, as it is very delicate and has few fallacies. Fehling's Test. This is, perhaps, the best-known test for sugar but it is inconvenient and frequently inaccurate for the reasons above mentioned. The formulae for the solutions are as follows: Solution A. Copper sulphate, 34-64 gm, Distilled water, q. s., ad., 500 c.c. Solution B. Rochelle salt, 173 gm. Sodium hydrate 50 gm. Distilled water, q. s., ad., 500 c.c. In performing the test with Fehling's solution equal parts of solutions A and B are taken and the mixture brought to a boil. The urine is then added, drop by drop, when a reduction of the copper solution will appear in the pres- 1 Or one-half the amount of the anhydrous salt may be used. THE URINE. 279 ence of sugar. The amount of urine added should rarely exceed 10 drops, at the outside 20, if a reduction is to be taken as typical and the solution should not be boiled for more than a few seconds after adding the urine. The addi- tion of larger volumes of urine will usually introduce errors from the factors mentioned under Trommer's test. This test has the same points of interest as has Trommer's test, so that it is wise to dilute the urine before making the test. It is more frequently used than is Trommer's test, but it is not as con- venient as two separate solutions must be measured out before the test can be applied. Haines' Test. Haines has introduced a modification of Trommer's test by adding glycerin, instead of Rochelle salt, to increase the amount of copper in solution. This test is much more convenient than Fehling's, the solution having the advantage of keeping almost indefinitely. It is, however, far less delicate than is Benedict's and is reduced by preservatives as well as by excess of many normal urinary constituents, especially by uric acid and creatinin. The composition of Haines' qualitative solution is as follows: Copper sulphate, 12 grams. Potassium hydrate, 45 grams Glycerin, 90 c.c. Water, q.s., ad., 1000 c.c. A perfectly clear, transparent, dark-blue liquid results which throws down a very slight reddish deposit of cuprous oxid on standing a week or more. This does not affect the value of the solution, as the clear blue solu- tion is simply decanted as required. Four or five c.c. of this solution are placed in a test-tube and gently boiled. Six drops of the suspected urine are added and the upper portion of the mix- ture brought to a boil and immediately removed from the flame. If sugar be present an abundant yellow or yellowish-red precipitate is thrown down; if no such precipitate occurs sugar is absent. The precautions to be observed in using this test are never to add at the outside more than 10 drops of urine and not to boil the mixture for more than one or two seconds after the addition of urine. Many other modifications of the copper test have been introduced, both for qualitative and quantitative purposes, but they do not have any advantages over those mentioned especially over Benedict's test. This latter is used daily in the writer's laboratory and has been found very reliable and serviceable as it does not react with the usual preservatives which are added to urine. When the other tests are used an error may creep into the report which can be corrected only by resort to fermentation and other confirmatory methods. This test is a modification of the original Bbttger test and is a distinct improvement. The reagent is prepared as follows: Four grams of Rochelle Almen-Nylander's Test. 280 DIAGNOSTIC METHODS. salt are dissolved in 100 c.c. of xo per cent, sodium hydrate solution with gentle heat, and as much bismuth subnitrate is added as will dissolve (about 2 grams). After the mixture is cooled the undissolved bismuth subnitrate is filtered off and the filtrate kept in a dark bottle, where it will remain permanent for a long period. To a few c.c. of urine in a test-tube is added one-tenth of the volume of this reagent and the mixture boiled for a few minutes. If glucose is present, the fluid will darken and a black precipitate of metallic bismuth separate out. This black precipitate must occur while the solution is being warmed and not after it has cooled. If only a small amount of sugar be present the phos- phates precipitated by the alkali may be slightly gray in color instead of the usual white. This test is somewhat more delicate than is the copper test, as it will show about 1/40 per cent, of sugar. This reagent is not reduced by uric acid, creatinin, pyrocatechin, hydro- quinon or homogentisic acid; but it is reduced by the conjugated glycuronic acid, excess of urinary pigment and pentoses. It is particularly necessary in this test that albumin be removed, as bismuth sulphid forms in the pres- ence of albumin, and may precipitate either in the form of a reddish or a dis- tinctly brownish, even black precipitate. The reduction observed after the administration of medicaments, such as rhubarb, senna, antipyrin, camphor, salicylic acid, salol, sulphonal, trional, quinin, eucalyptus, oil of turpentine, and chloral hydrate, is usually a brown rather than a black unless the solution stands for some time. The administration of saccharine usually results in a reduction of this test, while the copper tests are not affected by it. Sulphur bodies (methyl mercaptan) excreted after the patient has eaten asparagus give a distinct precipitate which may be more or less confusing. If the urine be ammoniacal the reaction may not appear in a characteristic way owing to the fact that the free ammonia is evolved and the alkalinity of the solution is reduced by the combination of the sodium ion with the acid radical formerly bound to the ammonium ion. This test is used by many as a routine test for sugar, as it is practically always negative with normal urine. The reducing action of the glycuronates and pentoses must, however, be remembered. As previously stated, the reduction tests do not absolutely prove the pres- ence of sugar. All that one can say is that a reducing substance is present and if the reaction be typical the probability is that the reducing substance is sugar. The fermentation test is, perhaps, the most certain of all the tests for glucose and depends upon the fact that only those sugars which contain three or a multiple of three carbon atoms are fermentable with yeast. Not all members of these groups of sugars will ferment with yeast, so that for absolutely scientific purposes fermentation will not differentiate them. Fischer's work along this line should be carefully read by anyone interested in the biologic properties of the various types of sugar. As the hexoses, which occur in the urine are Fermentation Test. PLATE VIII. Osazons. (Hawk.) Upper form, dextrosazon; central form maltosazon; lower form, lactosazon THE URINE. 281 practically limited to glucose and levulose, we are safe in saying that any sugar fermenting with yeast is one or the other of these monosaccharides, which may be differentiated by tests to be outlined later. The test is performed as follows: Ten c.c. of urine are placed in a test- tube and a piece of compressed yeast, which should be perfectly fresh, about the size of a pea is added and the urine gently shaken until the yeast is finely divided. This mixture is then poured into a fermentation tube, which is allowed to stand in the incubator for a few hours. Two control tests, using in one normal urine and yeast and in the other normal urine, yeast, and a trace of dextrose, are then made and placed in the incubator along with the suspected urine. The presence of sugar is indicated by gas (CO2) in the upper portion of the fermentation tube. The rapidity of formation of this gas depends upon the amount of yeast as well as upon the age of the yeast. The test in- dicates from o.i to 0.05 per cent, of sugar, especially if the urine be sterilized by previous boiling. The compressed yeast as usually purchased develops a certain minimal amount of gas in normal urine, so that the control test is necessary both to show whether " self-fermentation" is excessive and also whether the yeast is at all active in producing C02 from glucose when it has been added. It is necessary, moreover, that decomposition of the urine be prevented, either by previously boiling the urine, addition of a trace of sodium fluorid, or tartaric acid. This test has the great advantage that the other substances, which reduce copper and bismuth solutions, do not ferment. If precautions are observed to add the right amount of yeast, not to shake the yeast and urine violently enough to include much air, and to prevent bacterial decomposition, this test will positively show the presence or absence of glucose or levulose in the urine and no other substances. It is always wise to use this test either as confirma- tory or decisive in conjunction with the previous reduction tests. Phenylhydrazin Test. This test is much more delicate than any of the previous tests mentioned. Theoretically it will show sugar in the amount present in normal urine, but practically no definite reaction is observed. The principle of the reaction is the decomposition which occurs between the aldehyd or ketone group of the sugar molecule and the amino group of the phenylhydrazin. In this reac- tion, characteristic bodies known as hydrazones are first formed, which are converted, in the presence of dilute acetic acid and an excess of phenylhydra- zin, into crystalline bodies known as osazones. These latter bodies are char- acteristic of the sugar group, usually differing from one another, depending upon the original sugar from which they were formed. The osazones crystal- lize in definite forms, the purified crystals showing rather sharp melting-points. It is, therefore, essential not only that a crystalline body be obtained in this reaction, butthat the crystalline form and the melting-point of the crystal be that characteristic of the sugar suspected. The success of the test will depend largely on the relation of the sugar to the reagents, the best proportions being 282 DIAGNOSTIC METHODS. approximately one of sugar, two of phenylhydrazin, and three of sodium acetate. All of the members of the hexose and pentose groups show this reaction, as do many of those of the polysaccharide series. As will be seen from the reaction given below, those sugars which differ only in the space-relations of the atoms attached to the first two carbon atoms can possibly give the same osazone. For this reason we find glucose, levulose, and mannose as well as glucosamine giving exactly the same osazone (phenylglucosazone), showing a characteristic yellow needle-shaped crystalline deposit, with a melting-point of 204 to 2050 C. The reactions leading to the formation of this body are shown in the following equations: Twenty-five c.c. of urine are treated with a few drops of a solution of lead acetate and filtered to remove the albumin if present. To the filtrate, which must be acidified with acetic acid if it is not already acid, is then added phenyl- hydrazin hydrochlorate (1/2 to 1 gram) and about 2 grams of sodium acetate. The tube is then shaken thoroughly to mix the contents and is placed in a boiling water-bath for from one to two hours. Some workers recommend a shorter period, such as 20 minutes, but the writer has never been able to get as good results with the short heating. At the end of this time the material in the tube is filtered while hot and the filtrate allowed to cool slowly rather than as some advise to cool suddenly by immersion in cold water. If glucose be present a yellow crystalline deposit will appear, which under the microscope will show the characteristic needle-shaped crystals arranged in bundles or sheaves. This microscopic examination does not absolutely prove the crys- tals to be phenylglucosazone so that a melting-point determination must be made before definite proof is forthcoming. For this purpose the crystals must be purified by dissolving in a hot 60 per cent, alcohol and recrystallizing by adding water and evaporating the alcohol. A few of these crystals are then placed in a perfectly dry capillary tube and the melting-point determined by methods previously learned in organic chemistry. Neumann has applied Fischer's method of using this test to the urine. Five c.c. of urine are treated in a test-tube with 2 c.c. of a 50 per cent, solution of acetic acid saturated with sodium acetate and two drops of pure phenyl- hydrazin. This mixture is then evaporated by boiling to 3 c.c., after which it is cooled quickly and is then rewarmed and allowed to cool slowly. If glucose be present to the amount of 0.02 per cent., pure crystals of phenyl- glucosazone separate out in 5 to 10 minutes. If the urine has a high specific gravity and a low sugar content the crystals do not form so quickly. This THE URINE. 283 test is much to be preferred in well-equipped laboratories, but is not as conveni- ent as the use of the crystalline phenylhydrazin hydrochlorate for the general worker, as the fluid phenylhydrazin is quite irritating and is not as easy to work with. This test cannot be used for quantitative purposes as the yield is never complete. For qualitative purposes, however, it is to be especially recom- mended. When properly applied it is the most delicate and one of the most reliable tests at our disposal. If the melting-point of the crystals be determined the only confusing substances will be levulose, glucosamin, and mannose. The osazones formed from the other carbohydrates and the glycuronic acid compounds crystallize in somewhat similar form, but do not show the character- istic melting-point of 2040 to 2050 C. of the phenylglucosazone. It is true that the impure crystals melt at somewhat lower temperature, but if carefully purified from 60 per cent, alcohol they will melt at approximately 2040. Quantitative Methods. The solutions to be used in this test have been given under the qualitative test for sugar. The principle of the test is the determination of the exact amount of sugar necessary to decolorize a mixture of 5 c.c. each of solutions A and B. Providing chemically pure copper sulphate has been used in pre- paring solution A, the mixture is reduced by an amount of urine which con- tains 0.05 gram of glucose. If there is any question regarding the purity of the copper sulphate, the strength of the solution must be determined by titrat- ing against a known solution of chemically pure glucose, or of cane-sugar which has been inverted by heating with dilute acid (see Sutton's Volumetric Analysis). Fehling's Method. Technic. Five c.c. each of solutions A and B (p. 278) are carefully measured from a buret or pipet into an Erlenmeyer flask of about 250 c.c. capacity. Forty c.c. of distilled water are then added and the mixture boiled. The urine is now added in small quantities from a buret to this boiling copper solution until the blue color has entirely disappeared. It is advisable to add the urine rather rapidly as a preliminary test, it being practically impossible with a single titration to obtain accurate results. In doing this one adds a c.c. of urine at a time and observes whether the color still remains. If so, the titration is continued, adding from 1/2 to 1 c.c. at each period, until the color has dis- appeared. The number of c.c. used is then observed and a second test made as follows: To the boiling copper solution is added directly 1 c.c. less of urine than was used for complete decolorization in the preliminary test. The urine is then added two drops at a time, noting after each addition whether the color remains. In this way an accurate end-point may be reached. As the amount of urine used to decolorize the 10 c.c. of copper solution is equivalent to 0.05 gram of glucose, a simple calculation yields the percentage 284 DIAGNOSTIC METHODS. and total values for glucose in the urine. As this test is not accurate with high concentrations of the urine, it is necessary that the sugar content should range between 1/2 and 1 per cent. Dilution may, therefore, be necessary. A good working rule is to dilute a urine of a specific gravity of 1030 five times (by adding 4 volumes of water) and ten times if the specific gravity be above 1035. There is little danger of getting the dilution too great to show the characteristic reaction for sugar, as this test shows the presence of sugar when present to the amount of 0.08 per cent. In the use of this method, as well as of most reduction tests, albumin must first be removed if present in more than traces. Moreover, the presence of preservatives as well as excess of uric acid or creatinin will increase the reducing power of the urine as determined by this test, a considerable error being occasionally introduced unless overcome by dilution or removal of the interfering substances. This method1 appears to the writer to be the best one at our disposal for the quantitative estimation of sugar in the urine. Like the solution of Benedict previously mentioned, it contains alkali carbonates instead of hydrates. The principle of this method is as follows: The urine is treated with an excess of standard alkaline copper solution and boiled to bring out the reduction due to sugar. The amount of copper remaining in excess is then determined by titration with a solution of hydroxylamin sulphate (instead of with potassium sulphocyanate as used by Citron) and from this the amount of sugar is calculated. The solutions necessary in this test are as follows: (a). One hundred grams of potassium bicarbonate are dissolved in about 1300 c.c. of distilled water contained in a 2-liter flask. To this solution are added 500 grams of potassium carbonate and 400 grams of potassium sulpho- cyanate. Exactly 25 grams of pure copper sulphate (CuSO4 sH2O) are then dissolved in about 150 c.c. of warm distilled water. After cooling this solution is added quantitatively to the carbonate solution. Add water up to the 2-liter mark, allow to stand for 24 hours and filter. This solution is stable for about three months. (b). Two hundred grams of potassium sulphocyanate are dissolved in about 1500 c.c. in a 2-liter flask. 6.55 grams of hydroxylamin sulphate are dissolved in water and added quantitatively to the sulphocyanate solution. Add water up to the 2-liter mark and preserve the mixture in dark colored bottles. This solution is very permanent. Bang's Method. Technic. Ten c.c. of urine (5 or 2 c.c. diluted with water to 10 c.c. if more than 0.6 per cent, sugar is present) are measured into a 200 c.c. Jena Erlenmeyer flask and treated with 50 c.c. of the copper solution (a). Heat on a wire gauze to boiling for exactly three minutes. Cool quickly to room temperature by immer- 1 Biochem. Zeitsch., Bd. 2, 1907, S. 271; Bd. 11, 1908, S. 538; Bd. 32, 1911, S. 443. THE URINE. 285 BANG'S TABLE OF REDUCTION EQUIVALENTS Cubic centimeters of hydroxylamin solution Milligrams of sugar Cubic centimeters of hydroxylamin solution Milligrams of sugar o-75 60.0 25-5O 23-5 i .00 59-4 26.00 22.9 1-5° 58.4 26.50 22.3 2.00 57-3 27 .00 21.8 2-5° 56-2 27-5O 21.2 3.00 55-o 28.00 20.7 3-5° 54-3 28.50 20.1 4.00 53-4 29.00 19.6 4-50 52-6 29-5° 19.1 5.00 51-6 30.00 18.6 5-5° 5°-7 30.50 18.0 6.00 49-8 31.00 i7-5 6.50 48.9 31-50 17 .0 7 .00 48.0 32.00 16.5 7-5° 47-2 32.50 15-9 8.00 46.3 33 00 15-4 8.50 45-5 33-50 14-9 9.00 44-7 34-00 14-4 9-5° 44-o 34-5° 13-9 10.00 43-3 35-oo i3-4 10.50 42.5 35-5° 12.9 11.00 41.8 36.00 12.4 n-5° 41.1 36-5° 11.9 12.00 40.4 37-oo 11.4 12.50 39-7 37-50 10.9 13-00 39-o 38-00 10.4 i3-5o 38.3 38.50 9-9 14.00 37-7 39-0° 9-4 i4-5° 37-1 39-5° 9.0 15.00 36-4 40.00 8-5 I5-5O 35-8 40.50 8.1 16.00 35-i 41.00 7-6 16.50 34-5 41-5° 7-2 17 .00 33-9 42.00 6-7 17-5° 33-3 42.50 6-3 18.00 32-6 43-0° 5-8 18.50 32-o 43-5° 5-4 19.00 3i-4 44.00 4-9 19-5° 3°-8 44-5° 4-5 20.00 3°-2 45-0° 4-i 20.50 29.6 45-5° 3-7 21.00 29.0 46.00 3-3 21.50 . 28.3 46.50 2.9 22.00 27-7 47.00 2-5 22.50 27.1 47-5° 2.1 23-00 26.5 48.00 i-7 23-5° 25-8 48.50 i-3 24.00 25-2 49.00 0.9 24-5° 24.6 49-5° o-5 25-00 24.1 50.00 0.0 For every i/ro c.c. hydroxylamin solution used more than is given in the table, subtract o.i mg. if the reading be between 49 and 15, while if it be between 15 and 1 substract 0.2 mg. 286 DIAGNOSTIC METHODS. sion of the flask in cold water. Now titrate with the hydroxylamin solution (b) until a colorless mixture results. From the number of c.c. of hydroxylamin sul- phate solution used, calculate the sugar in milligrams by means of the accom- panying table. A simple calculation yields the percentage and total amounts. It has been found difficult to get a sharp end-point in this reaction if the urine be rather highly colored. Bang and Bohmansson1 have shown that the urine may be cleared by the following process. Twenty c.c. of urine are treated with 5 c.c. of 25 per cent. HC1 and 2 grams of blood charcoal. Shake a few times during five minutes and filter through a dry filter into a dry beaker. This filtrate is then used for the test as described above. Woodyatt and Helmholz2 and Andersen3 have called attention to the necessity of testing the charcoal by control experiments before using the above method of clearing. As some of the blood charcoals give very erroneous results, they recommend the use of Merck's preparation. A further advantage of using this clearing method is that the "auto-reduction" of the urine is reduced to a low limit owing to the almost complete removal of urochrom, uric acid, creatinin and glycuronic acid derivatives.4 Purdy's Method. It has been found that the addition of strong ammonia to the mixed copper and tartrate solutions makes the end-point much more distinct. Pavy, Sahli, Kumagawa and Suto, Kinoshita, and others have modified the original Fehling's solution in this way. It was shown, however, by Lowe that the substitution of glycerin for the Rochelle salt made a much more stable solution and one which could be originally mixed and kept for indefinite lengths of time. Purdy has succeeded in obtaining a copper solution which, in the writer's experience, is much to be preferred for general laboratory purposes over most of the other modifications of Fehling's solution. His formula is as follows: Chemically pure copper sulphate, 4.752 grams Potassium hydrate, 23.500 grams Strong C. P. ammonia (sp. g. 0.88), 350.000 c.c. Glycerin, 38.000 c.c. Distilled water, q.s. ad., 1,000.000 c.c. This solution is prepared by dissolving the copper sulphate and glycerin in 200 c.c. of distilled water, heating gently if necessary. The potassium hydrate is dissolved in a second 200 c.c. of water and mixed with the copper solution. When the mixture has cooled add the ammonia and bring the total volume up to 1 liter with distilled water. Thirty-five c.c. of this solution are decolorized by 0.02 gram of glucose. Haines has slightly modified this original solution of Purdy so that 10 c.c. 1 Zeitsch. f. physiol. Chem., Bd. 63, 1909, S. 443; Biochem. Zeitsch., Bd. 19, 1909, S. 281. 2 Arch, of Int. Med., vol. 7, 1911, p. 598. 3 Biochem. Zeitsch., Bd. 37, 1911, S. 262. 4 Bang (Biochem. Zeitsch., Bd. 38, 1912, S. 168) has recently advised the substitution of 2 c.c. of 95% alcohol for the 5 c.c. of HC1. THE URINE. 287 of the solution are decolorized by 0.01 gram of glucose. This titer is somewhat more convenient than that of Purdy as the calculation is distinctly simplified. The formula for his modification is as follows: Pure copper sulphate, 8.314 grams Pure potassium hydrate, 25.000 grams Glycerin, 40 000 c.c. Ammonia, 350 000 c.c. Distilled water, q. s. ad., 1,000.000 c.c. The principle of this test depends upon the fact that, in the reduction of cupric oxid in solutions of definite strength by glucose, the blue coloration disappears on the addition of a definite amount of glucose without any attendant precipitate, the reduced solution remaining transparent and colorless. Technic. Thirty-five c.c. of Purdy's test solution or 10 c.c. of Haines' solution are measured into an Erlenmeyer flask and 50 c.c. of water added. The flask is then closed with a doubly perforated rubber stopper, through one hole of which passes the stem of a buret containing the suspected urine and through the other a bent glass tube to conduct the fumes of ammonia away from the observer. The object of closing the flask with the stopper is to exclude the air and thus prevent reoxidation of the cuprous oxid. The contents of the flask are now brought to a gentle boil and the urine added, 1 c.c. at a time, shaking the flask after each addition until the fluid is completely decolor- ized. The number of c.c. used is then noted and a second test made, add- ing at once 1 c.c. less of urine than the total number of c.c. used in the first experiment. The last portions of the urine are added two drops at a time, allowing from three to five seconds to elapse between the addition of these separate portions. When the urine is completely decolorized the number of c.c. used is noted. As the amount of urine used is equivalent to 0.02 gram of glucose with Purdy's solution or 0.01 gram with Haines' solution, the per- centage may be obtained as in the previous method. It is necessary with this method that the urine be diluted for the same reasons as previously mentioned. In all of the copper tests for glucose the influence of preservative agents must be remembered. Thus chloroform, chloral hydrate, and formalin will all reduce copper solutions so that an error may be introduced unless these substances are removed. In selecting a method for the quantitative determination of sugar in the urine, the general worker should remember that the method used should be both simple and exact. With these points in view, the writer has little hesitancy in recommending Bang's method. If the urine be cleared as previously de- scribed, the results are all that could be desired, the error being extremely small. Many other modifications of the copper tests have been advocated, but the writer does not feel that they have any advantage from either the scientific or clinical standpoint. The method of Allihn, in which a copper solution is 288 DIAGNOSTIC METHODS. reduced and the precipitated cuprous oxid either weighed as such or further reduced to metallic copper in a stream of hydrogen, is not clinically available. The modifications of Rudisch, Rudisch and Celler, and of Gerrard and Allan do not give as accurate results as does the Bang method in the experience of the writer. The recent method of Benedict1 is apparently quite reliable. Polariscopic Method. In this test the urine must be absolutely clear and must contain no albumin. Moreover, it is essential that such substances as glycuronic acid and /?-oxy- butyric acid be removed before polarization as they will introduce considerable error in the reading. The best clearing agent is that, previously mentioned under Bang's test, of shaking the urine with HC1 and blood charcoal and filter- ing. Instead of this method the urine, acidified with acetic acid, may be treated with a solution of normal lead acetate, which will precipitate the albumin and remove excess of pigment. The mixture is filtered and the clear filtrate is used for the test. It must be remembered that a correction must be made in the percentage of sugar in the filtrate, if the urine be cleared with a solution of lead Fig. 83.-One form of Laurent Polariscope. (Hawk.) B, Microscope for reading the scale; C, a vernier; E, position of the analyzing Nicol prism; H, polarizing Nicol prism in the tube below this point. acetate. This may be done by adding 25 c.c. of a 10 per cent, solution of lead acetate to 75 c.c. of urine. The sugar in the filtrate will represent only three- fourth of that of the original urine. Basic lead acetate should be used with caution, as it precipitates various sugars. If the urine contain no albumin, magnesium oxid or silicic acid may be used as clearing agents. Various types of polarimeters have been suggested and most of them are quite satisfactory for the determination of sugar. One of these with its description is seen in the accompanying cut and legend. The principle upon which their use depends is the fact that optically active substances when in solution have the power of turning the plane of polarized light either to the right or the left. The zero point of the instrument is determined by observing the point at which the halves of the optical field have exactly the same degree 1 Jour, of Biol. Chem., vol. 9, 1911, p. 57; Jour. A. M. A., vol. 57, 1911, p. 1193. THE URINE. 289 of illumination, when the light passes through a tube either empty or con- taining an optically negative fluid. The point at which the graduated scale of the instrument and the vernier correspond is regarded as the zero-point. In the use of the polariscope any deviation of the plane of light passing through the polarizing Nicol prism will be noticed by a darkening of one portion of the field, so that the compensating or analyzing Nicol must be rotated until both parts of the field are equally illuminated. In this way one readily deter- mines how much the plane of polarized light has been deviated, by observing the degree of rotation necessary to bring the two portions of the field into equal illumination when the light passes through an optically active fluid. The most reliable instruments for general work are known as the "half-shadow" types, but they rarely find a place in the equipment of the general worker. The examination is usually made in a dark room, the light passing through the tube of the polariscope from a sodium flame. It must be remembered that the urine may contain substances other than glucose which rotate the plane of polarized light. The normal urine is slightly Fig. 84.-Diagrammatic representation of the course of light through the Laurent polariscope. (Direction reversed from that of previous figure.) (Hawk.) a, Bichromate plate to purify the light; b, the polarizing Nicol prism; c, a thin quartz plate covering one-half the field and essential in producing a second polarized plane; d, tube to contain the liquid under examination; e, the analyzing Nicol prism; / and g, ocular lenses. levorotatory, the degree varying between 0.05 and 0.18. Albumin is also levorotatory and will interfere markedly with the degree of rotation refer- able to glucose, unless it be very slight in amount or be entirely removed. Levulose, j3-oxybutyric acid, and the conjugated glycuronates are also levo- rotatory, the second being especially prone to interfere with the glucose rotation as it so frequently is associated with glucose in diabetic conditions. If the urine has been previously heated with acid before being examined in the polari- scope, glycuronic acid will also interfere, as it shows distinct dextrorotation. Cane-sugar as well as lactose are occasionally found in the urine and may lead to confusing results as both are dextrorotatory substances. It is, there- fore, necessary before any reliable results may be obtained with the polari- scope to remove the interfering substances, either as preliminary to the de- termination of glucose or subsequently, a correction being then made for the rota- tion of the interfering substance. It is to be said that the polariscope is more fitted for a clinical laboratory than for the general practitioner. The rotation of light when passing through optically active solutions is dependent upon several factors, among which we find the temperature at which the observation is made, the length of tube through which the light passes and the concentration of the active solution. By specific rotation of a fluid is 290 DIAGNOSTIC METHODS. meant the rotation observed when light passes through a solution containing 1 gram of the active substance per c.c. of fluid and placed in a tube one dem. in length. Thus we find the specific rotation of glucose in such a tube being (a)D = 4-52.74. It is, therefore, evident that the percentage of sugar in an unknown solution contained in a tube 1 dem. in length may be obtained by dividing the degree of rotation by 0.5274. Some of the tubes for clinical pur- pose are constructed of such a length (188.6 mm.) that one degree of rotation equals 1 per cent, of glucose. For a full discussion of the subject of optical activity of fluids the writer would refer to the work of Landolt.1 Having determined the zero-point of the instrument, the tube, which should be thoroughly cleaned and dry, is filled with the fluid to be examined. In filling this tube precautions must be taken not to include any bubbles of air, which is best done by filling the tube until a convex meniscus is observed and then sliding the glass disk over the end in such a way that the excess of fluid is shoved off. The metal cap is then screwed on tightly, but in such a way that undue compression is not exerted upon the glass disk. The tube is placed in position, the field distinctly focussed and the degree of rotation deter- mined by revolving the analyzing Nicol until the two portions of the optical field have exactly the same intensity of illumination. This will require consider- able experience, as the accuracy of the determination will depend not only upon the clearness of the fluid, the degree of focussing, the sensitiveness of the instrument, and the brightness of the light, but also upon the sen- sitiveness of the observer's eyes. Several readings should be made, turning the prisms from both direc- tions and observing the degree at which the fields are isochromatic, the average being taken as the final result. As the eyes soon become fatigued, they should be used for only a few seconds at a time. Having determined the degree of rotation, the percentage of sugar is then calculated as previously mentioned. This method, if carefully used and inter- fering substances avoided, is the most accurate method for the determination of glucose. It is, however, difficult, requires much experience, and is not as sensi- tive as the other methods, rarely detecting the presence of less than 0.2 per cent, of glucose. Theoretically, the percentage of sugar as determined by the polariscope should exactly agree with that obtained by titration of copper solu- tions, but in general work such is never the case, as interfering substances may be present or have not been completely removed. The writer must refer elsewhere for methods of correcting the readings of Technic. Fig. 85.-Einhorn sac- charometer. 1 Das optische Drehungs-Vermogen, Braunschweig, 1898. THE URINE. 291 the polariscope when interfering substances are present, as he does not think it wise to recommend the general worker to waste his time with this instrument when sufficiently accurate clinical results may be obtained by methods which are for him much easier and less liable to error. Fermentation Methods. The principle of this method has been previously outlined. It has been found that, if the precautions mentioned are observed, carbon dioxid is evolved quantitatively from glucose by the action of yeast. The most convenient method of applying the test is to use an Einhorn fermentation tube which is so graduated that the amount of CO2 evolved is directly read off in terms of per cent, of glucose. For this purpose the urine must contain less than i per cent, of glucose* the urine being diluted as previously described to bring the amount within this figure, multiplying the result, of course, by the degree of dilution. The urine, which should be acid in reaction, is shaken with a piece of com- pressed yeast about the size of a small pea, all precautions being observed as mentioned above. The mixture is then poured into the fermentation tube in such a way that no air-bubbles collect at the upper end of the tube. Controls as previously described are then made and the three tubes placed in the incu- bator at 37° C. In a few hours bubbles of gas (CO2) will collect at the top of the tube, the fermentation being practically complete overnight. The percent- age of sugar is then read direct from the calibration (1/4-1 per cent.) on the tube (the figures from one to five representing c.c. of gas and not per cent, of sugar). This apparatus is seen in the accompanying cut. Twelve c.c. of mercury are placed in the bulb of the apparatus. One-half c.c. of the urine to be tested is then floated upon the mercury and treated with a thick paste of com- pressed yeast diluted two or three times with water. The stopper is then carefully greased with vaseline and inserted so that the two apertures corre- spond. By tipping the apparatus a trifle the column of mercury in the long tube is then adjusted to the zero point of the scale. When this is done the stop- per is turned so that the holes no longer correspond and the weight is placed on the stopper to prevent leakage from the increased pressure of the gas liberated in the fermentation. The apparatus is then placed in the incubator and the extent of fermentation read off by noticing the height to which the column of mercury rises in the long arm of the instrument. The fermentation is usually complete within six hours. After removing the apparatus from the incubator it is allowed to stand in the air for a few minutes to adjust itself to room tem- perature, as the scale is graduated in this way. This method gives results which correspond very closely to those of titra- tion and is to be recommended for the quantitative determination by fermenta- tion methods. Lohnstein has also introduced a saccharometer which may be used with diluted urine. It is seen in the accompanying cut. It does not, in the writer's Lohnstein's Saccharometer. 292 DIAGNOSTIC METHODS. opinion, have any advantage over the above-mentioned apparatus as the prin- ciple is the same, although the urine must be diluted. This method has been recommended for the quantitative determination of sugar and is based upon the fact that the specific gravity of the urine is changed in a quantitative way when the sugar of the urine is fermented. The urine must be acid before applying this test. A piece of yeast about the size of a bean is added to the urine which is allowed to ferment at incubator Robert's Method. Fig. 86.-Lohn- stein's fermenta- tion tube for undi- luted urine. Fig. 87.-Lohnstein's fermentation tube for diluted urine. temperature until no further qualitative test for sugar is obtained. This will usually require from 24 to 48 hours, so that a trace of sodium fluorid should be added to the urine to prevent bacterial action. The specific gravity of the urine before it is subjected to fermentation is very carefully taken either with a very accurately standardized hydrometer or, preferably, with the pycnometer. After fermentation is complete the specific gravity of the fermented urine is determined in the same way. The difference in the specific gravity of the two specimens is then multiplied by 234 to obtain the percentage of sugar. Or, according to Purdy, each degree of specific gravity lost in fermentation corre- sponds to 1 grain of sugar per fluidounce. THE URINE. 293 If the specific gravity be accurately determined the results are correct within 0.1 per cent. As this method requires, however, the use of a pycnometer and a very accurate chemical balance, it can hardly be recommended to the general practitioner for his work, although some writers state that it should be preferred to the more uncertain titration with Fehling's solution? (&). Levulose (d-fructose). CH,OH- (CHOH)- CO-CH2OH. Levulose is found very widely distributed throughout the vegetable king- dom, especially in fruits. ■ Honey is almost a pure levulose. It may be found in the urine, transudates, or exudates, after a large intake of levulose-containing food or may occur spontaneously, when the subject has taken little such food. The levulosuria, reported by Zimmer, Ventzke, Czapek, Worm-Miiller, Seegen, Mauthner,. Cotton, Roehmann, Personne, Henninger, Marie, and Robinson, must be accepted with reserve, as the incompleteness of the methods then in use afforded no certain means of recognizing levulose (Neuberg2). The most authentic cases of true levulosuria are those of May, Schlesinger, Rosin and Laband, Lepine and Boulud, and Neubauer. This pure levulosuria occurs in both sexes and at all ages, the amount of sugar excreted being subject to variations from 2.7 grams per diem (Schlesinger) to 24 grams (Lepine). More common than pure levulosuria is its association with a glycosuria. This combination appears in all forms of diabetes, in the severe types levulose being practically never missed, according to Umber, especially when no restric- tion is placed on the carbohydrate intake. Neubauer's observations on this point are interesting. He finds that withdrawal of carbohydrates causes both levulose and dextrose to disappear from the urine. When levulose was given it was utilized, but when glucose was taken it was less completely assimilated, being excreted in part as levulose. The tolerance for levulose is, as a rule, less than that for glucose, so that we are not surprised to find that the administration of 100 grams of levulose to normal individuals is followed by a levulosuria in about 10 per cent, of cases, while no such effect may be observed in diabetes mellitus. Just why diabetics should tolerate levulose and not glucose is not clear, but we must remember that not all diabetics do. Here the question of individual tolerance must be considered, as Umber finds 25 grams of levulose excreted following an intake of 100 grams, while the writer has observed an excretion of 75 grams on the same intake. Strauss3 finds that an alimentary levulosuria occurs, after an intake of 100 grams of levulose, in 90 per cent, of cases of functional hepatic disturbance. This test would seem, therefore, to be a valuable indication of hepatic insufficiency, although it is not pathognomonic. Also alimentary galacto suria, following the administration of 40 grams of galactose to the fasting stomach, as originated by Bauer,4 has been shown by Reiss and Jehn5 and, 1 Simon. Clinical Diagnosis, Phila., 1907. 2Handbuch der Pathologic des Stoffwechsels. Berlin, 1907, S. 716. 3 Deutsch, med. Wochensch., Bd. 27, 1901, S. 757 and 786. 4 Deutsch, med. Wochschr., Bd. 34, 1908, S. 1505. 5 Deutsch. Arch. f. klin. Med., Bd. 108, 1912, S. 187. 294 DIAGNOSTIC METHODS. also, by Roubitschek1 to be inconstant. A decidedly positive excretion of galactose is observed only in cases in which the entire liver parenchyma is pre- sumably involved, as in the acute infections and intoxications characterized by catarrhal jaundice.2 Cases of cirrhosis,3 carcinoma, cholelithiasis and phos- phorus poisoning react either negatively or very slightly. The chemical reactions of levulose are very similar to those of dextrose. Owing to the presence of the ketone group in its molecule, it shows the same reducing actions as does the aldehyd group of the glucose. Like glucose it ferments, but not quite so readily. It is levorotatory, its specific rotation being (a)D = -91 degrees. Leo has reported the finding of a levogyrate car- bohydrate in diabetic urines, which he believes to be laiose. This differs from levulose in being unfermentable with yeast. Levulose forms exactly the same phenylosazon as does glucose, so that it is a matter of great difficulty to differ- entiate these bodies by the tests given above. Lobry de Bruin and Alberda von Eckenstein have shown that glucose and levulose may change, the one into the other, by the action of traces of alkali, acids, and neutral salts, such as sodium acetate. This point must be borne in mind, as it may explain why levulose is excreted in certain cases. This test has been advanced as one characteristic for the ketoses in distinc- tion from the aldoses. Ten c.c. of urine are treated with a few crystals of resor- cin and 5 c.c. of concentrated HC1. If the mixture be warmed a brilliant red color appears in the presence of a ketone (levulose) while no coloration is ob- served with an aldehyd (glucose).4 Muller has shown that glucosamin gives this test, while R. and O. Adler find a reaction in the presence of nitrous acid. If the mixture be heated too strongly or too long, mannose and maltose may also give a positive test. Adler finds that the use of acetic acid with a trace of HC1 gives better results than HC1 alone. It will be seen, therefore, that this test is not so characteristic as was believed, but it serves to distinguish levulose from glucose, which is the important point. If the red solution formed in this reaction be neutralized with sodium carbonate and extracted with amyl alcohol or, preferably, with acetic ether (Borchardt), the extract will have a yellow color with a faint green fluores- cence and becomes rose-red on the addition of alcohol. The spectrum of this solution shows a sharp line in the green between E and b, while if the solution be quite concentrated a second weaker line will be seen in the blue at F. Neuberg5 has recently shown that fructose forms a characteristic osazone Seliwanoffs Test. 1 Deutsch. Arch. f. klin. Med., Bd. 108, 1912, S. 226. 2 Jour. A. M. A., vol. 60, 1913, p. 287. 3 See Foster, Amer. Jour, of Med. Sci., vol. 143, 1912, p. 830. 4 Konigsfeld (Biochem. Zeitsch., Bd. 38, 1912, S. 310) shows that the HC1 must not have a final concentration greater than 12.5 per cent, and that the heating should not be longer than 20 to 30 seconds. If dextrose, also, be present in amount greater than 2 per cent., the possibility of its conversion into levulose must be remembered. See, also, Jolies, Ibid., Bd. 41, 1912 S. 331. 5Zeitsch. f. physiol. Chem., Bd. 36, 1902, S. 227; Ber. d. d. chem. Ges , Bd. 35, 1902, S. 959; Ibid., Bd. 37, 1904, S. 4616. THE URINE. 295 with phenylmethylhydrazin, while no such compound is obtained with glucose, mannose, or glucosamine. The test is, therefore, the most reliable and scien- tific one for the presence of levulose, although it is not clinically so acceptable as the Seliwanoff reaction. The formation of this osazone occurs according to the following equation. Technic. The urine is acidified with acetic acid and boiled to remove albumin if present. The mixture is then filtered and the clear filtrate, which must be acid, is evaporated in a vacuum, at a temperature not over 40 °, to a thin syrup. The reaction must remain acid during the evaporation so that a drop of acetic acid may be added if necessary. The residue is thoroughly extracted with 98 per cent, alcohol, using an amount of alcohol equal to one-half the original volume of urine. Filter and re-extract the residue with alcohol should any reducing action be observed in it. The alcoholic extracts are mixed and decolorized with animal charcoal. A portion of the extract is examined for its sugar-content by Fehling's test, all of the reduction being attributed to levulose. Methylphenylhydrazin is then added to the alcoholic solution, which should not measure over 30 c.c. The amount of the hydrazin to be added is in the proportion of 3 molecules for 1 of levulose or, in other words, for each gram of sugar a trifle over 2 grams of methylphenylhydrazin. The mixture is allowed to stand for a few hours in the cold and filtered if a precipi- tate forms. The filtrate is treated with 50 per cent, acetic acid, using the same amount of acid as of the methylphenylhydrazin, and sufficient alcohol is added to give a clear solution. The mixture is heated from three to five min- utes or, preferably, allowed to stand at 400 C. for 24 hours in an incubator. Crystals will usually separate out at the end of this time, but if not they will appear on the addition of a few drops of water. The crude product is purified by recrystallization either from a mixture of chloroform and petroleum ether or from hot water to which pyridin has been added. The yield by this method is 81 per cent, of the total sugar if pure solutions are used, while from urine it is but 50 per cent. The crystals of methylphenylosazone are delicate yellow, long, fine, needles, melting from 158 to 1600 C. This method can hardly find application in the hands of the general practitioner. Pure levulosuria is recognized by the levorotation of the urine, which possesses reducing properties and is capable of fermenting with yeast. After fermentation, the urine loses its reducing and optical properties. The presence of a levulosuria is indicated by a considerable difference between the results obtained by titration and polarization, providing a glycosuria be coexistent. 296 DIAGNOSTIC METHODS. The Seliwanoff reaction should be used as a routine in every case which shows both fermentation and reduction. The general practitioner will rarely have access to a polariscope and even then may be misled if other interfering sub- stances, such as albumin, glycuronic acid, and ^-oxybutyric acid be present. Do not presuppose that glucose is the only fermentable and reducing sugar of the urine, as the recognition of the true condition may make much difference in the treatment as well as in the prognosis. (c). Pentose. The pentose1 group of carbohydrates comprises eight possible stereo- isomers with the general molecular formula of C5H10O5. Three members of this group, rhamnose, fucose, and chinovose, are substituted pentoses, viz., methyl pentoses having the formula C6H12O5. These latter are the exceptions to the general rule that carbohydrates contain the same number of carbon atoms as of oxygen atoms, and have led to the more scientific method of classifying the carbohydrates according to the number of oxygen atoms rather than of carbon atoms. The pentoses are widely spread throughout the vege- table kingdom in the form of their anhydrids or in combination wTith other groups of atoms. They are also found as constituents of the nucleoproteins of animal tissues, being especially abundant in the pancreas. It has been found that the ingestion of large amounts of pentose-contain- ing food, such as apples, cherries, plums, beets, and the leguminous vegetables, leads to the excretion of pentose in the urine. This alimentary type of pen- tosuria is characterized by the presence of optically active xylose or arabinose and appears after the ingestion of small amounts, in some cases following an intake of as low as 50 mg. of the pure carbohydrate. In true diabetes mel- litus the urine frequently contains 1-xylose which probably arises from the breaking down of the pancreatic nucleoprotein. Alfthan states that the pentoses are constantly present in diabetic urine, so that it is highly probable that these sugars might be found whenever search was made for them in diabetic conditions. It is not strange that these carbohydrates are not more frequently reported, as their reducing action leads to confusion unless controlled by fer- mentation methods. The excretion of pentoses in diabetic conditions is not necessarily increased in direct proportion to the intake, as their absorption may be so slow that accumulation is not possible before oxidation has occurred. Salkowski and Jastrowitz2 reported in 1892 the finding of a pentosuria which was not associated with intake of pentose food nor with diabetes melli- tus. This type is known as idiopathic, essential, or intrinsic pentosuria, of which 24 cases were found by Janeway3 up to 1906. The peculiar thing of this type of pentosuria is that the sugar excreted is r-arabinose, an optically inactive pen- tose. This is the single exception in which an optically inactive pentose is found in all nature. This fact characterizes this type of pentosuria as an anom- 1 See Bendix, Die Pentosuria, Stuttgart, 1903. 2Centralbl. f. d. med. Wissensch., 1892, p. 337. 3 Amer. Jour, of Med. Sci., vol. 132, 1906, p. 423. THE URINE. 297 aly of metabolism sui generis. The origin of this urinary pentose is still unsettled. The source must be within the organism, as no inactive arabinose is taken as food, and if it be given in experimental cases it appears in the urine as d-arabinose. Moreover, it cannot be derived from the nucleoprotein as the pentose in these cases is 1-xylose. Neuberg suggests that galactose might be considered the source of this pentose, but no proof of this has been forthcoming. In the true idiopathic pentosuria the assimilation of other carbohydrates is unchanged and does not influence in any way the excretion of r-arabinose, although the active types of this pentose may be excreted at the same time in the urine. It is interesting to find that the pentoses taken in as food are excreted in different proportions by the diabetic and nondiabetic subjects. Thus von Jaksch observes that diabetics excrete from 49 to 82 per cent, of arabinose of the food and nondiabetics 1 to 47 per cent., while nondiabetics excrete from 19 to 55 per cent, of xylose and diabetics only a trace. The amount of pentose excreted in essential pentosuria has been reported as vary- ing between 0.08 and 1 per cent. Neuberg has recently shown that a certain amount of the r-arabinose is combined with urea in the form of a ureid, which does not reduce Fehling's solution until it undergoes hydrolysis with acid. For this reason he believes that the amount of pentose reported is in practically all cases 100 per cent, too low. Recently Luzzatto1 has reported the excretion of the optically active 1-arabinose entirely independent of the food intake. It is questionable whether this case is to be classified with the idiopathic pentosurias. The pentoses reduce copper solutions as do other carbohydrates, but the reduction is much slower, appearing during the cooling of the fluid. Ten c.c. of Fehling's solution are reduced by 0.0542 gram of pentose. They do not fer- ment with yeast and do not give a typical reduction with the Almen-Nylander test, the color being a gray rather than a black. In the true iodiopathic pen- tosuria no reaction is observed with the polariscope, while in the alimentary type a slight dextrorotation is usually noted, although von Jaksch reports the excretion of an inactive arabinose after the ingestion of active pentoses. These pentoses form more or less typical osazones which melt between 157 and 1600, but the reaction is not so easily produced.2 These pentosazones are readily soluble in warm water and show dextrorotation. The most characteristic chemical property of these types of carbohydrate is the formation of furfurol (C5H4O2) when they are distilled in the presence of acids. The color reactions given below are based upon the production of furfurol and the formation of distinct colorations on treatment with various reagents. Tollen's Test. A few c.c. of concentrated hydrochloric acid are saturated with phloro- glucin, care being taken to leave a small amount undissolved. This solution is then divided into two equal parts, to one of which is added 1/2 c.c. of the 1 Beitr. z. chem. Physiol, u. Path., Bd. 6, 1905, S. 87. 2 See Elliott and Raper, Jour, of Biol. Chem., vol. 11, 1912, p. 211; also, Klercker, Deutsch. Arch. f. klin. Med., Bd. 108, 1912, S. 277. 298 DIAGNOSTIC METHODS. suspected urine and to the other 1/2 c.c. of normal urine. Both tubes are then placed in a boiling water-bath for a few minutes, when an intense red zone will appear in the upper portion of the tube if pentose be present. After a few moments this red color will gradually spread throughout the fluid, while the control urine shows no marked change in color. It is advisable to remove the tubes from the water-bath as soon as the color appears, as the clearness of the reaction is interfered with by prolonged heating. The coloring matter is then extracted by shaking with amyl alcohol, when spectroscopic examina- tion will show an absorption band between D and E. This test reacts in the same way with glycuronic acid so that it has little value in differentiating pentose from the former substance. As a rule, the free glycuronic acid is not so easily split from its conjugated compound as is furfurol from pentose, so that the test is at least suggestive of pentose. Orcin Test. For this test the urine should be decolorized by heating with animal char- coal and filtering. Five c.c. of urine are treated with an equal volume of con- centrated hydrochloric acid and a few crystals of orcin are added. The mix- ture is then warmed approximately to the boiling-point, when a dark green color appears in the presence of pentose or glycuronic acid. The formation of a greenish-blue precipitate is very strong evidence of pentose rather than glycuronic acid. The pigment is then extracted with amyl alcohol, when spec- troscopic examination shows a characteristic absorption band between C and D. The presence of glucose may interfere with the reaction, so that it may be necessary to remove it by fermentation. Bial has modified this test in such a way that glycuronic acid is less apt to be a disturbing factor. His reagent consists of 500 c.c. of 30 per cent. HC1 to which are added 1 gram of orcin and 25 drops of 10 per cent, ferric chlorid solution. Four to five c.c. of this reagent are heated to boiling and removed from the flame. The suspected urine is then added drop by drop, not exceed- ing 1 c.c. in all, when a green color should appear almost immediately if pen- tose be present. The heat employed is hardly sufficient to split off glycuronic acid. Here also glucose, if present, should be removed by fermentation with a pure culture of yeast rather than with compressed yeast, as the bacteria possibly present in yeast may break up the pentoses at the same time. The osazone may be formed as previously given under Glucose. The glucosazone is separated from the pentosazone by digesting with water not over 6o° in temperature, the pentosazones being dissolved. If the pentosazone be treated with 20 c.c. of w'ater and 5 c.c. of concentrated hydrochloric acid and distilled, the distillate will give a beautiful test with Bial's reagent, which absolutely eliminates glycuronic acid and other interfering substances possibly present in diabetic urine. Quantitative Determination. Neuberg and Wohlgemuth1 have recently introduced a method by which 'Zeitsch. f. physiol. Chem., Bd., 35, 1902, S. 31 and 41. THE URINE. 299 the arabinose of the urine may be accurately determined. A preliminary determination of the sugar present is made by Purdy's solution, eliminating glucose by previous fermentation. If less than 1 per cent, of reducing sugar, which is assumed to be arabinose, is present, the urine must be concentrated in a vacuum so that the sugar content is slightly over 1 per cent. One hundred c.c. of urine are acidified with two drops of 30 per cent, acetic acid and evaporated on a water-bath to approximately 40 c.c. It is then treated with 40 c.c. of 96 per cent, alcohol, the mixture is allowed to stand for two hours, and is then filtered from the separated urates and inorganic salts. The residue is carefully washed with 40 c.c. of 50 per cent, alcohol. To the filtrate 1.4 grams of pure diphenylhydrazin are added and the mix- ture heated on a boiling water-bath for one-half hour, the alcohol being replaced as it evaporates. The mixture is allowed to stand for 24 hours and is filtered through a Gooch filter, using the mother liquor to transfer the precipitate. The crystals are then washed with 30 c.c. of 30 per cent, alcohol, and the Gooch with its contents dried at 8o° C. to constant weight. The amount of arabinose is obtained by multiplying the weight of the diphenylhydrazone by 0.4747 or by dividing by 2.107. Technic. Cammidge's Reaction. Recently Cammidge1 has found that the urine in cases of pancreatic disease contains a substance which gives an osazone when treated with phenyl- hydrazin. In his earlier work he was led to believe that this substance was possibly glycerin or a derivative. He advanced two reactions, the first of which he found to be due to a mixture of glycuronic acid and a true carbo- hydrate, while the second was apparently due only to the glycuronic acid. As his latest technic was supposed to eliminate all the glycuronic acid, he believed the mother substance of the osazone to be a pentose, probably derived from the nucleoprotein of the pancreas. This has been shown by many workers to be fallacious, the crystals not being of constant composition. The careful work done in recent years by Kinney,2 Whipple and King,3 Whipple, Chaffer and Fischer,4 Swan and Gilbridge,5 Wilson,6 and Roper and Stillman7 has proven conclusively that, although this reaction is given in some cases of pancreatitis, it is not pathognomonic as it occurs in many other conditions and may not, even, be present in pancreatic disease.8 Technic. The writer gives only the improved method known as "Reaction C" refer- lancet, vol. i, 1904, p. 782; Ibid., vol. 2, 1905, p. 14; Robson and Cammidge, Surgery of the Pancreas, London, 1907. 2 Amer. Jour, of Med. Sci., vol. 140, 1910, p. 878. 3 Bull. Johns Hopk. Hosp., vol. 21, 1910, p. 196. 4 Ibid., vol. 21, 1910, p. 339. 5 New York Med. Jour., vol. 91, 1910, p. 781. 6 Trans. Assoc. Amer. Phys., 1910. 7 Arch, of Int. Med., vol. 7, 1911, p. 252. 8 See Mayesima, Mitt. a. d. Grenzgeb. d. Med. u. Chir., Bd. 25, 1912, S- 403; also, Karas, Zeitsch. f. klin. Med., Bd. 77, 1913, S. 125. 300 DIAGNOSTIC METHODS. ring to the original work of Cammidge for reactions A and B. The urine to be tested should be a portion of the 24-hour specimen and must be freed from glucose and albumin by methods previously outlined. Forty c.c. of clear, filtered acid urine are acidified with 2 c.c. of concen- trated HC1 and boiled for 10 minutes. The mixture is then cooled and made up to 40 c.c. with distilled water. The excess of acid is then neutralized by the addition of 8 grams of lead carbonate and the mixture cooled if necessary. Filter off the resulting precipitate and treat the filtrate with 8 grams of powdered tribasic lead acetate to remove the glycuronic acid. Filter, treat filtrate with 4 grams of powdered sodium sulphate, heat to the boiling-point, and allow to cool. The lead sulphate is removed by filtration. Ten c.c. of the clear filtrate are made up to 17 c.c. with distilled water, 0.8 gram of phenylhydrazin, 2 grams of sodium acetate, and 1 c.c. of 50 per cent, acetic acid are added and the mixture boiled for 10 minutes. Filter while hot and make the filtrate up to 15 c.c. with warm water. The mixture is allowed to cool, when yellow crystals arranged in sheaves and rosettes may be observed under the high- power lens. (J). Lactose (C12H22OU). Lactose is found in the urine of women during the period of lactation and may be found in patients who have been on an exclusive milk diet for a long period. A distinct type of alimentary lactosuria is observed on account of the low assimilation limit for milk-sugar. In breast-fed children with gastrointestinal disturbance lactose associated with galactose may be found in the urine. In this case Langstein and Steinitz1 have shown that the excretion is not due to failure of the normal enzyme, but to an unknown derangement of the activity of lactase, which renders it incapable of splitting up the whole of the lactose, the remainder being absorbed unchanged. From the portion which is split up in the bowel the resulting easily assimilable glucose is utilized by the organism, while galactose partly escapes by the kidneys, on account of the much lower assimilation limit for this latter carbohydrate. Lactose in these cases is usually associated, therefore, with galactose (Neuberg). According to the work of Voit, an increase of lactose in the diet of a diabetic is associated with an increased output of glucose. The usual form of lactosuria is that observed in the parturient female. It is ordinarily first seen a few days after delivery of the child, but occasionally appears during the latter days of gestation, as Ney, Lemaire, and Porcher have shown. The amount of lactose excreted by the nursing mother equals 2 to 3 per cent., according to Naunyn, while McCann places the average at 0.35 per cent, for the first few days of the puerperium. Lactose may continue in the urine for some time, the actual amount depending upon the quantity of milk as well as its quality.2 If nursing is interrupted for any reason, more lactose will be found than when nursing is regular. Lactose reduces copper solutions, although somewhat less actively than 'Beitr. z. chem. Physiol, u. Path., Bd. 7, 1906, S. 575. 2 See Gronvall, Biochem. Zeitsch., Bd. 40, 1912, S. 145. THE URINE. 301 does glucose. It also shows a positive Almen-Nylander reaction. It has a strong rotatory power, its specific rotation being practically the same as that of glucose ( + 52.50). It does not ferment with yeast, although bacteria if present may hydrolyze it into its constituents, glucose and galactose, the former of which will show fermentation. It is, therefore, advisable when applying the fermen- tation test not to judge of a reaction which has progressed longer than a few hours. With phenylhydrazin it forms a lactosazon which appears in the form of sheaves of delicate curved needles much resembling bunches of yellow thread. These crystals melt at 2000 C. The test is not easily obtained unless the urine be concentrated to a small bulk and the residue extracted with alcohol, when the alcohol is evaporated and this residue taken up with water and the phenyl- hydrazin test then applied. Ten c.c. of urine are treated with an excess (3 grams) of lead acetate and boiled for a few minutes. The yellowish or brown solution is then filtered and ammonia added to the filtrate until a slight permanent precipitate remains. An intense brick-red fluid is obtained which later shows the deposition of a cherry-red precipitate with a colorless supernatant fluid. This test is not very delicate as it shows lactose only when present in amounts varying from 0.3 to 0.5 per cent. Glucose gives with this test a red solution, but a more distinctly yellow precipitate. Lactosuria is to be assumed when the urine possesses reducing properties and dextrorotation, but is incapable of fermenting with ordinary yeast within 12 hours. If the urine be boiled with 2 per cent, sulphuric acid and then neutral- ized, its optical activity will be increased and it will be capable of undergoing fermentation. It is to be remembered in testing for the amount of lactose by the use of Fehling's solution that 10 c.c. of this solution are reduced by 0.0678 gram of lactose instead of by 0.05 as in the case of glucose or levulose. Rubner's Test. Maltose has occasionally been reported in the urine, although many of the cases are questionable as the proper identification of the sugar was not thoroughly carried out. The most reliable cases appear to be those of Noble, von Ackeren, Rosenheim and Flatow, and especially that of Magnus-Levy. In thjs latter case the urine showed a considerable excess of rotation when compared with its reducing power. After inversion with dilute acid, by which each molecule of maltose was converted into two molecules of glucose, the rotation diminished and the reduction increased, so that the polarimetric and titration methods gave concordant results. The urine underwent complete fermentation with synchronous loss of optical activity and of reducing power. Calculations founded on these determinations showed that 1.5 per cent, of maltose and 2 per cent, of glucose were present. This seems to be a case in which the amount of maltose excreted exceeds all records (Neuberg). The cases in which maltose appears in the urine seem to be those of disease (e). Maltose (Cl2H22Ou). 302 DIAGNOSTIC METHODS. of the pancreas, especially those with interstitial lesions. It is possible that the Cammidge reaction previously mentioned is due to maltose rather than pentose, but at present the question is unsettled. Maltose reduces copper solutions, but not as strongly as does glucose. Ten c.c.of Fehling's solution are completely reduced by 0.0807 gram of maltose. It is much more strongly dextrorotatory than glucose and forms an osazone which crystallizes in large prism-like needles arranged in sheaves, and melts at 2070 C. This osazone is soluble in water and shows a dextrorotation, it being more distinctly identified by determination of the nitrogen content, which should equal 10.6 per cent. Maltose ferments with yeast only after inversion by heating with acid, the splitting products being two molecules of glucose. Other carbohydrates, such as dextrin, isomaltose, and saccharose, have been reported in the urine. These are extremely rare and need little comment in this place. In the case of cane-sugar the assimilation limit is so high that an alimentary saccharosuria could occur only after an enormous intake. Spontaneous excretion of cane-sugar has never been actually proven, but this sugar may be found in the urine of hysterical patients who have added it to deceive the physician. The so-called animal gum, first isolated by Land- wehr, seems to be a normal constituent of urine. Alfthan1 finds it is present in practically every case of diabetes to the extent of 1 to 37 grams per diem. This substance is probably not a definite chemical body, but a mixture of several. Little is known of its chemistry. Inosite was regarded for a long time as a carbohydrate, but it is now known to be a hexaoxyhexahydrobenzol with the formula C6H6(OH)6. This substance, has, therefore, nothing to do with true carbohydrate metabolism, but is discussed at this point as it has so long been regarded in this connection. Inosite enters into the composition of almost all animal tissues and occurs both in the optically active and inactive forms. A physiologic excretion of inosite is not infrequent, according to Hoppe-Seyler. It may occur in the urine in nephritis, diabetes mellitus and insipidus, and after a large intake of animal food. For its detection the writer must refer to works on physiologic chemistry. (/). Glycuronic Acid (CHO-(CHOH)4-COOH). Glycuronic acid is an intermediate product of the oxidation of carbo- hydrate, the CH2OH group being converted into CHO while the original CHO group is oxidized into COOH. This acid still retains the aldehyd group, in consequence of which it shows the same reducing action as does glucose. It seems to be characteristic of glycuronic acid that, when pro- duced naturally, it is never found in the free state, but only in the combined form as the conjugated glycuronic acid. It seems to be especially capable of combining with substances showing alcoholic or phenolic characteristics. The free glycuronic acid may be split off from its conjugated compounds by heating with acid and other hydrolyzing agents. The conjugated glycuro- nates are levorotatory while the free acid shows dextrorotation. Among 1 Ueber dextrinartige Substanzen im diabetischen Harn, Helsingfors, 1904. THE URINE. 303 the substances with which glycuronic acid combines we find chloral hydrate, butyl chloral hydrate, chloralamid, camphor, menthol, carbolic acid, resorcin, acetanilid, antipyrin, phenacetin, pyramidon, sandal-oil, morphin and cocain. The normal metabolism following intake of any of these substances is such that excretion of conjugated glycuronic acids will follow and may, therefore, lead to the assumption of sugar in the urine unless precautions are taken properly to differentiate these compounds. Besides conjugated gly- curonic acid of the above type we find a combination of urea with glycuronic acid as well as a certain amount of indoxyl, skatoxyl, phenol, and cresol in combination with this acid. Most of the products of bacterial decomposition in the intestine are excreted in combination with sulphuric acid, but some is in- variably present as a conjugated glycuronate. The origin and formation of glycuronic acid within the system is not entirely understood. It has been supposed to be derived from protein as especially advocated by Loewi, but Mayer has rather disproven Loewi's work, and shows that probably glycuronic acid is a direct derivative of glucose and that all carbohydrate oxidation must pass through the intermediate stage of glycuronic acid. The status of this question is very well summed up by Neuberg as follows: The formation of glycuronic acid out of protein is by no means excluded, nor yet from fat; but as it is difficult to eliminate the direct formation of glycuronic acid or its secondary development from previously existing grape sugar, Mayer justly contends that the question of glycuronic acid formation from these substances is practically included in the broader question of the formation of sugar from fat and protein. The exact point of conjugation of glycuronic acid is unsettled. It has been assumed by some to occur in the liver, while others find that the liver plays no part. It is probable that the synthesis takes place in various parts of the organism. It has been shown that the output of glycuronic acid may be increased in diabetes mellitus, in mild cases the unoxidized sugar being present largely in this form. Mayer1 advances the hypothesis of incomplete oxidation of sugar to explain its appearance in these cases.2 He shows that after the ad- ministration of glucose in amounts beyond the assimilation limit, an occasional excretion of glycuronic acid occurs with an equivalent diminution of the ethereal sulphates. It is possible that the substance conjugating with glycuronic acid is unknown and that we have the same results as though similar substances were introduced by mouth. As Mayer has advanced no direct proof of the correctness of his theory, the decision must be left for the future. Clinically, the question of the highest importance is whether the excretion of glycuronic acid is of any diagnostic value and whether it is of prognostic significance in diabetes. It does not seem wise to assume that a patient showing an occa- 1Zeitsch. f. physiol. Chem., Bd. 32, 1901, S. 518; Berl. klin. Wochensch., Bd. 40, 1903, S. 292 and 514. 2 See Conzen, Zeitsch. f. klin. Med., Bd. 75, 1912, S. 426. 304 DIAGNOSTIC METHODS. sional increase in the glycuronic acid excretion, which cannot be accounted for by intake or increased production of conjugating substances, will in the future show typical diabetes. Edsall does not believe in the value of glycuronic acid in the diagnosis of a latent diabetes, nor does Neuberg regard an increased excretion of glycuronic acid as the original derangement which may determine other deviations from health. Very few of the conjugated glycuronates show reducing action when treated with copper solutions using the precautions previously laid down. The chloral and camphor compounds are much more apt to produce typical reduction, but even these require heating for somewhat longer periods than does glucose. A diagnosis of a glycuronic acid excretion is based upon the following points: The fresh urine is levorotatory, but shows little or no reducing prop- erties and does not ferment. This same finding will be observed if £-oxybuty- ric acid be present so that a diagnosis may not rest on these findings alone. After being boiled with dilute acid for a period varying from one-quarter to three-quarters of an hour, the levorotation is changed to dextrorotation and the urine shows strong reducing powers. Such tests will not be given by /S-oxybutyric acid. In some cases after heating the urine with acid the action upon light may remain levorotatory or the solution may be optically inactive, on account of the fact that the conjugating substance may be levorotatory or that complete hydrolysis has not been effected. On heating the urine for some time with Bial's modification of the orcin test a positive reaction appears with the liberated glycuronic acid. This acid crystallizes with phenyl- hydrazin forming distinct yellow needles which melt at 114 to 1150 C. This test is, however, not readily obtained so that it is difficult to identify glycuronic acid by its osazone. It will be seen, therefore, that glycuronic acid is differen- tiated from the pentoses largely by its levorotation when in the conjugated state or dextrorotation when free. Tollens (Ber. d. d. chem. Ges., Bd. 41, 1908, S. 1788) has advanced a test which permits of a clear differentiation between glycuronic acid and pentose. To 5 c.c. of urine add 1/2 c.c. of a 1 per cent, alcoholic solution of naphthoresorcin and 5 c.c. of concentrated HC1. Warm over the free flame to boiling or place in the boiling water-bath for 15 minutes. Allow to stand for 4 minutes and then cool under running water. Add an equal volume of ether and shake vigorously. On separating, the ether layer shows, in the presence of glycuronic acid, a violet to blue coloration. Spectroscopic examination reveals a sharp absorption band at the D line. The test is not given by pentose. (See Tollens, Zeitsch. f. physiol. Chem., Bd. 56, 1908, S. 115; Jolies, Ibid., Bd. 81, 1912, S. 203; also, Neuberg and Schewket, Biochem. Zeitsch., Bd. 44, 1912, S. 502.) Goldschmiedt1 has introduced a test which Mayerhofer2 believes of great value to the pediatrician, as it indicates intestinal derangement better than indican. One-half to 1 c.c. of urine is treated with 2 drops of a 15 per cent. 1 Zeitsch. f. physiol. Chern., Bd. 65, 1910, S. 389; Bd. 67, 1910, S. 194. 2 Zeitsch. f. physiol. Chem., Bd. 70, 1911, S. 391. THE URINE. 305 alcoholic solution of a-naphthol. Carefully overlay 3 to 4 c.c. of concentrated H2SO4 with this mixture. A violet ring is observed, which changes to a dis- tinct emerald green on standing. In this test the urine must be free from nitrites and nitrates. Neuberg's Test.1 Five hundred c.c. of urine are treated with sufficient sulphuric or phosphoric acid to make the acidity from 1 to 2 per cent. This acidified urine is then heated in an autoclave for two hours at a temperature of 1150 C. The mixture is then cooled, neutralized with sodium carbonate, acidulated with acetic acid, and filtered. Two hundred and fifty c.c. of the filtrate are mixed with a hot aqueous solution of 5 grams of parabromphenylhydrazin hydrochlorate and 6 grams of sodium acetate. The mixture becomes cloudy at once, but on heating the cloudiness will disappear. As the mixture cools needle-shaped crystals will separate out and may be filtered off, the filtrate being again heated and cooled to obtain more crystals. This process may be repeated until no more crystals form. These crystals are then washed with distilled water followed by absolute alcohol, and are then recrystallized by dissolving in 60 per cent, alcohol and gradually evaporating. They are clear yellow in color, melt at 236° C., and show marked levorotation when dissolved in a mixture of pyridin (4) and absolute alcohol (6). This test is distinctive. (3). Acetone Bodies. By the acetone bodies3 we mean acetone, diacetic acid and /3-oxybutyric acid. The latter of these is the mother substance so that this group would better be called the /3-oxybutyric bodies. The chemical relation between these bodies is very close, the /3-oxybutyric acid being oxidized to diacetic acid, which then splits up into acetone and carbon-dioxid. This may be seen from the following formulae: CH3 - CHOH - CH2 - COOH /3-oxybutyric acid. CH3 -CO - CH2 -COOH Diacetic acid. CH3 -CO -CH3 Acetone. Formerly these substances were supposed to be derived from protein material, the /3-oxybutyric acid being formed from the /?-amino acids by desamidization and oxidation in the P position. This theory is, however, not generally held at present, being replaced by the more modern idea that the fats are the chief source of the acetone bodies. It has been found that in perfectly sound, well-nourished individuals the addition of fat causes only a very slight increase in the output of acetone bodies and, strangely enough, that butyric acid itself causes no acetonuria. On the other hand in normal individuals from whom the dietary carbohydrate has been removed or in a diabetic who is not utilizing what carbohydrate he may be allowed, a marked excretion of acetone bodies 1 Ber. d d. chem. Ges., Bd. 32,1900, S. 2395; Zeitsch. f. physiol, chem., Bd. 44,1905, S. 127. 2 See Waldvogel, Die Acetonkorper, Stuttgart, 1903; Magnus-Levy, Ergebnisse der klin. Med., Jena, 1908. 306 DIAGNOSTIC METHODS. may occur. While a portion of this acetone may possibly be derived from the carbohydrate groups of the protein molecules, it can hardly explain the enormous excretion in diabetes, as the amount of protein catabolism, as shown by the urinary nitrogen, is greatly insufficient to yield any such amount of acetone bodies. The fats are, therefore, the more probable source of these bodies. As long as the body is supplied sufficient carbohydrate or is able to oxidize a sufficient amount the acetone bodies of the urine remain low; but when the system is no longer capable of oxidizing the carbohydrates, the amount of acetone bodies increases to a marked extent. We see, therefore, that the older method of allowing diabetics practically no carbohydrate food may directly lead to aggravation of the symptoms which the withdrawal was supposed to remedy. In other words, the normal or the diabetic individual must have a certain amount of carbohydrate food in order that proper metabolism may be maintained. This is not the time or place to discuss the therapy or dietetic treatment of diabetes, but it must be remembered that the most successful diet is one which contains carbohydrates up to the point of tolerance. Even here we find that certain types of carbohydrates may be given to diabetics without increasing the glycosuria, while at the same time leading to a dimi- nution in the excretion of acetone bodies. Such diets are the oatmeal diet of von Noorden and the potato diet of Mosse. The condition arising from a surcharging of the blood with these acetone bodies is known as acidosis.1 For a long time it was supposed that the carbo- hydrates were not only accountable for glycosuria, but also for the acetonuria and acidosis noted in diabetes. In the advance of pathologic chemistry it has been shown that, instead of causing these latter symptoms and conditions, the carbohydrates in reality lessened them. This may be shown by the administration of a definite amount of sugar, especially in the milder types of diabetes, to patients from whose diet sugars have been previously excluded. The omission of sugars from the diet forces the organism to utilize its protein and fat and thus gives rise to an accumulation of nitrogenous and fatty metabolic products as well as to an increase in the acids of the body fluids. In this way an acidosis already present would be increased in intensity. If the carbo- hydrate-free diet be continued for some time, a readjustment takes place and the acetonuria may gradually diminish, as is instanced by the fact that certain races show no acetonuria even though on an absolutely carbohydrate- free diet. With regard to the proteins as the mother substances of these bodies, we must admit that their influence is to some extent a double one. In the first place protein tends to diminish the acetonuria on account of its carbohydrate content, those proteins containing the greatest number of carbohydrate groups not necessarily exerting the greatest effect either on this condition or on the glycosuria. With a diet excessive in protein the influence is, however, not of this sort. The sulphuric and phosphoric acids as well as the small amount of acetone bodies formed by the hydrolysis of the protein tend to increase 'See Straub, Deutsch. Arch. f. klin. Med., Bd. 109, 1913, S. 223. THE URINE. 307 an existing acidosis, while the carbohydrates formed in the splitting of these proteins may greatly increase an existing glycosuria. These points, together with the fact that the products of nitrogenous metabolism may greatly increase the osmotic tension of the blood and thus lead to disordered cell function, show us that proteins cannot be advantageous as an exclusive diet in diabetes. As is well known, the nitrogenous excretion is much more marked in a diabetic than in a nondiabetic owing to several factors. In the first place, the diabetic consumes more protein than the normal individual because his diet is limited as regards carbohydrates and must be made up to a requisite caloric value by protein and fat. Secondly, owing to the lack of the protein-sparing function of the carbohydrates, excessive protein is broken down and elaborated in order to furnish a portion of the energy necessary to maintain the body function. It must, however, be said that the diabetic protects himself for a time from the unusual loss of protein by the utilization of fat. Concerning the fats, it is to be recalled that, although formerly accredited with no power of influencing acetonuria, to-day they are regarded as directly affecting this condition to a great extent. This is true of the fatty acids, especially of the lower members, and not of the neutral fats. If the contention in Kastle and Loevenhart be true, that a reversible action of lipase converts the fatty acids and glycerin formed by a previous hydrolysis again into neutral fats, then the influence of fats on the acidosis is variable, or else we must assume a lack of lipase in the cells of the diabetic. We know that the fatty acids belong to the ketoplastic group (substances increasing excretion of acetone bodies), yet as Borchardt has recently shown this ketoplastic action is, doubtless, due to the union of the fatty acids with glycerin, thus withdrawing from the system the antiketoplastic body, glycerin, and enabling the remaining fatty acids to exert their influence on the formation of the acetone bodies. Fats do not increase an existing glycosuria as many experiments have shown, yet we must grant that a formation of sugar from fat does take place. Von Noorden speaks of a "facultative formation of sugar from fat," referring to the fact that the demand for sugar may become so great that this source is called upon to furnish its quota of carbohydrates. We must also remember that the syn- thesis of fat from disintegrated carbohydrate is much affected in diabetes. Were this not the case, a large part of the sugar, reaching the blood as such, would be synthesized by the fat-forming cells and glycosuria would be dimin- ished. Conceiving this latter function to be normal while the former is abnormal, we may readily see the close relationship between obesity and later diabetes. With an excessive diet of fat no more fat is oxidized than when the diet is low in fat. In the latter case, the body-fat is utilized to furnish the difference, while in the former the excess is deposited in the usual fat depositories. Besides the excretion of acetone bodies observed in diabetes, we find fever, carcinoma, inanition, lesions of the central nervous system, digestive disturbances, delayed chloroform poisoning, cases of pregnancy in which 308 DIAGNOSTIC METHODS. death of the fetus has occurred or in which persistent toxic vomiting is noted, and other conditions associated with an increased output of the acetone bodies. According to Mohr, most of the cases may be traceable either to limitation of carbohydrates in the food or to diminished power of utilizing them. In- creased protein catabolism may play a role in the pathogenesis of this condi- tion, but in the writer's opinion only a secondary one. A general statement should be that the excretion of the acetone bodies is little influenced by the amount of fat in the food of the normal individual, providing the carbohydrate content of the diet is good; but in pathologic conditions the factors influencing carbohydrate utilization are so numerous that fats play a much greater role than do proteins in bringing about an acetonuria. As /S-oxybutyric acid is the mother substance from which the other acetone bodies are formed by oxidation, we should expect to find, as we actually do, the severest cases showing large amounts of /?-oxybutyric acid and small or even no excretion of the other members of this group. As a rule, it may be said that the more acetone the less ^-oxybutyric acid but this is not always the case. In the diabetic coma we usually find large amounts of the first two members of this group while acetone may be absolutely lacking, and, on the other hand, we find in some of the milder types of diabetes acetone and no diacetic or oxybutyric acids. Dakin1 and Wakeman and Dakin2 have recently shown that the liver possesses two ferments by which the mutual interconversion of /?-oxybutyric acid and aceto-acetic acid may be effected, the one being an oxidizing process, the other a reduction process. The acidosis may, therefore, be more a result of the latter than of the former action. Chemically, acetone is dimethyl ketone. It shows, therefore, the reactions for this group of chemical compounds but is easily confused, both with the aldehyds and alcohols. The urine rarely shows typical reactions for acetone if the older tests are applied directly to the urine, so that it is necessary to distill and examine the distillate. In this process diacetic acid is split up into acetone and carbon dioxid, so that it is impossible to tell whether acetone was preformed or was produced by heating. From the clinical standpoint it is a matter of indifference, as acetone and diacetic acid are so closely related that their clinical significance is the same, acetone representing merely a further stage in the oxidation of this group of bodies. In these tests the urine must be perfectly fresh. If it is desired to eliminate the influence of the diacetic acid upon the acetone reaction, the urine may be alkalinized with sodium hydrate and extracted with pure ether. The ether removes the diacetic acid salt, enabling us to make a separate determination of the diacetic acid and acetone. (a). Acetone (CH3-CO -CH3). Legal's Test (Le Nobel's Test). To a few c.c. of the urine are added a few drops of a fairly concentrated solution of sodium nitroprussid and then sodium or potassium hydrate 1 Jour, of Biol. Chern., vol. 8, 1910, p. 97. 2 Ibid., vol. 8, 1910, p. 105. THE URINE. 309 until the mixture is strongly alkaline. A ruby-red color, later changing to yellow, appears in the presence of acetone. It will be remembered that this same test is given by creatinin, so that further modifications are necessary to permit of differentiation. If the ruby-red solution be treated with an excess of glacial acetic acid, the first red color will change into a carmine or reddish-purple color in the presence of acetone, while the same treatment with creatinin solutions yields a yellow, changing to green and finally to a blue coloration. As Le Nobel has found, ammonium hydrate does not give this reaction with creatinin, but with acetone, although the reaction is much slower in appearing. This test is given by diacetic acid, by alcohol, and by acetic aldehyd, so that it is not especially distinctive for acetone. It is perhaps better if this test is to be used at all that the urine be previously acidified and distilled, the distillate yielding a reaction which is more sensitive, according to Studer, although diacetic acid will thus be converted into acetone. Lange1 has modified this test as follows: To the suspected urine add 1/2 c.c. of glacial acetic acid and a few drops of a freshly prepared aqueous solution of sodium nitroprussid. Mix thoroughly and overlay-the mixture with 2 c.c. of concentrated ammonium hydroxid. At the point of contact a purplish-red ring is observed in the presence of acetone. Lieben's Test. To a few c.c. of urine2 or, preferably, of the distillate are added a few drops of concentrated sodium or potassium hydrate and a few drops of a solution of iodin in potassium iodid. On slightly warming the mixture yel- low crystals of iodoform will separate, which may be recognized by their characteristic odor as well as by their hexagonal shape when examined under the microscope. This test is given by alcohol as well as by aldehyds, and will show amounts of acetone varying between 1/100 and 1/1000 of a mg. This is a modification of the previous test and is much more specific, being given only by acetone. To the distillate from the urine are added a few drops of an alcoholic solution of iodin and the mixture treated with ammonia until a black precipitate of nitrogen iodid forms. On allowing the tube to stand for periods varying between 12 and 24 hours, this black precipitate disappears, leaving a yellow sediment of iodoform, which may be recognized as mentioned above. This test is less delicate than the original one of Lieben, detecting acetone when present in amounts of 1/100 of a mg. per c.c. of urine. Gunning's Test. Frommer's Test. Recently Frommer3 has introduced a test which seems to be distinctive 1 Munch, med. Wochs., Bd. 53, 1906, S. 1764. 2 Bardach (Zeitsch. f. physiol. Chem., Bd. 54,1908, S. 355) and, more recently, Rosenbloom (Jour. A. M. A., vol. 59, 1912, p. 445) have shown that protein and many of its derivatives prevent the formation of the characteristic iodoform crystals. Hence the distillate should always be used. 3 Berl. klin. Woch., Bd. 42, 1905, S 1008. 310 DIAGNOSTIC METHODS. for acetone and at the same time very delicate. It is based upon the fact that acetone reacts with salicyl aldehyd to form dioxydibenzoylacetone, accord- ing to the following equations. The alkali salt is distinctly red. The test may be performed as follows: Ten c.c. of urine are strongly alkalinized with potassium hydrate and 10 to 12 drops of a 10 per cent, solution of salicyl aldehyd in absolute alcohol are added and the mixture warmed to about 700 C. In the presence of acetone the fluid becomes yellow, then red, later purplish-red, and, on long standing, dark red. In the absence of acetone the color of the urine is practically unchanged. This test is said by Frommer to indicate the presence of one one-millionth of a gram in 8 c.c. of urine. Instead of applying the test as above, we may add about 1 gram of potassium hydrate (in the solid state) to 10 c.c. of urine and, without waiting for complete solution to occur, treat the mixture with 10 to 12 drops of the alcoholic solution of salicyl aldehyd and warm to 700. At the zone of contact of the alkali and the salicyl aldehyd an intense purplish-red ring is observed only in the presence of acetone. The writer has frequently used this test and finds it very satisfactory, as it does not react with diacetic acid unless the heating be long continued. It is quite as simple as the preceding tests and should find wide application. Quantitative Determination of Acetone. Many of the quantitative methods given for acetone are more or less complicated and at the same time do not yield absolutely accurate results. The writer selects, therefore, those proving most satisfactory in his hands. All of these methods, with the exception of Folin's, give the amount of preformed ace- tone as well as that derived, by distillation, from the diacetic acid. Huppert-Messinger Method. The principle of this method is the determination of the amount of iodin necessary to transform into iodoform the acetone derived in the distillation of the urine. Knowing this factor, a simple calculation yields the amount of acetone present. For this determination certain solutions are necessary: (i). A 50 per cent, solution of acetic acid. (2). A tenth-normal solution of sodium thiosulphate. In preparing this solution 24.8 grams of crystallized sodium thiosulphate (Na2S2O35H2O) are carefully weighed out and dissolved in distilled water, the solution being made in a volumetric flask and diluted exactly to the liter mark. (3). A tenth-normal solution of iodin. This solution requires exactly 12.685 grams of iodin in one liter. As iodin is difficultly weighable on account THE URINE. 311 of its volatility, it is advisable to weigh out approximately 13 grams of iodin and dissolve in approximately 1 liter of water to which has been added 25. grams of potassium iodid. This solution is then standardized by titrating against the previously made tenth-normal thiosulphate solution, using thin starch paste as an indicator and adding the iodin solution from a buret until the blue color of iodid of starch just appears. This determination is then con- firmed by duplicate estimations. Twenty c.c. of the iodin solution should be the equivalent of 20 c.c. of the thiosulphate solution, so that the necessary dilution of the iodin solution may be determined by the formula previously given under Determination of Chlorids in the Urine (page 185). One c.c. of the standard tenth-normal iodin solution equals 0.012685 gram of iodin and represents 0.967 mg. of acetone. Five hundred c.c. of normal acid urine or 100 c.c. of acetone-rich urine are treated with 2 c.c. of 50 per cent, acetic acid for every 100 c.c. of urine and distilled until 9/10 of the volume has passed over. The distillate is collected in a receiving flask, which is cooled with ice and which contains water to absorb the acetone. The flask is tightly closed with a doubly perforated stopper, through one hole of which passes the tube of the condenser reaching below the surface of the water and through the second opening a bulb filled with water to act as a safety bulb. The tube is then washed with distilled water and the fluid in the safety bulb is emptied into the receiving flask. It is important that the distilling flask be disconnected before the heat is shut off, as otherwise the fluid might suck back. This distillate is treated with calcium carbonate to remove any nitrous or formic acid which may have distilled over, and the mixture thoroughly shaken. This distillate is then acidified by the addition of 1 c. c. of dilute sulphuric acid (diluted eight times) and redistilled until one-tenth of the volume remains in the distilling flask. This second distillate is received in a flask arranged as in the previous distillation. It is then poured into a flask which can be closed with a tight-fitting glass stopper. Distillate and wash-water must not fill the vessel more than one-third full. A large excess of carefully measured N/10 iodin solution is added, the mixture well shaken, and strong sodium hydrate solution added drop by drop. The flask is then stoppered, shaken for one-fourth minute, and allowed to stand for five minutes. The stopper is then removed, the fluid clinging to it washed into the flask, and the fluid acidified with strong HC1. The excess of iodin is then determined by allowing N/10 thiosulphate solution to flow from a buret until the mixture is but slightly yellow, when a few c.c. of starch paste are added and the titration continued until the blue color just disappears. One c.c. of iodin solution used by the fluid corresponds to 0.967 mg. acetone, so that all that is necessary is to subtract the number of c.c. of thiosulphate used from the number of c.c. of iodine solution added and multiply by the above factor. The result is the acetone in the amount of urine taken. This method yields results which are from 5 to 10 per cent, low, so that Technic. 312 DIAGNOSTIC METHODS. slight variations in the acetone excretion, as determined by this method, have little significance. Folin's Method. This method1 yields only the preformed acetone occurring in the urine and does not regard the acetone which may be derived from the diacetic acid. If the urine be distilled and this method employed, the total possible acetone will result. It is somewhat more accurate than the previous method and is much more simple and less time-consuming. The writer finds it extremely serviceable and adopts it in all acetone determinations.' Twenty-five c.c. of urine are measured into an aerometer cylinder, similar to that used in Folin's ammonia apparatus, a few drops of io per cent, phosphoric acid, io grams of sodium chlorid, and a little petroleum are added. In the absorbing bottle which is fitted with an absorption bulb are placed 300 c.c. of water, 20 c.c. of 40 per cent, potassium hydrate, and an excess of tenth-normal iodin solution. The apparatus is connected with the filter-pump in the same manner as described under Ammonia and an air-current drawn through for one-half hour. The air should not be passed as rapidly as in the ammonia determination. Each worker should check his air blast by control estimations, using known solutions of acetone. In this way the acetone will be removed from the urine and converted, in the receiving flask, into iodoform. The contents of the receiving flask are acidified with concentrated HC1, using 10 c.c. of acid for every 10 c.c. of alkali previously used, and the excess of iodin titrated with tenth-normal sodium thiosulphate as in the previous method. If this method be employed with the distilled urine, subtract from the total acetone, obtained in the distillate, the preformed acetone, derived from the fresh urine, and obtain the acetone referable to diacetic acid. If this value be multiplied by 1.758 the result will be the amount of diacetic acid, as such, present in the urine taken. Shaffer has recently introduced a method for the determination of the total acetone, which is closly related to his method for estimating ^-oxybutyric acid and will be discussed in a later section. This substance is derived from ^-oxybutyric acid and is the precursor of acetone. As a rule, if acetone be present in large amounts, diacetic acid is also present and indicates a much graver condition than does the mere presence of acetone. If Folin's contentions are true, most of what we now call acetone is, in realty, diacetic acid, so that this latter substance should be tested for in routine urinary work. The remarks previously made regarding the appearance of acetone in the urine also hold for diacetic acid. This substance is very volatile and disappears from the urine in a relatively short time so that the tests should be applied to perfectly fresh specimens. (&). Diacetic acid (CH3-CO -CH2-COOH). 1 Jour, of Biol. Chem., vol. 3, 1907, p. 177. THE URINE. 313 Gerhardt's Test. To 10 c.c. of urine are added a few drops of 10 per cent, ferric chlorid solution. This is best added as long as a precipitate of phosphates occurs, these latter bodies being then filtered off. To the filtrate are added a few drops more of feme chlorid solution when the urine shows a Bordeaux-red color in the presence of diacetic acid. This color appears cherry-red by transmitted light and purplish-red by reflected light. Unfortunately, however, this test is not specific for diacetic acid. A red color is observed in the presence of cyantes, formates, acetates, phenol compounds, salicylates, the conjugated glycuronates of phenacetin, antipyrin, thallin, kryofin, and kairin, as well as meconic acid which may be excreted after intake of opium. It is necessary, therefore, to differentiate the red color produced by diacetic acid. If the urine be heated diacetic acid decomposes more or less completely into acetone and carbon dioxid, so that the red color due to diacetic acid will either disappear or become much weaker, while that with the other substances mentioned is not affected by heat. This same disappearance of color is noted if the mixture be allowed to stand from 24 to 48 hours. Schreiber recommends filtration of the urine through animal charcoal, using about 10 grams of charcoal to 100 c.c. of urine. Antipyrin, phenacetin, and kryofin are retained while sufficient of the diacetic acid passes through to permit of detection. The urine may even be acidified with sulphuric acid and extracted with ether. This ethereal solution, which contains the diacetic acid, is then shaken with water and ferric chlorid solution added. A deep red color will be seen in the watery layer in the presence of diacetic acid. If the red color, on addition of ferric chlorid, be due to the presence of meconic acid, it will disappear on the further addition of stannous chlorid or of alkali hypochlorites, while that due to diacetic acid is unaffected. This test is, perhaps, more frequently used than any other test for diacetic acid in the urine, but is by no means as serviceable as the following. Arnold's Test. For the performance of this test two reagents are employed: (i) a solution consisting of i gram of paraamidoacetophenon, 100 c.c. of distilled water, and 2 c.c. of concentrated hydrochloric acid; (2) 1 per cent, sodium nitrite solution. Fifteen c.c. of urine are treated with a mixture consisting of 10 c.c. of solution 1 and 5 c.c. of solution 2 and one drop of concentrated ammonia added. In practically all urine, whether it contains diacetic acid or not, a brownish-red coloration is observed which changes in the absence of diacetic acid into yellow if the mixture be treated with an excess of concentrated hydrochloric acid, while if diacetic acid be present the color changes to a beautiful purple on the addition of acid. If the mixture be shaken the foam also shows a distinct violet coloration. As this test is somewhat difficult in the presence of small amounts of diacetic acid, Lipliawsky1 has modified it as follows: Six c.c. of solution 1 1 Deutsch, med. Wochensch., Bd. 27, 1901, S. 151. 314 DIAGNOSTIC METHODS. and 3 c.c. of solution 2 are treated with the same volume of urine, a drop of ammonia is added and the mixture shaken, when it assumes a brick-red color. According to the probable contents of the urine in diacetic acid 10 drops to 2 c.c. of this mixture are treated with 15 to 20 c.c. of concentrated hydrochloric acid, 3 c.c. of chloroform and two to four drops of ferric chlorid solution. The test-tube is then closed with a cork and gently shaken for one- half to one minute. In the presence of traces of diacetic acid the chloroform assumes a characteristic violet coloration, while in the absence of this acid the color is yellow or light red. This test is positive for one part in 400,000 of water. Acetone and /Loxybutyric acid do not react with this test nor do the drugs previously mentioned under the Gerhardt Test interfere. If the urine is highly colored it is advisable to filter through animal charcoal. This acid, the mother substance of the acetone bodies, is found in the urine in extreme cases of the conditions described under Acetone. The amount excreted may vary from traces to as high as 100 grams. Kulz has reported an excretion of 225 grams in 24 hours. The free acid is practically never found in the urine, being excreted either as the ammonium, sodium, or potassium salt. These salts as well as the free acid are levorotatory and may be detected by the polariscope after previous fermentation of the carbohydrates. In the condition known as diabetic coma a specific intoxication with ^-oxybutyric acid is assumed as the causative factor. While this is undoubtedly true to a large extent it cannot be regarded as the only factor in diabetic coma, as administration of ^-oxybutyric acid in large quantities will not necessarily lead to such a syndrome unless largely retained. We do undoubtedly have an acidosis which acts by ultimately depriving the tissues of fixed alkalies and may be regarded, therefore, as of great importance in following a diabetic case. Were this acid, per se, accountable for the entire symptomatology, we should be able, by administering alkalies, to overcome the effect of the acid intoxication. In some cases such therapy is extremely beneficial, while in others it is practi- cally useless, as it cannot influence the formation of the toxic bodies and may not increase their elimination. As previously stated, the ammonia output of the urine is an invaluable guide in following the course of an acidosis, especially in diabetes. One gram of ammonia (NH3) is equivalent to 6.12 grams of /?-oxybutyric acid, especially when in excess of the amount directly due to the food. To detect /?-oxybutyric acid follow the method of Hart.2 To 20 c.c. of urine add 20 c.c. of water and a few drops of acetic acid and boil until the volume is reduced to about 10 c.c. (to remove acetone and diacetic acid). To the residue add sufficient water to make the volume 20 c.c. and place 10 c.c. in each of two test-tubes. To one of these add 1 c.c. of hydrogen peroxid, warm gently and allow to cool. Apply Lange's test for acetone (see p. 309) and allow (c). /LOxybutyric Acid1 (CH3-CHOH-CH2-COOH). 1 The more scientific name for this body is /?-hydroxybutyric acid. 2 Amer. Jour, of Med. Sci., vol. 137, 1909, p. 869. THE URINE. 315 the tubes to stand for a few hours. No reaction is observed in the tube which contains no hydrogen peroxid, while a distinct red zone is seen in the other. A large number of methods have been advanced for the determination of ^-oxybutyric acid in the urine. Among these we find the methods of Kulz, Tollens, Wolpe, Magnus-Levy, Bergell, Stadelmann, and Darmstadter. While some of these are accurate under certain conditions, so many precautions must be taken that widely varying results may be obtained. On the other hand, the method of Magnus-Levy, which is undoubtedly very exact, requires 24 hours, while that of Bergell depends to a large extent upon the condition of the powder which is extracted with ether. For these reasons the writer feels that certain newer methods are much to be preferred to the older ones. Quantitative Determination. This method1 is a modification of those of Magnus-Levy and of Bergell, but in the writer's hands has given much more satisfactory results. One hundred c.c. of urine are faintly alkalinized with sodium carbonate and evapo- rated in a porcelain dish to one-third or one-fourth of the original volume. The residue is then concentrated to about 10 c.c. on a water bath in order completely to remove the diacetic acid. This is then cooled, acidified with a few drops of hydrochloric acid, and made into a thick paste with plaster of Paris. This mixture soon begins to set, when it is stirred and broken up with a glass rod. This porous mass is then transferred to a Soxhlet apparatus and extracted with pure ether for two hours. The ether extract is evaporated, the residue taken up with water, decolorized with bone-black and filtered perfectly clear. The filtrate is then made up to 25 c.c. and its rotation determined with a polariscope. In determining the amount of /3-oxybutyric acid from its rotation we must make use of the following calculation. The specific rotation of the free acid is -24.12 in a decimeter tube. One division of the scale in the case of glucose equals 2 per cent., so that we may find the percentage of /3-oxybutyric acid by the following proportion: Black's Method. 2 : X :: 24.12 : 52.7. X = 4.37 per cent By a similar proportion it may be found that one division of the scale corre- sponds to 7.1 per cent, of sodium /?-oxybutyrate, whose specific rotation is-14.35. This method is just as exact as those of Magnus-Levy and Bergell and has the advantages that it is simpler, more reliable, and may be performed in shorter time. The greatest difficulty with this method arises in the deter- mination of the exact point of the scale at which the two portions of the polari- metric field are equally illuminated. Magnus-Levy has shown that a difference of 1/io of i° in reading the polariscope amounts to about three grams per liter of ^-oxybutyric acid, so that great care must be used with the polariscope in this method as well as in all others in which it is applied. 1 Jour, of Biol. Chern., vol. 5, 1908, p. 207. 316 DIAGNOSTIC METHODS. Shaffer's Method. This method1 seems to the writer to be the most desirable one that has been advanced. In the determinations which the writer has made by this method he has found it very reliable and less apt to give erroneous results if Shaffer's precautions are followed. Pribram,2 however, claims that the method shows a considerable error. The principle of the method is the oxidation of /3-oxybutyric acid to acetone and carbon dioxid and the determination of the amount of acetone thus evolved. The acetone and diacetic acid already existing as such in the urine are previously determined by this method, so that it serves as a general one for the acetone bodies. From 25 to 250 c.c. of urine, depending upon the amount of /?-oxybutyric acid expected, are measured into a 500 c.c. volumetric flask and an excess of basic lead acetate and 10 c.c. of concentrated ammonia are added. In selecting the amount of urine to be taken one should use sufficient to yield from 25 to 50 mg. of acetone derived from /?-oxybutyric acid. The solution in the flask is then diluted to the graduating mark, is thoroughly shaken, and filtered. Two hundred c.c. of the filtrate (representing 2/5 of the original volume of urine taken) are diluted with water to 500 or 600 c.c. 15 c.c. of concentrated sulphuric acid and a few grams of talcum are added, and the mixture distilled until 200 to 250 c.c. of distillate collects (distillate A). In this distillation, the distilling flask, which may be an 800 c.c. Kjeldahl flask, should be fitted with a dropping funnel and water run in to prevent the volume of fluid in the flask from becoming less than 400 c.c. This distillate (A) contains the preformed acetone and that from the diacetic acid as well as volatile fatty acids which may be present in the urine. To remove the fatty acids, especially formic acid, the distillate A is redistilled after adding 5 c.c. of 10 per cent, sodium hydrate solution. This distillate (T2) is then titrated with standard tenth-normal iodin and thiosulphate solutions as described in the Huppert-Messinger method for determination of acetone. The residue of urine and sulphuric acid from which A was obtained is again distilled, dropping in 400 to 600 c.c. of 0.1 per cent, to 0.5 per cent potas- sium bichromate3 solution. This bichromate must not be added faster than the distillate collects unless the boiling liquid turns a pure green color, indicating that the bichromate is being used up more rapidly. When about 500 c.c. of distillate (B) have collected, 20 c.c. of 3 per cent. H2O2 are added to the distil- late together with a few c.c. of sodium hydrate solution and this redistilled. This second distillate (B2) is then titrated with tenth-normal iodin and thio- sulphate solution. One mg. of acetone represents 1.794 mg. of /?-oxybutyric acid. 1 Jour, of Biol. Chem., vol. 5, 1908, p. 211. 2 Zeitsch. f. exp. Path. u. Ther., Bd. 10, 1912, S. 279 and 284. 3 Cooke and Gorslin (Jour, of Biol. Chem., vol. 10, 1911, p. 291) have recently advised the following modification. The residue of urine and sulphuric acid is diluted to 600 c.c. and 5 c.c. of a 5 per cent, solution of potassium bichromate are added. Distill down to 300 c.c. and add water through the dropping tube to 500 c.c. Distill until 500 c.c. have been obtained. Additional 5 per cent, bichromate is run in, a few cubic centimeters at a time, whenever the dis- tilling solution shows a greenish tinge. THE URINE 317 (4). Abnormal Pigments. (a). Blood Pigments. The principal blood pigment appearing in the urine is hemoglobin, which has been previously discussed under the heading of Protein in the Urine. Cer- tain derivatives of this pigment are found, however, in conditions in which hemoglobin does not appear (see hemoglobinuria). Hematoporphyrin. This is an iron-free derivative of hemoglobin and appears to be present in minute traces in normal urine. Pathologically, it has been found in cases of rheumatism, phthisis, Addison's disease, pericarditis, paroxysmal hemoglo- binuria, cirrhosis of the liver, exophthalmic goiter, croupous pneumonia, lead poisoning, syphilis, and many acute infectious diseases. Long-continued use of certain hypnotics, such as sulphonal, trional, and tetronal, is fre- quently associated with the appearance of hematoporphyrinuria. It is rather uncertain what is the causative factor in this condition. Urine containing hematoporphyrin is usually dark-red in color, but may vary from a brownish-red or port-wine color to a distinct Bordeaux-red. Ham- marsten has shown that this color is not entirely due to the hematoporphyrin, but partially to other abnormal pigments whose identity is not certain. In examining urine for the presence of hematoporphyrin the spectro- scopic method is practically the only one available. If this pigment be present in large amounts the urine may be directly examined with the spectroscope, showing the four bands of alkali hematoporphyrin discussed in the section on Blood. As this method is not always certain and not always easy of applica- tion, it would seem preferable to treat 50 c.c. of urine with 10 c.c. of 10 per cent, sodium hydrate solution. The precipitated phosphates carry down the pigments. This precipitate is then treated with 10 drops of concentrated hydrochloric acid and 15 c.c. of absolute alcohol. The solution is filtered if necessary and examined with the spectroscope when the two absorption, bands of acid hematoporphyrin will be observed. Normally, bile pigments do not occur in the human urine. As was pre- viously discussed in the section on Urobilin, biliary pigments may appear in the urine in conditions interfering with the passage of the bile into the intestine or when increased formation of biliary pigments from blood pigments has occurred and an associated obstruction of the bile-ducts is present. We see, therefore, that choluria occurs in every case in which there is obstruction to the outflow of bile into the intestine. Thus in catarrhal jaundice, biliary calculi in the common duct, carcinoma of the liver, and cirrhosis, bilirubinuria is frequent. Moreover, we may also find biliary pigments arising from purely hematogenous conditions, such as pernicious anemia, malaria, typhoid fever, arsenical poisoning, and yellow fever. Whether this latter type is not really hepatogenous in origin, as Stadelmann believes, is still un- (b). Biliary Pigments. 318 DIAGNOSTIC METHODS. settled, but it would seem more plausible to assume a primary breaking down of the red cells and a secondary insufficiency of the liver. The chief biliary pigment found in the urine is bilirubin, which is inter- mediate between hemoglobin and urobilin. On oxidation of this pigment, either in the system or in the methods of examination, various other pigments may arise. Thus we may find biliverdin, bilicyanin, bilifuscin, biliprasin, cholecyanin, and choletelin. In the fresh urine bilirubin is the only pigment noted, while any one of the others mentioned, especially biliverdin, may be found in older specimens. A bile-containing urine may show various shadings of color, ranging from a greenish-yellow, through yellowish-brown, deep brown, or greenish- brown, to a pure green. If the urine be shaken, a yellowish or greenish-yellow foam is observed, while in normal urines the foam is practically colorless. The presence of an excess of urobilin may also give a brownish foam. Urine containing bile always shows the presence of nucleoalbumin along with slight traces of serum albumin, so that bile should be regarded as a source of extraneous albumin. The sediment will usually be more or less colored by the biliary pigment, the casts in yellow fever, for instance, being usually distinctly bile-stained. Qualitative Tests. A large number of tests have been advanced for the detection of biliary pigments in the urine and many of them are distinctly unsatisfactory. If large amounts of bile be present, the urine, acidified with HC1, may be shaken out with chloroform which dissolves the bilirubin. If the chloroform be evaporated, rhombic crystals with rounded edges or distinct needles of a brownish-red color will be observed. These crystals will give the color tests mentioned below. This test has been described under other names, as those of Trousseau, Kathrein, Rosin, and Marechalt. A few c.c. of urine, acidified if necessary with acetic acid, are treated with a i per cent, alcoholic solution of iodin in such a way that the latter solution is superimposed upon the urine, forming a distinct line of contact. If bilirubin or other biliary pigments be present, a beautiful emerald-green color is observed at the point of contact. This test is especially recommended, but it is not very sensitive, indicating only one part of biliary pigment in 10,000 of urine. Certain drugs, especially antipyrin, may lead to the formation of a green color with this test. Thymol, if used as a preservative, may give rise to confusion with this and other bile tests. Smith's Test. Gmelin's Test. One or two c.c. of nitric acid are placed in a test-tube and the same amount of urine is allowed to flow from a pipet in such a way that a distinct contact line is formed. If the nitric acid contains a trace of nitrous acid and the urine biliary pigment, a distinct green ring will be observed at the line of THE URINE. 319 contact. In some cases, as the nitric acid oxidizes the pigment, a play of colors may be seen from green, through a blue, violet, and red, to a yellow. The primary green is the characteristic color, other colorations being occasion- ally due to pigments other than biliary. This test is supposed to indicate i part of bilirubin in 80,000 of urine. If the urine contains an excess of indican a deep-blue ring may be observed or the combination of this blue with the yellow of the urine may give a green. In the presence of skatoxyl a violet-red ring may be noted, while various medicaments may give colorations ranging through the entire spectrum. In such cases it is always advisable to extract the acidified urine with chloroform and apply the test either to the evaporated residue or to its aqueous solution. This test is, perhaps, more frequently used than any of the biliary tests, but requires considerable experience for its proper interpretation. If the urine be diluted the test is somewhat more distinctive. Many modifications of this test have been advocated, the most serviceable being the following: Bosenbach's Test. A large quantity of urine, which has been acidified with HC1, is filtered several times through a thick filter-paper, which will hold back the bile-stained elements of the urine. It is sometimes advisable to add a little milk of lime to the urine before filtering, instead of the HC1, as this will throw down the phosphates which will carry with them the biliary pigment. If the filter- paper and contents be dried by pressing with a second dry filter-paper and a drop of yellow nitric acid allowed to fall upon it, distinct rings will be seen, which will be colored as in the previous test, the green one being external. Nakayama's Test. This is a modification of the older Huppert test. Five c.c. of acid urine are treated with an equal volume of 10 per cent, barium chlorid solution and the mixture centrifuged. The barium chlorid precipitates the phos- phates and sulphates and carries down the biliary pigments. The superna- tant fluid is then poured off and 2 c.c. of the following reagent are added to the precipitate. The reagent consists of 99 c.c. of 95 per cent, alcohol, 1 c.c. of concentrated HC1, and 0.4 gram of ferric chlorid. If this mixture of precipitate and reagent be heated to boiling, a bluish-green or a brilliant green solution is obtained, which becomes violet or red on the addition of nitric acid. This test is said to indicate one part of bilirubin in 1,200,000 parts of urine. As the reagent in this test we use a mixture of one part of 25 per cent. HNO3 and 19 parts of 25 per cent. HC1. Before use one part of this reagent is mixed with four parts of alcohol. A few drops of the urine are added to this mixture, when it assumes a green color in the presence of biliary pigments. If the urine be treated as in Nakayama's test and the precipitate mixed with 1 to 2 c.c. -of the acid alcohol reagent and the whole centrifuged for a short time, a green solution is obtained if one part of bilirubin in 1,000,000 of urine be present. Hammarsten's Test. 320 DIAGNOSTIC METHODS. Bile Acids. The bile acids, taurocholic and glycocholic acids, are found in the urine in the form of their sodium salts. They may be found in small amounts in the same conditions in which the biliary pigments are present, but their amount is usually so small that they cannot be detected, as a rule, without being previously isolated. It seems to be fairly well established that the bile acids must be present to the extent of | per cent, before detection in the urine is possible. As the clinical significance of these acids is the same as that of the pig- ment and as their amount is so small that the methods of isolation require a fairly large volume of urine, the writer will refer to works on physiologic chemistry for such procedures. No absolutely reliable test is known for the detection of the bile acids in the untreated urine. There are, however, a few tests which are occasionally given providing the bile acids be present in sufficient amount. It need not be surprising if these tests usually fail, but occasionally one may be rewarded with a positive reaction. Hay's Test. This test depends upon the reduction of the surface tension of the urine in the presence of the bile acids. As advocated by Beddard and Pembrey, a pinch of powdered sulphur is sprinkled upon the surface of urine which should be preferably at a temperature not over 170 C. In normal urines the sulphur will float upon the surface, while if the urine contains bile acids the sulphur may sink at once indicating one part in 10,000 or may sink only after a few seconds to one minute, in this latter case indicating one part in 50,000. According to Sahli this test does not discriminate between biliary acids and biliary pigments, but clinically it is a matter of indifference which one is present. Phenol or aniline compounds lower the surface tension of the urine so that their presence may lead to wrong conclusions. Oliver's Test. This test is based upon the well-known property possessed by the bile acids of precipitating peptone when in acid solution. The reagent is as follows: Powdered peptone, 8.33 grams Salicylic acid, 1.12 grams Acetic acid, 2 drops Distilled water, 1 liter One to two c.c. of clear filtered urine are placed in a test-tube and treated with 5 c.c. of the reagent. In the presence of bile acids a decided milkiness appears at once, being the more intense the larger the amount of bile acids. Other tests, such as those of Pettenkofer and Udranzky, are only ser- viceable in the testing of the isolated bile acids, so that they must be looked for elsewhere. THE URINE. 321 (c). Melanin. In cases of melanotic tumors the urine not infrequently contains a chro- mogen (melanogen), which is converted into melanin on allowing the urine to stand or after adding alkalies or oxidizing agents. The urine when freshly voided is normal in color, but on exposure to air gradually darkens until it becomes distinctly black. This coloration is first noticed in the upper portion of the urine and gradually extends toward the bottom. If ferric chlorid be added to the urine the coloration may be somewhat intensified so that one may assume an excess of phenol derivatives. It is to be remembered that indican must first be split up with acid before giving the coloration with ferric chlorid; moreover, melanin is insoluble in chloroform while the indigo is readily soluble. (See Eppinger1.) The addition of the ferric chlorid may produce a black precipitate which is soluble in sodium carbonate solution, from which it may be reprecipitated by mineral acids. This pigment may also be found in some cases of chronic malaria2 so that it is not absolutely pathognomonic of melanotic tumors. {d}. Phenol Derivatives. As previously mentioned in the discussion of the variations in color of the urine, it may be dark colored either on voiding or after standing. While such colorations may be due to melanin, they are much more frequently trace- able to the presence of sulphuric acid in conjugation with phenol,3 paracresol, pyrocatechin, and hydroquinone. These substances are excreted especially in conditions associated either with increased intestinal putrefaction or with putrefactive processes elsewhere in the system. Aside from direct poisoning with these substances, they may be regarded as having much the same clinical significance as indican. Qualitative and quantitative tests for these substances are rarely of clini- cal importance. They may be roughly determined by estimating the amount of ethereal sulphates in the urine, when a large increase may be assumed to be referable to these bodies unless indican is greatly in excess. In certain conditions of disturbed protein metabolism, the urine contains pigments which cause it to turn dark on the addition of alkali or on standing. These urines are normal in color when voided and become black almost immedi- ately on the addition of alkali, whence the name alkapton bodies and alkaptonuria for their excretion in the urine. The substances causing this change of color are hydroquinon-acetic acid (as it has been prepared synthetically from gentisic aldehyd, it has been called homogentisic acid with the formula C6H3(OH)2(CH2COOH)) and uroleucic acid, whose structure has not been (e). Alkapton. 1 Biochem. Zeitsch., Bd. 28, 1910, S. 181. 2 Urriola (Interstate Med. Jour., vol. 19, 1912, p. 74) claims that a urinary excretion of blood pigment (probably hematin and not melanin) is invariably present as a pathognomonic sign of malaria. This may appear in the centrifuged specimen as a very abundant intense black pigment; a blue pigment, very constant but small in amount; or as an ochre pigment occasionally. 3 See Hensel, Zeitsch. f. physiol. Chern., Bd. 78, 1912, S. 373. 322 DIAGNOSTIC METHODS. absolutely settled.1 The former of these acids is present in all cases of alkap- tonuria, while the latter may be present in all cases, but in such small amounts that it remains undetected. Urines containing these bodies strongly reduce copper and ammoniacal silver solutions, while bismuth solutions are little affected. The so-called glycosuric acid of Marshall is probably identical with homogentisic acid and is not a definite chemical entity. This condition is observed at various periods of life and seems to be of theoretical more than practical interest, as of the 45 cases reported none ap- peared to be much disturbed in health, the condition usually being detected accidentally.2 The source of these substances is still uncertain. Baumann and Wolkow believe that the influence of specific bacteria in the intestinal canal upon the tyrosin, formed in the hydrolytic cleavage of the proteins, leads to direct formation of homogentisic acid. While this accounts for some of the ab- normal substance, it cannot account for the entire amount of 3 to 7 grams of homogentisic acid excreted in the 24 hours. From the work of Meyer, Falta, Langstein and Wohlgemuth it would appear that this condition depends not on an abnormal formation of homogentisic acid but on an incapacity for further oxidizing it when formed. Dakin3 believes that both factors play a part. However this may be, alkaptonuria has very few ill-effects, although transitory symptoms have been observed in occasional cases of diabetes, cirrhosis of the liver, tuberculosis, pyonephrosis and gastritis. The browning of the cartilage occurring in ochronosis is supposed by Albrecht and Zdarek to have some rela- tion to the alkapton bodies, although Langstein could not show the alkapton acids in the urine in such cases. Qualitative and quantitative tests for these acids are rarely necessary. The characteristic change of color on alkalinizing the urine is, at least, sug- gestive of homogentisic acid. Like all other hydroxy derivatives of benzol, this substance shows the Schiff reaction with ferric chlorid, producing a transitory blue coloration when present in amounts of 1 to 4,000. The reduction of copper solutions should not be mistaken for sugar, as homogentisic acid neither ferments nor shows any optical activity. The urine will usually show a rela- tively high acidity. As this substance is practically harmless in the system we should expect to find, as we actually do, no increase in the conjugated sulphates or glycuronates nor any increase in the ammonia content of the urine (Neuberg). ' (f). Ehrlich's Diazo Reaction. Under certain pathologic conditions the urine has been found to contain a chromogen which gives a deep red color to the urine when treated with diazo compounds and ammonia. This is known as the diazo reaction, the urinary substance causing it being somewhat uncertain. According to Bond- 1 See Morner, Zeitsch. f. physiol. Chern., Bd. 78, 1912, S. 306. 2 See Poulsen, Munch, med. Wochs., Bd. 59, 1912, S. 364; also, Baldwin, Amer. Jour, of Med. Sci., vol. 145, 1913, p. 123. 3 Jour, of Biol. Chem., vol. 9, 1911, p. 151. THE URINE. 323 zynski, the alloxyproteic acid is the causative factor, but this needs confirma- tion, as Clemens has apparently shown that the body producing the reaction is sulphur-free. In the performance of this test two reagents are used: (x) sulphanilic acid 5 grams, 50 c.c. of concentrated hydrochloric acid, and 1,000 c.c. of water; (2) 1/2 per cent, aqueous sodium nitrite solution. To a mixture consisting of 50 parts of solution 1 and one part of solution 2 is added an equal volume of the urine. The mixture is then shaken and about one-tenth of the bulk of ammonia added quickly, and the mixture thor- oughly shaken. The ammonia may be added in such a way that a line of contact forms, this latter method frequently bringing out a much more beauti- ful reaction. If the test be positive the entire urine will assume an intense red coloration or a colored ring will be observed at the point of contact between the ammonia and the mixture. With normal urine a distinct orange color may be observed. On shaking the mixture the foam will be more or less brilliant red in color, which is more characteristic than the red coloration of the mixture. On al- lowing the mixture to stand for 24 hours a green precipitate may be observed at the bottom of the, tube, which Ehrlich regards as especially characteristic of the true diazo reaction. This green precipitate is not always present and is not necessary for a positive reaction, the red coloration of the mixture and of the foam being more frequent. According to Greene, a mixture consisting of 100 parts of solution 1 to one of solution 2 renders the test more delicate. Instead of sulphanilic acid, paraamidoacetophenon may be used, as suggested by Friedenwald. The administration of certain drugs may markedly affect this test.1 Thus we find that naphthalin, chrysarobin, opium, and phenol derivatives give a reaction which is very similar to the true diazo reaction, but may be distin- guished, according to Wood, by the fact that color is more permanent in alka- line solutions, does not fade to any extent on the addition of a strong mineral acid, the foam is more yellow than in the typhoid reaction, and the green pre- cipitate does not appear on standing. Burghart has found that, following the administration of tannic acid, gallic acid, tannigen, and tannalbin, the diazo reaction disappears from the urine, the inhibiting effect being perhaps exerted upon the reagents used rather than upon the unknown factor which usually causes the coloration. According to Ehrlich, if the urine contains an excess of biliary pigment, a dark cloudy discoloration may occur, which is changed on boiling to a dis- tinct reddish-violet. He also finds on applying the diazo test in some cases that the urine and foam become yellow before the addition of ammonia. After ammonia is added the color changes to a lighter yellow. This reaction is known as Ehrlich's "egg-yellow" reaction, and is supposed to be due to the presence 1 Skorczewski und Sohn (Wien. klin. Wochs., Bd. 24, 1911, S. 1700) show that the urine of patients taking atophan gives this reaction. 324 DIAGNOSTIC METHODS. of urobilinogen. He regards this latter test as especially important in pre- dicting the crisis of pneumonia. This diazo reaction is never positive in health. It was formerly regarded as pathognomonic of typhoid fever, but it has been shown to occur in many other conditions. It is true that in typhoid fever it may be present as early as the third or fourth day and may persist for some time, reappearing if a relapse occurs, while the Widal test, as will be remembered, does not vary with a relapse. Moreover, the intensity of this reaction is somewhat parallel with the severity of the case, while the Widal reaction may not be present in the severest types of typhoid fever. A positive diazo reaction occurs fre- quently in measles, somewhat less frequently in pneumonia, miliary tubercu- losis, scarlet fever,1 diphtheria, and erysipelas, while in rheumatism and menin- gitis it is rarely obtained. Michaelis believes that the presence of a positive diazo reaction in pulmonary tuberculosis indicates a progressive condition with a grave prognosis.2 This reaction is, therefore, a valuable aid in diagnostic work. As it is present in about 80 per cent, of cases of typhoid fever, a negative reaction would not necessarily exclude typhoid nor would a positive reaction prove its presence. Its appearance in a relapse is of value, as the Widal test would give no information under such conditions. It is to be said, however, that the diagnosis should by no means rest upon this test any more than it should upon the Widal test. (g). Russo's Test. This test3 has been recently advanced and seems to have somewhat more diagnostic importance than has the diazo reaction. The technic is very simple and is as follows: Four drops of a i to i,oco aqueous solution of methylene blue are added to 4 or 5 c.c. of suspected urine. If the reaction be positive the mixture turns to an emerald or mint-green hue. A light green or bluish-green tint shows a negative reaction. The positive reaction is not affected by boiling the urine or by the previous ingestion of such compounds as phenacetin, salol, quinin, and calomel. The difficulty in the application of the test comes in the ability to recognize the various tints of green which may be present. With a little practice, however, a positive reaction may be readily detected, especially if a control test be made with normal urine. This test is shown as early as the second day of typhoid fever and persists throughout its course. The mint-green hue is first observed, the emerald- green tint appearing as the disease progresses. If the course is favorable the color tone becomes more and more bluish, while if unfavorable the emerald tint persists. This test is also given in measles, smallpox, chronic and suppurative tuberculosis, but is negative in varioloid, varicella, scarlet fever miliary tuberculosis, appendicitis, and malaria. 1 See Umber, Med. Klin., Bd. 8, 1912, S. 322; also, Woody and Kolmer, Arch, of Pediat., vol. 29, 1913, p. 18. 2 See Heflebower, Amer. Jour, of Med. Sci., vol. 143, 1912, p. 221. 3 Riforma Medica, Tm. 21, 1905; abstract Jour. A. M. A., vol. 45, 1905, p. 363. THE URINE. 325 This test is as simple and just as reliable as is the diazo-reaction, being especially valuable in differentiating a typhoid from a miliary tuberculosis. Rolph and Nelson1 have pointed out that urines containing bilirubin react to this test, so that this fallacy must be borne in mind. (h). Dimethylaminobenzaldehyd Reaction. This test, advanced by Ehrlich, is as follows: Prepare a 2 per cent, solution of p-dimethylaminobenzaldehyd in equal parts of concentrated HC1 and water. Add a few drops of this solution to 5 c.c. of fresh cold urine and allow to stand for a few minutes. A positive reaction is indicated by the appearance of a cherry-red color, which may be extracted with chloroform or epichlorhydrin. Heating facilitates the reaction, but here normal urine may give a slight reddish coloration. In the cold normal urine gives a greenish-yellow color. Apparently this coloration is due to the presence of metabolic products derived from blood pigments (urobilinogen compounds, see p. 243).2 It is to be expected, there- fore, that this reaction would be distinct in diseases of the liver and bile passages, although it is not constant even here. It is, also, observed in tuberculosis, pneu- monia, typhoid fever, and malaria. It has little clinical value. (i). Drug Reactions. After the administration of a large number of drugs, conjugated glycuronates or sulphates may appear in the urine which give rise to various colorations. The more important of these have been previously discussed so that the writer will refer thence for the treatment of this subject. The methods of detecting these various drugs, as such, must be found in works on pharmaceutical chemistry. Fig. 88.-Purdy electric centrifuge. The microscopic examination of the urine is important in 'every case. So varied are the elements which appear in the microscopic field that con- siderable experience is necessary before absolute interpretation can be made. Not infrequently the character of the sediment will change a diagnosis, as, for instance, when pus- or blood-cells are present in sufficiently large amounts to account for an albuminuria previously determined by chemical methods. IV. Microscopic Examination of the Urine 1 Medical Record, vol. 80, 1911, p. 373. 8 Fischer and Meyer-Betz (Zeitsch. f. physiol. Chem., Bd. 75, 1911, p. 232) show that the reaction depends on the presence of a dipyrrylphenylmethane pigment of the class of non-stable pyrrol derivatives, to which group the blood and biliary pigments, as well as urobilinogen, belong. 326 DIAGNOSTIC METHODS. It is, therefore, essential that the microscopic examination of urine form a part of the ordinary routine. In obtaining the sediment of the urine for microscopic examination, two methods are possible. In the first place, the urine may be allowed to stand in a conical glass for periods ranging from 12 to 24 hours. The sedi- ment originally present in the urine as well as that formed by chemical changes taking place during the standing will collect in the lowest portion of the glass and may be removed by a glass-tube drawn out to a somewhat small point. It is to be remembered that a sediment at the end of 12 to 24 hours may be entirely different from that originally present in the freshly voided specimen. The changes in the reaction of the urine will, necessarily, lead to the dis- solving of certain types of crystals and to the formation of other varieties. Moreover, casts, if originally present, may dissolve or disappear as a result of the reaction of the urine becoming alkaline. For these reasons it is ab- solutely essential, in the use of this gravity method, that preservatives be added to the urine. Among the preservatives which may be used to prevent bacterial action during the sedimentation, we find a small piece of camphor or a rather large crystal of thymol serving the purpose. Some workers add one-fifth the volume of a 4 per cent, solution of borax which is equally useful, but the addition of chloroform or formalin, does not serve as well in these cases, as the former does not completely preserve the casts and the latter introduces a crystalline compound of formalin and urea which may be confusing, as it is not unlike impure leucin. This method is not to be recommended for routine work, as it requires too long a period and as changes may occur which make it absolutely impossible to differentiate between a primary and a secondary sediment. Secondly, the sediment may be thrown down by the use of the centrifuge. This ap- paratus is seen in the accompanying cut. By the use of this method no preservative is needed, a deposit is obtained within three minutes in a much more concentrated form, and changes in the sediment do not take place. The writer would recommend, therefore, the use preferably of the electric centrifuge or, at least, of the type run by hand. Whatever method may be adopted for obtaining the urinary sediment, the next steps in the process are the same. A pipet, consisting of a glass tube drawn out to a point about one-half the diameter of the tube, is introduced to the bottom of the vessel containing the Fig. 89.- Sediment tube for Purdy cen- trifuge. Fig. 90.-Per- centage c e n t r i - fuge tube. THE URINE. 327 sediment, a finger being placed over the upper end to prevent fluid passing into the tube as it is introduced. When the tip of the pipet comes in contact with the deposit, the pressure of the finger on the upper end of the pipet is removed and the deposit allowed to flow up into the pipet. The finger is then placed tightly over the tube as it is withdrawn from the fluid. By placing the tip of the pipet in contact with a perfectly clean slide which is' absolutely free from scratches and by gradually rotating the pipet, a small portion of the sedi- ment is obtained. A cover-glass is then placed upon this drop avoiding any undue pressure, which might distort the organized elements of the sediment. Some workers dispense writh the cover-glass and use somewhat larger amounts of sediment, but the writer does not find this method as acceptable owing to the fact that the lens of the objective may dip into the fluid and thus give indefinite microscopic pictures. Moreover, the focus cannot be as accurately adjusted without the use of the cover-glass. In the examination of the microscopic specimen, prepared as above, the point of special importance to be observed is the proper adjustment of the light. The writer is accustomed to use the low-power objective in the pre- liminary examination. In this case it is essential that the light be shut off to a large extent, as the recognition of casts cannot be made in a brilliantly illuminated field. This examination with the low power has the advantages that larger visual fields are subject to inspection, casts are easily recognized and crystalline deposits, as well as morphological elements, are usually differen- tiated. After the preliminary examination with the low power, the final examination is made with the high-power dry lens. In this way elements which appear suspicious under the low power are more clearly brought out and differentiations made possible between various types of cellular elements. With the high-power lens it is, of course, essential that the field be somewhat more illuminated than when the low-power is used. While a mechanical stage is, at times, advantageous in the microscopic examination of the urine, the writer has found that the fingers serve practically every purpose in the manipulation of the slide under examination. Where the urine is to be examined for the presence of bacteria, stained specimens must be made and examined with the oil-immersion lens. Urinary sediments are classified into two divisions: (a) chemical or nonorganized and (b) anatomical or organized sediments. The nonorganized sediments exist in solution in normal urine and appear as deposits under conditions of excessive formation, excessive excretion, or of alterations in the urine affecting its solvent properties. The chief chemical sediments are uric acid and its salts, calcium oxalate, phosphates, sulphates, cystin, leucin, tyrosin, xanthin, fat, and fatty acids. The organized sediments are usually foreign substances and are not met with in normal urine. They consist of epithelial cells, pus corpuscles, blood-cells, renal casts, spermatozoa, infusoria, bacteria, and tissue fragments. 328 DIAGNOSTIC METHODS. (A). Unorganized Sediments. {a). Those Appearing in Acid Urine. This occurs as a sediment in the urine under 3 conditions: (1) great concentration, (2) high acidity, and (3) low temperature. The deposit differs from others in possessing a deep yellow or orange-red color, although some of the smaller crystals are occasionally colorless. The primary form of the uric acid crystal is that of the rhombic prism. Modifications of this, in the form of square plates, cubes, ovoids, dumb-bells, or whetstone crystals are sometimes noticed. A rare type, especially of the colorless crystals, is a per- fect hexagon which resembles cystin so closely that chemical means of differen- tiation must be used. The crystals may be single or grouped in rosettes or (1). Uric Acid (C5H4N4O3). Fig. 91.-Various forms of uric acid. 1, Rhombic plates; 2, whetstone forms; 3, quadrate forms; 4, 5, prolonged into points; 6, 8, rosettes; 7, pointed bundles; 9, barrel forms pre- cipitated by adding hydrochloric acid to urine. {Hawk.) fan-shaped masses. Occasionally typical needle-shaped crystals may be seen which are arranged in sheaves. The microscopic peculiarities of uric acid are usually such as to permit of its easy recognition. In some cases, however, it is wise to confirm the micro- scopic finding by the inurexid test as follows: Place a small quantity of the sediment in an evaporating dish and add a few drops of concentrated nitric acid. Evaporate on the water-bath to dryness, when a yellowish or reddish residue will remain. Allow the residue to cool and add a few drops of ammonium hydrate solution. In the presence of uric acid a distinct reddish purple color will appear. If water be added to this purple solution and the mixture evaporated to dryness the color disappears. This latter point is of importance as xanthin, which may resemble unusual types of uric acid in THE URINE. 329 microscopic appearance, also gives the murexid test, but the color does not disappear on heating with water. (2). Sodium Acid Urate (CsH3NaN4O3). This salt of uric acid forms the bulk of the "brick-dust deposit" or "sedi Fig. 92.-Acid sodium urate. (Hawk.) mentum lateritium" found when urine has cooled. In such cases the urine first shows a milky appearance and the sediment soon settles on the sides and bottom of the container. This deposit is usually in the form of irregular amorphous granules of a brownish or pink color. Occasionally the sediment Fig. 93.-Xanthin. (Hawk.) may be distinctly crystalline, occurring as prismatic needle-like crystals which are grouped in star-shaped, fan-shaped or dumb-bell-like clusters. (3). Potassium Acid Urate (CSH3KN4O3). This substance occurs only as a granular amorphous deposit. Owing to its greater solubility, it does not form as large an amount of the brick-dust deposit as does the sodium salt. 330 DIAGNOSTIC METHODS. These two latter sediments are occasionally associated with amorphous deposits of the calcium and magnesium acid urates. These are, however, rare and need not be separately considered. In detecting the presence of the urates in a deposit, a small portion of the turbid urine is poured into a test-tube and gently heated. If urates are present the sediment will completely dissolve. These salts also give the murexid test. This substance which is chemically closely related to uric acid is rarely found as a sediment in the urine. Its chief clinical importance is found in its appearance as a urinary calculus. It crystallizes in whetstone-shaped colorless crystals which resemble those of uric acid, from which it is differen- tiated by its solubility on heating and in hydrochloric acid as well as ammonia. It may be chemically recognized by Weidel's reaction. Place a portion of the suspected crystalline deposit in an evaporating dish and dissolve by warm- ing with a few drops of bromine water. Evaporate to dryness and place the dish containing the residue under a large beaker, allow the fumes of ammonia to fill the inverted beaker when a red or purplish-violet color will be produced in the presence of xanthin. (4). Xanthin (C5H4N4O2). (5). Calcium Oxalate (CaC204). This substance appears most frequently in acid urine, but may be found after the urine has undergone alkaline fermentation. If it occurs in acid urine it is associated with uric acid; if in alkaline urine, with the triple phosphates. The deposit is a colorless crystalline one having two distinct forms: (i) octahe- dral crystals (four-sided pyramids lying base to base); viewed from the side, these appear as squares crossed by two sharp lines, giving the so-called "en- velope" crystal. (2) Dumb-bell crystals in the form of ovoid or circular disks with round margins depressed at the centers. These latter often present radial striations. Emerson has called attention to a rare type of calcium oxalate crystal which appears in the form of flat plates with parallel sides and rounded ends, looking like superimposed sheets of mica. It is character- istic of the crystallization of this substance, as of most crystalline urinary deposits, that the crystals are practically always of the same type, variations rarely appearing in the same specimen of urine. Fig. 94.-Calcium oxalate. (Hawk after Ogden.) THE URINE. 331 These crystals are insoluble in acetic acid, but soluble in hydrochloric acid. This is a point of some importance, as it is occasionally difficult to distinguish microscopically between calcium oxalate and some crystals of triple phosphate. This latter crystal is soluble in acetic acid. These crystals may be chemically identified by dissolving them in hydrochloric acid, alkaliniz- ing with ammonium hydrate, and precipitating with ammonium oxalate. The appearance of cystin in the urine is known as cystinuria. This is a condition of perverted protein metabolism which is not well understood. It may make its appearance at any period of life and in either sex, perhaps somewhat more frequently in the male. It shows a remarkably frequent hereditary character, being observed in some cases through several genera- tions and in several children of the same parents. Many of these cases do not show any clinical characteristics, being present during their entire life without any apparent symptoms. In other cases, owing to the formation of calculi, frequent manifestations are noted and surgical intervention interposed. This condition is not very frequent, being reported only 180 times in the literature. "Prior to the time when cystin was found to be a product of the disintegra- tion of protein substances it had been conjectured, on the ground of its content in sulphur and nitrogen, that it might be a product of the intermediate pro- tein metabolism. The explanation which von Udranzki and Baumann gave (6). Cystin (C3H6NSO2)2. Fig. 95.-Cystin. {Hawk after Ogden.) of its excretion corresponds in principle to that which is familiar to-day for the appearance of conjugation products of glycuronic acid and of glycocoll in urine. Like the latter, it should be normally further oxidized, and only in the presence of definite bodies should it be intercepted. These authors considered the binding substances to be the previously mentioned diamins, especially putrescin and cadaverin, which they had found in the urine and in the feces of a cystinuric patient. The formation of the diamins was sup- posed to be brought about by specific bacteria in the digestive tract by an ex- traordinary chronic intestinal mycosis. The resorbed part of the cadaverin and putrescin was supposed to protect the cystin from combustion, just as benzoic acid does glycocoll, by entering into a loose combination which de- composes again after passing through the kidneys. Serious difficulties, how- 332 DIAGNOSTIC METHODS. ever, have opposed themselves to this interpretation of cystinuria. First, numerous cases have been described without simultaneous diaminuria, and, conversely, diaminuria occurs in malaria, and in the conditions brought about by the cholera vibrions and the Finckler-Prior bacillus, without cystinuria having ever set in, any more than it does when diamins themselves are administered." "Further investigations have shown that cystinuria is really a disturbance of the amino-acid metabolism. Of the end-products of protein hydrolysis which arise in the system, the cystinuric cannot avail himself in the normal way of the cystin, and in part excretes it; the remaining products of protein hydrolysis undergo their ordinary fate. If free monomolecular a-amino acids appear in places which are at present not well known, or if they occur there even in unusual amounts, then, unlike the normal individual, the cystin- uric is unable to burn them and they leave the organism unchanged just as cystin itself does. The basic diamino acids behave in practically the same way, except that the CO2 group is split off from them and we arrive at diam- inuria" (Neuberg). Crystals of cystin are rare in the urinary sediment. In some of the cases reported the cystin did not separate from the urine until this was acidified with acetic acid and allowed to stand for 12 hours. It crystallizes in two forms: (1) six-sided tablets having an opalescent luster and sometimes traced with fine lines of secondary crystallization; (2) four-sided square prisms lying separately or in stellate forms. These crystals are soluble in hydrochloric acid, alkaline hydrates, and insoluble in acetic acid. These tests differentiate it from uric acid. If the urinary sediment suspected of containing cystin be treated with strong'sodium hydrate solution and a few drops of benzoyl chlorid, and the mixture shaken vigorously, a voluminous precipitate of benzoyl cystin is obtained. Chemically leucin is a-aminoisobutylacetic acid. It occurs in the urine in conditions associated with more or less marked derangement of hepatic function (see amino acids). As found in the urine, leucin appears in the form of yellowish, highly refractile spherules, with alternating light and dark concentric layers and with radial striations. In the pure state it crystallizes either in thin, white, hexagonal plates or as scales or rosettes of irregular shapes. Leucin is soluble in water, acids, and alkalies, and insoluble, to a more or less extent, in alcohol. Not always do we find crystals of leucin in the sedi- ment when the urine contains this substance. If it be suspected, the urine should be evaporated to a small bulk and alcohol added to the residue, which may then be examined for the characteristic crystals. This leucin may be identified by Scherer's test as follows: Some of the solid residue obtained by concentrating the urine to a small bulk is evaporated with concentrated nitric (7). Leucin (C6H13NO,). THE URINE. 333 acid on a platinum crucible cover. With pure leucin the residue remains colorless, but as usually applied to the urine a yellowish residue obtains. This is heated with a few drops of sodium hydrate solution, when a yellowish or brown color will be observed. If further heating be applied the leucin will collect into an oily drop which rolls around on the heated surface. As leucin does not stain with Sudan-Ill it should not be confused with fat. Fig. 96.-Pure leucin. {Hawk.) Chemically, tyrosin is p-oxyphenyl-a-amino-propionic acid. As found in the urine tyrosin crystallizes in the form of fine colorless needles, which may appear black and are arranged in sheaf-like collections or rosettes. Like leucin, it may not crystallize out unless the urine be concentrated. Tyrosin (8). Tyrosin (C9HUNO3). Fig. 97.-Impure leucin. {Hawk after Ogden.) is soluble in water, acids, and alkalies, while it is slightly soluble in alcohol and insoluble in ether. As other crystals, which may appear in the urine closely resemble the tyrosin needles, it is advisable to confirm the microscopic findings by chemical tests. This may be done by evaporating the urine to a small bulk, removing the fluid and dissolving the residue in water. Mower's test may then be applied as follows: To this aqueous solution is added i c.c. 334 DIAGNOSTIC METHODS. of a reagent consisting of 1 c.c. of formalin, 55 c.c. of concentrated sulphuric acid, and 45 c.c. of water. If the mixture be heated to boiling a beautiful green color will be observed in the presence of tyrosin. Fig. 98.-Tyrosin. {Hawk.) (9). Calcium Sulphate (CaSO42H2O) This is a very rare sediment, appearing only when the urine is extremely acid. The crystals appear in the form of long, thin, rhombic plates or needles which may be single, but are more frequently observed in clusters. If the sediment be boiled with hydrochloric acid and barium chlorid added, a precipitate of barium sulphate will point to the presence of calcium sulphate in the sediment. Fig. 99.-Calcium sulphate. {Hawk after Hensel and Weil.) Fig. ioc.-Bilirubin (Haematoidin). {Hawk after Ogden.) Bilirubin or its isomer hematoidin may appear in the urine in conditions previously discussed. The type of crystal is either a brilliant yellow or ruby- red rhomb or a yellow needle. Rarely the deposit may be in the form of a (io). Bilirubin (C16H1SN2O3). THE URINE. 335 yellow granular sediment. Not infrequently small curved needle-like spines are observed projecting from the angles of the rhombic crystals. These crystals may be identified by extracting the acid urine with chloroform and applying the tests previously discussed under Biliary Pigments. (11). Hippuric Acid (C9H9NO3). This substance has been observed as a sediment, although rarely. It ap- pears in the form of semitransparent, colorless, four-sided prisms, or in long pointed rods or needles, occasionally in forms closely resembling those of the triple phosphates to be described later. These crystals are soluble in warm water, alcohol and ether and may be distinguished from uric acid by the fact that they do not give the murexid test. (12). Neutral Calcium Phosphate (CaHP042H20). This substance is found only in faintly acid or neutral urine. It is quite rare as a sediment, crystallizing in colorless needles or slender pyramids which group themselves together with their points in a common center to form rosettes or cross-shaped figures. These crystals are soluble in acetic acid and may be converted into calcium carbonate when treated with a strong solution of ammonium carbonate. Under normal conditions the urine contains no free fat, but amounts varying from traces to rather large excretions may be found under pathologic conditions.1 The excretion of fat in the urine is known as lipuria. This is characterized by the presence of small or large strongly refractile globules which may be stained black with osmic acid or red with Sudan-Ill. These globules are soluble in ether and may, therefore, be extracted from the urine by shaking out with this solvent. It not infrequently happens that the urine is contaminated with fat which may have been used in obtaining catheterized specimens or with fat coming from the bottle containing the urine. This may lead to a diagnosis of lipuria unless care be taken to exclude such a source. True lipuria has been observed in various conditions. Thus we may find after a large intake of fat in the diet or as a therapeutic agent the so-called "alimentary lipuria." Pathologically, it has been observed in various cachectic conditions, in crushing injuries, especially of the bones, in eclampsia, in chronic heart disease, fatty tumors, diabetes mellitus, tuberculosis, various affections of the pancreas and liver, nephritis, and after the use of various general protoplasmic poisons. In these cases the blood may also contain an excess of fat, although this has not been observed in all cases. In fatty degeneration along the genitourinary tract, fat droplets may be seen in the epithelial cells and in the casts. Free fat is rarely found in such conditions, but occasion- ally it may collect in droplets which float on the surface of the urine and then constitutes true lipuria. Occasionally flat superimposed plates with notched corners (cholesterin) may be seen and may be so numerous as to justify the term "cholesterinuria." (13). Fat. 1 See Sakaguchi, Biochem. Zeitsch., Bd. 48, 1913, S. 1. 336 DIAGNOSTIC METHODS. In conditions associated with infection by the filaria, large amounts of fat may be present, giving rise to the appearance of an emulsion. To this condition has been given the name chyluria. In this form the fat may be present occasionally in large masses resembling tallow, but more frequently is seen in finer clumps of globules. The appearance of the urine is much like that of skimmed milk, but may have a reddish tinge due to the presence of blood. On allowing the urine to stand, a cream-like mass of fat will rise to the surface. It is not unusual in such cases to find the ova or the parasite in the masses of coagulated material. The excretion of the fatty material Fig. ioi.-Cholesterin. (Hawk.) at times runs a somewhat cyclic course, being present during the day and absent at night or vice versa. Occasionally the excretion varies with the position of the patient, being somewhat more frequent when he is erect, and may be markedly increased after severe exercise. This condition should be taken simply as a symptom of filariasis. A nonparasitic type of chyluria has been observed, but its etiology is somewhat uncertain. It probably is closely related to the conditions above mentioned as causing true lipuria. It does have some relation to an increased fat diet and apparently is associated with exudation from the lymphatic vessels, as the cellular elements are largely lymphocytes. (ft). Those Occurring in Alkaline Urine. (i). Ammonium Urate (C5H3(NH4)N4O3). This sediment occurs most frequently in combination with amorphous calcium phosphate and triple phosphate crystals. It is the only urate deposit found in alkaline urine, but may occur in neutral urine. It appears as a crystalline deposit of dark brown spherical masses studded with fine spiculse, from which fact the name of "thorn-apple" crystals has been given to them. PLATE IX. Ammonium Urate, showing Spherules and '1 horn-apple-shaped Crystals. (From Ogden, after Peyer.) THE URINE. 337 Occasionally these spheres may show concentric or radial striations. Not infrequently one observes crystals having irregular shapes, such as those of a dumb-bell or a pear. Chemically, these crystals may be identified by dissolving in hydrochloric acid, when uric acid, which may be identified by the murexid test, will separate. If sodium hydrate be added to the dry sediment and heat applied, vapors of ammonia are given off. (2). Calcium Triphosphate (Ca3(PO4)2). This compound is frequently found in alkaline urine, as a white amorphous flocculent deposit arranged in irregular patches. This is the usual deposit which appears in the urine when it becomes alkaline after meals. In the so-called " phosphaturia" the urine is always turbid when voided so that the assumption was made that an excess of phosphoric acid was being excreted. Such is found not to be the case, as a deposition of the normal phosphates of the urine must occur when the reaction becomes alkaline. It is to be said in this place that no conclusion whatever can be drawn from the separation of a substance in the sediment as regards an increase in its excretion. So many factors influence the separation or nonseparation of a sediment that a finding should not be regarded as evidence of increased formation and excretion unless quan- titative chemical examination points in this direction. Calcium phosphate is soluble in acetic acid without evolution of gas, which test may be used to show the presence of this substance in the deposits. It may be absolutely identified as a calcium compound by dissolving in acetic acid and precipitating with ammonium oxalate; the phosphoric acid radical may be proven by dissolving the sediment in nitric acid and precipitating with ammonium molybdate. Theoretically this compound appears along with calcium phosphate as an amorphous deposit in alkaline urine. Its amount is, however, usually less than the latter compound. It is observed in rare cases as large, long, rhombic plates with beveled edges which closely resemble the crystals of triple phosphates. These crystals are found in cases in which not sufficient ammonia is present to form the true triple phosphate, and may be considered as transition crystals. (3). Magnesium Phosphate (Mg3(PO4)2). (4). Magnesium-ammonium Phosphate (Mg(NH4)PO4). The appearance of this substance in the urine is essentially characteristic of ammoniacal urine. It may very rarely be seen in amphoteric urine when ammonium salts are present in large amounts. The crystals belong to the rhombic system, appearing most frequently as triangular prisms or "coffin- lid" crystals. These may be shortened in the form of squares or one or more corners may be rounded or beveled. By refracted light a greenish tone is observed when these crystals are present. A second type of the "triple phos- 338 DIAGNOSTIC METHODS. phate" is that of a star-shaped feathery crystal with points not unlike fern- leaves. These crystals are easily soluble in acetic acid and may be identified by treating with sodium hydrate and warming when ammonia is evolved. Fig. 102.-Magnesium-ammonium phosphates. {Hawk after Ogden.) (5). Calcium Carbonate (CaCO3). This substance frequently occurs in alkaline urine in association with the amorphous phosphates. It may appear as groups of amorphous material or may form large spheroidal masses with concentric radiations. Occasionally Fig. 103.-Calcium carbonate. {Hawk.) it may be observed in dumb-bell like masses which resemble somewhat the same type of calcium oxalate crystal, from which it may be differentiated by the fact that it is soluble in acetic acid with production of CO2, while calcium oxalate remains undissolved. THE URINE 339 (B). Organized Sediments. (i). Mucoid material. Mucus is a constituent of practically every specimen of urine, in the form of the "nubecula." This appears in the form of small threads which branch and interlace in such a way that the entire microscopic field may be practically taken up by this material. In the meshes of the nubecular threads are observed the so-called "mucous corpuscles," which are practically identical with the ordinary leucocyte. Little significance is attached to this form of mucous threads unless a great increase is observed, when it indicates, as does mucin, a vesicle catarrh. The larger more pathologic types of mucus threads will be discussed later. Fig. 104.-Epithelium from different areas of urinary tract, a, leucocyte (for com- parison); b, renal cells; c, superficial pelvic cells; d, deep pelvic cells; e, cells from calices; f, cells from ureter; g, squamous epithelium from bladder; h, neck of bladder cells; i, epi- thelium from prostatic urethra; k, urethral cells; I, scaly epithelium; m, m1, cells from seminal passages; n, compound granule cells; o, fatty renal cell. {Hawk after Ogden.) Normally, the only epithelial cells found in the urine are the irregular flat cells from the bladder and urethra or the large flat epithelia seen in the urine of women and arising from the vagina. The presence of large num- bers of other types of epithelial cells is always pathological and denotes an inflammatory or destructive lesion somewhere along the genito-urinary tract. It is a matter of more or less difficulty absolutely to identify, in all cases, the source of the epithelium found in the urine. According to Heitzmann, the positive recognition is based largely upon the size of the cell, as the shape may vary from pathologic conditions as well as from the portion from which they are derived. As stratified epithelium is found in the pelvis of the kidney, the ureters, bladder, and urethra, it is to be expected that large flat cells, (2). Epithelial Cells. 340 DIAGNOSTIC METHODS. cuboidal cells, or columnar cells will appear depending upon the layer from which the cell is derived. As the simple epithelium exists in the uriniferous tubules, the prostate gland, seminal vesicles, and ejaculatory ducts, the recog- nition of such cells will limit their origin, the size being important in determin- ing the exact point from which they are derived. It is to be remembered, therefore, that the shape of the cell is of far less importance than is its size. The epithelial cells derived from the bladder are usually the large flat irregular cells commonly seen in all normal urine. They have a clear pro- toplasm and usually a small distinct central nucleus and are extremely granular. These flat epithelial cells may be single, in groups, or if the irritation is marked may occur in large sheet-like masses. The large cuboidal cells of the bladder epithelium may be seen in acute cystitis in which they are associated with large numbers of the flat cells previously mentioned. If the conditions become chronic the flat cells may entirely disappear and be replaced by cuboidal and by a few columnar epithelial cells. These latter cells are especially observed in the severe inflammatory processes in the bladder. The large, flat, squamous epithelial cells derived from the vagina are more frequently arranged in stratified groups so that their recognition is usually simple. As these types of cells denote simple desquamation, being pathologic only when present in enormous numbers, an absolute differentiation is of little consequence clinically and, if it be so, the clinical symptoms of the case will usually clear up the decision. The urethral epithelium very closely resembles that above described. The cells are large and irregular, being partly flat, partly cuboidal and partly columnar. The cylindrical types of urethral epithelium may occur in the form of longer, irregular, smaller types, than those of the bladder or vagina. This type constitutes the so-called "tailed" cells, which may be derived from the pelvis of the kidney and were at one time held to be indicative of a pyelitis. Sahli regards a preponderance of such tailed cells over the flatter and more regular types as distinct evidence of trouble in the renal pelves. As these cells may be derived from other portions of the urinary tracts it is unwise to make an absolute diagnosis on such a finding. The small polygonal cells as well as tailed cells may be derived also from the ureter so that our diagnosis would necessarily rest upon findings other than such epithelium. In the writer's opinion, it is a practical impossibility to make a positive diagnosis of a lesion in any specific portion of the genitourinary tract based entirely upon the appearance of the urinary epithelium. The points to be remembered are that we may have any type of epithelium and may have many variations in shape as well as in size. Such variations may be present in any portion of the urinary system, although distinctly renal epithelial cells are more frequently in the form of round or cubical cells somewhat larger than the leucocyte and containing a large vesicular nucleus. These latter renal cells are the only ones which seem to the writer distinctly diagnostic. They are differentiated from the similar cells arising from the ureters and prostate THE URINE. 341 gland by the fact that the latter cells are about twice the size of the pus-cell, being consequently larger than the true renal cell. Degenerative changes are frequently observed in these epithelial cells, even when examined immediately after voiding. The usual type of this de- generation is the presence of fat granules or globules, especially in the small renal cells. If the sediment be treated with Sudan-Ill these granules will appear distinctly red. (3). Pus-cells. A few leucocytes may be observed in practically every specimen of urine, especially in those from women, in which case they may be in large numbers and derived from the vagina. A marked increase, as recognized by numerous, indistinct, small, circular or irregular, granular cells, should be regarded as pathologic.1 To this condition the name pyuria has been given. The simple finding of a pyuria does not necessarily indicate the point from which these cells were derived. Severe inflammatory processes anywhere along the Fig. 105.-Pus corpuscles. 1, Normal; 2, showing ameboid movements; 3, nuclei rendered distinct by acetic acid; 4, as observed in chronic pyelitis; 5, swollen by ammonium carbonate. (Hawk after Ultzmannf genitourinary tract or the rupture of an abscess into the urinary tract will be associated with a pyuria, so that other features must be relied upon in deciding as to the source. As a rule, it may be said that the amount of pus is small in direct affections of the renal cortex, while disease of the urinary passages is associated with a larger number. If an abscess has ruptured into the pelvis of the kidney the number of cells may be enormous. If the pyuria be of renal origin, it will be associated with the presence of the small, round, renal epithelial cells as well as with tubular casts. Frequently leuco- cytes in small numbers are found adherent to the casts or they may even be 1 See Posner, Arch, of Diag., vol. 5, 1912, p. 269. 342 DIAGNOSTIC METHODS. so closely grouped as to give the name pus cast to such formations. If large numbers of pus-cells appear in the course of a chronic nephritis, they indicate either an acute exacerbation of the condition or a complicating process in some other portion of the urinary tract. The sudden appearance of very large numbers of pus-cells is especially indicative of a ruptured abscess. In inflam- matory processes in the pelvis of the kidney the amount of pus may vary within wide limits. In some cases the urine may be perfectly clear when voided, showing the presence of only a few pus-cells, while in others enormous numbers may appear. This paradoxical condition may be accounted for by the possibility of obstruction of the ureter on the affected side and the later forcing out of the large numbers of pus-cells. In pyelitis the urine is usually acid, which may serve as a distinguishing point from cystitis in which the urine is almost always alkaline. In tuberculosis of the renal parenchyma pus-cells appear very early and vary in number from a few to many thousands. This pyuria is usually con- stant, and is frequently associated with hematuria. The pus-cells in tubercu- losis are usually of the mononuclear type instead of the ordinary polymorphonu- clear form. This is not easily determined, as the degenerative processes make it somewhat difficult to distinguish the nuclear form. In such conditions the sediment should be frequently examined for the presence of tubercle bacilli and a portion inoculated into a guinea-pig. This is the only certain method of making a diagnosis of tuberculosis of the kidney. It is, perhaps, needless to add that for absolute differentiation a specimen obtained by ureteral catheterization must be examined. In cystitis the number of pus-cells appearing in the urine will vary ac- cording to the severity of the condition, the more severe the more pus-cells. In this condition the urine is alkaline and may be, when voided, glairy and ropy. In chronic cases of cystitis the pus-cells, although present in the bladder in large numbers, may be so degenerated by the alkalinity of the contents that practically no cells are recognizable. Here we find the appearance of a large amount of mucus, the urine being in some cases distinctly jelly-like. In inflammatory processes of the urethra pus may be present in varying amounts. In the acute conditions the number of cells is much more numerous than in the chronic types. The recognition of the causative factor, in most cases the gonococcus, will be treated of in a later section. As the acute condi- tion becomes subacute or chronic, the urine contains large numbers of the so-called gonorrheal threads which enclose numerous pus-cells. These will be treated in detail later. It is sometimes a matter of clinical importance to distinguish between an anterior and a posterior urethritis. This is best done by the so-called "two-glass" test. If the first portions of the urine be collected in a receiving vessel and the later portions in a second vessel, the urine in the first vessel will be cloudy while that in the second vessel is clear in the case of anterior urethritis; while in posterior urethritis associated with the anterior type the first portion will be cloudy and the second usually so, although at THE URINE. 343 times it may be clear. The reaction of the urine in both vessels will be acid unless a complicating cystitis has arisen, when the urine in the second vessel will usually be alkaline. The appearance of the pus-cells will vary depending upon the reaction of the urine. In acid urine their structure is very well preserved, the addition of acetic acid rendering the nucleus somewhat more distinct. Their usual form is that of the polymorphonuclear neutrophile, their size varying from 7 to 12 microns. If stained the vesicular character of the nucleus of the renal epithelial cell will absolutely differentiate it from the irregular type of pus-cell. In alkaline urine the cells swell up, lose their shape and become opaque. The addition of acetic acid usually clears them in such a way that the nucleus becomes visible, but occasionally does not. If the urine remains long in con- tact with the alkaline material in the bladder it becomes slimy, stringy, and gelatinous, owing to its large content in mucus. Albumin is always present so that it may be difficult to decide whether or not a true albuminuria exists. As the pus-cells may undergo such marked change when in alkaline urine and be converted entirely into a gelatinous mass in which corpuscles cannot be detected, certain tests must be applied for positive recognition of pus in such cases. Vitali's Test. Acidify the urine with acetic acid and filter. Treat the material on the filter with a few drops of tincture of guaiac, when a deep blue color will appear in the presence of pus. If the material is not filterable, as happens when the purulent material is extremely gelatinous, place a portion of this slimy urine in a test-tube and allow a few drops of tincture of guaiac to flow upon the surface. If pus be present a distinct blue line of contact will be observed. Donne's Test A portion of the urinary sediment in a centrifuge tube is treated with a few drops of concentrated solution of sodium hydrate. If pus be present an extremely viscid gelatinous mass will be obtained. If this mixture be heated, it will dissolve, according to Muller, with the formation of /3-nucleinic acid. If the pus-cells be treated under the microscope with a few drops of Lugol's solution, they will take a mahogany-brown color owing to the presence of glycogen. Such a procedure is not a part of the ordinary routine examination of urine. It is, however, sometimes advisable, as it permits of a decision regard- ing the presence or absence of a true albuminuria. If the latter exists, casts and renal epithelial cells will usually be present so that a diagnosis is often possible without a count of the cells. Enumeration of Pus-cells. 344 DIAGNOSTIC METHODS. Technic. A portion of the 24-hour specimen of urine is thoroughly shaken to bring the corpuscles into suspension. This turbid fluid is then drawn up to the upper (11) mark of the leucocytometer, a drop placed upon the glass slide and the cells counted as described under Blood. If more than 30,000 per cmm. are present, it is advisable to dilute five times with 3 per cent, sodium chlorid solution. For each 100,000 leucocytes per cmm. of urine 0.1 per cent, of al- bumin is assumed to be present, according to Wunderlich. (4). Red Blood-cells. The presence of red blood-cells in the urine is known as hematuria. This condition should be sharply differentiated from hemoglobinuria as the clinical significance is entirely distinct. Blood may be found in the urine in a variety of conditions. Thus in the more malignant types of the acute infectious fevers hematuria is frequently observed. Likewise, in scurvy, hemophilia, purpura, leukemia, and Werlhof's disease, the kidney may be so markedly affected that hematuria obtains. In the hematuria of purely renal origin we find both acute and chronic congestions as well as inflammatory processes in the kidney associated with this condition.1 In the more acute types of nephritis, hematuria is so common that the name "hemorrhagic nephritis" is frequently applied. Such cases are especially observed after poisoning with cantharides and phenol deriva- tives. The chronic parenchymatous type of nephritis is, according to Weigert, always hemorrhagic in type, the number of red corpuscles being an indication of the intensity of the process. In malignant growths of the kidney, tuber- culosis, renal calculus, and cystic degeneration of the kidney, hematuria is especially common; while in infection with certain parasites, such as the filaria, echinococcus, and the distoma hematobium, hematuria is relatively frequent although few cases of these conditions are seen. Hematuria may also be observed as a result of lesions or disease of any portion of the urinary tract. Thus stone in the ureter or urethra, tumors, ulcers, and parasites of the bladder, urethritis, prostatitis or injury during cathe- terization may also be associated with the appearance of red cells in the urine. A further type of cases in which hematuria occurs is known as the functional or idiopathic hematuria. In this class of cases no definite lesion has been found to account for the condition. It has been called " Gull's renal epistaxis," "essential renal hematuria," "angioneurotic hematuria," "renal hemophilia" and "renal aneurysm." The lesion, whatever it may be, is usually unilateral and the attacks appear at variable intervals. Some of the cases recover with- out any treatment after one or two profuse hemorrhages, while others require extensive treatment of the nervous system. In the diagnosis of a hematuria it is important to observe the appearance of the urine both with the naked eye and with the microscope. The urine 1 See Randall, Jour. A. M. A., vol. 60, 1913, p. 10. THE URINE. 345 is turbid, and varies from a light, hazy, "smoky" appearance to a bright- red or deep-brown color. The red cells appear in various stages of preserva- tion. In some cases the normal yellow color of the cell will be quite distinct while in cithers the color will be entirely washed out. If the urine be par- ticularly concentrated many crenated forms will also be observed. The blood-cells may exist singly and scattered, or may be grouped in large masses forming distinct clots or adherent to tube casts forming the so- called blood casts. In true renal hematuria the blood is intimately mixed with the urine, the individual corpuscles usually appearing as pale shadows or "ghosts." In hemorrhage from the bladder the urine may show the presence of blood-clots of irregular form and size. If the two-glass test be applied, the second glass will contain the more blood, while in hematuria of renal origin both glasses will show equal amounts. In some cases clots of blood in distinct casts are seen. In the chronic parenchymatous nephritis clots are rarely present, while in malignant disease of the kidney clots are relatively common. It is important in making a diagnosis from the presence of blood that extraneous sources of blood-cells be excluded. If the blood be of renal origin it will be associated with the presence of casts and epithelial cells while no such elements will be present from a hemorrhage lower down in the genito- urinary tract. Albumin will also be present in more or less amount. It has been stated that if the blood be derived from other than renal sources, the clear supernatant fluid in the centrifuge tube will be albumin-free. The writer has convinced himself that this is an error as he has practically always been able to obtain faint albumin reactions in cases of hemorrhage other than renal. (5). Casts. True casts are moulds of the uriniferous tubules. Their mode of forma- tion is not entirely clear. Undoubtedly a colloid substance is thrown into the lumen of the tubule and later solidifies forming a distinct cast of that particular tubule. In this process of hardening the material may enclose cells of different types which are, also, present in the tubule. Whether this coagulable material is derived from the blood as a transudate, whether it be a secretion of the epithelial cells which have become pathologic, or whether it be material arising from degeneration of renal cells is not at present settled, the latter source being the more probable.1 In the urine we find true casts of the renal tubules, as well as pseudo casts which have nothing in common with the true type of these pathologic formations. True Casts. Hyaline Casts. The true hyaline casts are pale, transparent, homogeneous cylinders with rounded ends. Their size may vary from a very small fragment to one several 1 See Erdman, Jour. A. M. A., vol. 59, 1912, p. 1952. 346 DIAGNOSTIC METHODS. mm. in length. In diameter they may be narrow or broad. As a rule, little difference is clinically made between these various types of hyaline casts, but the broader types seem to the writer to be somewhat more significant than do the narrower ones. The characteristics of the true cast are their cylindrical appearance, their sides being parallel and usually straight, although they may at times be observed in typical tortuous forms. They are never tapering at the ends, but may show an irregular outline at one or both ends, but the length and the parallel sides will usually differentiate them. The pure hyaline casts are perfectly homogeneous and free from granules. Such types are, however, not frequently observed as very fine granules may al- most always be detected embedded in the surrounding homogeneous material. Fig. 106.-Hyaline casts. One cast is impregnated with four renal cells. (Hawk.) There may be even inclusions of epithelial, renal, blood-, or pus-cells, so that the gradations between the pure type of hyaline casts and many of the other varieties are outlined with difficulty. It should be stated at this point that a distinction exists between hyaline casts with enclosures of cells to such an extent that the cast is named from the variety of cell present and the type of pseudo cast in which groups of such cells are massed so as to form an apparent cast, but which do not have any definite matrix. The true hyaline cast is soluble in acetic acid and may be stained yellow with Lugol's solution. These hyaline casts are not always easy to find in the sediment. In examining the urine for the presence of casts the light should be shut off as much as possible and a low-power lens used. With the use of the high power the field is much limited and one is not so apt to observe the cast as with the THE URINE. 347 low power. It is, however, always advisable to examine a cast, first seen with the low power, under the high power, so that the decision may be much more definite as to whether the cellular elements are real inclusions or simply material resting upon the true cast. The same is to be said regarding the presence of granules in the hyaline cast. Significance. Regarding the significance of hyaline casts in the urine, it is to be said that they occur in any condition in which the kidney is altered by circulatory, Fig. 107.-Granular casts, X 225. {Tyson.) toxic, or inflammatory disturbances. They are not pathognomonic of any one condition and may be found as a result of simple functional disturbance. A few hyaline casts may be found in practically every urine, providing suffi- cient search is made. Any undue strain, such as running for a car in one who is not used to such exertion, may be sufficient to add quite a number of hyaline casts to the urine. In many thousand urine examinations made in the writer's laboratory, it has been rather the unusual thing not to find an occasional hyaline cast. It would seem, therefore, that no significance what- 348 DIAGNOSTIC METHODS. ever should be attached to the presence of an occasional hyaline cast. When, however, these casts become very numerous they should then be interpreted as meaning a disturbance of the kidney, although the absence of other types would rather speak against a marked pathologic change. In diabetes mellitus, " showers " of casts are especially observed preceding the appearance of coma (Kulz). These are modifications of the true hyaline cast in the sense that fine or coarse granules are found in the matrix of the hyaline cast. Several types of granular casts are observed. The granules may be very fine, very coarse, or may be distinctly composed of degenerated epithelial cells. The fine as well as the coarse granules are undoubtedly derived from the renal epithelium, which has degenerated completely. The coarser the granules the more severe the inflammatory process. These granular casts vary in shape and in size, but are usually shorter than the hyaline type. To these granular casts may be attached various cells so that it is difficult to tell whether the cast is really a true granular or a cellular one. In some cases these cellular inclusions may undergo fatty degeneration giving a much higher re- fractility to the specimen. Not infrequently one ob- serves hyaline casts which are distinctly granular in one portion while the other is perfectly homogeneous. The so-called brown granular casts appear to be almost entirely degenerated epithelial cells, although the coloring matter is probably hemoglobin. The fact that the hyaline matrix cannot be distinctly made out does not argue against this type being truly hyaline in character, although the matrix is completely satu- rated with the pigment. Granular Casts. Waxy Casts. Fig. 108.-Epithelial casts. (Hawk.) This type is very refractile, transparent, and either perfectly colorless or showing a slight shade of yellow. Usually they are very long and broad and may be either straight or curved. The ends show a very distinct fracture while the cast itself may show a tendency to split transversely. Their appearance is, therefore, that of ordinary wax. They may have any type of cellular element attached and may show marked fatty degeneration. Some of these casts show the amyloid reaction while many of them do not. Waxy casts were at one time believed to be pathognomonic of amyloid de- generation of the kidney. It is true that they do appear earlier in this type of kidney lesion, but it is to be remembered that they occur in all varieties of chronic kidney disease. They are usually of bad prognostic omen as they in- dicate a very advanced process. PLATE X. Waxy Casts Treated with Iodine. (Tyson.) THE URINE. 349 Fibrinous Casts. These are very highly refractile, transparent, and always of a yellowish or brown color. They may be granular and have various cellular inclusions. Their shapes vary as do those of the hyaline types and they show a tendency to become fractured, the fracture usually being ragged, while in the waxy cast it is sharp-cut. Fibrinous casts usually appear in the acute renal conditions and disappear when this condition clears up. They do not have, therefore, the grave signifi- cance of the waxy cast and should be sharply differentiated. These casts are not composed of fibrin as their name would indicate, but are so called on account of their brownish color, which resembles fibrin. Fig. 109.-Fatty casts. {Hawk after Peyer.) Epithelial Casts These casts are true hyaline casts which include so many renal epithelial cells that the hyaline matrix may be lost. For the name epithelial cast to be accurate it is not necessary that more than a few cells be present. The cells may be well preserved or show marked fatty or granular degeneration. The nuclei of these cells are round and vesicular so that they may be easily recognized. Distinct gradations exist between the true epithelial cast and the coarsely granular and fatty cast. This type of cast is indicative of a severe destructive lesion of the kidney epithelium. These casts are masses of epithelial cells which have so markedly degener- ated that little is recognized beyond the original outline of the cell and the Fatty Casts. 350 DIAGNOSTIC METHODS. numerous fatty globules contained therein. They may be yellowish or black in color, the globules being soluble in ether and staining black with osmic acid or red with Sudan III. These are casts including large numbers of blood-cells. The casts are formed within the tubules of the kidney, the cells occasionally being very pale. These casts indicate a serious advanced lesion of the renal parenchyma. Blood Casts. These like the other types of casts arg true hyaline casts with enclosures of pus-cells. They are formed within the tubules of the kidney and usually Pus Casts. Fig. no-Blood, Pus, Hyaline and Epithelial Casts. {Greene.) a Blood casts; b, pus cast; c, hyaline cast impregnated with renal cells; d, epithelial casts. indicate an acute pyelonephritis. For the differentiation of these casts from epithelial casts it is advisable to add acetic acid to the sediment when the typical polymorphous character of the nucleus will distinguish the pus-cell from the epithelial cell with its vesicular nucleus. Moreover, the pus-cell is much more spherical than is the epithelial cell. Cylindroids. It is not infrequent to find in the urine formations which resemble the true hyaline casts to a marked degree. They, however, differ in the fact that at one or at both ends they taper off into a point which may be prolonged into a distinct thread. If, however, these ends are broken off, as may occur in PLATE XI. Mucous Threads in Urine. (Unstained Specimen ) THE URINE 351 the centrifugation, it is a practical impossibility to distinguish them from a hyaline cast. They are both found in the urine under the same conditions and their significance is practically the same. From the chemical standpoint they appear similar to the hyaline casts, their origin, therefore, being presum- ably in the renal parenchyma. If these bodies are true mucin and are insolu- ble in acetic acid, their origin is more probably in the bladder. A second type of cylindroid appears in the urine in the form of long tapering transparent shreds. They very much resemble ribbon which varies in diameter and may show under high power a distinctly fibrillar structure. These threads largely compose the nubecula. They are much longer than the hyaline cast Fig. hi.-Cylindroids. {Hawk after Peyer.) and considerably narrower so that confusion should not arise. In cases of gonorrhea one finds mucous shreds which may vary from a few mm. to i cm. in length and yellowish or pure white in color. In the meshes of these shreds one finds embedded large numbers of pus and epithelial cells. These should be sharply differentiated from the true cast by their larger size and typical mucoid character. Frequently they may be observed by the naked eye in large numbers. Pseudocasts. Not infrequently do we find in urine crystalline material arranged in masses much resembling casts. The most important of these are uric acid and the urates. It is true that any cast in a concentrated urine may become covered with urates so that the true nature of the cast becomes indefinite. If the slide be warmed these pseudourate casts will disappear, while the true casts will remain. Masses of bacteria, pus-cells, epithelial cells and blood- 352 DIAGNOSTIC METHODS. cells may so group themselves as to closely resemble true casts. As a rule such masses will show irregular outlines and no evidence of a distinct matrix. Moreover, the use of an old slide upon which there may be many scratches should be avoided as the writer has seen several instances in which supposed casts were found to be due to such scratches. Cylindruria. This is the name given to the appearance of casts in the urine. As a rule, it should be said that the presence of a few hyaline casts is not of particular moment unless associated with other evidences of marked renal disturbances. While albuminuria and cylindruria usually go hand in hand, yet we do find cases in which one exists without the other. It is undoubtedly true that casts indicate a disturbance of the renal epithe- lium. This, however, need not be anything more than disturbed nutritional or circulatory conditions. However, when the true hyaline casts are present in large numbers and when many other types of casts also exist, then a distinct patho- logic lesion of the kidney must be assumed.1 As a rule, the granular types of cast are observed in the chronic processes, while the cellular forms are more usually present in the acute conditions. This rule, however, is not invariable, so that it may generally be stated that no type of cast is pathognomonic of any single condition. In this connection we should remember that recent work, especially that of Cabot, has shown that it is unwise to base a diagnosis of a kidney lesion upon the finding even of both albuminuria and cylindruria. So much discrep- ancy was shown to exist between the urinary and autopsy findings that one must remain in doubt as to whether it is possible to make a definite diagnosis unless clinical symptoms other than urinary are made the basis of a diagnosis. On the other hand, some of the most typical cases of nephritis, as shown post- mortem, gave absolutely no indication in the urine that such condition existed. We are, therefore, face to face with the proposition that urinary examination must be in any case simply one of the diagnostic links. This fact is of special importance in life insurance examinations, as most companies absolutely refuse insurance to one who has ever shown albumin or casts in the urine. This would seem to the writer not only very short sighted, but base dupon an absolutely erroneous idea of the importance of albumin and casts in the urine of one who showed absolutely no clinical signs of renal involvement. Remem- bering that albumin and casts may not appear, even though the kidney be seriously affected, it would seem just as plausible to refuse life insurance because these substances were not present. In this connection the writer would say that only when the urine is considered as a whole may definite conclusions be made regarding any type of renal disease. The clinician, who is thoroughly familiar with the course of the case, is the only one capable of interpreting the findings of the laboratory, so that it should be an unvarying rule for a laboratory worker to avoid diagnostic remarks unless he is thoroughly en rapport with the 1 See Barringer and Warren, Arch, of Int. Med., vol. 9, 1912, p. 657. THE URINE. 353 patient. The writer does not wish to be interpreted as stating that a diagnosis of renal disease may never be made from an examination of the urine, but he wishes to impress upon his readers that both albuminuria and cylindruria may occur without direct kidney disease or may not appear when such is present. (6). Spermatozoa. Spermatozoa are frequently observed in the urine of healthy adults, especially after intercourse or nocturnal emissions. In females they may also be observed as an evidence of intercourse, which fact is of some importance in cases of suspected rape. Pathologically, they may be found in cases of marked constipation, when the pressure of the impacted feces upon the seminal vesicles may induce an emission. In occasional cases of cystitis, associated with stricture, these bodies may be observed as reported by Simon. In cases of epilepsy and hysteroepilepsy, as well as in spinal disease following vertebral fractures and dislocations, spermatozoa are not infrequent. Masturbation and venereal excess frequently lead to almost constant spermatorrhea. Their occurrence in cases of prostatitis will be discussed in the section on Semen. (7). Tissue Fragments. It is not infrequent to find shreds of tissue in the urine, which may throw some light upon a pathologic condition. In cases of carcinoma of the bladder more rarely of the kidney, true malignant tissue may be obtained, which may permit of a tentative diagnosis, although the material is usually too necrotic to make an absolute diagnosis possible, the principal finding being that of elastic tissue. (8). Bacteria. It should be stated in the beginning of this discussion that an examination of the urine for bacteria should be made only upon specimens obtained with the greatest possible precaution and preserved in absolutely sterile vessels. Soon after the urine is voided, especially if it remains in contact with the air, large numbers of saprophytic organisms may be found which, of course, did not exist in the original urine. In obtaining a specimen from the male it is not always necessary to catheterize the patient. If the surface of the glans and the orifice of the meatus be carefully washed with bichlorid solution followed by sterile water and the first portion of the urine voided be thrown away, the last portion may be collected in a sterile vessel and later put in work. With female patients, however, it is absolutely essential that catheterization be performed. The external genitalia and especially the orifice of the urethra are well washed with green soap and water. The opening of the urethra is then dried with sterilized cotton pads which are soaked in boracic acid. A sterilized glass catheter, whose external end is covered with a rubber tube about four inches long and large enough to fit loosely over the catheter is then inserted, care being taken that it touches only the orifice of the urethra. The urine is allowed to flow freely for a short time when the last portion is 354 DIAGNOSTIC METHODS. collected in a sterile vessel, the rubber tube being previously removed. (Kelly). Cultures are then made from the urine and the remainder centri- fuged in a sterile closed tube in order to throw down any bacteria which may be present. It is sometimes advisable, in order to diminish the specific gravity of the specimen, to add an equal volume of 95 per cent, alcohol and centrifuge the mixture. Practically all of the bacteria present will then be found in the sediment. The supernatant fluid is then removed by quickly inverting the tube and allowing the fluid to run out. The sediment, by this manipulation, will usually remain in the smaller portion of the tube. Smears are then made upon a glass slide and dried first in the air and then over the flame. It is not always the simplest matter to prepare smears which will remain after treatment with the staining solution, as the urea and salts of the sediment may be removed by the washing and carry with them the bacteria. If pus-cells are present, satisfactory smears are usually obtained; but if such conditions do not exist it is advisable to add a solution of egg albumin to the sediment before drying over the flame. The methods of staining the sediment for the various bacteria will depend entirely upon the organism supposed to be present. As a rule, a preliminary examination is very satisfactorily made by treatment with Loffler's methylene- blue solution, which stains practically all organisms. If the tubercle bacillus is suspected it may be detected in exactly the same way as outlined under Sputum by staining with carbol-fuchsin solution. Should one suspect the presence of the gonococcus1 this may be stained, as described in the next section, by Gram's method. Outside of these two types of organisms, it is almost impossible to differentiate the bacteria of the urine by staining methods. The peculiarities of the various bacteria upon culture media may be learned from works on bacteriology. If the urine be collected with the precautions mentioned above, any organ- isms found must be attributed to their presence in the urine as voided. In this connection we must remember that the presence of the tubercle bacillus does not necessarily indicate tuberculosis along the genitourinary tract.2 Tubercle bacilli are found in the urine in cases of miliary tuberculosis and have been re- ported in pulmonary tuberculosis, although it is more frequent to find them as evidences of local tubercular conditions. Of course, if large numbers of pus- and blood-cells be present along with tubercle bacilli, the diagnosis is usually certain. A word of caution is, however, necessary at this point. The smegma bacillus grows in abundance on the external genital organs and its morphological and staining characteristics may closely resemble those of the tubercle bacillus. If the proper precautions be observed, as they should be, no differentiation is necessary, bacilli showing the true morphological and staining character of the tubercle bacillus can be only this organism, as extraneous bacteria have 1 This must not be confused with the micrococcus catarrhalis, which not infrequently is the etiologic factor in genitourinary inflammations. 2 See Kielheuthues, Folia urol., Bd. 7, 1912, S. 191. PLATE XII. Cystitis due to Colon Bacillus. (Methylene Blue Stain.) THE URINE. 355 been avoided. In the clinical laboratory, however, one may never be sure whether the proper precautions have been taken, so that absolute methods of differentiation (see sputum) should be part of the technic unless the worker absolutely knows that contamination was avoided. A second point regarding the tubercle bacillus is that it may not be found by microscopic examination even after repeated attempts. Under such conditions the wisest course to pur- sue is the inoculation of a guinea-pig with the washed urinary sediment. The obtaining of the urine must in this case be absolutely accurately done by observ- ing every precaution to prevent contamination. The reason for this is not because contaminating organisms will cause lesions similar to those of the tubercle bacillus, but because such secondary invaders may so infect the animal that death results from causes other than those for which we are looking. The washed sediment is injected intraperitoneally and the animal kept under ob- servation for three weeks to one month unless death results previously. At the end of this time the animal is killed and a postmortem examination made for evidences of tuberculosis. The retroperitoneal glands, spleen, and liver are the especial organs to show such lesions. These organs should be sectioned, portions run through the regular pathologic routine, and sections examined microscopically. A finding of tuberculosis in this way is unequivocal and is the quickest way in the long run of making a positive diagnosis, although a single examination may show the presence of tubercle bacilli in the urine, but rarely such is the case. Having found the tubercle bacilli in the urine, we are confronted with the question of the part affected. As a primary tuberculosis of the bladder is rare we may usually assume the seat of the difficulty to be the kidney, although if evidences of cystitis be present a combination may exist. As the symptoms of genitourinary tuberculosis are frequently vesicle in origin, a kidney lesion may not be suspected, but should be assumed until the contrary is proven. Thanks to the introduction of methods of cystoscopic examination and espe- cially ureteral catheterization, we are in a better position to make a positive diagnosis of renal tuberculosis and exclude that of bladder origin. An inter- esting point regarding tuberculous cystitis is that the urine, although frequently containing large amounts of pus, is practically always acid in reaction. More- over, this pus is frequently sterile in tubercular cystitis. It is not the province of the writer, at the present time, to outline methods of differential diagnosis of various conditions; he will refer, therefore, to works on genitourinary diseases for the various types of cystitis and their clinical differentiation. Suffice it to say at this point that cystitis may be an ascending or a descending one and should always be correlated with the associated condition. It is usually a simple matter to determine the presence of the gonococcus in the urethral discharge, but it is far from easy to demonstrate this organism in the case of gonorrheal cystitis. Under such conditions the symptoms and the association with an existing gonorrhea would furnish the decisive clue. 356 DIAGNOSTIC METHODS. By this condition is meant the presence in the freshly voided urine of so many organisms that the urine is distinctly cloudy. These organises are usu- ally those associated with a mild cystitis or may be those of an existing general infection. In cases of persistent bacilluria, we may find a true renal origin, which is largely associated with the presence of the typhoid and colon bacillus. As has been well established, the typhoid organism may be excreted for months after the patient is convalescenent so that it becomes necessary to use strict measures to disinfect all urine of typhoid patients. The colon bacillus is at present assuming so much importance in clinical work that cases are being almost daily recognized which may be directly traceable to the colon bacillus and not to the typhoid as usually assumed. It is necessary, therefore, for the laboratory worker especially, and, where possible, for the general practi- tioner to be able to recognize each of these organisms when present either in the feces, urine, milk, or water-supply (see Feces). A second general class of cases associated with bacilluria are those of urethritis and prostatitis. Usually there is a secondary cystitis arising from the same organism or, at least, the resistance of the bladder has been so far reduced that these organisms which find their way into the bladder develop profusely therein. Regarding the presence of the various bacteria which may be found in the urine after it has been exposed to the air, the writer cannot take up space for their de- scription. Their number is extensive and their type protean. Reference must be made to general works. Bacilluria. (9). Parasites. Various types of parasites are observed in the urine. Thus the trichomonas vaginalis has been found by Kunstler. Miura, and Dock. Amebae have been found by Balz, Jurgens, Wijchoff, and by Musgrave and Clegg. Various portions of hydatid cysts are frequently observed, among which we find the echinococcus booklets and fragments of membrane. Nematode worms, especially the filaria sanguinis hominis, are present in cases of chyluria, while the anguillula aceti or "vinegar-eel" has been reported, especially by Stiles, while Billings and Miller report its presence as a possible contamina- tion from the bottle in which the urine was collected. Eggs of the schistosomum haematobium are not infrequently observed together with large numbers of red cells in cases of bilharziasis. This worm as well as its ova will be discussed in the section on Blood. Stuertz has reported the findings of the egg of eustrongylus gigas in the urine in a case of chyluria. Fig. i i 2.-Scolex and booklets of tasnia echinococcus in urine. PLATE XIII. Staphylococcus Cystitis. (Leishman Stain.) THE URINE 357 V. Calculi. Concretions of a more or less hard and dense character are prone to form in the urinary passages. These bodies are termed, according to their size and location, sand, gravel, stone, and calculi. Theoretically, these formations may consist of accretions of any of the various crystalline or amorphous sedi- ments previously mentioned, the type of stone depending upon the reaction of the urine to a large extent. These calculi are formed by the deposition of the crystalline material around a definite nucleus, which usually consists of organic material, suchfas fibrin, blood, desquamated epithelial cells, mucus, or even a crystal of uric acid or calcium oxalate. It is very difficult to decide as to the reason for the Fig. i 13.-Ova and miracidium of schistosomum hematobium, X 300: A, Ovum as seen in urine; B, the same after addition of water; C, miracidium. (Tyson after Railliet). deposition of this material in the form of a renal stone. The growth of the calculus takes place by accretion, the deposition of successive layers of material occurring around the original nucleus. The material of which the stone con- sists will usually be of one kind, so that we speak of uric acid or phosphate calculi, for instance, while, occasionally, mixed calculi may be formed by the deposition of two or more chemical combinations. The classification of urinary concretions is based on the chemical con- stituents of which they are composed. Before examining a calculus chemi- cally a thorough optical examination should be made, as this may give a definite clue as to its composition. After this preliminary examination, the calculus is ground to a fine powder and examined according to the following table of Heller: 358 DIAGNOSTIC METHODS. DOES NOT BURN. DOES BURN. The powder when treated with hydrochloric acid. With flame. Without flame. Does not effervesce. The gently heated powder with hydrochloric acid. The powder when moistened with a little potassium hydrate E f Abundant am- No ammonia, monia. T h e or, at least, only powder dis-traces of it. r solves in acetic Powder dis- v or hydrochloric solves in acetic e acid. This so- or hydrochloric s lution gives a acid. This so- c crystalline pre- lution gives an e cipitate with amorphous pre- s ammonia. cipitate with ammonia. E f f e r V e s c e s Flame yellow and continuous. Odor of burned feathers. In- soluble in al- cohol and ether. Soluble in po- tassium hydrate on heating. P r e c i p i t ated from alkaline solution by ace- tic acid with generation o f hydrogen sul- phid. Flame pale yellow, contin- uous. Odor of resin or shellac on burning. Powder soluble in alcohol and ether. Flame pale blue, burns a short time. Peculiar sharp odor. The pow- der dissolves in ammonia, from which six- sided plates separate on the spontaneous evaporation of the ammonia. Does not give the m u r e x i d test. The pow- der dissolves in nitric acid with- o u t efferves- cence. The dried yellow residue becomes orange on ad- dition of alkali, red if heat be applied. The powder gives the murexid test. The powder when treated with potassium hydrate gives. A strong No notice- ammonia able reaction reaction. for ammonia. T r i p le phos- phate mixed with an un- known amount of earthy phos- phates. Magnes ium and calcium phosphates. Calcium oxalate. Calcium carbonate. Fibrin. Urostealith. Cystin. Xanthin. Ammoniu m urate. Uric acid. HELLER'S TABLE FOR EXAMINATION OF URINARY CALCULI. On Heating the Powder on Platinum Foil it THE URINE. 359 Uric Acid Calculi. These are, perhaps, the most common renal stones. They are not always composed of pure uric acid, but are made up of a mixture of this substance with the urates. They are always colored, usually yellowish or brownish, but may at times appear distinctly red. They are fairly hard and usually show a rough irregular nodular surface, although at times this may be smooth. They fracture very easily and show, on cross section, a distinctly laminated structure, the layers frequently being of different colors, in some cases even being composed of deposits other than uric acid. If heated on a plati- num foil they are combustible, burning without a flame. They give the murexid test and do not liberate appreciable amounts of ammonia on treatment with sodium hydrate. The ammonium urate calculi occur rarely in the adult. They are small, yellow and very soft, being distinctly clay-like and easily powdered when dry. These stones give the murexid test and also give a strong reaction for ammonia on treatment with sodium hydrate. Ammonium Urate Stones. Calcium Oxalate Stones. Next to the uric acid calculus the oxalate stone is most frequently met.1 The smaller types of these calculi are practically colorless and have a smooth surface, while the larger ones are grayish, brownish or even black in color and have a rough nodular surface with sharp projecting angles. These stones frequently cause severe hemorrhage and much irritation when passing through the ureter and urethra. From their appearance they have been called the mulberry calculi. They are, perhaps, the hardest of the urinary stones. These calculi are insoluble in acetic acid, but soluble in hydrochloric acid without effervescence unless the powder is previously heated. It is rare to find these stones perfectly pure, admixtures with various other sediments leading to distinct concentric arrangement as shown on fracture. Stones composed of pure alkaline phosphates or triple phosphates are exceedingly rare. Usually the phosphatic calculi contain admixtures of ammonium urate, calcium carbonate, and calcium oxalate. The color of such stones may range from a white through yellow to some with distinct reddish tones. They are frequently of very large size, especially when formed in the bladder, are of a chalky consistency, and show a rough surface. These calculi are soluble in hydrochloric or acetic acid, such solutions giving reactions both for phosphoric acid and the alkaline earths. Phosphatic Calculi. Calcium Carbonate Calculi. Such stones are exceedingly rare. They are small in size, are distinctly 1 See Rowlands, Biochem. Jour., vol. 3, 1908, p. 346; Kahn and Rosenbloom, Jour. A. M. A., vol 59. 1912, p. 2252; also, Kahn, Arch, of Int. Med., vol. n, 1913 p. 92. 360 DIAGNOSTIC METHODS. chalk-like in consistency and color, and have a smooth surface. If treated with acid, carbon dioxid is evolved. Cystin Calculi. These stones are white or pale yellow in color, have either a smooth or irregular surface, and are soft and waxy in consistency. They vary in size occasionally being found as large as a hen's egg, although those of true renal origin are about the size of a pea. The formation of such calculi and their passage through the ureter and urethra constitute practically all of the untoward symptoms shown by subjects affected with cystinuria. Such stones burn readily if heated on a platinum foil, giving off a peculiar sharp odor. The powder is soluble in ammonia from which the characteristic hexagonal plates separate on allowing the ammonia to evaporate. Xanthin Calculi. These stones occur especially in children, although even here they are very rare. They are usually light brown in color, moderately hard, and vary in size from that of a pea to a tennis-ball. On cross section they appear amorphous and if rubbed take a polish much resembling that of wax. The powder shows the typical reaction for xanthin previously outlined. Urostealith Calculi. These masses consist of fat, calcium and magnesium soaps, and choles- terin. They are usually soft and may be somewhat irregular in shape. This material burns with a pale yellow flame giving an odor of resin. The dry powder is soluble in alcohol and ether, from which rhombic notched plates of cholesterin separate on evaporation. VI. Functional Diagnosis It is usually of great importance, especially in cases in which surgical intervention is contemplated, to know just exactly what the functional capa- bilities of the kidney are. If one kidney is to be removed, the question arises as to whether the remaining kidney can sufficiently accomodate itself to the increased work which must be put upon it. As such a variation has been occasionally found between the results of the chemical and microscopical examination of the urine on the one hand and the pathologic condition shown in the kidneys postmortem on the other, an attempt has been made to find delicate tests by which a true renal lesion might be indubitably determined and thus permit of an absolute diagnosis even though the urinary findings were or were not conclusive. It is granted that a more or less severe lesion of the kidney may exist and yet its functional capacity be almost normal. If this functional capacity can be determined by tests which are more or less simple, it is evident that such methods should form part of the daily routine of the practitioner. Before discussing some of these tests, which have been advocated for estimating the functional activity of the kidney, the writer would say that few THE URINE. 361 of them have proven entirely satisfactory. If we are to regard nephritis as a constitutional condition with secondary renal manifestations, it is self-evident that such tests can show only the manner in which a normal or pathologic kidney reacts toward certain substances and can have little reference to the primary etiologic factors. Cryoscopy. The method of determining the freezing-point of a solution is one of the most delicate of those of physical chemistry. As it is based upon the principle that substances in solution lower the freezing-point of the solvent in direct proportion to the molecular or ionic concentration of the solution, this method serves as a ready means of determining the molecular weight of a substance as well as the molecular concentration of a solution. For a successful outcome of a cryoscopic determination, the most assiduous attention to detail must be paid, so that this method certainly can find no place in the hands of the general practitioner or even in those of the laboratory worker, who has not been espe- cially trained along these lines. Such being the case, it is absurd to expect that slight variations in the freezing-point (△) of such a complex mixture as the urine can yield any valuable information, especially when one remembers that fluctuations wider than those shown under pathological conditions may be noted as a result of not observing such slight details as that of constantly agitating the urine during the cooling and freezing. The normal freezing-point of the urine varies between -0.9 and -20 C., an increase being known as hypersthenuria and a decrease as hyposthenuria. As will be seen in the discussion of this subject in the section on Blood, the results obtained by this method have been far from satisfactory, as nothing of distinct diagnostic value has as yet been derived from comparative studies by various workers. It would seem to the writer, therefore that for the present this method would better be left to the research worker than to be adopted by the general or special student, who should make use of methods which will yield results of more immediate value to him. For these reasons the writer must refer to other works for a detailed description of the method. Electric Conductivity. The remarks made later in the section on Blood regarding electric conduc- tivity are especially applicable to the urine. This test is altogether too delicate to be applied to such complex fluids as the urine with the hope that slight variations in the conductivity will show anything of importance. As the electric current is conducted only by dissociable compounds, this method can show absolutely nothing regarding the excretion of the nondissociable organic substances. Such being the case the writer can see no reason for resort- ing to such delicate procedures as the determination of the conductivity of the urine when no attempt is made, as may be observed in many of the experiments reported, to control the intake of the inorganic substances which would affect the conductivity of the urine. A little more attention to ordinary methods 362 DIAGNOSTIC METHODS. of chemical examination, with especial regard to ascertaining the intake and output of the patient, would, in the writer's opinion, yield much more valuable information than could be obtained by the rather uncertain urinary manipu- lations with the method of Kohlrausch. Chlorid Excretion. By the determination of the chlorid excretion in a given case under the influence of a specific intake of sodium chlorid it was hoped that some informa- tion might be derived as to the "glomerular sufficiency" of the kidney. This test, known as alimentary chloruria, advanced by Claude and Mantb1, has like the above yielded little of value. The exceptions to the general rule of pathologic retention are too numerous to permit of absolute differentiation. While it has been shown by Widal that a retention of sodium chlorid does occur in nephritis and that a restriction of the salt in the diet frequently does alleviate the edema and albuminuria, yet many factors not well understood fre- quently give rise to opposite results (see Chlorids in Urine). The general consen- sus of opinion regarding this test may be summed up by the statement that the chlorid excretion in cases of health and disease is of no distinct value in diag- nosing the functional activity of the kidney. Methylene Blue Test. It was hoped by the introduction of such tests that the capability of the kidney to excrete substances present in the blood could be determined by the ease with which artificially introduced substances were eliminated. Achard and Castaigne administer by the mouth o.r gram of methylene blue in a capsule or, preferably, 0.05 gram subcutaneously. This substance is excreted by the kidneys in the form of a colorless chromogen in 15 to 30 minutes after injection. This chromogen may be converted into the colored pigment by heating with acetic acid. In about five minutes after the excretion of the chromogen, the urine becomes greenish-blue from the excretion of the pigment itself. Under normal conditions the excretion reaches a maximum in from three to four hours and may last two to three days. If the pigment is not excreted within one hour after subcutaneous injection, pathologic conditions are supposed to exist. In some cases of chronic interstitial nephritis it was found that such injections were not followed by an excretion of the pigment until one to three hours had elapsed. This would be a valuable diagnostic method had it not also been shown conclusively that many cases of advanced interstitial nephritis show the same excretion of this pigment as normal individuals. Such being the case the test is of no more use than many clinical and routine urine methods. Phloridzin Test. This test is based on the assumption that phloridzin normally gives rise to a glycosuria through distinct alternations in the renal cells. In other words, a so-called "renal diabetes" is set up. The technic is as follows: 1 c.c. of a 1 Arch. gen. de m£d., Tm. 8, 1902, p. 129. THE URINE. 363 1:200 aqueous solution of phloridzin is injected subcutaneously. The urine is tested at intervals of 15 minutes for the appearance of sugar. Normally, sugar may be detected in one-half to one hour and may be present for as long as five hours. The quantity eliminated may vary from 0.5 to 3 grams of glucose. In nephritis the sugar is usually absent or below 0.5 gram. This test does not distinguish the various types of nephritis, yet it does usually indicate that renal activity is disturbed. It is, however, being rapidly displaced by the following test. Phenolsulphonephthalein Test. Rowntree and Geraghty1 have introduced a test for the functional activity of the kidney which is by far the most accurate at our disposal.2 The solution used is prepared as follows: 0.6 gram of phenolsulphonephthalein and 0.84 c.c. of 2/N NaOH solution (8 per cent.) are added to sufficient 0.75 per cent. NaCl solution to make 100 c.c. Each c.c. of this solution contains, therefore, 6 mg. of the dye. As the mono-sodium salt, formed in this solution, is slightly irritant locally when injected, 2 or 3 drops more of the 2/N NaOH are added, when the solution changes to a beautiful Bordeaux red color and becomes non-irritant. The technic is as follows: 20 to 30 minutes before applying the test, the patient is given 300 to 400 c.c. of water in order to insure a free urinary secretion, otherwise delayed time of appearance may be due to lack of secretion. Under aseptic precautions a catheter is introduced into the bladder and the bladder completely emptied. Noting the time, 1 c.c. of the above solution is administered subcutaneously in the upper arm by means of an accurately graduated syringe. The urine is allowed to drain through the catheter into a test-tube in which has been placed a drop of 25 per cent. NaOH solution and the time of the first pink- ish tinge is noted. In patients without urinary obstruction, the catheter is withdrawn at the time of appearance of the drug in the urine and the patient is instructed to void into a receptacle at the end of one hour and into a second receiver at the end of the second hour. A rough estimate of the time of ap- pearance may be made by having the patient void urine without the use of the catheter at frequent intervals. In prostate cases it is wise to have the catheter in place until the end of the observation. The catheter is corked at the time of appearance of the drug in the urine and the cork is removed at the end of the first and second hours, the bladder being drained each time. Each sample of urine is measured and the specific gravity taken. Suffi- cient 25 per cent. NaOH is added to make the urine decidedly alkaline in order to elicit the maximum color (a brilliant purple-red). This solution is now placed in a liter volumetric flask and distilled water added to the 1L. mark. Thoroughly mix the solution and filter a small portion for comparison with the standard solution. This standard consists of 3 mg. of phenolsulphonephthalein (or JJour. of Pharm. and Exp. Therap., vol. i, 1910, p. 579; Arch, of Int. Med., vol. 9, 1912, p. 284. 2See Conzen, Deutsch. Arch. f. klin. Med., Bd. 108, 1912, S. 353; Smith, Amer. Jour. Dis. of Child., vol. 5, 1913, p. 25; Pepper and Austin, Amer. Jour, of Med. Sci., vol. 145, 1913, p. 254; also, Rowntree, Fitz and Geraghty, Arch, of Int. Med., vol. 11, 1913, p. 121. 364 DIAGNOSTIC METHODS. 1 c.c. of the solution used for injection) diluted to 1 liter and made alkaline with 1 or 2 drops of 25 per cent. NaOH solution. This solution will retain its inten- sity of color for weeks. The comparison of color is made in the Duboscq or Sargent colorimeter. The known (standard) solution is adjusted to the 10 mm. mark and the intensity of the unknown solution is made to correspond by means of the usual manipulations. One may readily calculate the amount of drug eliminated as follows: If, for instance, the reading for the unknown solution is 20, it is evident that only 50 per cent, as much dye is present as is contained in the standard solution. That is, the excretion equals 50 per cent, of the 3 mg. of the standard or 25 per cent, of the 6 mg. injected. The readings may be then made on a second and a third voiding until the drug is completely eliminated. In normal cases it has been found that the time of appearance varied from 5 to 11 minutes and that 40 to 60 per cent, of the drug was excreted in the first hour and 20 to 25 per cent, in the second hour. The excretion of the drug does not run parallel to the excretion of water. The smaller the amount of urine in normal cases the greater the concentration of the drug. It is immaterial, as far as the excretion of the drug is concerned, whether the urinary output is 50,200,400 or more c.c. In pathological cases it has been demonstrated that the permeability of the kidney for this drug is decreased in both chronic parenchymatous and chronic interstitial nephritis, the decrease being most marked in the latter and varying with the intensity of the disease. In cases with obstruction in the lower urinary tract, this test may show renal involvement to such an extent that operation should be deferred, especially if the time of appearance is delayed beyond 25 minutes and the output of the drug is below 20 per cent, for the first hour. This point is of special value to the surgeon as it shows the danger of using surgical intervention until the kidney may become less insufficient. It would seem to be of value, also, in the study of the renal condition preceding uremia. To the obstetrician it may show the possible approach of eclampsia, as has been demonstrated in several cases. In unilateral and bilateral kidney diseases the absolute amount of work done by each kidney as well as the relative proportion can be determined by resort to ureteral catheterization followed by this test. The writer has no hesitancy in advising a careful study of this test in all cases with suspected or actual renal involvement. BIBLIOGRAPHY. i. Benedict. Influence of Inanition on Metabolism. Washington, 1907. 2. Blarez. L'Urine au point de vue clinique et medical. Paris, 1906. 3. Blumenthal. Pathologic des Harnes am Krankenbett. Berlin, 1903. 4. Chittenden. Physiological Economy in Nutrition. New York, 1905. The Nutrition of Man. New York, 1907. 5. Cohnheim. Chemie der Eiweisskorper. Braunschweig, 1906. 6. Croftan. Clinical Urinology. Chicago, 1907. 7. Daiber. Mikroskopie der Harn-Sedimente. Wiesbaden, 1906. 8. Fischer. Untersuchungen ueber Aminosauren. Berlin, 1906. THE URINE, 365 p. Gerard. Traite des Urines. Paris, 1906. 10. Hammarsten. Lehrbuch der physiologische Chemie. Wiesbaden, 1907. 11. Heitzmann. Urinary Analysis and .Diagnosis. New York, 1906. 12. Hoppe-Seyler-Thierfelder. Handbuch der physiologisch- und path- ologisch-chemischen Analyse, Berlin, 1909. 13. Krehl. Ueber die Storung chemischer Korrelationen im Organismus. Leipzig, 1906. Pathologische Physiologic. Leipzig, 1912. 14. Mann. The Physiology and Pathology of the Urine. London, 1904. 15. Naunyn. Der Diabetes Mellitus. Wien, 1907. 16. Neubauer und Huppert. Analyse des Harns. Wiesbaden, 1910. 17. Neuberg. Der Harn. Berlin, 1911. 18. von Noorden. Die Zuckerkrankheit. Berlin, 1912. 19. Ogden. Clinical Examination of the Urine. Philadelphia, 1903. 20. Pavy. Carbohydrate Metabolism and Diabetes. London, 1906. 21. Purdy. Practical Urinalysis and Urinary Diagnosis. Philadelphia, 1900. 22. Rieder and Delephine. Urinary Sediments. London, 1899. 23. Saxe. Examination of the Urine. Philadelphia, 1906. 24. Scott. Clinical Examination of the Urine. Philadelphia, 1900. 25. Spaeth. Untersuchung des Harns. Leipzig, 1903. 26. Sutton. Volumetric Analysis. Philadelphia, 1904. 27. Tyson. Bright's Disease and Diabetes. Philadelphia, 1904. 28. Wells. Chemical Pathology. Philadelphia, 1907. CHAPTER VIL SECRETIONS OF THE GENITAL ORGANS. I. Male Secretions. General Considerations. The normal secretion of the male generative organs is known as semen and is a mixture of the secretions of the prostate gland, the glands of Cowper, the testicles, and the seminal vesicles. It is a practical impossibility from the clinical standpoint to separate the different elements of the semen, so that this must be discussed as a whole. The semen is a white or slightly yellowish, somewhat thick and viscid liquid with a peculiar odor, somewhat resembling fresh glue, and showing a neutral or faintly alkaline reaction, a nonhomogeneous milky appearance, and a specific gravity greater than that of water. It is composed of semisolid material in the form of white masses floating in a limpid liquid and holding in suspension specific elements, derived from the secretory glands of the genital apparatus and from the desquamation of the various canals through which the semen passes. Semen contains about 6 per cent, of organic and 4 per cent, of inorganic matter. Its chief chemical characteristic is the presence of spermin (C2H5N)2, which is at least isomeric, if not identical, with diethylen-diamin, according to Ladenburg and Abel. This spermin, which is derived largely from the prostate gland, combines with the phosphoric acid radical to form spermin phosphate, which crystallizes in the form of four-sided spindles or prisms which may appear as flattened needles. In some cases these crystals resemble very closely the diamond-shaped double pyramids known as Charcot-Leyden crystals, which are found in the sputum. They are, however, of a different crystalline group and are soluble in formalin, while those found in the sputum are insoluble in this menstruum. These spermin crystals are known as Bbtt- cher's crystals. Miescher has studied the composition of the heads of the spermatozoa and has been able to isolate certain bodies, known as protamins, which are the simplest type of protein material. Microscopic Examination. The most important and characteristic constituent of semen are the sper- matozoa.1 These sexual elements consist of an anterior broader portion or head and a narrow thread-like tail. The former is oval or egg-shaped and meas- ures about 5 microns in length, 4 in breadth, and 2 in thickness. Just behind this pyriform head is a short cylindrical portion measuring 6 microns in length, 1 Koessler (Trans, of Chic. Path. Soc., vol. 8, 1912, p. 280) has shown that the sperma- tozoa may act as direct carriers of infectious agents, among these being the Spirochaeta pallida. 366 SECRETIONS OF THE GENITAL ORGANS. 367 which is known as the middle piece. This tapers somewhat to the point of union with the tail. This so-called tail is a thread-like posterior portion and is approximately 45 microns in length. In the freshly voided semen these tail portions show active undulatory whip-like motions, which persist for 24 to 48 hours, and even longer under proper conditions, and enable the spermatozoa to progress from point to point. Alkalies seem to favor this movement, while dilute acids inhibit it very rapidly. This movement of the spermatozoa is closely associated with their sexual activity, as the cells showing no movement Fxg. i 14.-Normal semen. when freshly voided may be regarded as possessing no functional power. For a discussion of spermatogenesis as well as of fertilization of the ovum the writer would refer to the admirable description of McMurrich.1 Besides these characteristic portions of the semen, large numbers of lecithin globules are seen, which give the milky appearance to the fluid. Espe- cially to be noted among the cellular elements are the so-called corpora amylacea which resemble very closely starch granules, having concentric striations, a finely granular center and occasionally a nucleus. These cells take a distinct blue color on treatment with iodin solution. Moreover, various epithelial 1 The Development of the Human Body, Philadelphia, 1907. 368 DIAGNOSTIC METHODS. cells are observed, which are derived from the several glands contributing to the composition of the semen. Some of these cells are distinctly granular, some contain fat globules and very closely resemble the colostrum corpuscles of the lacteal secretion, while some of the granules resemble myelin. In rare cases cylindrical casts are seen which simulate the hyaline casts of the urine, but as a rule they are larger and longer. These are supposed to be derived from the prostate gland and seminal vesicles. If the semen be allowed to stand for a few minutes, several types of crystal may be observed, especially the phosphate of spermin, ammonium-magnesium phosphate, fatty acids, and oxalate of calcium. This last crystalline component is especially noted in the urine in cases of spermatorrhea. Pathologic Variations. Direct pathologic variations in the semen are limited to two conditions. Either spermatozoa are absent or those present are nonmotile. The deter- mination of the presence of spermatozoa in the semen or in suspected stains is a matter of simple microscopic examination. With the question of the motility of such elements, when present, the conditions under which the examination is made may markedly influence the findings. If possible, semen should be examined, with regard to the motility of the spermatozoa, as soon as ejected, but if such is not possible the fluid must be kept warm until examination may be made. It is absolutely out of the question to make a positive diagnosis of true nonmotility of spermatozoa from examination of specimens, which have been allowed to cool. In some cases, if the time has not been too extended, warming may bring back the motile power of these cells, but in many cases it does not. It is, therefore, unjust and unwise to brand a man as sterile without absolute proof that such a condition really exists. If no spermatozoa are present, especially after several examinations, sterility is absolute. This condition is known as azobspermatism. According to Kehrer, 40 per cent, of cases of conjugal sterility are due to the absence of spermatozoa in the semen. It is, therefore, necessary that the ordinary gynecological idea that women are the responsible factors in the family sterility should be, at least, partly borne by the man, as it is unjust to the woman to blame her for faults existing in the husband. The writer does not wish to be interpreted as stating that sterility does not frequently exist in women, but he does desire to emphasize the point that many more men are sterile than is generally supposed and that the sterile women are in this condition largely through the results of gonorrheal infection through their husbands. Spermatozoa may be absent from the semen during convalescence from acute febrile conditions, valvular heart disease, and in general conditions associ- ated with lowered nutrition. On the other hand, the constant presence of spermatozoa in the urine, as well as in the semen, may be noted as a result of various pathological conditions as well as of venereal excesses or masturbation. To this condition is given the name of spermatorrhea. SECRETIONS OF THE GENITAL ORGANS. 369 Medicolegal Aspects. Not infrequently the physician is called upon to decide whether certain stains are due to spermatic fluid or whether assault has been committed. If the question is one of suspected rape, an examination of a drop of the vaginal fluid or of scrapings from the vulva or vagina will usually reveal the spermatozoa. Of course other signs, which are important from the medicolegal point of view, will be observed in the examination of the external organs. The stains usually subjected to medicolegal examinations for the presence of spermatic fluid have a grayish-yellow color, their size is somewhat variable, their contour usually irregular, and the linen upon which the stain is usually found is almost as stiff as if it had been starched. As spermatozoa are very resistant to the action of reagents as well as to putrefactive processes, they may be detected many years after the stain was made. It is, therefore, almost an impossibility to say by examination of a stain anything about the length of time the stain has been upon the cloth. A fragment of the linen, which shows the stain, is placed in a watch- glass and allowed to soak for one hour in 30 per cent, alcohol or in faintly alkaline water. It is then lifted from the solvent, placed in another watch- glass, and teased with needles in a solution of 1 per cent, eosin in glycerin. A few drops of this mixture are then placed upon a glass slide, covered with a cover-glass, and examined with a high-power dry lens. Spermatozoa, if present, will practically always be seen by this method. The heads are stained a deep red while the tails, which are usually broken off by the teasing, show a light reddish tint, which distinguishes them from the unstained vegetable fibers. Florence's Test. It not infrequently happens that spermatozoa may not be found, although the stain be due to spermatic fluid. The principle of this test is that spermatic fluid when treated with a solution of iodin in potassium iodid gives crystals which were supposed to be iodospermin. According to Bocarius, this substance is not iodospermin, but an iodin compound of cholin. This test would be given, therefore, by any substance containing cholin and cannot, for this reason, be distinctive for spermatic fluid. Such being the case, a negative result is of far greater importance than is a positive one. The reagent used consists of 1.65 grams of iodin and 2.54 grams of potassium iodid dissolved in 30 c.c. of distilled water. If a drop of spermatic fluid or of an aqueous extract of a suspected stain be treated with a drop of this solution and immediately examined under the low-power lens, long rhombic platelets of a dark brown color, fine needles, or lance-shaped bodies often grouped in rosettes may be observed. A positive reaction is seen many years after the formation of the stain so that a positive result is of value when other sources of cholin are excluded. Technic. 370 DIAGNOSTIC METHODS. Barberio's Test. Barberio has found that the treatment of a drop of spermatic fluid or an aqueous extract of a suspected stain with a saturated aqueous solution of picric acid gives immediately a precipitate of sharply refractile, yellow, ovoid or needle-shaped crystals which gradually increase in size. This test was supposed to be of much greater diagnostic importance than that of Florence, but the recent work of Fraenkel and Muller has shown that the crystals are not sufficiently characteristic to permit of an absolute diagnosis. They call attention to the fact that substances other than spermatic and prostatic fluids may give similar crystals, but that in such cases these crystals are isolated and form usually on the border of the drop, while with spermatic fluids the crystals are numerous and are formed throughout the specimen. These workers recommend this test for the recoghition of prostatic secretions or for the condition of azobspermatism, but caution the worker against making an absolute diagnosis from its presence in medicolegal cases. A negative result does not necessarily exclude the presence of semen. It will be seen, therefore, that neither one of the microchemical tests given above should be regarded as absolutely indicative of the presence of semen. It is much better practice to make repeated search for spermatozoa than absolutely to identify a stain as semen by the microchemical method. II. Female Secretions. (i). Vaginal Secretions. The normal secretion of the vagina is scanty, usually just sufficient to moisten the mucous membrane. It is clear or occasionally opalescent, semi- liquid in character, and is composed largely of mucus and epithelial masses. Its reaction appears variable. As a rule, it should be considered acid in the case of virgins, while in those who have borne children it is usually alkaline. Little is known regarding the chemical properties of this secretion. From the clinical standpoint the normal vaginal secretion is of importance owing to the fact that it possesses marked bactericidal properties. According to Krbnig, pus organisms introduced into the vagina of pregnant women disappear in from four to thirty-six hours. Whether this bactericidal power is due to the reaction of the secretion or to some unknown agent is unsettled. A remarkable fact seems to be that frequent irrigation of the vagina with water or antiseptic solutions decreases the bactericidal power. If this be true, it is questionable whether frequent douching is advisable. Microscopic Examination. Outside of the great number of large, irregular, stratified, squamous epithelial cells which are constantly found in the vaginal secretion, one observes mucous corpuscles, a few large mononuclear leucocytes, cellular debris, and numerous bacteria. The bacterial flora of the vagina is very extensive. SECRETIONS OF THE GENITAL ORGANS. 371 These organisms are normally saprophytic and rarely take on pathologic functions, but they may occasionally do so. Among these bacteria we find the colon bacilli, streptococci, staphylococci, and bacilli which are not unlike true diphtheria bacilli. None of these organisms are particularly important from the clinical standpoint and will be passed with mere mention. We do find, however, certain organisms which give rise to no particular clinical symptoms, but which are extremely confusing in the examination for the presence of the gonococcus. As will be learned from the later discussion of this organ- ism of Neisser, it appears in the form of biscuit or coffee-berry shaped diplococci, which are both intra- and extracellular and do not stain by Gram's method. The chief of these confusing organisms has been called the orchiococcus of Eraud and Hugounenq. This has the same morphology as the gonococcus, but is slightly larger, is negative to Gram's stain, and is rarely intracellular. It is, however, differentiated by the fact that it grows easily upon ordinary media, while the gonococcus requires special media. The more or less normal presence of this orchiococcus should be constantly in mind and a diagnosis of gonorrhea made only when clinical symptoms are present to point to the gonococcus. It is wise in all doubtful cases to resort to cultivation, as one may very much regret having made a diagnosis of gonorrhea when such did not really exist. The practitioner should be cautioned to take his smear high up, in the vagina, preferably from the region of the cervix uteri. If this be done as a routine fewer specimens will be found showing these confusing orchiococci. Smears are frequently sent to laboratories for examination which will not show the gonococcus even though present in large numbers in the vagina. It is not sufficient to take a specimen from simple purulent material which maybe present in the lower portion of the vagina as the organisms are frequently absent in these locations. Pathology. Physiologically, an increased vaginal secretion (blennorrhea) is seen during sexual excitement, preceding menstruation, and during pregnancy, when a very profuse secretion may be observed. If this secretion contains a large number of epithelial cells and leucocytes, as seen in nonspecific inflammations, it becomes more or less creamy in color and is then called leucorrhea. This type of leucorrhea should be sharply differentiated from the true pus secretions observed in the blennorrhagia of gonorrhea, as the former is not necessarily associated with pus formation. In pregnancy a slight catarrhal vaginitis is not infrequent, so that leucorrhea is more apt to appear at such times. If the inflammatory process becomes intense, large shreds of epithelium may be found and ulceration followed by vaginovesicle or vaginorectal fistulas may be observed. Such pathologic findings are usually the result of gonorrhea. In slight catarrhal conditions of the vagina yellowish-gray patches may be seen, which are due to infection with mycotic fungi. Blennorrhea. 372 DIAGNOSTIC METHODS. It is not infrequent to find the trichomonas vaginalis in the vaginal prepara- tions. This organism has been previously discussed in the section on Feces, to which the reader is referred. The oxyuris vermicularis as well as its ova have been reported in the vaginal discharge, but do not seem to have excited any pathologic changes. Purulent Secretions. True purulent secretions arising from the vagina are almost always due to the presence of the gonococcus. This organism is accountable for a large number of gynecological conditions, so that it is wise for the practitioner to be able to identify it both from its clinical manifestations as well as by its laboratory detection. In doubtful cases cultures should be made and a portion of the pus dropped .into the eye of a rabbit. The gonococcus itself may later be recovered from the conjunctival exudate. It is not to be assumed that the finding of the gonococcus in the vaginal discharge is necessarily evidence of a gonorrheal vaginitis or vulvovaginitis. It has been shown that the semen of the male as well as the urethral discharge may contain gonococci and that these may be introduced into a perfectly normal vagina without necessarily setting up gonorrhea. However, this is unusual. The gonococcus may arise from the urethra, the Bartholin glands, acute inflammatory processes of the uterus, or from a ruptured pyosalpinx. In any of these cases the gonococcus may be found in the vaginal discharge, so that the laboratory worker should be guarded in his diagnosis of a vaginitis. Further, suppurative processes which were originally due to the gonococcus may later take on a type of mixed infection or even become of the sterile type. This latter condition is especially observed in old Bartholinitis, metritis, and cystic salpingitis. The organisms usually associated with the gonococcus in the mixed infection are the streptococcus, staphylococcus, colon bacillus and pseudodiphtheria bacillus. In chronic metritis or salpingitis it is not infrequent to find the tubercle bacillus as the causative agent. Fetid Secretions. In these conditions the pus is usually chocolate colored, has a fatty ap- pearance, is extremely repulsive, is frequently sanguinolent and contains numerous degenerated cells as a result of marked leucolysis. This condition is especially observed in puerperal infection and may be extremely severe. (2). Uterine Secretions. Normally, the uterus has no secretion beyond a slight mucoid one which is recognizable clinically. In inflammatory conditions, during normal men- struation, or following abortion or parturition, certain types of discharges are observed which have some clinical importance. SECRETIONS OF THE GENITAL ORGANS. 373 Menstruation. Under normal conditions the menstrual fluid is at first mucoid in char- acter, but within a short time red cells appear and later the discharge takes on almost the character of pure blood. This menstrual fluid should be bright red in color, should contain no clots and should be discharged without causing active pain. This fluid contains red cells, leucocytes, and prismatic epithelial cells showing large areas of fatty degeneration. The duration of the men- strual period is variable, running between two and five days. The amount of blood lost under normal conditions averages about 200 grams, but may be much larger under pathologic conditions. Not infrequently do we find cases in which menstruation is associated with marked pain during more or less of the period of flow. This condition is known as dysmenorrhea and may be associated with the exfoliation of large shreds of mucous membrane, in some cases reported these shreds constituting almost a cast of the uterine cavity. To this latter condition is given the name membranous dysmenorrhea. For the pathologic significance of these abnormal types of menstruation as well as for a discussion of the condition associated with failure of menstruation (amenorrhea) the writer must refer to works on gynecology. By this term we have reference to the discharges from the uterine cavity during the puerperium. At first such discharges consist of blood, which may be in the form of clots, and decidual shreds along with epithelial cells which are probably of vaginal origin. This type is known as the lochia rubra or cruenta. During the next two or three days the discharges become paler and thinner, the red cells diminish and the leucocytes increase, while the decidual shreds may continue approximately the same. This type is known as the lochia serosa. After about a week the discharge assumes a grayish or yellowish color and a creamy consistency, the red cells diminishing rapidly and the white cells increasing markedly. Microscopic examination shows, besides the leucocytes and epithelial cells, numerous fat globules and cholesterin crystals. This discharge may continue during the remainder of the period of uterine involution and is known as the lochia alba. Under normal condi- tions the lochial discharge has a faint odor, but is never fetid. If a portion of the placenta or membranes have been retained, the lochia may assume a dirty brownish color and become extremely fetid. After the first two or three days numerous bacteria, such as staphylococci, streptococci and colon bacilli, may be present, but no untoward symptoms exist unless these, along with other saprophytes, give rise to a distinct puerperal infection or sapremia. The Lochia. Amniotic Fluid. This is a thin, whitish or pale-yellow fluid containing the constituents of ordinary transudates. The reaction is neutral or faintly alkaline, the specific gravity varies between 1002 and 1008, and the amount of solids rarely reaches 2 per cent. The albuminous bodies are principally vitellin, serum 374 DIAGNOSTIC METHODS. albumin, and traces of mucin, while glucuse is absent. Urea and allantoin are present in traces, while creatinin has occasionally been reported. The amount of amniotic fluid varies between 700 and 1,000 c.c. Under pathologic conditions, however, this amount may be increased or decreased, giving rise on the one hand to polyhydramnios or dropsy of the amnion and on the other to oligohydramnios. For a discussion of these pathologic varia- tions as well as for a treatment of the subject of pathologic changes in the membranes, the writer will refer to works on obstetrics. Abortion. The recognition of abortion is usually made by examination of the material discharged from the uterine cavity. Usually one finds blood-clots in which the villi of the chorion are present. These usually appear as club-shaped Fig. i 15.-Chorionic villi. (McMurrich.) masses with epithelial coverings, showing the characteristic capillary network. Moreover, decidual cells are usually present, and may be recognized by their large size, their round, polygonal, or spindle-shaped form, and their irregular and large nuclei with nucleoli. Vesicular Mole. This condition has been called dropsy of the villi of the chorion, hydatidi- form degeneration of the chorionic villi, cystic mole, and myxoma of the placenta. One of its most important symptoms is the expulsion through the vagina of the vesicles forming the degenerated mass. The mole is a mass of peduncu- lated vesicles which closely resemble a bunch of grapes or gooseberries. Each vesicle may vary in size from a millet seed to a large hazelnut and contains a fluid which is usually colorless and limpid, but may be reddish and some- SECRETIONS OF THE GENITAL ORGANS. 375 what dense. Microscopic examination of the tissue shows the peculiar myxo- matous degeneration of the chorionic villi. Carcinoma. It is not infrequent to observe, in cases of severe hemorrhage through the vagina, the appearance of occasional shreds, which on microscopic exami- nation show the characteristic appearances of carcinoma of the cervix or body of the uterus. The diagnosis of carcinoma, however, would better be made upon sections removed by the surgeon rather than upon shreds found in the hemorrhagic fluids. Although a diagnosis may be at times possible, it should be somewhat guarded unless the clinical symptoms are distinctive. CHAPTER VIII. THE BLOOD. I. General Considerations. The blood is perhaps the most important tissue in the body, inasmuch as it is at once the purifier and the nutritive source of the cell. Any normal or abnormal product of cellular activity finds its way ultimately into the blood, either to be taken up by the assimilatory organs or to be thrown out by the excretory ones. While but relatively few disease processes are as- sociated with diagnostic findings in this tissue, yet many are characterized by definite manifestations which are invaluable aids to the clinician. It is, therefore, of the utmost importance that we should have a proper knowledge of the normal blood in order better to understand the various phases which characterize abnormal blood and which give to hematology such an interesting and important position in diagnosis. While it is true that some of the methods involved in hematological examinations require definite apparatus and a large experience for their proper interpretation, yet the results obtainable from the ordinary routine blood examinations are so invaluable, being in some cases pathognomonic, that no practitioner should consider himself fitted to give his patients the proper service without being equipped with a clear working knowledge of the methods of examination and the findings of normal and of abnormal blood. It is essential to remember that certain physiologic as well as pathologic conditions influence the quantity and quality of the blood. So great are the effects of digestion, exercise, nervous factors, massage, cold, heat, sweats, dysentery, constitutional and specific diseases, that one does not wonder at the many conflicting reports of cases showing widely varying hematological findings. As Grawitz has pointed out, no conclusion should be drawn from an examination of the blood without taking into consideration the physiologic and pathologic condition of the patient. It is a pleasure to observe in these days the tendency toward a more rational and thorough study of the plasma, the so-called "intracellular fluid" of the blood. We have forgotten, in our enthusiasm over the many valuable findings obtained from histological investigations, that the relations of the fluid portions of the blood are, in some cases, of quite as much importance as are the variations in the cellular elements. It is necessary only to cite the work on lysins, precipitins, agglutinins, opsonins, etc., to show the value of a more extended study of the plasma or serum. Regarding the technic of blood examinations, the writer will have much 376 THE BLOOD. 377 to say later, but he wishes to impress upon his readers one point which has been well expressed by Turk, namely, an indispensable basis for the proper utilization of any diagnostic, prognostic, or therapeutic method of clinical examination is a knowledge of the absolute limitations of the method. Reliable results can, however, be obtained only by those who are thoroughly familiar with the principles as well as with the technic and the little "knacks" of the method used. It will be found, when the attempts are made to apply the methods outlined, that quite as much depends on the exactitude with which the separate details are carried out as upon the selection of the method itself. It is not to be expected that a first trial will yield exact results or that a few determinations will perfect one in the methods of examination. Experience is the only teacher that can equip one with the skill and power of interpretation necessary to cope with the many difficulties to be overcome in the hematological investigations. II. Physiology and Chemistry. (i). Blood Formation and Blood-forming Organs. While it is impossible in a general work of this character to go into great detail regarding the formation of the blood, yet it seems to the writer that a brief discussion of this subject is extremely valuable both to the student and practitioner. The following section, taken largely from Ewing and McMurrich, will outline the generally accepted views on these points. Red Corpuscles. According to Kblliker, the first blood-corpuscles have their origin, in embryonal life, in the embryonic heart and blood-vessels. They appear as nucleated colorless cells, which later develop into colored corpuscles by the appearance of hemoglobin in some of the cells of the mesodermal cord, which cells go to form the first capillaries. Upon the formation of these vessels the cells lie within them as nucleated reds. At this time there are no true leucocytes and none appear until after the complete formation of the red cells, which is advanced as an argument against Pappenheim's theory of single origin of red and white cells. It will thus be noted that the vessel wall and the primitive erythrocyte have a common origin in the meso- dermal cord, the peripheral cells going to form the endothelium of the vessel and the internal cells the corpuscles. Up to the end of the fourth or fifth week of embryonal life all of the red cells are nucleated; while from that time on the relation of the nonnucleated to the nucleated forms gradually increases until at birth few if any nucleated cells obtain. In later embryonal life (about the third month), the liver becomes the chief seat of blood formation. During the fifth month, the spleen and lymph- glands take up this work, and finally the bone-marrow becomes the seat of such activity. 378 DIAGNOSTIC METHODS. In extrauterine life, the bone-marrow is the chief point of formation of the red cells, but under pathological conditions the spleen and liver may assume their embryonic functions. It appears that the formation of nucleated reds in the adult is practically the same as in the embryo and that, at all periods of life, the red cell is the product of several series of mitoses of a colorless mesoblastic cell. The difficulty of tracing this series, from the large nucleated red cell to the colorless mesoblastic "mother-cell" in the marrow, has given rise to the diverse opinions now held regarding the ultimate development of the red corpuscles. Leucocytes. The earliest indications of the formation of leucocytes are seen in the presence of primary wandering cells, of mesodermal origin, which are found principally in the loose connective tissues of the early embryo. Though of mesodermal origin they are, from the first, quite distinct in morphology and, apparently, in function from the capillary endothelium and fixed con- nective-tissue cells. Their development has been traced by Ziegler to masses of mesodermal cells surrounding the cords from which the capillaries are formed. It thus seems that originally the parent leucocytes lie outside the vessels, into which they make their way by virtue of ameboid powers. Most observers find that the primary wandering cells produce, by mitotic division, one or more generations of colorless cells which gradually approach, in morphology, the early basophilic leucocytes of the circulation. Denys, Lowit, Ziegler, von der Stricht, and others claim that red cells and white cells develop from separate series of cells, which have become differentiated from the primary mesodermal cells with the first appearance of blood and blood-vessels. Kostianecki, Muller, Schmidt, Saxer, Pappenheim, and others believe that the primary wandering cell persists in the blood-forming organs as the parent of both red and white cells. Before the leucocytes begin to appear in the circulation, mitotic figures are abundantly seen in the primary wandering cells in various situations. These are gathered in groups, first in the loose connective tissues of various regions, where lymph nodes subsequently develop; but the chief seat of the production of the leucocytes is found in the embryonal liver. In both situa- tions the wandering cells are found in the lymph and blood capillaries, in the interstices of the connective tissues, and between the liver cells. In later embryonal life the process is gradually transferred from the liver to the lym- phoid and adenoid tissues, as indicated by the development of lymph nodes, spleen, marrow, and thymus. Under normal conditions, the reproduction of leucocytes, in the adult, is limited to the lymphoid structures both of the lymph-glands and bone-marrow. (2). Total Volume of Blood. The various methods which have been advanced for the estimation of the total quantity of blood in the body are subject to such wide variations THE BLOOD. 379 that they have yielded little exact information regarding this subject. The procedures advocated by Valentine, Vierordt, Buntzen, and Thibault have an error sufficiently great to exceed the physiological and pathological varia- tions of the blood. By these methods, the quantity of blood in the body has been estimated as equal to one-thirteenth of the body-weight. The so- called clinical methods of Quincke or of Tarchanoff are of purely theoretical interest, because certain factors, such as the appearance of the patient or the volume of the pulse, are taken into consideration in making a rather unreliable guess as to the total quantity of blood in the body (Buckmaster). By the use of a method, recently introduced by Haldane and Smith,1 the total volume of blood may be fairly accurately estimated. This method is based on the following points. The capacity of hemoglobin for oxygen and for carbon monoxid is identical. On the assumption that none of this latter gas is oxidized in the body, and that no substance in the blood, other than hemoglobin, unites with it, the experimenter is able to determine the CO capacity and hence the O capacity of the blood. This method has, however, little clinical application and will be left with reference to the original work. In the cases studied by Haldane and Smith by this method, the average value was 3,240 c.c. or, on the basis of a specific gravity of 1,060, about 3,434 grams. This yields, according to Smith, a figure ranging between one-six- teenth and one-thirtieth of the body-weight. In obese persons the volume of blood is less, proportionately, than in the more normal specimens of mankind. The question of the volume of blood in the body is of great importance in the study of the changes taking place in this tissue. It must be remembered that the number of red or of white cells in a cmm. of blood will depend upon the total amount of blood present. If for any reason the volume is diminished or increased, corresponding changes, in the inverse sense, will be observed in the number of the cellular elements per cmm. It seems to the writer, there- fore, that certain factors, not ordinarily taken into account in blood examina- tions, should be known before any definite report is made upon a blood count. It may be readily seen that a concentration of the blood, due to hemorrhage, diarrhea, sweating, etc., will lead to an apparent increase in the number of corpuscles. Certain abnormal blood counts, known to the writer, have un- doubtedly been due to variations in these points. Certain physiologic and pathologic conditions lead to definite changes in the volume of the blood, as such, or of some of its constituents. As Plehn2 has recently shown, the volume remains quite constant or is adjusted through the activity of the capillary endothelium and through the influence of the nervous system. However, definite changes of a more or less transitory nature do occur and exert marked influences on the results of blood examinations as well as upon many pathologic conditions. 1 Jour, of Physiol., vol. 20, 1896, p. 497; Ibid., vol. 25, 1900, p. 333. 2Deut. Archiv. f. klin. Med,, Bd. 91, 1908, S. 1. 380 DIAGNOSTIC METHODS. Oligema. By this term is meant a reduction in the total volume of blood, both as regards the liquid and the cellular portions. This condition is most frequently noticed after profuse hemorrhage and may be so marked that death results. In other cases, in which the hemorrhage is less extensive, the loss of blood is made up by osmosis from the lymph spaces into the capillaries and, later, by an increase of the cellular elements due to compensatory activity of the hematopoietic organs. Plethora. The opposite of the preceding condition is known as plethora, a state characterized by an increase in the total volume of blood. There has been much discussion as to whether a true plethora exists, but there can be little doubt that a transitory plethora vera may occur as a result of direct transfusion of blood, and also, according to Bergmann and Heissler, who have established the fact that there is a direct ratio between the volume of blood and the size of the heart, on the one hand, and the muscular development of the subject, on the other, as the result of increased muscular activity, provided the loss of fluid by perspiration is not excessive. Such a plethora disappears, of course, in a very short time. In this discussion we must distinguish between a serous and a cellular plethora. By the former is meant an increase in the volume of blood due to excessive quantities of its liquid and soluble constituents; while by the latter we understand an increase in the number of corpuscular elements, that is a polycythemia. Serous plethora is most frequently observed in organic lesions of the kidneys and of the heart, in which a diminished elimination of water and inorganic constituents is noted. This condition is usually of a transient duration, as the volume of blood is soon restored to normal by intra- capillary transudation and by diffusion. Osmotic effects must also be taken into consideration here, inasmuch as the salts will tend to diffuse out from the blood and will consequently draw water after them, giving rise, under certain conditions, to dropsical effusions of a more or less transient duration. Hydremia. Another condition of the blood is frequently observed, in which an in- crease in the quantity of the liquid constituents is observed. This is known as hydremia and is different from serous plethora, as the latter carries with it an increase in the saline as well as in the watery portion of the blood. In hydremia the specific gravity of the blood is reduced, while in serous plethora it is increased. As Engel1 has shown, the estimation of the coefficient of refraction of the blood serum is a reliable method for the clinical study of the subject of the water content of the blood. Hydremia may be produced by any factor which changes the normal relationship of the blood constituents in such a way that the watery portion 1 Magyar Orvosi Arch., Bd. 7, 1906, S. 555. THE BLOOD. 381 is relatively increased. It is in these more or less physiologic states that we are apt to observe the greatest variation in the blood counts. As the cellular elements are not simultaneously increased, the drop of blood under examination contains relatively fewer cells than normally. The most common physiologic causes of hydremia are the ingestion of large quantities of fluids, saline trans- fusions, and vasomotor dilatations as a result of exercise or nervous influences. In severe anemias we find the watery portion of the blood relatively increased. In dropsical states, following cardiac or renal lesions, we often observe such a condition, whose duration will depend, of course, upon the etiological factors of the trouble. Anhydremia. This is a condition characterized by a diminution in the liquid constituents of the blood. There is no change in the cellular elements and hence a blood count will show an erroneous increase in the number of cells. In this condi- tion the specific gravity of the blood is naturally increased. Anhydremia follows any condition which results in the excessive loss of fluid from the body, as, for instance, that following profuse diarrhea, polyuria, sweating, vomiting, and effusions into the various serous cavities of the body. According to Oliver, this state may be caused by influences which increase the arterial tension and hence bring about an increase in the passage of water from the vessels into the tissues. Thus, for instance, we may observe anhydremia following local and general exercise, massage, bathing, etc. (3). Volume Relations of Cells to Plasma. The study of the relationship between the cellular and the intracellular portions of the blood is a comparatively recent addition to the technic of blood examinations. This determination is based on the principle that the corpuscles may be thrown by centrifugal force to the distal end of a calibrated tube, while the plasma will collect in the proximal portion. If the tube be properly calibrated, the percentage relations of the cells and plasma may be readily ascertained. These ideas were used by Hedin in elaborating the earlier methods of Blix. Daland's Hematocrit. Fig. 116.-Daland's hematocrit. Daland has introduced a modification of the clumsy model of Hedin and has succeeded in simplifying the technic to such an extent that this method is directly applicable to clinical use. His instrument is shown in the accom- panying cut. 382 DIAGNOSTIC METHODS. One of the calibrated glass tubes is fitted with the rubber tubing and is filled with blood from the ear or finger. The forefinger, smeared with a little vaselin, is then placed over the beveled end of the tube and the rubber tubing withdrawn. Insert the tube into one arm of the frame, the other arm of which should carry the second tube filled in the same manner, in order to balance the instrument and to control the reading. Rotate the spindle for three minutes at such a rate of speed as will insure 10,000 revolutions per minute (80 revolutions of handle). In this way the corpuscles are separated Fig. i i7.-Hematocrit tube from the plasma and are distinguishable as a distinct column, which may be read off directly from the graduations of the tube. These divisions will give the percentage relations of the cells and plasma as the tube is divided into 100 equal portions, each division of the scale representing approximately 100,000 cells. This latter makes it possible to make a rather rough blood count with this instrument, but it is to be remembered that accurate results cannot follow, as we find such variations both in the size and elasticity of the cells in the different conditions in which the number of cells is most sought. Volume Index. Recently Capps1 has introduced the conception of volume index, that is the quotient of the volume per cent, as obtained with the hematocrit, and the blood count in terms of per cent. Sahli advises the use of the expression volume quotient or volume value for this factor. By means of the hematocrit the volume of the red cells, as compared with that of the whole blood, is taken. In normal cases this is about 50 per cent, which is reckoned as one. Hence the volume of the red cells may be obtained directly in percentage value. The red cells are then counted by the method to be later outlined, and the result is expressed in percentage by comparing this count with a so-called normal one of 5,000,000 red cells. By dividing the volume of red cells (in per cent.) by the per cent, of red cells, Capps obtains his volume index of the red cells. In normal conditions this quotient is one. According to Capps, an increase of this index is a constant factor in pernicious anemia. The color index never exceeds the volume index in such cases, which fact shows that there is no supersaturation of the corpuscles with hemoglobin. In primary and also in secondary anemia this factor is diminished. Here we find the color index often falling below the volume index. This method may be used in detecting various pathological conditions of the blood. According to Emerson, it is used in the Johns Hopkins Hospital 1 Jour. Med. Res., Vol. io, 1903, p. 367. THE BLOOD. 383 in ascertaining the presence of lipemia, cholemia, or hemoglobinemia. It would seem to the writer that it could be employed with advantage as a routine procedure, especially in hospital practice. The osmotic pressure of the plasma plays a great role in this determination, as the concentration of the blood may be such as to cause swelling or shrinkage of the cells. As Capps has found, normal red cells with a volume index of one have their discoplasm saturated with hemoglobin. Hence, if the hemoglobin index becomes greater than one, an enlargement of the red cells is indicated. Upon the other hand, the color index may fall, regardless of a corresponding lowering of the volume index. It follows, therefore, that, if the color index is above normal, the volume index must also be increased; while, if the hemoglobin index is below normal, the volume index is not necessarily diminished (Sahli). (4). Methods of Obtaining Blood. The method of obtaining blood for examination depends upon the amount desired and on the examination which is to be made. For ordinary routine work only a few drops are necessary, while for bacteriological investigations 2 to 20 c.c. may be required. In obtaining the blood no set rule can be laid down as to the proper place from which to take the specimen. We should select the part which promises the best results, avoiding naturally the points which are cyanosed, eczematous, edematous, hyperemic, cold, unduly heated, or, in other words, any part which is not normal. The ear usually furnishes the best results, in the writer's opinion, because its puncture is relatively painless, it is easily accessible (which point is often of importance in attempting to obtain blood from those who are comatose or who refuse to submit their hand for puncture), and because the patient, who may be easily affected by the sight of blood, can not see the drop. If the lobe of the ear is thick it is usually pricked on the flat side, but if it be thin it is well to make the puncture on the edge and parallel to the surface. Some workers prefer the palmar surface of the ball of the middle or second finger of the left hand and others advise pricking the arm over a small super- ficial vein. In cases in which repeated examinations of the blood are to be made the parts should be varied in order to prevent soreness and also to avoid anesthesias which occasionally follow repeated use of the same site of puncture. Quite a number of special forms of blood needles, as, for instance, those of Francke and of Daland, are on the market, and each of them has its ad- vocates. Many of them are in the form of special holders, which permit a puncture of a desired depth to be made without any danger of going too deep. These are to be recommended to those only who seem unable to control their stab. The writer does not find that the results obtained by the student are any better with such instruments than are those following the use of the ordinary Hagedorn surgical needle. In lieu of any better article, a clean steel pen with one nib broken will yield admirable results. The one thing to bear in mind 384 DIAGNOSTIC METHODS. in selecting a needle for blood work is that the point must have a cutting- edge and should not be round or sharp nor should it be too long or slender. Having decided upon the part from which the blood is to be taken, this surface is washed carefully with water and alcohol or ether and is then thor- oughly dried. Avoid any undue rubbing as this will cause hyperemia and will introduce an error into the work. It is, as a rule, unnecessary to sterilize the needle but if it seems advisable to do so on account of the patient's attitude, the sterilization is best done by heat, hydrogen peroxid or alcohol, allowing the needle to cool before making the puncture. In this latter part of the technic, much depends on the amount of blood desired as regards the use of a short quick stab or a slow steady puncture. The latter procedure usually yields more blood but is more painful. It is much better to prick the patient too deeply, going even through the lobe of the ear, than it is to subject him to repeated punctures. The part pricked should not be squeezed nor held in a position which will cause an abnormal circulation. If the puncture is successful, the blood will come out in good-sized drops. The first of these are wiped away and subsequent ones used for the examination. As often happens, the blood coagulates fairly quickly, so that the coagulum should be wiped off with a little alcohol followed by a dry cloth. It is im- portant to remember that a patient with hemophilic tendencies or history may bleed very easily from a very slight puncture. Care should, therefore, be taken to question the patient regarding the ease with which blood flows from a wound and also regarding the history of "bleeders" in the family. If considerable blood is desired, resort must be made to venous puncture. The site of this operation is usually the median basilic vein at the bend of the elbow. This vein may be made more prominent by tying a tight bandage around the arm, but the bandage should be removed before the blood is withdrawn, except when serum reactions are to be studied. The site of puncture must be thoroughly cleansed, using the precautions observed before any surgical operation. Likewise, the needle and the aspirating instrument must be absolutely sterile before puncture is made. The instrument best adapted for this purpose is, in the writer's experience, the Liier syringe, which is made of glass with a tightly-fitting glass plunger and adjustable platinum needle. From 2 to 20 c.c. of blood are withdrawn and immediately placed in work. The question of the bacteriological examination and of serum reactions will be discussed in a later section. Fig. 118.- Blood needle (5). Physical Properties. The blood must be regarded as a fluid tissue, consisting of a transparent liquid, the plasma or liquor sanguinis, in which are suspended the corpuscular THE BLOOD. 385 elements, erythrocytes, and leucocytes. Besides these latter cells we find two constituents, the blood plates of Bizzozero and the hemoconien (blood dust), which are hardly to be considered as true corpuscular entities. As it flows from the vessels, blood is a thick, viscid, red liquid, having a peculiar odor, a salty taste, and an alkaline reaction to litmus. If allowed to stand it shows, unless certain precautions are observed, the peculiar phen- omenon of coagulation. In this process the blood is separated into two por- tions, the cellular elements and the plasma, the latter changing, as the process goes on, into serum and the clot (fibrin), which holds the corpuscles in its meshes. In the diagram given below, adapted by Webster and Koch, may be seen the composition of the blood. Serum Albumin. /Serum* Serum Globulin. Glucose, extractives, calcium salts, sodium Plasma< and potassium chlorids, carbonates, phos- phates, etc. 'Fibrinogen (yields fibrin) Blood' Oxy-hemoglobin. 'Red Corpuscles Lecithin. Salts. Cellular Elements White Corpuscles Fibrin Ferment. Blood Plates. •Hemoconien The color of the blood is due to the presence in the erythrocytes of an iron-containing albuminous substance, hemoglobin, which has remarkable affinity for oxygen and other gases. This latter property enables this pigment to play one of the most important rbles in the body economy. Arterial blood is bright red in color, while venous blood shows a purplish-blue tint. These variations are due entirely to the relative proportions of oxygen and of carbon dioxid in the two types of blood. Many different shadings are observed, physiologically and pathologically, in the coloration of the blood, and each is due to some combination of hemoglobin with normal or abnormal substances. The presence of large numbers of red corpuscles in the blood gives rise (A). Color. 386 DIAGNOSTIC METHODS. to a characteristic opacity of this tissue. If, for any reason, such as admixture of blood with water, dilute salt solutions, urea, ether, snake venom, extract of mushrooms, etc., the blood loses its opacity, the change is due to the dis- solving out of the hemoglobin from the red cells. This is the well-known phenomenon of 'Taking" or, better, of hemolysis, about which we will have something to say later. The normal color of the blood is often changed in pathological conditions. Thus, in anemia the blood is pale and watery; in leukemia it may be milky; in diabetes buff-colored; while in poisoning with potassium chlorate it is choco- late-colored and in that with carbon monoxid it is bright red.1 (B). Odor. The odor of the blood is peculiar and indescribable. This halitus san- guinis is due to the presence of certain volatile fatty acids and may be rendered more distinct by the addition of concentrated sulphuric acid, which increases the volatility of these acids (Barruel's test). (C). Reaction. If we are to accept the teachings of physical chemistry, that the alka- linity of a solution is due to the presence of free hydroxyl (OH) ions and that its acidity depends on the surplus of free hydrogen (H) ions, we must grant that the blood is a practically neutral fluid. If, however, we have in mind the acid-combining power of the blood, we must regard the reaction of this tissue as alkaline. It is certain that the blood shows both acid and alkali combining powers due to the presence of protein constituents as well as to both acid- and alkali-reacting substances, the measure of such powers being dependent on the indicator used in the estimation. As the combining power for acids is greater, in the case of blood, than it is for alkaiies, the reaction, as judged by titrimetric methods, must be alkaline. The normal free or diffusible alkalinity of the blood is due to the presence of disodium hydrogen phosphate (Na2HPO4), sodium bicarbonate (NaHCO3), and sodium carbonate (Na2CO3). This total diffusible alkali constitutes, according to Brandenburg, about 20 per cent, of the entire alkalinity and may be measured by dialyzing against known alkaline solutions and observing the concentration at which the strength of the known solution does not change. This factor represents the so-called alkaline tension of the blood and remains fairly constant, in normal cases, at about 60 mg. of NaOH per 100 c.c. of blood, while in pathologic conditions, such as uremia, diabetes, etc., it is somewhat reduced. Besides this diffusible alkali, the blood contains nondiffusible alkali bound to the proteins. This portion represents normally about 80 per cent, of the total alkalinity, and is dependent largely on the cellular content of the blood, as the soluble protein constituents are not generally subject to wide variations. The fluctuations in this nondiffusible alkali are no doubt account- able for the great differences in the figures given for the alkalinity of the blood. 1 See Gaisbock, Med. Klinik, Bd. 8, 1912, S. 1906. THE BLOOD. 387 The subject of the reaction of the blood is one which should furnish, if properly studied by reliable methods, much valuable data upon subjects which are now very obscure. Unfortunately, however, most of the methods at our command are so unreliable and so inexact that the results have little more than a comparative value. As Moore and Wilson1 have pointed out, the titration methods do not give us the true neutrality of the blood, but rather the amount of alkali or of acid which may be added to it without raising the hydroxyl or hydrogen concentration above certain low limiting values. The reaction of the blood depends on the indicator used for the determination of the neutral point and cannot be definitely measured by any method which employs an indicator for such purposes. These writers have introduced the term "reactivity" to indicate the property, possessed by the blood, of combining with both alkalies and acids in such a way as not to raise its ionic composition. In selecting the methods for the quantitative estima- tion of the so-called alkalinity of the blood, I limit myself to those of Dare2 and of Salkowski,3 as neither depends upon the titration of the blood. The figures given by the older titration methods have such a wide range, being as low as 162 mg. and as high as 800 mg. of NaOH per 100 c.c. of blood, that the writer feels compelled to give them up for methods which seem to yield more exact and acceptable results. Method of Dare. This method is based on the fact that the absorption bands of the spectrum of oxyhemoglobin disappear at the point of exact neutralization of the blood. Although it has not been conclusively proven that oxyhemoglobin is not destroyed before the point of complete neutralization is reached, yet the results with this method are rarely so variable as are those with the titration methods. The apparatus (hemoalkalimeter) may be seen in the accompanying illustration. It consists of a special graduated tube (A), a blood pipet (B) which holds 20 cmm. (15 mg. by weight) of blood, a reagent pipet (D), and a hand spectroscope (E). The method of using this instrument is as follows: The tube (A), into which fits the pipet (B), is held horizontally in such a manner that the pipet may be filled by capillary attraction with blood from the finger or ear. With a medicine dropper, containing distilled water, the pipet is washed free of blood, using sufficient water to bring the diluted blood to the zero-point of the Fig. i 19.-Dare's hemoalkalimeter: A, alkalimeter tube; B, automatic blood pi- pet; C, air hole; D reagent pipet. 1 Biochem. Jour. vol. i, 1906, p. 297. 2 Phila. Med. Jour., vol. 11. 1903, p. 137. 3 Centralbl. f. d. med. Wissensch., Bd. 36, 1898, S. 913. 388 DIAGNOSTIC METHODS. tube A. This diluted blood is then thoroughly mixed by shaking and inverting the tube, after which agitation it is allowed to settle. In this process care should be taken to close the opening (C) with the thumb to prevent loss of fluid. The reagent pipet (D) is then filled with the following test solution (N/200 tartaric acid, each c.c. of which is equivalent to .0002 gram of NaOH) and is connected with the end of the blood pipet by means of rubber tubing. Tartaric acid, 0.375 gram. Alcohol (95 per cent.), 100.000 c.c. Distilled water, q. s. ad., 1,000.000 c.c. By compression of the bulb of the pipet the reagent is forced through the blood pipet into the tube A. Mix the fluids by inversion, care being taken to avoid loss of fluid. Adjust the spectroscope to the tube below the zero point after each addition of reagent and observe the spectrum for the presence of the two bands of oxyhemoglobin. This is best done by holding the tube a constant distance from an artificial light. This part of the technic is very tedious, but is necessary until the observer learns to associate the change in color of the mixture with the point of neutralization. If the spectroscopic bands of oxyhemoglobin still persist, more reagent is added and the mixture shaken as before, until these bands can no longer be seen. This is the point of neutralization and the end-point of the test. Read off, from the graduations on the tube, the number of c.c. of reagent used. The following table, representing the graduations on the tube, gives the equivalent of NaOH for every fraction of a c.c. of reagent used. It is calculated on the basis of 15 mg. of blood for every 2 c.c. of the N/200 tartaric acid. Dare considers the normal figure to be 2 c.c. of acid, which represents 266 mg. of NaOH per 100 c.c. of blood. C.C. of Milligrams of NaOH reagent. per xoo c.c. of blood. 3-o 376.0 2.8 360.0 2.6 345-o 2.4 3i9-o 2.2 292.0 2.0 266.0 1.8 239-0 i.6 212.0 1.2 176.0 1.0 i33-o o.8 96.0 o.6 79-o 0.4 53-o 0.2 26.6 THE BLOOD. 389 The writer has found that this method is capable of easy application and, with a little practice, will give very reliable and constant clinical results. Naturally the personal equation arises in the reading of the spectroscope, but this is easily corrected by experience. As the spectroscopic changes are some- times slow in appearing, one should not be in too great haste to finish his determination. The figures obtained with this method are somewhat lower than those with the Salkowski technic and are probably more nearly repre- sentative of the nondiffusible than of the total alkalinity. Method of Salkowski. This method, like the preceding one, has the advantage of avoiding direct titration of the blood, but has the objection that a considerable time is necessary to obtain certain results. The method is very simple and is based on the principle that ammonium salts are decomposed, in the presence of alkalies, with the liberation of free ammonia, which latter substance may be absorbed by a standard acid solution and determined by titration of the excess of acid. Twenty grams of finely powdered chemically pure ammonium sulphate (NH4)2SO4 are placed in the larger lower dish of Schlosing's apparatus (see Urine) and are dissolved in 20 c.c. of distilled water. Ten c.c. of N/10 sulphuric acid are placed in the upper dish. Pour into the lower dish, which contains the ammonium sulphate solution, 10 c.c. of blood, which has been measured in a cylinder previously washed with 1 per cent, solution of sodium oxalate to pre- vent coagulation of the blood. Mix the blood and ammonium sulphate solution thoroughly and cover the dishes as quickly as possible wfith a bell jar, which should be tightly fitted to the glass plate with vaseline. Allow the apparatus to stand at room temperature for five to six days, when the ammonia, liberated by the action of the alkali of the blood upon the ammonium sulphate, will have been taken up by the sulphuric acid. Titrate this acid solution with N/xo NaOH, using methyl orange or rosolic acid as an indicator, until the point of neutralization is reached. Subtract the number of c.c. of NaOH used in the titration from the original number of c.c. of H2SO4 taken and mul- tiply the result by 0.004006, which will give the amount of sodium hydrate equivalent to the ammonia liberated from the 10 c.c. of blood. To obtain the amount of NaOH per 100 c.c., multiply the above result by 10. This method is appealed to on account of its simplicity and its reliability, being certainly more accurate than the methods of direct titration. According to Waldvogel, the normal values obtained by this method are 350 to 400 mg. of NaOH per 100 c.c. of blood, being somewhat higher than those obtained with Dare's method owing to the fact that all of the alkali present, whether diffusible or nondiffusible, will react by this method. The reaction of the blood varies in certain physiologic states, as, for in- stance, during digestion. The alkalinity or the "reactivity to acids" is higher in man than in woman and the child; is at its minimum in the early morning, DIAGNOSTIC METHODS. 390 rises during the afternoon, and falls in the evening; is increased during digestion and falls after digestion is complete; is decreased by excessive exercise and also by a diet deficient in nitrogenous constituents. Many of the beneficial results of massage and baths are unquestionably due to an increase in this factor. It must be remembered that the normal balance between the acid and alkali constituents of the blood is easily maintained, otherwise the system would be in a state of constant disorder, subject as it is to the influence of acids, either produced within itself or taken into it from without.1 As Moore and Wilson have shown, the susceptibility of the living cell to increase in concen- tration of either hydrogen or hydroxyl ions is due to the fact that the protein of the cells possesses strong affinity for both these ions, forming feebly disso- ciated salts with them. Increased alkalinity of the blood is supposed to denote increased antibactericidal power of this tissue. For this reason we must assume some direct relationship between immunity and alkalinity. Moreover, the further study of the opsonic, lytic, and other properties of the plasma may reveal definite influences of the alkalinity upon such processes. It has been found that pathologic changes are very often evident in the reaction of the blood. Thus, for instance, we find in severe secondary anemia and in pernicious anemia a marked decrease in the alkalinity, while in chlorosis we observe little or no change. The degree of alkalinity is reduced in prac- tically all infectious diseases, in uremia, in diabetes, and organic diseases of the liver, etc., while in chronic diseases it may, at times, show an increase. The study of the conditions associated with an "acidosis" would be, undoubtedly, made much clearer by the application of reliable methods of estimating the reaction of the blood. (D). Specific Gravity. The specific gravity of normal blood varies between, 1,055 and the average being 1,060. Certain variations in this figure are observed depend- ing on the sex or age of the patient or upon the time and temperature at which the determinations are made. The most accurate method of determining the specific gravity is, of course, the use of the pycnometer. This method is open to the objection that it re- quires much more blood than can usually be obtained in routine work. In cases in which bleeding can be resorted to without detriment to the patient this method is the one to use, as it gives the most reliable and accurate results. The writer has used it in many cases of pneumonia, where the withdrawal of a certain amount of blood is often beneficial, and finds the results all that could be desired. Besides the quantity of blood (5 to 50 c.c.) which is required there is also necessary a very accurate chemical balance, else the results will be influenced by the inaccuracies in the weighings. The technic is as follows: weigh the pycnometer empty, filled with distilled water, and then filled with blood. Care should be taken to have the 1 See Lundsgaard, Biochem. Zeitsch., Bd. 41, 1912, S. 247. THE BLOOD. 391 vessel absolutely dry and clean before weighing it empty and before filling with either water or blood. Subtract the weight of the empty bottle from that of the bottle filled with blood and divide this figure by the difference in weight between the bottle filled with water and the empty bottle. The result will be the specific gravity of the blood, water being taken as unity. We should be careful in this determination to have the temperature of the water the same as that of the blood in order to insure accurate results. Method of Schmaltz. This method is a modification of the above and does not give quite as accurate results. It consists in the use of small tubes, which hold about i/io c.c. These tubes are constricted at the end to prevent loss of blood and are filled by capillary attraction. The determination is made in the same way as with the pycnometer. More or less manipulative skill is necessary in the handling of these tubes, but the results are sufficiently accurate for most purposes. The more frequently employed methods of determining the specific gravity are the so-called areometric ones. The principle of these procedures is the determination, by the use of accurate hydrometers, of the specific gravity of a liquid mixture, in the center of which a drop of blood will remain suspended. Method of Hammerschlag. This method is a strictly areometric one and consists in the use of a mixture of benzol and chloroform into which a drop of blood is introduced through a capillary tube. If the drop rises in the mixture benzol is added and if it sinks chloroform is used. The point at which the drop of blood remains sus- pended in the center of the perpendicular axis of the mixture is taken as the one representing the specific gravity of the blood. As the means of estimating the density of this mixture, we employ an accurately graduated hydrometer. It must be remembered that the fluid mixture should be well stirred after the addition of either benzol or of chloroform in order to insure uniform density throughout. As the mixture evaporates rapidly we must work quickly and should confirm our results by a duplicate determination. A further precaution should be to allow no bubbles of air to adhere to the drop. This is fairly well accomplished by the use of the capillary blood pipet. It is also essential that the temperature of the mixture should not vary to any appreciable extent. This method is simple and, with the precautions mentioned, will yield good clinical results. The specific gravity of the serum may be tested in this same way, first allowing the blood to coagulate in sealed tubes and then drawing off a drop or two of the separated serum. Naturally, the specific gravity of the blood is a measure of its concentration Fig. 120 -Pyc- nometer. 392 DIAGNOSTIC METHODS. and, hence, of its water content. We are, therefore, certain to find variations in this factor under the influence of any physiologic or pathologic changes, which are associated with fluctuations in the volume of blood. Physiologically, the specific gravity is higher in men than in women and children; is higher in venous than in arterial blood, and is lower after ingestion of large quantities of fluid in the food or after infusions. Clinically, we find that the specific gravity of the blood runs parallel to the number of corpuscles and to the amount of hemoglobin in the red cells. So striking is this ratio that it was formerly used to determine the percentage of hemoglobin in the blood. Any marked alteration in these constituents gives rise to a variation in the specific gravity. Thus, there is observed in anemia in which there is a lowered percentage of hemoglobin, and in those forms of secondary anemia which are characterized more particularly by diminution in the number of cells, a low figure for the specific gravity. In polycythemia, on the other hand, we find the specific gravity increased as a result of the increased corpuscular content of the blood. Pathologically, the specific gravity may run between 1,026 and 1,068. An increase is noted in practically all febrile diseases, in those conditions associated with cyanosis, and in disorders leading to obstructive jaundice. In conditions showing marked diuresis, diarrhea, or sweating an increase is likewise observed, but such changes are usually of slight duration, as the blood soon adapts itself to the condition by withdrawing liquid from the tissues to compensate for the loss in the above processes. In nephritis we may find either an increase or a decrease in the specific gravity, depending on the osmotic changes which take place in this disease. (E). Viscosity of the Blood. Freshly drawn blood has a greasy feeling, which is replaced by a stickiness as coagulation proceeds. This viscosity or internal resistance of the blood depends, to a large extent, upon the cellular content of the tissue and is distinct from the phenomenon of coagulation. Many methods have been advanced for the determination of this property of the blood, that of Determann1 being as clinically accurate and simple as any. So many uncertain factors influence the viscosity that the writer feels that this determination can add little to our clinical knowledge. As Burton-Opitz2 says, "Although the results have been gratifying in so far as normal viscosity- values have been established for the human blood, it seems doubtful whether this method will ever be perfected in a way that small variations can be accu- rately and safely recorded. Nor do the limitations lie wholly in the method. One of the gravest obstacles, encounted in establishing slight differential values, exists in the variability of the viscosity itself." It has been shown that the degree of viscosity (n) is influenced by cold, with- drawal and application of heat, the former factors causing an increase, while 1 Munch, med. Woch., Bd. 54, 1907, S. 1130; Die Viskositat des menschlichen Blutes., Wiesbaden, 1910. 2 Jour. A. M. A., vol. 57, 1911, p. 353. THE BLOOD. 393 the latter lowers it. Hirsch and Beck demonstrate that the lower the specific gravity of the blood the less marked is its viscosity. These results agree with those that indicate that the lower the specific gravity the lower the number of cellular elements. We may readily see, therefore, why the blood in anemia and leukemia shows such a slight tendency to become sticky or to form rouleaux. The researches of Rotky1 show that n varies between 5.02 and 5.52 under normal conditions (water being 1), is 1.69 in anemias, 3.34 to 5.58 in nephritis, 13.56 in febrile states and 16.93 in cyanotic conditions. According to Hess,2 the normal viscosity of male blood ranges between 4.3 and 5.3, while that of female blood varies from 3.9 to 4.9. The normal relation3 Hemoglobin -- ; = I 7 to 2 I. Viscosity (F). Coagulation of the Blood. It is impossible in this place to discuss the physics and chemistry of the process of coagulation, more than to say that this phenomenon is due to the conversion of the fibrinogen of the plasma into fibrin. This change takes place under the influence of a ferment, thrombase, which is present in the leucocytes in the form of prothrombase. This latter zymogen, through the influence of calcium compounds, is changed into thrombase, the active agent in bringing about coagulation. (See Howell4 and Bordet and Delange.5) It is occasionally of clinical importance to know the time of coagulation of the blood under certain conditions, as one of the normal processes of this tissue, when outside the vessels, is coagulation. This process takes place normally in from two to eight minutes, depending on several factors, among which are the length of time the blood is in contact with the tissues, the depth of the incision, the pressure with which the blood is expelled, the nature of the vessel into which the blood flows, and the temperature at which coagulation takes place. It has been found that certain variations in the normal coagulability are present at different hours of the day and are observed when blood is drawn from different parts of the body. The whole question of coagulation of the blood being so intimately associated with its chemical composition and ex- travascular coagulation being only significant of the changes which take place when blood coagulates intravascularly, I must leave undiscussed this phase of the question, referring the reader to works on pathology for a treatment of coagulation in its relation to thrombosis.6 In a general way the time necessary for blood to coagulate may be de- termined by taking a drop either from the finger or the ear and allowing it to fall upon a glass slide. Several drops are collected in this way and tested, at intervals of one minute, by drawing a broom straw lightly through each drop until a thread of fibrin is seen clinging to the straw. The time which 1 Zeitsch. f, Heilkde., Bd. 28, 1907, S. 106. 2 Deutsch. Archiv. f. klin. Med., Bd. 94, 1908, S. 404. 3 Austrian. Bull. Johns Hopk. Hosp., vol. 22, 1911, p. 9; see, also, Matsuo, Deutsch. Arch. f. klin. Med., Bd. 106, 1912, S. 433. 4 Amer. Jour, of Physiol., vol. 29, 1911, p. 187; also, Ther. Gaz., vol. 36, 1912, p. 95. 5 Ann. de 1'inst. Pasteur, Tm. 26, 1912, pp. 657 and 737. 6 See Barratt, Jour, of Path, and Bact., vol. 17, 1913, p. 303. 394 DIAGNOSTIC METHODS. elapses between the taking of the drop and the appearance of the fibrin rep- resents roughly the coagulation time of the blood. Instead of the straw, a white horse hair, which has been previously boiled in alcohol and ether, may be used, as advised by Vierordt. Wright's Method Wright1 has devised an instrument called the coagulometer (see cut). This apparatus consists of a reservoir containing a removable rack holding a thermometer and 12 calibrated and numbered capillary tubes. The determina- tions are made at 18.50 C. This method was formerly much used but, at Fig. 121.-Wright's coagulometer. (Da Costa.) present, is rarely employed clinically, so that the writer refers elsewhere for the technic. Rudolf's Method. This2 is a modification of the older one of Sabrazes3 and applies some of the points of the method of Kottmann.4 It is so simple and clinically reliable that the writer advocates it for general use. Thin glass tubes 1.5 mm. in diameter and about 18 cm. (7 inches) in length are used. A pint Thermos bottle is 1 The Lancet, vol. 2, 1902, p. 15. 2 Amer. Jour, of Med. Sci., vol. 140, 1910, p. 807; vol. 142, 1911, p. 481. 3 Folia Haemat., Bd. 1, 1904, S. 394. 4 Zeitsch. f. klin. Med., Bd. 69, 1910, S. 415. THE BLOOD. 395 employed, its ordinary cork being replaced by a triply perforated rubber stopper. In these perforations rest two brass tubes 7 inches long and just large enough in caliber to hold easily the glass blood-tubes. The third perforation contains a thermometer. The apparatus rests on its side in such a way as to prevent its rolling about. The bottle is filled with water at 200 C. and the stopper is inserted. The finger or lobe of the ear is punctured, the exact time of puncture being noted. Two glass tubes are partially filled from the same drop, the blood being made to run nearly to the far end of each tube, which is placed, as filled, in a brass tube of the thermostat. The protruding ends, the ones at which the blood entered the tubes, are then sealed with a spirit lamp. In about five minutes the first tube filled is drawn out of its holder by the left hand (covered with a glove to diminish the effect of heat of the fingers), is touched with a sharp file, broken across and the broken ends slowly separated. The tube is at once replaced in the thermostat. This technic is carried out with the first tube at intervals of 15 to 30 seconds, until a thread of fibrin appears between the broken ends. The time between the puncture and the appearance of the fibrin is the "coagulation time" of the blood. Tube No. 2 may then be used as a control. Normally the coagulation time averages eight and one-half minutes by this method. Method of Boggs. This is a modification of the older method of Russell and Brodie1 and gives a lower coagulation time than does the preceding method. It is based on the fact that the corpuscles, set in motion by a current of air in a moist chamber, move freely and independently of one another at first but, as coagulation pro- ceeds, clumping of the cells occurs, and finally an elastic radial motion of these cells obtains. This is observed under the low power of the microscope. The writer refers elsewhere for the technic (see cut). Normally, the coagulation time, as evidenced by the formation of fibrin, is between two and nine minutes.2 Anything above nine minutes means delayed coagulation. The importance of testing this function of the blood is observed more particularly in cases of suspected hemorrhagic diathesis, in which the period of coagulation is remarkably increased. In some cases of hemophilia it requires fifty minutes, while in certain of the purpuras from io to 15 or more (Emerson). Likewise, it is customary to test this factor in cases of long-standing jaundice, in which surgical intervention for obstructive lesions of the biliary passages is to be undertaken, as here also the time is much increased. Moreover, we find the coagulation time delayed or imperfect in cases of hemoglobinemia, asphyxia, and general dropsy. Poisoning from the bite of certain serpents, such as the cobra, is characterized by a greatly delayed coagulation of the blood as well as by hemolysis. On the other hand, we notice a quicker coagulation in conditions associated with stasis, repeated 1 Jour, of Physiol., vol. 21, 1897, p. 403. 2 Carpenter and Gittings (Amer. Jour. Dis. of Child., vol. 5, 1913, p. 1) show that the aver- age time for children is slightly longer than for adults. 396 DIAGNOSTIC METHODS. hemorrhages, transfusion, hunger, and also under the therapeutic use of calcium chlorid and of gelatin. In this connection see Cohen.1 The formation of fibrin, as the end-product of coagulation, is increased in some cases and diminished in others. An increase in the amount of fibrin in the blood (hyperinosis) is observed in acute inflammatory processes and in most infectious diseases. Hayem states that the density of the fibrin network, observed when blood is allowed to dry in thick smears on a glass slide, indicates the degree of resisting power of the individual against disease. The largest amounts of fibrin are observed in such conditions as pneumonia2 and acute Fig. 122.-Boggs' coagulometer: a, moist chamber; b, tip of tube through which air passes; c, cover which fits over moist chamber and which holds glass cone; d, pin-hole for escape of air; e f, cross section of cover c; g, tip of glass cone upon which is placed the drop of blood. articular rheumatism, but are seen to a lesser extent in parenchymatous inflam- mation, in inflammations of the mucous membranes and the skin, in the febrile stages of chronic suppurations, in hepatitis, influenza, diphtheria, acute gout, and erysipelas. A decrease in the amount of fibrin (hypinosis) may be observed in malignant growths, malarial fever, pernicious anemia, leukemia, and purpura. In parenchymatous nephritis the amount of fibrin is but slightly if at all increased, while in interstitial nephritis the increase may be notable (Da Costa). (G). Osmotic Pressure and Cryoscopy. The cryoscopic examination of the blood may be of some importance in the diagnosis of certain conditions, particularly in reference to the sufficiency or insufficiency of the kidneys as regards the elimination of the urinary solids. While this subject has not yielded as much information as was expected of it, a brief discussion seems essential. For a theoretical discussion of osmotic pressure and cryoscopy, I must refer the reader to works on physical chemistry. By cryoscopy from the Greek kryos, frost, and skopeo, to see, is meant the 1 Arch, of Int. Med., vol. 8, 1911, pp. 684 and 820. 2 Dochez (Jour, of Exp. Med., vol. 16, 1912, p. 693) finds the coagulation time lengthened in the acute stage of pneumonia. THE BLOOD. 397 determination of the freezing-point of a solution and the referring of this figure to the freezing-point of the solvent, which is regarded as o. Substances in solution lower the freezing-point of the solvent in direct proportion to their molecular (dissociated or nondissociated) concentration. In determining the osmotic pressure of a solution, and cryoscopy is one of such methods, it is important to remember that the proteins have practically no osmotic value. We have, therefore, in this cryoscopic method a means of ascertaining directly the molecular concentration of the body fluids. The determination of the freezing-point of the blood is best made by means of the Beckmann ap- paratus, which may be found in works on physical chemistry. As the osmotic pressure of the serum is equal to that of the plasma or of the whole blood, the serum is generally used for this determination. Withdraw from a vein, preferably the median basilic, about 30 c.c. of blood and allow it to coagulate in a clean closed vessel. Place the serum in the freezing tube of the Beckmann apparatus, adjust the ther- mometer and stirring rods, and proceed as directed elsewhere. The freezing-point of the blood is usually desig- nated as the small Greek delta (3), while that of the urine is given the sign △. Normally, blood freezes at-0.56° C., distilled water freezing at o. This point is subject to more or less physiologic variation depending upon the changes which occur in the blood after meals, exercise, baths, etc. This physiologic variation is transient and slight because the kidneys soon regulate the osmotic pressure (molecular con- centration) of the blood by withdrawal of the consti- tuents which have caused the abnormal tension. This close relationship between the kidneys and the blood is of great importance in the study of renal insufficiency, as many products of metabolic activity remain in the blood in cases of renal disturbance. This is more particularly true of the inorganic con- stitutents of the blood, as the organic elements do not materially affect its osmotic proportions. It is in those cases of renal insufficiency with a tendency to uremia that the most is to be expected from the cryoscopic examination of the blood, yet we find that Schoenborn,1 in the study of 88 cases, observed practically normal figures for the cryoscopic point of the blood. Engelmann,2 on the other hand, reported Fig. 123.-B eckmann apparatus. (Long.) 1 Wiesbaden, 1904. 2Mitteil, a, d. Grenzg. d. Med. u. Chir., Bd. 12, 1903, p. 396. 398 DIAGNOSTIC METHODS. a series of 36 cases in which the freezing-point averaged-0.664 C. It is true that in certain surgical conditions of the kidneys a parallelism is noticed be- tween the lowering of the freezing-point and the development of the uremic symptoms. In cyanotic conditions of any origin whatever we find the cryo- scopic point lower than the normal, owing to the fact that the CO2 is increased in amount in such cases. If the C02 be driven off, the freezing-point becomes normal. Hence we see why, in certain cases of uncompensated cardiac disease, the freezing-point has frequently been reported much lower than the normal standard. There is practically no disease of the blood itself about which a cryoscopic study will furnish any information. From these considerations, as well as from many others, which I have not enumerated, the study of the cryoscopy of the blood does not seem to the writer of sufficient importance to warrant its adoption as a routine method of blood examination. It is, certainly, of no value whatever to the general practitioner who does not have access to a thoroughly equipped laboratory. The time consumed in such examinations would be much better spent in a more thorough study of the fresh and of the stained specimens. While cryoscopy may give some idea of the osmotic activity of the kidneys, it adds to blood examinations a slightly enlarged laboratory record, whose interpretation is a matter of more or less difficulty and whose results rarely ever give any information, beyond that which may be more easily and more readily learned by other methods. It is a purely theoretical method of follow- ing the metabolic activity of the system arid reveals no signs of an oncoming uremia nor does it show us anything regarding the results of therapeutic measures. Naturally, variations in the osmotic pressure of the blood are of importance in the study of conditions associated with effusions into the various serous cavities, as such exudations are markedly influenced by the molecular con- centration of the blood. For a discussion of such phases of the subject I must refer the reader to works on pathology. (H). Electric Conductivity. In the effort to enlarge the scope of blood examinations, the clinician and, more particularly, the laboratory worker have taken advantage of everything offered for the furtherance of their aim. In so doing they have often over- stepped themselves and complicated the examinations by useless additions to their technic. By electric conductivity is meant the reciprocal of the resistance, which a certain amount of solution offers to the passage of an electric current of known strength between two platinum electrodes of given size and a given distance apart. This method is merely a measure of the number of electrolytes (both dissociated and nondissociated) in solution and is not affected by the nonelectrolytic organic substances of the blood, although, according to Hardy,1 xProc. Royal Soc., vol. 66, 1900, p. 95. THE BLOOD. 399 the proteins move to one or the other pole of the battery depending on the reaction of the fluid. The method usually adopted for estimating this factor is that of Kohlrausch, the resistance being balanced on a Wheatstone bridge against a rheostat, the point of equilibrium being determined by means of a telephone attachment. As this determination can give us no information regarding the retention of organic products in the system, it is of comparatively little value in the study of metabolic or of blood diseases. It is in those conditions which are associated with retention of inorganic constituents that this method should be of value, and yet we find, according to the work of various authors, that in advanced nephritis, in which the retention of chlorids has been assumed, the conductivity of the serum is not increased to any extent, although the freez- ing-point is lower than the normal standards. This proves, its seems to the writer, that in such cases the substances capable of increasing the conductivity- that is, the chlorids-are either not retained in the blood or are in such a com- bination as not to influence the conductivity of this tissue. We must, therefore, assume that the lowering of the freezing-point observed in such conditions is due more to the organic than to the inorganic constituents of the blood. As this determination is, in the writer's opinion, of so little practical value in blood work, a detailed description of the methods and results of observations will not be taken up. (6). Chemical Properties. The chemical examination of the blood resolves itself into a consideration of the composition of the whole blood, of the plasma, of the serum, and of the various cellular elements. It is evident that the composition of the whole blood will depend upon the relations in which the single constituents of the blood stand to one another.1 Naturally, physiologic influences are of great importance in the consideration of the composition of the fluid portion, while the cellular elements are less affected by these changes. In view of the recent work on opsonins and immunity, we must conclude that the chemical examina- tion of the fluid elements of the blood may yield much important information as soon as proper methods of research are evolved. Normally, the composition of the cellular elements remains almost con- stant and is but slightly affected by the varying composition of the fluid portions. As is well known, the corpuscles act more or less like semipermeable mem- branes, although Kahlenberg would have us believe that no such a condition is possible. At any rate, the membrane surrounding the erythrocyte allows the passage of certain inorganic and organic constituents of the plasma into the cells and permits the back-passage of certain constituents of the cells into the plasma. This question is intimately related to the subject of hemolysis, which will be discussed in detail later. It is true that a many-sided exchange exists between the red cells and the hypertonic plasma in which they float, such an exchange accounting very often for the alterations in the volume 1 See Bonninger, Zeitsch. f. exp. Path. u. Ther., Bd. u, 1912, S. 1. 400 DIAGNOSTIC METHODS. of the corpuscles and frequently being the cause of abnormally developed cells. This exchange is subject to certain restrictions. Thus we find that the corpuscles show a very high content in potassium salts and a very low sodium content, while the reverse conditions obtain in the serum. If free exchange took place between the cells and serum no such conditions could exist. It is known that potassium salts are closely associated with phenomena of growth and hence it might be possible to prove that cells, which are mor- phologically the older, show a less potassium content than do the newly develop- ing ones. It is evident that a quantitative analysis of the blood can be of only com- parative value as far as the blood as a whole is concerned. We should ascertain on one side the relationship of the plasma and blood-corpuscles to each other, and on the other side the composition of each of these two chief constituents. As there are many difficulties in the way of such determinations, we will not go into detail regarding the chemical composition of the different portions of the blood, but will discuss the blood as a whole. According to C. Schmidt, the composition of the blood is as outlined in the following table. This table, while giving the composition of the whole blood, may readily serve as one from which the composition of the corpuscles and serum may be obtained by reducing the constituents to parts per 1,000 of either corpuscles or serum. It is beyond the scope of this work to discuss the chemical properties of the blood or of its constituents in detail, but certain points which have clinical interest must be taken up. The large excess of chlorin in the serum of man as compared with that of woman and the excess of phosphoric acid in the serum of woman as com- pared with that of man is noteworthy. These variations may later be shown to have a great influence upon sexual differences as regards metabolism. Man's. - Woman's. Corpuscles • • • 5x3-°2 396-24 Water 349-690 272.560 Solids 163.330 123.680 Hemoglobin, proteins, and other organic bodies . 159-59° 120.130 Inorganic bodies. . . . 3-740 3-55° K20 1.586 1.412 Na2O 0.241 0.648 CaO MgO 0.320 0.485 Fe2O3 J Cl 0.898 0.362 P3OS 0.695 0.643 Fe2O3 THE BLOOD. 401 Man's Woman's. Serum 486.98 603.76 Water 439-O2° 55I-99O Solids 47.960 5i-77o Protein and other organic bodies 43.820 46.700 Inorganic bodies 4.140 5-O7O K2O o-i53 0.200 Na2O % 1.661 1.916 CaO MgO J- 0-533 0.608 Cl 1.722 0.144 p2o5 0.071 2.202 (A). Total Solids. The determination of the total solids and, hence, of the water content of the blood is often of importance, especially in cases of anemia. The method is as follows: Allow about 1 c.c. of blood to flow onto one of two previously weighed matched watch-glasses. Cover this with its mate and weigh them while in the moist condition. Separate the glasses and place them in a des- iccator over CaCl2 or H2SO4 for twenty-four hours, at the end of which time weigh them as before. The blood should have dried to a hard glassy mass in this time. The loss of weight will represent the water of the blood taken, from which amount the percentage may be calculated. Gumprecht and Stintzing and Biernacki advise the use of higher than room temperature, the former using 67° C. in an incubator, while the latter employs a heat of 100 to 1200 C. Drying by these latter methods is often associated with loss; hence the writer is accustomed to use room temperature (about 200 C.). Working in this way, many have found the dry residue (total solids) of the blood to be from 21 to 22.5 per cent., the water content being 77.5 to 79 per cent. This figure shows marked variation under the influence of such processes as diarrhea, excessive sweating, and exudation into serous cavities. In anemia the solids are much reduced, while in leukemia they are increased. (5). Blood Pigments. Hemoglobin. Ine normal color oi the blood is due to the presence ot the pigments, hemoglobin and oxyhemoglobin. Usually the former is not large in amount, while the latter is the predominant factor in the color. In blood after asphyxia we find a mixture of hemoglobin, pseudohemoglobin, and parhemoglobin; in arterial blood large amounts of oxyhemoglobin, and in venous blood a mixture of hemoglobin and oxyhemoglobin. Hemoglobin belongs to the class of bodies known as "chromoproteins'' and, owing to its power of combining with various gases and thus aiding in the gaseous exchange of the body, may be styled a "respiratory protein." It is easily decomposed into a protein, globin (about 96 per cent.), and a pig- 402 DIAGNOSTIC METHODS. ment, hemochromogen (4 per cent.), which contains iron and is easily changed in the presence of oxygen into hematin. The iron content of hemoglobin is the portion which enables this pigment to exert its peculiar vital power of oxygen transference. The amount of hemoglobin in normal blood is variable, depending on the age of the subject examined. Normally, 100 c.c. of adult human blood contain from 13 to 14 grams of hemoglobin (the average being about 13.77), which amount is not a constant factor at all ages. From the following table of Leichtenstern it will be seen that a definite age curve exists for this substance. Age. Grams per 100 c.c. of blood. i to 4 days, 19.329 to 21.160 8 to 14 days, 17.869 to 16.124 8 to 20 weeks, 15.362 to 12.928 6 months to 5 years, 10.971 to 11.373 5 to 15 years, 11.151 to 11.796 15 to 25 years, 13.034 to 13.870 25 to 45 years, 14.727 to 15.013 45 to 60 years, 12.484 to 13.150 Over 60 years, 14.79° Under certain conditions the hemoglobin is dissolved out from the red cells, leaving only their stromata behind. This gives rise to the condition known as hemoglobinemia which will be discussed in a later section. Accord- ing as hemoglobin is free or combined with certain gases, we have in the blood various derivatives of this pigment, each one of which has certain characteristics and a definite clinical importance. These derivatives are recognized by Fig. 124.-Direct-vision spectroscope. qualitative tests, especially by spectroscopic methods, and will be presented here; while the important tests used for the recognition of hemoglobin, either in the fresh state or as dried stains, will be taken up under the medicolegal discussion. Hemoglobin, also called reduced hemoglobin, is much more soluble than oxyhemoglobin, its solution in water being more violet or purplish than one of oxyhemoglobin of the same concentration. Such solutions of hemoglobin absorb the blue and violet rays of the spectrum to a less marked degree than do those of oxyhemoglobin, but they strongly absorb the rays lying between C and D. In proper dilution a solution of hemoglobin shows a spectrum with one broad not clearly defined band between D and E, lying toward the red end of the spectrum a little over the Fraunhofer line D (See plate). Absorption Spectra. PLATE XIV. Oxyhaemogh Haemoglobin. Carboxy- haemoglobin. Neutral Met- haemoglobin. Alkaline Met haemoglobin. Alkali Haematln. (from HA WKS "PHrS/OLOG/GAL CHFMIS TRK 7 Absorption Spectra. PLATE XV. Reduced Alkali Haematin or Haemochromogen. Acid Haematin in ethereal solution. Acid Haemato- porphyrln. Alkaline Haematopor phyrin. Urobilin or Hydro bilirubin in acid solution. Urobilin or Hydro- bilirubin -in alkaline solution after the addition of zinc chloride solution. Bilicyanin or Cholecyanin in alkaline solution. foOM HAWKS "PHKS/OLOG/CAL CHEMISTRY") THE BLOOD. 403 Pseudohemoglobin. According to Ludwig and Siegfried, blood, reduced by hyposulphites or by a stream of hydrogen to such an extent that the spectrum of oxyhemoglobin disappears, yields large amounts of oxygen when exposed to a vacuum. This loose combination of hemoglobin and oxygen, which gives the spectrum of hemoglobin,, is called pseudohemoglobin. Hammarsten considers it an in- termediate body between hemoglobin and oxyhemoglobin. Oxyhemoglobin. Oxyhemoglobin, also called hematoglobulin, is a molecular combination of hemoglobin and oxygen. The ability of hemoglobin to take up oxygen is a function of its iron content. When this factor is calculated as 0.33-0.40 per cent., 1 atom of iron in the hemoglobin molecule corresponds to two atoms (one molecule) of oxygen. This combination is a loose one and hence the quantity taken up will depend upon the partial pressure of the oxygen. This oxygen is set free when the oxygen pressure is reduced, thus giving rise to the characteristic property of oxygen transference. As Pfluger has shown, oxy- hemoglobin may, when it is gradually oxidized, act as an "ozone exciter" by the decomposition of neutral oxygen into the atomic form. It may also act as an "ozone transmitter" as in certain tests to be outlined later. A dilute solution of oxyhemoglobin or of arterial blood shows a spectrum with two absorption bands between the Fraunhofer lines D and E. One band, a, is narrower, but darker and sharper and lies on the line D, the other, ft, is broader, less defined and less dark and lies at E. As the dilution becomes weaker the band first disappears. By increased concentration the two bands become broader, the space between them smaller, and the blue and violet parts of the spectrum darkened. Other substances may give this same absorption spectrum, but oxyhemoglobin may be differentiated by its be- havior toward reducing agents, such as ammonium sulphid or Stokes' solution of ammoniacal ferrotartrate (see plate). Methemoglobin. This pigment is closely related to oxyhemoglobin, as it contains the same amount of oxygen and is isomeric with it. The oxygen is, however, not in loose combination and cannot, therefore, be utilized by the system. This coloring matter is formed by the spontaneous decomposition of blood, as observed in hemorrhagic transudates and cystic fluids, and occurs also in cases of poisoning with potassium permanganate, potassium ferricyanid, chlorates, nitrites, nitrobenzol, acetanilid, antipyrin, turpentine, sulphonal, and arsenic, and in cases of cyanosis with diarrhea. According to Jaderholm and Bertin-Sans, the absorption spectrum of methemoglobin, in aqueous or slightly acidified solution, is similar to that of acid hematin (see below), but is easily distinguished from the latter by the readiness with which it turns into that of hemoglobin on treatment with alkali and a reducing substance. Hematin, under the same conditions, gives the 404 DIAGNOSTIC METHODS. spectrum of an alkaline solution of hemochromogen (Hammarsten). Met- hemoglobin, in alkaline solution, shows two absorption bands, which are like those of oxyhemoglobin, but differ from them in that the band nearer E is stronger than the one at D. A third fainter band may be observed, according to Hammarsten, lying between C and D. Carbon-monoxid Hemoglobin. This pigment is a molecular combination of one molecule of hemoglobin with one molecule of CO. This combination is stronger than those of hemo- globin and oxygen. The oxygen of hemoglobin is easily replaced, therefore, by CO and, in consequence, the tissues suffer for want of oxygen. This pig- ment imparts to the blood a bright cherry-red color both in the venous and arterial circulation. The most common cause of formation of this pigment is the inhalation of coal gas or of illuminating gas. The characteristic color in these cases may disappear after a few hours or it may persist for days, depending on the severity of the case, being frequently found in the blood after death. Such blood shows an absorption spectrum similar to that of oxyhemoglobin, but the two bands lie more toward the violet end of the spectrum than in the case of oxy- hemoglobin. On the addition of a reducing agent to blood showing this spectrum, the absorption bands of carbon-monoxid hemoglobin are unaffected, while those of oxyhemoglobin are changed to those of hemoglobin. This spectroscopic test is not delicate as less than 20 per cent, of carbon monoxid hemoglobin in the blood is difficultly, if at all, detected. Carbon-dioxid Hemoglobin. According to Bohr, hemoglobin forms three molecular combinations with CO2, in which products, a, P, and y, one gram of hemoglobin combines with 1.5, 3, and 6 c.c. of CO2. The spectrum of these compounds is similar to that of reduced hemoglobin. If a large excess of CO2 be present the hemoglo- bin is decomposed and globin is precipitated. The absorption band under these conditions is probably referable to the presence of free hemochromogen. Sulph-hemoglobin. This pigment, formed by the action of hydrogen sulphid upon hemoglobin, is found in the corpuscles and not free in the plasma. Sulph-hemoglobinemia has been reported by several, especially by van der Bergh1 whose name is given to the condition known as "idiopathic enterogenous cyanosis" and character- ized by cyanosis, headache and marked constipation. Clarke2 has reported the first case in the United States. Spectroscopic examination of the diluted blood is essential to show the presence of this pigment. The spectrum shows 3 absorption bands, two similar to those of oxy-hemoglobin and a third in the red not so near the C line as is the third band of methemoglobin. Addition of dilute ammonium sulphid destroys the third methemoglobin band, while it does not effect that due to sulph-hemoglobin. 1 Deutsch. Arch. f. klin. Med., Bd. 83, 1905, S. 86; Berl. klin. Woch., Bd. 43, 1906, S. 7. 2 Medical Record, vol. 76, 1909, p. 143; vol. 78, 1910, p. 987. THE BLOOD. 405 Decomposition Products of Blood Pigments. As stated previously, hemoglobin and oxyhemoglobin are proteins, which are converted under the action of different physical and chemical means into a globulin-like protein, globin, and an iron-containing pigment, hemochromo- gen. From the table below, adapted from Webster and Koch,1 may be seen the relations of hemoglobin to its various derivatives and also to the bile pigments and to chlorophyll. Hematin. This decomposition product is found in any situation in which oxyhemo- globin is destroyed; thus, in the digestive tract where it is formed by the action of gastric or pancreatic juice on oxyhemoglobin, in old extravasations, in the stools after hemorrhage, in urine after poisoning with arsenical compounds, and in the blood of persons poisoned with nitrobenzol and acetanilid.2 Hema- tin is a dark brown or blue-black amorphous powder, insoluble in water, dilute acids, alcohol, ether, and chloroform, but readily soluble in acidified alcohol or ether and in dilute alkalies. Arnold has shown that there are three modifications of this pigment, a neutral, an alkaline, and an acid hematin, each showing spectroscopic differ- 1 Laboratory Manual of Physiological Chemistry, Chicago, 1903 2 Schumm (Zeitsch. f. physiol. Chern., Bd. 80, 1912, S. 1) reports hematinemia in a case of acute chromium poisoning. 406 DIAGNOSTIC METHODS. ences. The neutral hematin has little importance clinically and will be passed in this discussion. Acid-hematin solutions show four absorption bands, one between C and D, a second broad but not clearly defined band between D and F, which divides under certain conditions into two narrower bands, a fourth band between D and E which is nearer D and is very weak. Usually only the band between C and D and the broad band between D and F can be seen (see plate). Solutions of alkaline hematin show one absorption band between C and D which reaches out to some extent between D and E. If the alkaline hema- tin solution be reduced with ammonium sulphid, the spectrum of hemo- chromogen is observed, which shows two characteristic absorption bands, one very sharp and dark between D and E and a second paler and broader band covering the E line. Hematin forms very characteristic compounds with hydrochloric, hydro- bromic, or hydriodic acids. With HC1 hematin crystallizes with one mole- cule to form the compound hemin. The crystals are light or dark brown rhombic forms and are called, after their discoverer, Teichmann's crystals. Their formation is specific for blood, but the kind of blood cannot be deter- mined by their presence. As their detection is largely a medicolegal question, I will refer a discussion of the technic to the later section. When hematin is treated with concentrated sulphuric acid in presence of air, iron is split off leaving the pigment hematoporphyrin. If air be excluded the product yielded by such treatment is hematolin. This pigment is insoluble in water, but dissolves in alcohol, strong and weak alkalies, and in acids. It is isomeric with bilirubin, with which it is associated in the liver cells. In acid solution hematoporphyrin shows two absorption bands, one fainter and narrower between C and D and nearer D, the other darker, sharper, and broader in the middle between D and E. In dilute alkaline solutions this pigment shows four absorption bands. A band between C and D, a second broader band surrounding D with the broadest part between D and E, a third between D and E nearly at E, and a fourth broad and dark band between E and F. Hematoporphyrin is of great import- ance from the medicolegal standpoint as certain suspected stains may be identified only by its spectrum. For this phase see the later sections. Hematoporphyrin. Hematoidin. This ruby-red or reddish-yellow pigment is derived from blood coloring matter and like hematoporphyrin is iron free. It is found in old blood-clots, in hemorrhagic exudates, in sputum, and in feces. It is more abundant when the blood pigment is not much exposed to the action of living cells, as in the center of large extravasations and in hemorrhages into preformed cavities of the body (Ziegler). Hematoidin is identical with bilirubin and shows no absorption bands, but only a strong absorption of the violet to the green portion of the spectrum. THE BLOOD. 407 Hemosiderin. This yellow, orange, or brown pigment is a derivative of hemoglobin and contains iron. Unlike hematoidin, it is found more particularly in extrava- sated blood which has been subjected to the action of living cells. After a time this pigment changes into one (probably hematoidin) which contains no iron. Malarial Pigment. The older view of Ewing1 and others that this yellowish-brown or black pigment was melanin seems to have been disproven by the recent work of Brown2 who has shown that it is, probably, identical with hematin and is formed by the action of the malarial parasite (possibly of a proteolytic enzyme) upon hemoglobin. On spectroscopic examination of an alkaline alcoholic solution of this pigment, a single broad band is seen starting sharply at D and extending to the left, gradually shading into the red between C and D. The melanins, observed in the leucocytes in relapsing fever, melanotic sarcoma and Addison's disease, are quite distinct from this malarial pigment as shown by decolorizing and solubility reactions. A very large number of methods have been introduced for the deter- mination of the blood coloring matter. Some of these are extremely accurate, but are too complicated for general clinical purposes; others are less accurate, but are more applicable to our work. It may be said that, as a rule, the clinical instruments at our command will give comparative figures, providing their construction and their standardization is accurate. However, we find so much inaccuracy or even lack of proper standardization in some of these instruments, that no absolute comparison may be attempted between the results obtained by various workers using the same or different methods. Normal adult blood contains about 13.77 grams of hemoglobin per 100 c.c. of blood, this figure being subject, however, to variations at different periods of life, so that those instruments which are standardized against normal adult blood and which furnish the amount of hemoglobin in terms of per- centage of such blood cannot be absolutely accurate no matter how perfectly constructed or how accurately standardized, when they are used for the esti- mation of hemoglobin in the blood of a child or of an elderly patient. It is, therefore, much better practice to obtain the values in terms of the actual amount per 100 c.c. of blood. It is well to recognize, as Turk points out, that we are working with possible errors and can obtain comparative results only when we use the same exactitude and care in each of our estimations. The errors are essentially constant, but we must remember that the arbitrary color standards change as time goes on, necessitating a restandardization of our instruments if accurate results are to follow. The writer has seen errors of 50 per cent, arise from the use of an old von Fleischl instrument and one of 25 per cent, from a faulty standardization of Sahli's solution. Estimation of Hemoglobin. 1 Jour. Exp. Med., vol. 6, 1905, p. 119. 2 Arch, of Int. Med., vol. 13, 1911, p. 290. 408 DIAGNOSTIC METHODS. It must be insisted upon that we must not assume that a patient, who is pale and "anemic looking," is in reality anemic. Pallor depends not only upon the amount of pigment in the blood but also upon the delicacy and transparency of the skin and upon the superficial distribution and size of the blood-vessels. By remembering these points we may often save ourselves unnecessary chagrin on finding that the results of our determinations do not accord with the "anemic " expression of the patient. A simple puncture of the ear, allowing the blood to drop upon a clean linen towel (so-called "towel test"), will many times set us right and prevent a diagnosis of anemia without further examination of the blood. Such examinationless diagnoses are unwarranted and inexcusable. Direct Methods of Estimation. These methods are, owing to their complexity, not applicable to clinical work. The spectrophotometer of Hiifner is undoubtedly the very best and most accurate method, but I must refer to the original article for its description. Likewise the colorimetric double pipet of Hoppe-Seyler and the methods of Nebelthau and of Zangemeister must be neglected. Indirect Colorimetric Methods. These indirect methods employ a comparison of the blood solution with a second medium which approximates the blood in color. Two principles are possible in such methods. We may either have a fixed unchangeable medium of comparison, with which the blood is matched by constant dilution, or we may use a fixed blood solution of known strength and compare there- with the standard graduated color scale. Both of these principles have been used in such determinations, the variations being shown in the fact that instru- ments for this purpose have been introduced by Gowers, Sahli, Hayem, Malas- sez, Henocque, Bizzozero, von Fleischl, Miescher, Haldane, Grtitzner, Gartner, Dare, Oliver, and Tallqvist. It is evident, a priori, that these indirect methods must have greater errors than the direct. Two things are essential for any exactitude whatever. First, the medium of comparison must agree in the most complete manner with the various color tones which the blood shows at different percentage values of hemoglobin. Second, the standardization of the medium of com- parison must be extremely accurate. We must have no illusions regarding the exactness of the hemoglobin values obtained by these methods, as it is very difficult to construct two instruments of the same kind that will agree exactly. Certainly no two instruments of different make will agree, but com- parative results sufficiently accurate for clinical purposes are obtained by the use of the more reliable instruments. The personal equation in reading the color comparisons must be remembered, as some individuals show abnormal sensitiveness or lack of sensitiveness to shadings of red. It is an impossibility in this book to give in detail all of the methods advanced for the estimation of hemoglobin. I select, therefore, those that have proven most reliable in my hands. THE BLOOD 409 Hemometer of Fleischl-Miescher. Up to recent times the most frequently used of the instruments for the estimation of hemoglobin was the old von Fleischl instrument. With the introduction of the Miescher modification, this original form has been or should be less often employed. We avoid, therefore, a discussion of the older instru- ment, referring to other works which have included it. This new apparatus, made by Reichert, under the direction of Miescher, is similar in general appearance to the old von Fleischl. It has the same stand and the same scale principle, although this latter is standardized differently and graduated on a different basis. It differs, materially, in the method of measur- ing and diluting the blood, in the form of the comparison chamber, and in the meaning of the graduation of the scale (see cut and legend for its description). Fig. 125.-Hemometer of Fleischl-Miescher: R, Stage; T, milled head, which moves the color scale; m, opening in stage through which the instrument is read; M, mixing cell; D', cover glass; D, cap; PS, gypsum mirror from which light is reflected; mel, diluting pipet. This is similar in construction to the pipet of the Thoma-Zeiss hemo- cytometer, its calibrations, however, being different. The marks are 1/2, 2/3, and 1. Above and below each of these main divisions are two marks each corresponding to 1 /100 of the contents of the capillary tube. This device enables the worker to measure accurately the column of blood taken, in case he gets too little or too much blood in the tube. The relation of the capillary to the ampulla is such that blood, drawn to the mark 1 and diluted to the mark above the ampulla, receives a dilution of 200; if drawn The Diluting Pipet. 410 DIAGNOSTIC METHODS. to the mark 2/3 the dilution is 300; while the line 1/2 furnishes a dilution of 400. The diluent used is 1/10 per cent, sodium carbonate solution. This dissolves the stromata of the red cells furnishing a clear solution. Occasionally the diluent becomes turbid after standing some time, and should be freshly made and should contain no bicarbonate. The taking of the blood and the mixing with a diluent is done with the same precautions mentioned under the method of making the blood count. In choosing the proper dilution, one should select that which will enable him to use the central portions of the graduated scale. A dilution of 400 is usually applicable except in cases in which marked anemia is suspected when dilutions of 300, 200, or even 100 should be made. In making this last dilu- tion the erythrocytometer may be used. In doing accurate work, it is well to make a preliminary determination of hemoglobin in order to tell, the better, just what dilution would be advisable. This may be done by the Tallqvist method described later. The Comparison Cells. With this instrument two cells for holding the blood and diluent are furnished. One of these has a depth of 15 mm., the other one of 12 mm. the former is the standard cell, the latter the control, giving 4/5 the value of the larger cell. Their external appearance is similar to the cell of the old Fleischl, but their capacity is less Owing to their greater thickness. The dividing partition between the halves of the cells projects about 1/2 mm. above the borders, thus preventing any mixing of the fluids in the two portions. A grooved cover-glass is slid over the compartments without fear of mixing the fluids. If fluid is lost no error is introduced, as the dilution is uniform in the pipet and the depth of the chamber is definite. A diaphragm, with an opening 4 mm. wide, is placed over the cover-glass after this latter has been adjusted, the opening being so placed that its long axis is perpendicular to that of the vertical partition. The field of comparison is thus limited to 4 mm., which corresponds to about 30 of the scale, affording a comparison of a single tint of the scale with the color of the diluted blood. Graduation of the Scale. The slide for comparison of color is tinted with Cassius' golden-purple, as in the old Fleischl instrument. The graduations of this color scale are made by comparison with standard solutions of hemoglobin and not with the arbitrary standard of blood of so-called normal individuals. The scale shows the same divisions as that of the Fleischl instrument, but their significance is different in the Miescher modification. Here one does not read directly the percentage of hemoglobin, but must obtain the corresponding value by reference to a "table of calibrations," which accompanies each instrument. While the figure thus obtained is in terms of actual percentage (grams per 100 c.c.), it is quite different from the percentage figure given by the Fleischl instrument. THE BLOOD. 411 Method. After testing the chamber by placing in one compartment the diluting fluid in order to see that none runs into the other, the blood is drawn into the pipet to the desired point and diluted to the mark, all precautions mentioned under Blood Counting being observed. Thoroughly mix the blood and diluent by shaking and blow out the unmixed contents of the capillary tube. Fill one com- partment of the 15 mm. cell with the diluted blood so that a convex meniscus appears above the border of the chamber, the other compartment being filled with the diluent or with distilled water. Adjust the grooved slide and cap, place the cell in the central opening of the stand, and adjust the light. This latter portion of the technic is of considerable moment, as this instrument gives the best results when used in a dark room illuminated only by a small candle flame which is placed about 18 inches from the stand and to the side. In the absence of a dark room, a light-proof shield for the eyes may serve. This may readily be obtained by the use of a tube of stiff, dark paper, which fits over the comparison cell and shuts off the light of the candle from the field of vision. The observer should stand in such a position that he looks into the cell from the side and not from the front or back, the eyes being about one foot above the cell. In the comparison of the color tones the variations are much better seen by quick movements of the slide rather than by slow gradual changes. The eyes should be rested at short intervals to prevent fatigue and lack of sensitiveness to the different shadings of color. When the color is matched take the reading of the instrument, by observing what mark of the scale coincides with the notch on the edge of the opening above the scale, and control this reading by several duplicates. Remove the blood from the 15 mm. cell by means of the pipet and transfer it to one compartment of the 12 mm. chamber, tested as was the larger one. Adjust this chamber and make readings in the same way. These values should be only four-fifths of those obtained with the 15 mm. cell. This modification has the advantage of using different portions of the color scale and should give comparative figures. If there should be any variation, which should never be over 1 per cent., a correction may be made. This is possible only with the use of the "table of calibrations," which contains the series of scale divisions and the absolute amount of hemoglobin in mg. per 1,000 c.c. of blood, corresponding to each division of the scale when the 15 mm. chamber is used. Thus, if the scale shows 56, we find the value corresponding to this to be 447 mg. of hemoglobin in 1,000 c.c. of diluted blood. As the dilution may have been 400, we would have, in 1,000 c.c. of undiluted blood, 400 x 447 = 178.8 grams. As we wish to know the amount in 100 c.c. we have merely to divide by 10, thus obtaining 17.88 grams of hemoglobin. If we wish to get the percentage figures, as read on the old Fleischl, we divide the figure obtained with the Miescher, in this case 17.88, Calculation of Results. 412 DIAGNOSTIC METHODS. by the amount of hemoglobin corresponding to the 100 division of the scale of the instrument used, in this case 14, and we obtain 127.7 Per cent. This Miescher modification is our very best clinical apparatus for estima- tion of hemoglobin, giving results which are accurate within 0.2 to 0.5 per cent. It is open to the objections that it is bulky, expensive, and requires more time and practice for its use than is at the control of the busy practitioner. For hospital use, however, it is the instrument par excellence and should never be substituted by others. It has one disadvantage that each instrument is standardized, not against a known hemoglobin solution, but against a so- called "normal" instrument, which has been properly standardized. It is to be hoped that each instrument will, in the future, be properly calibrated and thus insure us against the possi- bility of errors arising from any change in the standardization of the "normal" instrument. Hemoglobinometer of Dare. The instrument, introduced by Dare, has the advantage of using un- diluted blood, and avoids any error consequent upon dilution. The prin- ciple of this instrument is as follows: The color of undiluted blood is com- pared by artificial light with that of a graduated glass scale colored with golden-purple, the 100-point of which is standardized against a solution of 13.77 grams of hemoglobin in 100 c.c. of serum. For a description of the apparatus see cut. Fig. 126.-Hemoglobinometer of Dare; R, milled wheel; s, case inclosing the color disc; t, movable wing, which is swung out- ward; rr, telescoping camera; v, aperture admitting light; w, capillary blood pipet; y, detachable candle holder; z, slot through which the percentage of hemoglobin is read. Method. Swing outward the movable screen, which serves as a cover for the case, adjust the camera tube, and fit the candle attachment in its place opposite the camera tube. The candle should be so adjusted that its upper end is flush with the top of the clips which hold it. If the wick be curved, it should be so turned that the intensity of the light is midway between the two apertures. See that the pipet, composed of the rectangular glass plates, is thoroughly cleansed and dry. The space between the plates is filled by applying the edge of the pipet to the side of a fairly large drop of blood. Adjust this pipet in its place and rotate the colored scale, by means of the milled screw, until the colors match. Hold the instrument steady to prevent the flickering of the flame as much as possible. No dark room is necessary, but it is advisable to point the instrument at some dark object and to avoid direct sunlight, as the shadings of color are not so easily matched by direct daylight. As soon as THE BLOOD. 413 the colors are matched make the readings and check the results with several duplicates. This reading is observed on the left side of the case in the small open space, the line which coincides with the beveled edge of the opening representing the percentage of hemoglobin, on the basis of a value of 13.77 grams of hemoglobin per 100 c.c. as 100 per cent. It is, therefore, easy to calculate the direct amount of hemoglobin in the blood examined. This instrument has the advantages that undiluted blood is used, that the scale of comparison is usually very accurately standardized, that it is con- venient, easy of manipulation, and rapid in giving results. Coagulation of the blood does not occur sufficiently soon to introduce an error, providing the reading is taken within a reasonable time. It is more convenient for general Fig. 127.-Method of filling the Dare blood-pipet. (Da Costa.) use than is the Miescher, is less expensive, can be used in a light room, and gives results second only to those of the Miescher. The disadvantages of this instrument are that an occasional faulty standardization may introduce errors, it costs much more than some of the instruments to be described, and it is not a long-lived instrument unless care is taken in handling it. In the writer's laboratory this instrument has given great satisfaction and can be recommended for general use on the ground of its convenience and ready application to clinical work. Hemometer of Sahli. This instrument is a new modification of the older hemoglobinometer of Gowers, and has so many advantages over the older instrument that this latter will be passed over. A modification of the Gowers instrument intro- duced by Haldane is simple and reliable, but has the disadvantage that coal gas is used in converting the hemoglobin into CO hemoglobin, and consequently, is not of easy application for bedside work. As Sahli has pointed out, a colored fluid under examination should not be compared with a different substance similar in color, but with a solution of known strength of the same coloring matter. His modification of the Gowers method employs an instrument constructed on exactly the same lines, but using a different standard of comparison. This standard of comparison 414 DIAGNOSTIC METHODS. is an acid hematin solution in a concentration corresponding to a i per cent, solution of normal blood. This solution is somewhat dark, as it is standardized against blood showing high hemoglobin value. As Sahli states, the standard fluid, as furnished with his hemometer, corresponds to a blood which shows with the Miescher instrument at a dilution of 200 a reading of 109, or an ab- solute quantity of 17.2 grams of hemoglobin per 100 c.c. of blood. The fine particles of hemin, which are in suspension, may adhere to the glass, especially if the instrument lies unused for some time. This process changes the color □f the standard to some extent, so that it is supposed that deterioration has occurred. This may be remedied by completely inverting the tube, without violent shak- ing, thus allowing the precipitate to diffuse uniformly. Occasionally one of these instruments is found to be improperly standardized, but this is rare. Method. This consists in diluting the blood with io times its volume of N/io normal hydrochloric acid. After a few seconds the fluid becomes dark brown from the formation of hematin hydrochlorid (hemin), which substance is not in solution, but in fine suspension. The blood is taken with the 20 cmm. pipet and blown into the graduated tube, which contains N/10 normal HC1 up to the mark 10. This acid may be accurately enough made by diluting 15 c.c. of concen- trated chemically pure HC1 to 1,000 c.c. Shake the mixture of blood and acid and dilute with ordinary water as soon as a clear dark brown color is visible. Add water until the shade of the mixture corresponds exactly with that of the standard solution, when the percentage of hemoglobin may be read off. The water should be added very carefully from a dropping pipet, as the accuracy, of the method depends upon adding the exact amount of fluid necessary. The comparison of colors may be made in any light, as the two solutions, being the same, will be similarly affected. This instrument is very conveniently gotten up, being mounted in a case carrying a white-glass plate to reflect the light to better advantage. It is simple, inexpensive, and is accurate within 2 per cent. The author would recommend this instrument to the general worker above all others, with the possible exception of the Dare, whose advantages have been mentioned. Fig. 128.-Hemometer of Sahli. {Greene) Hemoglobinometer of Oliver. In this method the color of the blood in a definite dilution is compared, by light reflected from a white surface, with a series of tinted glass standards. THE BLOOD. 415 Such a method has the advantage that the color of the diluted blood is com- pared with one single tint at one time. The standard glass disks correspond accurately, as determined by the tintometer, to the specific color curve of Fig. 129.-Hemoglobinometer of Oliver. {Coplin.} progressive dilutions of normal blood. Two sets of standards are furnished, one for use in daylight, the other by candlelight, the latter being preferable. Each set of 12 disks is mounted on plaster of Paris and enclosed in two wooden frames, six disks in each frame. They represent the color of solu- 416 DIAGNOSTIC METHODS. tions of hemoglobin with percentages from io to 120. To obtain interme- diate values, colored riders are used, each representing 2.5 and 5 per cent, of hemoglobin, when used with the disks from 70 to 120, but twice these values if used with the disks from 10 to 60. The blood is taken in a capillary pipet similar to that of von Fleischl, holding 5 cmm. of blood. The diluting chamber has a white background of plaster of Paris and, when filled with distilled water in which one pipetful of blood has been dissolved, yields a solution of 1 per cent. When filled the cell is covered with a blue glass cover. Method The pipet is filled with blood drawn by capillarity from a puncture of the ear or finger. Wash the blood from the pipet into the diluting chamber by means of water from a medicine dropper, fill the chamber with water, stir well with the handle of the pipet, and adjust the cover-glass. A small bubble of air should be included as an evidence that the chamber was not overfilled. Compare the color of this solution with that of the standard disks, using the light of a candle placed 3 or 4 inches away in such a manner that the light strikes the two chambers alike. It is well to use the camera tube to shield the eyes and enable one to compare the tints more accurately. If the color matches that of any of the disks, the percent- age is read directly from the case. If, how- ever, the color is intermediate between two disks, the riders must be superimposed on the disk of lower percentage, and a second com- parison made by using a corresponding plate of unstained glass as a rider on the diluting chamber. The adjustment of tints and riders must be continued until the colors match ex- actly. Naturally the variation will equal two per cent, as the ordinary riders are equivalent to 2 1/2 per cent, of hemoglobin. It is not necessary for this determination that the room be absolutely dark. This method is accurate within the limit mentioned (2.5 to 5 per cent.), but it is trying, time consuming, and does not equal in accuracy the methods of Dare and of Sahli. The instrument is expensive and the disks are prone to deteriorate. Fig. 130.-Tallqvist's hemoglo binometer. Tallqvist has introduced a method for the estimation of hemoglobin which is so simple that certain writers have been led to recommend it above Tallqvist's Hemoglobinometer. THE BLOOD. 417 other methods. The principle of this method is essentially the same as Oliver's method, although the application is entirely different. Tallqvist compares the color of the undiluted blood with that of a series of lithographed standard tints, which range by differences of io° from io to 100 per cent. These stand- ard tints were prepared by matching in water colors the tint of the blood of various patients (whose hemoglobin values had been determined with the von Fleischl instrument) when soaked into standard filter-paper. These tints were then lithographed and bound up with sheets of filter-paper, the combi- nation making a very simple and convenient book which may be easily carried in the pocket. Allow a fairly good-sized drop of blood to soak into a portion of the filter- paper by holding the edge of the sheet against the drop. Care should be taken to allow this to take place very slowly so that the color may be uniform through- out. If carefully done it 'will not be necessary to blot the stain, but some- times this is essential. As soon as the stain has lost its humid gloss compare its color with that of the lithographed scale. Do not allow the stain to become dry, as the color comparisons are disturbed by the coagulation. Hold the scale and the stained paper in such a way that the light (daylight only) is well reflected from the color scale. The percentage of hemoglobin is then read off directly by noticing the point of the scale with which the blood stain ex- actly corresponds. As this scale does not read closer than io per cent., the intermediate percentages must be judged by difference. Here personal factors are of great importance, some workers being so skilled as to detect small variations. This method can furnish, at best, only an approximate result and has nothing in common with the other methods discussed. It is to be advised only when some more reliable method is not at hand or when a rough estimate only is wanted. It is recommended by some writers as being as generally useful and accurate as any of the other methods, but the writer can see no advantage whatever in its use, as the results obtainable are, in his opinion, not reliable and are not as satisfactory to one who is not especially accustomed to such color comparisons. This test would better be limited to rough, pre- liminary, approximate determinations than to be used in the more careful estimations which clinical work requires. In the selection of a method for estimating the hemoglobin of the blood, the writer would advise the Fleischl-Miescher instrument for those who are in close touch with hospital or clinical laboratory facilities. For the general practi- tioner who must make his own determinations under varied circumstances, the Dare or the Sahli instrument may be unequivocally recommended, the latter having the advantage of cheapness along with accuracy and ease of manipulation. Method. 418 DIAGNOSTIC METHODS. Variations in the Amount of Hemoglobin. The percentage values are misleading, as the hemoglobin varies with age and, to some extent, with sex. For this reason it is preferable to obtain the absolute amount of hemoglobin, which is done directly by the Miescher and, by a very simple calculation, with the Dare or Sahli instruments. In those methods which read in direct percentages one may readily calculate the absolute values per 100 c.c. of blood by multiplying the percentages obtained by 0.1377. In estimating the true figures for the hemoglobin in the blood of women, it is necessary to add one-eighth to the percentage values as read, as female blood runs from 12 1/2 to 20 per cent, less in hemoglobin than does the blood of man; likewise for a child we should add one-seventh to the per- centage. In this way we correct the readings of the instrument which is calibrated against adult male blood. It has been found by comparative tests, that the blood of rural residents reaches the normal standards much more frequently than does that of their city brothers, this being due, no doubt, to the purer fresh air which the country dweller enjoys. A relative diminution in the amount of hemoglobin, as determined by the above methods, is known as oligochromemia or as achroiocythemia. This condition is usually associated with a decrease in the number of erythrocytes (oligocythemia), but in cases of chlorosis the diminution of hemoglobin is an absolute reduction, each cell showing less hemoglobin than normally and no oligocythemia being, as a rule, present. In pernicious anemia, on the other hand, each cell may show an absolute increase in hemoglobin, although the percentage value is reduced owing to the marked oliogocythemia present at the same time. Factors bringing about conditions of hydremia lead to a transient oligochromemia, while factors acting as etiologic units in the pro- duction of anhydremia lead to a reverse condition of polychromemia. Oligochromemia is observed in chlorosis, pernicious anemia, leukemia, and secondary anemias, following chronic infections, hemorrhage, malignant growths, and some constitutional diseases. It is noted in chronic nephritis, chronic enteritis, and mineral poisonings, especially those with lead and mercury compounds. It has been stated that low hemoglobin values some- times observed in cases which are to have surgical intervention are contra- indications to operative treatment as the anesthetics themselves may bring on a condition of oligochromemia. While it is wise to watch with extreme care cases showing less than 50 per cent, hemoglobin, yet surgical operations have been successfully performed on cases with a more marked diminution of the hemoglobin. Color-index. This term is used to indicate the amount of hemoglobin contained in each cell, as compared with the amount present when a normal number of erythro- cytes obtains. In other words, it is the quotient of the hemoglobin percentage divided by the percentage of red cells. This latter factor is obtained by dividing THE BLOOD. 419 the number of red cells, as found by the count, by the number reckoned as normal, namely, 5,000,000. A very simple method of getting this percentage is to multiply the number of hundreds of thousands of red cells by two; thus if 2,650,000 reds were counted we obtain 53 as the percentage of red cells. Sahli recommends the use of the term "hemoglobin quotient" or "hemoglobin value" for this factor, but the expression color-index has become so general that it will be hard to replace. Moreover, the latter term seems to convey a more definite idea to most of us than would the substitutes suggested. The color-index is normally one; that is, a hemoglobin value of 100 per cent, is associated with a blood count of 5,000,000 red cells. We find, how- ever, in the different anemias that this figure varies markedly. In those anemias, in which the reds are diminished to the same relative extent as is the hemoglobin, the index remains normal; while in those cases in which the hemoglobin is markedly reduced without a coincident decrease in the reds, the value is necessarily less than one. This latter condition is observed es- pecially in chlorosis and in splenic anemia, while in the pernicious types of anemia we find the diminution of the reds much greater, proportionately, than that of the hemoglobin, a high color index consequently obtaining. In such cases the index may run from 1.02 to 1.9 as in one case observed by the writer. Such variations are the rule, but are not invariable. We may find the various anemias showing, at times, very unusual color indices. As this factor is intended to show only the relations of the hemoglobin to the cells it must not be taken as absolutely diagnostic, but rather as merely significant. It must be remembered, moreover, that this figure cannot be absolutely exact as it is based on a purely arbitrary number of red cells as the normal value and as the instruments for estimating hemoglobin are often improperly standardized. The results are, however, comparable and often yield valuable clinical informa- tion, if individual, racial, and seasonal variations in both the amount of hemo- globin and the number of red cells are taken into consideration. (C). Proteins of the Blood. From the point of view of physiological chemistry a discussion of the various protein bodies found in the blood embraces, necessarily, those of the intracellu- lar fluid as well as those of the cellular elements. From the clinical stand- point, however, the discussion of this field is limited more or less to the proteins of the serum. I can, therefore, do little more than refer to the fact that the red cells contain, besides the hemoglobin which constitutes about 90 per cent, of their organic matter, a nucleoprotein which shows properties resembling those of both the globulins and albumins. The proteins of the white cells are still little dif- ferentiated. Miescher found five different forms of protein, showing various solubility and coagulation relations. Besides these the leucocytes contain, as characteristic proteins, nucleins, which are compounds of the phosphoric acid- containing nucleinic acid with simple albumins. The metabolism of the nucleins is an important factor in various clinical conditions, but I must refer elsewhere for such discussions. Little is known regarding the chemical composition of 420 DIAGNOSTIC METHODS. the blood-plates. Lbwit affirms that they are composed principally of globu- lin, while Lilienfeld asserts that their substance belongs to the class of nucleo- albumins. For the differentiation of the subdivisions of the proteins mentioned above, works on physiological chemistry must be consulted. In the process of coagulation, fibrinogen, a protein of the plasma, is converted into fibrin through the influence of thrombin (fibrin ferment), whose chemical nature is not absolutely settled; it belongs probably to the class of nucleo-proteins, possessing, however, many of the characteristics of the globulins. Blood serum, of both physiologic and pathologic types, contains two protein bodies, serum globulin and serum albumin. A third body, called by Chabrie albumon, has been assumed, but the researches of Drechsel and of Brunner have shown that this body is not preformed in the serum, but arises from serum albumin and serum globulin during the process of coagulation. According to Hammarsten, normal human blood serum contains 7.62 per cent, of total protein, while Schmidt gives this figure as 8.26 per cent. The researches of Reiss, Strauss and Chajes, and more recently of Engel have shown that the refraction coefficient of serum, in health, is between 1.3487 and 1.3517, corresponding to a percentage of 7.74 to 9.13 of protein. The introduction of the refraction coefficient into the study of the serum and plasma has furnished a method of clinically studying hydremia as well as one by which the water content and serum proteins may be easily estimated. These factors are of great importance in the study of the various anemias, as we know that the serum or plasma is of much more importance, from the pathologic stand- point, than are the variations in the cellular structures, which are simply manifestations of profound changes in the liquid portions of the blood. In this connection we must distinguish between an increase in the proteins of the blood (hyperalbuminosis) and a decrease in their amount (hypalbumin- osis). The former is observed whenever water is more rapidly withdrawn from the system, and so from the blood, than it can be supplied. Such conditions are furnished by marked diarrhea, cholera, profuse perspiration, and polyuria without extra intake of fluid. This increase in protein content is only transient and is a result of mere concentration of the blood, the proteins passing out in relatively less amount than does the water. If the above conditions are associated with true exudation, then, of course, a hypalbumino- sis of transient duration will replace the hyperalbuminosis. This decrease in the amount of total proteins is observed whenever direct loss of protein from the blood occurs. Thus Becquerel and Rodier observed a diminution of the proteins in Bright's disease, cardiac edema, and puerperal fever. Hoppe- Seyler noted a loss in melanosarcoma, while Schmidt, von Jaksch, Panum and Limbeck, and Pick have reported such a condition in severe anemias and leukemias. In most severe infections, as Ewing states, the proteins are but slightly reduced. Along with hypalbuminosis we often observe a hydremia which may be referable to a direct absorption of fluid from the tissue under the influence of a hypertonic state of the blood. ■ THE BLOOD. 421 In contrast to the constancy of the total protein values of the serum, we find the relationship between the albumin and globulin markedly disturbed at times. These two bodies are in reality not definite chemical compounds, but are separable, each into two distinct substances with different solubilities and precipitation constants. This fact may have great importance as further study is made, inasmuch as Pick has shown that pseudoglobulin has associated with it the antitoxins of diphtheria and of tetanus. The normal amount of total protein being taken as 7.62 per cent., Hammarsten has shown that this percentage is made up of 3.10 per cent, of serum globulin and 4.52 per cent, of serum albumin, the relationship of globulin to albumin beings as 1:1.5. This ratio is variable, running, according to Limbeck and Pick, globulin 16.9 to 38.3 per cent, of total protein, albumin 61.7 to 83.1 per cent, of the total albuminous content. Such a wide variation makes it difficult to establish any absolute relations in disease. Erben has shown that the albumin remains about normal in pernicious anemia, while the globulin is markedly reduced. The researches of Estelle, Hoffmann, Halliburton, and Mya and Viglezio show marked pathological variations not only in the total protein content, but in the so-called "protein quotient," globulin* The latter authors conclude among other deductions that the relation of the proteins is greatly changed in disease, especially in conditions associated with transudation and exudation, in the sense that the globulins1 are increased while the albumins are diminished. In such states globulin is seen to be relatively more resistant and less diffusible than is albumin (Gottschalk). That this view is not uniformly applicable is noted from the work of Freund who observed in the serum of nephritis a relationship of 1 111.3 between the globulin and albumin, instead of the normal 1:1.5. Ducceschi has reported an interesting observation on this point. He found that, during the period preceding the convulsions which follow thyroidec- tomy, a percentage increase of albumin as regards globulins obtains, while during the convulsions the reverse conditions are present. In this discussion I have taken no account of the total protein of the whole blood. In its determination we include not only the albumin and globulin of the serum, but also the hemoglobin of the red cells, the nucleo- proteins of the white cells, and the fibrinogen of the plasma. Traces of albumose and peptone (?) have been found in pathological conditions, the former possibly in normal states, while still other nitrogenous constituents are commonly determined with the proteins.2 The total protein of the blood has been given by von Jaksch as 22.62 per cent., while Limbeck states a percentage of 25. So much depends upon the method adopted for its estimation and so much on the physiological state of the blood that comparative figures are difficultly ob- tainable from the literature. Regarding the determination of the total protein and of the globulin and albumin, the reader is referred to the section on Urine. 1 See Epstein, Jour, of Exp. Med., vol. 16, 1912, p. 719. 2 See Folin and Denis for a discussion of protein metabolism as shown by blood examina- tions (Jour, of Biol. Chern., vol. 11, 1912, pp. 87 and 161; Ibid., vol. 12, 1912, p. 141; Ibid., vol. 14, 1913, p. 29). 422 DIAGNOSTIC METHODS. Regarding the question of the presence of peptone in the blood, as reported by von Jaksch, Freund and Obermayer, and Ludwig, much depends on the proper differentiation of the substance found. So much has been called peptone that is, in reality, albumose, that we are uncertain whether peptone was present or, if it were, whether it was not formed in the processes used or was not a postmortem product. Devoto and Wagner could not confirm the finding of true peptone when blood from the living subject was examined. It is rather strange that peptone, which reduces the coagulability of blood when added in small amounts, should not exert this power in the vessels during life were it really present. The fact, however, that peptone and albumose have both been repeatedly found in the urine in various condi- tions by competent observers points to the probability of the existence of these products in the blood, although they may be combined in such a way as not to be capable of easy detection. Further elaboration of our methods of detection and differentiation may clear up a much disputed field. Bywaters1 believes the so-called albumose of the blood to be identical with sero-mucoid. (D). Other Nitrogenous Constituents. Other nitrogenous bodies than those already discussed are found in the blood both in normal and in abnormal conditions. While these remaining azotized bodies are not as important as are the proteins, yet in some cases they have great clinical and experimental value, especially in metabolic studies.2 There is no question but that the variation in the total nitrogen of the blood is worthy of a more extended study, especially in the various constitutional diseases. Metabolism in febrile diseases is so intimately associated with increased protein disintegration that a proper study of the changes taking place under the influence of abnormal body temperatures should include an estimation of the total nitrogen as well as of the nitrogen-partition of the blood. The methods to be used in such study have been discussed in detail under The Urine to which section the reader is referred. Slight changes in the technic of certain determinations are necessarily adopted in blood work, owing to the preponderance of the proteins over the other nitrogenous constituents. Von Jaksch in his studies found the total nitrogen of normal blood serum to be 1.37 per cent., while that of the whole blood showed a value of 3.62 per cent. Variations in these figures are no doubt frequent, but have been rarely reported owing to lack of work along these lines. General conclusions can, therefore, not be drawn from the meager literature. Total Nitrogen. This is the most important nitrogenous constituent of the blood aside from the proteins already discussed. Its importance lies not so much in its Urea. 1 Biochem. Zeitsch., Bd. 15, 1909, S. 344. 2 See Folin and Denis (Jour, of Biol. Chem., vol. 11, 1912, p. 527) for new microchemical methods for these nitrogenous substances; also Lowy, Zeitsch. f. physiol. Chem., Bd. 79, 1912, S. 349. THE BLOOD. 423 pathological significance as in its relatively large amount. Traces of urea are normally present in the blood in amounts of 0.011 to 0.02 percent., accord- ing to Gottlieb and Schroder, while Schondorf has shown that during digestion this value may reach 0.61 per cent, and in pathological conditions, especially in uremia and other states complicated with a marked nephritis, it may reach even higher figures. This points to the accumulation of urea in the system whenever its elimination through the kidneys is interfered with. No further pathological significance can attach to such accumulations of urea as they are not sufficient to bring out the hemolytic effects noticed when blood is treated with relatively strong solutions of urea or to cause the marked diuresis which is observed following the administration of urea. The retention of urea in the blood has been held responsible by various workers for the condition of uremia, but more recent research has shown that such a view is untenable. In febrile conditions, associated with increased protein decomposition, urea is found in increased amounts in the blood. As the liver is the chief seat of formation of urea, we find in cases of marked insufficiency of this organ the urea content of the blood and also of the urine markedly reduced; instead of urea, its precursors, the amino acids and ammonium salts, being found in increased amount. For the methods of detection and determination of urea in the blood, I must refer to the section on Urine or to works on physiological chemistry. Like many other substances formed outside the kidneys but excreted by them, uric acid occurs in the blood in traces too small to yield reactions with the ordinary tests. According to Magnus-Levy,1 uric acid may . be detected in the blood only when it is present in the amount of i mg. in 100 c.c. of blood, as a minimum. The presence in the blood of an increased amount of uric acid is known as uricacidemia and is associated with the so-called uric acid or lithemic diathesis. Weintraud found 5 mg. of uric acid in 100 c.c. of blood following the ingestion of a large amount of sweetbreads. This points doubtless to the fact that the kidneys have a limited power of excreting uric acid when a surplus is suddenly poured into the blood from the digestive tract. Further, we observe in cases of severe nephritis, in which the eliminatory power of the kidneys is reduced, quite an accumulation of uric acid in the blood (3 to 6 mg. and over). Uricacidemia is likewise observed in conditions associated with insufficient aeration of the blood, as found in organic heart disease, emphysema, and exudative pleurisy. Febrile states have, per se, nothing in common with uricacidemia, although in some conditions associated with fever uric acid may accumulate in the blood, but not as a direct result of the increased temperature of the body. For some time, following the ideas of Garrod, an accumulation of uric acid in the blood was held to be pathognomonic of gout. He based his theory Uric Acid. 1 Berlin, 1906. 424 DIAGNOSTIC METHODS. oh the fact that uric acid was detected in the blood by the so-called "thread test" which is now known to be imperfect and unreliable. Although his con- clusions regarding the direct relation of gout to accumulation of uric acid have been proven by later writers to be ill-founded, yet we do find a lithemia in gout. This condition arises, however, from the increased nuclein metabolism in this disease, which is of pure endogenous origin as Schittenhelm has shown. This endogenous uric acid metabolism is not peculiar to gout, as accumulations arising from the same source are observed to as great, or even greater, extent in pneumonia, leukemia, and other severe anemias. The conclusions of Haig regarding the influence of excess of uric acid in the blood in various conditions seem to the writer absolutely unwarranted and unscientific. Klemperer1 and especially Magnus-Levy2 have reported the most careful researches on such accumulations of uric acid in gout, showing that no increase occurs preced- ing or during the attack, but that a marked accumulation obtains following the seizure. The method of detection as well as the chemical and physical properties of uric acid must be found elsewhere.3 Xanthin Bases. The xanthin, alloxur, or purin bases do not occur in normal human blood in appreciable amounts. Their close relationship to uric acid makes it decidedly probable that they are present, along with uric acid, in cases of gout, leukemia, emphysema, pleurisy, and nephritis, although the literature does not yield any extensive researches covering this point. This subject is discussed in considerable detail under The Urine, to which the reader is referred. Ammonia. It is highly probable that this substance is a normal blood constituent, being present in venous blood, according to Winterberg, to the amount of 1 mg. per 100 c.c. of blood. The presence of an increased amount of ammonia or of ammonium compounds in pathological blood is certain. The normal metabolism is so regulated that any excess of acids, introduced into the system from without or produced -within the body by increased decomposition of tissue, is neutralized up to a certain point by a corresponding increase in the ammonia produced in the breaking-down processes. While this subject has not been so carefully worked out in the case of the blood as it has in the urine, yet we are justified in assuming that a preceding increase of ammonium compounds must obtain in the blood in order to allow an excretion of such an excess in the urine. An increase in the acids of the system must, therefore, carry with it an increase in the ammonium compounds. If the acid intoxication (acidosis) is not too severe, the increase of ammonia is sufficient to neutralize the effects of the acid. However, in most cases showing this symptom-complex, the fixed alkalies are called upon to make up the deficit. This is the basis of the alkali-therapy in febrile states (Walter), in diabetic conditions (Naunyn), 1Deut. med. Woch., Bd. 21, 1895, S. 655. 2Zeitsch. f. klin. Med., Bd. 36, 1899, S. 353. 3 See Folin and-Denis, Jour, of Biol. Chem., vol. 13, 1913, p. 469. THE BLOOD. 425 in cases of cyclic vomiting (Czerny and Keller), and in pneumonic attacks (Billings). It has been shown by various workers that uremia is associated with an increased excretion of ammonia in the urine and probably with an accumula- tion in the blood. This ammoniemia is not to be confounded with the narrow idea of Treitz and of von Jaksch, that uremia is due to the increase of ammonia in the blood, as this view has been shown to be untenable. The increased amount of ammonia, both in the blood and urine, in uremic conditions is probably due to a direct insufficiency on the part of the liver to elaborate urea from its precursors. Moreover, a contributory cause may be found in an abnormal production of acids calling for an increase in ammonia as a neutralizing substance (Senator). Regarding the presence in the blood of other nitrogenous bodies, such as carbamic acid, creatin, hippuric acid, and amino-acids, the writer has something to say in the section on urine. (E). Carbohydrates. The presence of sugar in the blood, both of normal and of abnormal types, is a well-demonstrated fact. While this sugar is probably not in the free molecular state and is not dialyzable, it is uncertain in just what combination it does exist. The discovery of jecorin (a combination of glucose with lecithin) by Drechsel shows us that such combinations occur in the blood; yet it has not shown that this is the usual blood sugar. The sugar usually present in the combined state is glucose, but maltose is occasionally found in the blood of nursing mothers, while levulose or pentose may occur both in the blood and urine after intake of large amounts of food containing them. Glucose. The occurrence of glucose in the blood is termed glycemia. This hexose is found normally in quantities of i to 1.5 parts per 1,000. Under pathological conditions this amount may increase to as high as nine parts per 1,000, as Hoppe-Seyler reports in a case of diabetes. According to Claude-Bernard, when- ever the sugar of the blood reaches three parts per 1,000, diuresis and glycosuria occur as evidences of the systemic effort to control the hyperglycemia. This figure is probably too high, as the later methods1 of study have given a lower value for the normal sugar content and hence for the point at which excretion occurs. Indeed, Pavy believes that traces of free glucose in the blood will be immediately excreted. While a hyperglycemia may be produced by purely alimentary excesses, the most common condition associated with it is diabetes mellitus. Here of course many factors, such as the pancreatic, duodenal, hepatic, and myogenic influences come into action as causative or contributory agents in such production. Freund reports that the blood of patients suffering with carcinoma shows a strong reducing action due to the presence of sugar. As he was able to find 1 See Reicher and Stein, Zeitsch. f. exp. Path. u. Ther., Bd. io, 1912, S. 532; also, Herzfeld, Zeitsch. f. physiol. Chem., Bd. 77, 1912, S. 420. 426 DIAGNOSTIC METHODS. no such condition in the blood of patients with sarcoma he considers hyper- glycemia a differentiating point between the two malignant states. Trinkler confirms these findings and adds that the blood of nephritics shows a very small amount of sugar, while that of febrile cases shows an increase. Such conditions as the preceding are prone to be associated with the appearance of reducing substances in the blood, but there is some doubt as to whether sugar is accountable for such reactions, although increased -protein disintegration occurs, which in itself might give rise to a transient hyperglycemia. A hyperglycemia is observed following the removal of the pancreas or after perversion of its function, as von Mering and Minkowski have shown. Recently Pfliiger1 has been able to demonstrate a marked glycosuria and hence an hypothetical hyperglycemia, following removal of the duodenum. Both the preceding conditions give rise to diabetes and cannot be discussed in this place. Several blood tests advocated as diagnostic of diabetes have been advanced, especially by Williamson and Bremer, but these will be dis- cussed in a later section. No doubt can be raised as to the presence in the blood of reducing substances other than glucose. According to Paul Mayer, we must assume that both normal and abnormal metabolism are associated with the conjugation of various aromatic substances with glycuronic acid. These glycuronates easily split up into their constitutent elements and show reducing properties. Whether sugar, in its normal metabolism, must invariably pass through the intermediate stage of glycuronic acid is debatable, yet the theory is a fascinating one and fits into many obscure points. This polysaccharid undoubtedly appears in the blood singly or in combi- nation with albuminous bodies. Salomon, Frerichs, Lepine, Ehrlich, and Gabritschewsky have reported it, while Caminer was unable to find it. Hup- pert obtained it in quantities ranging from 0.114 to 1.56 grams per 100 parts of blood. Much depends on the method used to isolate and determine this sub- stance, as it is easily lost by careless manipulation. Certain properties of the granules found in the blood, both extra- and intracellularly, have led many to be- lieve that glycogen is present as a characteristic in many conditions. There is much reason, however, to assume that these granules, which stain brownish with iodin, are not glycogen, but rather albuminous bodies of indefinite com- position. For a discussion of this subject see the treatment of iodophilia. Glycogen. This carbohydate has not as far as I am aware, been found in normal blood, but has been reported by Freund in that of tubercular patients. Cellulose. (F). Fats and Fatty Acids. The presence of free fats (palmitin, stearin, and olein) in the blood has been frequently observed both in health and disease. The relative quantities 1 Pfliiger's Archiv., Bd. 118, 1907, S. 267 THE BLOOD. 427 of these fats vary in different animals and are subject to wide variation in the same animal under influence of diet. Bonninger gives the fat content of normal human blood as 0.75 to 0.85 per cent., while Engelhardt finds 0.101 to 0.273 Per cent., these latter figures being practically confirmed by Becquerel and Rodier. The presence of an excess of free fat in the blood is referred to as lipemia. The physiologic variations are more notable than are the pathological, being observed after ingestion of a meal rich in fats, in breast-fed children, in preg- nant women, and in the obese. Pathologically lipemia more or less permanent may be observed in various conditions, as in acute and chronic alcoholism. d.iabetes, arteriosclerosis, chronic nephritis, phthisis, carbon monoxide and phosphorus poisoning, gout, typhoid fever, fat embolism following injuries of the long bones, pneumonia, leukemia, acute infections, cachexia from inani- tion or malignant disease, hepatic diseases, and malaria. It has, therefore, little differential diagnostic value. The extent of the lipemia may vary from the presence of isolated fat droplets to the overloading of the blood to such a degree that it becomes salmon-colored, turbid, and milky. This fat may be either the normal fat which has been transported from different parts of the body or may be that abnormal to the body arising from excess of fat in the diet. The fat is soluble in ether and stains black with osmic acid and red with Sudan-Ill. Regarding the quantitative estimations of fat, I must refer to other sections. Concerning the presence of fatty acids in the blood little is known. Traces of volatile acids are sometimes present, but probably not as normal constit- uents. Gaglio, Spiro, Irisawa, and Berlinerblau report sarcolactic acid as a normal finding, while Zweifel finds an excess of this acid in the blood and in the urine in cases of toxemia of pregnancy. An excess of fatty acids in the blood is known as lipacidemia. Von Jaksch found fatty acids in the blood in cases of diabetic coma, leukemia, acute yellow atrophy of the liver, while Hougounenq reports the presence of ^-oxybutyric acid in the cadaveric blood of a diabetic. It is rational to assume that in all those conditions associated with acidosis, fatty acids are present in the blood, as these may be detected in large amount in the urine in these states. The occurrence in the blood of demonstrable amounts of acetone is known as acetonemia. Deichmiiller and von Jaksch have found a substance giving the reaction of acetone in various conditions, especially in fevers, which finding has been confirmed by Reale. Whether acetone is a product of normal intermediary metabolism and, as such, is found in many physiologic and pathologic states, must be found in the section on Urine. (G). Acetone. The conditions in which the biliary constituents, especially the pigments and acids, are found in the blood is termed cholemia.1 It is usually stated that these elements are not found in normal blood, but Croftan has shown (H). Biliary Constituents. 1 See Lehndorff, Prager Med. Wochs., Bd. 37, 1912, S. 495. 428 DIAGNOSTIC METHODS. that the bile acids are observed in the blood of healthy subjects. This is not unexpected as they are completely absorbed from the intestines and are re- excreted in the bile (Weintraud). Pathologically, both the acids and the pigments are found in the blood in any condition associated with their appearance in the urine. Oftentimes they may be found in the blood when no reaction for them is obtainable in the urine. While the bile acids exert marked toxic effects, such as hemolysis, the biliary pigments show, as Bouchard, de Bruin, Lugli, and Colosanti have demonstrated, certain harmful influences. Flint ascribes the toxic effects observed in cholemia to cholesterin, but this idea needs confirmation. The most usual condition showing cholemia is jaundice. Whether the blood changes observed in the various types of jaundice are due, primarily, to the cholemia or to a primary hemolysis or to a combination of these effects as a result of intoxication is an unsettled question. These phases of the blood variations in jaundice will be discussed later.1 (Z). Inorganic Constituents. The inorganic composition of the blood shows quite a marked variation both under physiologic and pathologic influences. This variation applies both to the cellular and intracellular constituents of the blood. Changes in the molecular concentration as well as changes in the concentration of specific inorganic combinations are observed in the blood of the two sexes, as may be seen by consulting the table on page 400. Regarding the special significance of the different inorganic constituents little is known, but chemical analyses of the blood ash in health and disease have shown that the pathological variations are more important as regards the chlorids, phosphates, and the iron compounds.2 Chlorids. Physiologically, a certain (about six parts per mille) concentration of sodium chlorid is necessary to hold the proteins in solution, as well as to maintain the proper osmotic tension of the serum. The larger the proportion of plasma, the greater the percentage of chlorids in the blood. This is true only within certain limits as the NaCl-content remains practically constant, no matter how large an amount is ingested. This constant value is regulated by the increased or diminished renal excretion. In anemias the chlorids of the blood are usually high, yet, according to Limbeck, cases are occasionally found in which normal amounts of sodium chlorid obtain. In pneumonia, in which a diminished urinary excretion of chlorids is observed, the blood does not show an excess of chlorids but may even show a decrease owing to the effects of the exudative process. Diminished ingestion of food, vomiting, diarrhea, and general exudative processes may be associated with a decrease in the chlorids of the blood, but this is merely 1 See Pel. Deutsch. Arch. f. klin. Med., Bd. 106, 1912, S. 239; also, Brugsch and Retzlaff, Zeitsch. f. exp. Path. u. Ther., Bd. n, 1912, S. 508. 2 See Weissbein and Aufrecht, Internat. Beitr. z. Path. u. Ther. d. Ernahr., Bd. 4, 1912, S. 22. THE BLOOD. 429 temporary. In cases of marked nephritis, associated with retention of chlorids, the blood may show an excess of sodium chlorid, but this condition soon disappears through the influence of the increased edema. This phase of the question, originally advanced by Widal and Javal, has been previously discussed. Phosphates. These compounds exist in the blood as neutral or alkaline salts of sodium, potassium, calcium, and magnesium, as well as in organic combinations in the red and white cells in the form of lecithin and nuclein. The inorganic phosphates are concerned, at least in part, with the solubility of the proteins. Just what variations in the amounts of the organic and inorganic phosphorus compounds occur in health and disease is unknown. It is certain that a definite phosphorus metabolism exists and that this is characterized, to a great extent, by variations in the normal relationship between the compounds of the alkali and alkali-earth groups. Just what these relations are in physiologic and pathologic states experiment must determine. That the system retains compounds of phos- phorus more energetically than it does any other mineral constituents is proven. As the phosphates of the blood and of the urine come both from the food and from the breaking down of the nuclein-containing protein material of the system, large variations are possible under the influence of many factors. The subject of the metabolism of phosphorus is taken up under The Urine. Iron. The iron of the blood is found principally in the hemoglobin which contains about 0.42 per cent, of Fe. It is also found in traces in the plasma, and, ac- cording to Hammarsten, in the nuclein compounds. The attempt to estimate the amount of hemoglobin by determinations of the blood iron have proven failures, as no definite relations exist between them. All of the blood iron is not in the form of colored compound? and, moreover, some of the derivatives of hemo- globin are iron-free. Biernacki has shown that the direct quantitative estimation of iron yields higher results than could be obtained by computation from the percentage of hemoglobin, this finding being confirmed by Jolies and Jellinek. According to Schmidt and Becquerel and Rodier, the amount of iron in the blood ranges between 0.056 and 0.058 per cent. Female blood shows a some- what lower value, just as it does for hemoglobin. For the estimation of the iron of the blood, I must refer to other works for the details, as the method is too cumbersome for general clinical use. The principle of the method is based on the comparison of the colors of a known solution of iron treated with am- monium sulphocyanate solution, with that of a solution of blood iron, obtained by incinerating blood, fusing the ash with potassium bisulphate, dissolving the fused mass in water, and treating it with sulphocyanate solution. The instrument used for the colorimetric tests is called Jolies' ferrometer, the amount of iron in 1,000 c.c. of blood being obtained by reference to a table accompanying the instrument. If it is desired to obtain the actual percentage by weight, the calculation must include the determination of the specific gravity of 430 DIAGNOSTIC METHODS. the blood. Having this latter factor, we may use the equation G : V :: 100 : X, in which G represents the specific gravity and V the percentage by volume of iron which is obtained from the table. The hemoglobin value may be . . . ... . . Til TT1 TOO X ^1 . 1 * 1 H i found by use of the formula of von J aksch, Hb = - , in which M equals 0.42 the percentage of iron by weight. (J). Blood Gases. The gases existing in the blood are oxygen, carbon dioxide, and nitrogen, the latter having little importance in the body functions, its amount corre- sponding to that which would be absorbed by an equal volume of distilled water, namely, 1.8 volumes per cent. The amount of oxygen and of CO2 varies widely, depending on the arterial or venous character of the blood, upon the velocity of the blood flow, upon the temperature, amount of exercise, etc. Oxygen occurs principally in chemical combination with hemoglobin, but a small amount, about one-fourth per cent., is present in solution in the plasma. About one-tenth of the C02 is held in solution in the blood, while the remaining nine-tenth is apportioned as follows, one-third loosely bound to the alkalies and hemoglobin of the corpuscles, two- thirds held in chemical combination with the alkalies and proteins of the plasma. The following table made up from the figures of Setschenow, Ludwig and Sczelkow, and Hammarsten, will show the relation between these gases. A rterial. Venous. Oxygen, 21.6% by volume. 6.8% by volume. Carbon dioxid, 40.3% by volume. 48.0% by volume. Nitrogen, 1.8% by volume. 1.8% by volume. (A). Ferments of the Blood. The fact has been well established by Jacdbi that the various organs and tissues of the body contain ferments which are proteolytic as far as the corresponding tissues are concerned, but are usually inactive when applied to the proteins of other organs. In other words, these ferments are autolytic, but not, as a rule, heterolytic. Much experimental work of former and latter years has shown that many metabolic processes, associated with the building up and breaking down of various tissue elements, are influenced to a great extent by the presence of ferments arising from many sources. It is not unreasonable to assume that the blood, like other tissues, contains various ferments which have to do with general and special metabolism. The work of Schbnbein on the oxidases of the blood, of Hanriot and, indirectly, of Castle and Loevenhart on the lipolytic ferment, of Lepine, Seegen, and Spitzer on the glycolytic ferment shows that such properties are resident in the blood. The work of Ascoli and Moreschi, Jochmann and Muller, and of Stern and Eppenstein has opened up an entirely new field of work on the proteolytic properties of the leucocytes. This proteolytic ferment action is both autolytic and heterolytic and is influenced to a great extent by the checking THE BLOOD. 431 action of the antiferments, which have been so well studied by Opie. Further, it has been shown that differences exist between the proteolytic properties of the many varieties of leucocytes and that the ferments are not always heterolytic (Mosse). It would lead me too far afield to discuss this subject in detail; hence, I must be content with reference to the presence of these substances.1 Much benefit will be forthcoming from a further study of such properties of the blood and of certain constituents of the blood. It is to be recalled that ferment action may be accountable for the influences of the toxophore and other groups of Ehrlich's complement, but this phase must be discussed later. (7). Enumeration of Red and White Cells. This section of hematological technic is, perhaps, the most perfected and most frequently employed. The red and white cells are usually counted, as the enumeration of the platelets has little practical or scientific value at the present time and the latter technic is not sufficiently perfected to admit of conclusions. Various instruments have been introduced for the purpose of counting the corpuscles in a given volume of blood. Most of them are based on the principle that a layer of diluted blood, of a certain depth, covering a certain known space, shows a definite number of corpuscles for every drop of blood used. The general method of making the count consists in diluting the fresh blood in definite proportions with some indifferent fluid and counting, under the microscope, the number of cells in a drop of this diluted blood, which is contained in a small glass cell on the floor of which is ruled a series of micrometer squares of known dimensions. The cubic contents of the cell and the degree of dilution of the blood being known, the number of corpuscles counted in any given number of squares of the ruled area may be taken as a basis for calculating the total number of cells in a cmm. of blood. Strong and Seligmann dispense with a special counting chamber and enumerate the cells in a definite quantity of blood diluted in exact proportions with a diluent stain and mounted as a permanent specimen. Einhorn and Laporte use a somewhat similar method and arrive at very good comparative results. The normal number of red cells in the adult male is, approximately, 5,000,000 per cmm. of blood, while in the female it is somewhat lower, namely, 4,500,000. Marked variations in this figure are observed in pathological conditions and will be discussed in a later section along with the treatment of the physiological factors which influence the number and appearance of these erythrocytes. In the normal adult the number of white cells varies between 5,000 and 10,000, the average being 7,500. This figure is subject to both physiologic and pathologic influences to a greater extent than are the red cells. Many instruments, such as those of Hayem, Gowers, Malassez, Thoma- Zeiss, Alferow, and Durham, have been introduced, but the most universally used and the best adapted for such investigation is, in the writer's opinion, 1 Smithies (Jour. A. M. A., vol. 59, 1912, p. 539) has called attention to the presence in human blood-serum of an agent (probably a ferment) which shows peptid-splitting properties. 432 DIAGNOSTIC METHODS. that of Thoma-Zeiss. This combines certain modifications of the mixing pipet of Malassez, the counting chamber of Hayem, and the micrometer rulings of Gowers. It gives us, therefore, a most complete instrument for such work. Hemocytometer of Thoma-Zeiss. The blood counter, originated by Thoma and constructed by Zeiss, is all that could be desired. Leitz, Reichert, and other makers of optical goods have introduced similar types of counters, but our experience with them has not been so good. So much depends upon the accuracy with which pipets are graduated and upon the trueness of the rulings of the counting chamber that we recommend the general use of the Zeiss instrument. This instrument consists of a pipet for mixing the blood to a certain dilution, a counting chamber by means of which a layer of known depth and area is obtained, and a special cover-glass for the chamber. Fig. 13 i.-Thoma-Zeiss counting chamber. Pipets The original form of the Thoma apparatus included but one pipet this being used for the dilution when both red and white cells were to be counted. Experience has shown, however, that the dilution given by this pipet is, in most cases, too great to permit of accurate counting of both types of cell.1 A modification has, therefore, been introduced to allow of greater accuracy by giving a lower dilution and by furnishing a larger number of cells in the counting chamber. Erythrocytometer. This is the original pipet of Thoma. It consists of a graduated capillary tube (A) opening into a dilatation (B), at the opposite end of which is a shorter glass tube (C), graduated with a line marked 101 and to which is attached a rubber tube with an ivory mouth-piece. This pipet is so graduated that the capacity of the ampulla (B) is exactly 100 times that of the capillary tube from its point to the line marked 1 and 200 times that from the point to the line marked 0.5. Other lines, both below and above this latter point, are calibrated on the tube, each line representing one-tenth of the capacity of the capillary. In some of the pipets, especially that of the Miescher hemoglobinometer, two smaller marks each representing one one-hundredth of the length of the tube are 1 See Ellermann, Deutsch. Arch. f. klin. Med., Bd. 109, 1913, S. 378. THE BLOOD. 433 calibrated on each side of the major divisions. By this means the dilution of the blood may be definitely known if the tube be not filled exactly to the point described. Those pipets with obtuse ends are much to be preferred to those with the more acute angles. In the ampulla is a small glass bead which is of service in properly mixing the blood with the diluting fluid. This pipet may be used in counting both the red and white cells and is all that is abso- lutely necessary when one uses the Zappert or Turk count- ing chamber. In general work, however, it is used only in the count of the red cells as the extreme dilution given is better adapted to this estimation than it is to the count- ing of the white cells. Leucocytometer. This is a graduated capillary tube similar to the erythrocytometer in construction but having a larger lumen, which will permit of lower dilutions. It is graduated into ten divisions, with the marks 0.5 and 1 representing these measures of the total capacity of the capillary. Above the ampulla is a graduation, 11, which is so calibrated that blood, drawn to the mark 1 and diluted to 11, gives a dilution of 10, while if drawn to the point 0.5 the dilution is 20. Those forms of this pipet which have the lower end tapering to a fine point giving a gradually decreasing lumen, are much to be preferred to the older models. As the caliber of this instrument is relatively large, the student is cautioned against using too great suction in making the dilution, and also against placing too large a drop on the counting surface.1 Fig. 132.-Diluting pipets; Erythrocytom- eter; leucocytometer with Croy's attach- ment. The Counting Chamber. This consists of a heavy glass slide, A, on which is cemented a thick glass ring, B, the surface of which is highly polished. This ring surrounds a circular table of glass, 1), the height of which is just o.i mm. less than that of the surrounding ring. Between this glass table and the inner edge of the ring is a small ditch, C, to catch the drop which may run off from the table and to prevent its flowing up between the ring and the cover-glass on the other side of the ditch. On the central glass table cross at right angles 21 parallel lines, equidistant, and between the extremes of which is exactly 1 mm. Hence we have an area of 1 square millimeter divided into four hundred small equal squares. Through each fifth row of squares is ruled an extra line, which is not a boundary but is 1 My colleague, Dr. C. C. Croy; has devised a most excellent and useful pipet attach- ment (see above cut), which prevents leakage of the contents during transportation. This consists of a strong rubber band with two disks of any kind of thin metal folded on the rubber in such a way that the ends of the pipet come in contact with the rubber and press against the metal disks. 434 DIAGNOSTIC METHODS. merely an aid in keeping one's position in the ruled area. Indicated, not bounded, by these extra lines, the square millimeter is divided into 16 units of 25 small squares each" (Emerson). As the ruled surface of one square millimeter is divided into 400 small squares, each small square has an area of 1/400 square mm. The height of Fig. 133.-Ruled surface of Thoma-Zeiss counting chamber. (DaCosta.) the column of blood being 1/10 mm., the cubic contents of each small square is 1/4000 cmm. This counting chamber of Thoma does not permit of the counting of a sufficiently large number of leucocytes, especially when the dilution has been made in the same degree as is used for the enumeration of the red cells. In order to overcome this difficulty and also to give a larger ruled area in which the leucocytes may be counted, Zappert has modified the original ruling in such a way that a counting surface of 9 sq. mm. is afforded. This modification has been improved by Ewing and by Turk in such a way that the four large corner squares, each of 1 sq. mm., are subdivided into 16 smaller squares, each of which is equal in area to the total 25 smallest squares of the Fig. 134.-Turk's ruling of the counting chamber. THE BLOOD. 435 Thoma chamber. This latter modification is much the best and is used exclusively by the writer. Its advantage in counting both red and white cells will be appreciated when the student compares it with the older chamber. The sixteen central squares are used in counting the erythrocytes, while the entire area may be used in the enumeration of the leucocytes. Simon has recently introduced a different modification of the Thoma ruling, which is extremely simple and should prove very satisfactory. Lack of experience with it prevents the writer, however, from making comments upon its advantages and possible disadvantages. Before use the counting chamber should be well washed with water and carefully dried. Precautions should be taken to see that no lint is left on the surface of the glass ring and that no alcohol or ether are used in the cleaning process, as these substances loosen the cement with which the glass table is fastened to the slide. The Cover-glass. This is made of heavy polished glass with accurately planed surfaces. The ordinary cover-glasses should never be employed with the counting- chamber as they are often uneven in surface and do not fit tightly to the slide. Moreover, these ordinary cover slips are so thin that the capillarity of the drop of blood may bend them down to some extent. The cover-glass must be as carefully cleaned and dried as is the chamber. Diluting Fluids. In order that the blood may be properly examined, it must be diluted with a solution which will at the same time prevent coagulation and hemolysis and will preserve the corpuscles intact. There are numerous formulae for such solutions and the choice is largely a matter of experience. Much will depend on whether the red and white cells are both to be counted at the same dilution or whether two different pipets are used in making the dilutions. Personally the writer prefers the use of two pipets using different diluents, but other workers use one pipet and one diluent. The red cells may be destroyed by certain fluids leaving the white cells intact or the white cells may be colored by the same diluent used in counting the reds. Hayem's Solution. This solution preserves the red-cells permanently and permits the corpuscles to settle slowly, thus furnishing an even distribution of the cells. Moreover, it will keep almost indefinitely and does not permit of the development of yeast spores which so readily multiply in many of the other diluents. The writer can recommend this diluent as the most generally useful of the prepara- tions advised. It is made up as follows: Mercuric chlorid, 0.500 gram. Sodium sulphate, 5.000 grams. Sodium chlorid, 1.000 gram. Distilled water, 200.000 c.c. 436 DIAGNOSTIC METHODS. Owing to the presence of mercuric chlorid, this fluid cannot be mixed with an aniline coloring substance to stain the leucocytes and is, therefore, not applicable to the combined counting of the red and white cells. Toisson's Fluid Sodium chlorid, 1.000 gram. Sodium sulphate, 8.000 grams. Neutral glycerin, 30.000 c.c. Distilled water, 160.000 c.c. Methyl violet 5 B., 0 025 gram. The addition of methyl violet serves to color the leucocytes and permits of their recognition along with the erythrocytes. Occasionally this fluid hemolyzes the red cells and thus invalidates the count. Moreover, it easily becomes infected with yeast spores which develop profusely in it. For this reason it is advisable to filter the fluid before use, each filtration, however, weakening it, so that it becomes after a time useless. Other diluents, such as the solutions of Pacini, Lowit, Petrone, Acquisto, Marcano, and Edington, have been used, but the more generally applicable ones are those mentioned above. Toisson's fluid is particularly useful when the count of reds and whites is to be made in the same specimen. The coloring of the leucocytes is not necessary for their recognition, but it is a convenience to one who is not making blood counts frequently. Hayem's solution is the very best diluent at our command for general purposes. If it is desired to count the white cells alone, and this is always wise, a one per cent, solution of acetic acid, to which is added gentian violet to bring out the white cells a little more clearly, may be used. This solution destroys the red cells and thus gives only the white cells in the preparation. For this reason the addition of the gentian violet is unnecessary. Yeast cells develop in this solution with more or less readiness, hence one should employ only freshly made solutions, as these yeast cells resemble, to some extent, mono- nuclear leucocytes and may introduce an error into the count. Method of Counting the Corpuscles. With this process of counting the cells, whether red or white, there are five steps to be taken: 1. Obtaining the blood. 2. Diluting and mixing the blood 3. Filling the counting chamber. 4. Counting the cells. 5. Cleaning the apparatus. Erythrocytes. (i). Obtaining the Blood. As previously stated, the blood may be drawn from a puncture of the ear or finger. Personally, the writer always uses the ear, unless some valid THE BLOOD. 437 reason exists for not doing so. As soon as a good-sized drop appears, which is obtained without pressure or constriction, the tip of the pipet is placed in the drop and is supported by a finger of the left hand, which holds the ear in position. The blood is drawn by suction to the mark 0.5, in cases in which anemia is not suspected, or to the mark 1 in such cases, as a routine the former mark being preferable. As the student will find, some practice is needed to stop the column of blood exactly at the point desired. If the blood be drawn a little too far, the tip of the pipet may be rubbed with the finger or the excess may be shaken down by tapping the tip against a towel or, preferably, the blood may be drawn to the next mark and the necessary correction made in the dilution. Unless this error can be corrected, the pipet must be cleaned or a second one used. In those pipets which have extra marks, indicating 1/100 of the length of the capillary, we have a measure which may prevent the necessity of using a second test, as the correction may be made in the final calculation. In any event, if the blood be not accurately measured, we must reject the deter- mination. Moreover, the work should be done rapidly to prevent coagulation of the blood. Hence, too much time should not be spent in adjusting the column of blood.1 As soon as the column of blood is adjusted at the desired height of the capillary, the tip of the pipet is carefully wiped with the fingers to remove any adherent blood and is immediately dipped into the diluting fluid in such a way that no portion of the blood is lost. Turk recommends closing the end of the pipet with the finger and exerting slight suction on the closed tube in order to prevent any loss on immersion of the pipet. I have never seen it necessary to use this measure, providing ordinary care is employed in applying the suction as soon as the pipet touches the diluent. The diluting fluid should stand ready in a small dish or bottle and should be carefully examined before use to see that no flocculi or spores are present. The diluent should be at once drawn into the capillary by suction. The fluid rises slowly in the tube, the pipet which is held vertically being rotated between the finger and thumb of the left hand as the fluid rises. By thi§ rotation, the diluting fluid is mixed with the blood at once and bubbles of air, which often cling to the inside of the tube, are avoided. The glass bead in the ampulla serves the purpose of thoroughly mixing the blood and diluent. As the column approaches the mark ioi on the upper end of the pipet, care should be taken that the aspiration is not too strong. While the error, intro- duced by drawing the diluted blood beyond this mark, is not as great as it is in drawing the undiluted blood to the mark 0.5, yet both should be avoided. When this upper mark is reached, withdraw the pipet from the diluent, close the tip with the finger, bend the rubber tubing down over the other end, and close this with the thumb. Some prefer to remove the rubber tubing at this (2). Diluting and Mixing the Blood. 1 See Maddox (Jour. A. M. A., vol. 60, 1913, p. 663) for the description of a new device aiding in filling the pipets. 438 DIAGNOSTIC METHODS. juncture, but this is not. at all necessary. Shake the pipet vigorously for at least one minute to insure a thoroughly uniform mixing of the contents. If for any reason the count is not to be made at once or if it be desired to carry the blood to the laboratory for examination, the rubber tubing may be removed and the ends of the pipet closed by means of a rubber band. Naturally, before the mixture can be used it must be again thoroughly shaken. If the examination is to be done at once, proceed as follows. Blow three or four drops from the pipet in order to remove the column of fluid which has remained in the capillary and has not mixed with the blood. If Toisson's fluid has been used as the diluent, it is better, before blowing out the drops, to allow the pipet to lie horizontally for ten to fifteen minutes in order to permit of the staining of the leucocytes. Shake the tube well, blow out the drops as stated, and adjust a small drop on the counting table. (3). Filling the Counting Chamber. As a rule, it is better, unless Toisson's fluid has been the diluent, to fill the counting-chamber at once as errors may creep in by allowing the pipet to stand, even though the later mixing may be thorough. A small drop, the size of which can be learned only by experience, is blown onto the center of the ruled area of the counting chamber. This drop should not be so large as to run over into the moat, but should be large enough practically to cover the glass table. A drop which is either too large or too small will introduce an error into the count. Adjust the cover-glass at once. This is a point in the technic which requires considerable practice and must be mastered before accurate results can be obtained. Emerson's advice on this point is admirable, "grasp the cover-glass by two diagonal corners, place a third corner against the slide with the edge of the glass ring as a fulcrum, and hold it in that position by a finger of the left hand. By now raising the finger the cover is rotated onto the drop radidly and also in such a way that no air-bubble is left." Breathing upon the cover-glass before it is adjusted is often servicable in making this preparation. If there has been no dust on the slide or cover-glass and they are ■ perfectly clean, the student will observe, providing the adjustment has been properly made, a beautiful band of colors known as Newton's rings, which are due to a phenomenon of interference of light. If these rings do not appear, they may often be brought out by firm pressure on the edges of the cover-glass. If they are not persistent, after the pressure has been removed, the adjustment of the two polished surfaces must be assumed to be imperfect and the preparation rejected. It happens at times that these rings are difficult to obtain and some workers state that they are not necessary for an accurate count. The writer prefers, however, to reject those slides which do not show these diffraction rings than to run the risk of including a possible error. These rings may best be seen by holding the slide on a level with the eyes in such a way that the light is totally reflected from the surface of the cover-glass. After the slide is properly adjusted allow it to stand for three or four minutes before proceeding with THE BLOOD. 439 the next step of the determination, in order to insure the proper settling of the corpuscles upon the counting table. (4). Counting the Cells. Before any count of the cells is attempted, the entire surface covered by the blood must be examined with the low-power lens to ascertain whether the distribution of the cells is uniform throughout. If not, the slide should be rejected, even though the points mentioned previously have obtained. It is much better to stop proceedings at this point than to attempt to equalize an uneven distribution by a larger counting area. If the specimen proves satis- factory the count may then be undertaken. The lenses, best suited to this purpose, are Bausch & Lomb |, Zeiss D, Leitz 6 or 7, and Reichert |. As the student becomes accustomed to cell-counting, he may use a lower objective. This has the advantage of bringing the entire field of one sq. mm. into focus. A mechanical stage is of some convenience in making the count, but the fingers answer quite as well after some practice in manipulating the slide. The unit of counting surface is a matter of individual preference. Sahli recommends a unit of four small squares, Grawitz and Simon use sixteen of these squares, while Cabot advises the use of 36. In common with Turk, Ewing, Da Costa, Wood, and Emerson, the writer prefers the unit of 25 small squares, as this is the unit of the ruling and the calculation is much simpler than with the other units. In order to simplify the process of counting, some routine method must be used. These methods depend on the worker, but the usual procedure in the writer's laboratory is as follows: Adjust the slide upon the table of the microscope so that the upper left-hand corner of the central ruled area of 16 large squares of the Turk chamber is brought into the field in such a manner that one may count the cells in the small squares from left to right. If the low- power objective is used, the complete group of 16 units will lie within the field of observation and may be easily examined. With the higher power, which the student may be forced to use in his earlier work, the optical field is neces- sarily limited. Count the total number of cells lying within the 25 small squares of the single unit. In doing this count the upper row of the unit from left to right, drop down a row, and count the cells from right to left and so on until the cells in the whole unit have been counted. The accompanying cut will indicate the method to be followed. In making this count, cells which touch the right hand or lower boundaries of the unit are disregarded, while those which touch the upper and left-hand line are included in the count of that square. After counting the cells in the upper left-hand unit, count those in the remaining fifteen units, thus covering a field of 16 units of 25 small squares each, making a total of 400 small squares counted. Turk recommends the counting of eight units as a minimum, Emerson advises the counting of the four corner units in each of two separate preparations, while Da Costa and others count the cells in four groups of units from above downward and repeat with four units not 440 DIAGNOSTIC METHODS. adjacent. It seems to the writer that the error is less the larger the area covered, and he, therefore, advises the beginner to make the total count of the sixteen units, although it is not supposed that one will try to cover up defects in technic by this larger counting area. A variation of more than 25 cells in the count of the varions units should be taken as evidence that the distribution is not perfect. In such cases the count would better be rejected on the ground of inaccuracy. After the technic is mastered and the worker has discovered just exactly wherein his error lies, a count of 8 or 4 units will suffice. It is advisable, where accurate scientific results are indispensable, to clean the slide and make a second count with a fresh drop of blood so that one may have a check on his work. For clinical purposes, however, the count of four units will ordinarily be sufficient. Fig. 135.-Plan of Counting the Cells. {Da Costa.) The small squares are examined in the order indicated by the arrow In ordinary counting it is not necessary to differentiate the red cells from the leucocytes, as the error thus introduced is small and may be disregarded. In cases, however, which show large leucocyte values, this error will be quite appreciable and must be overcome by the following method. All of the cells observed may be counted as erythrocytes and the reduction made for the number of leucocytes as obtained by the special leucocyte count to be described later. In the use of Hayem's diluting fluid, the leucocytes are not colored, while the erythrocytes retain their normal yellow color. The leucocytes appear, by good illumination of the field, of a bluish tone, are somewhat larger than the red cells, and are characterized by a sharper border. These facts will enable one after some experience to distinguish the white from the red cells. Toisson's fluid stains the leucocytes blue, and may, therefore, be used to outline these cells. As a rule, it is better to learn to recognize the leucocytes by differences in refraction than to rely on the staining qualities of these cells. In some cases THE BLOOD. 441 the methyl violet of the diluent colors some of the red cells so that they cannot be easily distinguished from the white ones. The number of cells in the blood is invariably reported as the number contained in a cubic millimeter of blood. In making the calculation of this number it is necessary to know the number of units counted, the number of cells in these units, the area of each small square of the unit, and the degree of dilution of the blood. Thus, if 16 units, each of 25 squares, have been counted, the total number of small squares is 400, each having a cubic area of 1 /4000 cmm. The total volume of the units counted is, therefore, 1/10 cmm. It is evident that the number of cells in 1 cmm. of blood is 10 times that in the area counted over if the blood were undiluted. But the count is always made with diluted blood and we must, therefore, take this factor into account. With a dilution of 100 multiply the number of cells in 1 cmm. by 100, and with a dilution of 200 multiply by this factor. Thus, if 2,500 cells were counted in the 400 small squares and the cubic contents of the units gone over was 1/10 cmm. then 1 cmm. of diluted blood would contain 25,000 cells. As the blood was diluted 200 times, the total number of cells in one cmm. of undiluted blood is 5,000,000. A very simple way of remembering this calculation is to multiply the number of cells counted in the total area of 400 small squares by 1,000 if the dilution was 100, and by 2,000 if the dilution was 200. If the total area of 16 units be not counted, the method of calculation is the same but the factors are variable. This method goes as follows: Multiply the number of cells counted by the degree of dilution and this result by the cubic contents of each small square (4,000). Divide this result by the number of small squares counted. Thus, to calculate the number of cells in a cmm. of blood when 100 small squares (4 units) were counted at a dilution of 200, , . . , „ . . . ... 625 x 200 x 4000 the count being 625 cells, the equation is as follows: -- = 100 5,000,000. Leucocytes. In counting the leucocytes much depends on the sort of ruled slide at the disposal of the worker. With the old Thoma chamber at least five different drops must be examined in order that a sufficiently large number of leucocytes may be counted, while with the Turk cell a counting area of 9 sq. mm. is afforded for each drop. The more leucocytes counted so much the less is the error. It is usually sufficient to count the white cells in a single drop, using the Turk chamber, but for scientific purposes three or even four drops would better be examined. If it is desired to count the leucocytes in the same specimen as the red cells the procedure is as follows: Prepare the drop of blood exactly as described for counting the red cells, using Toisson's fluid as the diluent and the erythrocy- tometer as the diluting pipet. After the reds have been counted, enumerate the whites in the entire ruled area of this chamber. In this process the leucocytes will be found to be stained a faint blue. It is often advisable, in case of low 442 DIAGNOSTIC METHODS. leucocyte counts, to repeat this process with a second drop of blood. The calculation by this method is very simple. As the entire ruled area of the Turk chamber covers a surface of 9 sq. mm., each equal to the central area used in counting the red cells, we have the equivalent of 3,600 small squares in the ruled surface. Multiply this figure by the number of drops used to obtain the total number of small squares covered by the count. Thus, if 54 leucocytes were observed in two drops (7,200 small squares) and the dilution was 200, , . . 54 x 200 x 4000 , ,, then we have the equation, = 6,000 cells. 7200 A second method of calculating the number is to consider each sq. mm. of the surface of the Turk chamber as a unit. If, then, the number of cells counted in two drops (18 units) be 54, we divide this number by the number of units counted, 18, and multiply the result by 10 (the cubic contents of each . . , , , ..... - 54 X IO X 200 unit) and then by the dilution. Thus, = 6,000. It is usually preferable in counting the leucocytes to use the special leuco- cytometer previously described, as this gives a smaller dilution, and consequently a larger number of leucocytes to the counting surface. In using this pipet, the blood is drawn to the mark 1 and the diluent, 1 per cent, acetic acid, added to the mark 11. This diluent destroys the red cells and brings out the leucocytes clearly. The dilution of the blood will thus be 10. If a large increase in the number of leucocytes is anticipated, it is better to use a dilution of 20, drawing the blood to the mark 0.5, as a routine the dilution of 10 being, however, preferred. In counting the white cells with a dilution of 10, using the Turk chamber, it is not as a rule necessary to go over the entire counting field, but more accurate results will obtain if this be done. A count of at least 250 cells is advisable, while one of 1,000 is more to be preferred in scientific work. The method of calculation is the same as given above. Thus if 540 cells were counted in the 9 sq. mm. of the ruled surface, the dilution being 10, we have 54° x 10 x 10 = 6,000. If we have not used the total 9 sq. mm., but have used 4 or any other number which may be considered sufficient by the worker, the divisor in this division will be the number of units counted. If only a Thoma chamber be at hand for counting the cells, two methods are available. By the first, several drops may be gone over counting the total number of cells in the ruled sq. mm. of the cell. In the second method we find the cubical contents of each visual field and then count the leucocytes in many of such fields. The method of computing these factors is given by Stengel1 to whose work the author refers the reader who may not have access to a Zappert or a Turk chamber. In making this leucocyte count great care must be exercised to have the diluent fresh and free from yeast spores, which so freely develop in such mixtures. If this factor be observed, all the cells seen may be counted as 'Twentieth Century Practice of Medicine. New York, vol. 7, 1896, p. 271. THE BLOOD. 443 leucocytes, but occasionally nucleated red cells may be confusing, especially if these be present in large numbers. The physiological condition of the patient should always be considered in making a report on a leucocyte count, as such factors as digestion and exercise influence this count to a great extent. The normal error in making a leucocyte count, with a-count of 200 and more leucocytes, is about 5 per cent., while in the case of a red count it should not be over 3 per cent,. Careful work with special attention to all the details mentioned will often reduce this error to a lower figure. Naturally, the error in counting the leucocytes will be much reduced by using the Turk chamber and giving the blood a dilution of 10. There are certain errors due to faulty construction both of the pipet and of the counting chamber which remain constant in the same apparatus. Hence it is wise to procure *he very best equipment possible and to test the different portions of the pipet for such errors. In this way a very appreciable difference may be obviated. Not all of these blood counters show such variations, but some do, and it is, therefore, a matter of moment to know your tools. (5). Cleaning the Apparatus. This is the last step in the technic of making a blood count. While perhaps not as important as some of the other steps, yet if not properly carried out it will introduce errors which may prove very annoying. It is readily seen that an unclean pipet or counting chamber will interfere with the proper manipulations as described above. After the cover-glass has been removed from the slide, wash out the chamber with distilled water and dry thoroughly by means of a clean piece of linen. Place the slide in its proper receptacle, so that it may be conveniently found when desired. This may seem a small point, but the author has seen too many slides lost by carelessness in putting them away. The cover-glass is rubbed clean and dry and put in the case with the slide. Wash out the pipet with water until all of the blood is gone. In some cases fine clots will be observed sticking to the side of the tube. Under such conditions remove the adherent blood by means of a fine wire or, if this is not effective, draw a little strong potassium hydrate solution into the capillary and allow it to act on the clot. After the tube is apparently clean, wash it out with alcohol and ether. Blow a stream of air through the tube by means of compressed air or the suction pump. Be certain that the apparatus is perfectly dry and clean, and that the glass bead in the ampulla is freely movable as the tube is shaken, before putting the pipet in its case. Recently Durham has introduced a modification of the older instruments for blood-counting. This embodies the principles of the various methods, but substitutes a self-measuring capillary pipet for the suction pipet of the Thoma apparatus, and special mixing vessels for the dilution of the blood. This device makes it possible for one inexperienced in blood-counting accu- Durham's Hemocytometer. 444 DIAGNOSTIC METHODS. rately to measure the blood and diluting fluid and thus eliminates the error possi- ble with the older pipet. The direct count is made in the counting chamber of Thoma. This capillary of Durham is more easily cleaned than that of Thoma, thus giving an advantage in cases where several blood examinations are to be made in a limited time. While this method of diluting the blood has the advantages above mentioned, the writer has never been able to convince himself that the use of this apparatus gives the student any better results than Fig. 136.-Cross-section of Durham's Blood Pipet. (Da Costa.) T, Glass tube; N, rubber nipple; p, lateral perforation in nipple; c, cork in which a capillary pipet is fitted. he obtains with the Thoma instrument. After some experience, and this is needed in any method, the dilution of the blood with the Thoma pipet is quite as easily made as with the Durham modification, and the results of the count give quite as close checks as do those with the newer method. Oliver's Hemocytometer. This instrument was intended to furnish a more accurate method of count- ing the red cells than was given by the older instruments. The method is based on a principle entirely different from that of the older instruments and does not afford an actual count of the cells. If blood be diluted with a fluid, which preserves the corpuscles, in a rectangular test vessel composed of longitudinally striated glass, each striation of the glass will act as a lens project- ing an image of a candle flame viewed through the sus- pension of opaque particles of the blood, providing the suspension is of a sufficient dilution to permit of the almost unobstructed passage of the rays of light. At the proper dilution, these images of the candle flame will form a bright streak horizontally across the tube. Ex- periments have shown that the development of this bright line, on dilution of the blood with Hayem's solution, is an accurate measure of the percentage of red cells in the specimen examined. The dilution of the blood is made in a rectangular glass cylinder by means of a capillary pipet, which is washed out with Hayem's solution. The cylinder is graduated into divisions from io to 120, each division representing 50,000 red corpuscles. As this instrument has little clinical value, owing to the fact that the error is very great in cases in which the blood is diseased, I refer the reader to other works for a description of the method in detail. In the study of the physiologic variations of the red cells, this method affords very accurate results Fig. 137.-Oliver's Hemocytometer. (Greene.) THE BLOOD. 445 giving figures which would be lost with the Thoma instrument. It must be remembered that, in this method, ^s in many others the personal equation plays a large role and may account for serious error which is not found with the Thoma instrument. Emerson has shown that the variations in the results may be as much as 2,000,000 cells when this instrument is compared with the Thoma in counting the blood in primary anemia. Baumgarten has proven that varia- tions in the size of the cell as well as deformities of the cell will introduce a serious error into this method. Ellermann and Erlandsen1 as well as Dunger2 have recently introduced new methods, for making accurate counts of the white cells, which may prove very advantageous. Counting of the Blood-platelets. The technic of counting these cellular elements has been imperfect and the results variable. The methods advanced are both direct and indirect. The technic of Determann and of Brodie and Russell belong to the latter class, while the method of Helber is a direct one. Recently Wright and Kinnicutt2 have introduced a method which is simple, exact and reliable. The technic is as follows: The blood is diluted 1:100 by means of the pipet used for counting the red cells and the counting is done in the Thoma-Zeiss counting-chamber, using all the precautions previously discussed. The specially thin cover-glass of Zeiss, with central excavation, is used to render the platelets clearly visible. The diluting fluid consists of 2 parts of a 1:300 aqueous solution of "brilliant cresyl blue" and 3 parts of a 1:1,400 aqueous solution of potassium cyanid. These two solutions should be fairly fresh, kept separate, mixed and filtered just before taking the blood. After the counting-chamber is filled, it is left at rest for 10 to 15 minutes in order to allow the blood-platelets to settle to the bottom of the chamber and be more easily and accurately counted. The platelets appear as sharply outlined, round, oval or elongated lilac-colored bodies, some of which form a part of small spheres or globules of hyaline sub- stance. The red cells are decolorized and appear as "shadows," while the nuclei of the leucocytes are stained a dark blue and their protoplasm light blue. This method shows a normal platelet count of 225,000 to 350,000 per cu. mm. No constant relations seem to obtain between the variations in the number of platelets and of the leucocytes. According to Determann, the ratio between the red cells and the blood plates is, on the average, 22:1. The pathologic variations in the number of these cellular elements will be discussed in detail in a later section. III. Morphology of the Blood. Before any examination which is concerned with the study of the morpho- logical characteristics of the blood can be made, it is essential that all of the glassware which comes in contact with the blood should be absolutely clean and dry. The glass slides as they come from the shops, are often coated 1 Deutsch. Arch. f. klin. Med., Bd. 98, 1910, S. 245. 2 Munch, med. Woch., Bd. 58, 1911, S. 1131. 3 Jour. A. M. A., vol. 56, 1911, p. 1457. 446 DIAGNOSTIC METHODS. with substances which are removable with difficulty. Moreover, these slides are not in all cases perfectly level on both surfaces. It will need but one ex- perience with an uneven slide to convince the worker that it is a loss of time to attempt the use of such slightly convex or concave slides. The cover- glasses should be of the very best quality of glass, should be as thin as possible (number o), and three-fourths inch square. The seven-eighths inch square covers as also the larger rectangular ones are not as desirable for blood work, especially in the examination of fresh specimens. The slides and covers should be cleaned with soap and water followed by water and alcohol. In some cases it is necessary to soak them in concentrated hydrochloric acid for some hours and then wash with water, alcohol, and ether, or the ordinary acid-alcohol may be used. After being cleaned they should be kept either in 95 per cent, alcohol or, preferably, polished with a clean linen cloth or a piece of tissue-paper and kept in dust-proof receptacles. It is a wise precaution invariably to polish the slides and covers before use, as dust particles are prone to collect even under the best conditions. As a rule, it is better to use only new Cover-glasses and not attempt to clean them after use. The slides may, however, be cleaned by boiling with a strong alkali solution, washing with hot water acidified with hydrochloric acid, then with hot water, alcohol, and ether. After the cover-glasses have been polished it is the best practice to handle them only with forceps, as moisture is almost certain to collect on them if the fingers be used. This is not only better technic, but more rapid work may be done with their use. Two kinds of forceps are necessary in such work. The first is one for holding the cover firmly, being found as the locking forceps of Ehrlich or the cross-point forceps, while the second is the ordinary pinch forceps with which the second cover-glass is handled in making smears. (i). Examination of Fresh Blood. The examination of fresh blood is a very important part of hematological work and should be a routine procedure in every case possible. If the blood cannot be examined for several hours after being taken, it is wise not to attempt the study of a fresh specimen, as so many changes will occur in such slides that no certain findings obtain. The information obtainable from such examin.ation of fresh blood often supplements that which one may derive from a study of the stained specimens, as some peculiarities, such as the ameboid movement of the leucocytes or the motility of the malarial parasite, maybe studied only in this way. Technic. Assuming that the slides and cover-glasses are clean and dry, the ear is punctured as previously described. Wipe away the first few drops of blood and touch the center of a cover-glass, held with the pinch forceps, to the top of the next drop, which should be about the size of a small black-headed pin. If this drop be too large the layer of blood will be too thick to permit of proper examination. Care should be taken that the cover-glass does not touch the PLATE XVI. Fresh Normal Blood. (Zeiss Ocular 4, Objective DD.) THE BLOOD. 447 skin. Drop this cover onto a slide, which has been wTarmed by rubbing or by passing through a flame. If the glassware be clean, the drop will spread evenly in a thin circular layer, not quite to the edge of the cover-glass. Under no circumstances should pressure be used to thin the layer or to readjust the cover after it has settled on the slide, as artefacts may be easily introduced in this way. The slides thus prepared are examined first with a low-power lens to obtain an idea of the even distribution of the cells over the entire area. The detailed study is carried out with the 1/12 oil immersion lens, but it should be remembered that a smaller magnification may give a better general survey. These preparations will keep long enough for the purposes of exami- nation, but if one wishes to preserve the blood fresh and uncoagulated for a longer period it is well to enclose the cover-glass with vaselin or paraffin or to use the ordinary hanging-drop chamber as suggested by Rosin and Bibergeil. It is sometimes desirable, especially in the study of malarial parasites, to use a warm stage or a warm chamber. If, however, the specimen is examined soon after its preparation, no such precaution is necessary, provided the room is not cold. In order to judge of the changes which abnormal blood may show in the fresh state, one must be thoroughly familiar with the appearance of normal blood. This latter knowledge can be obtained only by frequent study of fresh normal specimens and not from any text-book description. To attempt to learn without microscopic study the size, shape, color, and refraction of the red and white cells, the relation of the blood-plates to fibrin formation, the number of the various cells and their ratio to one another, would be absolute idiocy. An examination of the fresh blood as described above gives information regarding the presence or absence of the malarial parasite, the spirochete of relapsing fever, the filaria, and trypanosomes. It affords evidence of increased or decreased rouleaux formation, number, deformities, and degenerations, as well as of the amount of hemoglobin of the red cells; the presence of a leucocytosis or of a leucopenia and of ameboid movement of the leucocytes. However, care must be taken to avoid premature conclusions from such study and to institute further examinations of the stained specimen to clear up doubt- ful points. The observer must be on his guard lest he mistake the normal Brownian movement in the protoplasm of the cells for ameboid or parasitic movement. Curious phenomena are observed in the fresh specimen as the blood dries and should not be misinterpreted. The various characteristics of fresh blood will be taken up in detail later. The introduction of the ultra- condenser or dark-field illuminator has opened up a field of great possibilities in the examination of specimens of fresh blood, especially when malarial parasites or spirochsetae pallidae are suspected. (2). Preparation of Smears. To prepare blood smears, which are to be later examined in the stained condition, one may spread the blood in capillary layers on slides or between 448 DIAGNOSTIC METHODS. cover-glasses. The former method is the one used in the writer's laboratory and has given excellent and satisfactory results. A fair-sized drop of blood is collected on one end of a clean dry slide, held between the thumb and second and third finger of the left hand. A second slide is held in the same manner by the right hand, but at an angle of 45 degrees to the first one and touching the drop of blood. Allow the blood to spread out by capillarity along the edge of the second slide. As soon as this occurs, draw the drop of blood along the first slide with a clean sweep, exerting little pressure with the second slide and maintaining the angle of 450 between the two slides, allowing the second slide to rest rather upon the blood than upon the slide (see cut). In the Fig. 138.-Preparation of smears with two glass slides. {Da Costa.) process, as recommended by some writers, the second slide is gradually drawn into a position perpendicular to the first one. This procedure does not yield, in the writer's hands, as good results as the former method, as it is more difficult to maintain equal pressure, the smear being as a result too thick or too thin in places. Instead of a slide, a cigarette paper may be used as a spreader and gives good results. This method of making blood smears has the advan- tage of offering a large surface for examination, of making smears which are fairly uniform after some practice, of dispensing with the necessity of mounting the specimen, and of permitting the fixation of the smear in the free flame. It is less expensive than the method to be described later and permits of the Fig. 139.-Preparation of blood smear with cigarette paper. {Da Costa.') cleaning and later use of the slides. The beginner may find, on examining his early specimens made by this method, that the leucocytes collect at the distal end of the smear and that the general surface contains few white cells. This is due to the use of undue pressure in making the smear, and may be avoided by proper attention to this detail. A second method which has many advocates is the use of two cover- glasses. One clean, dry cover-glass, which should not be too large (preferably three-fourths inch square or the larger rectangular slips), is held by the Ehrlich or cross-bladed forceps or, as some advise, between the thumb and first finger of the left hand. The other cover, held in the pinch forceps or between the THE BLOOD. 449 thumb and first finger of the right hand, is touched to the drop of blood as it wells from the puncture in the ear. This second cover is then dropped at once upon the first in such a way that the corners of the two glasses do not coincide. If the glasses are clean the blood spreads out evenly in a thin capillary layer be- tween them. As soon as the spreading is complete, the two covers are drawn apart, in a line parallel to the plane of their surfaces, by a steady, quick motion, being sure to avoid lifting them apart. This manipulation can be learned only by practice and never from any description. If the fingers are dry and if care be taken to touch only the corners of the covers the forceps need not be used in separating the covers, but it must be remembered that moisture will Fig. 140.-Ehrlich forceps cause changes in the specimen. It is, therefore, advisable to use the forceps in this part of the technic unless the rectangular slips be used. As soon as the covers are separated, they are allowed to dry in the air or by waving them two or three times to and fro. They should be at once placed in a clean, closed receptacle, such as a Petri dish, until ready for the later complete fixation and staining, as dust will collect upon them or flies may attack them if left in the open air. It is rarely necessary to fix these smears at once, but with some stains such treatment is advisable. This second method of making the blood smears is more difficult than the first, is not so reliable, does not give as great a surface for examination, and always shows the lower cover-glass better and more uniformly spread than the upper. Fig. 141-Pinch forceps. Smears made by either of these methods should be uniform throughout with the exception of the edges, which should never be used as they are too thick for allowing definite conclusions to be drawn. The red cells should lie on their broad surface, should not be in rouleaux except at the edges, and should not show deformities due to errors in technic. The leucocytes frequently collect at the edges of the specimen if too great pressure be used in making the smear, while the platelets always collect at the point first touched by the second slide or cover-glass. The preparation of thin even smears is necessary for the proper carrying out of the later technic. Those specimens which are irregular or are too thick would better be discarded, as the time 450 DIAGNOSTIC METHODS. consumed in studying such specimens will not be compensated for by the results obtained. It is much better to make several smears than to be content with a few bad ones. A little experience with poor smears will convince the worker that it is advisable to use great care in preparing them, especially if a differential count is to be made or if a study of the degenerations and deformi- ties of the cells is to be undertaken. (3). Fixation of Smears. Before any staining of the cellular elements of the smear takes place, the protein constituents of the blood must be coagulated by exposing the air-dried film to the action of a high degree of heat or to that of various chemical reagents. The selection of the method of fixation will depend to a great extent upon the stain to be used later. Fixation is always essential if aqueous stains are used, while it is not so necessary if strong alcoholic solutions are employed. In the use of the different modifications of the Romanowsky stain, the fixation is done by the methyl alcohol employed as a solvent for the various stains. As a general rule, fixation by heat is preferable to that by chemicals, as artefacts are less prone to appear, providing the degree of heat is carefully regulated. Fixation by Heat. This method, which is the most difficult to use and which is at the same time the best, is the only one which is reliable when Ehrlich's triple stain is to be employed. The principle is as follows: The air-dried specimen is subjected to the action of a temperature of no0 to 1500, for a more or less varying length of time, depending on the experience of the special worker. The lower the temperature the longer must its action be exerted. The apparatus most frequently employed is the copper plate introduced by Ehrlich. This is an unpolished triangular plate of copper about 3 mm. thick, 30 to 50 cm. long, and 10 cm. wide, which is held in position by vertical standards. It is heated by an alcohol lamp or a Bunsen burner placed under the narrower end until the temperature of the plate becomes constant, the parts nearer the flame being naturally warmer than those more remote. The temperature of the different portions of the plate may be readily found by determining the points at which water (100), toluol (no), xylol (140), or of turpentine (150) boil. The slides or cover-glasses are then placed, smeared side up, at the desired point (the outer margin of the glass being three-fourths inch from the boiling-point and toward the flame). Just how long a period is necessary, at the temperature selected, to give a perfect fixation, will depend upon the age of the specimen and upon the condition to be studied. It is good practice to place several slides at the point desired for a period of one hour and then remove a slide at intervals of 15 minutes thereafter. One of the specimens is sure to be good by this method and the remaining ones may, therefore, be properly heated. Freshly made specimens require longer heating than the old ones, while normal blood requires a longer exposure than abnormal specimens. As a rule, the specimens require, when the triple stain is to be THE BLOOD. 451 used (and this is the one most frequently employed with heat fixation), from one to one and one-half hours at a temperature of i io, although some workers use only a few minutes' (one to three) exposure to such temperatures. The higher the temperature the less time is essential. Rubinstein uses a point at which a drop of water does not boil, but assumes the spheroidal state (so-called Leidenfrost phenomenon) and places the slides, with the smeared side down, upon the plate at this point for one-half to three-fourths minute. Some workers, as Pappenheim, use this point, but place the smeared side upward. Instead of the copper plate, one may use the ordinary drying oven or a Victor Meyer heater. It is the writer's custom to use a cop- per drying oven heated by a gas flame regulated by a thermostat. The slides, with the smeared side downward, are placed on a glass plate whose temperature is measured by an accurate thermometer. The temperature is allowed to increase gradually to about 8o°, from which point it is more quickly raised until the desired stage is approximated, when the heating must continue slowly, the final temperature being maintained for 15 minutes. The best temper- ature for fixation, in the writer's experience, is no° to 1200 for 15 minutes when the staining is to be done with eosin-methylene blue or eosin- hemotoxylin, while for the tri-acid stain a temperature of 1200 to 1250 for one and one-half hours should be maintained. Some workers heat to 1600 rapidly and then allow the films to cool to 300, when fixation is complete in about 15 minutes. Engel and Cabot recommend, in the absence of other equipment, the passing of the smear through the flame several times. Such treatment often yields excellent results, but is uncertain and requires much experience. It is to be remembered that too rapid changes of temperature are to be avoided, as shrinking or splitting of the cells will occur under such conditions. So much depends upon proper fixation that a little more time spent in obtaining good specimens will shorten the time necessary for the future examination. Proper staining depends on proper fixation, especially with the tri-acid stain, in which cases the over- or under-heating is evident in the color tone of the erythrocytes. Fig. 142.-Oven for fixing blood films. (Da Costa.) Chemical Fixation. (i). Absolute Alcohol. Allow the alcohol to act for five minutes to one hour, or boiling alcohol for one minute. The slide is simply covered with the fixative and left for the desired period. Much depends upon the stain to be used as to the time neces- sary for good fixation. If an alcoholic stain, five minutes is sufficient; if a 452 DIAGNOSTIC METHODS. watery or alkaline one, a longer time is essential. After fixation drain off the alcohol and allow the residue to evaporate in the air or wash with water and dry between sheets of filter-paper. If alcoholic stains are to be used the moist specimen may be directly passed through the flame. This fixative is unreliable if a study of the neutrophile granules is to be made, but it brings out the nucleas structures fairly well. (2). Nikiforoff's Method. Cover the smear with equal parts of absolute alcohol and ether and allow the fixative to act from one-half to two hours. After the fixation allow the fixative to evaporate or wash with water and dry. Some workers advocate short fixation especially where the malarial organism is to be studied. This fixative brings out the degenerations of the red cells in good shape. (3). Methyl Alcohol. This fixative is used absolutely pure for three to five minutes. It is the most generally applicable chemical fixative and gives beautiful specimens. If used in combination with stains, as in the Romanowsky methods, it gives as perfect preparations as absolute alcohol in one-half hour and brings out the neutrophile granulations in better outline. A longer fixation than three to five minutes does no harm; and a shorter one, especially if the fixative is the solvent for the stain, will give fair results, the outline of the cells being sharper the longer the fixative is allowed to act. (4). Formalin. This fixative may be used as a i per cent, solution in 95 per cent, alcohol and allowed to act for one minute, as Benario advocates, but the writer has had much better success with the 0.25 per cent, solution in 95 per cent, alcohol for one minute, as Futcher and Lazear suggest. Allow the fixative to act for one minute, wash in water, and dry between filter-paper. Some writers advocate the direct treatment of the fixed smear with the stain without an intermediate washing and drying, but I have never found the results as good by this method. Instead of the formalin solutions, the vapors may be used by placing the specimen under a bell jar with a few drops of 40 per cent, formalin and allowing the fixative to act for one to five minutes. The normal staining of the cells is not always as clear as could be desired after the use of such a fixative. A large number of inorganic fixatives have been advised, among them being mercuric chlorid, chromic acid and osmic acid, but these fixatives are much more apt to produce artefacts than are the others mentioned above. While these latter fixatives give good demonstrations of the nuclear structures and of mitotic figures, the granules are very imperfect; but chromic acid has many advantages as a fixative in the study of the chromatin elements. THE BLOOD. 453 (4). Staining Methods. G4). General Considerations. Since the work of Witt, we have recognized that the color of an organic sub- stance is due to the presence of two definite atomic complexes in the molecule of the colored substance. The first of these, the chromophore group, is responsible for the chromogenic properties of the substance, while its influence as a dye is increased by the presence of the second or auxochromic group. The color of the compound is the more intense the more of these groups are present. Groups or atoms which intensify the color of the substance are called, by Schutze, bathochromes, while those which reduce the color are called hypsochromes. While the dyes, the so-called anilin dyes, used in hematological work are all aromatic derivatives, it must not be assumed that such are alone characterized by staining qualities, as many simple aliphatic derivatives show a definite color and exert some staining property, depending onthe presence of the two groups above mentioned. The chief chromophore groups are the CO (carbonyl) group, the CS group, CN, HCN, the - N = N-(azo group), - N. /° the | /O (azoxv group), NO (nitroso group), -| (nitro group), -NZ ' XO and the - N = SO group. The most important auxochrome groups are the NH2 and the OH groups, the former being a much more powerful one than the latter. These auxochrome or haptophore groups make possible the union of the stain with the tissue as a direct salt formation. Substances, which possess a chromo- phore group and are colored or intensified by the presence of the auxochrome group, are called chromogens. It is, therefore, evident that the effect of an auxochrome depends not only upon its own nature, but also upon that of the chromogen. Through the introduction of H by means of hydrocarbon radicals, new auxochromes are produced and the color becomes intensified, the effect being so much the greater the higher the molecular weight of the substituting hydrocarbon radical. As a large number of colored substances have a tendency to form tautomeric compounds, the salt formed by combination with the auxochrome groups may have a different constitution from that of the free base or acid. Many chromophore groups are also capable of forming salts, but only with strong acids or bases. In this formation of salts, by union with the auxochrome or chromophore groups, we must remember that halochromia may be seen. By this is meant that uncolored or weakly colored substances may combine with acids to form salts without the color being due to the presence of a chromo- phore group. It is essential that the chromophore group possess a double bond of union, as oscillation in one portion of the molecule is thus possible. As the chromophoric as well as the auxochromic radicals may have acid or basic tendencies, it is manifest that the reaction of the substitution product will depend upon the interrelations of the acid and the basic radicals. The best ayes are, therefore, obtained by substitution in one direction, that is, by accumulating either basic or acid groups separately. 454 DIAGNOSTIC METHODS. It is thus seen that we have two general classes of dyes, the acid and basic, depending upon the preponderance of the total acid or basic groups in the molecule of the dye. It has been shown by Ehrlich that these acid and basic dyes, may be so combined that a third one results, showing neither acid nor basic properties. This class of dyes, known as neutral dyes, is of the greatest importance in hematological work. In them we have not only the staining properties due to the original chromophore and auxochrome groups of both the acid and basic dyes, but also those due to the union of the component groups in the neutral dye. Such dyes are hence called polychrome dyes and are usually soluble in an excess of one of the component mother dyes, generally the acid one. It must be understood that in speaking of a dye as acid, basic, or neutral, we do not refer so much in our staining work to the chemical reaction of the dye, but to the portion of the dye to which the staining is due-that is, to the specific auxochrome and chromophore groups. Among the basic stains we find methyl green, methylene blue, amethyst violet, neutral red, dahlia, pyronin, thionin, fuchsin, methyl violet, Bismarck brown, alum hematoxylin, and safranin. Most of these stains, depending on the strength of their component groups, color the nuclear substance of both red and white cells, as also the cytoplasm and certain granules of abnormal red and of some normal white cells, the staining being influenced, as Matthews has shown, by the alkaline reaction of the tissues. The most important acid stains are eosin, acid fuchsin, orange-G, indulin, nigrosin, aurantia, and salts of picric acid. These dyes color the red cells and the eosinophile (oxyphile) granulations of the leucocytes. The neutral portion of the dyes color the so- called neutrophile granules of the leucocytes. In the process of staining, it is a question whether we have to do with purely chemical processes as Knecht's theory assumes or whether the solid-solution theory of Witt or the mechanical theory of surface attraction are accountable for the phenomena observed. It is probable that the salts of the dyes are de- composed by the cells and that new compounds result from the union of the acid and basic stains and the various chemical entities of the cells. Yet we have instances in which the dye is simply stored in the cell without any chemical union taking place. We must, however, account for the elective character of certain stains by a purely chemical activity, as, for instance, chromatin, which undoubtedly consists of nucleinic acid, always takes a basic stain, even though a neutral compound is used as the staining agent. It is to be remarked that a neutral stain does not color all of the acidophile or basophile substances of the cells of the same tint. Thus, eosinophile granules are differently colored from the oxyphile hemoglobin of the red cells. The neutral portion of the stain has nothing to do with the staining process beyond the coloring of the neutrophile substances in the protoplasm of certain leucocytes. In the selection of a stain for general hematological work, it is necessary to use a compound stain or, at least, two simple stains, one after the other A single simple stain colors only a few of the elements and affords no general THE BLOOD. 455 idea of the blood picture. Indeed, to obtain definite conceptions of the finer structure of the blood, it is necessary to study several slides stained by different methods. It is usual to select for routine work a stain which will reveal the greatest amount of information regarding the histological characteristics of the blood. This is the so-called panoptic staining and is to-day generally accomplished by the use of the various forms of neutral stains. In making the stains for one's self or in buying them in the market, one must be certain that only chemically pure pigments are used and that the solution is made accord- ing to the formula with the purest solvent obtainable. It will need only one experience with a poorly made stain to convince the worker of the importance of this detail. (B). Methods of Staining. In his work the author uses, for routine purposes, the Wright's stain, the eosin-methylene blue, and eosin-hematoxylin stains. These stains have the advantage of giving clear-cut pictures of practically all of the important blood elements along with simplicity of technic. As it is not always possible to secure or even to make a Wright or an Ehrlich triple stain which will give reliable results under all circumstances, the writer feels that the general worker would better use the eosin-methylene blue stain for his daily work. The many modifications of the Romanowsky stain have their advocates, and it must be said that they yield reliable results when they are working properly, but no one can say when they will go wrong or how to make them work right when once they do give poor results. Eosin-Methylene Blue. A number of methods of using these two simple stains have been advocated by such workers as Chenzinsky, Ehrlich and Lazarus, von Willebrand, Plehn, Aldehoff, and Gabritschewsky, but the writer has found the method advocated by Mullern1 to be the most generally applicable and, if properly applied, the most reliable modification. By it we are able to stain all of the blood elements, including the neutrophile granules and obtain, thus, a panoptic picture whose findings are not excelled by those of the various modifications of the Roman- owsky stain. Technic. (a) Fixation of the smear in pure methyl alcohol for three minutes. (5) Preliminary staining in 1/2 per cent, alcoholic (70 percent.) solution of Griibler's "french pure" eosin from three to five minutes, (c) Wash in distilled water and dry between filter-paper. (J) Lay the slide in a carefully measured and well-mixed solution of 20 drops of 1/4 per cent, aqueous solution of methylene blue (B. pat.) and ten drops of the above eosin solution for one- half to one minute, (e) Wash quickly and briefly with distilled water and dry at once between filter-paper or over the flame. (/) Mount if desired in Canada balsam or examine directly with a high-power lens. 1 Grundriss der klinischen Blutuntersuchung. Wien, 1909. 456 DIAGNOSTIC METHODS. This stain shows the red cells and the eosinophile granules of the leuco- cytes of a bright red tone, the neutrophile granules pink to bright red (distin- guished from the eosinophiles by their smaller size), the nuclei, mast-cell granules, bodies of the lymphocytes, platelets, malarial organisms, trypanosomes, and filaria varying shades of blue. The preliminary staining with eosin serves the purpose of bringing out the neutrophile granules more clearly while the basophile granulations are of course unaffected by such treatment. It is probable that the later staining with the eosin-methylene blue mixture has the same characteristics as the neutral stains to be mentioned later. While this stain has the advantage of simplicity, reliability, and panoptic power of staining, it is somewhat inferior to other stains in bringing out some forms of the malarial organism, owing to its lack of chromatin staining qualities. The writer has found that the eosin and methylene blue above mentioned are the best to use in this process and that Turk's advice, regarding the use of fairly fresh eosin solutions and of old methylene-blue solutions, is warranted. The blood prepa- rations should not be over a few days old to show the best pictures, as those over a week old may show a diffuse plasma staining and the neutrophile granules not clearly differentiated. If the nuclei do not stain well with the methylene blue as many writers claim, make a second preparation fixing somewhat longer in methyl alcohol. By following out these precautions the worker will be rewarded with beautiful specimens. This method is to be recommended for all routine examinations. This stain is especially important in cases in which the nuclear structures are to be studied. It stains the nuclei beautifully, showing their finer structure, karyokinetic figures, and pycnotic qualities, as well as the basophile granules of both red and white cells. The solutions required are: (1) 1/2 per cent. Griibler's blood eosin in 70 per cent, alcohol. We may use, with equally good results, the eosin mentioned in the previous method. (2) Delafield's hema- toxylin, the formula of which is Eosin Hematoxylin. Hematoxylin crystals, 4 grams Alcohol (absolute), 25 c.c. Ammonium-alum crystals C.P., 52 grams. Distilled water, 400 c.c. Glycerin, C. P., 100 c.c. Methyl alcohol, C. P., 100 c.c. Rub up the hematoxylin crystals with the alcohol until they are dissolved and place the solution in a loosely-corked glass bottle, allowing it to stand exposed to the light for four days. Dissolve the ammonium-alum in the water and allow it to stand exposed in the same way for four days. At the end of this time mix the two solutions, shake thoroughly, and filter at the end of three hours. Add the glycerin and methyl alcohol to the filtrate and allow this to THE BLOOD. 457 stand overnight. Filter the mixture, place it in a clear bottle, and allow it to ripen, exposed to the light for six weeks, when it is ready for use. Technic. Stain the specimen with the eosin solution for one-half minute, and wash in water. Without drying place the slide in the hematoxylin solution for one to three minutes, the time varying with the particular stain and with the experience of the worker. Wash with water, dry, and mount. This stain does not give as good results as does the former method, but is to be especially recommended when the nuclear structures are to be studied. Ehrlich's Triple Stain. In the literature of hematology we find the expressions "triacid" and "triple stain" used synonymously. The triacid stain, as originated by Ehrlich, was a mixture of equal parts of saturated solutions of indulin, nigrosin, and aurantia, and was used to differentiate the eosinophile granules. When used synonymously in these days, the triple stain is always meant. The composition of this latter stain is as follows, made up of two acid and one basic stain: Saturated watery solution of orange-G, 13-14 c.c. Saturated watery solution of acid fuschin, 6-7 c.c. Distilled water, 15 c.c. Alcohol (absolute), 15 c.c. Saturated watery solution of methyl green 00, 12.5 c.c. Alcohol (absolute), 10 c.c. Glycerin, 10 c.c. The pigments must be chemically pure and the solutions must be added in the order named, the methyl-green solution being added drop by drop with constant stirring. This stain is difficultly prepared, being usually a. failure. The writer would, therefore, advise the worker to obtain it already made; and even then it may not prove satisfactory. The correct stain should have a russet-brown color and should not show any trace of a precipitate. It improves up to a certain point with age but, after a time, precipitates invariably occur, making it useless. It should never be filtered or shaken, the solution to be used being taken from the center of the bottle with a dropping pipet. Technic. The smear should be fixed by heat as previously described. After proper fixation, cover the smear with the stain and allow it to act for one to ten minutes as the experience of the worker with the particular stain may indicate. Wash with distilled water, dry, and mount. One advantage of this stain is that it cannot overstain, those films appearing so being underheated, while those understained are overheated. 458 DIAGNOSTIC METHODS. This stain shows the red cells of a buff or orange color, without the slightest shade of red (a reddish tint is given with underfixed specimens while a yellow tone is shown by those over-heated), the nuclei of the leucocytes a dark green, those of the normoblasts black, the neutrophile granules a lilac color (though some occasionally show a reddish tinge), and the eosinophile granules of a crimson tone. This stain was introduced as specific for the neutrophile granules, but its disadvantages are too numerous to warrant its recommendation as a routine stain. It is a poor nuclear stain, does not show the structure of the normal mononuclear leucocytes, does not stain the basophile granules, nor the malarial or other parasites. For a reliable preparation, showing those features for which it is especially adapted, a proper fixation is an absolute essential. On account of the lack of nuclear staining with the triple stain, Pappen- heim has substituted methylene blue or methylene azure for the methyl green, and eosin for the acid fuchsin. The writer has had no experience with this modification and cannot, therefore, speak regarding its value. These stains are very numerous, each having its advocates. They are easy to use, contain a reliable fixative, and give satisfactory results, but are not always obtainable or easily prepared. They are the stains which give, perhaps, the best panoptic results and are especially serviceable in the study of the malarial organism and other parasites as many of them contain chroma- tin-staining elements. The granulations of the leucocytes are not as well marked in all cases as they should be, so that for a complete study of the various types of granules in the cells several stains would better be used. Romanowsky had found that the addition of a watery eosin solution to an aqueous methylene-blue solution, until an insoluble precipitate began to form, gave rise to new staining properties of the solutions, in the sense that the chromatin substance of malarial organisms was stained a beautiful red. The specific staining properties of this mixture were later found to be due not to a combination of eosin and methylene blue, per se, but rather to the formation of a new compound between eosin and an impurity or decomposition product in the methylene blue, namely, methylene azure. Jenner in his stain makes use of a methyl-alcohol solution of the isolated precipitate, the eosinate of methylene blue, which lacks the red chromatin staining element. The same may be said of the May-Grunwald stain. In making up these polychrome stains it is not general to use the pure methylene azure and eosin, but rather solutions of methylene blue containing a variable amount of the methylene azure to which eosin is added. Recently Wilson,1 in a careful study of the methylene-blue-eosin combinations, has shown that very little methylene azure and methylene violet exist in the stains as commonly employed. He finds evidences of at least four staining bodies in Polychrome Methylene-blue-eosin Stains. 1 Jour, of Exp. Med., vol. 9, 1907, p. 645. THE BLOOD. 459 such mixtures, namely, the eosinate of methylene blue, eosinate of methylene violet, eosinate of methylene azure, and eosinate of thionin. Wright's Stain. Preparation. To a 0.5 per cent, aqueous solution of sodium bicarbonate add methylene blue (B. X. or "medicinally pure") in the proportion of 1 gm. of the dye to each 100 c.c. of the solution. Heat the mixture in a steam sterilizer at ioo° C. for one full hour, counting the time after the sterilizer has become thoroughly heated. The mixture is to be contained in a flask, or flasks, of such a size and shape that it forms a layer not more than 6 cm. deep. After heating, allow the mixture to cool, placing the flask in cold water if desired, and then filter it to remove the precipitate which has formed in it. It should, when cold, have a deep purple-red color when viewed in a thin layer by transmitted yellowish artificial light. It does not show this color while it is warm. To each 100 c.c. of the filtered mixture add 500 c.c. of a 0.1 per cent, aqueous solution of "yellow- ish, water-soluble" eosin and mix thoroughly. Collect the abundant precipi- tate, which immediately appears, on a filter. When the precipitate is dry, dis- solve it in methyl alcohol (Merck's "reagent") in the proportion of 0.1 gm. to 60 c.c. of the alcohol. In order to facilitate solution, the precipitate is to be rubbed up with the alcohol in a porcelain dish or mortar with a spatula or pestle. This alcoholic solution of the precipitate is the staining fluid. It should be kept in a well-stoppered bottle because of the volatility of the alcohol.1 Technic. Cover the film with a noted quantity of the staining fluid by means of a medicine dropper. After one minute add to the staining fluid on the film the tame quantity of distilled water by means of the medicine dropper and allow she mixture to remain for two or three minutes, according to the intensity of the staining desired. A longer period of staining may produce a precipitate. Eosinophilic granules are best brought out by a short period of staining. The quantity of the diluted fluid on the preparation should not be so large that some of it runs off. Wash the preparation in water for thirty seconds or until the thinner portions of the film become yellow or pink in color. The experience of the worker may be such as to demand longer decolorization with any particular specimen. When the desired degree of differentiation is reached, dry the speci- men quickly between filter-paper, mount if desired, and examine first with the low-power lens to observe the staining effects, and then with the high-power lens for the more minute study. When searching for malarial parasites, the decolorization would better be of short duration, as the chromatin suffers to a great extent in this process. This stain shows the red cells colored either orange or pink (depending on the time of decolorization), nuclei of leucocytes blue or dark lilac, neutro- 1 Jour. A. M. A., vol. 55, 1910, p. 1979. 460 DIAGNOSTIC METHODS. phile granules lilac, eosinophile granules red or pink, fine basophile granules deep blue, large mast-cell granules purple, protoplasm of the lymphocytes robin's-egg blue, blood-plates deep blue or purple, bacteria blue, malarial and other parasites blue, the chromatin element varying from lilac to ruby- red to black. Polychromatophilia and granular degenerations are well shown, the granules being blue. This stain is very useful in studying lympho- cytes, mast cells, blood-plaques, and the finer structure of the malarial organism; but the leucocytic granules, at times, are not sufficiently differentiated. The writer has, however, found this stain very well adapted for making a differential count as it brings out, in clear relief, the most important characteristics of the various types of leucocytes. When a good Wright stain is at hand it can be recommended as a most serviceable one for routine work. Giemsa has shown that the complicated methods of preparing the dyes may be dispensed with by the use of the pure staining substance extracted from the polychrome methylene blue, namely, the methylene azure. If this pure pigment be used the stain becomes a pure chromatin one. It is advisable, therefore, to use Griibler's azure II, a combination of equal parts of methylene azure and medicinal methylene blue. The formula of the Giemsa stain is as follows : Giemsa Stain. Azur II-Eosin, 3.0 grams. Azur II, 0.8 gram. Glycerin (C .P.), 250.0 c.c. Methyl alcohol (C. P.), 250.0 c.c. Grind up the dyes in the alcohol and then add the glycerin. Fix the films in methyl alcohol, and stain for five minutes in a mixture of 14 drops of the stain to 10 c.c. of distilled water. If necessary, a trace of sodium carbonate may be added to the water to intensify the basic stains. Wash in water, dry, and mount. It is to be remembered that oil of cedar will bleach these specimens quite rapidly, so that it is advisable, if the films are to be kept, to stain them upon the slides so that mounting becomes unnecessary. Strong light should not be allowed to act on these stained films for any length of time as the chromogen stain fades rather quickly under these conditions. The various elements are stained as with the Wright stain, but the neutrophile granules are often not well defined.1 Technic. 1 Pappenheim (Technik der klin. Blutuntersuchungen, Berlin, 1911; Folia Haemat., Bd. 13, 1912,8. 339) advocates as the best and most serviceable stain the combination of the Giemsa stain as follows: Fix the blood film by covering it with May-Grunwald stain (Wright's answers as well) for 3 minutes, add to this stain on the slide an equal amount of distilled water and allow it to act for 1 minute. Pour off the fluid but do not wash. Now cover the stained film with dilute Giemsa solution (15 drops of stain to 10 c.c. of distilled water) and allow to act for 15 minutes. Wash thoroughly and dry, but not over the flame. Mount, if desired, in neutral Canada balsam. The result is extremely panoptic. THE BLOOD. 461 Specific Stains for Malarial Organisms. While the above stains all give good results with the malarial organisms, enabling one to make a diagnosis from an examination of the smear, yet certain other stains are often of advantage in that little else is stained and thus the confusion arising from indefinite staining is obviated. Add to 100 c.c. of 2 per cent, carbolic acid 20 c.c. of a saturated solution of thionin (Lauth's violet) in 50 per cent, alcohol, and allow this mixture to ripen for a few days. Fix the specimen by the formalin method given above and stain the smear for 10 to 15 seconds. Wash in water, dry, and mount. These preparations do not keep indefinitely, usually fading within a year. The plasmodia are shown as deep purple, irregular masses enclosed in the faint green red cells. The hyaline forms show as reddish-violet ring-like bodies. The basophile granules and the nuclei are the only other elements showing any particular staining qualities. Thionin Stain (Futcher and Lazear). This is essentially a Romanowsky stain and is made as follows: Add two or three drops of i per cent, aqueous eosin solution to 2 c.c. of water. To this diluted eosin solution polychrome methylene blue is added, drop by drop, until the red color of the eosin is only faintly visible (the polychrome methylene blue is a 1 per cent, solution of methylene blue chemically changed by heating with 1/2 per cent, sodium carbonate solution for two days at 550 C.). Nocht's Stain. Fix the film in methyl or ethyl alcohol and lay it face downward in the above mixture for five to ten minutes. Wash with water, dry and examine. The picture is the same given by Wright's stain, except that the nuclear trans- formations and the chromatin substance are better differentiated, being stained a bright red color. The many other stains for the examination of fixed specimens will be passed over, as the writer finds the above stains applicable to practically all routine work. Many writers prefer different stains from those mentioned, but these stains do not seem to have any advantages which would warrant their use to the exclusion of those given above. Special stains used for bringing out certain granules and deformities will be discussed under later headings. Technic. Although the examination of stained specimens of the blood is usually made with the dried and fixed smear, excellent results obtain when the fresh blood is stained without previous fixation. It is true that a "vital" staining of the blood-cells does not actually take place, as the dyes are decolorized by the reducing and oxidizing action of the living cells. However, a "post vital" staining, that is the staining of whole cells or portions of the cells, after their removal from the circulation and before the death of the cell results, may be accomplished in several ways. Vital Staining. 462 DIAGNOSTIC METHODS. We may either add to the fresh drop of blood a few crystals of the stain, as advised by Arnold, and note the staining of certain leucocytic granules and nuclei and recticular structure of many erythrocytes; or we may first dry a staining solution upon the slide, cover this dry stain with a drop of fresh blood, adjust the cover-glass, and seal this to the slide with wax. Some workers prefer to use the hanging-drop slide in the preparation of these fresh specimens. The stains which may be used for vital staining of the nuclei, granules, and plates are methylene blue, toluidin blue, thionin, neutral violet, Capri blue, Nile blue, brilliant-cresyl blue, Janus green, and paraphenyl blue. Of the proto- plasmic stains we have fuchsin, acridin red, pyronin, safranin, and neutral red. While the results derived from this method of staining are not as numerous as those of the more usual method, yet it affords much valuable information regarding the vital properties of the cells and regarding the normal structure and the circulatory changes of the cells.1 This method has been used extensively by Ito, Rosin and Bibergeil, Levaditi, Cesaris-Demel, and Pappenheim. Their results lead us to assume that further study will give us much valuable information concerning details of structure both of the cell and of the nucleus. It may be possible by this method to differentiate between certain forms of degeneration of the cell which are now known only indefinitely under the names of metachromatic and polychromatic staining.2 (5). Erythrocytes. (X). Appearance and Structure. In fresh, normal blood the red cells, or erythrocytes, appear as thin, flattened, homogeneous, biconcave, nonnucleated, discoid bodies with a sharply defined regular outline and a clear semitransparent center. In some cases they may appear distinctly cup or bell-shaped. The cells show, when examined singly, a pale greenish-yellow color but, when more thickly grouped, exhibit a reddish tint. The degree of color in these cells depends upon their hemoglobin content, the clear central area becoming larger and the entire cell becoming paler as the hemoglobin decreases. The loss of hemoglobin may be so great as to lead to the formation of the so-called "pessary" form, in which only the periphery of the cell is apparent. This central pale area varies much in individual cells and is not at all evident in those which are flattened out. In dry specimens, if thinly spread, the cells are circular, the normal biconcavity is obliterated, and a uniform stain is observed. If the smear be thick the central pale area is observed. In chlorosis and secondary anemias, the color of the cells is usually uni- formly paler than normally, while in pernicious anemias the color may be even deeper than normal. These are the usual but not invariable pathologic 1 See Luzzatto and Ravenna (Folia. Haemat., Bd. 13, 1912, S. 102) for a discussion of the numerical relations of the granular and nongranular erythrocytes as shown by this method. 2SeeRoss (Induced Cell-reproduction and Cancer, Philadelphia, 1911 and London, 1912) for the use of agar films containing stains and activating salts (auxetics) in this connection. THE BLOOD. 463 findings. In malarial conditions, discolored cells are often observed in the fresh specimens, the bronzed or "brassy" tone often drawing attention to the presence of a parasite of the estivoautumnal type or of the quartan form. The red cells show a marked tendency to cohere to one another in more or less regularly arranged piles, forming long rows (rouleaux), like rolls of coin piled up face to face. The exact cause of this phenomenon is unknown, but it may be dependent on the presence of the fatty membrane surrounding the cell, as Peskind's findings show that the red cells are enveloped by a layer com- posed of lecithin, cholesterin, and a nucleoprotein.1 Incertain pathologic condi- tions this normal rouleaux formation is increased and in certain ones is de- Fig. 143.-Normal blood showing rouleaux formation and fibrin network. {Da Costa.) creased. The diminished rouleaux formation is observed in conditions associated with increased viscosity of the cells, as observed in most inflammatory diseases and in the anemias due to malignant disease. This hyperviscosity of the cells has not at present much clinical significance, but may be shown to be of importance when our knowledge concerning the normal viscosity of the blood becomes more extensive. « The structure of the erythrocytes is still an unsettled point in hematology. Neither membrane nor stroma have been fully demonstrated yet after the hemoglobin has been removed from the cells by hemolysis, a stroma may be definitely seen which could hardly be called an artefact. Schafer assumes that the hemoglobin is held in firm combination by chemical union with other albuminous constituents of the cell and supported by a stroma similar to that of Peskind's outer membrane. That some sort of an outer membrane does exist would seem to be proven by the experiments on hemolysis which have 'See Wiltshire, Jour, of Path, and Bact., vol. 17, 1913, p. 282. 464 DIAGNOSTIC METHODS. shown that many substances penetrate the cell producing hemolysis, while others in the same concentration have no such effect. It is hard to believe that a selective vital activity is at work here, as the results follow too closely the laws of physical chemistry as applied to diffusion and osmosis. The corpuscles are very elastic and contractile so that rapid and marked temporary distortions of shape are possible under the influence of variations in the composition of the circulating plasma. As the blood dries various changes in the appearance of the red cells are observed. These changes, known as crenation, are due to the evaporation of water and depend upon the quantity of air which comes in contact with the specimen as well as upon the length of time this influence acts. "The develop- ment of one or more small, bright, highly refractile spots in the body of the cell and a slight indentation of the periphery of the cell are the most conspicuous indications of beginning crenation. As the process goes on, more and more of these hyaline points develop, until finally the whole surface of the corpuscle becomes thickly studded with glistening bead-like spines. As the stroma becomes drier and drier, its typical biconcavity and sharply-cut outline are lost, contracting strands of the stroma are seen to extend from point to point among the beaded projections, the periphery of the cell changes to a cogged rim, and finally the cell becomes shrunken and shriveled up into a small, many-starred asterisk. Some of the erythrocytes become fragmented and small bits of their stroma are observed to break off and float through the plasma. Others become progressively paler and paler, as the hemoglobin is dissolved out, until complete decoloration occurs. Still others become distorted into designs of every conceivable shape so that their resemblance to the normal cell becomes more remote" (Da Costa). These changes must not be confused with those occurring as a result of pathological changes. Crenation is often induced more rapidly than in normal blood, in the blood of persons suffering from acute infection and from chronic diseases. True ameboid movement of the red cells is sometimes seen as a result of a high-grade anemia. The average diameter of these normal human erythrocytes (normo- cytes) is 7.5 /1, the normal variations being between 6 and 9 // (a micromillimeter, 1 /1000 of a millimeter). The size varies depending on the method of preparing the specimen and also upon the osmotic tension of the plasma. Although dwarf and giant cells may occur to a slight degree at all ages, the normal infant blood shows these variations more markedly (3.3 to 10.3 fi, according to Hayem). According to Hamburger, the cells are slightly larger in the venous than in arterial blood. Gram states that the size of these cells varies with climatic conditions, being greater in those of the northern cooler countries than in those of the southern warmer climates. The variations in the size of the red cells is very slight in the different sexes. Pathologically, variation in the size of these cells is a common and (5). Size and Shape. PLATE XVII. Types of Red Cells. (Wright's Stain. Zeiss Ocular 4, Oil Immersion Objective.) i-Normal Cells. 2-3-4-Normal Cells as seen with different Focus. 5-So-called "Pessary Form." 6-7-8-Polychromatophilic Cells. 9-Macrocyte. IO-12-I3-POIKILOCYTES. 11-Poikiloblast. 14-15-Punctate Basophilia in Red Cells. 16-17-Normoblasts. 18-Normoblast with Pycnotic Nucleus. 19-Punctated Normoblast. 20-23-Megaloblasts. 24-Gigantoblast. THE BLOOD. 465 important feature. Generally speaking, variations in the cellular size indicate a severe and chronic anemia, while in the more mild acute forms of anemia such variations are unusual. The average size is said to be increased in jaundice, cholera, lead-poisoning, leukemia, congenital heart disease, and cretinism. Types of Pathological Erythrocytes. Variation in the normal size of the red cells is indicated by the term anisocytosis. This term does not include the misshappen red cells which appear in the blood as a result of degeneration or of mechanical injury and to which are given the name of poikilocytes. These are cells under the normal size, the variations being between 1 p and 6 p, the usual representatives being about 3.5 p. It cannot be stated at present whether these cells are mere schistocytes (fragments of larger cells) or are perfect cells of degenerative origin. They occur normally in the blood of embryos and infants, but are rare in that of the adult except in pathological conditions. It is certainly true that these undersized cells may arise from purely physical causes, as a result of increased osmotic pressure of the plasma as well as from division of undersized mother cells. This latter phase is made possible by the appearance, especially in pernicious anemia, of nucleated reds of corresponding size. Pathologically, these smaller cells are observed in all severe anemias. As a rule, they stain deeply and evenly, but in some cases of pernicious anemia these cells are deficient in hemoglobin and show irregular staining, yet many of them may have an increased hemoglobin content. Occasionally, but not invariably, we find these undersized cells in chlorosis, in which the hemoglobin is deficient. At times these cells may show polychromatophilia, but this is not the rule. According to Tallqvist, an increase in the number of these micro- cytes (microcytosis) is an indication of rapid destruction of blood. Microcytes. These are cells above the normal size, the variations being between 10 and 20 /z. Those cells from 9 to 12 /z are called macrocytes; those between 12 and 16 p are known as megalocytes; while those above 16 /z are termed gigantocytes. These cells are of regular shape, of even staining qualities, and generally without a well-defined central clear area. The larger size of these cells may be partly due to the swelling incident to a lowered osmotic tension of the blood, but more probably this increase in size is traceable to the origin of the macrocytes from the large nucleated reds of the bone-marrow. These cells may show an excess or deficiency in hemoglobin, the former characteristic being observed in the primary pernicious anemias, while the latter is evident in the secondary forms. Morris and Thayer1 report ameboid movements in these cells. Pathologically, the presence of these various forms of macrocytes, giving Macrocytes. 1 Arch, of Int. Med., vol. 8, 1911, p. 581. 466 DIAGNOSTIC METHODS. rise to the condition of macrocytosis, indicates a severe and unusually chronic anemia. They are most frequently seen in pernicious anemia, in which the largest cells are sometimes the darkest and some of the microcytes are exceed- ingly pale. These large cells do occur, however, in leukemia, cholemia, and chlorosis, being frequently pale or "chlorotic" and "dropsical." These "dropsical" cells are not sufficiently numerous, however, to change the volume index, as Capps has shown, of a secondary anemia to one shown in true per- nicious anemia. These are misshapen red cells of large or of small size, the varieties of such deformities being numerous. The presence of poikilocytes in the blood is known as poikilocytosis (first described by Damon) and is closely related to crenation as in both cases the cells may be similarly misshapen. The former is a pathologic process demonstrable the moment the blood is taken, while the latter is a physiological process appearing only after the blood has been in contact with air for some time. Poikilocytes arise in several ways. First, faulty technic, especially pressure on the cover-glass of the fresh specimen, will give rise to fragmentation of some of the corpuscles into small spherical masses, dagger-shaped bodies, and small elongated rods (pseudo bacilli of Hayem). Such fragmentation is indicative of lowered vitality and feeble powers of resistance of the cells. If the cover-glass be moved after the cells have spread, a large number of them will be distorted into oval or pear-shaped forms, the long axes of which usually point in the same direction. Secondly, true poikilocytes, or cells misshapen while in the circulation, are probably due to ameboid motion of a portion or the whole of a cell or to alterations in the plasma. These misshapen cells are usually pear-shaped with a budded projection at one or more poles. Cells may be seen which resemble a tennis-racket, a kidney, tomahawk-blade, dumb-bell, or anvil, while oval forms are especially observed in pernicious anemia and are considered by Cabot of diagnostic importance. Poikilocytosis is an indication of severe anemia with degenerative changes in the red cells; although it is not characteristic of any single disease, it is found more fre- quently in pernicious anemia and leukemia. Poikilocytes. Nucleated red cells may be considered pathological at any period of extrauterine life, although they are usually found in the blood of the child during the first few days of life. These nucleated reds are always found in the bone-marrow, the normal and large forms being quite distinctive. The large form is the oldest and gives rise by cell division to the smaller cell or the normoblast. It is probably true that the nonnucleated red cells are derived from the nucleated form, but the denucleation takes place before the normal cells reach the blood. Just how we are to explain the disappearance of all trace of the nucleus from the normal erythrocyte is a question, but the general (C). Nucleation. THE BLOOD. 467 concensus of opinion seems to be that the nuclear material gradually fades within the cell, although some slight evidence of extrusion of the nucleus can be advanced: Although our ordinary methods show no such evidence, King1 has recently introduced a method by which he demonstrates stainable remnants of a nucleus in practically every red cell in the normal adult blood. Normoblasts (Trachyochromatic Erythroblasts). These are nucleated red cells similar in size, shape, and color to the normocytes. They do not usually show a biconcave form and do not unite in rouleaux. The protoplasm of the cell is usually regular in outline, stains more intensely than does that of the normocyte and frequently shows evidence of polychromatophilia, although it is normally orthochromatic. In myeloid leukemia, cells are frequently observed in which the protoplasm presents a ragged outline and may even, in some cases, be so degenerated as to show only a small fragment attached to the nucleus. These latter cells are practically always polychromatophilic. This type of small cell might be called a microblast, which is the rarest form of erythroblast and corresponds in size to the microcyte. Moreover, we may find such cells or fragments of cells attached to a nucleus, such cells corresponding in size and shape to the poikilocytes and being termed poikiloblasts. The nucleus of the mature normoblast (Howell's mature nucleated red) has a diameter of about one-third that of the cell, is densely stained, homogeneous, sharply defined, spheroidal in shape, and without any decided chromatin net- work (the so-called pycnotic nucleus). It is situated rather toward the per- iphery than in the central portion of the cell and is surrounded by a clear zone shading off into the cellular protoplasm. Occasionally the nucleus is observed resting upon a margin of the red cell or even extruded entirely from it probably as a result of degeneration of the surrounding protoplasm (Pappenheim). These nuclei often show amitotic figures, being subdivided into two or more lobes or fragments which may be connected by strands of chromatin. These mature forms may show all gradations in which the chromatin network becomes more and more evident, until we reach the very immature forms (Howell's immature nucleated reds). These latter cells are somewhat larger than the mature forms and are somewhat lighter in color; the nucleus is relatively larger and is composed of delicate faintly basic chromatin fibers radially arranged, and frequently showing mitotic figures. These two types of cell are the forerunners of the normocyte and their appearance in the adult blood is indicative of increased activity of the hemato- poietic organs, especially of the bone-marrow, either as a result of a poor condi- tion of the blood itself or as a direct disease of the blood-forming organ. These cells are most commonly seen in the milder forms of anemia, chlorosis, and acute anemia from hemorrhage, inanition, or organic disease. In the severe types of anemia they are constantly met with and are usually associated with the larger megaloblasts. During the course of severe anemia, especially in chlorosis, there 1 Jour, of Med. Res., vol. 24, 1911, p. 91. 468 DIAGNOSTIC METHODS. obtains a periodic increase of normoblasts and of leucocytes lasting several days. This is followed by a marked increase in the normal number of red cells, giving rise to the condition described by von Noorden as a "blood crisis." The normoblasts disappear, the blood count falls off, and a second crisis may obtain. This condition is considered as a transitory attempt on the part of the bone-marrow to regenerate the blood. Such a crisis is not always a sign of improvement, but it does demonstrate that the type of blood formation is becoming physiological and that recovery may follow. Megaloblasts (Amblyochromatic Erythroblasts). These are cells of larger than normal dimensions, corresponding in size to the macrocytes and varying from 9 to 20 microns in diameter, if exceeding 20 microns they are known as the gigantoblasts of Ehrlich. Occasionally some cells are seen which are no larger than a normoblast, but in which the nucleus shows definite characteristics which should enable one to classify the cell as a megaloblast (Pappenheim). The protoplasm of the megaloblast often appears swollen and enlarged (dropsical). The cell is usually circular or oval, but it is easily deformed, giving rise to irregular-shaped bodies. Although it usually contains an excess of hemoglobin, it may show a deficiency. It is usually polychromatophilic, the shade of cellular staining varying from yellow to purple. These various color tones may not be regular, but may be varied by tintings of almost any shade in the same cell. The nucleus of the megaloblast is very large, varying between 6 and 10 microns. This may be situated either centrally or somewhat peripherally. It shows a great variety of forms appearing as a vesicular body with intra- nuclear network and nodal thickening, but without nucleoli. It may rarely be pycnotic, may show mitoses or many stages of karyorrhexis with fragmentation, vacuolation, fading of the segments of the nuclei, as well as minute subdivisions into fine basic staining particles widely scattered in the cell (Ewing). It is frequently poorly defined and shows feeble basic staining qualities. It may be sharply differentiated from the body of the cell by a distinct white margin which is thrown into relief by the deeper staining of the nuclear and cell-subs- stances. Occasionally the nucleus is overlooked owing to the polychromato- philic properties of the cell, which do not always permit of clear differentiation of nucleus and cell substance. Careful examination of the nuclear structure, with its wide-open meshwork, should, however, prevent the mistake of classi- fying this cell as a large lymphocyte. The clinical significance of the megaloblast is more or less in doubt. As such cells are foreign to the blood of an adult and as they are not present beyond 15 per cent, of the total nucleated reds in the marrow, according to Emerson, it must be assumed that their presence in considerable number is indicative of a reversion to an embryonal type of blood formation, or at least denotes an arrested development of normal cells and in consequence an increased production of these abnormal types. Megaloblasts are, therefore, evidences THE BLOOD. 469 of degeneration of blood-forming organs, while normoblasts are significant of regeneration of blood. A few of these cells, along with a larger number of normoblasts, has no special significance in cases of severe anemia, being found in small numbers in any variety of anemia. If, however, the majority of the nucleated red cells be megaloblasts, especially if gigantoblasts be present and unequal mitoses be observed, a diagnosis of primary pernicious anemia seems justified. It may be considered that the number of erythroblasts has no special significance as regards the severity of a particular case, but that it indicates merely the effort on the part of the bone-marrow to overcome the effects of blood destruction. The appearance of megaloblasts may, therefore, be regarded as an evidence of incomplete formation of the younger elements and, not necessarily, as an unfavorable sign. In the anemia following infection with bothriocephalus latus, the specific toxins produce a megaloblastic degeneration of the bone-marrow, so that the blood picture may assume the characteristics of a severe pernicious anemia. Ehrlich and Lindenthal have Reported a case of nitrobenzol poisoning, in the later stages of which the megaloblasts out-num- bered the erythroblasts. Thayer1 is the first to report the finding of ameboid movements in a megaloblast in a case of Addisonian anemia. (P). Number of Red Cells. The normal number of red cells is generally regarded as 5,000,000 per cmm. in the blood of the adult male, while the normal value for the female is 4,500,000. These figures are purely arbitrary, but they serve as an approximate basis upon which one may form his opinion as to the probable normality of a specific speci- men of blood. We frequently observe marked variations in the red count under the influence of both physiologic and pathologic conditions. These variations may be due to an actual increase or decrease in the number of cells or to a change in the volume of the plasma as discussed under Total Volume of Blood on page 378. It is not uncommon to find much higher counts than these normal ones in healthy individuals, especially in those living the "simple life," so that any special count must be a law unto itself and must be considered normal or abnormal only after considering all the factors which may influence the number of cells. Physiologic Variations. (1). Sex. The variations due to sex are not so marked as those from other causes. While the adult woman almost invariably shows a lower blood count than does the adult man, yet we occasionally find the girl showing, before puberty, a somewhat higher count than the boy of corresponding age and also the woman, after the menopause, showing somewhat higher values than does her brother of similar years. It can hardly be doubted that menstruation, as well as pregnancy and lactation, have some influence in lowering a count in woman at certain periods of life, yet these conditions are simply transient and would not seem to have as much bearing on the question as does the somewhat more hydremic state of the normal plasma of woman. ^rch. of Int. Med., vol. 7, 1911, p. 223; Ibid, vol. 8, 1911, p. 581. 470 DIAGNOSTIC METHODS. (2). Age. The number of the red cells varies more or less with the age of the subject examined. The highest values are usually observed at birth, when the count may run as high as 7,500,000, as in a case observed by the writer, the hemoglobin being practically always over 100 per cent, at this time; but, as a rule, a lower count is noted, averaging about 6,000,000. These high figures are due, to some extent, to the concentration of the plasma at birth through the loss of body fluid before a compensatory intake. The count gradually falls during the first few days and becomes fairly constant about the tenth day, when it is rare to find nucleated red cells. The number of cells is then stated to be somewhat reduced until the age of puberty, when a gradual increase occurs until about 40 years of age, after which a slight decrease may be observed in man. These variations, observed at different ages, are to be regarded as slight and influenced, to some extent, by the many modifications of the plasma as the result of growth and development as well as to the gradual decrease of functional activity after the middle periods of life. (3). Altitude. An increase in the number of cells has been observed under the influence of a higher altitude. The count has heen shown to increase at the rate of approximately 50,000 cells per 1,000 feet of ascent and to diminish, within 36 hours, at a corresponding rate, this increase or decrease being more marked the more sudden the ascent or descent. Just what factors are at work in causing these changes is undecided.1 The rise is too rapid to be entirely accounted for by new formation of blood-cells under the influence of diminished oxygen tension, and the fall is not accompanied by signs of destruction of the reds. Weinzirl considers the increased count of high altitudes due to the lowered temperature at these elevations. This factor would seem to have some influence, as counts are frequently observed in which a variation is noted at places showing the same elevation. Change of residence from warm to a cold or from cold to a warm climate may lead to an increase or decrease,2 as the case may be, in the number of cells varying from 500,000 to 2,500,000. (4). Nutrition. The general nutritive condition of the subject has more or less influence upon the number of red cells per cmm. This statement must not be interpreted to mean that the obese person shows a higher blood count than does his sparer brother. It is well known that obesity is an indication of poor assimilation of food; hence we should expect to find, as we really do, that the muscular person has a somewhat higher count than the obese subject and well-developed robust individuals a larger value than do the poorly nourished patients. Diet has, of course, much to do with the general nutrition, so we find the meat eaters averag- 1 See Cohnheim, Kreglinger, Tobler, and Weber, Zeitsch. f. physiol. Chem., Bd. 78, 1912, S. 62. 2 Chamberlain has, however, shown (Philip. Jour. Sc., Sec. B., vol. 6, 1911, p. 467) that the average number of red cells in the tropics is little different from that in temperate zones. THE BLOOD. 471 ing somewhat higher in their percentage of red cells than do vegetarians. Immediately following a hearty meal we may observe a temporary decrease in the number of red cells, but this soon returns to normal, owing to the rapid adjustment of the water content of the blood. (5). Exercise. Active muscular exercise produces a transient increase in the red count due both to the increased blood-pressure and to the concentration of the blood through the loss of water by perspiration. Physical exercise, taken to the point of producing fatigue, may produce a marked diminution in the number of red cells, due probably to the fact that regeneration of new cells cannot keep up with the destruction of the old cells. Passive exercise, in the form of massage, has a transient influence in producing an increase in the number of the red cells, owing to its increasing the general circulatory tone. Careful investigation of the influences of both cold and hot baths has shown that an increase in the number of reds occurs under the action of these two conditions. The increase as a result of a cold bath may be as great as 2,000,000 cells, due to capillary stasis as a result of vasomotor constriction. Likewise, a hot bath will increase the number of cells by causing dilatation of the per- ipheral vessels and a consequent increase in the amount of blood at the point from which the specimen is taken. When marked perspiration follows either a hot or cold bath the blood becomes concentrated and, as a result, an increase in the number of red cells will obtain. (6). Baths. Any drugs which cause rapid loss of fluid from the body, as for instance emetics, purgatives, diuretics, and diaphoretics, will cause concentration of the blood and hence a coincident increase in the number of cells, providing the change is sufficiently rapid and the blood examination is made before compensation occurs. Among the drugs which increase the number of red cells we find iron and arsenic, both of which are particularly valuable in anemic conditions, the former in chlorosis and the latter in pernicious anemia. The compounds of mercury and of lead, on the other hand, have a destructive action upon the red cells, so that we find these drugs causing a diminution in the number of cells. (7). Therapeutic Measures. Pathologic Variations. (1). Oligocythemia. This is a condition characterized by a diminution in the number of red cells. It is usually associated with a decrease in the percentage of hemoglobin and with a slight reduction in the total volume of blood, although this latter factor is not invariably present. This condition is found, to a more or less degree, in all forms of anemia and may be temporary or permanent. The extent of the decrease in the number of red cells varies from 500,000 cells to 472 DIAGNOSTIC METHODS. a reduction of 4,000,000. This diminution in the number of cells is usually an indication of the severity of the anemia, the most marked decrease being observed in the pernicious types, Osler reporting probably the lowest count recorded, namely, 100,000 red cells. While this oligocythemia is usually associated with oligochromemia, yet we find in chlorosis that the diminution of red cells is not as marked as is the reduction in the amount of hemoglobin, the color index in this condition being usually low. In some cases of chlorosis, however, the oligochromemia may keep pace with the oligocythemia. In pernicious anemia, on the other hand, we have a marked reduction in the number of cells and coincidently a large de- crease in the amount of hemoglobin, the result being that we have a very high color index as the usual sign of this condition. Cases showing the loss of a large amount of blood, as the result of hemorrhage have naturally an oligocy- themia, but the percentage of hemoglobin may not necessarily be reduced to any great extent. This gives rise to the condition of secondary anemia, in which the color index may be high. In these cases of hemorrhage a sudden reduction to the point of 1,000,000 or less cells is usually followed by a fatal result, although even here recovery has been made possible by rapid infusion of salt solution and compensatory activity of the blood-forming organs. In leukemia we find that the red cells are not usually diminished to a very great extent, the oligo- cythemia being generally more marked in the lymphatic than in the myeloid variety. Occasionally, however, we do find very low counts, in both varieties of leukemia. These same statements apply to splenic anemia, although the count here very rarely reaches a lower point than 2,500,000 cells. A large number of conditions, aside from direct blood diseases, cause a diminution in the number of these red cells. Thus we find that the toxins of certain of the specific fevers, such as typhoid and pneumonia, may cause a marked anemia, but this is not necessarily the rule. Acute infection with pus organisms is frequently observed to cause an extreme and rapidly progress- ive anemia, the destruction of the red cells being in some cases very extensive. In malignant disease we usually find, especially if cachexia is present, a very extensive anemia which may lead to a diagnosis of the pernicious type. It is to be remembered, however, that a secondary anemia may assume all of the characteristics of the primary pernicious type, so that we should be on our guard as to the cause of an anemia and not be content with the simple findings of such a condition. (2). Polycythemia. This is a condition characterized by an increase in the number of red cells and is sometimes called polyglobulia.1 Whether this increase is actual and permanent or whether it be simply apparent, due to concentration of the blood or to unequal distribution of blood in the peripheral vessels, is still an unsettled question. It is true that many physiological causes previously 1 Also erythremia. THE BLOOD. 473 enumerated do bring on a polycythemia, so that these should be remembered whenever a blood count is to be made. From the pathologic standpoint we find an increase in the number of red cells occasionally following active blood regeneration after hemorrhage. Likewise, we observe an increase in the number of cells in phosphorus poisoning, acute yellow atrophy of the liver, and in certain cases of general hepatic insufficiency. Why such an increase of cells occurs in these conditions is very hard to say. It has been found that the solids of the plasma are not increased, so that we are not warranted in assuming a concentration of the fluid elements. A polycythemia of uncertain origin is also seen in cases of poisoning with carbon monoxid and illuminating gas.1 In a condition called by Osler autotoxic enterogenous cyanosis, which is characterized by a marked increase in the number of red cells and also by enlargement of the spleen, we have reason to assume that a direct real increase in the number of red cells occurs, but the pathogenesis of the condition is uncer- tain. It has been traced to improper aeration of the blood and occurs also in congenital heart disease, mitral lesions in the adult, and in pneumonia and acute miliary tuberculosis. While this high blood count must be definitely admitted, its importance from a clinical standpoint is still unsettled. It is uncertain whether the rapid increase in the number of reds is due to an attempt on the part of the blood-forming organs to overcome the influences of the toxic substances upon proper oxidation in the system; yet it is certain that relief very frequently follows such an increase in the number of red cells, although in Osler's disease or, as it is called by some, Vaquez's disease, death frequently follows without any relief from the condition.2 (E). Staining Properties of the Red Cells. The normal red cell, like all living cells, is incapable of being stained with anilin dyes; that is, it is achromatophilic. Previous to staining, fixation of the cell must take place, the protoplasm being killed in this process. The so- called vital staining has been considered previously. The normal fixed red cell has a marked affinity for various dyes of the acid type, such as eosin, orange-G, acid fuchsin, and indulin, and is therefore called acidophilic or oxyphilic. This cell normally takes up but one color from a mixture of dyes, and is called, therefore, monochromatophilic. Polychromatophilia. Under various pathologic conditions we find red cells which show a tendency to take up the basic stains, and for this reason are called basophilic or poly- chromatophilic cells. The tint of such abnormal red cells varies from that of a light indefinite shade of the basic stain to a dark distinct tone. Just what factors are at the bottom of this change, which is called polychromatophilia or polychromasia, is difficult to say. The normal acidophilic tendencies of the cell are due to the presence of hemoglobin and, in consequence, the normal 1 See Karasek, Trans, of Chic. Path. Soc., vol. 8, 1911, p. 173. 2 See White, Lancet, vol. 92, 1912, p. 7; also Lucas, Arch, of Int. Med., vol. 10, 1912, P- 597- 474 DIAGNOSTIC METHODS. staining properties will depend upon the relative richness of the cell in this pigment. As hemoglobin is always acidophilic, we cannot assume that the polychromatophilic properties of the abnormal red cells are due to variations in the hemoglobin, but must content ourselves with the belief that these changes have something to do with the protoplasm of the cell. This condition, which has been termed by Ehrlich anemic degeneration, is characterized by a diffuse basic staining property of the red cells. These basophilic cells are somewhat larger than the normal ones, show less bicon- cavity, and often are abnormal in shape. The megaloblasts are practically always polychromatophilic and the normal erythroblasts of the bone-marrow usually show this degeneration. This fact would seem to point to the probability of this condition being a sign of regeneration of the blood as the erythroblasts are more prone to be polychromatophilic the younger they are. Two distinct forms of polychromasia have been found. The first of these, known as the polychromatophilic degeneration of Gabrits chew sky, is the diffuse basophilic staining of the cells which is found in various forms of anemia and extensively in the cells of the normal bone-marrow. The second of these, the polychromasia of Maragliano, is shown in severe anemias and in toxemias and appears as a more punctate basophilia, being closely related to the basophilic degeneration of Grawitz, which will be discussed later. While the polychromatophilia of the diffuse type is considered by Ehrlich as evidence of a degenerative process, that is as a coagulation necrosis of the discoplasm in consequence of which this takes up the albuminous principles of the plasma while it loses its power of retaining hemoglobin, yet there are evidences showing that this may be a regenerative process. Polychromatophilia is often seen in cells which are undergoing degeneration, especially in myelogenous leukemia, pernicious anemia, severe secondary anemia, especially those following malig- nant disease, eruptive fevers, malaria, and after various poisons, such as that of snake venom. The forms of partial polychromatophilia, which have been described under various names, such as vacuolization, pseudonucleation, globular decolorization, and more commonly Maragliano's endoglobular degeneration, are seen in normal blood in from 30 to 70 minutes after the specimen is made. They are usually found in the center of the cell, but may be near the periphery. There may be several such areas in a single cell, but the more common form is the single area of degeneration, which is usually round, may be elliptical, and may resemble a vacuole. These are shown quite distinctly in the unstained specimen and in this condition very frequently are seen in rapid motion. This motion is not of the true ameboid type, but is due more to the gradually progressing coagulation and consequent constriction of the protoplasm. These areas, whether in the fresh or stained specimens, may be mistaken for malarial parasites, so that the worker should be on his guard lest he make a wrong diag- nosis without sufficient evidence. They differ in size from the parasites, and on focusing the specimen are characterized, usually, by a change from a smaller THE BLOOD. 475 to a larger form, which variation is not evident with the malarial organism. The longer one searches in the unstained specimen for malarial organisms the more apt a mistake in diagnosis is to be made, as these " Maraglianos " develop quite rapidly. In the stained specimen these areas of coagulation necrosis will show a basophilic staining quality, while the chromatin element of the malarial organism will differentiate this body from the more usual degenerative area. While the preceding conditions of polychromatophilia are to a certain extent true degenerations, yet they are more properly grouped under the heading of atypical staining reactions of the red cell. Other forms of degeneration, which are likewise characterized by abnormal staining qualities, are nevertheless different in many ways. (F). Degenerations. Basophilic Degeneration of the Red Cells. This condition is distinct from the polychromatophilia above described and is known as the punctate basophilia of Grawitz. It is characterized by the appearance, in the body of the red cell, of granules of varying size, which stain with the basic dyes. The cell may be dotted throughout with these gran- ules or may show this degeneration only in parts. These granules are not observed in the fresh unstained specimen and are not increased, as are the Maragliano areas, by allowing the blood to stand. These cells are observed in pernicious anemias, leukemia, in the toxemia of malignant states and especi- ally in cases of lead-poisoning. The size of the granules may vary from small dots to large granules, showing a diameter of one or more microns. Their origin is much in doubt, but they are generally considered to be areas of coagulation necrosis associated with either degeneration or regeneration of the red cells. They are quite distinct from the granular basic "stippling" observed in malarial conditions, but are probably evidences of direct toxic effects upon the protoplasm of the red cells. Their greatest clinical importance is probably in lead poisoning, where they may be the only signs of abnormality in the blood picture. They may vary in number from day to day, showing in some cases five or six in one microscopic field, but, as a rule, they are present in much fewer numbers, as one observes them only after examining several visual fields. As a rule, they appear very early in cases of lead-poisoning, in one case observed by the writer within three days, and they may be present in the blood of a lead worker for several years after his exposure to the effects of the lead. They are usually the first sign of anemic change and usually persist longer than do the other abnormalities of the blood.1 Occasionally one observes in the red cells curious ring-like bodies which are in shape very much like the hyaline malarial ring with a circular refractive center. They change their shape in a peculiar way, much resembling the Ring Bodies. 1See Petry, Biochem. Zeitsch., Bd. 38, 1912, S. 92. 476 DIAGNOSTIC METHODS. undulatory movements of the hyaline body. They do not increase in number or grow larger on standing as do the Maraglianos and they are observed in a large number of conditions, such as measles, pernicious anemia, and severe secondary anemia. Two types of these ring bodies must be distinguished, the first is usually more peripherally situated, occasionally has a definite crescentic shape or has an appearance much resembling that of a Maragliano, and is found especially in cases of measles. This finding has lead to the assump- tion that these bodies were the organisms causing measles. The second form has been described by Cabot1 and more recently by Schleip2 in cases of perni- cious anemia, leukemia, and secondary anemias. These latter bodies are larger than the former, more distinct, irregular in shape, frequently forming figure-eight bodies and are usually stained bright red with Wright's stain, but may occasionally show a blue shading. Herrick3 has observed thin, elongated, sickle-shaped red cells, some of them nucleated, in a case of severe anemia. Various other forms of degeneration of the red cells occur, as for instance the appearance of rod-like areas resembling bacilli,4 which may keep up a con- stant vibratory motion carrying them through the entire substance of the cell. This finding should not confuse one in making a diagnosis of the presence of bacteria. Another form of degeneration, known as Ehrlich's hemoglobinemic degeneration, has the appearance of a small dark cell lying upon a larger paler one. These are probably areas of condensed protoplasm with the hemoglobin distinctly separated from the stroma. Such cells appear, occasionally, in certain types of malaria and are shown as corpuscles in which the hemoglobin is apparently condensed around the parasite. This degeneration is occasionally seen in nucleated red cells and may give the appearance of a microblast lying upon a macrocyte. It is best seen in cases of pernicious anemia and may explain some of the "acidophilic granules" of the red cells which have been described (Emerson). (G). Isotonicity and Resistance of the Red Cells. Normally, the relations of the cellular to the fluid portions of the blood are such that the hemoglobin and other constituents of the red cells are held intact within the limiting membrane of the cell. Any change in the osmotic pressure of the plasma is certain to be manifested by variations both in size and in the composition of the cellular elements. We find, therefore, as previously discussed under the heading of Osmotic Pressure, that the inorganic constituents of the plasma are the factors upon which the volume of the red cell depends to a large extent. By an isotonic solution, as applied to the blood, we mean a solution of such a strength as to preserve the corpuscles and to prevent passage of water, salts, and organic bodies from the plasma into the corpuscle or from the corpus- 1 Jour. Med. Res., vol. 9, 1903, p. 15. 2 Deut. Archiv. f. klin. Med., Bd. 91, 1907, S. 449. 3Arch. of Int. Med., vol. 6, 1910, p. 175. 4 See Roth Zeitsch. f. klin. Med., Bd. 76, 1912, S. 23. PLATE XVIII. Ring Bodies in Red Cells. (After Schleip.) THE BLOOD. 477 cle into the plasma. In other words, an isotonic solution is one whose osmotic pressure is equal to that of the contents of the red cell. While the limiting structure of the red cell is indefinitely known, it seems to have certain properties which would lead us to assume that it is a semipermeable membrane, that is one which will permit the passage of organic compounds or of salts or their ions in both directions. This apparent selective activity is dependent both upon the laws of osmosis and diffusion through a semipermeable membrane and upon the basic ideas of Ehrlich's side-chain theory. While other figures have been used at various times to represent the strength of a solution isotonic with the red blood-corpuscle, the one which seems more nearly to duplicate the actual condition is a 0.9 per cent, solution of sodium chlorid. This is or should be the normal salt solution which is used in physiological work and in transfu- sion. Loeb has shown that a properly balanced solution containing the chlorids of sodium, potassium, and calcium more nearly represents a proper "physio- logical salt solution" than does the simple 0.9 per cent, sodium chlorid solution. Such an isotonic solution will preserve the corpuscles, will prevent the passage of hemoglobin from the cells into the plasma, and will not permit of shrinkage or swelling of the cells. Solutions which are stronger than the one above mentioned are called hypertonic solutions, while those of a lower concentra- tion are termed hypotonic solutions. In the first of these shrinkage of the cells will occur, while in the latter phenomena of swelling will be observed. These facts are of importance in the study of the volume of the corpuscle and should not be forgotten in reporting variations in size of the cells. The erythrocyte normally shows a certain amount of resistance to variations in the osmotic pressure of the blood and to the hemolyzing effects of various substances. It is true that the osmotic pressure of the plasma may be raised by the introduction of certain substances into the blood without causing phenomena of shrinking, swelling, or hemolysis to occur, yet we must not assume that the cell is not influenced by such changes in osmotic pressure. Simple variations in osmotic pressure come more nearly to the foreground in the explanation of these changes, especially of those occurring in most disease processes of a chronic type, while in the more acute infectious types associated with toxemia the biologic theory must be invoked. As yet little of value has been forthcoming from the study of the resistance of the blood-cells to variation in osmotic pressure, as the scope of the examination has not been sufficiently widened by application of accurate experimental methods. The writer is forced to say, from the results of his own experiments, as well as those of others with the more usual method of Hamburger,1 that no reliable data are at hand. It is highly probable that the little understood biologic influences are clinically much more important than are those due to variations in osmotic pressure, as this latter process cannot explain the hemolytic results of the toxins or of certain definite chemical compounds which bring about these results in concentrations which have little influence upon the tonicity of the blood. 1 Wiesbaden, 1902. 478 DIAGNOSTIC METHODS. (ZZ). Variations of the Red Cells in Childhood and in Old Age. As previously stated, the average number of red cells is much higher than the normal figure at birth, gradually decreasing until the age of puberty when a gradual increase occurs again until middle life, after which a steady decrease is observed. It is to be expected that variations in the color tones (hemoglobin content) of these cells will vary normally with the age of the subject, being very high in the early days of life, gradually diminishing as puberty approaches, in- creasing until the middle periods of life, and steadily declining as the age increases. Aside from the number of cells, the blood in childhood shows little variation in the characteristics of the red cells, but the variations in the plasma are more or less marked, as the synthetic and retrograde products of tissue metabolism are much more evident in the period of development than during the middle periods of life. As the subject becomes older the blood is laden with products of retrograde tissue change and the blood-forming organs become less active. It must be remembered that the various organs of the body contribute certain factors to the blood and may, therefore, all be considered as blood-forming organs in a definite sense. The ductless glands, in particular, pour into the blood certain substances which markedly influence the proper correlation of these various organs. Just what these substances are is not well established, so that the introduction of the word "hormone" by Starling does not at present have a specific meaning. As old age approaches variations become evident in the plasma, and through such influences changes in the morphology of the red cells are more or less frequent. Besides these plasma changes, gradual atrophic degeneration of the blood-forming organs takes place, so that we may find red cells showing the various types of anemic degeneration and may even see nucleated red cells as an evidence of the attempt on the part of the bone-marrow to overcome the gradual destruction of cells. The more notable variations in the red cells are observed in their size, owing to the hydremic state which is almost always characteristic of the plasma in advanced periods of life, but associated diseased conditions will, of course, give their characteristic changes which have no place in our discussion at this point. (Z). Functions of the Red Blood-cells. The only well-recognized function of these cells is their oxygen-carrying power, which is dependent upon their hemoglobin content. Other oxygen catalysts, such as oxidase, peroxidase, and catalase, are found more largely in the leucocytes and play a significant role in the function of the whole blood. These will be discussed later. As is well known, the fluid portions of the blood are much more influenced by the various organs of the body than are the cellular elements, although the red cells will naturally be affected by variations in the hematopoietic organs, such as the bone-marrow, liver, and spleen. Our knowledge regarding the variation in the plasma is so uncertain that we have been led to interpret the changes in the histological appearance of the blood as DESCRIPTION OF PLATE XIX- 1,2, 3, 4. Small Lymphocytes. Contrast the faintly colored protoplasm of these cells in the triple stained specimen with their intensely basic protoplasm in the film stained with eosin and methylene- blue, 17 and 18. The cell body of 1 is invisible. Note the kidney-shaped nucleus in 4. 5, 6. Large Lymphocytes. With this stain the nucleus reacts more strongly than the protoplasm; with eosin and methylene-blue (19, 20), on the contrary, the protoplasm is so deeply stained that the nucleus appears pale by contrast. This peculiarity is also observed in the smaller forms of lymphocytes. 7, 8. Transitional Forms. Note the modera^^H^tau: an^indented nucleus, and the almost hyaline non- granular protopla:^.c 8 with the myelocyte, 7, Plate 1 V, these cells dif- fering chiefly in tlBBI^nyelocyteicontains neutrophile granules. 9, 10,11. Polynuclear Neutrophiles. These cells are cMH»M0ed ^y^a polymorphous or polynuclear nucleus, sur- rounded by a cell IS^mP^W'ith fine neutrophile granules. In 11 the nuclear structure is obviously separated into four parts; in 9 it is moderately, and in 10 markedly, polymorphous. 12,13. Eosinophiles. The nuclei are not.unlike those of the polynuclear neutrophile, except that they are somewhat less convoluted, and poorer in chromatin, staining less intensely. The protoplasm is filled with coarse eosinophile granules, the characteristics of which are clearly illustrated by 13, a " fractured " eosinophile. 14. Eosinophilic Myelocyte. Compare with 15. 15, 16. Myelocytes. {Neutrophilic.) These cells are morphologically similar to 14, except that they contain neutrophile instead of eosinophile granules. Note that the granules of the myelocyte are identical with those of the polynuclear neutrophile. A dwarf form of myelocyte is represented by 16. (Triacid Stain.) 17,18. Small Lymphocytes. Note the narrow rim of pseudo-granular basic protoplasm surrounding the nucleus and the pale appearance of the latter. 19,20. Larg Lymphocytes. Budding of the basic zone of protoplasm is represented by 20, Both of these cells belong to the same type as 5 and 6. 21,22. Large Mononuclear Leucocytes. Compared with 19 and 20, these cells have a decidedly less basic protoplasm, but a somewhat more basic nucleus. In the triple stained film these differences cannot be detected, so that they must be classed as large lymphocytes. 23. Transitional Form. The distinction between this cell and 24 is not marked; the nucleus of the latter simply being somewhat more basic and convoluted. 24,25,26,27. Polynuclear Neutrophiles. With this stain these cells show a feebly acid protoplasm, and lack granules. Note that the more twisted the nucleus the deeper it is stained. Compare with 9, 10, and 11. 28, 29. Eosinophiles. Compare with 12 and 13. 30. Eosinophilic Myelocyte. Compare with 14. 31. Basophile. {Finelygranular.) This cell is characterized by the presence of exceedingly fine 5-granules, staining the pure color of the basic dye. The nucleus is markedly convoluted and deficient in chromatin. The cell here shown was found in normal blood. 32, 33, 34, 35, 36. Mast Cells. The granules take a modified basic color, as shown by their royal-purple tint in this illustration. Note their unusually large size and ovoid shade in 35. their peculiar distribution in 35 and 36, and their irregularity in size in 32 and 36. With the triacid mixture these granules, as well as those of the finely granular basophile, 31, remain unstained, showing as dull-white stippled areas in the cell body. The nuclear chro- matin of the mast cell is so delicate and so freely stained that it is barely visible. These cells were found in the blood of a case of spleno-medullary leukemia. {Eosin and Methylene-blue.) PLATE XIX (1-16, Triacid Stain; 17-36, Eosin and Methylene-blue.} The Leucocytes. (E. F. Faber, fee.} (From Da Costa's "Clinical Hematology.") THE BLOOD. 479 significant of certain general pathological conditions. It is true that in some cases we do have direct disease of the blood-forming organs and, in consequence, variations in the cellular structures are of diagnostic importance, yet in the large majority of cases the preliminary changes must be resident in the plasma, those in the cellular structure being simply incidental. It may be, therefore, necessary to assume a functional activity on the part of the red cells to overcome biologic changes in the plasma and if so we must attempt to discover just what plasma changes we are to regard as significant of the various conditions shown by the red cells in each true blood disease. (6). Leucocytes. (A). Appearance. In the fresh specimen the leucocytes appear as colorless, highly refractive bodies, somewhat larger than the red cells, and showing a definite nucleus. These white cells show distinct ameboid movement in the fresh specimen by virtue of which they are able to surround a foreign body and enclose it within their own protoplasm. This is the well-known property of phagocytosis which is such an important factor in Metschnikoff's theory of immunity. In contradistinction to the red cells, these white cells show many variations both in size and in shape. As the relations between the nucleus and the cellular protoplasm is quite distinct in many of these forms, some writers have been led to classify the white cells according to the peculiarities of the nucleus. Other writers classify them into granular and nongranular forms, as many of these cells show distinct granulations which are more clearly differentiated in the stained specimens. As more or less confusion exists regarding these different variations, it seems more rational at present still to classify these white cells with Ehrlich, who combines both the nuclear and granular characteristics in his classification. (B). Leucocytes in Normal Blood. (1). Lymphocytes. The cells of this class which occur normally in the blood are of two kinds, (a) the small cell, about 5 to 8 microns in diameter and (&) large cells showing a diameter of from 8 to 10 microns. Both of these cells have a very large nucleus which is usually centrally located but may have an eccentric position. It may be compact or coarsely reticulated and is not always as strongly stained as is the surrounding cytoplasm. Nor is it as deeply stained as is the nucleus of the normoblast, although it is very rich in chromatin. The nucleus of the larger cell may show very irregular staining properties. The nuclei of both of these cells is usually circular in outline, but may show an oval or a kidney- shaped structure, in which at times may be seen a distinct nucleolus. The protoplasm of these cells is usually in the form of a narrow rim surrounding the nucleus and showing a strongly basophilic homogeneous cytoplasm. The older cells may at times show a difficult staining property, may even be acido- philic, and occasionally exhibit a net-like structure in which are observed a 480 DIAGNOSTIC METHODS. few granules scattered throughout the protoplasm of the cell. This granulation, which occurs in about one-third of the cells seems to have little clinical signifi- cance, except as a probable sign of age of the cell, this granular stage being regarded as the end-point of their development. These granules are probably not representatives of true granulation, but seem to be more referable to nodal points of the reticular structure of the protoplasm. These cells usually constitute from 20 to 25 per cent, of the leucocytes, their absolute numbers being from 1,200 to 2,000 per cmm., but the relative proportion of the small and large form being unsettled. The larger types of these cells are more closely associated with increased functional activity of the lymphoid tissues and should probably be considered the older cell of the two. According to Pappenheim, the large lymphocyte represents the mother cell from which all other leucocytes as well as erythrocytes are indirectly developed as a result of heteroplastic differentiation. Variations in technic of pre- paring the slides w'ill often lead to differences in size of these cells, the thinner preparations showing these cells as larger forms, while the thicker preparations present the smaller type of these cells. These cells are beautifully colored by the hematoxylin-eosin stain and by the various modifications of the Romanowsky stain, although the boundary between the nucleus and the protoplasm is not always clearly outlined by these latter stains. The granules, which are occasionally found in these cells and which are in size between the a and the e granules of Ehrlich, show most clearly in preparations stained by the Giemsa stain and appear distinctly azurophilic. These granules do not appear at all with the triple stain. These cells are frequently increased and frequently diminished, the former condition being spoken of as lymphocytosis, the latter as lymphopenia. A relative increase of the number of these lymphocytes is regarded as more or less characteristic of typhoid fever, especially when associated with a diminution in the total number of white cells. In lymphatic leukemia these cells are present in large numbers and are indicative of marked disturbance in the lymphatic structures. (2). Large Mononuclear Leucocytes These cells, which are supposed to be derived from the spleen and are called, therefore, splenocytes, are as a rule two or three times as large as a red cell (12 to 15 microns), and show a nucleus which may be large and round, but is more frequently oval in shape or may be indented, forming the so-called kidney type of nucleus. It is to be remembered that we may have both large and small types of this large mononuclear leucocyte, the smaller forms being distinguished from the lymphocyte by the relation of the protoplasm to the nucleus. These nuclei are usually eccentric in position and are not always sharply outlined, they are poor in chromatin, but are strongly basophilic, although less so than are the nuclei of the lymphocytes. The protoplasm of these cells is very abundant in relation to the size of the nucleus and is very THE BLOOD. 481 clear, hyaline, and nongranular in appearance. This protoplasm is much less basophilic than is the nucleus and shows a very fine reticulum with nodal thickenings which are somewhat more strongly basophilic and may give the appearance of granulation.1 These cells should not, however, be regarded as granular cells. Occasionally vze find cells, derived from the above, which have been called by Ehrlich "transition forms," which are the largest of all the white cells. The nucleus is pale, often deeply notched giving it usually the appearance of a saddle-bag, and shows the characteristics of the large mononuclear type, being distinguished from the polymorphous nucleus of the neutrophile cells by its greater thickness and by its diminished intensity of staining. The extensive granulation of the polymorphonuclear leucocyte should make any mistake in diagnosis impossible. The protoplasm of these transitional cells is very abundant and shows faint basophilic properties and may even have a few fine granules, which are neutrophilic in character, in the neighborhood of the nucleus. These two forms of cell constitute between 3 and 5 per cent, of the leucocytes, their absolute number varying between 200 and 400 per cmm. Pappenheim considers that these large mononuclear cells develop directly from the large lymphocytes and then pass into the transition forms, which are the final developmental types. (3). Polymorphonuclear Cells. (a). Polymorphonuclear Neutrophiles. These cells, sometimes called the finely granular cells of Schultze, are from two to three times the size of the red cell (10 to 12 microns) in diameter, the size depending upon the extent to which the cell is flattened by pressure. These cells are always smaller than the large mononuclear type and quite small specimens are frequently found in cases of myeloid leukemia. They are the most sharply characterized cells of the blood, their protoplasm being relatively great and showing slightly acidophilic staining properties. The re- ticular portion of the protoplasm is very slightly basophilic, showing occasionally nodal thickening or granules when stained with the methylene blue dyes. Throughout this protoplasm are scattered fine dust-like granules, the e granules of Ehrlich, which are not all of the same size and which stain with the neutral principle of the dyes. In some cases these granules may gradually diminish so that they may be apparently absent or at least undetected. The triple stain is the characteristic stain for these granules, showing them of a distinct lilac color; but various other acid dyes, such as eosin, give various tints to them, so that they are recognizable as distinctly reddish granules in specimens stained by any of the Romanowsky modifications. The nuclei are elongated and constricted and may appear in the form of a bent rod or a mass of interwoven fine fibers, showing a thin chromatin-rich structure, or a distinctly S-shaped formation with oval thickenings. Frequently these nuclear masses appear 1 Bergel (Deutsch. Arch. f. klin. Med., Bd. 106, 1912, S. 47) has shown that these cells play a large role in the hemolysis and lipolysis noted after injection of cells and lipoids. 482 DIAGNOSTIC METHODS. as if separated into two or more distinct nuclei, but as a rule these masses are connected by fine bands and should, therefore, be considered rather as polymorphous than as polynuclear. The nucleus shows a reticular structure with nodal thickenings and is very basophilic. These forms constitute from 65 to 75 per cent, of the total number of cells, their absolute number averaging 5,000 per cmm., the percentage ranging from 20 to 40 per cent, in the first years of life when the lymphocyctes are increased. When outside of the blood-vessels they form the usual pus-cells and are extremely active phagocytes. These cells are derived from the mononuclear neutrophilic myelocytes which are normal habitants of the bone-marrow. They run their course as such and are not transformed into other cells, their granules diminishing in patho- logical conditions or with age, such change being associated with degeneration in the nucleus. Perinuclear Granulation. Sometime ago Neusser reported a finding of basophilic granules in certain leucocytes, especially the mononuclear and polynuclear type. These granules surrounded the nucleus and even appeared attached to it, showing a variable size and being more or less refractive. He regarded these perinuclear granulations as characteristic of the uric acid diathesis, but the work of Futcher and of Simon show that these granules are undoubtedly artefacts which can be produced by heating and by variation in the staining and appear in both health and disease, having no relation whatever to the output of purin bodies in the urine. Arneth 1 in his study of the blood both in health and disease has been led to classify the polymorphonuclear neutrophiles into five classes depending upon the number of nuclear lobes. His divisions seem to be fairly constant in health, but show great variation, especially in infectious conditions. His first subdivision, known as Class 1, is divided into (a) M cells, which are mononuclear forms identical with Ehrlich's myelocyte, (b) W cells, representing forms which show only a slightly indented nucleus, the indentation never extend- ing beyond the middle of the nucleus; this cell forms what is called the meta- myelocyte, (c) T cells, in which the indentation of the nucleus is deeper than in the W cell, but there is no distinct separation into isolated loops, this form constituting the true polymorphonuclear type. The first two varieties of Class 1 are usually seen only under abnormal conditions, although the W cell may be found to the extent of 0.2 per cent in healthy conditions. The cells of Class 1 are usually present, according to Arneth, to the extent of 5 per cent., Simon giving this percentage as from 4 to 9. The second class of Arneth embraces cells with two distinct nuclear fragmentations and shows three subdivisions: (a) 2 K cells, neutrophiles whose nucleus consists of two round Arneth's Classification. 1 Jena, 1904 THE BLOOD. 483 nuclear portions; (b) 2 S cells, neutrophiles whose nucleus consists of two distinct S-shaped forms; (c) 1 K 1 S cells, neutrophiles whose nucleus consists of one round nuclear portion and one S-shaped division. These cells of Class 2 constitute, according to Arneth, 35 per cent, of the total neutrophiles, while Simon's figures are 21 to 47 per cent. In Arneth's figures we find the 2 S cells forming about 23 per cent, of the total of this class. The third class has three nuclear divisions and is subdivided into four parts, as follows: (a) 3 K cells, (&) 3 S cells, (c) 2 K 1 S cells, (J) 2 S 1 K cells. Arneth gives the percentage of Class 3 as 41, the 2 K 1 S and the 2 S 1 K subdivisions, each representing approximately 16 per cent, of the total of this class. Simon's figures for this class range from 33 to 48 per cent. The fourth class comprises cells with four nuclear divisions showing five subgroups, as follows: (a) 4 K cells, (&) 4 S cells, (c) 3 K 1 S cells, (J) 3 S 1 K and (c) 2 K 2 S cells. This class, from Arneth's figures, shows a percentage of 17, the 4 K, 3 K 1 S, and 2 K 2 S types being largely in excess of the other forms. The fifth class comprises cells with five or more nuclear subdivisions and may be arranged into five groups, this class representing about 2 per cent, of the total neutrophiles. It is probable that these various classes represent the gradual development of the polymorphonuclear neutrophile, the older the cell the greater the tendency to reach Class 5, while in conditions associated with new formation of cells, as in infectious conditions, we find the percentages of the earlier classes being increased, that of the later ones diminished. Minor and Ringer1 have recently emphasized the prognostic value of this method in pulmonary tuberculosis. (J). Polymorphonuclear Eosinophiles. These cells, sometimes called the coarsely granular cells of Schultze, are somewhat smaller than the preceding, varying in size from that of a lympho- cyte to that of the neutrophile. Their protoplasm is usually somewhat less in amount than is that of the neutrophile and may not be distinct. It is filled with coarse, round, or slightly oval granules about one micron in diameter, which are very refractile and appear in the fresh specimen distinctly black, w'hile in the stained smear they take up the acid portion of the dye and are called, therefore, acidophilic, oxyphilic, or eosinophilic cells. These granules are the a granules of Ehrlich, and are sometimes associated with the p granules of Ehrlich which are about the same size or a little smaller than the a granules jand take both the acid and the basic stains, although with the ordinary staining solutions they appear, like the a forms, stained with eosin. The nuclei of these cells are coarsely reticulated, are larger and thicker than those of the neutrophiles, are usually bilobed, and more frequently show distinct separation of these lobes. These nuclei do not stain very deeply with the nuclear dyes, so that they may be rather indistinct. These cells ar^ probably derived from the ,mononuclear eosinophile myelocytes of the bone-marrow and constitute from two to four per cent, of the total number of leucocytes, their average number being from 100 to 200 cells. 'Amer. Jour, of Med. Sei., vol. 141, 1911, p. 638. 484 DIAGNOSTIC METHODS. (c). Polymorphonuclear Basophiles. These cells, frequently called mast cells, resemble the neutrophile cells in the fresh specimen, but show quite distinct characteristics in the stained form. Their protoplasm is much the same as that of the neutrophile and shows the same relation toward the nucleus which does not, however, so frequently form distinct lobes as does the nucleus of the neutrophile. The size of these cells averages about 10 microns, but shows a marked variation, being very small in myeloid leukemia. These cells are characterized by their granules, the v granulation of Ehrlich, which are more irregular in size than are the neutrophile granules and are not so extensively scattered through the protoplasm. These granules show the peculiar property of metachromasia, being colored red with violet dyes and with the blue dyes violet, although with absolutely pure methy- lene blue they take a blue shading. In cases of myeloid leukemia these granules are particularly soluble in water and may, therefore, not be seen if aqueous stains are used, while in normal blood the granules are usually water-fast. These cells stain best with Ehrlich's dahlia stain, or Turk's iodin solution, and take a tone which is not strictly basophilic, but resembles more that of mucin, on which account they have been called mucinophiles. Whether these granules are all of the same significance is questionable. The true mast cell granules are known as the y granules, while other basophilic granulations have been found by Ehrlich and are known as the delta (3) granules.1 These latter granules are found in the large mononuclear cells, especially in the lymphocytes, and do not stain with Ehrlich's dahlia stain. Whether these latter bodies are true granules or nodal thickenings is at present uncertain. True mast cells probably originate in bone-marrow from a granular mononuclear type corre- sponding to other types of myelocytes (Pappenheim). These cells constitute about 1/2 per cent, of the total leucocytes, averaging between o and 50 per cmm. Ehrlich's dahlia stain for these mast cells is made as follows: Distilled water, 100 c.c. Saturated absolute alcohol solution of dahlia, 50 c.c. Glacial acetic acid, 10 c.c. The specimens are heated or fixed by alcohol and are then stained in the above stain for 5 to 10 minutes, the mast cells appearing of a distinctly violet tone. Turk's lodin Method. The specimens are fixed with heat and are then first stained in a 1 per cent, alcoholic methylene blue solution, warming the slide very carefully until it steams. It is then allowed to cool, washed quickly in water, and dried between filter-paper. The slide is then covered with a solution of iodin in potassium iodid of the strength 1:300. Allow this solution to act not longer than 1/2, minute, pour off and mount in the following syrup: 1 Huguenin (Zentralbl. f. allg. Path. u. path. Anat., Bd. 23, 1912, S. 725) reports the presence of sudanophile granules in the mast cells in an obscure case. THE BLOOD. 485 lodin, i gram. Potassium iodid, 3 grams. Distilled water, too grams. Gum arabic q.s. to make a syrup. The mast-cell granules appear, when treated in this manner, very distinctly outlined and colored black. The nuclei are brownish in color, the erythrocytes yellowish-green, and the polychromatophilic erythrocytes dark green, the neutrophile and eosinophile granules faintly yellow. (C). Leucocytes in Pathological Blood. (1). Myelocytes. Under this heading we must regard any cell of the bone-marrow as a myelocyte, but for diagnostic purposes we have reference more to mononuclear cells which are distinctly granular. While the granulations of these cells are usually either neutrophilic or eosinophilic, we may rarely find, especially in myelogenous leukemia, cells which show basophile granulations. The size of these myelocytes varies from that of the red blood-corpuscle to that of a large mononuclear cell. (a). Neutrophile Myelocytes. This myelocyte may be either large or small, in the first case being known as Cornil's myelocyte (amblyochromatic type), a cell which is much larger than the polymorphonuclear leucocyte (at least 15 microns in diameter), and showing a round, pale, eccentric nucleus which stains feebly but not diffusely. This cell shows many distinct neutrophile granulations and has a narrow zone of basophilic protoplasm surrounding the nucleus. It is found almost entirely in myelogenous leukemia and has been called by Pappenheim the heteroplastic promyelocyte. The second type of neutrophile myelocyte is known as the Ehrlich myelocyte (trachyochromatic type of Pappenheim) which is a medium- sized cell with a pale central nucleus which stains deeply but not diffusely. This cell shows extensive neutrophilic granulation of the protoplasm, which is faintly oxyphilic, and has a nucleus which is either perfectly round, oval, or indented, but is never lobed nor pycnotic. A distinction between the myelocyte and the polymorphonuclear leucocyte should be based entirely upon the structure of the nucleus, all those cells with round, oval, or kidney-shaped nuclei which occupy at least one-half of the cell and show neutrophile granulations but no diffuse staining of the nucleus must be called myelocytes. It is to be remem- bered, however, that this type develops into the polymorphonuclear neutrophile, so that in abnormal blood we may have all gradations between these two types. (6). Eosinophile Myelocytes. These cells are exactly analogous to the preceding, with the exception that the granules of the more mature form show distinct eosinophilic tendencies. The younger forms of these granules may show a purplish-violet or even blue color, owing, as Simon states, to the fact that the young eosinophilic granule is 486 DIAGNOSTIC METHODS. physically cyanophilic and chemically amphophilic, whereas the mature granule is physically erythrophilic, but chemically absolutely oxyphilic. The size of these cells is more or less variable, so that it is probable that we have the two types of eosinophile myelocytes, corresponding to the large and small neutrophile myelocytes. These cells occur more frequently in leukemia, in association with tumors of the bone-marrow and in the pseudoleukemic anemia of children. These myelocytes may be of variable size, but are characterized by the large centrally-located nucleus, which is not clearly defined from that of the surrounding slightly basophilic protoplasm. The granules are distinctly basophilic and in some cases are very numerous, while in others they may be widely scattered through the protoplasm. These cells are practically never seen except in cases of severe splenomyelogenous leukemia, in which they may reach as high as 47 per cent, with an absolute count of 140,000 cells (Taylor). (c). Basophile Myelocytes. (2). Irritation Forms. These cells vary in size from a lymphocyte to a large mononuclear cell, resembling more nearly the former. They are mononuclear, nongranular cells, thus differing from the myelocyte which is always granular. The nucleus is round and eccentrically placed, showing a very slight chromatin network and staining with the triple stain of a bluish-green color while with the Roman- owsky dyes the color is a pale blue. The protoplasm is stained a deep brown with the triple stain and is thus differentiated from other forms of cells. With the methylene blue dyes the protoplasm appears more deeply stained than does the nucleus. These cells were first described by Turk and would seem to have the same significance as do the myelocytes, namely, an indication of marked activity of the bone-marrow. Pappenheim regards them as plasma cells and largely derived from the lymphocytes. (3). Degenerated Forms. Occasionally we find in normal blood degenerated leucocytes which stain poorly and show no granules. These may be even so much degenerated that they show as the so-called basket cells or "shadows." This condition is very frequently seen in severe infectious diseases, while in the acute leucocytoses which occur under many influences diminution in the number of neutrophilic granules as well as swelling and fragmentation of the bodies of the leucocytes is very common. The changes in the staining qualities of the nucleus seem to be the most significant of the lesions in the acute type of degeneration of the leucocytes. In chronic degeneration of the leucocytes we find hydropic degen- eration, which is frequent in the blood of chlorosis and, when the nuclei are involved in this degeneration, seems to be limited to certain cases of leukemia. Besides such changes we find fatty degeneration as well as glycogenic degenera- tion. This fatty degeneration is characterized by the appearance of fat globules in the leucocytes which stain with osmic acid and with Sudan-Ill. For this PLATE XX. lODOPHILIA THE BLOOD. 487 latter reason they have been styled sudanophiles, and have been very carefully studied by Buttini and Comesatti as well as by Cesaris-Demel. Iodophilia. In many pathologic conditions, especially in acute infectious diseases and in those associated with all types of sepsis, a so-called glycogen reaction or iodophilia may be demonstrated in the bodies of the leucocytes, as well as in certain extracellular granules. The technic of this method is as follows: An unfixed dry blood smear is exposed to the vapor of solid iodin until it is stained a brownish color. After the specimen is stained it is mounted in the syrup described on page 485 and is examined with an oil-immersion lens. The blood of normal individuals, stained by this method, shows the protoplasm of the leucocytes of a bright yellow, while the nucleus takes on a much lighter tint. In pathologic blood two types of reaction can be noted. The intra- cellular one, which is of greater clinical importance, shows a more or less marked diffuse brown color of the entire protoplasm of the leucocyte, or the protoplasm contains reddish-brown granules which may be more or less distinct. The extracellular reaction is evident in the blood plates, while the intracellular type is more particularly confined to the neutrophiles, although the mononuclear leucocytes may occasionally be tinged brown. Much difference of opinion exists as to the nature of this brown-staining substance, Ehrlich regarding it as glycogen, while Czerny considers it as an antecedent of amyloid, and Goldberger and Weiss regard it as peptone. Kaminer regards this reaction as a degenerative change and not as an evidence of regeneration. While this reaction has little value in differentiating infectious or septic conditions one from the other, it is sometimes of importance in making a diagnosis between purulent and nonpurulent affections, being present in the former and absent in the latter. These granules have been found by Hofbauer in pernicious anemia, secondary anemia, and leukemia, but not in chlorosis or pseudoleukemia. This reaction is observed in pneumonia, but is seldom seen in tuberculosis, typhoid fever and diphtheria. It is, however, not to be regarded as dependent upon infection, as it occasionally obtains in non-infectious conditions. (D). Differential Counting of the Leucocytes. By a differential counting of the leucocytes is meant the counting of the different varieties of the leucocytes found in the stained smear and the calcula- tion of each type in terms of percentage. The technic of this method is that previously outlined and consists in making an even smear upon a glass slide and staining it with any of the stains previously mentioned, noting that the triple stain does not bring out the granulations of the leucocytes with the exception of those of the neutrophiles. It is self-evident that the larger the number of leucocytes counted the greater will be the possibility of arriving at true percentage relations. It is wise, therefore, to count at least 250 of these cells, and in many cases, to extend this to 500. If the smear is even and the leucocytes well distributed throughout, 100 cells will frequently suffice. 488 DIAGNOSTIC METHODS. For a differential count a satisfactory classification is an absolute essential. As none of the systems at present advanced are entirely adequate, we still use the classification of Ehrlich, which is as follows: Small mononuclears, large mononuclears (including the transitional), polymorphonuclear neutrophiles, eosinophiles, basophiles (mast cells), and myelocytes. The characteristics of these cells have been previously given, but may be summed up in this connec- tion. By a small mononuclear is meant any nongranular mononucleated cell smaller than a'polymorphonuclear neutrophile. A large mononuclear is any nongranular cell with a round or oval nucleus and larger than a polymorpho- nuclear neutrophile. A cell of the same description and size but with an indented nucleus is a transitional form. The polymorphonuclears are cells which are about 10 microns in size and show a diffuse granulation, which may be either neutrophile, eosinophile, or basophile. It should be remembered that the mast cells appear as nongranular forms when the triple stain is used, so that the characteristics of the nucleus in its relation to the protoplasm must be borne in mind. The percentage relations of these cells in normal blood are as follows:1 Percentage. Number per cmm. Small mononuclears, 20-25 1200-2000 Large mononuclears, 3-5 200-400 Polymorphonuclear neutrophiles, 65-75 5000 Polymorphonuclear eosinophiles, 2-4 100-200 Polymorphonuclear basophiles, 0-1/2 0-50 In the writer's laboratory differential staining is usually carried out with the use of the Wright or Giemsa stain as he has found the usual triple stain quite unreliable. Although the granular differentiation is not as distinct as could be desired, yet one soon becomes accustomed to the staining of various cells so that it is not a matter of great difficulty to distinguish the various types. In many cases it is not the easiest matter to distinguish myelocytes from the small mononuclear types, but careful study will usually clear up any obscurities which exist. The writer cannot go into the variations which these leucocytes suffer in pathological conditions, but must leave the percentage relations in disease to later subdivisions. (E). Number of Leucocytes. The normal number of leucocytes in a cmm. of blood has been given various figures. As a rule, it may be said that anything above 10,000 leucocytes per cmm. should be considered pathological, the normal variation running from 5,000 to 9,000 cells. In estimating the normal number of white cells, both in health and in disease, a large number of factors which influence these cells must be taken into consideration. Thus vasomotor phenomena, varia- tions in the volume of the plasma, inflammatory processes, state of digestion, age, variations in different parts of the circulatory system, and many different 1 See Bunting, Amer. Jour, of Med. Sci., vol. 142, 1911, p. 698; also, Galambos, Folia Haemat., Bd. 13, 1912, S. 153. PLATE XXL Polynuclear Leucocytosis. (Wright's Stain.) THE BLOOD. 489 disease processes usually bring about an increase in this number, while many pathological conditions are associated with a reduction. An increase in the number of white cells is usually spoken of as a leucocytosis, but it must be remembered that such an increase may be purely physiological and should be sharply differentiated from a pathological increase which is clinically the more important. This increase is usually referable to the increase in the number of the polymorphonuclear neutrophiles, while an increase in the other varieties of cells is spoken of as a lymphocytosis, myelocytosis, esoinophilia or an eosino- philocytosis, or a mixed leucocytosis. A diminution of the number of polynuclear neutrophiles is designated as a leucopenia, which is the more usual form in which reduction of the cells occurs. It is to be remembered that these conditions may be transitory and symptomatic pointing to a purely physiological process, while a more permanent and more marked increase or decrease in the number should be considered pathological. It is rare that we find the increase limited absolutely to one variety of cell, the increase in the others being less marked. The absolute number of these cells is much more to be regarded than their percentage relation, as with an increased number of leucocytes the actual number of some of these varieties may be increased, although the percentage may be diminished; while with a low leucocyte count the percentage may be increased and the absolute number diminished. It is wise, therefore, to report not only the percentage relations of the leucocytes, but also the actual number of these cells per cmm. This will correct mistaken ideas as to an apparent increase or decrease in any partic- ular variety of cell. Thus a differential leucocyte count may show a percentage of 50 for the neutrophiles and at the same time an actual number of 10,000 per cmm., giving rise to confusion as to the actual relation of these important leucocytes. Leucocytosis. (Polymorphonuclear Neutrophiliosis). As stated above, anything above 10,000 leucocytes per cmm. should be regarded as a leucocytosis. This is, however, relative and should not be considered pathological without taking into consideration all of the physiological and pathological influences. Regarding the theoretical cause of leucocytosis, little is definitely known. The influence of infectious processes is such as usually to increase the number of leucocytes in the blood, as an attempt on the part of the system to overcome by phagocytosis the action of the bacteria of the various diseases. Yet we find in some of these infectious processes, notably in typhoid fever, a marked reduction in the number of the white cells, although the bacillus typhosus is present in large numbers in the blood at the same time. We must, therefore, assume some specific influence upon phagocytosis and chemotaxis, as a general infection is not necessarily associated with an increase in the number of white cells, but is dependent more upon the specific nature of the infection. The work upon opsonins and vaccine therapy may open up an entirely new field in our study of this subject. The leucocytosis shown in 490 DIAGNOSTIC METHODS. noninfectious conditions is still a matter of much dispute and has probably more to do with variations in the plasma than with direct increase in the number of white cells. According to Limbeck,1 the following classification of the various leuco- cytoses is the most comprehensive: Classification. (i). Physiological leucocytosis. (a). Leucocytosis of digestion. (b). Leucocytosis of pregnancy. (c). Leucocytosis of the new-born. (2). Pathological leucocytosis. (a). Inflammatory leucocytosis. (b). Leucocytosis associated with malignant tumors (cachectic leucocytosis). (c). Posthemorrhagic leucocytosis. (d). Agonal or antemortem leucocytosis. (3). Leucocytosis following medicinal and therapeutic measures. (4). Leucocytosis from other causes. (1). Physiological Leucocytosis. (a). Leucocytosis of Digestion. Normally a leucocytosis of the polynuclear type will be observed beginning about one hour after a meal rich in proteins and will reach its maximum in from three to five hours.2 The actual figure reached varies in different persons, being usually an increase of about one-third, the maximum often reaching 15,000 cells, but more usually not much over 10,000. In this increase the small mononuclear cells may be absolutely as well as relatively increased. In some persons this leucocytosis of digestion does not appear, which fact may be referable to marked torpidity of the intestines, to a prolongation of the process of digestion, or to a large absorption of fluids. It has been found that a highly albuminous diet has a much more marked influence upon this leuco- cytosis than does a diet of vegetables and fat. The rapidity of absorption and of digestion must be taken into consideration as factors which influence the appearance or nonappearance of the leucocytosis. In children the increase in cells is much more marked than in the adult, due probably to the increased digestive and absorptive powers, providing the food taken is, as it should be, easily digestible and absorbable. Various pathologic conditions influence this digestion leucocytosis. Thus Muller has found that in cases of carcinoma of the stomach a digestion leuco- cytosis is rarely observed after heavy meals. There are, however, a few cases of gastric cancer in which a slight increase has been observed. This failure of leucocytosis is probably due, as Schneyer shows, to lessened absorption as 1 Jena, 1896. 2 See Brasch, Zeitsch. f. exp. Path. u. Ther., Bd. 10, 1912, S. 381. THE BLOOD. 491 a result of involvement of the lymphatics rather than to a malignant stenosis of the pylorus, as the benign stenoses are usually associated with a leucocytosis. That the lack of leucocytosis is due more to diminished absorption than to lack of digestive power is shown by the usual occurrence of a digestion leucocy- tosis in ulcer of the stomach, chronic gastritis, and in dyspeptic conditions. Just why carcinoma, with or without stenosis, should be associated with a normal or subnormal number of leucocytes after digestion is hard to explain, in view of the fact that these other conditions are associated with a leucocytosis. Practically all absorption of food material occurs from the bowel and we would naturally expect to find disorders of the intestinal canal leading more frequently to a normal leucocyte count than to a leucocytosis. Very little detailed work has been done upon the influence of enteric troubles upon the leucocyte count, but the writer has seen several cases in which a digestion leucocytosis did not occur and in which the findings both ante- and postmortem were entirely related to the bowels. The examination of the blood may, how- ever, clear up a diagnosis of carcinoma, but will not always permit of a differen- tial diagnosis between this condition and pernicious anemia. It is to be remembered that a leucocyte count must be frequently made in order that this test may be of any value whatever and that a considerable rise only is to be taken as evidence of a leucocytosis. Patients with cancer either of the stomach or of other viscera frequently show a leucocytosis as the result of cachexia and this should be remembered in the interpretation of a leucocyte count following digestion. (b). Leucocytosis of Pregnancy. It has been shown that from 50 to 75 per cent, of cases of pregnancy are associated with a leucocytosis, averaging about 15,000 cells per cmm. This is especially true of primiparae, but is often shown in the multiparae. Just what cause can be given for this rise is uncertain. It is more than likely that a condition of slight intoxication is present due to the overloading of the blood of the patient with the products of metabolism of the fetus. These substances are not normal to the blood of the woman and in consequence act as foreign bodies which may attract, by their chemotactic influence, the white cells and bring about an increase. As these cases of pregnancy show an ab- sence of a digestion leucocytosis it has been assumed that the increase in leucocytes is due to a prolonged digestion leucocytosis, but this does not seem probable. The changes in the breasts and in the uterus during this period have suggested these organs as the factors influencing this leucocytosis, but such inferences do not seem to be directly warrantable. The leucocytosis of pregnancy is a mixed leucocytosis, all of the various types of leucocytes, with the exception of the eosinophiles being increased. After the birth of the child the leucocytes gradually diminish in number and normally reach their usual values in from four to fourteen days after delivery. Should complica- tions, such as postpartum hemorrhage or septic fever arise, the leucocyte 492 DIAGNOSTIC METHODS. count may remain high until these complications have subsided. In multi- parae these changes in the number of leucocytes are not so marked as in the case of the primiparae, but a slight rise always occurs. This fact has been attributed to the lessened reactivity of the organisms to the influence of the toxic substances thrown into the blood from the cells of the fetus. (c). Leucocytosis of the New-born. As Askanazy has found, the blood of the fetus shows a diminution in the number of leucocytes, owing to the fact that there is no function as yet estab- lished for these cells in utero. In the blood of the new-born, however, a leuco- cytosis running from 15,000 to 20,000 cells may be observed, these figures going as high as 40,000 under the influence of the first feeding. As the weight of the child begins to diminish these cells are markedly reduced in number to about 8,000, and are subsequently increased to 10,000 as the child begins to gain in weight. The high leucocyte count at birth is more probably due to the rapid blood formation than to a concentration of the blood, although this latter factor as well as that of the influence of digestion must be taken into consideration. The increase in the number of cells is chiefly limited to that of the small mononuclear cell, the differential count of the leucocytes showing during the early periods of life from 40 to 60 per cent, of the total number. This leucocyte count, if taken at the moment of birth, will not, however, vary much from that of the adult, the change becoming more marked in the early periods of life. (2). Pathological Leucocytosis. In most cases of inflammatory nature, of acute infections and general febrile diseases, there is observed an absolute increase of the polymorphonuclear neutrophiles, which runs more or less parallel to the temperature. This increase may run from 10,000 to 50,000 cells and diminishes as the influence of the inflammatory process is diminished. In general it may be said that leucocytosis represents the reaction of the individual to the disease. A high count may mean a vigorous reaction to the infection; a low count may mean either a poor reaction and hence an unfavorable condition of the patient, or it may indicate a very mild degree of infection with a normal reactivity of the patient. It must be said that all diseases of an infectious nature are not necessarily associated with a leucocytosis. For instance, pneumonia shows a leucocytosis which runs parallel to the degree of virulence; measles, influenza, malaria, and tuberculosis are rarely if ever associated with a leucocytosis unless complications arise or the conditions become very severe. In typhoid fever we usually find a leucopenia which, however, is associated with a relative lymphocytosis; if complications such as perforation arise a leucocytosis may appear. Among the conditions causing leucocytosis are acute lobar pneu- monia, the count running between 20,000 and 100,000, depending upon the severity of the infection and the degree of resistance against the infection. (a). Inflammatory Leucocytosis. THE BLOOD. 493 Acute articular rheumatism, diphtheria, acute cerebrospinal meningitis, follicular and suppurative tonsillitis, scarlet fever, mumps, rabies, erysipelas, ulcerative endocarditis, small-pox, cholera, general pus infections of the serous membranes and of the mucous membranes, acute bronchitis, and many other conditions cause a leucocytosis of the polymorphonuclear type. The varia- tions in these many conditions will be discussed under the Pathology of the Blood. The exact cause of this leucocytosis following inflammatory processes is still a matter of much discussion. The bone-marrow has been shown to be markedly increased as regards cellular proliferation in the early stages of the inflammation and may be so markedly changed as to cause permanent de- rangement in the functions of this blood-forming organ. That leucocytosis has much to do with 'immunity both from the standpoint of phagocytosis and from that of Ehrlich's side-chain theory cannot be questioned. A leuco- cytosis must represent the attempt on the part of nature to rid the blood and the system of the bacterial and toxic products of the disease. Whether phago- cytosis as such or under the influence of various opsonins is the direct cause of immunity and of recovery from any specific infection must be left to a later chapter. Suffice it to say that the work upon Antitoxins in the acute in- fections and of the vaccines in the chronic suppurating types of disease is bear- ing remarkable results. This variety of leucocytosis is the least uncertain. The cases of carcinoma and of sarcoma show a leucocytosis which is not definite for the particular kind of cancer, but is more usual with the latter than with the former type. The leucocytosis is usually one of the polymorphonuclear type, but is frequently associated with an increase in the number of mononuclear cells. This leuco- cytosis has no direct relation with the situation of the tumor and is not always present in all cases of cancer. Whether this leucocytosis be due to an intercur- rent infection is a question which must be left for more detailed work, but it would seem wise to accept Ewing's statement that a marked leucocytosis in the course of a cachexia, from tertiary syphilis, tuberculosis, nephritis, and in the majority of cases of carcinoma, should suggest a search for a complicating infection, while in the sarcomata a leucocytosis is much more common as a direct result of the disease. (&). Cachectic Leucocytosis. (c). Posthemorrhagic Leucocytosis. A well-marked leucocytosis, which may begin in from ten to fifteen min- utes and may reach as high as 20,000 cells within an hour, has been often ob- served following extensive acute hemorrhages. The leucocytosis in these cases bears a general relation to the extent and rapidity of the loss of blood and usually disappears or diminishes long before regeneration of the blood has occurred. This leucocytosis is of the polymorphonuclear type and is referable rather to the sudden outflow of lymph which occurs as a compensation for the 494 DIAGNOSTIC METHODS. loss of fluid than to a new production of cells, as this latter process does not take place for some time. In hemorrhages which are slight and long-continued as in cases of gastric or intestinal ulcer, the duration of the leucocytosis as well as its extent is very brief. Stassano and Billou assume from their work that a hypoleucocytosis follows a severe hemorrhage while a true leucocytosis is observed after the loss of small quantities of blood. These findings seem to be rather doubtful, except where the hemorrhage has been so extensive as to cause death. (d). Antemortem Leucocytosis. This form of leucocytosis has been questioned, especially by Arneth, but there seems to be little doubt that such a leucocytosis may occur if death does not take place too rapidly. In some diseases the leucocytes do not fall in number, but in some a distinct rise is noted which has been attributed by Ehrlich to the accumulation of white cells along the periphery of the blood- vessels as a result of slowing or stasis of the circulation. This type of leuco- cytosis is usually of the polymorphonuclear variety and may support the view of Limbeck that antemortem leucocytosis, when it does occur, is the result of a terminal infection, although it cannot explain those cases which show a lymphocytosis rather than the ordinary leucocytosis. The character of the antemortem leucocytosis must depend largely upon the precedent condition and will be associated with antemortem dissemination of bacteria, ante- mortem hyperpyrexia, vasomotor paralysis, serous exudation, and many other causes whose influences are not well established. (3). Leucocytosis Following Therapeutic Measures. (a). Drugs. It has been found following administration of tonic drugs, ethereal oils, myrrh, turpentine, camphor, peppermint, quinin, and other drugs, that a leucocytosis of a more or less extent occurs, which is probably referable to the same cause as is digestion leucocytosis. Many of these drugs if applied locally for the purpose of counterirritation were shown to have the same effect. Extracts of tissues, especially those containing large amounts of nucleinic acid and of purin substances, produce an extensive leucocytosis, which fact has been taken advantage of in the administration of the nuclein substances as therapeutic remedies. In the case of those drugs which destroy the red cells, such as the coal-tar antipyretics, chlorates, and illuminating gas, a normal number of white cells is usually observed, although Simon wrongly states that a hyperleucocytosis follows the use of such drugs. After the pro- longed use of chloroform or ether a polynuclear leucocytosis is generally observed which is usually of short duration. In this connection it is well to remember that an increase in the number of leucocytes of 10,000 to 15,000 above the normal value of the individual should be regarded as evidence of infection if this increase is sustained for more than a few hours. THE BLOOD. 495 (b). Baths. After a cold bath the leucocytes of the polynuclear variety have been shown to be increased from 100 to 300 per cent. This is true only if the bath is of moderate duration, a prolonged cold bath taken to the point of exhaustion will diminish rather than raise the number of white cells. Experiments on the result of hot baths have shown just the reverse condition, namely, a hot bath of short duration produces a decrease, while one of long duration causes an increase in the number of white cells. Massage in itself following either a hot or a cold bath tends to increase the number of leucocytes. (c). Exercise. Prolonged muscular exercise, as taken in the form of gymnasium work, as applied in the various therapeutic movement treatments, and also as given in the more violent athletic contests produces a rise in the number of leucocytes which is temporary and is characterized more by an increase in the number of polynuclear cells than of any other variety of the leucocytes. Cyanosis, passive hyperemia obtained by the method of Bier, shock whether physical or mental, injection of various toxins, such as Koch's tuber- culin, the various autovaccines, and injection of various organic principles, such as peptone, pus, organ extracts, etc., have all been shown to produce a polynuclear hyperleucocytosis, either by causing stasis of the circulation or by increasing the chemotactic power of the serum. The exact explanation of chemotaxis is still in doubt and we find that some substances are positively and others negatively chemotactic, thus accounting for the lack of phagocy- tosis in certain conditions, such as pneumonia, although in these same condi- tions we may have a leucocytosis as a direct result of an infection. From the work of Rosenow we learn that the phagocytic action of the leucocytes in pneumonia is increased by the addition of an extract of the active pneumo- cocci; whether this extract increases the chemotactic power of the bacilli in the blood or whether certain restraining influences are overcome is still unsettled. Future work upon phagocytosis and upon the factors governing increased op- sonic power in the various infections may show us what causes are effective in such conditions. (4). Leucocytosis from Other Causes. In the previous discussion of leucocytosis we have had reference more to an increase in the number of polynuclear cells, this type being more properly called polynuclear hyperleucocytosis. It is very rare that this condition exists in the absolutely pure state, being frequently associated with an absolute or relative increase in the number of lymphocytes. Mixed Leucocytosis. By this is meant an increase of both granular and of nongranular cells of various types, its more common interpretation, however, being a leucocytosis characterized by an increase in the number of neutrophile myelocytes. This neutrophilic myelocytosis is best seen in leukemia in which the absolute number of myelocytes may reach 150,000, the increase in no other condition, according 496 DIAGNOSTIC METHODS, to Ehrlich, rising above 1,000 cells. The increase in the myelocytes in leuke- mia is not limited to those of the neutrophile variety, but is commonly associated with a marked increase in the number of eosinophile types as well as with an increase in the number of mast cells. In pernicious anemia we may also find a large increase in the number of these myelocytes. These cells when in- creased in number are clinically significant of exhaustion of the bone-marrow, but should never be interpreted in this way unless they remain while the leu- cocyte count is falling; in other words, they are significant only when constitut- ing a large percentage of the leucocytes. Lymphocytosis. A relative or absolute increase in the number of lymphocytes is of fre- quent occurrence and has occasionally some significance. In judging of this condition the actual number of cells present as well as their numerical rela- tions to the other varieties of leucocytes must be considered. The term lymphocytosis must always be applied to an absolute increase in the number of these cells, the normal number being between 1,200 and 1,500 per cmm. Physiologically we find this condition in infants and during a digestive leucocy- tosis in the adult. The child shows at birth from 50 to 65 per cent, of these cells which percentage gradually diminishes, reaching the normal percentage (20-25) about the age of puberty. Pathologically this condition is observed in poorly nourished children, those showing the "constitutio lymphatica," rachitis, whooping-cough, gastro- intestinal disturbances in the child, cervical adenitis, splenic tumors, in most infectious diseases of children, and especially in lymphatic leukemia, in which as high as 90 per cent, of a largely increased leucocyte count is referable to the small lymphocytes. In the splenomyelogenous form of leukemia and following removal of the spleen we observe a steady increase in the number of lymphocytes continuing during the first year. It must be remembered that a leucocyte count may be low and yet a lymphocytosis exist. This is shown in typhoid fever, amebic dysentery, chlorosis, pernicious anemia, scurvy, and other conditions in which the leucocyte count is low, the granular cells being dimin- ished in number, but the lymphocytes being increased.1 In the congenital and also secondary acquired syphilis we find a lymphocytosis which must be re- ferred to involvement of the lymph-glands. The worker must not be led into a mistaken interpretation of his blood-findings, as enlarged lymph-glands may give the same picture of lymphocytosis as is seen in Hodgkin's disease, chronic and acute lymphatic leukemia. Eosinophilia. By this is meant an absolute increase in the number of eosinophilic cells. The average number of these cells is between one and two hundred per cmm., hence the term eosinophilia should be limited to those cases showing counts 1 Borchardt (Deutsch. Arch. f. klin. Med., Bd. 106, TQ12, S. 182) shows that a lympho- cytosis obtains in a large majority of cases of disease of the thyroid, hypophysis and suprarenal gland. This is oftentimes associated with a leucopenia. Cabot (Amer. Jour, of Med. Sci., vol. 145, 1913, p. 335) calls attention to the occasional occurrence of a lymphocytosis in infections usually associated with a polynucleosis. PLATE XXII. Eosinophilia. (Wright's Stain.) THE BLOOD. 497 something above 250. We find a physiological eosinophilia during childhood, the average increase being about 1 to 2 per cent, above the normal adult find- ing. No physiological relations have been established between eosinophilia and sex, pregnancy, menstruation, digestion, or old age; however racial distinc- tions are sometimes shown by eosinophilia, the natives of southern China showing between 15 and 20 per cent, of the leucocytes as eosinophiles. We observe variations in the number of eosinophiles in various affections of the bone-marrow. Thus in splenomyelogenous leukemia, we may find these cells increased as high as 30,000 per cmm. According to Ehrlich, a diagnosis of this form of leukemia is warranted only when we have an increase in the number of eosinophiles, but it cannot be doubted that cases are relatively frequent in which these cells are not increased. In sarcoma of the bone-mar- row as well as in osteomyelitis and osteomalacia, we find these cells usually increased in number. In chlorotic conditions we may find the eosinophiles moderately increased, while in secondary anemias especially those following infection with parasites a marked eosinophilia may be observed. After ex- tirpation of the spleen as well as in cases of chronic splenic tumor we find an eosinophilia which may run from 10 to 40 per cent, higher than the normal, constituting as high as 40 per cent, of the total number of leucocytes. Whether or not a pure disease of the lymph-glands is associated with the eosinophilia is unsettled, but if metastases have extended to the bone, an eosinophilia of marked degree is usual. True bronchial asthma is always associated with an increase in the num- ber of eosinophiles, one case of Billings showing 54 per cent, of the total leuco- cytes as eosinophiles. This relationship in asthma is of importance from a diagnostic standpoint as asthmatic attacks from other causes are not asso- ciated with an eosinophilia. It is true that in emphysema a marked eosino- philia does occur so that we may find, in cases in wrhich this condition complicates a tuberculosis, that an eosinophilia obtains. Tuberculosis of the lungs or of other tissues does not show an eosinophilia unless complicated by emphy- sema, cachexia, or secondary infection.1 A large number of skin diseases are associated with eosinophilia, the highest count being reported by Zappert in a case of pemphigus showing 4,800 cells per cmm. The occurrence of this condition in skin diseases depends not so much upon the nature of the lesion as upon its extent, intensity and lack of healing tendency. Many skin lesions are known to be produced by toxic agents which have special chemotactic influences over the eosinophile cells; this fact may account for the eosinophilia in such conditions. The most usual skin diseases showing this increase are pemphigus, eczema, psoriasis, urticaria, purpura, scleroderma, lupus, leprosy, herpes zoster, and general gouty affections. It is a general rule that the infectious fevers are not associated with eosino- Pathological Eosinophilia. 1 See Ameth, Deutsch. Arch. f. klin. Med., Bd. 108, 1912, S. 323, for the relations of the eosinophiles in pneumonia. 498 DIAGNOSTIC METHODS. philia, but with the acute polynuclear leucocytosis.1 In scarlet fever, however, we find that eosinophiles are frequently markedly increased, while in acute rheumatism and in malaria these cells are usually present in more or less increased numbers.2 In gonorrheal infections the eosinophile leucocytes are very frequently marked in the discharge in the early days of infection, diminish- ing in number as the number of neutrophiles increases, and increasing again as the discharge clears up. It is generally believed that the eosinophile cells in the blood are increased coincidently with their increase in the gonorrheal discharge. It has been shown that the eosinophiles are usually increased in all forms of ovarian disease with the exception of cancer. In syphilis a uniform increase of the eosinophiles has been observed, but some cases do not show such a regular finding. Outside of the splenomyelogenous form of leukemia and the true bron- chial asthma, infection with various parasites is accompanied by the most pronounced eosinophilia. Any parasite from the harmless pin-worm to the most malignant uncinaria may cause an eosinophilia (Emerson). This eosinophilia is not necessarily constant nor does its extent bear any relation to the severity of the infection. A differential diagnosis between typhoid fever and trichinosis is frequently possible on the basis of the marked eosinophilia in this latter condition, although eosinophilia does not always obtain here. Mast-cell Leucocytosis. In myelogenous leukemia we find these cells increased to a very large extent, often outnumbering the eosinophile cells. An increase of these cells is generally regarded as the sole isolated pathognomonic sign of this disease. Their increase may be as high as 20 per cent, of the leucocytes. These cells have been found in a few cases of cancer, tuberculosis, syphilis, and other lesions of the skin, while in bone disease complicated by septic infection a few reports of positive findings have been made. This is a condition characterized by a reduction either in the total number of leucocytes or in one or more groups of leucocytes. The most usual condition showing a leucopenia is typhoid fever in which we find the polynuclear cells markedly diminished, thus lowering the leucocyte count, while the large mononuclear cells are relatively increased. Anything below 5,000 cells is regarded as a leucopenia. It must be considered in this connection that the typhoid leucopenia remains only when the disease is limited to the intestinal canal; when other organs become involved a leucocytosis supervenes. The count in typhoid fever may run as low as 2,000 cells, while in tuberculosis of the lymph-glands it may reach 500 cells. In cases of starvation or malnutrition the leucocyte count is always low and in chronic intoxication with the heavy metals, morphin, alcohol, and cocain, we observe a very low count, as a rule. Leucopenia. 1 Schlecht and Schwenker, Deutsch. Arch. f. klin. Med., Bd. 108, 1912, S. 405, show the relationship of eosinophilia to anaphylaxis. See, also, Herrick, Arch, of Int. Med., vol. 11, 1913, p. 165. 2 See Ebbell (Norsk Mag. f. Laegeviden., Bd. 73, 1912, S. 617) for a very naive theory of the relation of eosinophilia to the parasitic diseases. THE BLOOD. 499 A general rule to be observed in typhoid fever is that an increase in the number of white cells following a low leucocyte count is evidence of complications or of a mistaken diagnosis. It must be remembered that a relapse in typhoid may bring on a hyperleucocytosis, but this must occur during the afebrile period, otherwise a leucopenia will remain. In measles also we observe a leucopenia following the eruption and a leucocytosis preceding the eruption. This leucopenia affects the polynuclear neutrophiles while the lymphocytes are relatively increased, with disappearance of the eosinophiles. In cases of uncomplicated influenza we usually find a diminution in the number of leucocytes, although a normal number may obtain. This enables one to dif- ferentiate influenza from pneumonia, in which latter condition a marked leucocytosis is practically always present. In pernicious anemia and in splenic anemia we find a marked leucopenia during the active periods of the disease, the count running, as a rule, between two and three thousand cells. (F). Variations in Infancy and Childhood. In the first few days after birth the leucocytes may reach as high as 20,000, while in the nursing period the cells average 13,000. In this connection the influence of a digestion leucocytosis must be remembered as the cells may reach as high as 25,000 of which the lymphocytes constitute about 60 per cent. As the age of the child increases, the number of leucocytes gradually diminishes until the age of puberty, when the average number is about 8,000. As the development of the child progresses, the number of lymphocytes is gradually diminished and that of the neutrophiles correspondingly increased. The relations between the lymphocytes and the neutrophiles must be constantly remembered in making a blood count, as the percentage of lymphocytes is high at first and then gradually diminishes until the age of puberty; the neutrophiles during the same period are at first low, reaching the normal 65 to 70 per cent, about the age of 15. There are no marked differences in the morphology of the granular and of the nongranular cells of the child as compared with those of the adult. The characteristic changes in the leucocytes of the blood of the growing child are more related to the changes consequent upon development and show no signs of degenerative changes. This state- ment has reference of course to the normal healthy child, the variations under the influence of the various infectious diseases being quite marked. These latter changes will be discussed under a separate heading. (G). Functions of the Leucocytes. The functions of the leucocytes are largely related to their powers of overcoming the effects of bacterial processes. These results are accomplished both through their ameboid powers and their characteristic property of phago- cytosis. This latter property is largely influenced by chemotaxis as well as by the indefinite increase of so-called opsonins. Just what we are to regard as the basis of the opsonic index of the leucocyte is unsettled, but it is certain that the same leucocyte may show marked variations in its power of absorbing 500 DIAGNOSTIC METHODS. bacteria of different types. Whether this opsonic power is related in any way to the various assumptions embraced by Ehrlich's side-chain theory must be left for a later section. Besides the above important functions of the leucocytes, these cells show oxidizing, reducing, and fermentative powers, all of which are of more or less importance in the study of immunity and of reaction to bacterial infection. As stated (p. 478) the oxygen catalysts of the blood include oxidase, peroxidase, catalase and hemoglobin. To these ferments we owe many chemical tests for blood and pus, which are destroyed if the speci- men be previously boiled, while hemoglobin is still capable of inducing certain oxidations. Kastle and Amoss1 have studied the peroxidase activity of the blood in health and disease, while Kastle2 has made an exhaustive study of the oxidases. Winternitz and Meloy,3 Winternitz and Pratt,4 and, especially, Winternitz, Henry and McPhedran5 have turned their attention to catalase. The technic of these latter workers is as follows: 0.025 c.c. of blood is taken from a puncture of the ear by means of a specially graduated pipet. This is immediately diluted with 10 c.c. of distilled water, giving a dilution of 1: 400 Five c.c. of this diluted blood are placed in each of two 100 c.c. salt-mouth bottles, one being used for the test, the other being held as a check. In one is placed a small vial containing 5 c.c. of neutralized commercial hydrogen peroxid (3 per cent.). The large bottle is connected with a gas buret, to collect the gas formed by the action of the catalase upon the H2O2. The bottle is agitated for one minute, readings being taken every 15 seconds. Eighty per cent, of the normal cases studied show a liberation of 14 to 17 c.c. of oxygen in 15 seconds, this being constant for the same individual for a period of many months. The importance of this test rests in the variations from day to day or in any change from the normal activity of the individual's blood. The amount of gas liberated is, therefore, a personal rather than a general factor. Tscher- noruzki8 has recently shown that 1 gram of leucocytes is capable of liberating 312.9 c.c. of molecular oxygen under the influence of catalase. The diagnostic value of this test in disease is as follows: In typhoid fever there is no change in the early stages, but toward the third week there is a gradual fall accompanying the anemia. During the course of lobar pneumonia a slight decline may be noted. Diabetes mellitus shows no change. In diseases of the thyroid, the catalytic activity is not constant from day to day, in hyperthyreosis a tendency to increase being noted while in hypothyreosis a lower level than normal is observed. In renal disease we find no change as long as no marked renal insufficiency obtains, but the activity becomes irregular from day to day and, as uremia approaches, becomes much lower than normal. With retention of urine due to obstruction of the lower urinary tract a marked decline is observed. Little change is noted in cardiac cases even in the severe 1 Bull. 31, Hyg. Lab. U. S. Pub. Health and Mar. Hosp. Serv., 1906. 2 Bull. 59, Hyg. Lab. U. S. Pub. Health and Mar. Hosp. Serv., 1909. 3 Jour, of Exp. Med., vol. 10, 1908, p. 759. 4 Jour, of Exp. Med., vol. 12, 1910, p. 1 and 115. 5 Arch, of Int. Med., vol. 7, 1911, p. 624. 6 Zeitsch. f. physiol. Chem., Bd. 75, 1911, S. 216. THE BLOOD 501 types. The toxemias of pregnancy are separable into two classes by this test; without renal involvement eclampsia shows no change but, when renal in- sufficiency obtains, a marked decline prevails. This test should be of much value in diagnosis. (7). Blood-plates. These bodies which have been called "third corpuscles" are probably not true cellular entities. Hayem has considered them as the direct forerun- ners of the erythrocytes and has styled them, therefore, "hematoblasts." These bodies have been variously called plaques by Osler, blood-plates by Bizzozero, and by Arnold fragments of cells. These so-called corpuscles are small colorless bodies containing no hemoglobin. They are about 3 microns in diameter, round, oval, or rod-shaped without any biconcavity. They appear bluish, homogeneous or occasionally granular, and stained lightly by both basic and acid dyes. They do not contain any nucleus or membrane and become hyaline and glassy as soon as removed from the vessel, but on standing they become pale and unite to form granular masses. They are very sticky and adhere very extensively to one another, forming masses from which fibrinous threads radiate. This fact has lead to the belief that they have an important part in the formation of fibrin. (See Duke.1) Specimens of the platelets are best obtained by puncturing the tip of the finger or the ear through a 10 per cent, solution of sodium metaphosphate. In this way the blood becomes at once mixed with the fixing fluid and the drop may then be placed on a slide and covered with a cover-glaso. It has been customary to call anything a platelet which is smaller than a red-blood cell and which does not contain hemoglobin. The term platelet, however, should be reserved more particularly for those bodies which have a peculiar bluish refractility, no nucleus, show marked cohesive properties, and soon disintegrate. (a). Appearance The normal size of the blood-platelet averages about 3 microns, although Preisich and Hein have reported them as high as 7 microns. As a rule, their size varies inversely as their number. Some of these bodies show clear areas either in the center or on one side, or on the whole periphery; others become crescents, triangles, spindles and threads (Emerson). (&). Size. (c). Number. The normal number of platelets per cmm. is approximately 250,000, this number varying in the same person at different times of day. The physio- logical factors influencing the number of these cells are not well understood and in pathological conditions we may find large or small numbers of these cells. It is generally accepted that they are increased in anemias from any cause and may be related to the red blood-cells as 1 to 10. They are usually normal in chlorosis, increased in severe secondary anemia and decreased in per- 1 Jour. A. M. A., vol. 55, 1910, p. 1185; also, Arch. f. klin. Med., vol. 10, 1912, p. 445. 502 DIAGNOSTIC METHODS. nicious anemia. In splenomyelogenous leukemia we find a large increase of these bodies, while in the lymphatic leukemia these cells are diminished. They are increased in chronic diseases associated with cachexia and malnutrition, being more marked in cancer and nephritis than in anemia, due to other causes. The more acute, more severe, more threatening the disease, the fewer the number of platelets,1 so that we have a direct relationship between severity of disease and number of these bodies. The method of counting these cells has been given in a previous section to which the reader is referred. Just exactly what significance is to be attached to their increase or decrease is uncertain, but it must be remembered that there is a certain relationship between their number and the severity of general conditions. (t7). Staining Properties. These bodies stain very much like nuclear material with a basic stain, but also take the acid dyes under certain conditions. With the Wright stain these cells are seen as distinctly blue bodies grouped in numbers from 1 to 10 and seem to be composed of nucleus and protoplasm. The apparent nucleus consists of rows of blue or reddish dots occasionally arranged in spherical masses, while the indefinite poorly-defined protoplasm-like substance seems swollen to almost the size of a red corpuscle. As a rule, however, these bodies appear as grouped bluish masses of indefinite structure and outline. (A. Function. The function of these cells is very indefinite. It is possible that they have much to do with the formation of fibrin and may be the source of the so-called fibrin ferment, thrombogen.2 If we are to regard these bodies as derived from the leucocytes this function of fibrin formation is acceptable, but if they are to be considered as derivatives of red cells or as true in- dependent bodies such an hypothesis is untenable. No facts of any clinical value have so far been forthcoming from the study of the blood plates, and it may be possible that they are really artefacts as Lbwit has claimed. (8). Hemoconien. By this term we have reference to the presence in normal blood of very fine granules which are actively motile, not truly ameboid, but with motion more of the Brownian type. These granules, also called blood-dust by Muller, are small, round, colorless granules which vary in size from very fine dust-like particles to some as large as 1 micron in diameter. Their exact chemical nature is uncertain as they do not show, according to Muller, either the reactions Of fat or of albumin. The general idea prevails that these bodies are the ex- truded granules of the leucocytes, as they resemble in size and in staining qualities those of various leucocytic types. The number of these granules is uncertain, 1 See Duke, Arch, of Int. Med., vol. n, 1913, p. 100; also, Port and Akiyama, Deutsch. Arch. f. klin. Med., Bd. 106, 1912, S. 362. 2 Bayne-Jones (Amer. Jour. Physiol., vol. 30, 1912, p. 74) has shown that the platelets aid in the clotting of blood in two ways: (1) by setting free prothrombin, which is later activ- ated to thrombin, and (2) by liberating a thromboplastic substance which neutralizes the anti-thrombin normally present in the blood. THE BLOOD. 503 but apparently the relation of these granules to the red cells, as shown by the ultra condenser, is about 50 to 1. Although no clinical significance is known to attach to these granules, the writer has observed marked variations in their number in various pathological conditions, but cannot at present draw any conclusions from such observations. (9). Morphology of the Blood-forming Organs. As the bone-marrow is of such importance in the production of both red and white cells, it would seem advisable briefly to discuss the histology of this tissue. In it we find practically every cell which occurs in the blood both in normal and abnormal conditions, and also many transitional forms between various groups of cells. The following brief outline of the histology of the bone-marrow is taken from a recent paper of Dickson.1 Varieties of Bone-marrow. (1). Primitive or Embryonic Marrow. A delicate interlacing network of mucoid cells, which later in the process of development go to form the connective-tissue framework or adenoid reticu- lum of the tissue. (2). Red "Lymphoid" or Formative Bone-marrow Proper. This is found in the adult in the short and flat bones, sternum, ribs and vertebrae, and to a varying extent in the ends of the long bones. This is the most important variety as in it are found the red and the majority of the white cells of the blood, and, according as one or the other of these series of cells predominates, the type of marrow may be classified as erythroblastic or leuco- blastic, a varying admixture of these two types being practically always found in any given case. (3). Fatty or Yellow Marrow. This is found mostly, as age advances, in the central part of the long bones and is formed by a process of physiological transformation or degradation of the connective-tissue elements, together with the gradual disappearance of most of the blood-forming cells of the red marrow. (4). Fibroid Marrow. This is found in old persons, especially if debilitated by long-standing disease, and is characterized by the proliferation of the connective-tissue ele- ments and by the progressive sclerosis of the marrow, followed by the gradual disappearance of the hemopoietic cells of the tissue. (5). Gelatinous Marrow. This is essentially a retrogressive change in the tissue and is in no way identical with that type already described as primitive or embryonic marrow. This change has been described by a previous writer as a chronic condition only, but has been frequently found by Dickson as an acute change in many of the acute infectious fevers and allied diseases. 1 Jour, of Path., and Bact., 1907. 504 DIAGNOSTIC METHODS Cytology of the Bone-marrow. In this discussion little more will be taken up than a brief enumeration of the various varieties of cells, as the more important ones of these have been treated of in other sections. (Z). Blood-forming Cells. (a). Nongranular Cells with Basophile Protoplasm. (1). Large. (2). Small. (a). Cells similar to but smaller than the large variety. (^). Cells identical in appearance and in staining reactions with the small lymphocyte of the blood. Of these there are also probably two varieties. (Z>). Granular Cells. (1). Neutrophile. (a). Myelocytes with large rounded or oval nuclei. These have been definitely traced by Dickson to nongranular or hyaline cells in which the granules may be seen gradually develop- ing. There are two types known as the larger and the smaller neutrophile myelocyte. (^). Intermediate cells with indented or horse-shoe-shaped nuclei. These are developed from the myelocytes (a) and in turn go to form (?). Polymorphonuclear cells or adult leucocytes which pass out into the blood stream. (2). Eosinophile. The same types of these cells with eosinophile granulations are found in the marrow, as have been previously tabulated under the heading of neutrophile cells. (3). Basophile. (a). Mast cells. The three types above discussed are also observed in the basophile mast cells of the bone-marrow. (/?). Cells resembling the eosinophile myelocytes but with granula- tions staining with the basic dye. These cells are probably altered eosinonhiles. (A}. Leucocvte Series. (B). Hemoglobin-holding Series. (i). Normoblasts Normocytes or ordinary red corpuscles. , . ,, ... ( Normoblasts, normocytes. (2). Megaloblasts \ , , . . . , . . . ( Megalocytes (entirely pathological in the adult). (II). Giant Cells. (i). Mononucleated or megakaryocytes. (2). Multinucleated or polykaryocytes. THE BLOOD. 505 (ZZZ). Cells of Connective-tissue Type. (1). Fat cells. (2). Cells of the reticulum. (3). Various forms of phagocytic cells. (4). Ordinary connective tissue cells. {IVY Endothelial Cells. (1). Found in their normal position in the vessel wall. (2). Found proliferating and taking on phagocytic functions. Reactions of the Bone-marrow in Disease. Many of these reactions are connected with the production of the so- called inflammatory leucocytosis and take place with great rapidity. Other varieties of change are intimately concerned with the production of the hemoglo- bin-holding series of cells. These reactions may, according to the type of cell involved, be summarized as follows: (Z). Leucoblastic. (1). Neutrophile (2). Eosinophile. (3). Basophile (4). Hyaline or non-granular. (ZZ). Erythroblastic. (i). Normoblastic. Megaloblastic. The above brief outline will show the reader that we have in the marrow all possible types both of red and of white cells and that any variation in the normal activity of the marrow will result in the overloading of the blood with cells of a particular type depending upon the kind and extent of the affection. The histology of the other blood-forming organs, as the liver, spleen, and hemolymph nodes may be found in any text-book on histology. IV. Pathology of the Blood. (I). Special Pathology. Under the head of the special pathology of the blood we have to consider the conditions which are manifested directly by changes in the composition and cellular structure of this tissue. In very few of the blood diseases proper is the blood picture so characteristic that a definite diagnosis is always possible, but in a few of them certain changes are more frequently found and more often lead to a presumptive diagnosis. The pathological conditions in the blood may be considered either primary or secondary, but it should be remembered that severe secondary changes may so closely simulate those found in primary conditions that differentiation is almost impossible. It is probably true that all pathological changes of the blood are really secondary, but in a certain number of these states the etiologic factors are so obscure that we can do no more than interpret the blood findings as primary conditions. 506 DIAGNOSTIC METHODS. (A). Anemia. This is a condition characterized by a deterioration, both qualitatively and quantitatively, in one or in all of the blood constituents. It is usually char- acterized by a diminution in the percentage of hemoglobin (oligochromemia) and by a decrease in the number of red cells (oligocythemia), but we should regard as essential factors in the anemic condition a reduction in the total volume of blood (oligemia) as well as a reduction in the amount of protein (hypalbuminosis). As much more attention has been paid to the first two factors than to the latter ones, anemia has come to mean a reduction in the amount of hemoglobin with a more or less extensive reduction in the number of red cells. Associated with these conditions we have, in the severer types of anemia, variations in form, size, and structure of the red cells as well as definite changes in the relationship of the different white cells. In the study of the anemic conditions we must differentiate the primary from the secondary form, by this we mean a differentiation between those forms which have no demonstrable cause from those types whose etiology is more or less secondary to other pathologic conditions. Primary Anemias. This form which has no demonstrable cause is difficultly separable from the secondary form as well as from certain other primary forms, such as primary pernicious anemia. It will probably be shown to be a true secondary anemia and must be sharply differentiated from the pernicious type from the point of prognosis. These cases are only recognizable when they are typical in form and are frequently not amenable to any form of treatment which we may nstitute. The rule, however, is that these forms of anemia yield more or less promptly to proper dietetic and therapeutic treatment. It is probable that the question of prognosis in these cases must depend largely upon the amount of degeneration of the red cells which occurs.1 In these cases we find that the diminution in the number of red cells is usually parallel to the reduction in the amount of hemoglobin, so that a high color index will usually obtain. More or less degeneration, as evidenced by the appearance of poikilocytes, normoblasts, Maragliano's polychromatophilia, etc., will be observed depend- ing upon the severity of the case. The leucocytes are usually normal in number and in differential relations, while the blood plates are usually increased. The changes in the plasma in this primary anemia are not characteristic, although we do observe a diminution in the specific gravity which runs parallel to the oligochromemia. (1). Simple Primary Anemia. Chlorosis is a primary anemia occurring almost exclusively in girls about the age of puberty and characterized by a marked reduction in the amount (2). Chlorosis. 1 See Pollitzer, Zeitsch. f. klin. Med., Bd. 75, 1912, S. 367, for a discussion of the types of regeneration and degeneration in anemia; also, Wichern and Piotrowski, Deutsch. Arch. f. klin. Med., Bd. 106, 1912, S. 533. PLATE XXIII. Chlorotic Anemia. (Wright's Stain.) THE BLOOD. 507 of hemoglobin and a slight change in the number of red cells. Clinically, this state is evidenced by the appearance of a wax-like changing into a greenish tone of the skin and a sky-blue coloration of the cornea. Some cases may show a variety of colors of the skin. Certain changes are observed in the digestive and generative organs and certain general abnormalities are seen which are due to lessened production of blood-cells and to diminished oxidative and fermentative powers of the system. This form of primary anemia differs from all other forms in the absence of blood degeneration, as very rarely marked degenerative signs appear in the blood picture. The blood finding is not absolutely characteristic for this clinical entity, as it is simulated by many anemias of the secondary type. Clinically, this disease is so sharp that a diag- nosis is often possible without a blood examination. The chief characteristics of the blood in this condition are: (i) Reduction in the hemoglobin. This may run as low as 20 per cent., giving a color index of 0.5. Secondary anemias rarely reach such a low level. (2) Variations in the number and size of the red cells. The number of red cells is not reduced to a very great extent, the average being about 4,000,000, although counts as low as 1,000,000 have been reported. When these low counts do occur some complication should be suspected. Ordinarily the size of the cell is diminished, although we frequently find large "dropsical" cells which are due to absorption of fluid from the hypotonic plasma. These latter cells are usually few in num- ber, the great majority of cells being smaller than the normal size. Poikilocytes and degenerated reds rarely occur except in the severer forms of this disease, while chromatophilia is usually regarded as a sign of active regeneration of the blood. When nucleated reds occur, which is a rare finding, they are practically always of the normoblastic type and rarely appear of the megaloblastic form. The leucocytes in this condition are usually normal both in size and number and degenerative forms are rarely seen. It has been stated that the eosinophile cells are much increased in this condition, but the writer has found that their ratio is very rarely above the upper limit of the normal figure for these cells. The platelets are usually about normal and are usually large in size. In chlorosis we find certain variations in the physical and chemical prop- erties of the blood. The specific gravity is usually reduced in proportion to the reduction of the hemoglobin and may fall as low as 1028. The plasma is usually much diluted so that a condition of oligemia may be considered more or less characteristic of chlorosis. Whether or not a hydremia, as indirectly manifested by a diminution in the amount of albumin, obtains, is debatable. Chlorosis belongs to the class of primary anemias and as such has no definite etiology. Various conditions such as hypoplasia of the arterial system, intestinal autointoxication, disturbances of the nervous system, such as a vasomotor neurosis, have been advanced to explain this condition, but none of them are tenable in all cases. A great importance must attach to hygienic conditions, poor food, and mental depression, especially about the age of puberty, as this disease is usually apparent under these conditions at this time 508 DIAGNOSTIC METHODS. of life. A further factor which must necessarily bear upon the etiology of this condition is the defective power of absorption of iron compounds. Patients afflicted with chlorosis improve rapidly under the administration of iron, but not unless the digestive and absorptive powers improve at the same time. It does not concern us here as to the dynamics of the absorption and the effect of iron, but it should be accepted as an axiom that no therapeutic effects may be expected from iron unless the absorptive power is made better. Just exactly what the pathological conditions are which are accountable for the functional insufficiency of the bone-marrow is uncertain, but it must be re- called that gross as well as microscopic pathologic changes are not neces- sary to produce functional disturbance in any organ. It should be remembered that we may have various types of chlorosis which show different prognostic characteristics: (i) Those in which the red cells are very slightly reduced (about 4,000,000), a marked diminution in the amount of hemoglobin, a low color index, and no change in the size and shape of the cell. Such cases usually recover promptly without showing any relapse. (2) Cases in which the red cells are below 4,000,000, which show a marked diminution in the amount of hemoglobin with a very low color index and which give very slight evidences of degenerative changes in the red cells. These cases are usually characterized by marked prostration, but usually recover more or less promptly although relapses frequently occur. (3) Cases in which the red cells are reduced as low as 2,500,000, a reduction in the hemoglobin with a very low color index, marked changes in the shape and size of the red cells. These cases respond slowly to treatment and have a bad prognosis (Ewing). As these cases of chlorosis convalesce, we observe an increase in the number of red cells to a point somewhat above the normal and a later increase in the hemoglobin content of each individual cell. These changes are usually evident in from eight to ten days after institution of treatment, but as a rule a much longer time is necessary for any marked change to be observed in the number of the cells or in their hemoglobin content. The changes in the plasma are usually the first to appear and should be considered essential for the proper regeneration of the blood in chlorotic conditions. This disease belongs in the group of curable conditions and usually has a good prognosis, but the susceptibility to intercurrent acute diseases during this period is much increased so that we should always bear in mind the proba- bility of complications arising in the convalescent period. (3). Progressive Pernicious Anemia (Biermer's Anemia). This term is applied to a form of severe anemia which, in spite of all treatment, progresses steadily toward death. It results from defective hema- togenesis and excessive hematolysis and is characterized by definite changes, both numerically and morphologically, in the red cells and by characteristic changes in the bone-marrow (Ewing). This condition has frequently been described as the result of infection with certain intestinal parasites so that it PLATE XXIV. Blood in Pernicious Anemia. (Wright's Stain.) THE BLOOD. 509 cannot in all cases be considered a truly primary disease, although the larger majority of cases show no known etiology. According to Herter, certain anerobic bacteria, found in the large intestine, produce a substance of marked hemolytic power, which penetrates the intestinal wall and enters the portal circulation. In some cases of idiopathic purpura hemorrhagica the blood picture is that of pernicious anemia, but may be distinguished by the absence of megaloblastic change and by the prominence of hemorrhage. The blood picture in this disease is not absolutely characteristic, as certain forms of secondary anemia may show similar findings. The chief characteristics of the blood in primary pernicious anemia, as also in those severe types of secondary anemia which simulate this form, are (i) signs of rapid blood de- struction, such as degenerated reds, endoglobular degenerations, polychroma- tophilia, increased iron compounds in the serum and corpuscles, an increase of iron in the liver and spleen; (2) poikilocytosis; (3) a high color index resulting from a marked diminution in the number of red cells and a correspondingly less degree of diminution in the amount of hemoglobin; (4) megaloblastic blood formation. This latter indicates a direct reversion to the embryonal type of blood formation, in which the presence of megaloblasts as direct pre- cursors of megalocytes is observed. The red blood-cells in this condition are few in number being reduced to as low as 1,000,000 cells, counts of 500,000 having been observed without the patient suffering any marked in- convenience. Naegeli1 reports a case showing 138,000 reds. This fact should be taken as evidence that the oligocythemia in pernicious anemia is not alone accountable for the symptomatology. This count may remain stationary, may show slight decrease, but usually progresses slowly until death ensues. The average diameter of the red cells is somewhat increased in pernicious anemia. While many of them may be normal in size and many very small, the cells average from 4 to 13 microns in diameter. A pernicious anemia is a distinct large cell anemia. A macrocytosis is much more character- istic of this disease than of any other, 70 per cent of the cells in these cases being of this type. Microcytosis is rare but may occur to such an extent that the average size of the red cell may be about normal. Poikilocytes are very com- mon and often show extreme shapes and are frequently numerous in number. Polychromatophilic degeneration is very extensive in this form of anemia. While both normoblasts and megaloblasts occur in large numbers in perni- cious anemia, the megaloblasts usually outnumber the normoblasts. This megaloblastic increase may be considered pathognomonic of this anemia, as a preponderance of these large nucleated red cells does not occur, except in rare cases, in the other varieties of anemia. The hemoglobin may be markedly reduced showing values rarely above 50 per cent, and often as low as 10 per cent. The color index is always high; being more frequently above one than below. A low index is found, according to Ewing, in the chronic cases, while the acute forms are more frequently associated with a high index. If im- 1 Leipzig, 1908. 510 DIAGNOSTIC METHODS. provement occurs the index is always lowered, an increasing index denoting a bad prognosis. In this pernicious form of anemia we find the leucocytes practically al- ways diminished in number, averaging about 4,000, a condition which almost never obtains in a secondary anemia. Their number usually runs parallel to that of the red cells, a leucocytosis pointing probably to a complication. As the case improves the neutrophile cells increase in number, the low leucocyte count being, as a rule, due to their diminution. The percentage of the non- granular mononuclear cells varies inversely to that of the granular form. As the disease progresses the percentage of the nongranular cells increases, while the granular cells are diminished. This disease shows, therefore, a high lym- phocyte count which is relative and not rehl, being due to the diminution in number of the polymorphonuclear cells. The lymphocytes may constitute as high as 50 per cent, of the leucocytes, the eosinophiles may reach as high as 10 per cent., the myelocytes 2 per cent., while the mast cells may run as high as 3 per cent. Degenerations of all kinds are observed in the leucocytes, but nothing characteristic of pernicious anemia is found in the white cells. The blood-platelets are largely decreased and may be as low' as one- twenty-fifth of the normal number. Von Limbeck and Sahli claim that these cells are increased in number, but the usual finding is one of diminution. (4). Splenic Anemia. This is a form of chronic anemia characterized by idiopathic enlargement of the spleen without any involvement of the lymph nodes. A large number of conditions may be responsible for the anemia of the splenic type, so that a direct etiologic factor should be looked for in all cases. Among the conditions which may give rise to this type of anemia, we find gummata of the spleen, large round-cell sarcoma of the spleen, chronic splenitis of the malarial type, and splenomegaly associated with cirrhosis of the liver. This latter condition is known as Banti's disease and its etiology is uncertain.1 While it is true that the splenic lesions do not differ essentially from certain stages of the lesions in ordinary pseudoleukemia, yet we do not find in this condition any involvement of the lymph-glands, although the spleen may be enormously enlarged. The blood picture in this condition is character- ized by a relatively high red cell count, a marked reduction in the percentage of hemoglobin, and a consequent low color index. The white cells are rarely increased, a leucopenia being rather the rule. When we do find an increase in the number of white cells a relative lymphocytosis occurs usually associated with an increase in the number of basophiles. Poikilocytes and nucleated red cells are very uncommon while degeneration may occasionally be observed. Whether or not we have an increase or a decrease in the number of leucocytes, a relative lymphocytosis associated with enlarged spleen and no involvement of the lymph nodes must be considered characteristic of splenic anemia. 1 See Banti, Berlin, klin. Wochs., Bd. 48, 1911, S. 2328; Stein, Amer. Jour, of Med. Sci., vol. 144, 1912, p. 856; also Umber, Munch, med. Wochs., Bd. 59, 1912, S. 1478. PLATE XXV Blood in Leukanemia. (Wright's Stain.) THE BLOOD. 511 In Banti's disease we find an enormous increase in the size of the spleen and associated with this an extensive cirrhosis of the liver. Along with these factors we find a high-grade toxogenic protein decomposition, which is associated with very high values for the total nitrogen of the urine and of the output of purin bases. The number of erythrocytes diminishes corresponding to the degree of anemia, while the hemoglobin percentage is more markedly reduced. The leucocytes are either normal or more frequently diminished in number, a relative lymphocytosis existing as in the pure type of splenic anemia. (5). Anemia Infantum Pseudoleukemica. Von Jaksch 1 has described a rare form of anemia seen in children which is characterized by enlargement of the spleen, liver, and lymph nodes. The most striking points in this condition are the great diminution in the number of red cells, one case showing only 820,000; numerous nucleated red cells; diminution of hemoglobin; the leucocytes always increased in number, being from 20,000 to 50,000, as a rule, and displaying a remarkable variety of form and frequently attaining to unusual size. The morphological changes in the blood resemble both those seen in leukemia and in pernicious anemia, the disease passing either into one or the other of the previously mentioned conditions. The blood findings are not alone sufficient to warrant a diagnosis of infantile pseudoleukemia, but are significant when taken in conjunction with the clinical findings. This condition has been classed by some with the leukemias, as the blood findings are occasionally more prominent among the leucocytes, while by others it is classed with pernicious anemia owing to the frequent changes in the red cells. Von Leube considers this anemia a mixed form of pernicious anemia and of leukemia. Luce2 regards this as a symptom of many conditions rather than an independent blood disease. In the majority of these cases the changes are more evident in the red cells, marked diminution in number (as low as 200,000 per cmm.) associated with extensive destruction of red cells with all irregular and unusual types of these cells being observed. The diminution in the amount of hemoglobin is great, but the color index is usually high just as we find it in pernicious anemia. The white cells show an extensive disturbance in the neutrophile and eosinophile blood picture with a large increase in the number of large lymphocytes. The number of white cells is usually increased, but not to as great an extent as in leukemia. The changes in the red cells and in the hemoglobin usually precede those in the white cells, so that the early stages may show us a typical picture of pernicious anemia, while later examinations may lead to the diagnosis of the mixed condition. (6). Leukanemia. 1 Wien. klin. Woch., Bd. 2., 1889, S. 435 and 456. 2Deut. Archiv. f. klin. Med.. Bd. 77, 1900, S. 215. 512 DIAGNOSTIC METHODS. (7). Aplastic Anemia. Ehrlich1 has reported a rapidly progressing anemia accompanied by hemorrhages into the mucous membranes, associated with hyperplasia of the bone-marrow, and not showing the ordinary changes in the blood which are supposed to accompany pernicious anemia. This type has been called aplastic anemia and has not been frequently reported. The red cells are usually markedly reduced in number, being as low as 790,000 in a case reported by Wood. The hemoglobin may fall as low as 11 per cent, as reported by Muir, while the leucocytes are usually normal in number but showing much reduced percentages of the polynuclear neutrophiles. In this condition we find an enormous increase in the number of lymphocytes, the percentage being in Lipowski's case 93, the remaining 7 per cent, being neutrophiles. No nucleated reds have been found in the blood and only a very few in the marrow itself, which is fatty, almost white and contains few neutrophiles and no eosinophiles.2 Secondary Anemia. By secondary anemia we mean one in which etiological factors seem sufficient to explain the variations observed in the blood. The principal variation seems to be more directly observed in the reduction of the hemoglobin, although the number of red cells is coincidently reduced, but not to the same degree as is the hemoglobin. In the mild cases the color of the blood is but slightly paler than normal, but in secondary anemia of a severe type the color may resemble the watery drop observed in cases of typical pernicious anemia. Cabot suggests a classification of the secondary anemias as follows: (x) Mild cases, showing a normal count of red cells, but having the hemoglobin diminished. (2) Moderate cases in which the count is normal, but the cells show signs of moderate degeneration, abnormal staining qualities, and a diminished tendency to rouleaux formation. (3) Severe cases, in which the count is not much reduced but in which the hemoglobin is very much lessened and the cells show marked qualitative and quantitative changes. (4) Very severe cases with a slightly lessened blood count, a marked diminution of the hemoglobin, and evidences of degeneration and destruction of the cells as well as evidence of regeneration. In some of the severer types of secondary anemia the blood picture may so closely resemble that of pernicious anemia that a diagnosis is possible only through the careful investigation of the etiology of the condition. The general points observed in secondary anemia are as follows: A variable decrease in the amount of hemoglobin, a reduction in the number of red cells less in degree than the diminution of hemoglobin and a subnormal color index, which rarely reaches the low grade shown in chlorosis. The Blood Picture. 1 Charit^-Annalen, Bd. 13, 1888, S. 300. 2 See Hirschfeld, Folia Ha?mat., Bd. 12, 1911, S. 347. THE BLOOD. 513 lowest color index is seen in those secondary anemias following cancer, severe hemorrhage, and gangrenous processes. The specific gravity of the blood is reduced corresponding to the degree of reduction of hemoglobin. The rapidity of coagulation is increased depending upon the grade of oligochromemia and of oligocythemia. The reduction in the number of red cells may be very marked as, for instance, in a case of von Limbeck the number was 306,000. The red cells show a lack of hemoglobin, frequently appearing as the pessary forms. Polychromatophilia is quite common, but bears no direct relation to the hemoglobin content of the cell. Only in the severer cases do we find poikilocytes, although anisocytes, especially microcytes, are frequently present, the larger "dropsical" cells being less common than in chlorosis. Nucleated erythrocytes are frequently seen in some cases, while in others even of severer grade of anemia they are absent. These nucleated reds are, as a rule, of the normoblastic variety, megaloblasts being exceedingly rare. The leucocytes vary in number depending on the cause of the anemia, from a leucopenia, which is rare, to a leukemic condition. The increase in the number of leucocytes is more frequently in the polynuclear neutrophiles, the lymphocytes being rarely if ever increased, while the eosinophiles, although not increased, are usually at the upper limit of the normal value for these cells. In some cases of the severe chronic types of secondary anemia we may find a lymphocytosis, but this is rare. The blood-platelets are usually increased in number, in some cases being two and one-half times the normal values. (1). Acute Hemorrhage. The character of the anemia following an acute hemorrhage will depend upon the type of the hemorrhage, that is whether the loss of blood occurred suddenly and at one period, or whether slowly and at intervals.1 The loss of one-half of the total volume of blood at one time is usually fatal, as Panum has shown. If the loss be less following one large hemorrhage, regeneration takes place in from five to thirty days, depending upon the amount of blood lost. Regeneration is quickest in men between the ages of 20 and 40, slower in women, and slowest in children. Immediately following a hemorrhage the blood picture will be normal qualitatively as there has been no time for the morphologic changes to take place. Shortly following the hemorrhage lymph pours into the blood to re- store the volume and maintain the pressure, so that the blood count and the hemoglobin diminish usually to the same degree. As the formation of new cells goes on the color index decreases, as the new cells are, as a rule, deficient in hemoglobin. These cells are more or less easily degenerated, showing vari- ation both in their shape and staining qualities. The number of cells reaches the normal much sooner than does the percentage of hemoglobin, so that we may find for weeks evidences of marked anemia. The most frequent causes of acute hemorrhagic anemia are traumatism, 1 See Milne, Jour, of Exp. Med., vol. 16, 1912, p. 325; Deutsch. Arch. f. klin. Med., Bd. 109, 1913, S. 401. 514 DIAGNOSTIC METHODS. ectopic pregnancy, abortion, gastric and duodenal ulcers, uterine tumors, pulmonary tuberculosis, and hemorrhagic pancreatitis. (2). Chronic Hemorrhage. By a chronic hemorrhage we mean one in which repeated hemorrhages follow one another so closely that the blood has no time to regenerate before a second loss of blood occurs. This rules out of consideration those cases in which repeated hemorrhages occur but at sufficiently long intervals to permit of regeneration of the blood. This latter condition, although a chronic hemor- rhage, gives the picture described under Acute Hemorrhage. In the chronic hemorrhage we find the red cells markedly reduced, the hemoglobin very much diminished and usually a marked leucocytosis. The red cells are usually small and pale, show a low color index, and usually few nucleated forms, although the picture may rarely assume the pernicious type. Regeneration in this form of anemia is slow, as the blood-forming organs seem to lose their power of regenerating the blood after repeated hemorrhages. The most frequent causes of chronic hemorrhage are scurvy, epistaxis, hemorrhoids, intestinal ulcers, gastric and other carcinomata, and intestinal parasites. (3). Inanition. In the discussion of the anemia of inanition, it must be remembered that other factors than starvation are necessary in its causation. That starvation alone will not cause anemia is shown from the examination of the blood of Cetti who fasted ten days. His blood showed about 6,000,000 red cells, a small diminution in the percentage of hemoglobin, and a leucocyte count of 4,200. Although the changes in the cells and the pigment of the blood are not marked following starvation, it is true that we have a loss of albumin of the plasma and a diminution in the total volume of blood. This is taken as evidence by Grawitz that a true anemia occurs. Such changes are not striking if the days of fasting are alternated with days of slight nourishment. It is not so much the quantity of food as the quality which is of importance in bringing about the anemic conditions. As is well known, the foods containing iron are the principal sources of the hemoglobin of the blood, and these are frequently, owing to disturbed gastric and intestinal functions, poorly digested and assimi- lated. Although the amount of iron contained in the ordinary food is sufficient under the best conditions to maintain the hemoglobin content of the blood, yet the methods of food preparation, as well as the abnormal methods of rapid eating, are important factors in the poor assimilation of the iron of the food. The lack of sunlight as well as impure air are contributory factors in causing anemia through their influence upon the general body functions. No tissue, whether animal or plant, can flourish in air w'hich does not have sufficient oxygen to support the combustion processes of the system. This statement needs no retraction in the case of anerobic bacteria, as it has been definitely shown that these organisms obtain the oxygen necessary for their develop- THE BLOOD. 515 ment from the culture media upon which they grow, although they are inca- pable of developing in an atmosphere of pure oxygen. Overwork, especially when associated with worry, has great influence upon all of the functions of the system. For this reason overwork has been credited with the power of producing anemia as well as many other serious systemic disturbances. How- ever, it is rare to find an authentic case of anemia which can be traced directly to overwork without the mental influence of worry nd the coincident nervous strain from this latter cause. (4). Intestinal Parasites. The anemia caused by intestinal parasites may be of such a severe grade as to resemble very closely the type of pernicious anemia. In many cases it is an impossibility to make a differential diagnosis between these types with- out the finding of an intestinal parasite which will clear up the diagnosis. In these cases the blood picture returns more or less quickly to normal after removal of the parasite in question, while in the pernicious anemia of un- known origin the progress of the disease is always toward a fatal termination. It is probable that the cause of the severe secondary anemia due to the presence of the intestinal parasite is a result of the toxic condition set up by the absorp- tion of the hemolytic toxins elaborated by the parasite. A very severe anemia of the secondary type is frequently seen as a result of decomposition of the in- testinal contents and in cases of chronic constipation in which the direct toxic agent is at present unknown. The most common intestinal parasites causing the severe types of secondary anemia are (i) uncinaria duodenale, (2) strongy- loides intestinalis, and (3) bothriocephalus latus. The first of these causes an anemia which is very closely related to that shown by miners and tunnel diggers, and seems to be much more prevalent in the southern part of the United States, although it occurs in many different countries. The blood count may fall below one million red cells and the hemoglobin may be as low as 15 per cent., while all varieties of degenerative changes may be seen in the erythrocytes. In the anemia following infection with the bothriocephalus we find very marked similarity with the primary pernicious type. One-half to two-thirds of the nucleated reds in this variety may be of the megaloblastic type and yet may disappear within two to three weeks after the worm has been expelled. It is uncommon to find in secondary anemias caused by the parasites above men- tioned any marked eosinophilia, which is so common in cases of infection with many of the other forms of intestinal parasites. The anemia of the severer type seems to prevent a chemotaxis toward eosinophile cells. It is still very much of a question whether the blood in febrile cases shows the characteristics of a secondary anemia, as the result of the temperature increase. It has been shown that increased temperature, in itself, does not always produce anemia, although we do have marked destruction of the red (5). Fever. 516 DIAGNOSTIC METHODS. cells and a coincident loss in the amount of hemoglobin of the remaining cells. So great is the influence of the toxin of the febrile condition that it is highly probable that the anemia so frequently observed in febrile cases is due to a combination of causes, rather than to a specific effect of the increased tem- perature. The changes in the white cells in febrile cases are not always of the same character, the variations being dependent upon the specific causative factor of the fever. As the blood changes in the acute infectious fevers are of more or less importance, they will be discussed later under separate headings. The anemias which are secondary to both the acute and chronic infections are probably directly due to the influences of the toxins upon the blood and blood-forming organs. The condition of general nutrition as well as the state of digestion, especially in the chronic states, such as tuberculosis, leprosy, and syphilis, must be regarded as important factors in the causation of these second- ary types. (6). Blood Poisons. There are a very large number of compounds which produce, when taken in toxic doses, very marked changes in the qualitative and quantitative composition of the blood. As is well known, iron compounds in therapeutic doses increase the amount of hemoglobin in the red cells and also increase the number of red cells up to a certain point. Many of the effects which are attributed to iron compounds may be due to the improved hygienic and dietetic conditions which usually prevail during the administration of these substances. Yet the therapeutic results following the administration of iron are such as to make it certain that a specific influence of this drug is present in anemic conditions, especially of the chlorotic type. Many compounds produce a very marked secondary anemia, the most important of these being alcohol, opium, lead compounds, cocain, and acetani- lid. While others, such as arsenic, nitrobenzol, nitroglycerin, phenacetin, and poisonous mushrooms, cause dissolution of the red cells with marked hemoglobinemia. The anemia following the use of lead, either in toxic doses or after its slow absorption from constant contact with it in the arts, is of great practical importance. The causes of this lead anemia are rather complex. The lead compounds have a direct action on the red cells and on the blood- forming organs as well as upon the gastrointestinal tract and the eliminative organs. While the anemia shows no especial characteristics as regards the number of red cells and the amount of hemoglobin, yet the peculiar granular degeneration and the polychromatophilia are sufficient to differentiate this type from most of the other secondary anemias. The basophilic degenerations of the reds is more marked in lead anemia than in almost any other condition and usually runs parallel to the severity of the clinical symptoms of the case. In arsenical poisoning also we occasionally find a slight amount of granular degeneration of the red cells, but the hemoglobinemia in this latter condition will differentiate it from lead anemia. THE BLOOD. 517 (B). Leukemia. Although acute forms occur, leukemia may be regarded as an essentially chronic condition which is characterized on the one hand by definite changes in the lymphatic and myeloid tissues of the body and, on the other hand, by certain peculiar changes in the number and relations of the various cellular constituents of the blood. These latter conditions must be regarded as purely symptomatic of the preceding states and not as the direct pathological condi- tion in themselves. Leukemia has been classed as a primary anemia, although the changes in the blood are here more directly related to variations in the white cells than to characteristic changes in the red corpuscles; yet we do find a diminished red count as well as a diminution in the amount of hemoglobin in this condition. It is a disease marked by the constant presence in the blood of granular mononuclear or polynuclear cells, or an increase of the non- granular cells with round nuclei. While the leucocyte count is almost in- variably increased to a marked extent, we find cases in which the number of cells is normal, but we find great deviations from the normal relations of these white cells. While the tendency is becoming more and more general to regard this condition as a single entity, manifested by various blood pictures, yet we find the cells grouping themselves together in such definite ways that we are justified in dividing leukemia into three general types, with transitions from one to the other form. These types are (i) splenomyelogenous leukemia or "myelemia," (2) lymphatic leukemia or "lymphemia," (3) mixed leukemia. Each of these types shows a distinct blood picture which permits of the classi- fication of the condition studied. This condition was formerly subdivided into two distinct types-the true splenic and the myeloid leukemia. However, practically none of the cases reported could be definitely classed under either one of these headings, as the blood picture was always referable to disturbance in both the spleen and marrow. This type of leukemia is characterized by a marked increase in all of the granular cells, especially of the neutrophile, eosinophile, and basophile types, while the nongranular cells are not so characteristically increased. (1). Splenomyelogenous Leukemia The gross appearance of the blood is normal even though the leucocytes are increased to an enormous extent. In extreme cases it may appear pale and opaque and does not flow from a puncture as readily as normal blood. In making smears of such blood, the preparations appear granular and are not readily spread so that the future examination is rendered somewhat difficult. Gross Appearance. Red Cells. As a rule, these cells are diminished in number, but the oligocythemia is of a mild degree, the average count being about 3,000,000, although it may run 518 DIAGNOSTIC METHODS. as low as 1,500,000. As a general rule, the red cells diminish in proportion to the increase in the number of white cells. Occasionally we find cases in which an oligocythemia persists with a normal or slightly increased leucocyte count. Such a condition might lead to the diagnosis of pernicious anemia, unless the differential count was carefully studied. The red cells are usually pale and of the chlorotic variety. Very little degeneration of the red cells is observed, microcytes and macrocytes are rare, but a few poikilocytes are seen in practically all cases. Polychromatophilia is more or less common and cells showing basophilic granulations appear with more or less frequency. Normo- blasts are very common in this condition, yet their absence does not rule out the diagnosis of leukemia. Megaloblasts and gigantoblasts are frequently seen and are sometimes many in number, although they rarely if ever exceed the normoblasts in number. Hemoglobin. The hemoglobin is reduced to a somewhat greater extent than is the number of red cells, the color index being about 0.6 and the average percentage of hemoglobin about 40. Leucocytes. In this condition we find the leucocytes increased, as a rule, to a very marked degree, counts running as high as 750,000 having been seen by the writer. Osler gives as his average for the white count 298,700, while the average may vary to a slight extent at different periods of the day. Some of the cases show a uniformly high count, others a moderate count and a few others a low count of about 100,000 cells. It is this increase in the number of white cells which gives the blood its peculiar opacity in this condition and may make a diagnosis possible by mere inspection. In some cases of leukemia we find a normal count of white cells while in others the count may be similar to that of a simple leucocytosis. It is the differential count in combination with the large increase in cells which should be considered characteristic, rather than a simple increase in itself. Differential Count. These cells are large mononuclear cells with neutrophile granules. They are present in large numbers, averaging about 35 per cent, of all the leucocytes present. A diagnosis of leukemia is almost always possible when we have such an extreme neutrophile myelocytosis along with an extreme leucocytosis. These myelocytes appear in two forms: (1) the large myelocytes of Cornil, which may be as large as 30 microns in diameter and have a large, pale, eccen- tric nucleus which is poor in chromatin. These cells are seen only in spleno- myelogenous leukemia and in some of the secondary leukemias of children and must be regarded as practically pathognomonic of this condition. (2) Small myelocytes about the size of the normal polynuclear leucocytes with Neutrophile Myelocytes. PLATE XXVI Blood in Spleno-myelogenous Leukemia. (Tri-acid Stain.) THE BLOOD. 519 a centric, round nucleus staining deeply with the various aniline dyes. We find all gradations between these large and small myelocytes, sometimes observing a few which are about the size of the red cell. The granulations of these cells are sometimes numerous, but may be entirely lacking so that they may be indistinguishable from the large lymphocytes unless the pale quality of their nuclei is remembered. The degenerative changes in these myelocytes are few in number and are limited, as a rule, to the hydropic form usually seen in chlorosis. Polynuclear Neutrophile Leucocytes. These cells are relatively diminished, their average, according to Cabot, being about 46 per cent., although an absolute increase is present, amounting to as much as 60,000 to 75,000 cells. Marked variations in the size of these cells are common, some of them being very large, some very small and no defi- nite relation existing between the numbers of the large and small cells. These variations in size of the polymorphonuclear cells are rarely if ever seen in ordinary leucocytosis. It is very common to find cells with irregularly-shaped nuclei and with more than one form of granule, which may vary in tint depend- ing on the method of fixation. Marked degenerative changes in these cells are very common. Thus we find their stickiness is increased, their nuclei usually pale and frequently showing karyokinetic figures. All grades of vari- ations in the granulations may be observed, the granular cytoplasm being oc- casionally replaced by a homogeneous highly refractive material. The eosinophile cells are usually much increased in the splenomyelogenous form of leukemia, but their percentage relations to the other leucocytic forms are practically normal. Their number may run from 3,000 to 100,000, the average absolute number being about 12,000, while their percentage is about five. The total number of these cells per cmm. greatly exceeds that found in any other condition so that we accept, with Ehrlich, such an increase as pathognomonic of splenomyelogenous leukemia. These cells appear in all modifications, some of them being very small, while some of them are very large. The eosinophile form of myelocyte occurs in large numbers, but never is as numerous as is the neutrophile myelocyte. We occasionally observe all forms of transition between the myelocyte and the eosinophile leucocyte, the eosinophile myelocytes occasionally forming the majority of the eosinophile cells. The granulations of these cells may be of uniform size and staining quality or there may be some basophile granulations among the eosinophiles while the granulations themselves may vary greatly in size. Ewing considers eosinophile myelocytes with granules of unequal size and density of stain as pathognomonic of myelocythemia. Eosinophiles. Basophiles. According to Ehrlich, we always find an increase in the number of mast cells in leukemia, their absolute increase being in some cases greater than 520 DIAGNOSTIC METHODS. that of the eosinophiles and is always proportionately higher. This increase is so marked as to constitute a very reliable diagnostic feature of the blood. The number of basophiles may run as high as 140,000 (Taylor), while the percentage may vary from 5 to 47. Lymphocytes. The number and proportions of lymphocytes in the splenomyelogenous leukemia vary in different cases and at different times in the same case. As a rule, their percentage is reduced averaging about 10, while an absolute in- crease is usually present, this increase having no uniform relationship to the stage or character of the disease. These cells vary much in size, the large cells usually outnumbering the small ones. Large mononuclear cells with very faint cytoreticulum and vesicular nucleus occur in large numbers in this form of leukemia and seem to have no special significance, although they may be mistaken for the large lymphocyte or for the myelocyte. Degen- erative changes are observed in both the small and large lymphocytes in leu- kemia; thus the nuclei of the small cells may become incurved and bilobed or even trilobed, while the cell body remains basophilic (Rieder). Points in Diagnosis. An excessive leucocytosis, with a large proportion of neutrophile mye- locytes, the presence of a large number of eosinophile myelocytes and of basophile cells, the presence of atypical cells, both of the mononuclear and polynuclear variety, and large numbers of nucleated red cells are the chief characteristics. Any one of these points may fail for a time, but will usually be evident at some stage of the disease. The large size of the myelocyte is much more characteristic than the mere presence of these cells, so that we should confine our diagnosis to cases which show irregularity in size, staining qualities, and degenerative reactions of these cells rather than to those show- ing merely an increase. The presence of the large number of eosinophiles, especially those showing granules of irregular size and staining qualities, is a very important point to be remembered in the diagnosis of this condition. (2). Lymphatic Leukemia (Lymphemia). In this form of leukemia we observe a marked increase in the number of mononuclear nongranular cells in distinction from the previous form of leukemia in which the increase is rather in the number of the granular types. While a variety of the mononuclear nongranular cells are present, there is usually observed a predominance of one particular form and size, in some cases the small mononuclear cell with a narrow ragged rim of protoplasm, in others the cells of the large lymphocyte type, and in others large cells whose protoplasm is basophilic or in some cases distinctly acidophilic. Red Blood-cells. In this form of leukemia we find a much greater anemia than in the splenomyelogenous form, although we may observe a normal red count for PLATE XXVII. Lymphatic Leukemia. (Tri-acid Stain.) THE BLOOD. 521 some time. The number of cells varies between 1,500,000 and 4,000,000, while the average percentage of hemoglobin is about 37 per cent. Nucleated red cells are rare in this condition, yet in the severer cases we may find them as numerous as in the splenomyelogenous type. All forms of degeneration noted under the previous type of leukemia are occasionally seen in this latter form. Leucocytes. The leucocytes are, as a rule, increased, the average being about 145,000, according to Osler. In this form we may find aleukemic periods which may last for a considerable period of time, the count usually rising just before death, Differential Count. According to Grawitz the cases of lymphatic leukemia may be divided into (i) those in which the increase of leucocytes is especially in the small mononuclear variety, (2) those showing an increase in the medium-sized cells with basophilic homogeneous protoplasm, and (3) those in which the cells which predominate are very large and usually degenerated. All these forms may occur together and may vary in the same case at different times. These mononuclear cells may constitute as high as 99 per cent. (Osler) of the total number of leucocytes. These leucocytes show in a very large number of cases much degeneration either of the protoplasm or of the nucleus, very few of the cells showing mitosis which is so common in the splenomyelogenous form. In this type of leukemia polymorphonuclear cells are rare, eosinophiles usually absent, and myelocytes and basophiles rarely if ever present. This type of leukemia is not easily amenable to diagnosis, especially in differentiating it from some cases of sarcoma in which the blood may show a similar picture. In some cases we find a lymphatic leukemia with a considerable number of myelocytes both of the eosinophile and neutrophile type. This has led to the differentiation of a "mixed leukemia," which does not seem to be advis- able as we may find myelocytes in the pure lymphatic type of this disease. (3). Acute Leukemia. This form of leukemia is characterized by its brief course (from six to eight weeks), by the severity of its symptoms, the frequency of the hemorrhagic diathesis, rapidly developing cachexia, and death. This condition occurs chiefly in young people and is usually of the lymphatic type,1 although a few cases of the myelogenous variety have been reported (Billings and Capps). In all cases the anemia is extreme, the red cells usually running below 1,000,000 and the hemoglobin as low as 10 per cent. There is no type of cell which is characteristic of this form, although the cells are much more uniform in size than in the chronic states of this disease. In some cases nearly all of the cells have a basophile protoplasm, while in others they show acidophilic properties. Nucleated reds are usually rare, although they may be present in fairly large numbers. The drop in the count of red cells is usually sudden and denotes 1 See Strauch, Amer. Jour, of Dis. of Child., vol. 5, 1913, p. 43, for a discussion of this condition in children. 522 DIAGNOSTIC METHODS rapid blood destruction. The leucocyte picture resembles closely that of acute infections. (C). Pseudoleukemia Under the heading Pseudoleukemia have been grouped a great variety of diseases, which have in some cases the external appearances of the disease, such as the glandular swelling, splenic tumor and progressive cachexia without, however, showing the blood picture which is so characteristic of leukemia. On the other hand, we find conditions which have little in common with the clinical findings of leukemia and yet show a blood picture similar in some respects to that of leukemia. It is difficult to group all of these cases under one heading, as the blood-picture is not characteristic for any one of these conditions, but it seems wise to the writer to discuss certain of these states under the above heading of Pseudoleukemia. (1). Hodgkin's Disease. This condition, first described by Hodgkin in 1823, is characterized by chronically progressing cachexia with enlargement of the lymph-glands and spleen. It has been called, synonymously, lymphatic pseudoleukemia, lymphosarcoma, malignant lymphoma, and lymphatic anemia. It has nothing in common with the glandular tuberculosis, and gummatous lymphoma of syphilis, and may be sharply differentiated from these conditions. The blood characteristics in this disease are more particularly those of a true cachexia, the red cells showing a diminution in the number which may be as low as 2,200,000, but which is usually between 3,000,000 and 4,000,000. The more severe and pronounced the signs of anemia and cachexia the lower the number of cells, the number in these cases running as low as 1,500,000. Morphologic- ally, the red cells show much less deviation from the normal than in other severe anemias, the average size of the cell being usually normal, degenerations of the red cells appearing only in the very severe conditions, microcytes and macrocytes, as well as nucleated erythrocytes, being very unusual except in the late stages. The hemoglobin content runs parallel to the number of red cells, being the lowest in those cases showing very low counts. The leucocytes are slightly increased, averaging about 12,000. This failure of a leucocythemia enables us to differentiate Hodgkin's disease from a true lymphemia. The differential count of the leucocytes may show a relative lymphocytosis, the relation of the lymphocytes to the polynuclear cells being as three to one instead of the normal one to three.1 Ehrlich and Pinkus consider this relative increase of lymphocytes characteristic of true pseudoleukemia in contradis- tinction to sarcomatous and other lymphomatous conditions. According to Grawitz, an increase in the leucocytes is associated with an unsatisfactory course of the disease, while a diminution in the number is observed as the disease progresses toward convalescence. 1 Bunting (Bull. Johns Hopk. Hosp., vol. 22, 1911, pp. 114 and 369) considers an increase in the number of transitional cells together with increase of blood platelets as quite character- istic of true Hodgkin's disease. THE BLOOD. 523 Cases have been reported which would seem to point to the transition of Hodgkin's disease into a true leukemia, so that we may find irregular prog- ress of a pseudoleukemia as an evidence of a transitional stage. (2). Tuberculosis of the Lymph-glands. Why this condition has been classed as a pseudoleukemia is questionable, as the blood picture shows nothing beyond a secondary anemia with cachexia or may show even a normal red and white count. A large increase in the number of leucocytes, which is so characteristic of leukemia, is rarely seen, although a true leukemia may arise in the course of a glandular tuberculosis. The differentiation of this condition should be based upon examination of an excised gland, which will show distinct tuberculous lesions and usually will contain the tubercle bacilli in demonstrable num- bers.1 In other cases we may find a simple lymphoid hyperplasia without any distinct inflammatory changes and without demonstrable bacilli. Such nodes should be tested by inoculation experiments as advised by Ewing. Heredity plays a great role in the diagnosis of these tubercular conditions, while a scrofulous child should always be looked upon with suspicion. A splenic tumor appearing coincidently with the glandular swelling would speak rather against tuberculosis and in favor of a lymphatic pseudoleukemia. The diagnosis of this condition by examination of the blood alone is a practical impossibility. It is this type of case that is especially amenable to diagnosis by the use of the various tuberculin tests. (3). Lymphosarcoma. The lymphosarcomata usually run their course either as primary benign lymphomata or as the malignant sarcomata. The blood findings in these conditions show nothing beyond a slight anemia with nothing characteristic in the appearance of the white cells. The diagnosis must be based entirely on the examination of the excised gland or tumor. (4). Gummatous Lymphoma. An exact diagnosis of this syphilitic swelling of the lymphatic glands is at the present time a matter of more or less difficulty. The previous history, as well as other manifestations of syphilis, must be studied and a careful search made for the presence of the Spirochaeta pallida. It is to be said that these organisms have been frequently reported in lymphatic enlargements which seem to have no direct relationship to purely syphilitic conditions. The interpretation of one's findings is of the utmost import as artefacts, which commonly appear in preparations of broken-down glandular tissue, resemble very closely the spirochaete. The blood condition shows nothing characteristic and can have only incidental diagnostic importance. Application of the Wassermann serum test might throw much light on the diagnosis. (II). General Pathology. (a). Blood Changes Following Surgical Intervention. Under this heading the writer will not attempt to take up all of the surgical 1 Resort should be made to the antiformin method, if ordinary staining processes are not conclusive. 524 DIAGNOSTIC METHODS. conditions, as the vast majority are not associated with any direct hematological characteristics. As a rule, it may be said that pus formation anywhere in the system will cause a leucocytosis. The degree of this leucocytosis averages about twice the normal standard, but may greatly exceed this figure in individual cases. It must be said, however, that trivial as well as extensive pus formations may be accompanied by normal or even subnormal values for the leucocytes. This is due to the facts that small pus foci do not cause any systemic reaction and extensive pus formation may overcome the power of the system to react against the infection. If the pus cavity is well encapsulated, the absorption of the toxin from this focus is necessarily limited so that we may find no leucocy- tosis, even though a very large pus cavity is present. Thus we find in localized peritonitis following appendicitis that a leucocyte count may be normal, but may suddenly increase to a marked extent as an indication of the rupture of the cavity and an extension of the process. A general rule is that a distinct increase in the number of cells in excess of the figure originally obtained is indicative of the extension of pus formation and should put the surgeon on his guard as to operative interference. If the absorption of toxic material from a focus of pus formation is great enough to produce a systemic effect upon a patient as manifested by a high leucocyte count, we find an anemia characterized by a marked diminution in the hemoglobin and the number of red cells, which is parallel in intensity to the severity of the poisoning. Cases are frequently found in which a low leucocyte count prevails, although clinical evidence of severe sepsis is at hand. In these cases a differential count of the leucocytes should be made in all cases, as the low leucocyte count may throw one off his guard unless this precaution be taken. Should the polynuclear neutrophiles form 80 to 95 per cent, of the total leucocytes, a severe infection is indicated, even though the leucocyte count may be subnormal. If a leucocyte count does not diminish or even return to normal within one to two days after operation, this should be taken as evidence that a reinfection has occurred or that the pus cavity has not been properly drained. Recourse must, therefore, be had to measures to overcome the secondary infection. It has been found that administration of ether and chloroform causes a leucocytosis which usually lasts from 24 hours to 48 hours and which may interfere with the interpretation of a blood examination. This point must be borne in mind in the examination of blood of cases which have shown high leucocyte values prior to operative procedure. The differential count in such cases, however, will show only slight variations in the proportions of the dif- ferent types of cells so that one may judge as to the cause of the secondary leucocytosis by determination of the polynuclear cells. The number of red cells as also the amount of hemoglobin are very markedly reduced in some cases following the administration of an anesthetic, so that a direct secondary THE BLOOD. 525 anemia may be the result. This fact has led to the refusal by many surgeons to resort to operative procedure in cases which show as low as 30 per cent, hemoglobin prior to operation. As the blood changes in surgical conditions, which are not accompanied by pus formation, are not especially characteristic and are more especially associated with the diseases of the special organs, the writer will refer such discussions to other headings. (&). Blood in Constitutional Diseases (1). Diabetes Mellitus. In diabetes mellitus the changes in the cellular content of the blood are not very marked. The leucocytes may be subnormal, normal, or slightly in- creased, usually a very slight leucocytosis being observed. The amount of hemoglobin is usually reduced to a very slight extent, while the number of red cells may be slightly increased. One of the most striking peculiarities of the blood in diabetes is the presence of an excess of fat (lipemia). Microscopic examination usually reveals the presence of the extracellular globules, but in some cases fat is present in sufficient amount to permit of demonstration by macroscopic methods. Glyco- gen has been found both in the plasma and in the leucocytes of diabetic blood and shows the peculiar characteristics discussed in the section on lodophilia (p. 487). Certain peculiarities of the blood in diabetes have led to the in- troduction of tests supposed to be characteristic for such blood. These tests are occasionally of diagnostic value, although a diagnosis may, as a rule, be made even when these tests do or do not obtain. Bremer's Test.-This test is based upon the fact that diabetic blood does not stain to any appreciable extent when treated with certain aniline dyes. Thick smears of the blood are made upon slides and are fixed by dry heat. These smears are then covered with a 1 per cent, aqueous solution of Congo red and allowed to stain for a few minutes, after which they are rinsed in water and dried. Diabetic blood will be stained either a faint yellow or not at all, while normal blood will be colored a bright red. A 1 per cent, solution of Biebrich scarlet will stain the diabetic blood intensely while the normal blood is unstained. Bremer's original staining solution was made up as follows: Saturated watery solution of eosin and of methylene blue are mixed in equal proportion when a precipitate forms which is filtered, washed, dried and powdered. To 24 parts of this powder are added six of powdered methylene blue and one of eosin. One-twentieth of a gm. of this mixed powder is dissolved in 10 c.c. of 33 per cent, alcohol and forms the staining solution in which the specimens are stained for four minutes. The diabetic blood stained by this solution has a greenish tint, while normal blood is reddish violet. Similar reactions have been found in normal blood, in leukemia, in exoph- thalmic goiter, in Hodgkin's disease, in multiple neuritis, and in some cachectic 526 DIAGNOSTIC METHODS. conditions, but the reaction in all these cases is very inconstant. According to Bremer, cases of renal diabetes do not give this reaction, which is more characteristic of the pancreatic type of the disease. Williamson's Test.-This test is performed as follows: Two drops of blood (20 cmm.) are dissolved in four drops (40 cmm.) of water and to the solution is added 1 c.c. of a 1 to 6,000 aqueous methylene blue solution. To this is added 4 drops (40 cmm.) of 6 per cent, solution of liquor potassae and the test-tube placed in boiling water for four minutes. Diabetic blood will decolorize the solution, while normal blood leaves it a deep blue. The same effect is observed by using diabetic urine instead of the blood. Little iniormation is available as to the variations in the number ol the red and white cells in the gouty condition. The recorded observations show that acute gout has little effect upon the number of red cells and upon the amount of hemoglobin, while chronic gout may be accompanied by an anemia which is more directly referable to causes other than the gouty condition itself. The leucocytes are usually increased in the acute attack, while in the chronic form the leucocytosis is of a more moderate grade. Neusser in working upon the blood of gouty patients found many polymorphonuclear leucocytes whose nuclei were surrounded by basophile granules-the so-called perinuclear baso- phile granules. These he considered diagnostic of the uric acid diathesis, but Futcher and Simon have found them in many other conditions, while Ehrlich regards them as artefacts. While the chemistry of the blood in gout has been the subject of much investigation for a long period, nothing of diagnostic importance has been found in the chemical properties of the blood. The excess of uric acid has been shown not to be pathognomonic of gout, as it is present in many other conditions which are clinically far removed from the gouty state. (2). Gout (3). Addison's Disease. This disease is usually associated with a severe grade of anemia, the number of red cells being reported as low as 1,120,000 while the percentage of hemoglobin is coincidently reduced. The leucocytes are usually diminished, but may be slightly increased, while the relative proportions of the different cells are not markedly changed. As the disease progresses unfavorably, a relative lymphocytosis may be observed, but this is not always the case. (4). Rickets. The state of the blood in rickets varies with the extent and severity of the primary disease and is markedly affected by complications. Cases are re- ported in which the red cells are practically normal and in which the hemoglobin was only very slightly reduced. This disease is not associated with any special type of anemia, although the hyperemia of the bone-marrow might be expected to yield a large number of nucleated red cells. The usual condition of the THE BLOOD. 527 blood in rachitic children is of the type of simple chlorotic anemia. A grave secondary anemia is seen in many cases in which there are serious complica- tions. The leucocytes in practically all cases of rickets are increased, but may not exceed the normal limits for the child. As is usual in the blood of a child, the lymphocytes are increased while the eosinophile cells are often relatively numerous.1 Just exactly what the cause of the leucocytosis in rickets is must be left to the realm of hypothesis, as neither the gastroenteritis nor the hyperplastic splenitis are sufficient to explain all cases. (5). Myxedema. In this disease we usually find an anemia of the secondary chlorotic type along with a moderate leucocytosis. The number of cells is usually somewhat diminished, although their size is usually increased. The proportion of the different leucocytes does not vary, although a few myelocytes are sometimes seen in the blood, which is numerically normal in other respects. Although many studies of the chemistry of the blood in myxedema have been made, little knowledge has been forthcoming as to the exact cause of this toxemia. It is highly probable that the changed activity of the thyroid gland in this disease influences other organs to such an extent that slight anemia is the result and we should, therefore, assume that this anemia is more a secondary one than a primary result of thyroid insufficiency. (c). Blood in Acute Infectious Diseases. In a study of the blood in acute infectious diseases, we must remember that there are certain general rules which apply to all of such diseases with a very few exceptions. The interrelation of the fever with the resistance of the system in general is so close that it is hard to say in any given case whether certain changes in the blood are or are not due to the increased temperature in itself. There can be little doubt that a high temperature working over a considerable period of time will destroy large numbers of red cells and will bring on various changes in the blood which might be misinterpreted. A rather extensive concentration of the blood along with a progressive loss of albumin is observed in practically all conditions associated with fever. We should, therefore, expect to find the number of red cells increased at the outset of such condition, owing to the concentration of the fluid portion, while a dis- tinct anemic condition may become evident only after the lapse of some time. Fever in itself does not have a large influence upon the number of leuco- cytes, but it may be stated as a general rule that most infectious diseases (the exceptions being malaria, typhoid, tuberculosis, influenza, and measles) are associated with an increase in the number of white cells. This is not an invariable rule, as will be seen under the discussion of the various infectious types. The leucocytosis so commonly seen associated with infection is no doubt due to the action of the bacteria themselves and of their products upon the leucocytes. The positive chemotaxis which bacteria and their 1 See Ostrowski, Folia Hsemat., Bd. 13, 1912, S. 305. 528 DIAGNOSTIC METHODS. toxins exert upon the leucocytes is very marked. Moreover, as the blood becomes laden with these abnormal products, new leucocytes are thrown into the circulation to aid the old ones in their phagocytic action. Just what substances are accountable for the increased opsonic power of the serum in any specific infection must be left undecided for the present. In the following discussion of the various infectious diseases, the writer will not attempt to give more than a brief discussion of the blood changes in these separate conditions, leaving associated questions to other writers. (i). Pneumonia. This disease is, hematologically, one of the most definitely characterized of all the infectious diseases. While the physical findings of this condition are largely local, the systemic effects are so marked that definite changes are seen in the blood both in the early and in the later stages of the infection. While showing so many characteristic findings in the blood, it is, at the same time, one of the most obscure in its relations to opsonins and to phagocytosis. Just why the virulent pneumococci should be so little capable of phagocytosis and why the addition of attenuated cultures of pneumococci or of extracts of virulent organisms should increase this phagocytic power of the leucocytes is at present very uncertain. We must, therefore, leave the discussion of this phase of pneumonia as well as of other infectious diseases to the section on Bacteriology of the Blood. In pneumonia we find in the early stages that the blood is somewhat concentrated owing to the action of the increased temperature, while this concentration gradually increases as the exudate forms. Such a condition can lead only to an increased count of both the red and the white cells. As the disease progresses, the number of red cells shows a slight but a steady decline, which points not only to a destruction of the red cells, but to a diminished formation. This decrease in the number of red cells is occasionally seen only at the time of crisis, while in the cases in which the diminution is gradual the period of diminution does not usually exceed ten days. It will be seen, there- fore, that the red cells in pneumonia may be about normal in number and at the same time an anemia may be present which becomes evident only after the disease has progressed for some time. The red cells are, as a rule, normal in appearance, but an occasional polychromatophilic cell may be seen, especially in the severe cases. Rarely normoblasts may be observed and very rarely megaloblasts. The hemoglobin usually shows a greater reduction than does the number of red cells, which decrease may become evident only after the fever has sub- sided. A reduction in the hemoglobin below 60 per cent, is very unusual in pneumonia of the pure type. Pneumonia is one condition in which the leucocyte count may prove of great value. A leucocytosis appears in most of the cases, being absent in very mild cases as well as in those very severe ones which show very feeble resistance of the organism toward invasion by the pneumococci. Rieder's THE BLOOD. 529 observations are very interesting on this subject. He has found that the leucocytosis of pneumonia is more a function of the intensity of the infection and the degree of resistance toward this infection than it is of the fever or of the extent of the exudate. A leucocytosis which may reach 12,000 to 20,000 appears very early in the course of this disease, and is usually evident at the time of the chill or immediately following. A steady increase is sometimes observed in the number of the white cells so that the maximum is usually reached just before the crisis. It is to be said that rapid extension of the disease as well as continuous high temperature may cause much irregularity in the count, cases being reported in which the leucocytes are high at first and steadily diminish as the patients grow worse. Others may show a sudden increase in the number of cells as the time of crisis is approached.1 According to Ewing, when the leu- cocytes increase slowly they usually diminish slowly and the disease defervesces by lysis. The degree of leucocytosis in pneumonia may reach any stage be- tween the normal figure and that of 115,000 as reported by Laehr. The increase in the number of white cells in pneumonia is largely referable to increase in the number of the polynuclear neutrophiles, these cells con- stituting as high as 97 per cent, of the total number of white cells. Associated with this polynuclear leucocytosis we have a marked diminution of lympho- cytes, while the large mononuclear cells usually persist in considerable numbers. The eosinophile cells are always much reduced at the height of the leucocytosis, so that we may not be able to find a single one after very prolonged search. Cabot has reported a case in which the lymphocytes constitued 66 per cent, of a total of 94,600 white cells, but such a finding is not the usual one fol- lowing infection with the pneumococcus. As defervescence goes on the polynuclear cells diminish very rapidly, while the lymphocytes increase and the large mononuclear leucocytes become very numerous, reaching as high as 16 per cent, in a case reported by Turk. The eosinophile cells usually appear about the time of crisis, but occasionally their appearance is postcritical. The degenerative changes seen in the leucocytes in pneumonia are in no way different from those observed in other infectious diseases. It will be noted, from the above remarks, that the blood changes in pneu- monia are those of a mild anemia associated with a high-grade polynuclear leucocytosis and a distinct lymphopenia. Too much reliance must not be placed on the blood finding in a case of pneumonia, owing to the fact that many abnormal cases are present and show results different from the above which can be interpreted only by a complete study of the complications in any special case. As a general rule, it is to be said that an absence of leucocy- tosis is strong negative evidence against pneumonia, while leucocytosis may serve to differentiate this condition from typhoid fever and malaria with which it might be confounded, especially where the systemic and cerebral symptoms are more pronounced than are the local pulmonary changes. 1 Dick (Jour, of Inf. Dis., vol. io, 1912, p. 383) shows that proteolytic ferments develop in the blood about the time of crisis. 530 DIAGNOSTIC METHODS. (2). Typhoid Fever. This condition, like the preceding, is very often associated with such marked systemic disturbance that the local intestinal manifestations are obscured and the diagnosis rendered somewhat difficult. While typhoid fever is a purely infectious condition and subject to the ordinary laws governing such cases, yet we find for some reason that the invasion of the blood by the specific organism is not associated with a leucocytosis, although the febrile rise may be very marked. The study of the characteristic serum reaction for typhoid fever as well as of the bacteriology of the blood must be left for a later section, the discussion here being limited to the changes in the microscopic appearances of the blood. In this condition we find the total volume of blood very much diminished in the early stages, both as a result of the high temperature and the diarrhea and repeated hemorrhages which may occur at any stage of the disease. This concentration of the blood leads to an initial polycythemia which may last for two or even three weeks However the characteristic change in the red cells is one of a slight and gradual decline, the number of these cells not usually falling below 4,000,000. One must be on his guard in an examination of the blood in any infectious disease lest he conclude from a slightly increased count that no anemia is present. It is a very safe precaution, although rarely followed, to determine the specific gravity of the blood so that one may compute the degree of concentration. In this way he may be able to show that the number of cells normally present in such a concentrated blood is much higher than in the case of suspected typhoid fever which he is examining. At any rate, it is wise to make frequent determinations of both the number of cells and of the hemoglobin, as a reduction in both of these elements takes place gradually as in the cases reported by Thayer and Da Costa. The reduction in the amount of hemoglobin is in some cases very marked, being as low as 50 per cent, in one case observed by the writer. The repeated hemorrhages which so often occur in typhoid fever may cause marked varia- tion in this value. The morphological changes which occur in the red cells are not very marked, as a rule, but may be very severe in case much blood is lost by frequent hemorrhages. Polychromatophilia is more or less frequent and irregularity in the size of the red cells is occasionally seen, while nucleation of these cells along with formation of a few megaloblasts may occur in severe hemorrhagic cases. The leucocytes are usually normal in number in the early stages of uncom- plicated cases, but any complication may cause a polynuclear leucocytosis which may be confusing to the worker. The behavior of the leucocytes is very variable in the early stages, so that one should never rest his diagnosis of typhoid fever upon a negative leucocytosis. As the disease progresses, the leucocytes show a gradual reduction, especially in the number of the polynu- clear cells, which reduction continues until the disease has reached its highest point, after which they slowly increase. THE BLOOD. 531 The more severe the action of the typhoid toxin the lower is the leucocyte count, the reduction not usually going below 2,500 cells, the majority of cases showing a count between 4,000 and 6,000. It is not an uncommon thing to observe a leucocytosis during the later course of typhoid fever and it is not always easy to explain such a condition. Marked hemorrhage, cold baths, severe diarrhea, and usually perforation may account for the increase in the number of cells, but we do not always find a leucocytosis after such conditions. In typhoid fever we find quite characteristic changes in the relations of the various types of leucocytes. During the first week the neutrophile cells do not, as a rule, increase, while the lymphocytes, especially of the medium-size variety, show a progressive rise. The lymphocytes are rarely below 25 per cent, of the total number of cells and may reach as high as 65 to 70 per cent. The eosinophile cells are usually low in number during the febrile period, but reappear about the time of defervescense. It will thus be seen that the characteristic changes in the blood of typhoid fever are a slight anemia, together with a leucopenia and a relative, and in some cases absolute, lymphocytosis. Such characteristics are the usual ones of typhoid fever, but it is to be remembered that suppurative processes do not always produce a leucocytosis, nor is a leucopenia always present in typhoid fever. For the relatively certain differentiation of typhoid fever from other diseases it is necessary to perform the Widal test, and even this may not always be present. A discussion of this test will be given in a later section. In this condition we find the usual effects of fever manifested in a slight con- centration of the blood, leading in the early stages to a polycythemia. The usual change, however, in the red cells is one of gradual reduction in numbers to as low as 3,000,000 cells and occasionally much lower. The hemoglobin also suffers quite a diminution, so that the anemia may reach quite a severe grade. The leucocytes in scarlet fever usually increase in number one or two days before the appearance of the rash and continue to increase until quite a marked leucocytosis, ranging from 10,000 to 50,000 cells, becomes evident at the time of the complete eruption. The degree of the leucocytosis cor- responds as a rule with the severity of the disease and in some cases is diminished at the time of the eruption, but usually continues for several days and may even extend for weeks after the temperature has subsided. The increase in the number of leucocytes is largely referable to the poly- nuclear cells, these constituting from 85 to 99 per cent, of the total number. The lymphocytes diminish in the early stages of the disease, but later rise to normal or slightly above normal figures. The eosinophile cells are usually normal or even subnormal at first, but steadily increase as the disease pro- gresses and reach a degree of 10 to 20 per cent, in the second or third week, after which they slowly decline. These rules are not invariable in scarlet (3). Scarlet Fever.1 1 See Koessler, Jour. A. M. A., vol. 59, 1912, p. 1528, for a discussion of the recent advances regarding scarlet fever. 532 DIAGNOSTIC METHODS. fever, but a severe leucocytosis appearing prior to the period of eruption of an in- fectious fever is practically always suggestive of this disease. In some cases the polynuclear cells diminish about the end of the first week and the lymphocytes and eosinophiles rapidly increase, leading to a later secondary leucocytosis.1 (4). Measles. This condition shows in itself nothing particularly characteristic in the blood, but the absence of definite findings is of great importance in its differen- tiation from scarlet fever with which it might be confounded. The red cells in this disease are not found to be greatly changed, although a slight reduction in their number is usual. A loss of hemoglobin is practi- cally always noticed, so that we have a distinct anemia which will vary depend- ing on the complications which may arise in the course of the disease. The leucocytes are usually normal or slightly reduced in number at the outset of the disease, being the lowest at the height of the eruption when the figure may reach as low as 2,500 cells, returning to the normal within a few days after subsidence of fever. A complicating bronchitis may cause a moder- ate leucocytosis of 8,000 to 16,000 cells, but this should not lead one to a mis- taken diagnosis, as the clinical symptoms of both scarlet fever and of measles should be well-established at the time of the complicating bronchitis. An eruptive fever in the second or third day of its course should be considered scarlet fever, or at least scarlatina, if a leucocytosis is present, while if the disease be measles the number of leucocytes will be normal or even subnormal in the absence of extensive bronchitis. (5). Variola. This condition is associated with more or less extensive destruction of the red cells. In the early stages of the disease the red count may be slightly above normal, owing to the concentrating effect of the fever, but later the red cells will show a sudden reduction. This reduction is especially noticeable in the cases associated with extensive pustulation, when the septic process has such a marked influence in destroying the red cells. The hemoglobin is usually reduced in degree parallel to the diminution in red cells, so that we may have an anemia ranging from the mild to severe type. In most of the cases of small-pox we find a distinct leucocytosis which may run from 10,000 to 20,000 as a rule, but has reached as high as 41,000 in the severe cases. This leucocvtosis beeins with the anoearance of the vesicle. 1 Dbhle (Centralbl. f. Bakteriol., Bd. 61, 1911, S. 63; Ibid, Bd. 65, 1912, S. 57; Miinch. med. Wochs., Bd. 59, 1912, S. 1688) has announced the almost constant presence in the poly- morphonuclear leucocytes of "inclusion bodies." These are observed near the margin of the cell as rod or coccus forms, which stain readily a faint bluish-green tint with the Giemsa or other blood stains. While these bodies are observed in about 95 per cent, of scarlet fever cases in the early days of the disease, the number diminishes gradually. They are also found in many cases of diphtheria, pneumonia and tuberculosis, yet they are very infrequent in cases of serum sickness or of scarlatiniform rashes. Their presence is, therefore, not pathognomonic but their absence almost excludes scarlet fever. See Kretschmer, Berl. klin. Wochs., Bd. 49, 1912, S. 499; Deutsch, med. Wochs., Bd. 38, 1912, S. 2163; Nicoll and Williams, Arch, of Pediat., vol. 24, 1912, p. 350; Ahmed, Berl. klin. Wochs., Bd. 49, 1912, S. 1232; Kolmer, Amer. Jour. Dis. of Child., vol. 4, 1912, p. 1; Granger and Pole, Brit. Jour. Child. Dis., 1913, p. 9. THE BLOOD. 533 increases as the exudate becomes purulent, and reaches its height when sup- puration becomes extensive; that is, the degree of leucocytosis usually runs parallel to the severity of the septic process, the count returning gradually to normal as the suppuration subsides. The leucocytosis in small-pox is usually of the lymphocytic type, the number of these cells varying from 35 to 45 per cent, of the white cells. Associated with the increased lymphocyte value we find from 5 to 10 per cent, of the large mononuclear leucocytes and usually an average of 3 per cent, of neutrophile myelocytes, in some cases these latter running as high as 16 per cent. Eosinophiles and basophiles are occasionally observed, especially in the hemorrhagic form of this disease. It will thus be seen that small-pox causes a leucocytosis which may reach even the degrees given by scarlet fever, but the differential count as shown by the large percentage of lymphocytes in the former and the greatly increased number of polynuclear neutrophiles in the latter should make a mistake in diagnosis impossible. It is to be remembered that complications of true abscesses with the pustules of small-pox may increase the percentage of polynuclear cells in this disease, but never to such a degree as is shown in scarlet fever. An examination of the blood of children, who have been vaccinated with small-pox virus, shows a distinct leucocytosis of the polynuclear type, reaching as high as 20,000 cells. This leucocytosis usually begins on the third or fourth day after inoculation and gradually subsides until the end of the period of vaccination. It is rather hard to understand why the leucocytosis in this condition should not take the same form as in the true small-pox, but no ex- planation is at present available. The blood in cases of varicella1 seems to show the same characteristics as that of very mild cases of variola, or of vaccinia. The slight leucocytosis is usually of the polynuclear type, reaching a degree of 15,000. The large mononuclear cells which seem to play such an important role in the differential leucocyte picture of variola are for the most part absent in varicella, and myelocytes are practically never found in this condition. The number of red cells in diphtheria seems to be slightly increased, owing to the marked concentration of the blood in this condition. In practi- cally all cases of diphtheria the number ranges from a high normal value to as high as 7,800,000 reported by Cuffer. As the disease progresses, especially after the temperature has fallen, the number of red cells is diminished and a coincident decrease in the percentage of hemoglobin is observed. This slight anemia is not evident in the early stages of the disease due, no doubt, to the abnormal concentration of the blood. In this condition, like most infections, we find a leucocytosis ranging between 25,000 and 50,000, the higher the leucocytosis the more grave the prognosis. In one case reported by Felsenthal the leucocytes numbered 148,000, but this is very unusual, as the grade of leucocytosis is ordinarily (6). Diphtheria. 1 See Mensi, Gazz. d. Osp., Tm. 33, 1912, p. 1625. 534 DIAGNOSTIC METHODS. proportional to the extent and depth of the membrane. The leucocytosis is usually of the polynuclear type, the lymphocytes being also slightly in- creased. In some cases a lymphocytosis of 60 per cent. (Ewing) has been observed, but this is not the usual finding. The eosinophile cells are reduced in number, but are relatively more numerous than in pneumonia. It is in diphtheria that we find quite marked degenerative changes in the leucocytes, "the leucocyte shadows" and increased acidophile tendency of the neutrophile granules being especially worthy of mention. (7). Pertussis. The recent work of Barach1 and others has so modified our ideas of the changes in the blood in pertussis that I can do no better than to give his summary upon these points: "In the early stages of this disease there is a leucocytosis with increase of all the forms; then a small-cell lymphocytosis becomes conspicuous and continues to increase when the other forms have reached their limit. The large lymphocytes follow the course of the small ones, but they reach their greatest numbers after the small cells have reached theirs. During the stage of active lymphocytosis, bilobed, small lymphocytes are frequently seen as well as numerous degenerated large lymphocytes, especially the basket forms. Then comes the simultaneous falling of the leucocytosis and lymphocytosis, while the polynuclears begin to resume their normal proportion. A little later the mast cells are observed more frequently, and an occasional myelocyte may be seen. While the leucocytosis and lymphocytosis continue to fall by lysis, an eosinophilia is noted; this continues for a variable time, after which the blood formula resumes its normal proportions. During this entire cycle the transitionals seem unaffected. If we were to speak of the first and second half of the blood cycle in this disease, we would say that in the first half the lymphocytes are the prominent factors and in the second half the polynuclears and the eosinophiles. Clinically, leucocytosis is present at about the time the child first coughs; as the coughing goes on, the leucocytosis increases and the lymphocytosis becomes very marked. Churchill believes that a lymphocytosis exists in the prespasmodic stage and is of extreme importance in early diagnosis. The height of the leucocytosis is reached in the spasmodic stage, sometimes early, and sometimes in the latter part, the sickest children showing the highest grade of leucocytosis. About the time that a marked improvement is noted in the child the leucocytosis has decreased, the polynuclears have increased and the eosinophilia is present." The degree of leucocytosis in this condition varies between 25,000 and 5IJ5O. (8). Acute Rheumatism. In this disease we find that the red cells are quite markedly destroyed, causing, very frequently, a reduction of 2,500,000 cells. This reduction is 1 Arch of Int. Med., vol. i, 1908, p. 602. THE BLOOD. 535 not always evident in the early stages of the disease, as the blood becomes very much concentrated by the marked sweating which is such a prominent symptom of the disease. The hemoglobin suffers more than the red cells so that we may find the percentage of hemoglobin as low as 60 per cent, in this condition. This anemia is one of the characteristic signs of acute rheumatism and continues well on into convalescence, the hemoglobin not being as quickly restored as are the red cells. A largely increased formation of fibrin has been observed and may have some diagnostic importance. The leucocytes are increased in proportion to the severity and acuteness of the disease, the grade being usually moderate and the type being poly- morphonuclear. In the very mild cases we may find no leucocytosis and one reaching 20,000 or more is, according to Turk, always associated with com- plications. As the fever diminishes the leucocytes'return to normal and are not as much affected by the subsequent attack as by the previous initial one. The eosinophile cells are absent only in the early stages, while they are present in moderate amounts later in the disease and show a distinct increase after defervescence (Loeffler). It would be impossible in a work of this character to discuss in detail the blood changes in all diseases, whether infectious or noninfectious. The writer, therefore, has selected under the acute infectious diseases those which show the more characteristic changes in the blood and those in which an examination of the blood is more frequently called upon to aid in diagnosis. Many phases of these diseases have been left to the chapters on bacteriology of the blood, while those not taken up at all in this work must be looked for in the many books on general medicine. (J). Blood in Chronic Infections (1). Tuberculosis. The earlier studies of the blood of tubercular patients reveal the fact that the blood may show practically no changes which are comparable with the pallor of the skin and the degree of the emaciation of the subjects affected. An anemia is often seen of the very highest type, but usually one of moderate degree is present and even may not exist at all. The degree of anemia is independent of the localization of the disease, although pulmonary affections are more frequently associated with high-grade anemia than are other tubercu- lar conditions. It is to be remembered that pulmonary tuberculosis is so frequently associated with extensive hemorrhage that one may not wonder at the severe anemia present, yet we find cases in which the regeneration is very rapid after severe hemoptysis. As a rule, a mild anemia of the chlorotic type prevails; that is, the count is practically normal with the hemoglobin some- what reduced. Occasionally we find a slight lymphocytosis, especially of the smaller cells, and only when a secondary infection prevails does the leuco- cytosis take on the polynuclear type.1 The lymphocytosis in tuberculosis is 1 See Miller and Reid, Arch, of Int. Med., vol. 9, 1912, p. 609; Miller, Lupton and Brown, Amer. Jour, of Med. Sci., vol. 143, 1912, p. 683; Ringer, Ibid., vol. 144, 1912, p. 561. 536 DIAGNOSTIC METHODS. so common that we usually find even in the sputum an excess of the small mononuclear cells in the pure tubercular affection of the lungs. In tubercular infection of the meninges we practically always find a leucocytosis, but with this exception uncomplicated tuberculosis is not asso- ciated with an increase in the number of white cells. It is highly probable that the increased percentage of the mononuclear cells is more closely associated with the poor nutrition which the tubercular patient shows than it is with any specific effect of the bacillus tuberculosis. Occasionally we find the eosinophile cells somewhat increased, especially in pulmonary conditions with cavity formation, but it must be remembered that a slight eosinophilia will obtain if tuberculin therapy is being used in such cases. (2). Syphilis. According to Becquerel and Rodier, a moderate grade of anemia is to be found in the majority of cases of syphilis, becoming more pronounced as the disease progresses. This anemia of syphilis is of the chlorotic type, but may increase until the pernicious type becomes established. As a rule, the reduction in the number of red cells is moderate, being rarely below 3,000,000 cells. The hemoglobin is usually relatively more decreased than are the cells, and the application of mercury in the treatment of this condition frequently lowers this percentage still further, establishing an anemia which is directly referable to the mercury. This reduction in the number of red cells and in hemoglobin becomes more marked as secondary symptoms appear, so that a diagnosis of an initial lesion becomes established by a later examination of the blood. The increase in the number of leucocytes is largely limited to the secondary and tertiary stages of this disease, as the leucocytes usually remain normal up to the time of the eruptive stage. The increase in the secondary stage, which may reach as high as 20,000, is largely in the number of the small and large lymphocytes, but the eosinophile cells may be increased to as high as 5 per cent, in some cases. In very severe cases a progressive polynuclear leucocytosis is observed. As the tertiary stage comes on the leucocytosis usually persists, but the lymphocytosis becomes less distinct and constant through the increase in the number of polynuclear cells. Justus' Test. Justus has found, in studying the blood of patients suffering with florid syphilis, that injection or inunction of preparations of mercury cause a reduction in the percentage of hemoglobin of from io to 20 per cent, for a period of a few hours or days. After a certain time, varying with the general condition of the patient and the severity of the symptoms, the hemoglobin increases again. This test can hardly be considered diagnostic of syphilis, as the mer- cury salts all cause an anemia which may be directly traceable to their hemo- lytic action upon the red cells. In the blood of patients suffering with congenital syphilis we always find a THE BLOOD. 537 distinct anemia, associated with a slight leucocytosis, especially of the lymph- ocytic type. The red cells in this condition show many changes, such as poly- chromatophilia and nucleation, while the changes in the white cells often resemble the picture of a mild grade of leukemia. For a discussion of the causa- tive factor, the Spirochaeta pallida, see the section on Parasitology (p. 556). (3). Leprosy. The blood in leprosy is quite different from that in either of the two pre- vious conditions. The usual rule is a very slight reduction in the number of red cells, although cases have been reported with a red count of 1,900,000 and a blood picture of pernicious anemia. The hemoglobin does not seem to be reduced to any extent, the percentage usually being relatively higher than the number of red cells, so that a high color index almost invariably obtains. The leucocytes are rarely increased in number, being usually sub- normal, with a relative increase in the number of lymphocytes, their percentage reaching as high as 47 per cent., according to Winiarski. (4). Carcinoma. Although discussed under the heading of Chronic Infections, carcinoma has at present no etiological relation to such conditions. This is one of the most important causes of anemia, owing to the frequent hemorrhages and the mechanical effects of the growth as well as to the unknown toxin, which may produce severe constitutional symptoms, even though the growth may have become latent. The anemia of malignant disease usually runs parallel to the progressive cachexia. The grade of anemia may vary, depending upon the location of the tumor, from a very mild chlorotic anemia to one with a perfect picture of pernicious anemia. It is natural to suppose that the more malignant the disease the greater will the blood changes be, so that we should expect to find the rapidly growing cancers which form numerous metastases usually associated with the more extreme blood picture. That this is not always the case is shown by the following statement of Emerson: "Our cases with rapidly developing metastases, with large nodules, are those with a slight chlorotic anemia; those which simulate pernicious anemia are more often those with few objective signs of cancer, an insignificant-looking little nodule." It is possible that this paradox may be explained by the fact that the development of the cancer is so rapid that the toxin has not had sufficient time to cause the blood changes which the more slowly developing growth may bring about. The changes in metabolism are much more marked in the slowly developing cancers than in the more rapidly growing ones, so that we might assume that the same rule applies to the blood changes as expressions of the general systemic disturbances. When the anemia of cancer develops it is usually more severe than in any other chronic disease. The chief changes are at first in the size, shape, weight, and degeneration of the red blood-cells; later, as the cachexia develops the red cells are often as low as 2,500,000 or even as low as in pernicious anemia, 538 DIAGNOSTIC METHODS. 1,000,000. The hemoglobin is always reduced in amount, but is rarely as low as in chlorosis, the average, according to Cabot, being about 58 per cent. The hemoglobin value seems to be lower in cases of visceral cancer than in those of peripheral type. In a majority of the cases a moderate leucocytosis obtains which is never seen in the benign tumors unless these be complicated by suppuration. This leucocytosis depends largely upon the amount of hemorrhage from the tumor and upon the position of the cancer. We find in carcinoma of the stomach and uterus, in which hemorrhages are very frequent, quite an extensive leucocytosis, while in cancer of the esophagus a leucopenia may obtain. The larger and faster the tumor grows the greater will be the degree of leucocytosis, a condition which is the reverse of that usually found in the case of the red cells. The leucocytosis of cancer is usually of the poly- nuclear type, but this may not be over 45 percent., in which cases the lymphocytes are relatively increased. The eosinophiles are rarely as much diminished as in other conditions, but they are not always increased. Myelocytes are, perhaps, more frequently found in cancer than in the other types of anemia, excepting pernicious anemia and leukemia. The degree of cachexia is often very extensive in cancer, but is not always so closely related as one would suppose to the changes in the blood. In those cases in which the cachexia is due to a combination of malnutrition with intoxi- cation by the malignant toxin the blood changes are naturally very severe, but, as previously stated, the more fulminating types of cancer are not associated either with great cachexia or with severe changes in the blood. Cachexia, therefore, seems to be more a function of the chronicity of cancer than of its malignancy. The changes in the blood in the numerous specific diseases of various organs show nothing characteristic in themselves. It is to be expected that in all chronic diseases of whatever organ, a slight anemia may be present owing to the effects of such disorders upon the general metabolism. These changes are, however, of the general type of simple anemia and are usually rapidly remedied by the application of the ordinary therapeutic agents. It is true that disease of the general organs causes changes in the composition of the fluid portions of the blood, which may bring about a secondary change in the cellular content. So closely correlated are the various organs that disorder of one necessarily brings about a disturbance in the normal functions of the others, so that we may find either the normal products of these organs lacking in the blood or abnormal products poured into it through the perverted metabolism induced by primary disease of the correlated organ. It would be useless to outline the blood changes in these general diseases, as nothing is found that could be used as a reliable diagnostic aid in case the blood were examined. It can only be by a careful study of the plasma that variations may prove of value. Our knowledge of such changes in the blood is, however, too meager to warrant any discussion. (e). Effects of Splenectomy. The effects of splenectomy are usually the combined results of severe THE BLOOD. 539 hemorrhage, a preexisting anemia, of the loss of functions of this organ, and of intravenous infusion which has been performed following the operation (Ewing). In comparatively healthy subjects splenectomy has often been performed without affecting the blood more than does any other abdominal operation. The most marked changes in the blood are seen in those cases in which the organ has been removed for rupture or idiopathic enlargement, the loss of blood and the shock of operation giving rise to a considerable degree of secondary anemia. The red cells are frequently restored to normal in one to three months, but in some cases which progress less favorably the anemia may be more per- sistent. The restoration of hemoglobin does not take place, as a rule, as rapidly as does that of the red cells. Following the operation we usually observe a polynuclear leucocytosis which may run, as in one case observed by the writer, as high as 75,000 cells. This leucocytosis usually lasts from one to two months, but may persist for several months, in which case the polynuclear cells are re- placed by lymphocytes. Eosinophilia usually develops early and has been ob- served in some cases two or three years after operation. In some cases, especially those suffering from extensive hemorrhage, a very profound anemia character- ized by great diminution in the number of red cells, the presence of poly- chromatophilic and degenerated cells, nucleated red cells, and a high degree of leucocytosis is observed. The leucocytes in these cases may take on the picture of an acute leukemia, but this condition is transitory as the blood improves more or less rapidly. A leucocytosis or permanent lymphocytosis are probably the only specific effects of splenectomy.1 V. Parasitology of the Blood (1). Malaria (Paludism; Hemamebiasis) Malaria is a disease caused by the entrance of an animal parasite into the blood and its development within the red blood-corpuscle. This parasite was first studied by Laveran and belongs to the class of sporozoa. It was not until the recent work, especially of Grassi, Ross, and Nuttall, that we were enlightened as to the source of this invader. It is at present well estab- lished that the malarial parasite runs its sexual cycle (sporogone) within the body of the anopheles mosquito (Anopheles maculipennis). The old idea that malaria was an air-borne disease, the contagion arising from stagnant pools in swampy regions must now be replaced by the modern mosquito theory. It is true that the anopheles lays its eggs upon the surface of almost stagnant water and that the larvae hatch in these places. The eggs are boat-like in shape (each separate, the groups being arranged in ribbons) and float upon the surface, while the larvae lie just below the surface and are in a plane parallel with it. These facts have led to the adoption of the modern methods of prevention of malaria by covering the surface of such stagnant pools with oil, which prevents access of air to the larvae and, in consequence, causes death.2 'See Musser, Arch, of Int. Med., vol. 9, 1912, p. 592. 2 See McCoy, Pub. Health Rep., vol. 27, 1912, p. 1029. 540 DIAGNOSTIC METHODS. In a discussion of malaria it must be remembered that there are several types of this disease, depending on (i) the kind of parasite causing the infection and (2) the period at which the various groups of the same parasite run their asexual course in the host. We have, therefore, to discuss the three types of infecting organism, each of which is a protozoan form and is found in the red cells. The cycle of development of the tertian organism is approximately 48 hours, so that with a single infection paroxysms will occur on alternate days. With the quartan organism the cycle of development requires 72 hours, while with the estivo-autumnal form it is variable, running from 24 to 72 hours. It is, of course, possible to have infection with more than one form of parasite or with more than one series of the same parasite, so that we may have daily exacerbations through infection with any one of these three types of parasite. In the study of the blood in malaria there are definite changes both in the red and white cells and in the hemoglobin. These, however, are of second- ary importance to the study of the parasite causing the disease. We will, therefore, discuss the parasite before taking up the changes in the cellular com- position of this tissue. In an examination of the blood for the malarial parasite a study of the fresh specimen is always desirable when possible, as the peculiar ameboid movements of the parasite as well as the rapid oscillatory movements of its granules can, of course, not be seen in the fixed specimen. Moreover, the peculiar brassy tone of the red cell and the irregularity in shape and size of these cells may best be studied in the fresh specimen. The beginner, however, will find that a stained specimen will yield much more definite results, providing his staining technic is good than will a study of the fresh specimen, as the slight refractility both of the cell and of the parasite in the fresh specimen makes it difficult in every case to get the proper illumination of the specimen. In the hands of an expert the examination of fresh blood is practically all that is required for a diagnosis in the average case, and when the parasites are moderately numerous the beginner can scarcely make a mistake. It would seem, therefore, inadvisable to rest a diagnosis upon an examination of the fresh specimen in cases in which no organisms are found, but to control this examination by a careful study of a stained specimen in which one may frequently be surprised at the number of parasites to be seen, although a negative result has been observed in the unstained specimen. A wTord of caution, however, is necessary at this point. Frequently one observes in stained specimens many artefacts due to deposition of staining pigments upon the red cell, while in the fresh specimen areas of coagulation necrosis are not infrequently seen, so that the untrained observer may assume the presence of malarial organisms. For an absolute diagnosis of malaria it is necessary to find intracellular organisms, and not to be content with a single examination in doubtful cases.1 1 Bass has recently succeeded in cultivating the various types of malarial plasmodia. See Bass, Jour. A. M. A., vol. 57, 1911, p. 1534; Bass and Johns. Jour, of Exp. Med., vol. 16, 1912, p. 567; Lavinder, Jour. A. M. A., vol. 60, 1913, p. 42. The parasites are cultivated only in the red cells of human blood and are destroyed by the leucocytes as well as by the serum. See, also, Sinton, Ann. of Trop. Med. and parasitol., vol. 6, 1912, p. 371 and Olpp, Munch. Med. Woch., Bd. 59, 1912, S. 2623. PLATE XXVIII. The Tertian Parasite. 1. Normal erythrocyte. 2, 3. 4> 5- Intracellular hyaline forms. 6, 7. Young pigmented intracellular forms. In 6 two distinct parasites inhabit the ery- throcyte, the larger one being actively ameboid, as evidenced by the long tentacular process trailing from the main body of the organism. This ameboid tendency is still better illustrated in 7, by the ribbon-like design formed by the parasite. Note the delicacy of the pigment granules, and their tendency toward peripheral arrange- ment in 6, 7, and 8. 8. Later developmental stage of 7. In 7, 8, and 9 enlargement and pallor of the infected erythrocyte become conspicuous. 9. Mature intracellular pigmented parasite. 10, 11, 12. Segmenting forms. In 10 is shown the early stage of sporulation-the develop- ment of radial striations and peripheral indentations coincidentally with the swarm- ing of the pigment toward the center of the parasite. The completion of this process is illustrated bv it and 12. 13. Large swollen extracellular form. Note the coarse fused blocks of pigment. (Com- pare size with that of normal erythrocyte, 1.) 14. Flagellate form. 15. Shrunken and fragmenting extracellular forms. 16. Vacuolation of an extracellular form. Note.-The original water-color drawings were made from fresh blood specimens, a Leitz ^,-inch oil-immersion objective and 4 ocular, with a Zeiss camera-lucida, being used. (E. F. Faber, fee.} (From Da Costa's "Clinical Hematology ") THE BLOOD. 541 Examination of Fresh Blood. The technic of making a fresh specimen of suspected blood is the same as that previously outlined and consists in touching a perfectly clean cover- slip to a drop of blood and allowing the cover-slip to fall upon a clean glass slide. The quantity of blood should be rather small so that the red cells may be distinctly separated from one another. The examination is best made by the use of a 1/12 immersion lens. The youngest form of the tertian parasite as it appears in the red cell resembles very closely the spore of the parent rosette. It is a small, compact, colorless, nonpigmented disk (hyaline form) about 2 microns in diameter and shows an undulating outer rim of basophilic protoplasm which encloses a single large nuclear body which does not stain with methylene blue but shows a distinct chromatin stain with any of the modifications of the Roma- nowsky stain. This nuclear body is usually surrounded by a clear space which does not take the stain and which has been termed by Gautier "the milky zone." The parasite has a very rapid ameboid movement and shows a great number of changes in shape and position. It sometimes assumes a typical ring-like form which is usually a little thicker at one point, from which the name "signet ring" has been given. Occasionally several of these rings may be seen within a single blood-cell. After about 12 hours the corpuscle increases slightly in size, becomes somewhat paler, but still has the sharp, smooth, round outline of the normal cell. At this stage the ameboid powers of the organism are very marked, so that many pseudopodia may be seen, connected to the larger part of the organism by very thread-like pale and rather indistinct bands of union. This gives the appearance of disconnected globules of protoplasm, which is very slightly refractile. At this period (12 hours) pigment (hematin) appears in the parasite in the form of very fine light brown granules which have a very rapid dancing motion and are clustered especially at the ends of the pseudo- podia. The organism continues to increase in size and, at the same time, the host becomes somewhat larger, paler, but still round in outline. At the end of 24 hours the organism fills about one-third of the cell, is still ameboid and shows increased pigment, which is somewhat darker in color and is less actively motile, being distributed throughout the substance of the parasite. In this form one may occasionally see the nucleus as a globular body at the end of a pseudopod. In the last half of the cycle of development of the tertian organism the growth is much more rapid than in the first half, the parasite being fully developed within 40 hours. The cell at this time is about one and a half times its normal size and is so little refractile that its outline can scarcely be seen. The organism is from eight to ten microns in diameter, is round, and is even less refractile than is the corpuscle. The pigment is much more abundant at this time and is still evenly distributed throughout the organism. The next stage in the development of the organism is known as the pre- (a). The Tertian Organism (Hemameba vivax; Plasmodium vivax). 542 DIAGNOSTIC METHODS. segmenter stage. The cell becomes practically invisible, the pigment collects in one or more irregular clumps throughout the organism, the granules moving in irregular lines. At this time the periphery of the organism shows slight crenation and refractive dots appear irregularly in the periphery of the organism. The line of demarcation between the presegmenter and the segmenter is very slight. The corpuscle is now practically eliminated and the organism becomes more dense and highly refractile. The refractive dots which were visible in the presegmenter stage are now seen to be in the center of lines of separation which pass from the irregular crenated border down toward the center of the organism, thus marking off future segments, which are as a rule from 15 to 20 in number. As development proceeds the segments become more sharply defined until the clumps form into discrete circular masses with a distinctly refractile spot in the center. The pigment in these segment forms seems to be left in masses between the segments without any definite arrangement. Each segment now splits off from the mother segmenter cell and becomes free in the blood in the form of the original hyaline type which becomes attached to a red cell and soon enters it to pass through the various stages discussed. It is to be remembered that the hyaline forms do not modify their host, either in shape, color, or size, such changes being observed only at the time when the pigment first becomes evident. The preceding is a concise description of the cycle of development within the cell (asexual generation or schizogone),\>\it^e occasionally find tertian forms other than hyaline which are extracellular. These seem to be of two types, the degeneration forms and the gametocytes, cells capable of sexual development. The degeneration forms or, as they are sometimes called, extruded intracellulars, are sometimes the only ones seen in the specimen. These are parasites which have passed from the cell and have died, the organism sometimes appearing as if it had passed out through a very fine hole. If it be entirely extruded, the hemoglobin leaves the cell with it and only a shadow of the red corpuscle remains behind. However, this does not always occur, so that we meet with typical dumb-bell-shaped organisms in the plasma. If the blood be observed while this process of extrusion is going on, the pigment will still be extremely active but it gradually becomes quiet as the organism dies. The organism may break up into fragments forming several pigmented spherical masses or it may become deformed and vacuolated, constituting the so-called "speculating" forms. The gametocytes are found at all times in the blood after the infection has been established for a few days. These are not in reality extracellular forms as one sees them in the stained specimens surrounded by the shell of the corpuscle. These gametocyctes are of two forms, the macrogamete or female cell and the microgamete, the male cell, which is one flagellum of the micro- gametocyte (parent male cell). The macrogametes are large organisms, pale, indistinct, and three or four times as large as the red cell. Some of them show no trace of the corpuscle, their pigment is abundant and exhibits very active movements. Their nucleus is about three and one-half microns in diameter and PLATE XXIX. The Quartan Parasite. I. Normal erythrocyte. 2. Intracellular hyaline form. 3. Young pigmented intracellular form. Note the coarseness, dark color, and scantiness of the pigment granules. 4, 5,6, 7, Later developmental stages of 3. Note the peripheral distribution of the pigment in all the parasites from 3 to 8. (Compare size and color of the erythrocytes in 5, 6, and 7 with 7, 8, and 9, Plate VI.) 8. Mature intracellular form. Note that the stroma of the erythrocyte is no longer demonstrable. 9, 10, 11. Segmenting forms. In 9 are shown the characteristic radiating lines of pigment. (Compare with io, 11, and 12, Plate VI, and with 10, 1 r, and 12, Plate VIII.) 12. Large swollen extracellular form. (Compare with 13, Plate VI.) 13. Flagellate form. (Compare with 14, Plate VI.) 14. Vacuolation of an extracellular form. (E. F. Faber, fee.} (From Da Costa's "Clinical Hematology.") THE BLOOD. 543 is sometimes seen in the fresh specimen; either its outline is distinct or its size and shape may be recognized, as it is the only portion of the parasite which is not invaded by the pigment granules. The function of these macrogametes seems to be to continue the life of the organism within the mosquito after the organism has become fertilized by the male element or microgamete. The microgametocyte is smaller than the macrogamete, being eight to ten microns in diameter. Its pigment is very active, but soon forms a circle around the center and becomes stationary. Occasionally this pigment may become even more active than before, the margin of the cell may undulate and several flagella protrude. These flagella are the microgametes and are two to three times the length of a red blood-cell and often contain pigment granules which enable them to be followed when they break loose from the parent cell. After these flagella separate, the parent cell is seen as a small cell with central motionless pigment. This process of flagellation is not seen in the fresh specimen, but occurs 15 or 20 minutes after the blood has been drawn, which would point to the fact that such a process does not occur within the body. Normally, this change takes place only in the stomach of the mosquito. (b). The Quartan Organism (Hemameba malariae). This organism is much more rare than is the tertian. Its cycle of develop- ment requires 72 hours so that the normal paroxysm occurs every fourth day. If two groups are causing the infection there will be two days with paroxysms, one free day, and then two days of paroxysm following. If more than two groups are introduced we may have daily chills and fever, but only when the groups are large enough in number to cause a paroxysm. According to Ross, 250,000,000 organisms are necessary before a chill follows. The small hyaline forms of the quartan parasite are not distinguishable in the early stages from those of the tertian type, but are easily recognized at the time pigment appears, as the granules of the former are coarser, darker in color, and not so actively motile. As the parasite grows in size the corpuscle becomes smaller and stunted with an irregular crenated margin. The protoplasm of the organism is more refractile than that of the tertian organism and hence the outlines of the pseudo- podia are more easily seen, although the parasite is less actively motile. In 24 hours the red cell is quite small, crenated, and distinctly brassy in color. The organism is round or oval, quite distinct, slightly ameboid, and its pigment blackish-brown in color and gathered at the periphery, especially on one side, thus differing from the tertian organism in which the pigment is scattered throughout the organism. The pigment granules have practically no motion at this stage of the development. As development proceeds the parasite fills from one-third to one-half of the cell, becomes rounder, and loses its ameboid power. The protoplasm is very distinct and highly refractile. During the third day only a rim of the cell is left and this usually takes on a dark, brassy tone. The organism is at this time full grown and is about seven microns in diameter. The pigment now passes from the periphery of the organism toward the center in definite radial lines, giving a wheel-like formation, with the 544 DIAGNOSTIC METHODS. pigment granules forming the spokes. Later the pigment collects in the center and we have the formation of the presegmenter form. Following this the organism becomes opaque, refractive dots appear in a single regular circle about the periphery, and crenations of the border appear with these dots as a center. Lines of division start from these crenations and run to the center, forming from six to twelve segments, like the petals of a flower, giving rise to the name "daisy," "marguerite," or "rosette" form. These quartan segments are much more perfect than are the tertian forms, and later separate to form the hyaline types which take up the development as outlined above. This whole cycle of the development of the quartan organism takes place in the peripheral blood as it does in the tertian organism, but the number of segmenter forms is much more numerous in the quartan type than it is in the tertian form. This is probably due to the fact that a large number of the tertian forms accumu- late in the internal organs. The gamete forms are not as frequently seen as are those of the tertian organism. They are similar in appearance, but somewhat smaller than those of the tertian organism and give rise to flagellation in the same manner. The extracellular forms are occasionally found, but not so frequently as those of the tertian form. The distinguishing marks between these two types of organisms may be summarized as follows: The cycle of development of the tertian organism is 48 hours, while that of the quartan is 72. The quartan organism is smaller, more refractile, less ameboid, and its pigment is coarser, darker, less motile, and more peripheral in position. The corpuscle infected by the quartan organism is smaller, shrunken, crenated, and more brassy. The presegmenter and segmenter stage are much more distinctive in the quartan than in the tertian type and more of the segment forms of the former are found in the peripheral blood, although the number of segments of the quartan type are less than those of the tertian parasite (Emerson). (c). The Estivo-autumnal Parasite (Plasmodium precox; Hema- tozoon falciparium.) This is the most dangerous type of malarial infection. The duration of the cycle of development varies from 24 to 72 hours. In infection with this organism the members of the same group do not always develop in the same unity, so that we may find at times an intermittent fever, but one which becomes more and more continuous. The hyaline forms are similar to those of the tertian and the quartan types, but are slightly smaller and assume the "signet-ring" form much more commonly and maintain it longer. This early form of the estivo-autumnal parasite is distinguishable from the tertian by the shrinkage of the red blood-cell and from the quartan parasite by the smaller dimensions. In some cases the rings do not show the thickening of one segment, but remain of a uniform fine caliber throughout. These rings which do not show the distinct "signet-ring" type nearly always present two nuclear bodies lying at opposite poles or close together. Occasionally these PLATE XXX. The Estivo-Autumnal Parasite. i. Normal erythrocyte. 2, 3. Young hyaline ring-forms. 4, 5, 6. Intracellular hyaline forms. In 4 the parasite appears as an irregularly shaped disc with a thinned-out central area. In 5 and 6 its ameboid properties are obvious. 7. Young pigmented intracellular form. Note the extreme delicacy and small number of the pigment granules. (Compare with 6, Plate VI, and with 3, Plate VII.) 8, 9. Later developmental stages of 7. to, 11, 12. Segmenting forms. 13, 14. Crescentic forms at early stages of their development. 15, 16, 17, 18, 19. Crescentic forms. In 15 and 19 a distinct " bib" of the erythrocyte is visible. Vacuolation of a crescent is shown in 18, and polar arrangement of the pigment in 17. 20. Oval form. 21, 22. Spherical forms. 23. Flagellate form. 24. Vacuolation and deformity of a spherical form. 25. Vacuolated leucocyte apparently enclosing a dwarfed and shrunken crescent. 26. Remains of a shrunken spherical form. (From Da Costa's "Clinical Hematology.") (E. F. Faber, fee.) THE BLOOD. 545 rings appear as if unfolded and stretched across the cell like a thread, the nuclei appearing at irregular intervals. These rings may at times lose their refractility and become ameboid. As the parasite develops, a slight amount of pigment appears, usually seen as one or two granules which are motionless, as a rule, and are located at the periphery of the parasite or at the inner edge of the biconcavity. The cell is very commonly much shrunken, crenated, and brassy, even in the early stages. Some cells, which do not contain parasites, show the same injurious effects of the organism. The parasite at this time occupies about one-fifth of the cell. The infected cells now usually disappear from the circulation and continue their development in the lymph-glands, especially in the spleen. In some cases, however, the parasite does continue its development in the peripheral blood, but this is rare. In such blood or in that obtained from the spleen, the pigment appears much increased and seems to be rather coarse and dark in color, thus resembling very closely the quartan organism at this stage. It seems to be a general rule that the more malignant the type of estivo-autumnal malaria, the fewer older forms are seen in the peripheral blood, although numerous young parasites are present. In some cases the hemoglobin becomes concentrated around the parasite, leaving an almost colorless ring at the periphery of the cell. The cycle of development in the internal organs seems to take place within the macrophages, which are best studied in the fresh specimen. The parasite develops to about five microns in size, which is about half the size of the cell, and when full-grown has its pigment all in the center, never diffusely scattered as in the tertian organism or peripherally located as in the quartan type. This form is rarely seen in the peripheral circulation. The segmenters vary in size from 2 1/2 to 5 microns in diameter, the process of segmentation being similar to that of the other organisms giving rise to the formation of 15 or 16 very small segments. Certain characteristic forms of this type of malaria appear in the per- ipheral blood from about the seventh day of infection and in the internal organs as early as the fifth day. These forms are known as the crescents and the avoids. The crescents are slightly longer than the red blood-cells and show a distinctly crescentic shape with rounded ends, although irregular forms are at times observed. They are very refractile and usually show a fringe of the degenerated red blood-cell, which is more abundant in the concavity of the crescent and forms the so-called "bib." The pigment is large in amount and is massed at the center of the crescent, occasionally in the form of a sheaf or ring. The granules are usually coarse and rod-shaped. These crescents very frequently change their shapes, becoming oval, dumbbell-shaped, or circular and then may resume their original crescentic form. In the circular types no trace of the corpuscles is seen and the protoplasm is not as distinctly refractive as is that of the crescent. In this form of malaria we also find pigmented leucocytes, both the polynuclear neutrophiles and the large mononuclears assuming this function. In these phagocytic cells one may see masses of pigment or even parasites, especially the segmenting and flagellating forms. These pigmented cells are also seen in the other form of malaria, but only just after the chill, 546 DIAGNOSTIC METHODS. while in the estivo-autumnal form they may occur at any time during the infection. Examination of Stained Specimens The technic of making preparations of malarial blood for staining is practically the same as that outlined previously. Precaution must be taken to make thin smears so that the parasite may be brought out more clearly. The stains to be used will depend largely on the experience of the worker, but the writer would recommend the thionin and Nocht stains above the others, although the Wright and Giemsa stains will frequently give beautiful pictures. Stains which have been k^pt for some time are not always reliable, so that it is well to have fresh specimens of the stain on hand for use. If the blood has been kept for some time before staining a diffuse plasma staining with methylene blue will be observed. The Tertian Parasite. The young hyaline form consists of a mass of blue protoplasm usually grouped in ring-form with a mass of reddish-violet stained chromatin, usually situated at the thinner portion of the ring and extending for a large part within the clear achromatic or vesicular part of the parasite. These hyalines are two to three microns in diameter. There is some discussion as to what portion of the parasite the name nucleus should be applied. Some give this term only to the chromatin staining part, while others include both the chromatin and achromatic portion. However this may be, it is necessary for the recognition of the tertian organism that the blue protoplasm and the red chromatin be both observed. There are frequently artefacts in the blood which resemble very closely these hyaline forms so that the worker must be constantly on his guard. Such artefacts are the Maragliano degenerations so commonly seen in red blood- cells and also the cases in which blood-platelets lie upon a red blood-cell. Any structure which lies upon a red blood-corpuscle appears surrounded by a colorless zone, while the true malarial ring is in direct contact with the hemoglobin of the cell. Moreover, such artefacts will not show the chromatin staining portion which is so characteristic of the hyaline ring. In the specimens examined at the end of 24 hours, one will observe that the achromatic area has become somewhat larger, while the chromatin portion seems to be grouped in more irregular masses, some cells appearing to have several nuclei. In the full- grown parasite the chromatin breaks up into a cluster of fine granules which are scattered diffusely through the cell in the form of strands and masses. These chromatin clumps separate into from 15 to 20 dense round masses, around which the protoplasm collects with them as a center. The protoplasm at this stage is distinctly achromatic and is always so in the segmenting cell. The distinct, achromatic, milky zone surrounds each segmentary chromatin clump, while the general protoplasm shows a diffuse faintly basic staining. The pigment which must not be confused with the chromatin is pushed toward the periphery and, after segmentation is complete, collects in masses near the center. It is to be recalled that at the time the pigment collects in the center PLATE XXXI. Tertian Malarial Parasite. (Wright's Stain.) THE BLOOD. 547 in the fresh specimen there is no distinct evidence of segmentation, although this segmentation shows quite distinctly in the stained specimen. The sexual development of the parasite is easily followed in some cases in the stained specimen. According to Stephens and Christophers the young gamete is characterized by the position of the chromatin as it lies in the center of the vacuole instead of at the edge. During the development the gamete is occasionally filled with basophile particles which are known as Plehn's karyo- chromatophilic granules or Schiigner's granules. The full-grown macrogamete contains an abundance of protoplasm which stains a deep blue and a small amount of chromatin in a compact mass, which is peripherally placed and surrounded by a thin vacuole-like area. The pigment of these female cells is uniformly distributed throughout the cell and the inclosing red blood-cell can be seen only with difficulty. The chromatin is much more voluminous in the microgametocytes, but is looser and centrally placed in a large achromatic zone arranged in the form of a band which stretches clear across the cell. The protoplasm is in the form of a ring around the nucleus and stains more of a grayish-green color than does the bluish protoplasm of the macrogamete. The Quartan Parasite. The structure of the quartan parasite resembles very closely that of the tertian form, but in the hyaline type the chromatin mass is less distinct and is in the form of an irregular clump of granules in the older forms, while in the younger a cluster of fine granules without any distinct achromatic zone is seen. As development proceeds the parasite generally takes a form extending across the cell and usually occupies the larger portion of the red cell which has become shrunken and irregular in shape. The segment forms are much more distinct in the quartan type than in the tertian and show much more regular and geometric lines of cleavage with the chromatin exactly in the center of the cre- nated surface. The pigment granules are coarser and much more distinct than in the tertian form and are more peripherally located. The Estivo-autumnal Parasite. The hyaline forms in this type show the chromatin in two or more masses or filaments. The protoplasm is scantier than in the other forms and remains so throughout the cycle of development of this parasite. A very characteristic appearance of the hyaline rings of this type seems to be the thickening of the protoplasmic layer opposite the chromatin mass. The gamete forms are distinctly spherical, being of the same thickness all the way round. Their nucleus forms a portion of the ring, but does not project as in the schizonts (the asexual parasites). The red blood-cells in these sexual types usually show no coarse granular stippling. The crescent forms, which are characteristic of this type of malaria, show the chromatin in a loose network which occupies the larger portion of the cell, has little blue staining protoplasm, and has its pigment scattered throughout its body. This is the male form and is somewhat kidney-shaped and is shorter and broader than the female type. The female 548 DIAGNOSTIC METHODS. crescent is longer and narrower, its chromatin more or less compact and centrally located, its pigment in a ring around the nucleus or in a clump at the center, while its protoplasm is more or less extensive and takes a distinct bluish tinge. We also find two types of the circular form. The microga- metocyte is smaller than the red cell, distinctly spherical in shape, with its chro- matin in the center in a large irregular mass, or in several dense masses near the periphery. These masses containing chromatin material are later extruded and form the flagella or microgametes. The macrogamete is two or three times the size of the microgametocyte, is often of a triangular shape, and has abundant blue-staining protoplasm. The chromatin is in a single mass at the periphery and is surrounded by a circle of pigment. The examination of the stained specimens does not give as great an opportunity for study of the developmental cycle of these parasites as does the examination of a fresh specimen. The conditions found in the fresh blood resemble more nearly those found in the stomach of the mosquito than in the circulating blood, so that many pictures seen in the fresh specimen are practically never found in the stained slide. It is further to be said that we do not always find malarial organisms either in the fresh or stained specimen, although the patient may be at the time suffering from malaria. It may be stated, as a rule, that in all well-marked initial attacks of malarial fever the parasite may be found in the blood if it be examined within 18 hours of the chill. The energetic use of quinin has so much influence upon the ameboid types of the parasites, that the blood may fail to show any of these organisms, although the patient may die from the effect of the infection. It is safe, how- ever, to state that there is practically no case of malaria in which parasites may not be found in the blood, if frequent and repeated examinations are made. It is, therefore, absolutely necessary to make many examinations of the blood, both in the fresh and stained state, before a negative diagnosis can be sustained. Development of the Organism Within the Mosquito (Sporogony). The cycle of development of the malarial organism has been more closely followed in the mosquito in the case of the estivo-autumnal parasite. For any development to occur within the body of the mosquito it is necessary that the macrogamete become fertilized, so that the course in the mosquito is one of sexual development. The microgametocytes throw out their flagella (microgametes) and the macrogamete ripens in the stomach of the mosquito by casting off karyosomes (polar bodies consisting of chromatin), and in so doing causes the formation of a slight mound at one portion of the organism through which the free flagellum enters. This process occurs in from one to one and a half hours after the mosquito has bitten a patient infected with malaria. The nuclear material of the macrogamete and microgamete then unite. The cell then forms a distinct motile spindle shape called the vermiculus or ookinet. The size of this fertilized macrogamete is from 20 microns up and may be found in about 48 hours after the blood has been ingested. T.his motile form is found PLATE XXXII. Estivo-autumnal Parasite. (Wright's Stain.) THE BLOOD. 550 Fig. 144.-Cycles of the Malarial Parasite. (Deguy et Guillaumin.) I, Beginning of the development of the parasite as the ameboid intra-cellular body. I, II, III, and IV represent the endoglobular cycle or schizogony. I, II, II', II", III", and III' represent the extra-cellular cycle or sporogony. In II', there are two free gametes (g), one microgametocyte (m), one microgamete (m'), and the union of a macrogamete and a microgamete (w"). HI', free gametes; III", fertilized macrogamete, taken up by the mosquito, A. In B, C, D and E it becomes encysted in the gastric musculature forming the zygotes (z and z'). In E, the sporozoits are formed and in F they are thrown out by the saliva of the mosquito. In G, these are free to enter the cell forming the ameboid bodies. 550 DIAGNOSTIC METHODS. only in the stomach of the mosquito. The vermiculus then bores its way through the epithelial cells of the intestinal wall and becomes encysted between the intestinal epithelium and the elastic layer, which forms the membrane of the spore cyst (zygote, oocyst, sporoblast). This zygote increases rapidly in size and the nucleus divides rapidly. In its growth it bulges outward from the intestinal wall forming pendulous tumors into the body cavity which growth may vary from four and a half to ninety microns in diameter. This stage is associated with the appearance of much pigment and is called the medium zygote or medium sporoblast stage. The protoplasm gathers around the divided nuclei, forming daughter cysts which are connected by bridges of protoplasm forming the stage known as large zygote, large sporoblast, or large oocyst. In each of these divisions the nucleus divides many times, the daughter nuclei remaining on the surface of each daughter cyst. The protoplasm now collects around each daughter nucleus, the first forming spherical cells which then elongate into threads lying parallel over the remains of the sporoblasts. These threads are called sporozoits and have an elongated nucleus. The final length of these sporozoits is about fourteen microns and their width about one. Their protoplasm is thick, homogeneous, and very refractive. They are sometimes present to the number of 10,000 in some zygotes, but more frequently are not so numerous. As the oocyst becomes larger it bursts into the body cavity, the sporozoits of each cyst ripening at about the same time. These sporozoits "wander at first free, but soon collect in the salivary gland of the mosquito. They are motile and move with a bending and gliding motion. When they are inoculated into the blood of man by the bite of the female mosquito they attach themselves to the red blood-corpuscle and finally penetrate it to form the initial hyaline type of the organism. The period of incubation after the bite of the mosquito is usually between the eighth and twelfth day, when the first chill will appear, although the exact time of appearance of the initial symptoms will depend upon the number of sporozoits introduced into the circulation (Emerson). The anopheles is the only type of mosquito which is at present known to be the host of the malarial organism and to give rise to the development of the gameto-schizonts (the sexual cells), which the bite of the female animal introduces into the blood cells which are known in their future development as schizogones. The sexual cycle within the mos- quito is known as the sporogone, which Rowley-Lawson1 has recently shown may occur in the circulating blood of the human host. General Changes in the Blood in Malaria. There are few conditions which lead so rapidly to such an extreme reduction in the red cells as does acute malaria.2 An acute attack may reduce the red cells to as low as 500,000 cells as reported by Kelsch. Frequently a reduction of 1,000,000 is observed during the first day, with a progressive reduction as the time goes on. In the afebrile period of the disease a continuous fall 1 Jour, of Exp. Med., vol. 13, 1911, p. 263. Ibid., vol. 17, 1913, p. 324. 2 Rowley-Lawson (Arch, of Int. Med., vol. 9, 1912, p. 420) believes that this severe anemia is due to the migration of the plasmodia from corpuscle to corpuscle. THE BLOOD. 551 is observed, but this is much less rapid. The regeneration of the cells is very active so that an increase in the number of cells has been observed directly after an attack in some cases. In cases of chronic malaria the red cells are commonly reduced to as low as 583,000 (Kelsch), while when attacks occur only at intervals and are promptly stopped by quinin no reduction in the red cells may follow (Marchiafava). In cases of moderate severity the usual changes of secondary anemia are present in the red cells. Polychromatophilia and granular degeneration of the reds progresses steadily, while the hemo- globin content of the cells may be markedly reduced. Frequently cases are seen in which the anemia takes on the absolute pernicious type, so that the parasites seem to have been massed in the bone-marrow. As Ewing states, "there can be no doubt that the tendency of the estivo-autumnal parasite to be massed in the bone-marrow, in both ameboid and crescentic phases, and the excessive demand on red-cell production arising in the disease render pernicious malaria an extremely favorable condition for this disturbance of the structure of the marrow and the development of specific megaloblastic changes." Besides the changes which can be directly referred to anemia or toxemia, changes in the size of the cell are quite constant, the tertian parasite causing from the start swelling of the cell and progressive loss of hemoglobin, while the quartan and estivo-autumnal forms cause the red cell to shrink and take on a peculiar brassy tone. The leucocytes do not show very characteristic changes. In the acute malarial attacks of average severity the absence of leucocytosis is of considerable corroborative value, although a slight leucocytosis amounting to about 10,000 with an increase in the percentage of polynuclear cells has been observed by Billings and others. Except during the three or four hours immediately following a chill malarial blood usually shows a diminished number of leucocytes with a distinct relative lymphocytosis, which finding is that seen in typhoid fever. In the more severe estivo-autumnal attacks a definite leucocytosis has been distinctly observed, especially in the hemoglobinuric or black-water type of malarial infection. The extent of the leucocytosis varies between 10,000 and 35,000 although many attacks fail to cause any distinct increase. During the afebrile periods the eosinophile cells are usually increased and may be observed throughout the course of the attack. Neutrophile myelocytes are occasionally present and rarely eosinophile myelocytes. Pigmented leucocytes are seen in the majority of cases, especially in the severe and fatal cases, the pigmented leucocytes being more closely related to the severity of the paroxysms than to the extent of the deposits in the various viscera. These pigmented or phagocytic cells include mononuclear and polynuclear leucocytes and a few endothelial cells. The large and small mononuclears usually contain pigment or rosettes, while many of the polynuclear leucocytes also contain the parasites. These phagocytes may contain, besides parasites and malarial pigment, hematoidin, hemosiderin, red blood-cells, leucocytes, and occasionally an unknown crys- talline pigment. 552 DIAGNOSTIC METHODS. (2). Relapsing Fever (Famine Fever). The cause of this peculiar fever is the spirillum of Obermeier and is not a member of the class of bacteria but, more properly speaking, belongs to the class of spirochete. This organism is between 16 and 40 microns in length and about one micron in width, but is subject to considerable variation in size. It is thin, sharply curved, and appears to be structureless. It takes a deep chromatin stain and also stains with methylene blue in from two to five minutes. It is seen in the blood only during the febrile period of the disease and at that time is actively motile with a rapid wavy motion, much resembling the move- ments of a coiled spring in its stretching and collapsing. It moves rather slowly among the corpuscles, but does not disturb them to any extent. Cases have been reported in which these spirochete are present in the blood 24 hours Fig. 145.-Spirillum of Obermeier. (Pitfield.') before the chill, but they are usually to be found in larger numbers at the time of the rise in temperature, increasing rapidly from day to day. The fever, as a rule, continues about six days, at the end of which time these parasites leave the blood. Strangely enough these organisms have been found in varying numbers in different parts of the circulation, while there does not seem to be any strict parallelism between their number and the height of the fever. Loewenthal has applied the agglutination test to the blood of suspected cases and finds the reaction positive in 85 per cent, of the cases in the periods in which the parasites are absent. The changes in the blood are not characteristic in this condition. The red blood cells seem to be diminished to a slight extent for several days after the attack, but increased during the afebrile period. The hemoglobin may be reduced to as low as 50 per cent, so that a very distinct anemia may be present. The leucocytes seem to be distinctly increased in this disease, the most marked leucocytosis occurring just after the crisis. Cases of relapsing fever are practically never found in the United States, unless imported through the medium of emigrants from Russia and especially India. The cases reported by Wellman show that relapsing fever, as found THE BLOOD. 553 in West Africa, may arise from the bite of a tick transmitting the spirillum of Obermeier. (3). Sleeping Sickness. This very interesting condition which is so prevalent in Central and West Africa seems to be due to an actively motile fusiform flagellate known as the trypanosoma Gambiense, which can be found in the blood free in the plasma (never intracorpuscularly), moving with a screw-like motion among the red cells which it does not seem to disturb. This parasite doubtless has a sexual development, its host being the common fly, Glossina palpalis, while closely related trypanosomata are transmitted by the bites of various flies, especially one of the seven varieties of the tsetse fly. This organism is from two to three times as long as a red blood-corpuscle (18 to 25 microns) and 2 to 21/2 microns wide, having a flagellum anteriorly and an undulating membrane extending its entire length.1 In the fresh blood specimen these parasites should be looked for with only a medium magnification. These parasites vary much in number sometimes being absent for a long period and then suddenly reappearing in large numbers. Symptoms of the disease seem to bear little relation to the number of parasites in the peripheral blood, so that in some cases it may be necessary to examine the fluid in the edematous areas or even to puncture the cervical lymph-glands.2 When these parasites are stained with a polychrome dye they show a rather large red nucleus about the middle, a centrosome staining intensely in a vacuole-like area near the blunt posterior end, and a line of chromatin taking a dense red stain running down the edge of the undulating membrane and terminating in the flagellum which is also stained red. The protoplasm of the body takes a distinct blue stain. Fig. 146.-Trypanosoma gambiense. (Da Costa.) 1 See Thomson and Sinton, Ann. of Trop. Med. and Parasitol., vol. 6, 1912, p. 331. 2 Wolbach and Binger (Jour, of Med. Res., vol. 27, 1912, p. 83) show that the trypano- osmes are not confined to the blood-vessels and lymphatics, but invade the tissues. 554 DIAGNOSTIC METHODS. The parasite contains no pigment and, therefore, obtains its nourishment from the plasma and not from the red cell. This disease may take an acute course, but as a rule is exceedingly chronic, running for years, but becoming fatal as soon as the parasite reaches the cerebro- spinal fluid. The true sleeping sickness appears only when the cerebrospinal fluid is invaded and seems to be, according to the recent work of Koch, directly amenable to treatment with atoxyl. In examining the cerebrospinal fluid for these parasites it is best gently to centrifuge the fluid for five minutes, after which the sediment may be examined under a vaselined cover-glass. There are many other types of trypanosomata, but the Gambiense form is the more important. This is pathogenic toward man, but cannot be distin- guished from the trypanosoma of the tsetse fly which is so fatal to the horse and mule (trypanosoma Brucei), thatof the surra disease (trypanosomaEvansi), or that of dourine (trypanosoma Equiperdum). (4). Kala-azar. Through the researches of Donovan, Leishman, and Ross, parasites have been demonstrated in the blood which are probably directly associated with the condition known as kala-azar, tropical splenomegaly, piroplasmosis, cachexial fever, and dum-dum fever. The organism has been called the Leishman-Donovan body, and is a small oval, round, or oat-shaped body from 2 1/2 to 3 microns in diameter. These bodies have a definite cell outline and contain two chromatin masses, a larger one, a nucleus which is almost round or oval and stains faintly, and a smaller bacillus-shaped centrosome which stains deeply and is directed almost at right angles to the axis of the nucleus. These two chromatin masses are both in the long axis of the cell, the larger one being at the periphery. Many of these forms are vacuolated and the outline of the cell cannot always be seen, although these two masses thus arranged are distinctive. They are easily stained with the various polychrome dyes and are best studied with the highest lenses. These bodies probably represent a stage in the development of a trypanosome as shown by the work of Leishmann and Statham. They are not found in the circulating blood as a rule, but they have occasionally been reported in the form of intracellular bodies in fatal cases. They are easily seen in the blood obtained by splenic puncture and also in the granulation tissue taken from the ulcers. Many are found in the mesenteric lymph-glands, bone-marrow, and liver. Some of these bodies lie free, but most of them are intracellular, either in the leucocytes, endothelial, or splenic cells, and frequently in large masses in the macrophages.1 The changes in the blood are those of a moderate anemia, associated with a leucopenia with a relative and absolute increase in the number of the large mononuclears. The average leucocyte count is about 2,000. 'Darling (Jour, of Exp. Med., vol. n, 1909, p. 515) has reported the finding of an intracellular parasite, the histoplasma capsulatum, as the cause of Histoplasmosis, an infectious disease of the Canal Zone much resembling Kala-Azar. The parasite is small, round or oval, 1 to 4 n in diameter, possesses a polymorphous chromatin nucleus, basophilic cytoplasm and achromatic spaces all enclosed within an achromatic refractile capsule. THE BLOOD. 555 (5). Filariasis. This is a condition associated with the presence of filariae in the blood {filaria sanguinis honiinis). While many of these filariae1 are known, the most common one is the filaria Bancrofti (filaria nocturna). These are from 270 to 340 microns (0.2 to 0.3 mm.) long and from 7 to n microns broad. They are enclosed in a sheath which is considerably longer than is the parasite and shows fine cross striations. The anterior end of the parasite is abruptly rounded and has a six-tipped prepuce and a sharp fang, while the posterior end tapers for about two-fifths of the length of the parasite. The median axis of the parasite is granular. The movement of these parasites is distinctly progressive at first as seen under the microscope, but they soon become motion- less, appearing to attach themselves to the glass slide at their anterior end. Strangely enough these embryos appear in the circulation only toward evening, their numbers gradually rising to a maximum about midnight and diminishing Fig. 147.-Filaria bancrofti. {Da Costa.) toward dawn. During the day they are found in the internal organs, especially the lungs. The forms appearing in the blood are practically all embryos, as the adult types lie in the lymphatics where they obstruct the lymph flow. The obstruction in the lymph-glands may also be brought about by the eggs, which are 25 to 38 microns long by 15 broad. The embryos reach the general circulation only through the thoracic duct. The female filaria is 85 to 150 mm. long, with a distinct neck, a head with a simple, minute, terminal mouth, and a plain cylindrical body covered by a striated cuticle and tapering toward the neck and tail. The tail ends bluntly and has a small depression surrounded by two lips. The male is about 80 mm. long, without a neck and having a tendril-like tail rolled into one or more spirals. 2 Wellman and Johns (Jour. A. M. A., vol. 59, 1912, p. 1531) have succeeded in culti- vating the filaria immitis on artificial media. 556 DIAGNOSTIC METHODS. Like the malarial organisms, the filaria has an intermediate host in the mosquito, both of the culex and anopheles variety. The embryos, which are taken up by the bite of the mosquito, cast off their sheath in about one hour in the stomach of a mosquito. Some of these embryos die at this stage, but others bore actively through the intestinal wall to the muscle, where they remain. In the next two or three days the embryo becomes larger and its alimentary tract develops. On the seventh day the worm is about 1 1/2 mm. long and is perfectly developed. It now travels toward the head and takes its position in the labium, whence it enters the blood of its new host during the biting by the insect. A large number of these adult forms is neces- sary to cause very severe cases and many years may pass before any symptoms are manifest. In examining the blood for the filaria, it is best to take a specimen late at night and to make a very thick, fresh specimen which should be examined with a low power. Besides the ordinary anemia which may develop in such cases, we find a very striking eosinophilia which may run from 4 to 17 per cent. A very characteristic finding in such cases is the condition of hematochyluria followed by chyluria. This hematochyluria seems to be due to rupture of the varicose lymph-vessels of the bladder, as these form a large part of the collateral circulation when the thoracic duct is occluded (Emerson). Such attacks may occur for years and be separated by long intervals. Their onset is spontaneous or following exertion and is usually associated with pain and fever. The urine shows the presence of blood, chyle (as high as 3.8 per cent, fat), and embryos. Many other forms of filariae are known, but this Bancroft type seems to be the more important. While this disease occurs endemically in the tropics there are undoubtedly many cases in this country. It is, therefore, wise in a case showing lymph tumor, elephantiasis, and hematochyluria, especially when pain and fever and enlarged spleen are present, to examine the blood for the filaria Bancrofti. (6). Syphilis. The search for the causative factor of syphilis has extended over a period of many years and various agents have been advanced from time to time, but none of them has remained fixed as the distinct etiologic factor. Through the work of Schaudinn and Hoffmann in 1905, a parasite has been found which seems to be so constantly associated w'ith syphilitic lesions, whether primary, secondary, or tertiary, that it is highly probable that the causative unit has been found. It is true that in many conditions which at present bear no known relation to syphilis, such as pseudoleukemia, similar parasites are found in the enlarged glands, but whether or not these are the true spirochaete pallidae remains to be seen. Zeit has very exactly summed up the points which must be met before the Spirochaeta pallida is definitely accepted as the etiologic factor of syphilis, as follows: "If we are to consider the Spirochaeta pallida as the unquestionable THE BLOOD. 557 etiologic factor of syphilis it will be necessary, however, to explain, by further exhaustive investigations, the discrepancies which the cytorrhyctes defenders have been most active to point out and still uphold, of which I will mention the following: The difficulty of finding spirochaetes in cover-glass smears from tissues which are teeming with silver-stained spirochaetes. The morphological similarity of the Pallida type and certain mouth spiro- chaetes which I have named pseudopallida. The practically total absence of spirochaetes in the highly infectious organs of slaughtered syphilitic monkeys, although the organs of the macerated syphi- litic fetus are teeming with them. The Jancke experiment of successful infection with syphilitic virus which was filtered through a Pasteur filter, whereas the sprochaetes do not pass the filter. The fact that pallidae can be kept alive for days and weeks, whereas syphilitic virus loses its virulence in 6 to 8 hours. The destruction of pallidae when kept in glycerin, although the virulence of syphilitic virus can be preserved for days by glycerin. The presence of typical pallidae in the skin of the macerated pig fetus. The only cultures (not pure cultures) which ever were produced of the pallida by Levaditi, grown in coilodium sacs in the peritoneal cavity of monkeys and rabbits, did not produce syphilis in monkeys. By fulfilling Koch's classic postulates in the way of culture, the whole question could be definitely and finally decided with one stroke. Everything speaks for it that the pallida is the etiologic factor of syphilis, and many of the discrepancies can easily be explained by considering spirochaete pallidae as protozoa and not bacteria." The Spirochaeta pallida1 (treponema pallidum) derives its name from its low refractive power and the difficulty with which it takes up anilin dyes. It has a very delicate structure, usually presenting io to 40 deep spiral incurvations in the larger specimens or only a few in the smaller ones. Its length varies between 4 and 10 microns and its width does not exceed 1/2 micron. The organism has been demonstrated in the circulating blood, in the scrapings obtained from the chancre, in the incised papules, in smears from the mucous patches, and in the fluid aspirated from the inguinal glands. It seems to be easily demonstrable in the blood from a splenic puncture, while in the congenital forms it is found in the internal organs and in the peripheral blood. A characteristic difference between this Spirochaeta and some other types (spirochaeta buccalis), with which it might be confused, is that its ends lie above and below a longitudinal line drawn through the center of its curvatures, while in the other forms the ends lie on the projection of such a line. The organism moves in an oscillatory manner about its longitudinal axis, its movements being winding bending and whipping, while in the spirilla the longitudinal axis remains 1 See McDonagh, Brit. Jour, of Dermat., vol. 24, 1912, p. 381 and Ross, Lancet, vol. 2, 1912, p. 1105. 558 DIAGNOSTIC METHODS. rigid. Schaudinn demonstrated the existence of a flagellum at each end, while the other spirochaete have an undulating membrane. These organisms are seen only with great difficulty in the specimens of fresh blood, but thanks to the introduction of the ultra-condenser (the dark- field illuminator) we are in a position to see these organisms, both in the splenic and peripheral blood, although considerable practice is necessary properly to adjust the light. These organisms do not take anilin dyes readily, so that special methods have been advanced for their demonstration in smears. A very good stain for them is the Goldhorn stain. The smears are fixed with pure Fig. 148.-Spirochaste pallidas and refringens. {Pitfield.) The darker ones are the refringens. methyl alcohol for 15 minutes and are then covered with the stain (polychrome methylene blue) for three to five seconds, when the excess is drained off. The specimens are then slowly introduced into clean water with the film sides down. Keep the slide in this position for four to five seconds and then shake in the water to remove the excess of the dye. The spirochsete appear of a violet color. This violet tint may be changed to a bluish-black by covering the specimen with Gram's iodin solution for 15 to 20 seconds, after w'hich it is washed and dried as usual and the examination made wdth the immersion lens. The writer has also found the use of the Giemsa stain very reliable, especially when the staining is continued for 18 hours (see Exudates). Other stains, such as that of Levaditi, have been advocated, but they do not seem to THE BLOOD. 559 give any better results and are more complicated. For staining the spirochaete in tissues the Levaditi stain is admirable. The examination of the blood is very often disappointing, owing to the fact that few spirochaete may be present in the specimen. Better results are obtained by examination of specimens from a curettage which has been carried sufficiently far to allow serum to appear. This serous fluid is then spread upon slides and treated in the usual manner (see Exudates). Fig. 149.-Ultra-condenser of Reichert. Cultivation of Treponema Pallidum As previously stated, the causative factor of syphilis can be definitely settled only by fulfilling Koch's postulates regarding pure cultures and the production of the disease by means of these pure cultures. Schereschewsky,1 Muhlens2 and Hoffmann3 were able to cultivate the pallidum but were unable to reproduce syphilitic lesions by means of their cultures. Bruckner and Galasesco4 and Sowada5 reported the successful reproduction of the lesions by injection of their "young impure cultures" but, as neither investigator was able to grow a second generation of these so-called cultures in any medium, their results are questionable. Recently Noguchi6 has succeeded not only in producing many gen- erations of pure cultures of the pallidum but, also, in reproducing the syphilitic lesions by use of these pure cultures. His work is of the greatest importance as it absolutely establishes the Spirochaeta pallida as the etiologic factor of syphilis. 1 Deutsch, med. Woch., Bd. 35, 1909, S. 835, 1260 and 1652. 2 Deutsch, med. Woch., Bd. 35, 1909, S. 1261; Klin. Jahrb., Bd. 23, 1910, S. 339. 3 Zeitsch. f. Hyg. u. Infectkr., Bd. 68, 1911, S. 27. 4 Comp. rend. soc. de biol., Tm. 68, 1910, p. 684. 5 Deutsch, med. Woch., Bd. 37, 19x1, S. 682. 6 Jour, of Exp. Med., vol. 14, 1911, p. 99; Ibid., vol. 15, 1912, p. 90; Ibid., vol. 15, 1912, p. 201; Ibid., vol. 16, 1912, p. 211. 560 DIAGNOSTIC METHODS. His technic is as follows: As the material for obtaining the cultures he uses the spirochactae-containing testicular tissue of rabbits, which have been inoculated with human syphilitic material. From this first generation in rabbits any number of generations may be derived by transmitting the strain, at appro- priate intervals (four to six weeks), from rabbit to rabbit. By this means the Spirochaetae are almost free from banal bacteria. The culture media is the following: 16 c.c. of serum-water (3 parts dis- tilled water and 1 part serum of sheep, horse or rabbit) are placed in tubes 20 cm. long and 1.5 cm. wide. These tubes are then sterilized by fractional sterilization at 100 C. for 15 minutes on each of three successive days. Then a small piece of freshly removed sterile rabbit tissue (preferably kidney or testicle, although heart muscle may be used but not liver) is placed in each tube which is then incubated at 370 C. for two days and examined for sterility. To each tube a layer of sterile paraffin oil is added to shield the medium from contact with air and to prevent evaporation. The method of cultivation must be strictly anaerobic in obtaining the first generation of pallida. Noguchi employs a combination of hydrogen gas, vacuum and pyrogallic acid in an anaerobic apparatus, which is fully described in his article. He believes the following conditions essential in obtaining his first generation: (1) the presence of suit- able fresh sterile tissue in serum water, (2) strict anaerobiosis, (3) a slightly alkaline reaction as furnished by the serum and tissue, and (4) a temperature of about 35 to 370 C. When once adapted to the artificial serum-water tissue medium, the pallidum grows well under less strictly anaerobic conditions. In fact, certain strains even grow well in a medium containing agar, provided suitable fresh tissue is placed low down in a high column of medium (serum water or serum agar), and is covered with paraffin oil. Out of ten strains of treponema pallidum, six have been cultivated in the above medium. The difficulty in this work seems to be to obtain the first generation. The culture may be purified by per- mitting it to grow through a Berkefeld filter, which the pallida pass about the fifth day. Certain strains grow together with bacteria along the stab canal in a serum-agar tissue medium. But while the bacteria do not grow out into the surrounding medium, the pallida grow out gradually in a pure state. The characteristics of this pure culture may be obtained from the literature. The morphology and motility are quite typical, difficulty being experienced in distinguishing these points from those of specimens taken from human lesions. These pure cultures produce typical lesions when injected into animals, thus completing the chain of evidence.1 1 In his later work Noguchi has cultivated the organism directly from the lesions in man, using a high cylindrical layer of solid media consisting of two parts of 2 per cent, slightly alkaline agar and one part of ascitic or hydrocele fluid, at the bottom of which has been placed a fragment of sterile tissue. The small pieces of tissue taken from the lesion are inserted deep into this media and the growth carried on as outlined above. See, also, Baeslack, Jour, of Inf. Dis., vol. 12, 1913, p. 55. Noguchi has, also, cultivated various types of spiro- chetse, which closely resemble the pallida, and has thus shown their morphological variations (Jour, of Exp. Med., vol. 15, 1912, p. 466; Ibid., vol. 16, 1912, pp. 194, 199, 261 and 620; Ibid., vol. 17, 1913, p. 89). THE BLOOD. 561 (7). Yellow Fever. This infectious noncontagious febrile disease is caused by a specific organism not yet discovered. This etiologic factor is, however, blood-borne, as shown by the fact that injection of blood from a yellow-fever patient into a healthy subject causes the disease. Whatever the organism be it must be very minute, as the blood serum retains its infecting power after passage through a Berkefeld filter. The various organisms which have been held responsible for this disease, among them the Bacillus X of Sternberg and the Bacillus icteroides of Sanarelli, have all been abandoned and even the myxococcidium stegomyiae of the Yellow Fever Commission is at present considered as foreign to yellow fever. In 1881 Finlay advanced the hypothesis that yellow fever was transmitted to man only through the bite of a mosquito of the Culex group, the stegomyia fasciata. The United States Commission, consisting of Reed, Carroll, Agra- monte, and Lazear, furnished the experimental proof that this hypothesis was valid and showed that the unknown organism of the disease required a period of 12 days' development in the body of the mosquito before it could be trans- mitted from the stegomyia as an infecting agent. A second U. S. Commission, consisting of Parker, Pothier, and Beyer with the help of Smith stated in 1903 that yellow fever was due to a parasite of the sporozoan type, the myxococ- cidium stegomyiae, which developed in the stegomyia. This organism has never been found in the human body, hence its schizogony (asexual development) is unknown. The French Commission, Marchoux, Salimbeni, and Simond, as well as Schaudinn and Carroll, do not believe that this organism has any- thing to do with yellow fever. Seidelin1 has reported the finding of a parasite, a protozoon, in the blood, which he believes the causative factor. As this disease is beyond question blood-borne, its hematological changes are of some interest. Jones shows that anemia is infrequent, that fibrin formation is deficient, that the globucidal action of the serum is greatly increased, and that both cholemia and hemoglobinemia occur. The red cells show little variation in number, Pothier never finding them below 4,280,000. The hemoglobin suffers considerable loss, being usually between 50 and 75 per cent. This loss is restored very slowly in convalescence. Albertoni draws attention to the lack of parallelism between the percentage of hemoglobin and the specific gravity of the blood, the latter falling much more than the former. Morphological changes in the reds are rare. An occasional normoblast may be seen. The leucocytes range between 4,660 and 20,000, the higher the count the more favorable the prognosis. In this leucocytosis the polynuclear neutro- philes are in higher proportion, only rarely being normal. Eosinophiles are few and myelocytes only occasional. 1 Jour, of Path, and Pact., vol. 15, 1911, p. 282 562 DIAGNOSTIC METHODS. (8). Rocky Mountain Spotted Fever (Tick Fever). This disease is not to be confused with typhus fever or epidemic cerebro- spinal meningitis, to both of which the term "spottedfever" has been occasion- ally applied. Wilson and Chowning have reported that the blood of man affected with Rocky Mountain spotted fever shows the presence of an erythrocytic parasite which they call Piroplasma hominis. These parasites are ovoid in form, have ameboid motility, and are unpigmented. Three forms of these intra- cellular ovoids were found, (i) a small, nonmotileform, i to 2 microns in length by 1 micron in width; (2) a larger actively ameboid form, 3 to 5 microns in length by 1 to 1.5 microns in width, and showing a dark granu- lar spot at one end; (3) a twin form, consisting of two pear-shaped bodies, lying with their tapered ends approaching and bearing a granu- lar spot at each end. These bodies stain best with the polychrome dyes. Through the work of Ricketts and King it has been definitely estab- lished that the parasite of spotted fever finds its host in the wood tick (dermacentor venustus (Banks) or, as Stiles states, the dermacentor andersoni). There is apparently no cycle of development in the tick as an intermediate host. Ricketts ad- vances much evidence against the piroplasma as the causative factor, but is unable to say that such might not be the etiologic unit. He1 has recently succeeded in finding an ex- tra- and intra-cellular pleomorphic polar staining bacillus, which is ex- tremely small and is constantly present in the blood of infected animals as well as in the infecting tick. Whether this is the true etiologic factor is still unsettled, but Ricketts believes it is.2 The red cells in this disease are reduced to about 4,000,000, while the hemo- globin content may be as low as 50, thus giving a low color index. Degenerations and atypical staining qualities are rare. The leucocytes are increased to 12,000 or Fig. 150.-Schistosomum hematobium; male with female in gynecophoric groove. {Tyson after Loos.) 1 Jour. A. M. A., vol. 52, 1909, p. 379. 2 See McClintic, Pub. Health Rep., vol. 27, 1912, p. 732, for an excellent bibliography of the subject; also, Rucker, Ibid., vol. 27, 1912, p. 1465. 563 THE BLOOD. more and show nothing differentially abnormal, except a slight increase of the large lymphocytes. (9). Distomiasis (Bilharziasis). This is a chronic parasitic disease due to the deposition in the tissues of the eggs of the worm, Schistosomum hemalobium, also called Bilharzia hematobia, gynecophorus, distomum hematobium, distoma capense, and Thecosoma. It is a very common condition in Africa, but has been found but six times in America, according to O'Neil. Infection appears to be more commonly induced by drink- ing unfiltered infected water, but occasionally it may come through the skin. The adult parasites inhabit the blood of the portal vein and the vessels of the pelvis, rectum, and bladder. The male is smaller and thicker than the female, is 12 to 15 mm. long and 1 mm. broad, is flat and so folded as to form a gynecophoric canal which holds the female. The female is about 20 mm. long and 0.25 mm. thick and is the active agent in the infection. The eggs of the parasite are oval or spindle-shaped, measure about 0.16 mm. in length and 0.05 mm. in breadth, and have a distinct spine-like projection from the posterior end or from one side. These ova are particularly frequent in the urine of such cases. Occasionally they may be found in the circulating blood. VI. Bacteriology of the Blood. From the standpoint of differential diagnosis and treatment of a con- dition known or thought to be of infectious origin, it is frequently of especial importance carefully to examine the blood for the presence of various bacteria. Moreover, in many conditions, as for instance in typhoid fever, cultures made from the circulating blood often give positive findings long before other tests are obtainable. The bacteriological study of the blood requires not only a thorough working-knowledge of the principles underlying the bacteriological technic but, in many cases, special training, also, in the isolation and differentiation of the suspected organism. In this discussion the writer can not go into great detail but must limit himself to more or less general points of clinical value. Technic. The success of this work depends partly upon the obtaining of a suffi- cient quantity of blood and partly upon the selection of the culture-media for the future development of the organisms present. The amount of blood re- quired varies from 5 to 20 c.c., so that it must be obtained by venous puncture. The vein usually selected is the median basilic or the median cephalic at the bend of the elbow, but, if these be not available, the veins of the dorsum of the hand may be used, although the latter are less satisfactory. The site of puncture should be carefully cleansed, using all the precautions taken for any surgical procedure. As a rule, contaminations from the skin are rare providing this cleansing has been adequate. In some cases, especially in very obese patients, it is necessary to expose the vein by incision of the skin and subcutaneous fat over the vein. The instrument best adapted for venous puncture is, in the writer's opinion, 564 DIAGNOSTIC METHODS. the Liier syringe, which is made entirely of glass and has a tightly-fitting platinum needle. Sterilization is best done by the use of dry heat to 1500 C. for one hour, boiling in 1 per cent, sodium carbonate solution for 45 minutes, or using steam under pressure in an autoclave for 15 minutes. These instruments should never be sterilized with carbolic acid or bichlorid solution owing to the danger of inhibiting the growth of the suspected organisms. It is well, also, to sterilize a pair of forceps to use in fitting the needle on the syringe. The properly sterilized outfits should be kept in long test-tubes, plugged with cotton, the point of the needle being protected by resting upon a pad of sterilized cotton. When the above precautions have been taken, an elastic band or a towel may be fastened about the arm above the point of puncture, to produce con- striction of the vessels and distention at the point of puncture. The skin may be anesthetized with a spray of ethyl chlorid, if the patient be very susceptible to the slight pain caused by the puncture. The needle is plunged into the vein against the direction of the blood current, care being taken that the point of the needle is sharp as, otherwise, the vein may roll about beneath it. Very slight aspiration is necessary to draw the blood into the syringe, as the blood tends to flow into the instrument through its own force. The usual amount withdrawn is 10 c.c. but occasionally 20 c.c. are preferable, especially if the patient is plethoric and can stand the loss easily. When the desired amount is obtained, the band- age is removed before the needle is withdrawn. After withdrawal from the vein, the needle is quickly removed from the syringe and the blood divided among several tubes containing agar melted and cooled to 400 C., the tip of the syringe being passed through the flame before inoculating each tube. In deciding as to the amount of blood to add to the agar, a general rule is to increase the amount of blood when the feeble-growing organisms, such as the pneumococcus and gonococcus, are suspected. The usual proportions are 2 or 3 of blood to 5 of agar. It is, probably, preferable to add the blood to fluid media, such as bouillon or litmus-milk, using 1 or 2 c.c. of blood to 100 c.c. of media in order to overcome the normal bactericidal effects of the blood.1 The blood and fluid media are then directly incubated at 370 C. In the case of the agar tubes, the blood and media are mixed by shaking as quickly as possible, poured into Petri dishes and placed in the incubator. It is advisable to use both solid and fluid media. Anaerobic cultures are pre- pared as usual, using, preferably, a combination of a vacuum with alkalinized pyrogallic acid solution to insure absence of oxygen. If, after incubation for 24 hours, the plates show only a few surface col- onies, contamination is to be considered; while if the colonies be deep and are observed in several plates, contamination may be ruled out. This must not be interpreted to mean that any colonies will, necessarily, develop within 24 hours as the plates may have to be kept for 5 to 10 days before a negative result may be certain. However, the contaminating organisms usually show up within 1 Hiss and Zinsser (Text-book of Bacteriology, p. 126) recommend the addition of a small piece of calcium carbonate to the liquid media in order to neutralize any free acid formed in case the pneumococcus is suspected. THE BLOOD. 565 24 hours. The fluid media will become cloudy if any organisms are present. From the plate colonies and the fluid media sub-cultures are then made upon different media and every expedient possible used to isolate and identify the offending organism. Usually the infecting organism may be obtained in pure culture as a mixed infection in the blood is very uncommon. The writer must refer elsewhere for the details of this work. Organisms Found in the Blood. 1. Bacillus Typhosus. Instead of transferring the blood, withdrawn from the vein, to bouillon or melted agar, one may add it directly to sterilized ox-bile in the propor- tion of i or 2 c.c. of blood to 5 of the bile. The blood may, if desired, be obtained in this case from a puncture of the ear (the puncture being prefer- ably made through a coating of collodion and the blood allowed to drop directly into the tube of bile). The mixture of bile and blood is allowed to incubate at 370 C. for 12 to 15 hours, when transfers of a few loopsful are made to the Drigalski-Conradi or modified Endo media discussed under Feces. The char- acteristic colonies are then sub-cultured and the agglutination tests made as described in the next section. According to Kolle and Hetsch1 this bacillus of Eberth-Gaffky shows the following characteristics: it is an actively motile bacillus with numerous flagellae, stains wfith the ordinary dyes but is Gram-negative, produces no indol in bouillon (colon bacillus does), produces no gas in glucose-bouillon (colon bacillus does), produces no acid nor any coagulation in litmus-milk (colon bacillus produces much acid, while the bacillus faecalis alcaligenes is a strong alkali-former), produces no change in neutral red-agar (colon bacillus and paratyphoid bacillus B cause fluorescence of the solution and gas-formation, while the bacillus faecalis alcaligenes and the dysentery bacillus produce no effects), produces dew-drcp colonies on litmus-lactose-agar with no change in color of the media (colon bacillus produces marked red coloration), produces acid and coagulation of the casein in litmus-nutrose-glucose solution, while no such effects are observed in litmus-nutrose-lactose solution, shows characteristic agglutination with pure culture of typhoid bacillus. Ordinarily the agglutination tests are all that are necessary after the organism has been grown on the Drigalski-Conradi or other media. How'ever, it is to be remembered that the freshly isolated organisms show less tendency to agglutinate than do the organisms which have been grown for some time or have been transferred through several generations. The importance of blood cultures2 in cases of typhoid fever may be esti- mated by the experimental evidence that 70 to 80 per cent, of cases are positive, oftentimes many days before the agglutination tests are positive. It has been further shown that the organisms tend to diminish in the blood during the 1 Die Experimentelle Bakteriologie. Berlin, iqii. 2 This work is fast becoming an absolute necessity as the employment of the preventive inoculation for typhoid practically destroys the reliability of the Widal test of the serum of those previously inoculated. 566 DIAGNOSTIC METHODS. course of the disease until they practically disappear from the blood about the end of the third week. 2. Bacillus Paratyphosus. In certain cases, which clinically closely resemble typhoid fever, although running a somewhat milder course, organisms may be isolated from the blood, which are intermediate between the true bacillus typhosus and the bacillus coli communis. The clinical differentiation of these cases is often difficult, although the sudden onset of vomiting, diarrhea, chill, and fever of paratyphoid are somewhat different from the more gradual onset and step-like increase of temperature in typhoid; while the temperature-curve, the somewhat different early characteristics of the feces, the early appearance of herpes, the usual lack of splenic enlargement, etc., may serve as differential points in favor of para- typhoid. Aside from this type, we find sporadic or epidemic cases in which the symptomatology is that of a severe gastroenteritis, the so-called "infectious meat-poisoning" usually classed as ptomaine poisoning, which is due to the bacillus paratyphosus. These organisms are widely spread in the animal king- dom and may be found even as a contamination in water supplies. Not in- frequently it is found in man, as a pure saprophyte, no clinical symptoms trace- able to its presence being observed. It may be present in the pus of otitis media, orchitis, cholecystitis, osteochondritis, arthritis, lymphadenitis, etc., also, as a secondary invader, in scarlet fever, measles, pneumonia, meningitis, etc. The examination of the blood is made as previously described, transfers and sub-cultures being made upon the various media. These paratyphoid bacilli are, apparently, of two types, known as "A" and "B". Both types differ from the bacillus typhosus in that they produce gas in glucose media and show different agglutination reactions. Type A behaves in most other respects like the typhoid bacillus, except that the former causes fluorescence in neutral-red agar. Type B is differentiated from the typhoid bacillus and from type A of the paratyphoid bacillus by not coagulating litmus-milk but in producing acid (red coloration) and, after 5 to io days, in causing a conversion of this red color to a deep blue (due to alkali formation). The typhoid, colon, paratyphoid A and dysentery bacilli never cause a blue coloration of the milk, while the blue color due to the bacillus faecalis alcaligenes is produced at once. The peculiarities of the agglutination reaction with this organism will be discussed in the next section. 3. Bacillus Coli Communis. This organism is, probably, not a unitary organism, as many types of the species exist. The colon bacillus is found in all warm-blooded animals and in many cold-blooded ones as a normal habitant of the intestinal canal. These bacilli show a marked antagonism for the putrefactive bacteria, so that, physio- logically, they should be regarded as conservative organisms. Owing to the wide occurrence of this organism, it is frequently found as a contamination of food and water supplies unless special precautions are taken to exclude it. Although ordinarily a harmless type, yet, from the pathologic stand- THE BLOOD. 567 point, the colon bacillus is generally regarded as a pathogenic organism under certain conditions. In this connection we must distinguish an endogenous from the exogenous colon infections. The first are found in cases of wound infection and, more especially, in infections of the lower urinary tract into which the bacilli have gained entrance through the urethra; while the latter are observed under the following conditions. So long as the natural protective and bactericidal agents of the body are intact and so long as the intestinal epithe- lium is unimpaired, the colon bacilli do not pass into the organs or serous cavities of the body. However, in the course of any disease, such as typhoid, cholera and severe non-specific enteritis, in which the epithelium is destroyed, the colon bacillus may pass from the intestinal tract, either by direct contiguity of tissue or through the blood current, and assume a pathogenic role as a secondary invader. Moreover, should the patient be suffering with a chronic wasting disease which lowers the general resistance to infection, the continuity of the epithelium may still be preserved and the bacillus become pathogenic in many organs. Thus we find infections due to the bacillus coli communis in the kidney, urinary bladder, gall-bladder, peritoneal cavity, ovary, uterus, and other organs. Pus pockets, which may have caused little clinical disturbance, may spread, become chronic and induce a septicemia, of which the etiologic factor may be determined only by resort to blood culture. The biologic properties distinguishing the colon bacillus from the typhoid and paratyphoid bacilli are as follows: The colon bacillus is very little, if at all, motile; it produces gas in glucose media and indol in bouillon or peptone media; it coagulates milk and produces acid with red coloration in litmus milk; its colonies on litmus-lactose agar are deep-red in color (the color diffus- ing into the surrounding media), while those of typhoid and paratyphoid are colorless; in neutral-red agar it causes fluorescence and marked gas pro- duction, while the typhoid bacillus produces no change in the media; it does not show characteristic agglutination reactions, while the typhoid bacillus does. 4. Pneumococcus. This organism may be recovered from the blood in practically all cases of lobar pneumonia due to infection with this type, although it must be remem- bered that other organisms may be the etiologic factor in some cases. At times the pneumococcus may be obtained as early as 12 hours after the initial chill, while in other cases it may appear as late as 48 hours after the crisis. Rosenow obtained positive results in 160 out of 175 cases, Prochaska 48 out of 50, while Cole and Lenhartz report only 30 per cent, of positive findings.1 The finding of the pneumococcus in the blood does not, necessarily, indicate an un- favorable prognosis, although a large percentage, 40 to 50, of the cases in which it is found result fatally.2 The earlier the organisms are found the more grave the prognosis. The number of organisms found by culture is not as large as in other types of blood infection, so that Fraenkel is lead to believe that the prognosis is bad when a large number of colonies develop from 1 c.c. of blood. 1 See Lyall, Jour. A. M. A., vol. 58, 1912, p. 1841. 2 See Hastings and Boehm, Jour, of Exp. Med.; vol. 17, 19x3, p. 239. 568 DIAGNOSTIC METHODS. The infection may spread through the lymphatic channels to the blood, so that we may find these organisms in the bile, urine, milk and other body secretions. Moreover, we may have many infections in other organs as a com- plication of the original pneumonia. Oftentimes, we find the pneumococcus as the cause of primary patho- logical conditions in many organs, when there has been no antecedent pneu- monia. Thus we may have a pneumococcic otitis media, bronchitis, pleuritis, peritonitis, meningitis, endocarditis, pericarditis, orchitis, tonsillitis, cystitis, sal- pingitis, enteritis, conjunctivitis and keratitis. The causative factor may be dis- covered only by examinations of the various excretions related to the special condition. While the blood culture may be of little clinical value in the diagnosis of lobar pneumonia as the clinical findings are usually so clear-cut, yet it may be the only means at our disposal in solving the mystery of an obscure infection in-other parts of the body, as for instance in a malignant endocarditis.1 The blood is taken as previously outlined and is transferred to agar, blood-agar, plain, serum or ascitic bouillon, or litmus-milk. The organisms which are especially confusing from the standpoint of differentiation are the various types of streptococci. On blood-agar plates the pneumococcus shows an especially characteristic greenish hemolytic colony. The streptococcus viridans produces the same green zone, while the streptococcus pyogenes shows no green color but a large colorless hemolytic zone. Blood bouillon shows no hemolysis with either the pneumococcus or streptococcus viridans, while a marked burgundy-red color appears in the presence of streptococcus pyogenes. The differentiation between the pneumococcus and streptococcus viridans is made by the use of bile acids. Bouillon cultures of pneumococcus are cleared while those of the viridans are unaffected. Further, the pneumococcus ferments inulin solutions, while most strains of streptococci do not produce this result. The detection of the capsule and the agglutination reactions clear up the diagnosis. Rosenow's Capsule Stain. This method2 is, in the writer's opinion, the best at our command for this purpose. The technic is as follows: Make a thin smear on a perfectly clean slide or cover-glass. If the material, such as sputum, is too thick, add enough distilled water so that it can be spread evenly by means of a piece of fine tissue or cigarette paper. In cases of cultures (blood-agar, serum, glucose or Loeffler's blood-serum being preferable), remove a small amount of the growth from the surface of the medium and at once mix thoroughly with a loopful of serum on the slide, or, better still, make a rather dense suspension in a few drops of distilled water and then mix an equal quantity of this suspension with serum, and spread by means of tissue paper. As the smear becomes nearly dry cover for ten to twenty seconds with 5 to 10 per cent, aqueous solution of tannic acid; wash in water and blot; stain with carbol (saturated alcoholic solution gentian violet 1 part, 5 per cent, aqueous carbolic acid, 4 parts) or anilin gentian violet 1 See Rosenow, ILL Med. Jour., vol. 21, 1912, p. 425; Jour. Inf. Dis., vol. ir, 1912, p. 94; also, Cole, Jour. A. M. A., vol. 59, 1912, p. 693 for a discussion of immunity in pneumonia. 2 Jour, of Inf. Dis., vol. 9, 1911, p. I. THE BLOOD. 569 half a minute to a minute, heat over flame but do not boil; wash in water again; Gram's iodin solution for one-half to i minute; decolorize in alcohol (95 percent.); stain for 2 to 10 seconds, depending on the thickness of the smears, with saturated alcoholic (60 per cent.) solution of Griibler's eosin; wash in water and blot finally, clear in xylol and mount in balsam or examine directly. If the organism, like the Bacillus mucosus, is Gram-negative, the bacillus may be stained with Loeffler's or aqueous methylene blue. The pneumococci are stained deeply brownish-black, sharply differentiated from the capsule, which is stained pink. Beautiful results are also obtained with the streptococcus mucosus. In the thickest part of the smear the space occupied by the capsule may be perfectly clear; elsewhere in the smear, if properly made, where the conditions are suitable for absorption of eosin, the capsule is stained deeply pink; not rarely a clear refraction zone (often mistaken for the capsule in former methods) may be seen peripherally to a dis- tinctly stained, often large capsule. In case of sputum in which the cocci are embedded in a more or less tenacious mucus, the capsules, at times, are not rendered stainable by the above method. In that case it is well to fix and stain simultaneously with the 2 per cent, aqueous tannic acid, 4 parts, and saturated solution of gentian violet, 1 part. This modification often gives beautiful results. The cocci, however, decolorize easily and the tannic acid-gentian- violet may be followed by carbol-gentian-violet and then the usual procedure. Ordinary carbol-fuchsin, diluted five to ten times, and aqueous eosin (50 per cent, saturated solution) may also be used to stain the capsule although the saturated alcoholic (60 per cent.) eosin has given the best results. Decolorization after the modified Gram procedure of tannic acid-fixed smears is more rapid than in the case of heat-fixed smears, which fact should be borne in mind. By the use of this method it has been determined that the capsule of the pneumococcus and allied organisms is not difficult to preserve or readily soluble in water, as is generally believed. To stain the capsule is a problem of render- ing it stainable rather than one of preservation. The reactions which accomp- lish this appear to be colloidal reactions. 5. Streptococci. In general infections with any of the varieties of the streptococcus, blood cultures reveal the offending organism in a large percentage of cases, de- pending, of course, upon the location and severity of the process. There are many varieties of streptococci, of which some are saprophytic and some pathogenic. Pathologically, streptococci are found in the blood in many con- ditions. They may obtain entrance to the blood by rupture of a streptococcic abscess into a vessel. Infected thrombi, as met with in puerperal sepsis, may form the source of the invasion. Long-standing infections, such as empyemas, may so lower the resistance of the patient that a septicemia may result. An especial point of invasion is through wounds, the infection here often being extremely dangerous. A further type is the cryptogenetic septico-pyemia in which the source of infection can not be traced. 570 DIAGNOSTIC METHODS. The organism may be obtained from the blood in practically all cases of mycotic endocarditis due to this streptococcus. In puerperal sepsis the results are not so frequent, only about one-fourth of the cases being positive. The cryptogenetic types associated with metastatic lesions in the joints and organs show a much larger percentage of positive findings. In scarlet fever we find, according to Hektoen, that streptococci may occasionally be found in the blood in cases which run a short, mild, and uncomplicated course; they occur with relatively greater frequency in the more severe and protracted cases in which there may develop local complications and signs of general infection, although they may not be demonstrable in fatal cases. The streptococci may also be found in diphtheria, measles and small pox especially in the fatal cases, in which they act as secondary invaders. It is to be emphasized that the prog- nosis is not, necessarily, bad when streptococci are found in the blood, although there is a large mortality. The morphology of these organisms is variable and the varieties numerous.1 The pathogenic types, isolated directly from pathologic lesions, appear to con- sist of rather long chains of more than 8 pairs of cocci when grown in bouillon, while the non-pathogenic types form shorter chains. Involution forms are common in specimens from culture. Some of them are encapsulated, two organisms being enclosed. This may lead to confusion with the pneumococcus but the differential points given under the pneumococcus will serve to distinguish them. Usually the microscopic examination of the stained specimen will be all that is necessary, but occasionally careful work is needed to identify them. In doubtful cases, 1/2 to 1 c.c. of the blood withdrawn by venous puncture maybe injected intraperitoneally into a mouse, when a general streptococcic septicemia will ensue. Many other organisms have been and may be isolated from the blood. Thus the staphylococci are not infrequently the cause of a bacteriemia arising from an endocarditis, osteomyelitis, or furunculosis. The gonococcus is often found in cases associated with gonorrheal endocarditis, arthritis, parotitis, myositis, etc. The meningococcus, the influenza bacillus, bacillus mucosus, bacillus of anthrax and of glanders have been obtained from the blood in suitable cases. The tubercle bacillus has been repeatedly found in cases of both acute and chronic tuberculosis. However, much recent work since the publication of Rosenberger, who attempted to prove that tubercle bacilli were present in the circulating blood of incipient tuberculosis and that a primary bacteriemia was always present in these cases, has absolutely refuted this finding and has em- phasized the importance of eliminating the possibility of the presence of acid- fast bacilli in the reagents and, even, in the distilled water used. It is probably true that the organism at the bottom of every infectious disease of known origin may be obtained from the blood at some stage of the infection. It is important, therefore, that the clinical examination include blood cultures in any case of obscure origin, as in this way treatment and prog- 1 See Winslow, Jour, of Inf. Dis., vol. io, 1912, p. 285; Jungmann, Deutsch. Arch. f. klin. Med., Bd. 106, 1912, S. 283; and Bergey, Jour, of Med. Res., vol. 27, 1912, p. 67. See p. 39 for a discussion of streptococcic sore-throat. THE BLOOD. 571 nosis will be markedly influenced. Churchill and Clark1 have recently called attention to the importance of this work in children. VII. Serum Pathology. This section of hematology is very closely associated with pathology and biochemistry and can, therefore, be taken up only in brief outline.2 Our knowledge of the great importance of the fluid portion of the blood is so rapidly increasing that soon the changes in its cellular composition will be of secondary importance. As a matter of fact, changes in the number of the cells as well as in the percentage of hemoglobin must be dependent, to a certain extent, upon the more obscure changes which are taking place in the plasma in various diseases. Our ability to fathom the secrets of the many physical and chemical changes of the plasma has been so slight, that we have hitherto neglected to take into considera- tion anything but the changes in the cells, which can be so easily studied by the various methods previously outlined. Our knowledge of the various types of immunity to infection and of the many factors concerned therein has increased to such an extent in recent years that a complete discussion is out of place here. By the elaboration of the side-chain theory of Ehrlich and of the opsonic theory of Wright, we have come somewhat nearer to a proper realization of the import- ance of the serum in all infections as well as in many diseases in which great metabolic disturbance is evident. For that matter one can hardly imagine a condition in which the blood plasma may not show some characteristic change, inasmuch as the nutrition of the entire body can come only through the blood. When one considers the close correlation of the various organs he may see at once that pathological changes in any of the viscera may result in an abnormal blood, which may show no variations at present capable of detection. For these reasons one hails with delight any advance in serum pathology and could but wish that his knowledge might more rapidly increase. Ehrlich's Side-chain Theory.3 The early work of Ehrlich, published in 1885, advanced a theory to account for various phases of immunity, especially of the action of the blood in produc- ing antitoxins against various poisons elaborated by infectious agents. It is to be said that no such formation of antitoxin against the ordinary medicinal poisons has been found. According to this theory, the protoplasm of the cell consists of a central group of molecules (Leistungskern), in which the inherent vital characteristics 1 Amer. Jour, of Dis. of Child., vol. i, 1911, p. 193. 2 See Kraus and Levaditi, Handbuch der Technik und Methodik der Immunitats- forschung, Jena, 1909-1911; Citron, Die Methoden der Immunodiagnostik und Immuno- therapie, Leipzig, 1912; Dieudonne, Immunitat, Schutzimpfung und Serumtherapie, Leipzig, 1911. 3 In this connection see Sachs, Die Haemolysine, Wiesbaden, 1902; Kraus and Levaditi, Handb. d. Tech. u. Meth. d. Immunitatsforschung, Jena, 1907-1911; Adami, Principles of Pathology, New York, 1910; Dieudonne, Immunitat, Schutzimpfung und Serumtherapie, Leipzig, 1911; Kolle and Wassermann, Handbuch der pathog. Mikroorg., Jena, 1912; Kolle and Hetsch, Die Experimentelle Bakteriologie, Berlin, 1911. 572 DIAGNOSTIC METHODS. of the cell are located and whose integrity is necessary for normal cell life. At different portions of the cell certain other molecular groups are attached exactly as side-chain groups are attached to the benzene nucleus of organic chemistry. These groups or, as Ehrlich styles them, side-chains are capable of uniting with various material which is brought into intimate relationship with the cell structure. Such materials are foods, toxins, and other injurious agents. In order that food may be taken up by the cell it must possess certain groups which will enable it to combine with the groups in the side-chain of the cell. It must be, in other words, homologous or, as Ehrlich states, must bear the same relationship to the side-chain which Fischer has applied in his assump- tion of the "key in the lock" hypothesis regarding ferment action upon the various types of hexoses. In the nomenclature of Ehrlich the side-chains are Fig. 151.-Illustrating the mechanism of the toxin-cell union by the intermediation of receptors. (Da Costa.) styled receptors and the group of the food or of the toxin which combines with these receptors is known as the haptophore group. These receptors, as well as haptophores, possess specific affinity, uniting with one another only when homologous. It has been found that a toxin molecule has certain injurious effects upon the cell; it is, therefore, necessary to ascribe this action to other than the two groups above mentioned, as the union of a haptophore with a receptor would form an inert substance. Ehrlich, therefore, assumes the presence in the toxin molecule of a second group which he styles the toxophore group, which exerts the untoward effect upon the cell. The toxophore group in itself can unite with the cell only through the medium of its haptophore group. As the cell becomes irritated by the presence of the toxophore group, it endeavors to overcome this by a new formation of receptors. According to the strength of the irritation, many more receptors will be found than can combine with the toxin material present, so that many of the extra receptors pass out into the circulation in the form of free receptors. This is a graphic explanation of the fact that the cell when irritated by toxic material elaborates substances from THE BLOOD. 573 its own protoplasm, which have a neutralizing effect upon the toxic substance. These free receptors (haptines) form the antitoxins. They combine only with homologous toxin material and are, therefore, specific. Welch believes that these antitoxins have a second function beside that of neutralization of toxin, namely, an irritating one upon the bacterial invaders so that these organisms are forced to elaborate similar substances for their own protection. Toxins which have been deprived of their toxophore group are known as toxoids and can combine with the receptors of the cell, exerting no untoward effect upon the cell. These toxoids may also unite with antitoxin through the medium of their hapto- phore group. Occasionally toxins are incompletely combined with the anti- toxins; that is, the antitoxic material is not sufficient in amount completely to neutralize the toxin, so that such toxins may still combine with the cells and exert a modified poisonous effect. Such attenuated toxins are known as toxones. Fig. 152.-Illustrating the elaboration and action of antitoxin. (Da Costa.) It has been found that the injection into animals of bacteria, various body cells, and certain secretions of some animals, as for instance, snake venom, gives rise to the development of specific antibodies in the blood serum of the animal so treated, the substances injected being named antigens? Such blood or serum will be found to have a lytic (destructive) action, upon cells similar to those injected. Such sera are specific; that is, they act only upon the kind of cell used in the injection. The term hemolysis has been introduced to express the destructive action upon the erythrocytes shown by the dissolving out of the hemoglobin from the red cell. The stroma or discoplasm of the red cell is a membrane which shows peculiar relations to diffusion of various materials into the cell and to the passage of hemoglobin and other cellular material from the cell. Its chief function seems to be to prevent, as far as possible, any loss in hemoglobin. If this membrane becomes permeable, then we must assume the action of some toxic material. The term hemolysis has reference merely to the abnormal loss of hemoglobin and not to any disturb- ance beyond increased permeability of the stroma. The stroma of these cells 1 See Pick in Kolle and Wassermann's Handb. d. Path. Mikroorg., Bd. i, 1912, S. 685, who shows that true antigens are always albumin-containing colloids. 574 DIAGNOSTIC METHODS. remains behind and may be seen in the centrifuged specimen as the so-called shadows. Hemolysis must, therefore, be considered as a sign of protoplasmic death. Substances (hemolysins) bringing about such change belong, neces- sarily, in the class of blood poisons. Such hemolysins are increased or lowered temperature, various inorganic compounds, such as distilled water, ammonium salts, and organic compounds, such as urea, bile acids, ether, alcohol, chloroform, solanin, saponin, and digitalin. The saponins are among our strongest hemolytic substances, acting in dilutions of 1 to 100,000. Besides these we have various secretions, such as those of the cobra, spider, and the bees which are active hemolytic agents. It has been observed by various workers that hemolysis is prevented when the serum is heated to 6o°. This points to the fact that some substance is destroyed which is of great importance in this process. In addition it has further been found that the renewing of the activity of the old heated serum by adding a supply of fresh isologous serum will restore the hemolytic activity. It is evident, therefore, that there are two factors which must be taken into consideration, one is a thermostable (heat-resisting) substance, while the other is thermolabile (destroyed by heat). To the first of these Ehrlich has given the name amboceptor and to the second the name complement. The amboceptor has been shown to have two haptophore groups, with one of which it unites to the receptor of the cell and with the other to the complement. The hapto- phore group which unites with the cell is known as the cytophile, while the one uniting with the complement is termed the complementophile. For hemolysis, therefore, we must have the cell receptor, the amboceptor, and the complement. The complement has also been shown to have two groups analogous to those of the toxin molecule. The first is the haptophore group, while the second is the zymophore group, through which the destructive action upon the cell is manifest. Fig. 153.-Illustrating the mechanism of hemolysis. (Da Costa.) THE BLOOD. 575 The amboceptor is formed within the body as the result of cellular hyperactivity aroused by the irritant action of the toxic material. The comple- ment is probably derived, for the most part, from the leucocytes, and acts very much as an enzyme. It can exert its toxic action only when united with the cell by means of the amboceptor, so that free complement has no injurious effect. It has been frequently observed that the red cells are more resistant than normally, while in many cases they appear less resistant to hemolysis. This is explained by the side-chain theory very much as it explains the formation of antitoxin. These antihemolysins are formed within the blood plasma after Fig. 154.-Illustrating the Mechanism of Antihemolysis. (Da Costa.) A, Interference of anticomplement with complement-amboceptor union. B, inter- ference of antiamboceptor with amboceptor-cell union. C, antiamboceptor-amboceptor union. D, anti-complement-complement union. inoculation with hemolytic material. The hyperactivity of the cell causes it to throw off two types of such bodies, namely, anticomplement and anti- amboceptors. The former combines with the haptophore group of the comple- ment and the latter with the cytophilic group of the amboceptor, each of these combinations making it impossible for the necessary union of cell, amboceptor, and complement to occur. It has been found that frequently the serum of an animal, which has been injected with certain bacteria or with certain body cells, shows the peculiar property of agglutinating or clumping such bacteria or cells when these latter are added to it. This condition is known as agglutination and the agents bringing it about are styled agglutinins. These substances are developed in the blood of the animal during the process of adaptation toward the presence of such foreign material. Agglutinins from the standpoint of the side-chain theory are free receptors, having a haptophore group -which imites with the 576 DIAGNOSTIC METHODS. receptor of the cell or of the bacterium, and cause agglutination through the presence of a second group known as the zymophore or agglutiniphore group. Agglutination, therefore, does not require the presence of a complement. In some cases, intraperitoneal injection of body fluids or of solutions of certain proteins into animals brings about a condition which enables the blood serum of such animals to cause a precipitation of the protein to which the animal has been adapted. This fact has been taken advantage of in formulat- ing a medicolegal method for the detection of blood of different animals and will be taken up in a later section. The precipitins consist of free receptors combining, by means of their haptophore group, with the receptor of the cell and exerting their precipitating effect through the medium of their zymophore or precipitinophore group. Phagocytosis. According to the theory of Metschnikoff, the leucocytes are capable of incorporating into their substance materials which are foreign to the blood in which they circulate. This process is known as phagocytosis and is one of the greatest protective measures which the system has for its fight against bacterial invasion. When the blood becomes laden with bacteria, as in the various infectious diseases, we find a leucocytosis in practically all cases, the exceptions having been previously noted. This is the natural sequence if the system is to rid itself of these invaders. The leucocytes are drawn by chemical attraction (chemotaxis) toward the bacteria and attempt to swallow them by throwing pseudopodia about them and drawing them into the proto- plasm. This is successful in many cases, while in others it is not, so that the question of ascendency of the leucocyte or of the bacterium will depend upon the degree of phagocytosis. Strangely enough much variation is shown in the susceptibility of different organisms to phagocytosis. Thus the pneumococcus at times is very difficultly amenable to phagocytosis, while at others it is easily acted upon. The recent work of Rosenow1 has thrown much light upon the mechanism of this action. Opsonins. Realizing that there was a more definite basis for phagocytosis than was embraced in the older conceptions, Wright2 introduced the idea of opsonins to designate the presence in the blood serum of substances which render the various bacteria subject to phagocytosis. The normal blood serum contains such opsonic material for the various bacteria with which it may be infected, but this varies greatly toward the different organisms. Thus we may find individuals showing much more opsonin (a higher opsonic index) toward one organism than toward another. This explains, in a way, the well-known fact that different people are variably susceptible to the same disease, while the same individual may be strongly resistant toward infection with one organism, but easily a victim of another infection. 1 Jour, of Inf. Dis., vol. 3, 1906, p 683. 2 Proc. Royal Soc., vol. 62, 1903. p. 357. THE BLOOD. 577 Regarding their clinical nature very little is known. There seems to be a certain amount of evidence which points to the fact that these opsonins belong to the class of globulins, while Simon and Lamar1 have shown that they are apparently associated with the euglobulin fraction. Quite as little is known of the structure of the opsonins, so that it is at present doubtful in -what position to place them in the side-chain theory. According to Hektoen,2 they may contain a haptophoric group which unites with the bacterial or other receptors and also an opsonipherous group which brings about changes in the cell, making it capable of phagocytosis. According to Savtchenko and Dean, the opsonins should be regarded as amboceptors, while Greig-Smith looks upon the process of opsonification as the first stage of agglutination. All of these theories must wait for future confirmation. The opsonins are thermolabile and are usually destroyed by heating for ten minutes to 6o° C. They occur in all classes of vertebrates and show here a peculiar characteristic, namely, that the serum of different animals is capable of activating various organisms for phagocytosis by leucocytes of animals of different species. This would bring the opsonins into the same field as the agglutinins, precipitins, and hemolysins. From a clinical standpoint, opsonins are frequently found to be diminished in certain bacterial infections. It is, therefore, conceivable that the resistance of the patient or, in other words, his phagocytic power might be increased by the addition of substances which could act as opsonic material. Such substances are the bacterial vaccines of Wright, suspensions in physiological salt solution of dead cultures of the organism to which the patient shows a diminished power of phagocytosis. The relation of the phagocytic power of the patient, as evidenced by the number of organisms which a definite number of leucocytes takes up under the opsonifying influence of this serum, as com- pared with the same condition in the case of the serum of a normal individual, is known as the opsonic index toward the organism investigated. The number of bacteria taken up by the leucocytes of the normal individual is taken as one. According to Wright, the injection of a dose of vaccine is followed by a decrease in the opsonic index (the negative phase), which is of variable degree and duration, depending upon the dose. This negative phase is followed by an increase in the opsonic power of the leucocytes (the positive phase), which is associated with improvement in the condition of the patient. The various doses of the vaccine should be so administered that it is never given during a negative phase. While a low opsonic index is the rule in chronic cases, high indices may be observed with active systemic manifestations of acute cases. As a rule, it is more generally beneficial to use a vaccine prepared from the discharges of the patient than it is to use an already prepared vaccine of the same organism. The reason for this is that so much difference exists in the 1 Johns Hopkins Bull., vol. 17, 1906, p. 27. 2 Jour. A. M. A., vol. 46, 1906, p. 1407; Ibid. vol. 48, 1907, p. 1739; Ill. Med. Jour., vol. 13, 1908, p. 9. 578 DIAGNOSTIC METHODS. various strains of the same organism that little or no result may follow the administration of a stock vaccine. These autovaccines are very easily prepared by one having bacteriological facilities and can be administered by the attending physician. It is not always easy to gauge the dosage of a vaccine in any particular case, as the idiosyncracy of the patient plays such an important role. It is better practice, therefore, to start with a small dose and obtain no reaction than it is to give an overdose which will prove very harmful. Failure to observe this precaution is, in the writer's opinion, the reason that tuberculin so soon fell into disrepute and is now being restored to such great favor. It has been found by various workers that the administration of a proper dose of a vaccine is followed by certain local symptoms, such as swelling, redness, and tenderness at the point of injection, while certain systemic symptoms, such as rise in temperature, general malaise, and pains in the joints, are practically always evident. It has, therefore, become the practice of many of our workers, to gauge the results of vaccine treatment more by the reaction than by the determination of the opsonic indices of their patients. The routine of such treatment is to give an injection of bacterial suspensions about once a week, oftener only in cases which have shown no reaction following a previous injection.1 One great advantage of an autovaccine over a stock vaccine is that the former contains all of the organisms bringing about the infection. Affections which are amenable to treatment by vaccine therapy are practically always of the mixed variety so that a stock vaccine will increase the opsonic index and thus clear up only the infection with that specific organism. Recent work, especially in cases of gonorrhea, has shown that autovaccines do not always give better results than stock vaccines, for the reason that the patient has become so accustomed to his own autogenous material that a difference in strain is sometimes very beneficial. As a rule, the beginning dose of a staphylococcic vaccine is 500,000,000 organisms, while that of gonorrhea is approximately 40,000,000. There does not seem to be at present a well-established dosage for the streptococcic vaccine,2 so that each case must be a law unto itself. The most amenable cases for treatment by bacterial vaccines seem to be cases of long-standing suppurative conditions induced by the various pus-forming organisms. Some obscure infections, which the writer has seen treated with the colon vaccine, have cleared up in a surprising manner. The general principles of such work are as follows. For the opsonic determinations four factors are essential: (1) the patient's serum, (2) the normal control serum, (3) the washed leucocytes, and (4) the bacterial 1 Recently the use of typhoid vaccine as a preventive immunizing inoculation has become very widespread. The doses are given at ten-day intervals; the first contains 500 million bacteria and the second and third 1 billion each. A local reaction usually subsides within 48 hours, while a general reaction is not always present. See Wright, Lancet, vol. 2, 1896, p. 807 ; Brit. Med. Jour., vol. 1, 1897, p. 16; Leishman, Jour. Roy. Inst. Pub. Health, vol. 18, 1910, p. 394; Firth, Jour. Roy. Army Med. Corps, vol. 17, 1911, p. 495; Russell, Bost. Med. and Surg. Jour., vol. 164, 1911, p. 1; Nelson and Hall, Jour. A. M. A., vol. 57, 1911, p. 1759; Davis, Ibid., vol. 58, 1912, p. 537; Russell, Ibid., p. 1331; Maverick, Ibid., p. 1672; Engelbach, Interstate Med. Jour., vol. 19, 1912, p. 537; Albert and Mendenhall, Amer. Jour. Med. Sci., vol. 143, 1912, p. 232; Callison, Ibid., p. 350; Williams, Ibid., p. 352. 2 See Rosenow, Ill. Med. Jour., vol. 22, 1912, p. 676. THE BLOOD. 579 emulsions. One volume of washed corpuscles is drawn into a capillary tube and is followed by one volume of the patient's serum and then by one of the bacterial emulsion. The contents of the capillary are then thoroughly mixed by blowing the material into a dish and redrawing it into the capillary. A similar mixture is then made using the control serum. Both capillary tubes are now incubated for 15 minutes at 370 C. A drop of the mixture is blown onto a glass slide, a smear is made, fixed, and stained with any of the poly- chrome dyes or with special bacterial stains. The bacillary index is then determined by counting a series of 50 to .100 polynuclear leucocytes and observing the number of organisms in each. The phagocytic index represents the average number per leucocyte. The opsonic index is then calculated by dividing the patient's value by that of the normal control. Diagnostic Use of Vaccines. The vaccines discussed above are used almost entirely for therapeutic pur- poses. While it is highly probable that any of the vaccines might give char- acteristic reactions in cases infected with the same organism, yet they have not been put to extensive diagnostic use. However, certain suspensions of organ- isms are, at present, widely employed in this way. It is evident, from the preceding discussion, that the serum of a patient contains bodies acting as neutralizing agents against the toxins of the infecting bacteria. This property varies with the individual and with the severity (dos- age) of the infection. Patients, who have acquired immunity against certain infectious diseases, either by passing through an attack of the infection or by use of protective vaccination (small-pox or typhoid fever), or who are in the process of acquiring immunity, that is, are at the time passing through an active infection, show a certain changed reactivity to infection with the same organism. In other words, such patients are not insensible to reinfection, but show a changed reactivity, as von Pirquet1 has shown, especially as regards the time, quality and quantity of the reactions observed. This is the condition known as Allergy and is shown only by patients immune or becoming immune (infected) to the suspected organism. Normal individuals do not show the reaction so quickly nor so intensely as do the infected or immune subjects, providing the dosage is properly regulated, as the former have sufficient antibodies to over- come the effects of the amount of toxin, which will produce local or general reactions in the infected patient.2 These are the basic principles underlying the diagnostic use of vaccines. Tuberculin Reactions. Method of Koch. This is the earliest method for the diagnostic use of tuberculin. The normal temperature variations of the patient are first determined for 24 hours previous to the injection. The preparation used is Koch's old tuberculin, 1 Arch, of Int. Med., Vol. 7, 1911, p. 259. 2 See Hektoen, Jour. A. M. A., vol. 58, 1912, p. 1081; also, Vaughan, Amer. Jour, of Med. Sci., vol. 145, 1913, p. 161. 580 DIAGNOSTIC METHODS. which is a water and glycerin extract of the cultures of the tubercle bacillus, of such a strength that i c.c. is equivalent to i gram of tuberculin. The injection may be made into the deeper tissues either in ihe intra- scapular or gluteal region by means of a properly sterilized hypodermic syringe. The amount of tuberculin to be injected depends on the age and condition of the patient. As a rule 1/2 to 1 mg. is given to a robust adult, while a weak patient or a child receives 1/10 to 3/10 mg. These strengths are easily made by diluting the original solution with the proper amount of 1/2 per cent, carbolic acid solution. The temperature variations should then be followed, obser- vations being made every two hours. A rise of 1/20 C. (9/100 F.) within a few hours or, even, as late as 36 hours is taken as a positive result, although some observers require a more marked rise before admitting a positive finding. Few other symptoms are noticed but, if the dose has been large, marked malaise, chill, high fever, and local reactions may obtain. If no rise in temperature is observed after the first injection, a second or a third one may be given, the dcyse being doubled each time but exceeding, under no circumstances, 10 mg. It is possible that a single injection of 5 or 10 mg. may give more diagnostic results than repeated doses up to this amount, although it must be admitted that such large doses may react, in rare cases, positively in normal individuals. A positive result is very strong evidence of an active tubercular condition. A negative result is less reliable as some tubercular conditions, such as peritoni- tis or encapsulated foci elsewhere, may not react positively. In this method use is made of a salve consisting of equal parts of old tuberculin and lanolin. A small bit of this salve is rubbed into the skin, pref- erably in the thoracic or abdominal region. After 24 to 48 hours small pale nodules, 1 to 2 mm. in diameter, are seen at the point of inoculation if the case be tubercular. A more severe reaction may be observed in the form of reddish miliary nodules. These nodules are limited to the point of application of the salve and disappear in 24 to 48 hours. Occasionally, some exudation may arise from the intense reactions, but this is very rare. This method is very simple and is very reliable. Method of Moro. Method of von Pirquet. This method1 is very widely used, especially in cases of infection in children. The inner side of the forearm is cleansed with alcohol and ether and allowed to dry. Two drops of old tuberculin are placed on the skin at a distance of 6 to 8 cm. from each other. The skin covered by these drops is then punctured by a large needle or gently scarified. The tuberculin is allowed to remain for 10 to 15 minutes, when the clothing may be replaced. It is wise to make a third puncture or scarification between the two tuberculin points as a control upon the patient's reaction, no tuberculin being used here. A slight swelling is usually produced which becomes surrounded by a light reddish border in 1 See Wachenheim, Amer. Jour, of Dis. of Child., vol. 4, 1912, p. 27; also, Austrian, Jour, of Exp. Med., vol. 15, 1912, p. 149. THE BLOOD. 581 a few hours, and disappears in 24 hours. If the reaction be positive, the area of puncture swells within 24 to 48 hours to a papule, reddish in color and 10 to 15 mm. in diameter. The center may be pale and be surrounded with small serous points. By comparison with the normal point, a reaction may be more clearly demonstrated. The swelling and redness begin to disappear within 48 hours, a slight pigmentation usually following this positive reaction. No general disturbance is noted as a rule. If the reaction be repeated, a marked positive result may be observed as an evidence of hypersensibility, the so-called anaphylactic reaction. This may, also, be noticed in the other methods, even though a much smaller dose be used than in the initial application. This method was advanced almost simultaneously by Calmette and Wolff- Eisner. One or two drops of a 1 to 100 dilution of old tuberculin in physiologic salt solution are placed in the conjunctival sac, the lids being held apart for a few seconds to permit of absorption of the tuberculin. Only one eye is thus treated, the other being used as the control. A positive reaction appears in 6 to 24 hours and may assume one of the following types: (1) Mere reddening of the caruncle and the inner surface of the lower lid; (2) the bulbar conjunctiva may take part in the process; (3) suppurative conjunctivitis with marked injection of the palpebral conjunctiva; (4) hemorrhagic conjunctivitis of the palpebral and bulbar portions with profuse fibrino-purulent exudation. An associated rise in temperature may be observed in the severe types. Contra- indications to this test are found in any inflammation of the eye or conjunctiva, in the previous treatment of the eye with tuberculin, and recent application of tuberculin elsewhere. It would not seem that this method is advisable owing to the great discomfort which may arise and, especially, as it is not without danger of severe injury to the eye.1 Method of Calmette. Luetin Reaction. Noguchi2 has recently announced "A Cutaneous Reaction in Syphilis," which may prove of great value in the diagnosis of this condition as soon as the inoculating material may be obtained for general use. His material is prepared as follows: Pure cultures of the treponema pallidum (see p. 559) are allowed to grow for periods of 6, 12, 24, and 50 days at 370 C. under anerobic conditions. One was cultivated in ascitic fluid containing a piece of sterile placenta, and the other in ascitic fluid agar also containing placenta. The lower portion of each solid culture was cut out and the tissue removed. These agar columns, con- taining large numbers of spirochete, were carefully ground in a sterile mortar. This paste was then gradually diluted by adding, little by little, the fluid culture, until the emulsion becomes perfectly liquid. This mixture is heated to 6o° C. 1 Austrian (Bull. Johns Hopk. Hosp., vol. 23, 1911, p. 1) advocates a similar test for ty- phoid fever. The antigen is a " typho-protein " prepared from a mixed culture of 80 different typhoid strains. Positive results are obtained in 95 per cent, of cases of typhoid while it is negative without exception in practically all controls. Its great advantages are that it is a bedside test and is given in the early stages of the disease. 2 Jour, of Exp. Med., vol. 14, 1911, p. 557; Munch. Med. Wochs., Bd. 58, 1911, S. 2372. 582 DIAGNOSTIC METHODS. for one hour in a water-bath and 0.5 per cent, carbolic acid was added. When this mixture is examined under the dark-field microscope 40 to 100 pallidas may be seen in each field. This suspension is called luetin. The skin of the upper arm is sterilized with alcoholic sublimate solution before the injection. The amount of luetin injected is 0.05 c.c. This injection is intradermic, that is in the skin as superficially as possible. In normal cases, there appears, after 24 hours, a very small erythematous area at and around the point of injection. No pain or itching is experienced. This slight reaction gradually recedes within 48 hours and leaves no induration. In certain cases, the reaction may reach a stage of small papule formation after 24 or 48 hours, after which time it commences to recede. In positive cases the following types of reaction occur: (A) Papular Form.-A large, raised, reddish, indurated papule, usually 5 to 10 mm. in diameter, makes its appearance in 24 to 48 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 follow- ing three or four days, after which the inflammatory processes begin to recede. The color of the papule gradually becomes dark bluish-red. The induration disappears within a week, except in certain instances in which a trace of the reaction may persist for a longer period. This 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. (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 obtains. Within the next 24 hours, the papule changes into a vesicle filled at first with a semi- opaque serum that later becomes definitely purulent. Soon the pustule rup- tures. The margin of the broken pustule remains indurated, while a crust quickly forms, which falls off within a few' days. The induration soon disap- pears leaving almost no scar. This reaction was found almost constantly in tertiary types as well as in secondary or hereditary forms which had been treated with salvarsan. (C) Torpid Form.--In rare instances, the injection sites fade away to almost invisible points within three or four days, so that they may be passed over as negative reactions. Sometimes these spots suddenly light up again after 10 days or so and progress to small pustular formation. No marked constitutional symptoms have been observed after the use of the luetin. In most positive cases a slight rise in temperature took place lasting for one day. The conclusions of Noguchi,1 based upon a study of 400 cases, are as 1 See, also, Ziegel, Arch, of Int. Med., vol. 9, 1912, p. 520. THE BLOOD. 583 follows: Luetin produces a cutaneous reaction in syphilitic and parasyphilitic patients that is most constant and severe in the tertiary and hereditary affections. During the primary and secondary stages, the reaction is infrequent, and when present is of mild degree. An exception has been found in cases in which energetic treatment has been or is being carried out and in which clinical signs of syphilis are absent. Such cases may show a severe reaction, especially those treated with salvarsan. In certain cases of old infection in which no treatment has been taken and in which no symptoms have appeared for many years, and in the course of which miscarriages have occurred, this reaction has failed to appear. Despite the absence of symptoms, mothers who have young syphilitic children have usually given the reaction. It remains to be determined in how far this reaction can be used to supplement the Wassermann reaction. It appears probable that the Wassermann reaction is more constant in the primary and secondary, and the cutaneous reaction in the tertiary and latent forms of syphilis.1 It further appears that the Wassermann reaction is more directly and immediately affected by antisyphilitic treatment than is the cutaneous reaction.2 , VIII. Sero-Diagnosis. In the attempt to enlarge our diagnostic procedures advantage has been taken of many of the principles previously discussed. The methods evolved are among those most frequently employed and are reliable within certain small limits which will be mentioned later. They are all based upon certain peculiar properties possessed by the serum of patients with infectious diseases and are, therefore, of special importance in differential diagnosis. Fig. 155.-Bacillus typhosus at beginning of Widal test. (Da Costa.} A. Agglutination Reactions. Since the work of Gruber and Durham and of Pfeiffer and Kolle in 1896, the fact has been well established that the serum of an animal, which has been 1 See Robinson, Jour, of Cut. Dis., vol. 30, 1912, p. 410. 2 Irons has recently perfected a local skin reaction in cases of generalized gonorrheal i nfections, using as his vaccine a glycerine extract of autolyzed gonococci. After a few hours a'papule is formed with a surrounding area of hyperemia, the maximum reaction being reached in 24 hours. Irons classifies as positive the reactions showing a total diameter of 5 mm. or over; nagative when 3 mm. or less (Jour, of Inf. Dis., vol. 11, 1912, p. 77). 584 DIAGNOSTIC METHODS rendered immune to certain bacteria, shows the property of agglutinating or clumping homologous bacteria, owing to the production of immune bodies, the so-called agglutinins. This reaction takes place in high dilutions of the serum and is practically specific for any given organism. Widal,1 in the same year, applied this principle to the diagnosis of typhoid fever, showing that clumping and loss of motility of the typhoid bacilli occur when a suspension of the actively motile types is treated with an homologous immune serum at such a dilution that normal or non-homologous serum does not react. 1. Gruber-Widal Test. Fig. 156.-A Pseudo-Widal reaction. {Da Costa.) Cultures. The cultures of the typhoid bacillus must be fresh and must show many actively motile organisms in the hanging-drop specimens, if they are to be used in the microscopic method. It is advisable always to use cultures containing Fig. 157.-A positive Widal reaction. {Da Costa.) about the same number of bacilli, as the reaction is somewhat quantitative. Thus, if few bacilli be present, they may be clumped by a small amount of agglutinin both specific and non-specific, while if the number be very large, the specific agglutinin may not be sufficient to cause marked clumping and loss 1 Bull. Med., Tm. io, 1896, pp. 618 and 766. THE BLOOD. 585 of motility of the organisms. Great differences toward agglutination exist in different strains of the typhoid bacillus, so that it is essential always to employ, for diagnostic purposes, strains which have passed through several generations on artificial media. A stock culture of the bacillus is kept in sealed tubes of nutrient agar in a cool dark place and, from this stock, fresh agar cultures are made every few weeks as occasion may require. From these stock agar cultures, bouillon tubes are inoculated and incubated for 15 to 24 hours. These latter are used for the agglutination tests. These bouillon cultures should be equally cloudy throughout and show no pellicle formation or gross clots of bacilli. It is wise to make up fresh bouillon cultures every other day so as to be certain of the maximum motility of the organisms. In the use of the macroscopic method cultures killed by heat, formalin, carbolic acid, thymol, etc., may be employed. Many laboratories throughout the country supply these cultures for general work. Ficker's "Typhus diagnos- tikum" is one of the widely used types. For self-evident reasons it is wise to have some standard strength of these cultures. Hastings advocates a mixture of 5 c.c. of 5 per cent, carbolic acid solution, 10 c.c. of glycerin, and 85 c.c. of physiologic salt solution to which are added the organisms scraped from the surface of two 24-hour agar-slant cultures of bacillus typhosus (the bacilli being rubbed into the mixture with a spatula). Bass advises a suspension of 10,000 million dead typhoid bacilli per c.c. in 1.7 per cent. NaCl solution to which 1 per cent, formalin is added. Obtaining the Blood. In this procedure much depends upon the custom of the worker as to the method followed. As this reaction is a pure serum one, it is much more logical to use the serum than the whole blood, if one has any choice in the matter. The simplest and best clinical method is the use of a small capillary tube (made from 1/4-inch glass-tubing) with a central bulbar enlargement into which the blood (15 to 20 drops) is drawn by capillarity from a puncture of the ear. The tube is then laid flat until coagulation occurs and the serum separates. To hasten this process, the end of the tube, which contains no blood, may be sealed in the flame and the tube centrifuged. If it is desired to send the specimen to the laboratory, seal both ends and pack in such a way as to insure it against breakage. In this connection it may be well to recall that Lyons1 has shown the possibility of obtaining a culture of the bacillus typhosus from the clot settling in such tubes. This may be done by the method previously outlined. An- other method of obtaining the serum is to apply a cantharides blister which will furnish enough serum in 6 to 12 hours for many tests. Venous puncture may, also, be employed, but this is rarely advisable as sufficient serum may be more easily obtained in the above ways. Some workers use the whole blood instead of the serum. If this be done, the leucocytometer may be employed, using physiologic salt (0.9 per cent.) 1 Arch, of Int. Med., vol. 4, 1909, p. 64. 586 DIAGNOSTIC METHODS. solution as the diluent. Draw the blood from a puncture in the ear to the mark 1 and the diluent to 10. This gives a dilution of the whole blood of 1 to 10 or 1 to 20 of the plasma, as the plasma forms approximately 50 per cent, of the whole blood. This dilution may then be used directly for the later work by mixing with an equal volume of the typhoid suspension. More frequent, possibly, in the hands of workers in municipal or the larger clinical laboratories, is the employment of dried blood sent to them on glass slides, bits of paper, mica, etc. Such specimens may be easily transported but can not yield anything but an approximate result, as an accurate dilution is out of the question unless very careful weighing of the blood is carried out. A large drop of the dried blood is dissolved in 10 drops of water, the dilution being called 1 to 10. If the blood be upon paper, the dilution must be made largely by guess, experience teaching an approximate dilution from the color of the mixture. These dilutions will, of course, be doubled when the test is applied. The methods of diluting the whole fresh blood and the dried blood have already been given. As diluting pipets for the serum, one uses small (1/4 inch) glass tubing which has been drawn into a long capillary at one end. A rubber bulb fits over the larger end and provides the suction. It is good practice always to have a number of these pipets on hand so that they may be discarded or thoroughly cleaned before a second use. If the same pipet be employed for different cases, a contamination may arise to such an extent that erroneous re- sults obtain. A few drops of the serum, collected as above, are drawn into the capillary tube by suction. From this tube the serum is then dropped into the mixing vessels, preferably watch crystals. Two dilutions are usually made in the writer's laboratory namely 1 to 25 and 1 to 50. Allow 1 drop of serum to fall into each of two watch crystals, the remaining serum in the capillary being blown into the original collecting tube. Now add, from the same pipet used in dropping the serum in order to insure drops of the same size, to one of these drops of serum 24 drops of physiologic salt solution and to the other 49 drops of the same diluent. Mixtures of these diluted sera with an equal volume of typhoid suspension, as employed in the test, give final dilutions of 1 to 50 and 1 to 100. Dilution of the Serum. The serum (preferably), whole blood, or dried blood is diluted as previously described. Mixtures are then made of these dilutions with an equal volume of typhoid suspension. By means of a platinum loop of stiff wire, a loopful of the diluted serum (i to 25) is placed on the center of a cover-slip. After burning the serum from the loop, a loopful of the typhoid suspension is transferred to the same cover-glass and the two fluids thoroughly mixed by means of the loop. The dilution of the serum will then, evidently, be double its original dilution, or 1 to 50. The ordinary hanging-drop preparation is then made by inverting the cover-slip over the hollow of the slide, which has been ringed with vaseline. Microscopic Reaction. THE BLOOD. 587 Instead of using the hanging-drop method, some prefer to place the cover di- rectly upon the slide. Usually the specimen does not dry sufficiently to intro- duce errors, but the results are certainly not as characteristic by this procedure in all cases. The specimen is best studied with the high-power dry lens, using a strong light. The organisms will be seen, in specimens properly lighted, as free actively motile bacilli moving with their rapid darting motion through the field. In the course of 45 to 60 minutes, or often in much less time, the organisms will be seen to be collected in clumps and to be motionless, except for a slight vibratory movement. This is a typical positive reaction, namely marked clumping with complete loss of motility in one hour at a dilution of 1 to 50. Such a reaction is practically never due to any other disease than typhoid fever, if the culture of the bacillus be a pure one. In some cases marked agglutination with incomplete loss of motility obtain. In the writer's opinion this can be interpreted only as a doubtful or incomplete reaction. The probability is that typhoid fever is present, but one must admit the possibility of group agglutinins, due to the paratyphoid bacillus, as well as agglutinins of other infections. When dried blood is used, a slight tendency to agglutination may be observed, due to the presence of small parti- cles of fibrin, so that in such cases the loss of motility should be demanded even more rigorously than when serum is used, before a positive result is reported. It has been shown that the serum of many healthy patients or that of subjects suffering with diseases other than typhoid fever frequently show a reaction similar to that described above. This statement is true only when the question of dilution of the serum is taken into account. The characteristic Widal reaction is shown at a dilution of 1 to 50 or higher, while in other diseases, with the exception of paratyphoid, the reaction practically dis°npears (the typhoid bacillus being, of course, used) at a dilution of 1 to 20. The element of time is also of great importance. In typhoid fever, the reaction may occur instantly at 1 to 10 or 20, and in one-half to one hour at 1 to 50, yet a period ranging from one-half to one hour is necessary with other sera at a dilution of 1 to 10 or 20. On the other hand, there are cases of undoubted typhoid fever that show no agglutination at 1 to 10 or 20, while at 1 to 50 this is marked. What occurs here is uncertain, some crediting the paradox to bacteriolysis of the organisms at the former dilutions. Macroscopic Reaction. This method is based on the fact that the agglutination of typhoid bacilli by a potent homologous serum is visible to the naked eye, the clumps of bacilli settling to the bottom of the test-tube leaving a clear supernatant fluid. If living cultures of bacillus typhosus are employed, the best technic is the following: Make up the serum dilutions as previously described, using x to 50 and 1 to 100 dilutions. To these diluted sera, contained in small sterile test- tubes (11 cm. long by 8 mm. wide), is added a loopful of typhoid bacilli scraped from the surface of a 24-hour agar culture. The loop is rubbed against the 588 DIAGNOSTIC METHODS. inner surface of the test-tube until the bacteria have been suspended in the serum. The turbid suspension is then thoroughly shaken until it becomes uniformly cloudy, no clumps of bacteria being seen, and is placed in the incu- bator at 370 C. for one hour. The tubes are then removed and examined by strong transmitted light. The agglutinated bacteria may easily be recognized by their granular flocculent appearance. General clumping should obtain in one hour at a dilution of 1 to 50, while complete precipitation (settling) should follow in 24 hours, a clear supernatant fluid being left behind. This is the positive reaction. A negative result is shown by no clumping and little, if any, loss in the turbidity of the mixture. For the general practitioner, who desires to use the macroscopic method, the employment of cultures killed by heat or chemical means is recommended, as it obviates the necessity of an incubator, culture media, or microscope and permits him to have on hand material for his tests. Any of the suspensions of killed organisms may be used, Ficker's and Hastings' being widely employed. Dilute the serum as previously described, the dilutions in this case being 1 to 25 and 1 to 50. To this diluted serum (1/2 c.c.), contained in the small test-tubes, add an equal volume of the suspension of killed bacteria and examine as above mentioned. The clumping should be observed in one hour and complete clearing should obtain in 24 hours, if the reaction be positive. This method gives very reliable results and is very simple. Method of Bass and Watkins. These workers1 have introduced a modification of this macroscopic agglu- tination test, which is simple, reliable, and quick. It has the advantage that it can be carried out at the bedside, the result being known in two to three minutes. Little equipment is required. The suspension consists of 10,000 million killed typhoid organisms per c.c. in 1.7 per cent. NaCl solution to which 1 per cent formalin is added. Allow a full drop of blood to fall into 4 drops of water and mix thoroughly. Instead of this, one may make a blood-smear (using approximately 1/4 drop of blood) and dissolve this on the slide with 1 drop of water, the mixture being stirred with a tooth-pick or similar substance. With this diluted blood (1 to 4) mix an equal amount of the above suspension of typhoid bacilli on a glass slide. Tilt the slide from side to side so as to keep the mixture flowing back and forth. If the reaction is positive, a grayish mealy sediment appears within one minute, usually in less time. This sediment appears in the fluid around the edges and tends to collect there. If the agitation is continued, the clumps increase in size for 2 to 3 minutes. If the reaction does not appear in this time, it will not show up at all. When the reaction is negative, no agglutination occurs and the mix- ture remains as clear and unchanged as when placed on the slide. Recently Michaelis2 has advocated a new agglutination reaction for the bacillus typhosus as well as other bacteria. This is based on the fact that bacteria are agglutinated by acids, the rate and degree being dependent upon 1 Arch, of Int. Med., vol. 6, 1910, p. 717. 2 Folia Serologica, Bd. 7, 1911, S. 1010. THE BLOOD. 589 the concentration of hydrogen ions present. This test promises much in the future as it depends entirely upon chemical factors which are more easily and more exactly controlled than are the uncertain biologic units with which we usually work. As this test is too complicated for the general worker, the writer refers to the literature for the details.1 The time at which a Widal reaction may appear in the course of a typhoid is of some importance. The earliest date of appearance of a positive reaction is rather hard to determine, as workers differ in their interpretation of the beginning of the disease. Definite reports of positive results as early as the second day have been given by Fraenkel and others. Although this reaction may appear before the rose spots, splenic tumor, or diazo reaction, yet it is practically never found when the blood shows a normal or an increased number of leucocytes, a leucopenia with a relative lymphocytosis being associated with a positive Widal in practically all cases. A differential blood count may, therefore, be of great value when the means of making an agglutination test are not at hand. Wood was able to obtain positive results in only io per cent, of his cases in the first week of the disease. The majority of observers are agreed that the first appearance of this reaction is in the second week of the disease in most cases. In a certain number, it may not appear till the third or fourth week the highest agglutinations frequently being seen toward the end of the disease and in the early stages of convalescence. It is very rare not to obtain a positive reaction at some stage of the disease, providing the tests are made at dif- ferent intervals. It is possible that cases reported, in which no Widal reaction was observed throughout the course of the disease, may be due to faulty technic or to insufficient tests. A single negative test, of course, proves nothing. The persistence of the Widal reaction is likewise variable. Little seems to depend upon the severity of the attack as to the time of disappearance of the reaction. Usually this reaction persists for one to two months, although it may obtain for four to six months or even for years. It is undetermined whether these latter cases are due to reinfection rather than a continuation of the agglutination powers of the serum. Reports are extant of cholelithiasis in which typhoid bacilli were isolated from the gall-bladder as long as eighteen years after an attack of typhoid, the Widal reaction being positive in these cases. In this connection it is to be said that "typhoid-carriers" do not show a positive Widal as a rule. Time of Appearance and Disappearance of the Reaction. Specificity of the Reaction. From the standpoint of the biologic theory, the typhoid bacillus is agglutin- ated only by an homologous immune serum. Practically, we find that this is the case within certain limits, if all conditions are properly met. A positive Widal reaction is not infrequently met with in cases of jaundice, in which no specific disease is clinically evident. However, it is certain that most, if not all, 1 See Beniasch, Zeitsch. f. Imnun.-Forsch., Bd. 12, 1912, S. 268. 590 DIAGNOSTIC METHODS. these cases of jaundice are actually secondary to a typhoid cholecystitis with its accompanying sequelae. Such patients may have an "ambulatory" typhoid or may have no clinical signs other than a positive Widal. Further, in many cases of infection with bacteria of the colon group, a "collateral agglutination" may obtain, that is agglutination of several members of the group may occur with one serum. Cases are undoubtedly met with in which the infection is due to the paratyphoid bacillus and in which the agglutina- tion of the typhoid bacillus is stronger than the paratyphoid; while, on the other hand, cases of true typhoid may show a stronger reaction with bacillus para- typhosus than with the bacillus typhosus. According to Lentz, the time of such associated agglutinations is important in this connection. The para- typhoid bacillus is agglutinated more quickly by a paratyphoid serum than is the typhoid bacillus by the typhoid serum. A positive reaction, all conditions outlined above obtaining, indicates infection with the bacillus typhosus in practically all cases. This test can not be regarded as of great value in the early stages, as only a very small percentage of cases show early agglutination reactions. It is especially serviceable in the differential diagnosis from the second week of the disease on to convalescence. In the ambulatory types or in localized typhoid infections, especially in the gall-bladder, this test is of great value. A negative reaction should not be interpreted as excluding typhoid, as many typical and severe cases show no reaction until late in the disease.1 Like all laboratory tests, it must be used only in connection with the clinical findings, if the best results are to follow. The attending physician is the one to interpret the result, not the laboratory worker. Other organisms, beside the bacillus typhosus, show such agglutination re- actions. The technic is the same and the results comparable, the same specific- ity being shown. It is especially in the realm of infections with members of the colon group that these tests are, frequently, uncertain, yet, here also, careful vari- ation in the dilution of the serum and observation of the time and degree of ag- glutination will lead to a positive result, which will be of the greatest clinical value. The organisms, frequently identified by these reactions, are the paratyphoid bacilli, cholera spirillum, Shiga bacillus, bacillus of plague, malta-fever, and tuberculosis, and the diplococcus of pneumonia. 2. Diseases Other than Typhoid Fever. This reaction is based upon another fact well established in serum path- ology. It has been proven that the serum of an animal, injected with blood or blood-serum of another animal, shows the property, when added to an homolog- ous serum, of precipitating the albumin of this serum in the form of a light flocculent precipitate. This same peculiarity has been observed after injection B. Precipitin Reaction. 1 The widespread use of typhoid vaccine as a preventive inoculation has introduced a large possible error into this reaction. The serum of injected (immune) patients shows a positive Widal reaction very quickly. This condition persists for a varying period, so that blood cultures or Austrian's ophthalmo-reaction (see p. 581) should be used in such cases. THE BLOOD. 591 of exudates or transudates or of pure proteins both animal and vegetable. These reactions are, therefore, specific for the type of protein injected, within certain definite limits. Although this test has been used in the detection of certain albuminous substances excreted in the urine (see p. 249), yet its chief interest to the laboratory worker is its application to the medico-legal detection of blood and blood-stains. The work of Wassermann and Uhlenhuth has shown that blood-stains may reveal their origin from any particular species of animal by application of this test. It is evident that one must, of course, prove that such stains are really due to blood by other methods, as this biologic test is merely for specific pro- teins and not for blood as such. These latter tests will be given in a later sec- tion. In this discussion the writer quotes largely from Uhlenhuth and Weidanz.1 The Anti-serum. This is the serum of the animal immunized against the proteins of the blood, exudates or secretions of a particular species of animal. These proteins are usually those of the human as this type is of especial importance in medico- legal investigations. However, in the larger laboratories where many different serological tests are being frequently made, animals are kept on hand, which have been immunized against the proteins of most domestic animals, so that an absolute differentiation and identification of a stain may be made. If this be not done, the preparation of the animals for yielding such immune serum is a matter of too long a time to permit of such rapid examinations as many medico- legal cases demand. The preparation of the anti-serum is as follows: A rabbit is injected intra- peritoneally with 5 to io c.c. of whole blood or serum. At intervals of three to five days other similar injections are given until six or eight in all have been made. About a week after the last injection, a few drops of blood are drawn from a puncture in the ear of the animal and are allowed to clot in a small test- tube. The potency of the serum is then tested by permitting a few drops of the separated serum to fall into a small test-tube containing about i c.c. of a i to 1000 dilution of dried homologous blood in physiological salt solution. In this preliminary potency test as in the later precipitin test it is important that the dilution of the solution under investigation should be about i to 1000. One may recognize this proper dilution by the fact that the solution forms a layer of foam on shaking and, further, on heating and adding to i c.c. of the solution a drop of 25 per cent, nitric acid, only a slight opalescence obtains. If the addi- tion of the anti-serum to the 1 to 1000 dilution of dried blood causes a turbidity in 1 to 2 minutes, the serum is sufficiently active for the test and the animal may be bled. The blood is now withdrawn, either in small amount as desired for im- mediate use or the animal may be completely bled. The blood is collected in wide-mouth test-tubes, which should be then plugged with cotton and placed in a slanting position until the blood is coagulated. The serum, which must be 1 Prakt, Anleitung z. Ausfiihrung des biol. Eiweissdifferenzierungsverfahrens, Jena, 1909. See also Leers, Die forensische Blutuntersuchung, Berlin, 1910. 592 DIAGNOSTIC METHODS. clear and sterile, is removed to small test-tubes, which are plugged with cotton, sealed with paraffin, and stored in a cool place. This is the anti-serum. It should be sufficiently potent to produce a turbidity almost at once, at latest in one to two minutes, in a 1 to 1000 dilution of homologous blood in 0.9 per cent. NaCl solution. Solution of the Blood or Stain. As a preliminary precaution, it is to be emphasized that all vessels and instruments used should be thoroughly clean and sterile and that all fluids be absolutely clear. If the blood or exudate be dried upon solid material, such as glass, knife blades, stone, wood, etc., it is carefully scratched off with a sterile instrument. This material is powdered and dissolved, as far as possible, in chemically pure physiologic salt solution contained in a sterile test-tube. If only traces of the stain are present, one may form a well of wax about the spots and place in this some of the salt solution. It is absolutely essential that no other solvent than 0.9 per cent. NaCl solution be used, as otherwise erroneous reactions may occur or the usual precipitin reaction may not obtain. If the stain has penetrated clothing or other soft material, the spot is cut out with scissors, is finely divided, teased with a needle and placed in a small watch-glass or test-tube with a small amount of salt solution for one hour, or, if the stain be very old or has been exposed to marked changes in the weather, for 24 hours in a cool place. The fluid obtained in either of the above ways is filtered, first through hardened filter-paper (Schleicher and Schull, Nr. 575, 603, or 605) and then through a small Berkefeld filter. This filtrate, which must be perfectly clear should be neutralized to litmus either by the addition of sodium carbonate or tartaric acid solutions, as the case may be. Further, this filtrate should show distinct foam-formation on shaking. This is not always the case if the stain has been exposed to the sun and dust or has dried either upon fatty material or iron. Under these conditions, the precipitin reaction may not be seen in the usual time, but will obtain, as a rule, in five minutes. As previously stated, the dilution of this physiologic salt extract of the stain should be approximately 1 to 1,000. This may be tested by the formation of only a slight turbidity on the addition of 1 drop of 25 per cent, nitric acid to 1 c.c. of the filtrate. If this reaction be too marked, dilute with salt solution and test again. If sufficient stain can be scratched from the surface of the material upon which it is found, this may be weighed and proper dilution made. Such procedures are usually impossible, so that we use the above approximate method. The reaction is carried out in the small test-tubes, which should be ap- proximately of the same size and thickness, should be thoroughly clean and sterile, should be numbered and arranged in a series of io tubes in a small rack. In tubes i and 2 are placed 1 c.c. of the above solution of the stain to be tested. Technic of the Test. THE BLOOD. 593 In tube 3, 1 c.c. of a 1 to 1,000 dilution of known fresh human blood in physiologic salt solution. In tube 4, 1 c.c. of a 1 to 1,000 dilution of dried human blood in 0.9 per cent, salt solution. In tube 5, 1 c.c. of sterile physiologic salt solution. In tubes 6, 7, 8, and 9, 1 c.c. each of a 1 to 1,000 dilution of fresh or dried blood of such domestic animals as may be easily obtained, as chicken, dog, horse and sheep blood. In tube 10, 1 c.c. of a physiologic salt extract of the material upon which the stain was found. To each of the above tubes, with exception of tube 2, is now added from a carefully graduated pipet o. 1 c.c. of the anti-serum, which has been tested for its potency as above described. To tube 2 is added 0.1 c.c. of normal clear rabbit serum. The anti-serum must be carefully added in such a way that it runs down the side of the tube and does not drop directly into the fluid. After this addition the tube must not be shaken. The reaction is carried out at room temperature. If the reaction be positive, a distinct clouding or turbidity will be observed immediately, or at latest within two minutes, in tubes 1, 3, and 4 (all of which contain human blood), while the other tubes should remain perfectly clear. If the anti-serum has been properly added, the turbidity may be visible as a distinct contact ring which gradually increases upward. A negative reaction is shown by the lack of turbidity in any of the tubes. Specificity of the Reaction. Although this test is extremely delicate, yet it may happen that a stain, definitely identified as a blood-stain by other tests, may not react typically to the precipitin test, owing to the small amount of protein matter in the salt extract. This is especially the case if the protein has been rendered insoluble by heat or destroyed by chemical agents or by putrefaction. Even in such cases a positive reaction may occasionally be obtained, Lhlenhuth and Beumer reporting positive results after blood had been subjected to putrefaction for two years. As far as the influence of time upon this reaction is concerned, it may be said that blood-stains over 50 years old have been detected. The whole point, in this connection, is whether any trace of soluble protein remains. If so, the test will react positively, providing the dilution and the time of reaction be carefully regulated to suit the lessened concentration. Heterologous precipitin reactions do not occur if the conditions of the test are properly maintained. If more concentrated blood solutions are used and the amount of anti-serum increased, such erroneous reactions may obtain. It is true that confusing reactions may arise in the differentiation of the blood of the closely related animals such, for instance, as the horse and donkey; goat, sheep and ox; dog and fox; hare and rabbit; chicken and pigeon; and especially man and the anthropoid apes (orang-outang, gorilla and chimpanzee). However, 594 DIAGNOSTIC METHODS. the point of great medico-legal importance is that such close differentiations are rarely called for, as the question at issue is, usually, whether a given stain is or is not due to human blood. As the apes are relatively scarce in this country, confusion is seldom probable. This reaction may, therefore, be regarded as practically specific for the especial type of blood protein with which the immune serum reacts (all details of the test being properly regulated), only the proteins of the very closely related animals responding to the test in a way which might leave the slightest room for doubt as to the origin of a stain or of a protein. The substance on which the stain is found may, often, exert some retarding action on the test. Stains upon cloth or paper usually show prompt results, although those upon wall-paper may, at times, react atypically owing to the influence of the chemical of the dyes. Strong alkalies or acids prevent the reaction, but these are thrown out by neutralization before the test is applied. Tannic acid, derived from stains upon leather may prevent the reaction, al- though this is not always the case. Iron rust may retard but does not prevent the reaction. If the stain be upon soft clay, mortar, lime or fresh plaster so that an intimate mixture has occurred, the reaction will probably not obtain, according to Graham-Smith and Sanger, but if the stain be upon hard well- dried mortar no influence is seen, as demonstrated by Biondi. As previously stated, this active immune serum reacts promptly with the pro- tein elements of exudates, sputum, pus, urine, feces, nasal and bronchial secre- tions, vaginal, lochial and seminal fluids, etc., so that a positive precipitin reaction means merely that a protein is present, which has the same origin as that against which the animal furnishing the anti-serum was immunized. Whether or not this is a blood protein is, in no way, a part of the test. This must be determined by other tests. C. Complement Fixation Test. A full theoretical discussion of the points involved in this biologic test would lead me altogether too far afield. However, some of the more important facts, bearing upon it, must be mentioned in order to insure a proper understand- ing of the principles upon which the test is based. As previously stated, the serum of an animal, which has been injected with bacteria or with washed blood corpuscles of an animal of a different species, acquires the property of dissolving homologous cells through the development of antibodies or immune bodies, which are specific for the cells injected. This process is known as bacteriolysis or hemolysis. Such sera are known as lytic; bacteriolytic, if the action is directed toward bacteria, hemolytic, if toward blood cells. It is to be remembered, in this connection, that fresh blood serum of many species is hemolytic for the red cells of some, but not all, other species, the degree being relatively small. Moreover, we find that the blood of the same species may be markedly hemolytic for the homologous cells, as evidenced by the effects occasionally observed after transfusion. It is, therefore, evident that we may have in any serum both the immune and the natural hemolytic agents, the former being much the more effective. THE BLOOD. 595 It has been shown that hemolysis, by means of an immune serum, may be prevented by heating the serum to 55 or 56° C. for one-half hour, the process being known as inactivation. Addition of a small amount of a fresh isologous serum will restore (reactivate) the hemolytic activity. Hence, two separate factors must be concerned in hemolysis, the one a thermostable (heat-resisting) substance, the other a thermolabile (destroyed by heat) product. The first of these is called the amboceptor, the latter, the complement. Amboceptor is usually still active in serum which has been kept for a long period, wThile com- plement disappears from the serum in a short time even though it be kept on ice. The function of the amboceptor is to sensitize the cells, while that of the complement is to dissolve the cells so sensitized. Hemolysis is impossible without the combined action of both factors, either one singly having no effect. Complement can combine with the cell only through the medium of amboceptor, while this union does not take place unless amboceptor and antigen (the blood cells) are homologous. In other words, the cells can be hemolyzed only by homologous serum in the presence of complement. It is evident, on the other hand, that hemolysis must occur when the proper hemolytic trinity is present. The amboceptor and complement bear definite quantitative relations to one another in this process, an increase of one factor permitting the use of a less amount of the other. Complement may combine itself equally well with bacteriolytic or hemo- lytic amboceptor providing the homologous antigen be present. If, therefore, wre mix bacterial emulsions with the corresponding inactivated immune serum and complement, this latter must attach itself (become fixed) to the bacterial cell through the medium of the amboceptor, so that the later addition of hemo- lytic amboceptor and washed cells will not result in hemolysis, owing to the fact that the complement originally present has been fixed to such an extent that none is left to combine with the hemolytic amboceptor. This phenomenon, discovered by Bordet and Gengou, is known as "fixation of complement" or as "deviation of complement." It is decidedly a quantitative as well as a qualita- tive reaction. Had complement remained free or been added in excess of the amount necessary to act upon the bacterial cell, a certain amount of visible hemolysis must have occurred, and might, even, have been complete, had sufficient hemolytic amboceptor been present. These quantitative elements are the most important ones in the application of the test to be discussed later. This test has been used for diagnostic purposes in many infectious diseases; for instance, in typhoid fever by Widal and Lesourd as well as by Hirschfeld, in cerebrospinal meningitis by Cohen, in whooping cough by Bordet and Gengou, in scarlet fever by Besredka and Dopter and Foix and Mallein, in systemic gonococcic infections by Muller and Oppenheim, Bruck, Albarran and Jungano. Its most important and frequent application is in the sero-diagnosis of syphilis, as advanced by Wassermann, Neisser and Bruck and, independently, by Detre. It is evident that the application of this complement fixation test leads, 596 DIAGNOSTIC METHODS. just as in the case of the agglutination and precipitin tests, to the identification of an antibody, provided one has the homologous antigen, or vice versa, the antigen and antibody necessarily being homologous and, hence, specific, the one for the other. It had been shown by Klebs, Metschnikoff and Roux that syphilis could be transmitted to monkeys by inoculation with human syphilitic material (antigen). Wassermann and his coworkers then found that the serum of these syphilitic apes contained an antibody not found in normal serum. Having this antibody, it was an easy step to show its relation to syphilitic antigen and to apply the complement fixation test in the detection of antibodies in the serum of patients infected with syphilis. As the cultivation of the treponema pallidum has only recently been achieved by Noguchi, no direct method of detecting these antibodies is available.1 As indirect methods must be used, the antigen employed in the early studies was an extract of syphilitic tissues in the active stage of the disease, as these contain large numbers of the causative spirochete. Later workers have shown that this antigen need not be specific, as alcoholic extracts of certain normal organs, crude tissue lecithin, or, even, salts such as sodium glycocholate or oleate, may serve quite as well in the reac- tion. Although this fact lessens the true specificity of the test, from the biologic standpoint, there is no doubt of the clinical value of the reaction. The so-called syphilitic antigen must be either a normal substance or one possibly associated with an unknown product which actually produces a specific result. However this may be, it is a well established fact that, if the serum of a patient infected with syphilis be treated with antigen and complement, no hemolysis occurs on the later addition of hemolytic amboceptor and red cells because the complement is bound to the antigen through the medium of the homologous amboceptor; while if the patient is non-syphilitic marked hemolysis occurs as complement is free to act with the hemolytic system. This is the basis of the Complement Fixation Test as applied to syphilis by Wassermann and his colleagues. Al- though the antigen need not be specific, yet the test itself is remarkably specific being given by only a few non-syphilitic conditions as will be seen later. In the actual technic of the test, many modifications have been advised, the variations consisting either in the antigen or in the cells to be hemolyzed or in both. Five essential factors enter into the reaction whatever be the modifica- tion employed. These are (x) syphilitic antigen, (2) syphilitic antibody, (3) erythrocytes (antigen), (4) hemolytic amboceptor and (5) complement. These factors must be accurately adjusted to one another, as serious errors may arise if the quantitative relations are not observed. The writer will describe only the original Wassermann and the Noguchi methods, as these are, in his opinion, the most reliable ones to follow. a. The Wassermann Reaction Preparation of Antigen. Since the discovery of the fact that the antigen used in this test was not absolutely specific, in the biologic sense, for the syphilitic antibodies, many different antigens have been advocated. The writer selects those, which have 1 See foot-note on page 599. THE BLOOD. 597 promised the best results, although he believes that of Noguchi is by far the best and most reliable one introduced, especially if used in conjunction with a control test with one of the aqueous extracts. a. Aqueous Extracts. 1. Wassermann's Method.1 The liver or spleen of a syphilitic fetus is ground up in a mortar with sand or, preferably, minced in a meat grinder. The hash is placed in a dark colored flask and mixed with 0.9 per cent, sodium chlorid solution containing 0.5 per cent, phenol. The proportion is 1 gram of substance to 4 c.c. of the salt solu- tion. This mixture is shaken for 24 hours in a shaking machine and is then allowed to settle or is centrifuged. The reddish-brown opalescent supernatant liquid, which is the antigen, is decanted into dark bottles, tightly stoppered and kept on ice. On standing, a precipitate forms, which must not be disturbed by shaking. This aqueous antigen appears to be rather unstable and shows, sooner or later, some inhibition (anticomplementary action) with normal serum, although Citron2 finds such antigens active even after n months. Before use the activity of this extract, as well as of all others, must be quantitatively determined by the method later outlined. The usual doses are 0.2 (maximum) and o. 1 (minimum) c.c. As a general rule, we use only such an amount of antigen, the double dose of which shows no binding of complement. 2. Method of Marie and Levaditi.3 The liver of a syphilitic fetus is rubbed up to a fine pulp and dried in vacuum to a powder, which is very stable. As desired for use, extract this powder with physiologic salt solution in proportion of 1 to 4. Clear by centrif- ugalization and keep in small test-tubes on ice. This antigen gives much better results than the preceding after some time, as it is not so unstable. b. Alcoholic Extracts. The liver of a syphilitic fetus is hashed and placed in a bottle with a few glass beads. Nine c.c. of absolute alcohol are now added for each gram of liver pulp. Shake in a shaking machine for 24 hours and filter through filter- paper. The filtrate (antigen) is preserved in dark flasks and may be used in doses of 0.2 to 0.3 c.c. This extract may be, in itself, both hemolytic and anticomplimentary. If so in dosage used, the results are not to be relied upon. 1. Method of Bruck.4 The liver of a syphilitic or normal fetus is hashed and extracted for 24 hours with five times its weight of absolute alcohol. Filter through filter- 2. Method of Porges and Meier.5 1 Deutsch, med. Wochs., Bd. 32, 1906, S. 745. 2 Die Technik der Bordet-Gengouschen Komplementbindungsmethode. in Handbuch of Kraus and Levaditi, Jena, 1909, p. 1076; Die Methoden der Immunodiagnostik und Immunotherapie, Leipzig, 1910. 3 Ann. de 1'Inst. Pasteur., Tm. 21, 1907, p. 138. 4 Die Serodiagnose der Syphilis, Berlin, 1909. 6 Berl. klin. Wochs., Bd. 45, 1908, S. 731. 598 DIAGNOSTIC METHODS. paper and distill the alcohol from the filtrate in vacuum at 400 C. A syrupy residue remains which is suspended in physiologic salt solution containing 0.5 per cent, phenol, using 100 parts of salt solution to 1 of syrup. Filter through filter-paper. The usual dose of this antigen is 0.2 to 0.3 c.c., although less quantities may be used as determined by methods of standardization. Instead of this antigen, these workers use, also, a 1 per cent, solution of commercial lecithin in phenol-containing physiologic salt solution. This solution is stable if kept in the ice box. It has the disadvantage that the smaller doses may inhibit hemolysis, while the larger ones may, themselves, cause hemolysis. The stronger luetic sera require a larger dose (0.1 c.c.), while the weaker sera use less (0.01 c.c.) of lecithin. Hence we employ an average dose of 0.05 c.c., which may be compensated by the use of four to six times the titrated dose of amboceptor and complement. 3. Method of Landsteiner, Muller and Poetzl.1 The muscular tissue, free from fat, of a guinea-pig heart is rubbed up in a mortar and extracted with 95 per cent, alcohol, using 50 c.c. for every gram of substance. Heat for several hours at 6o° C. and filter through filter-paper. The filtrate is stable at room temperature for long periods. It is used in doses of 0.3 to 0.05 c.c. 4. Method of Michaelis and Lesser.2 Mince syphilitic or normal liver, add a few glass beads and shake with ten times the volume of absolute alcohol for several hours. Allow the mass to settle out for 24 hours, pipet off the solution and preserve in dark flasks in the ice box. When ready to use this antigen, dilute one part of the stock solution with 4 parts of physiologic salt solution and use 1 c.c. of this dilution for the test. A pre- cipitate forms when this emulsion is allowed to stand. As this precipitate is an essential part of the antigen, shake the mixture before using for the test. c. Acetone-insoluble Extracts. As it has been shown that the alcoholic extracts of various organs contain variable amounts of "syphilitic antigen," Noguchi3 advocates the use of the alcohol and ether soluble but acetone insoluble extract of organs. This frac- tion contains lecithin and other phosphatids. Method of Noguchi. Extract a mashed paste of liver, heart, or kidney of man, ox, guinea-pig, rabbit, or dog with 10 parts of absolute alcohol at 370 C. for several days. Filter through filter-paper and collect the filtrate. The latter is then brought to dryness by evaporation with the aid of an electric fan or in a vacuum at 400 C. Take up the dried residue with ether and allow the turbid solution to stand in a covered dish over night in a cool place. Decant the clear ethereal solution into a clean beaker and concentrate by evaporating the most of the ether. Mix 1 Wien. klin. Wochs., Bd. 20, 1907, S. 1565; Ibid., Bd. 21, 1908, S. 282. 2 Berl. klin. Wochs., Bd. 45, 1908, S. 301. 3 Serum Diagnosis of Syphilis, Philadelphia, 1912. THE BLOOD. 599 this concentrated ethereal solution with 10 volumes of pure acetone. Allow the precipitate to settle and decant off the supernatant fluid. This antigen is a light brownish precipitate which gradually becomes sticky on exposure to the air. This product usually has no hemolytic action upon human red cells, but may occasionally show this property. Further, it may, also, show an anti- complementary action. Three-tenths gram of this acetone-insoluble fraction is dissolved in 1 c.c. of ether and is mixed with 9 c.c. of methyl alcohol. If any precipitate forms or is left undissolved, remove it by centrifugation. This stock solution remains unaltered for a long period. For the actual test an aqueous emulsion is prepared by mixing 1 c.c. of the stock solution with 9 c.c. of physio- logic salt solution. This produces a clear opalescent solution containing 0.3 per cent, of the original lipoids. One-tenth c.c. of this emulsion is used in the test later described. This emulsion is kept on ice as it is unstable in this form.1 Antigen Paper.-Instead of keeping a stock solution of antigen as above described, Noguchi has introduced the use of paper impregnated with antigen for the test. These papers are especially serviceable to the general worker who may wish to employ this test but they do not hold their properties longer than three months. These are far inferior to the liquid antigen but may be used in an emergency. Weigh out about 1.2 grams of the sticky extract and dissolve in about 20 c.c. of ether. Have ten sheets of filter-paper, 10 by 10 cm. in dimen- sions laid one upon the other in a clean glass dish. Pour over these the lipoid solution and saturate the paper evenly. Separate each sheet as quickly as possible and lay flat on a clean sheet of unbleached muslin. Within ten min- utes the paper is ready for use. Before assigning the dimensions for each tube in the fixation test, the antigen paper should be titrated. This is done as follows: Cut the paper into equal width, say 5 mm., and use increasing lengths of this strip for standardization, starting with 1 mm., 2 mm., etc. The principle of this standardization will be discussed later. The strips may be marked in sections, each representing the required dimensions, and put into sealed tubes for preservation. The Syphilitic Antibody. The immune serum may be withdrawn from the patient by the method given under blood cultures. The blood, thus obtained, is allowed to coagulate in a large test-tube. The serum is then drawn off with a pipet into small test- tubes and inactivated by heating to 56° C. for one-half hour. In the writer's laboratory, the custom is to obtain the blood from a rather deep puncture in the ear, the lobe being milked until sufficient blood is obtained (2 or 3 c.c.). The blood is collected in small test-tubes, which are immediately centrifuged and the serum drawn off. If the blood is to be sent to the laboratory, our instructions are to allow the blood, obtained as above, to coagulate in a 1 Noguchi has recently (Jour. A. M. A., vol. 58, 1912, p. 1163) prepared a specific antigen from cultures of the treponema pallidum. The most interesting finding, based upon compara- tive studies with this antigen and with the lipoid antigens, is that there is no indication, in cases of active syphilitic orchitis, of a sufficient amount of specific antibodies to bind complement with the pallida antigen, although a strong positive Wassermann reaction obtains. See, also, Craig and Nichols, Jour, of Exp. Med., vol. 16, 1912, p. 336. 600 DIAGNOSTIC METHODS. small tube placed in the ice box. The serum is then drawn off by means of a sterile pipet and placed in a second tube, which may be sent to the laboratory. Specimens coming from a distance are usually unsatisfactory unless the serum is separated from the cells before shipment, as many such specimens hemolyze to a considerable extent before being placed in work. The serum is inactivated by heating to 56° C. for one-half hour either immediately after centrifugation or on receipt of the specimen. This inactivation is absolutely essential provided aqueous or alcoholic extracts are used as antigens, because these antigens contain various proteins, which are liable to give non-specific proteotropic fixation with active sera. Noguchi has shown that this property is lost if the serum be inactivated. More- over, with his antigen (acetone-insoluble lipoids) no such proteotropic fixation occurs, so that active sera may be employed in his test, although the inactivated serum works quite as well. In the Wassermann test, the corpuscles to be hemolyzed are those of the sheep. The blood may be obtained from an abbatoir or sheep may be kept in the animal rooms of the laboratory and bled from the ear as desired. The blood is received directly in a more than equal amount of 1 per cent, sodium citrate solution in physiologic salt solution or it may be defibrinated by shaking with glass beads or by whipping with wires or bristles. It is then centrifuged and the supernatant fluid drawn off. The cells are thoroughly mixed with physio- logic salt solution, again centrifuged and the clear fluid withdrawn. This process is repeated two or three times until the corpuscles are free from serum. The sedimented corpuscles are now drawn up with a graduated pipet and added to 19 volumes of salt solution. A 5 per cent, suspension is thus obtained, of which 1 c.c. is used in the test. If the centrifuge tube be calibrated, one may use this directly in making the suspension. Such suspensions may be used for two or three days, especially if kept in the ice box, but they should be discarded if any trace of hemolysis appears in the tube. Suspension of Red Cells. Hemolytic Amboceptor. The hemolytic serum is obtained from rabbits which have been injected with washed sheep corpuscles. If the blood is received from the abbatoir, it is usually contaminated to such an extent that it must be sterilized by heating to 6o° C. for one-half hour. The cells are obtained as above described and washed very thoroughly with salt solution. When free from serum, their volume is made up to that of the original defibrinated blood with physiologic salt solution. The immunization is best carried out by injecting a rabbit, intraperitoneally, with 2, 4, 8, and 12 c.c. of cells at intervals of four or five days. Ten days after the last injection a few c.c. of blood are obtained from an ear vein of the rabbit and the serum tested for its hemolytic effect upon sheep corpuscles. If it produces hemolysis in dilutions of 1 to 600 or more, the animal is bled from the carotid artery, the blood being collected in a series of sterile test-tubes. These are placed in the ice box, the blood allowed to coagulate and the serum is THE BLOOD. 601 drawn off. This serum is then inactivated and put into small sterile tubes, which should be sealed and kept on ice. This material keeps well, but its strength must be determined at frequent intervals. It is wise to inject several animals at the same time, so that one may surely yield a satisfactory serum of high hemolytic power. Complement. Normal guinea pig serum is used as complement in all modifications of this test. The animal is anesthetized and the blood drawn from the carotid, as above described, into sterile test-tubes in which it is allowed to coagulate. The serum is then drawn off with a sterile pipet, placed in small sterile test-tubes and kept on ice. As complement deteriorates rapidly (being of little use after 48 hours) the blood is, preferably, withdrawn by aspiration directly from the heart. The animal will usually survive the loss of 5 to 10 c.c. of blood, providing the heart has not been badly lacerated by repeated punctures. After two weeks the animal may be used for a fresh supply of complement. It is wise to have quite a number of these animals on hand so that fresh material may constantly be available. It is absolutely unnecessary to sacrifice an animal each time a few tests are to be made. When used complement is diluted 1 to 10 with physio- logic salt solution. Standardization of Reagents. This standardization must always be preliminary to the actual performance of the test. Everything depends upon the proper adjustment of the various factors entering into the test, so that especial care must be taken to find out just what quantities of each reagent to use. Without such precautions, little, if any, value can be attached to any report. Prepare a series of dilutions of the inactivated serum beginning with i to 300 and extending to 1 to 3,000. This may be readily accomplished by diluting 0.1 c.c. of serum to 30 c.c. with physiologic salt solution and using this 1 to 300 dilution as the basis of further dilutions. Arrange a series of small test-tubes and place in each 0.25 c.c. of these diluted sera, beginning with the highest con- centration and extending to the lowest. Now add 0.25 c.c. of a 1 to 10 dilution of complement and 0.25 c.c. of a 5 per cent, suspension of sheep corpuscles to each tube. Bring the total volume up to 1.25 c.c. with physiologic salt solution. All tubes are placed in the incubator at 370 C. for two hours and are then allowed to stand over night in the ice box. The highest dilution (smallest amount of amboceptor) in which complete solution of the cells occurs is taken as the strength of the serum. This is called one amboceptor unit. In performing the Wassermann test, two units of amboceptor are used. The titer of the ambocep- tor remains fairly constant, but it is wise invariably to determine its value when fresh complement is to be used, owing to the marked influence of these factors on one another. Titration of Hemolytic Amboceptor. 602 DIAGNOSTIC METHODS. Instead of using the above method of standardization, one may resort to the drop method of Landsteiner, Muller and Poetzl, which simplifies the technic but, in so doing, impairs the accuracy of the result, in the writer's opinion, as the drops from different capillary or other pipets are seldom of the same size. By this method one determines what dilution of amboceptor in 1 drop dose will hemolyze in one-half hour, in an incubator, 1 drop of 50 per cent, suspension of sheep's corpuscles with 1 drop of complement, 10 drops of salt solution being added as diluent. Twice the strength of the titer is used in the test. Titration of Complement. Having obtained the value of the amboceptor unit, one may use decreasing amounts of complement with one unit of amboceptor in the same way as in the above titration, the dilution of the complement running from 1 to 2 to 1 to 20. Complement is usually used in a dilution of 1 to 10, 1 c.c. being employed in the test. This is, however, open to objections as it is not absolutely quantitative. Noguchi has pointed out the fact that variation in the amount of complement used influences the amount of amboceptor. "If less than one unit of ambo- ceptor is used hemolysis will always be incomplete, even with more than one unit of complement. Likewise, if with one amboceptor unit there is combined less than one unit of complement, hemolysis can not be complete. If with more than one unit of amboceptor there be used less than one unit of complement, hemoly- sis may be complete or incomplete according to the relative amounts of each factor used.1 In the presence of many units of amboceptor hemolysis may be complete when but a small fraction of the complement unit is present." Titration of Antigen. The degree of dilution in which the antigen is to be used must be deter- mined for each extract by testing it both against normal and syphilitic sera. Its concentration must be such that it does not prevent hemolysis with normal sera, while it will promptly inhibit hemolysis in presence of complement with syphilitic sera. Set up two parallel series of tubes. Into each tube of one series place 0.25 c.c. of a 1 to 5 dilution of known syphilitic serum and into each tube of the second series place 0.25 c.c. of a 1 to 5 dilution of normal serum. Dilutions of the antigen are now made, with physiologic salt solution as the diluent, begin- ning with 1 to 1 and running up to 1 to 20 or higher if necessary. Into the first tubes of the above series place 0.25 c.c. of the first antigen dilution, into the second tubes of each series place the second dilutions and so on until all are accounted for. To all of the above tubes add 0.25 c.c. of complement (1 to 10) and incubate at 370 C. for one hour. Now add 0.25 c.c. of hemolytic ambocep- tor in twice the maximum dilution found to cause complete hemolysis (tw'o units) and 0.25 c.c. of a 5 per cent, suspension of sheep corpuscles. Place in an incubator for two hours and then in ice box over night. No hemolysis should be observed in the tubes containing syphilitic serum except in those containing 1 See Kromayer and Trinchese, Med. Klin., Bd. 8, 1912, S. 404 and 1670, who use an absolute minimum of complement in their "refined" Wassermann test. THE BLOOD. 603 the higher dilutions, while in the tubes with normal serum hemolysis will occur in all tubes except those with the lower dilutions of antigen. Hemolysis in the tubes with syphilitic sera is due to the normal hemolytic power of the antigen, while the lack of hemolysis in the tubes with normal sera is due to the anti- complementary action of the antigen. In carrying out the Wassermann test, antigen is used in one-half the lowest dilution (largest amount of antigen) which gives complete hemolysis with normal sera, this being taken as one antigen unit. Thus if a dilution of 1 to 5 gave complete hemolysis, a dilution of 1 to 10 would be used in the actual test. If one uses the drop method, as mentioned above, 1 drop of antigen (undi- luted) and 1 drop of complement (undiluted) are mixed with 10 drops of salt solution. The mixture is incubated one-half hour and 1 drop of amboceptor and 1 drop of a 50 per cent, suspension of corpuscles are added. Hemolysis should occur within one-half hour. If it does not, anticomplementary action is marked and less antigen must be employed in the test. Having determined the strength of the antigen, a portion of the stock solu- tion is diluted with salt solution so that 1 c.c. equals one unit. In testing the antigenic properties of this antigen, one may use it with a large number of posi- tive and normal controls, or, preferably, run it in parallel series with an antigen of known value. If the results agree, it may then be substituted for the old antigen. The titer being known, the antigen may be used for subsequent tests without redetermination of this value. Technic of the Test. Original Wassermann Method. All of the glassware used in the test must be kept thoroughly clean and dry, but need not be sterile. The pipets, which should be both graduated to o.oi and o.i c.c., should, preferably, be used with only one reagent. If they are employed with others, they must be thoroughly rinsed in salt solution between such applications. After using, wash them well in salt solution (never with soap and water), allow them to stand in a tall cylinder of distilled water for a few hours and dry them in an oven. The test-tubes used in this test are about 14 cm. in length by 1.5 cm. in diameter, while the somewhat smaller ones (1 by 10 cm.) may be employed in the Noguchi modification. Before actually performing the test, the reagents are prepared and stand- ardized as previously described or the amboceptor may be titrated at the begin- ning of the test, as it is not added until the preliminary incubation of the serum with the antigen and complement. The sera of the patient and of the known positive and normal controls must all be inactivated by heating to 56° C. for one-half hour.1 The complement is diluted 1 to 10 with physiologic salt solu- tion and the suspension of washed sheep corpuscles prepared. When these preliminaries are attended to one may proceed with the test itself. 1 Gramenitzki (Biochem. Zeitsch., Bd. 38, 1912, S. 501) and Fenyvessy (Ibid., Bd. 40, 1912, S. 353) have shown that if the serum be not heated to complete inactivation, the comple- ment is restored on allowing to stand for some time. 604 DIAGNOSTIC METHODS. Arrange a double row of test-tubes, the front tube of each set to receive the antigen and the back row the control tube without antigen. One set of tubes must be placed for the unknown serum, one set for the positive and one set for the normal control; while other controls are arranged as spoken of later. 1. Into both tubes of each set place 0.2 c.c. of the serum to be tested. 2. To all tubes, add 1 c.c. of complement (diluted 1 to 10). 3. To tubes in front row only, add 0.2 c.c. of antigen (aqueous extract of syphilitic liver). Make up volume of each tube to 3 c.c. with salt solution, mix contents by shaking and incubate at 370 C. for one hour. 4. Add to each tube 1 c.c. of antisheep amboceptor (two units). 5. Add to each tube 1 c.c. of 5 per cent, suspension of sheep corpuscles. Mix the contents of the tube thoroughly, incubate at 370 C. for two hours and place the tubes in the ice box over night. At the end of this period the results are read. Controls. i. Positive syphilitic serum. To be run as above. 2. Known negative serum. To be run as above. 3. A tube containing complement and two units of antigen, to prove that no ahticomplementary action of the antigen obtains. 4. A tube containing complement, to show that the hemolytic system is active. 5. A tube containing antigen alone, to show absence of lytic power of antigen. All controls are brought up to 3 c.c., incubated and, then, amboceptor and sheep corpuscles are added as above. If the test is properly conducted, hemolysis will occur in every control tube excepting in the tube containing syphilitic serum and syphilitic antigen (the positive control) and in the tube of control 5. In the series of unknown sera, no hemolysis will occur in the tubes containing syphilitic serum. This lack of hemolysis, partial or complete, is, of course, a positive reaction. Citron1 advises the use of two tubes for serum and antigen, one tube con- taining 0.2 c.c. of serum and 0.2 c.c. of antigen as given above, while the other tube contains 0.1 c.c. of serum and 0.1 c.c. of antigen. In this way the results may be represented somewhat more quantitatively, as the degree of inhibition of hemolysis varies according to the amount of syphilitic antibody present. If this amount be large, complete inhibition occurs in the tube containing 0.1 c.c. of serum and 0.1 c.c. of antigen as well as in tube No. 1, the result being ex- pressed as + + + +. If inhibition is incomplete in tube 2 but complete in tube 1, the result is + + + . Both of these degrees are strongly positive reactions. If tube 2 is completely hemolyzed while tube 1 is completely inhibited, the result is + +. Incomplete inhibition in tube 1 and complete hemolysis in tube 2 is expressed as +. These two latter results are classed as weakly positive reac- tions. If tube 1 shows doubtful inhibition while tube 2 shows complete hemo- 1 Handbook of Kraus and Levaditi, vol. 2, 1908, p. 1105. THE BLOOD. 605 lysis, the result is + . Complete hemolysis in both tubes is, of course, a negative result.1 This is not to be confused with Noguchi's test in which a human hemolytic system is used. The modification here suggested uses a different antigen and different quantities of the same reagents as employed by Wassermann. This is the method, followed in the writer's laboratory, when the sheep hemolytic system is used, a control being run with a specific watery extract as antigen in cases showing affections of the central nervous system. 1. Set up the double row of tubes as in the preceding test using 0.2 and 0.1 c.c. of serum to be tested. 2. To all tubes add 0.1 c.c. of an undiluted complement. 3. To front row only, add 0.1 c.c. of acetone-insoluble antigen. Make up volume of each tube to 1.5 c.c. with physiologic salt solution. Mix contents thoroughly and incubate at 370 C. for one hour. 4. Add to each tube 1 c.c. of antisheep amboceptor (two units). 5. Add to each tube 0.5 c.c. of a 10 per cent, suspension of sheep corpuscles. The total volume is now 3 c.c. Mix contents thoroughly and incubate at 370 C. for two hours. Allow to stand four to six hours at room temperature and read results. The interpretation is as given above. Noguchi's Modification. Drop Method. This is the method of Landsteiner, Muller and Poetzl and is frequently employed. The writer can not convince himself that this method is an accurate one owing to the possible variations in size of the drops, which are delivered from a capillary pipet. 1. Set up the double row of tubes as in the preceding tests and add to each 10 drops of physiologic salt solution. 2. Add 1 drop of serum to each of the two tubes used for the sera tested. 3. To front row only, add 2 drops of antigen. 4. To all tubes add 1 drop of complement. Mix thoroughly and incubate for one hour at 370 C. 5. Add 1 drop (two units) of standardized antisheep amboceptor to each tube. 6. Add 1 drop of a 50 per cent, suspension of sheep corpuscles to each tube. Incubate for two hours. Allow to stand at room temperature and read results. The method of interpretation is as above. It has been found that human serum contains a variable amount of natural antisheep amboceptor.3 This, of course, increases the effect of the immune Method of Noguchi.2 1 Ivy (Jour. A. M. A., vol. 59, 1912, p. 432) advocates the use of the Duboscq colorimeter as a means of estimating the exact degree of hemolysis, the control tube of each case being taken as the standard of 100 per cent. Ninety per cent, and over of hemolysis is regarded as negative. 2 Jour, of Exp. Med., vol. 11. 1909, p. 392; Serum Diagnosis of Syphilis, Phila., 1912. 3 See Dexter and Cummer, Arch, of Int. Med., vol. 9, 1912, p. 605; also, Bailey, Ibid., P- 551- 606 DIAGNOSTIC METHODS. antisheep amboceptor used in the tests previously described. One may, there- fore, obtain a negative Wassermann test under these circumstances even though a large amount of complement is fixed. Noguchi has overcome this possible error by introduction of a human hemolytic system to replace the sheep hemo- lytic system of the original test. He employs, also, a smaller amount of reagents and uses either fresh or inactivated serum of the patient. If preferred, one may use larger quantities of patients' serum, than outlined by Noguchi; if so, the reagents are increased in the same proportion. This method of Noguchi is the one preferred by the writer and is the one followed as a routine in his laboratory, as it has been found somewhat more reliable than the original Wassermann test in obscure cases.1 Collection of Serum. Noguchi obtains the patient's blood from a puncture on the ventral side of the last joint of the middle finger. The finger may be compressed by squeezing before the puncture and, afterward, massaged to press out the blood, this being collected in small tubes drawn out into capillary points. In the writer's laboratory the blood is obtained from a puncture of the ear, the lobe being milked and the blood allowed to fall directly into small sterile tubes. The blood is immediately centrifuged and the serum is drawn off. The serum may or may not be inactivated; if the former, a larger quantity (four times) is used in the test. Corpuscle Suspension. The suspension may be prepared with the blood of the patient or with that of the examiner. If with that of the former, it is especially important that every trace of serum be removed by careful washing. The standard amount of cor- puscle suspension is 1 c.c. of a 1 per cent, or 0.1 c.c. of a 10 per cent, suspension for each tube. If the patient's blood is to be used, enough may be obtained for many tests as follows: Fill a graduated centrifuge tube to 9 c.c. with 2 per cent, sodium citrate in physiological salt solution. Allow the blood to drop in until it fills the tube to 10 c.c. Mix and centrifuge. Pour off the supernatant fluid and fill up to 10 c.c. with fresh salt solution. Mix thoroughly and again centri- fuge. This may be repeated two or three times, in order to insure removal of every trace of serum. The deposited corpuscles may now be suspended either in 100 c.c. of salt solution (making a 1 per cent, suspension) or in 10 c.c. (giving a 10 per cent, suspension). This procedure, while thoroughly reliable, is not followed by the writer. It is our custom to obtain the corpuscles from a punc- ture of the finger of the wrorker, allowing 3 drops of blood to fall into 12 c.c. of salt solution in a centrifuge tube. This gives, approximately, a 1 per cent, suspension. The washing is done as above. This latter method enables us to obtain fresh suspensions whenever desired. 1 See Coca and L'Esperance, Arch, of Int. Med., vol. n, 1913, p. 84, for a slight modi- fication of the Noguchi technic. See, also, Miller, Interstate Med. Jour., vol. 20, 1913, p. 145. See Cecil and Lamb (Arch, of Int. Med., vol. 11, 1913, p. 249) for the value of this test in testing the serum of cadavers. THE BLOOD. 607 Hemolytic Amboceptor. This is prepared, as previously outlined, by injecting a rabbit, intraperi- toneally, with increasing doses of washed human corpuscles, instead of with sheep corpuscles as in the Wassermann method. After the animal is immu- nized, the blood is collected and allowed to coagulate. The serum is then with- drawn and titrated as previously described. Amboceptor Paper.-This has been introduced by Noguchi to simplify the test and to enable those not in close touch with clinical laboratory facilities to perform the test. While this paper may be used, the writer advises those doing any number of these tests to place their reliance more in the liquid amboceptor. The serum, obtained as above, is poured over sheets of filter-paper (Schleicher and Schiill's No. 597), cut into squares of 10 x 10 cm. Allow all the sheets to become evenly wet and absorb the excess with another sheet of paper. These are then dried at room temperature by placing each square separately upon a clean sheet of unbleached muslin. When dry, the sheets are cut into strips 5 mm. wide and are then standardized as follows: Take a series of tubes con- taining 1 c.c. of the 1 per cent, erythrocyte suspension and add to each tube 0.02 c.c. (as one unit) of complement. Now add the strips in increasing lengths, as 1 mm., 2 mm., 3 mm., etc., and incubate for two hours. The shortest strip causing complete hemolysis in this time represents one unit of amboceptor. These strips are then marked into sections of twice this length (two units) and cut off when the actual test is to be made. The papers should be kept dry and sealed. The papers on the market are quite variable and are not always to be relied upon. As previously stated, these papers should be used only when the liquid amboceptor is not at hand. Complement. Guinea-pig serum is used as complement in this system as in the others. The dilution is 40 per cent, made by mixing 1 c.c. of complement with 1 1/2 c.c. of physiologic salt solution. The method of obtaining it has been given above. Complement paper may be prepared but it is very unsatisfactory and should not be used if the liquid form can possibly be obtained. Antigen. The antigen, used in this system, is prepared as previously described (p. 598). The stock solution is a 3 per cent, methyl alcohol solution of the acetone- insoluble lipoids. From this a 0.3 per cent, emulsion is prepared for the test by mixing 1 c.c. of this stock solution with 9 c.c. of physiologic salt solution. It seems advisable always to control this antigen by running a control test with a watery extract of specific organs, especially if an affection of the central nervous system obtains. Technic. i. Arrange the double row of tubes with same controls as in the other tests. Place in all tubes i capillary drop (0.02 c.c.) of the fresh serum to be tested. If the serum has been inactivated use 4 drops (0.08 c.c.). One may use 0.2 c.c. of fresh (not inactivated) cerebrospinal fluid. 608 DIAGNOSTIC METHODS. ORIGINAL WASSERMANN TEST. Human serum Complement (guinea-pig serum i to io) Antigen >e to 3 c.c. with and incubate at Antisheep amboceptor two units Five per cent, suspension of sheep corpuscles r two hours. :e box id Hemolysis Unknown serum Unknown serum c.c. 0.2 o. 2 c.c. i i c.c. 0.2 each tut contents >ne hour c.c. i T c.c. i i te at 37° fo ) stand in it night. Re Depends on sera Complete. None. Complete. Complete. Complete. Complete. Complete. None. Positive control O . 2 I O . 2 I I Positive control O . 2 I ke up volume solution. M 37 I I 42 Negative control Negative control Anticomplementary control. Hemolytic control 0.2 0.2 I I I 0.2 0-4 I I I I I I I Mix. Incu Allow ove Antigenic control 0.2 NOGUCHI'S MODIFICATION (SHEEP HEMOLYTIC SYSTEM). Human serum Complement (guinea-pig serum undil.) c o .S3 c < c.c. O . I O.I O.I O.I 0.2 O.I Make up volume of each tube to i. 5 c.c with salt solution. Mix contents and in- cubate at 370 for one hour Ten per cent, suspension of sheep corpuscles Antisheep amboceptor. Two units Mix. Incubate at 370 for two hours. Allow to stand four to six hours at room temo. Read Hemolysis Unknown serum Unknown serum Unknown serum Unknown serum Positive control Positive control Negative control Negative control Anticomplementary control. Anticomplem e n t a ry control. Hemolytic control c.c. 0.2 0.2 O. I O. I 0.1 0.1 O . 2 0.2 c.c. O. I O. I O.I O. I O.I O . I O.I O.I O.I O.I O.I c.c. °-5 o-5 o-5 o-5 o-5 o-5 o-5 o-5 o-5 o-5 o-5 c.c. 1 1 1 1 1 1 1 1 1 1 1 Depends on sera. Complete. Depends on sera. Complete. None. Complete. Complete. Complete. Complete. Complete. Complete. NOGUCHI'S MODIFICATION (HUMAN HEMOLYTIC SYSTEM). Human serum Complement (guinea-pig serum i to 40) Antigen One per cent, suspension of h uman corpuscles ncubate for 7° Antihuman amboceptor for two hours, r temp, for ead Hemolysis 0 c.c. c.c. c.c. c.c. c Ri Unknown serum 0.02 0.1 0.1 I 2 units Depends on sera. Unknown serum 0.02 0.1 I 2 units 2 -o 0 Complete. Positive control 0.02 0.1 0.1 I 2 units None. Positive control 0.02 0.1 I 2 0 2 units g m 5 Complete. Negative control 0.02 0.1 0.1 I 2 units .2 2*" Complete. Negative control 0.02 0.1 I 2 units *2 £ A 0 Complete. Anticomplem e n t a r y 0.1 0.1 I § 2 units Complete. control. Hemolytic control 0.1 I 2 units Complete. 609 THE BLOOD. 2. Add to each tube o.i c.c. of 40 per cent, complement, or in a great emergency two units of complement paper. 3. To the tubes in front row of each set, add 0.1 c.c. of antigen emulsion. 4. To both tubes of each set, add 1 c.c. of 1 per cent, washed human corpuscles. Mix thoroughly and incubate for one hour at 370 C. 5. Add to each tube 2 units of antihuman amboceptor in liquid or paper form. Mix and incubate for two hours at 370 C. Allow the tubes to stand at room temperature for a few hours and read results. The results of this test are interpreted as in the preceding ones. This human system is more adjustable and reliable than is the sheep system. Nogu- chi has pointed out certain variable factors which may obtain and which may require a proper adjustment to insure proper results. These variables are, it is to be understood, greater in the sheep system, and are far less easily controlled. I quote directly from Noguchi. (1) One sometimes meets with instances in which hemolysis is complete within 10 to 20 minutes, and in which the positive control tubes with antigen undergo, sooner or later, gradual hemolysis. Such rapid progress of hemolysis at first mentioned is a sign of imperfect reaction. If the test is properly made, hemolysis proceeds gradually, and is complete in the water-bath within half an hour or thereabout. The causes of this acceler- ated hemolytic process are either an abnormally weak resistance of the blood corpuscles, or an exceptionally high activity or insensitiveness to fixation of the complement employed; or it may be the result of all these acting together.1 It happens occasionally that the serum of certain guinea pigs contains an abnor- mally active complement. In order to establish this point, and thus to remove this source of error, one has only to make the test with a quantity of comple- ment which corresponds exactly to two complement units. (2) There are sometimes encountered instances in which hemolysis remains incomplete even in the control tubes in which there is no antigen. Here the causes of the imper- fect reaction are found either in the weakness of the complement, or the ambo- ceptor used, or both.2 Usually the cause is the weakness of the complement, which, owing to its great lability, is likely to deteriorate. The activity of the amboceptor is far less subject to external influences which bring about its deterioration, and it is therefore extremely rare to find that the imperfection in the reaction arises from this source. In testing several specimens of serum at one time it happens occasionally that some specimens are slower in completing the hemolytic reaction than others. The cause of this slowness is not present in the complement or amboceptor, but in the specimens themselves.3 In such cases the specimens are found to contain anticomplementary substances which react with and reduce the activity of the complement. To remove this source of error, it is necessary to heat the serum to 550 C. for 20 minutes and use four drops for the test. The difficulty may be obviated in some cases by collecting specimens of serum to be tested just before meal-time because the anticomple- mentary substance is closely associated with the absorption of the chyle into the 1 See Bailey, Jour, of Exp. Med., vol. 15, 1912, p. 470. * See Barratt, Jour, of Path, and Bact., vol. 16, 1912, p. 363. 3 See Bronfenbrenner and Noguchi, Jour, of Exp. Med., vol. 15, 1912, pp. 598 and 625. 610 DIAGNOSTIC METHODS. circulation soon after the meal. (3) The quality and quantity of the antigen can also be sources of error. If one uses poor antigen, either there will be no positive reaction at all, or weak positive reactions will be entirely overlooked. If, on the other hand, an excessive amount of unfractionated crude antigen is employed, certain nonspecific weak reactions may become manifest, or a false positive reaction even may be obtained, as the result of the action of anticom- plementary substances sometimes contained in preparations of the antigen. These sources of error can be entirely excluded by choosing an antigen that has been carefully prepared and standardized. Diagnostic Value of the Wassermann Test. As stated above, the Wassermann reaction, or any of its modifications, is not a specific immunity reaction. While the antibody is, presumably, syphilitic in origin, the antigen is a physiologic lipoid and not a product of the treponema pallidum. In other words, the complement fixation test, as applied to the diagnosis of syphilis, is not the result of an interaction between homologous (specific) antigen and antibody. Nevertheless, we find that the reaction does have a certain specificity from the diagnostic, if not from the biologic, standpoint. As a general rule it may be stated, at the outset, that a positive Wassermann test, all proper precautions and controls being made, indicates syphilis in practically all cases. It is true that certain nonspecific conditions, especially leprosy and, occasionally, scarlet fever,1 give positive reactions, but the large majority of cases of other diseases, reported as giving positive results, are those in which syphilis has not been absolutely excluded. In such cases the syphil- itic infection may be latent instead of the active cause of the trouble at the time of making the test, yet it must be reckoned with in deciding as to the value of this test from the diagnostic point of view.2 On the other hand, it is to be distinctly understood that a negative Wasser- mann test does not exclude syphilis, a fact that must be taken into consideration when the question of marriage is involved. Such cases must be judged largely by the clinical manifestations and history rather than upon the negative char- acter of this test. Varying percentages of positive results in known syphilitic and parasyphilitic conditions have been reported by various workers. Some of the more enthusiastic ones claim that a negative Wassermann reaction means that the case is not syphilitic. This view, however, must be regarded as untenable in the face of the large numbers of negative results in the literature and in the daily experience of those working in this field. The following table, compiled from Boas3 and Noguchi4 shows the reported cases of syphilis with the percent- ages of positive results.5 1 See Kolmer, Jour, of Exp. Med., vol. 14, 1911, p. 235. 2 Field (Jour. A. M. A., vol. 58, 1912, p. 1681) has called attention to the occurrence of positive Wassermann tests in cases of lead poisoning. 8 Die Wassermannsche Reaktion., Berlin, 1911. 4 Serum Diagnosis of Syphilis, Philadelphia, 1912. 8 See Lucas, Amer. Jour. Dis. of Child., vol. 3, 1912, p. 259; Churchill, Ibid., p. 363; Kaplan, Med. Record, vol. 81, 1912, p. 1132; Bates, Arch, of Int. Med., vol. 10, 1912, p. 470; De Buys, Amer. Jour. Dis. of Child., vol. 5, 1913, p. 65; Milne, Amer. Jour, of Med. Sci., vol. 145, 1913, p. 197; Wolbarst, New York Med. Jour., vol. 97, 1913, p. 378; Grindon, Davis, Greiner and Weiss., Interstate Med. Jour., vol. 20, 1913, p. 221. THE BLOOD. 611 Condition Number of cases Positive results Primary syphilis 974 56.5 per cent. Secondary syphilis 2762 88.0 per cent. Tertiary syphilis 830 80.0 per cent. Early latent syphilis 1233 51.0 per cent. Late latent syphilis 861 47 .0 per cent. Congenital syphilis 125 94.5 per cent. Cerebrospinal syphilis 64 47 .6 per cent. General paralysis 498 88.1 per cent. Tabes dorsalis 360 70.0 per cent. Considerable discussion has arisen in the literature regarding the relative merits of the Wassermann and Noguchi systems. As a general rule it is found, by one running these two systems in parallel, that the positive results are some- what higher with the Noguchi system. The following table, taken from Noguchi, gives the figures for the cases reported, in which the two tests were made upon the same serum. Condition Number of cases Wassermann positive Noguchi positive Primary syphilis 208 88 per cent. 94 per cent. Secondary syphilis 669 92 per cent. 98 per cent. Tertiary syphilis 455 74 per cent. 83 per cent. Latent syphilis 305 54 per cent. 68 per cent. Congenital syphilis 79 98 per cent. 98 per cent. Cerebrospinal syphilis 55 73 per cent. 80 per cent. Under the influence of a vigorous course of mercurials, a positive Wasser- mann test may be made to disappear. This must, however, not be interpreted as meaning that the syphilitic infection has been cured. It is almost daily observed by those doing large numbers of these tests that a patient, having the test made while under active treatment with mercury, will frequently give a negative test; while the same patient may show a positive reaction if the mercury be withdrawn for two or three weeks prior to the test. On the other hand, we occasionally find cases in which the treatment has resulted in a positive Wasser- mann, prior tests having been negative. These facts must be borne in mind by those who may attempt to establish a diagnosis 'without full knowledge of the conditions under which the test is being made. Mercurial treatment may render the infection latent, so that a negative reaction may obtain until later active manifestations appear and become associated with a positive test. This variability has been shown by Craig,1 who reports 52 cases treated, with 32 positives and 20 negatives after treatment ranging from two weeks to twelve years. Patients well treated give negative reactions, while those receiving inadequate treatment, no matter for how long a period, show positive results. Effect of Treatment on the Reaction. 1Jour. of Exp. Med., vol. 12.1910, p. 726; Arch, of Int, Med., vol. 8, 19U, p. 395. Craig and Nichols (Jour. A. M. A., vol. 57, 1911, p. 474) and Craig (Ibid., vol. 60,1913, p. 565) call atten- tion to the influence of alcohol in converting a positive test into a negative one within 24 hours. 612 DIAGNOSTIC METHODS. The reaction is, unquestionably, affected by mercurial treatment, but some cases persist in giving a positive reaction in spite of treatment. The reaction may return shortly after treatment, while it is not, by any means, settled that the disappearance of a reaction justifies one in stopping treatment. After treatment with "606" (salvarsan, dioxy-diamino-arseno benzol), the results are variable. In promptly cured cases, a positive reaction may dis- appear within two weeks or it may require four to five weeks to obtain a negative result. A positive reaction may persist for considerable time and then disap- pear, reappearing again after an uncertain period. Cases are reported, just as with mercurial treatment, in which a prior negative is followed by a positive reaction on instituting the arsenic treatment. The exact status of this question must be regarded as unsettled. Many variations of all possible types may be observed so that no absolute statements may be made as to the probable in- fluence of treatment upon this reaction. Certain it is, however, that a negative reaction, following a positive one, does not of itself indicate a permanent cure.1 The complement-fixation test is, theoretically, applicable to any in- fectious disease. Practically, however, the results obtained have not been equally satisfactory in all types of infection or invariably reliable in the same type. This is largely due to variations either in the strains or activity of the antigens employed. The principle of the test is the same as described above, no matter what the condition under investigation. Attempts to use this method, as a diagnostic procedure, have been made in typhoid fever,3 tuberculosis,4 meningococcic infections,5 systemic gonorrhea,® infection with pyogenic cocci,7 cholera,8 pertussis,9 small-pox,10 and other infections, but the results have not been conclusive. In the majority of cases a positive reaction in these conditions indicates a focus of active infection somewhere in the system, provided every technical precaution be taken; while a negative result does not exclude these infections. Owing to the many un- certain factors, especially the antigenic ones, a much higher percentage of p. Diseases Other Than Syphilis.2 1 According to Nichols (Jour. A. M. A., vol. 58, 1912, p. 603) the Wassermann test should be repeatedly negative for at least one year after-treatment before a cure may be assumed. 2 See Ludke in Handbook of Kraus and Levaditi, Erganzungsband 1, 1911, S. 518; Zeitsch. f. klin. Med., Bd. 72, 1911, S. 545. 3 See Rodet and Fabre, Jour, de physiol, et de path, gen., Tm. 14, 1912, p. 86; and Kolmer Jour, of Med. Res., vol. 26, 1912, p. 373. 4 See Chevassu, Presse m£d., Tm. 20, 1912, p. 173; Mollers, Deutsch, med. Wochs., Bd. 38, 1912, S. 745; Oyuela, Bull, de la soc. d'etudes scient. sur la tuberc., Tm. 2, 1912, p. 71; Zweig, Berl. klin.Wochs., Bd. 49, 1912, S. 1845; Hammer, Munch, med. Wochs., Bd. 59, 1912, S. 1750. 8 See Sophian and Black, Jour. A. M. A., vol. 59, 1912, p. 527. 6 See Schwartz, Amer. Jour. Med. Sci., vol. 144, 1912, p. 369; Schwartz and McNeil, Ibid., p. 815; Lenartowicz, Tygodnik lekarski, Bd. 7, 1912, S. 368; Dermatol. Wochs., Bd. 55, 1912, S. 1179; Lespinasse and Wolff, Ill. Med. Jour., vol. 23, 1913, p. 26. 7 See Kolmer (Arch, of Int. Med., vol. 9, 1912, p. 220) for complement-fixation as applied to streptococcus antibodies. 8 See Amako and Kojima, Zeitsch. f. Chemother, Bd. 1, 1912, S. 94. 9 See Delcourt, La Pathol, infant., Tm. 9, 1912, p. 1. 10 Teissier and Gastinel, C. R. Soc. de Biol., Tm. 73, 1912, p. 264. THE BLOOD. 613 doubtful or even negative results obtain in known positive cases than one ob- serves when the test is applied to suspected syphilitic cases. Likewise one finds that this complement-fixation test has been used in the diagnosis of infection with various animal parasites. Thus, in echino- coccus disease,1 Weinberg and Parvu, Laubry, Lejars and Ghedini have ob- tained excellent results. In helminthiasis2 and trichinosis3 positive results frequently obtain. A further application of this test has been made to the diagnosis of pregnancy,4 the antigen employed being derived from extracts of the chorionic villi. This application has very little value as the results are too uncertain and variable. D. Sero-diagnosis of Pregnancy.5 Abderhalden has recently introduced a biologic test for pregnancy, which depends upon the presence of certain ferments found in the serum of pregnant animals. This test appears to be almost absolutely certain in its results and should prove invaluable to the gynecologist and obstetrician. It will be recalled that the injection of foreign protein into the system re- sults in the production of certain well-established biologic properties in the serum of the animal so injected. Among these properties one finds the forma- tion of precipitins, agglutinins, anaphylactogens, etc. Abderhalden has shown that the parenteral injection (subcutaneous, intraperitoneal or intra- venous) of foreign protein or carbohydrate brings about the appearance in the blood serum of proteolytic or amylolytic ferments, as an indication of the effort of the system to protect itself against the possible toxic effects of this non-hy- drolyzed material. Such protective ferments are apparently specific and are never found in the serum of normal (non-injected) animals. The blood will be seen, therefore, to have acquired definite digestive properties apart from that resident to a more or less extent in the leucocytes. A further step, taken by Abderhalden and his pupils, has been the demonstration that substances, 1 Parvu, C. R. Soc. de Biol., Tm. 72, 1912, p. 462; De Sandro, Rif. Med., Tm. 28, 1912, p. 313; Thomsen and Magnussen, Berl. klin. Wochs., Bd. 49, 1912, S. 1183; Hahn, Munch, med. Wochs., Bd. 59, 1912, S. 1483; Hertz, Deutsch, med. Wochs., Bd. 38, 1912, S. 1418; Barsony and Egan, Zeitsch. f. klin. Med., Bd. 76, 1912, S. 269. 2 Barit, Inaug. Dissert., Konigsberg, 1912. 3 Strobel, Mtinch. med. Wochs., Bd. 58, 1911, S. 672; Romanovitch, Ann. de 1'inst. Pasteur, Tm. 26, 1912, p. 351. 4Fieux and Mauriac, C. R. Soc. de Biol., Tm. 68, 1910, p. 848; Ann. degynec. et d'obst., Tm. 39 (2. S. 9), 1912, p. 257; Jour, de med. de Bordeaux, Tm. 17, 1912, p. 184; Graff and v. Zubrzycki, Mtinch. med. Wochs., Bd. 59, 1912, S. 574; Quintella, Ann. de gynec. et d'obst., Tm. 39 (2 S. 9), 1912, p. 408. 5 See Abderhalden, Freund and Pincussohn, Prak. Ergeb. d. Geburtsh. u. Gynak., Bd. 2, 1910, S. 367; Abderhalden, Schutzfermente des tierischen Organismus, Springer, Berlin, 1912; Abderhalden, Handb. d. biochem. Arbeitsmeth., Bd. 5, 1911, S. 575 and Ibid., Bd. 6, 1912, S. 223; Abderhalden and Kiutsi, Zeitsch. f. physiol. Chem., Bd. 77, 1912, S. 249; Abderhalden, Ibid., Bd. 81, 1912, S. 90; Ibid., Bd. 82, 1912, S. 109; Mtinch. med. Wochs., Bd. 59, 1912, S. 1190, 1305,1939, and 2172; Deutsch, med. Wochs., Bd. 38,1912, S. 2160; Abderhaldenand Weil, Berl. Tierarztl. Wochs., Bd. 28, 1912, S. 665; Frank and Heimann, Berl. klin. Wochs., Bd. 49, 1912, S. 1706; Franz and Jarisch, Wien. klin. Wochs., Bd. 25, 1912, S. 1441; Veit, Zeitsch. f. Geburtsh. u. Gynak., Bd. 72, 1912, S. 463; Petri, Zentralbl. f. Gynak., Bd. 37 ,1913, S. 235; Schwarz, Interstate Med. Jour., vol. 20, 1913, p. 195; Henkel, Arch. f. Gynak., Bd. 99,1913, S. 56; Lindig, Munch, med. Wochs., Bd. 60, 1913, S. 288; Abderhalden, Ibid., S.411 and 462. 614 DIAGNOSTIC METHODS. which are native to the system but foreign to the blood, may arise from physio- logic or pathologic changes within the system itself and, after absorption into the blood, produce quite as definite a response as if the material were introduced from without. Schmorl and Veit have shown that syncytial cells and portions of the chori- onic villi may become detached from the placenta and enter the maternal cir- culation, while other placental products (of an unknown nature) presumably also are either absorbed or washed into the blood current. As these sub- stances are, in part at least, protein in nature, we should have, if the theory holds, a production of protective proteolytic ferments in the maternal blood, which are capable of digesting placental protein and are more or less specific. If, therefore, the serum of a suspected animal be treated with placental protein, hydrolytic cleavage must occur with the formation of products capable of more or less easy detection, provided the serum contains such ferments. These are the principles which form the basis of the test as originated by Abderhalden. Two distinct methods of detecting the presence of these ferments have been devised. The first, the optic method, is capable of very wide application to the diagnosis of different conditions and should prove extremely useful in solving many problems of great clinical importance. It requires, however, considerable skill and technical ability as well as rather expensive apparatus. The second, the dialyzation method, is much simpler both in technic and in necessary equipment. Both of these methods require the most assiduous at- tention to the various details given, if any dependence is to be placed upon the results of the tests. The basis of this test is as follows. A solution of placental peptone in physiologic salt solution has a definite power of rotating the plane of polarized light. Likewise, the serum, both suspected and normal, has a similar action. The degree of rotation, however, of either remains permanent for some time at 370 C. If a solution of peptone and normal serum be mixed and the degree of rotation of this mixture determined, no appreciable change will be observed between the initial and final polarimetric readings. If, however, a solution of placental peptone, whose polarizing action is known, be treated with a serum containing the specific ferments above mentioned, digestion of the peptone occurs with the formation of products showing rotatory powers sufficient to change the initial rotation of the mixture to quite an extent. These changes may be observed at different intervals and interpreted as described later. 1. The Optic Method. The fresh placenta is made blood-free by cutting it into small pieces and placing these under running water for about 15 minutes. Dry the pieces be- tween folds of filter paper and place them in about 5 times their weight of 70 per cent, sulphuric acid. Allow the mixture to stand for 3 days at room tem- perature and shake the container frequently. At the end of this time, place Preparation of Placental Peptone. THE BLOOD. 615 the container in ice-water and dilute the contents with io volumes of distilled water, stirring constantly. Remove the sulphuric acid by adding approxi- mately the calculated amount of finely powdered barium hydrate and complete the precipitation with a known solution of this salt, stirring the mixture con- stantly. When the reaction of the mixture becomes neutral to litmus paper, filter off the barium sulphate. If the filtrate be turbid, refiltration is necessary until a perfectly clear filtrate is obtained. The separation of the barium sul- phate is much facilitated by the use of the large centrifuge, if such be at hand. Wash the precipitate with a large amount of cold water and combine the fil- trate and washings. Test the mixture for both barium and sulphuric acid. If either be present, it must be removed. Now evaporate this barium- and sulphuric acid-free solution to dryness on the water-bath, under reduced pres- sure at a temperature not exceeding 40 or 500 C. It is wise to test the evapo- rated material at several intervals for the presence of either barium or sul- phuric acid, as these sometimes appear on concentrating the mixture. It is important that these be removed, as their presence will result in further hydro- lysis of the peptone and, in consequence, will lessen the value of the final prod- uct. A thick yellow syrup or a foamy mass remains after this evaporation. The product may be used in this form, but it is preferable to purify it, if reliable results are to be invariable. This yellowish residue is dissolved in methyl alcohol with the aid of heat and the hot solution is poured into absolute ethyl alcohol. The peptone is thrown down as a yellow powder, which is soluble in water to a clear yellowish solution of weakly acid or amphoteric reaction. This powder is not hygro- scopic. A further purification is still advisable. Dissolve the above yellow powder in water up to a 5 per cent, solution and add 10 per cent, solution of phosphotungstic acid as long as a precipitate forms. Filter and wash several times with water. Rub up this precipitate in a mortar with some water and twice its weight of barium hydrate. Filter again and remove the excess of barium from the filtrate with sulphuric acid. Filter off the barium sulphate and evaporate the filtrate to dryness under reduced pressure at 40 to 500, as outlined above. This product is snow white and is permanent. It is absolutely essential for the successful application of the optic test that the placental peptone be as pure as possible. The same product is not always obtained by the above method, as the hydrolysis may proceed further than the peptone stage. Such products are unsuitable for the test. It is wise, therefore, to work as quickly and as carefully as possible with a large amount of placental substance, so that one may obtain an appreciable amount of pla- cental peptone. If the product be found serviceable, it may be kept for years. A further point to be considered in the use of a prepared peptone is that the solution of this placental product must give absolutely no turbidity with the serum to be tested. Such a finding is not infrequent, owing to the probable presence of precipitins in the product. Such a peptone cannot be used. A further important property of the prepared peptone must be its power 616 DIAGNOSTIC METHODS. of rotating the plane of polarized light. The degree must not be too small or the product will prove of little value. It will be seen, therefore, that the prepa- ration of a serviceable and proper placental peptone is a matter of considerable difficulty and is paramount to the successful performance of the test. One may preserve the peptone, prepared as above, either in the solid state or in the form of a solution. The advantage of a solution is that one has on hand a large amount of material, which will give good comparative results, as the solution is permanent. Abderhalden formerly used solutions of 0.5 to 2.5 per cent, strength. He now advises a 10 per cent, solution of the placental peptone in physiologic salt solution. This solution must be absolutely clear and colorless. If not, filter through thick paper or a Berkefeld. Preserve this clear solution by overlaying its surface with toluol. When required for the test, the solution is withdrawn by a pipet dipping below the toluol. If care be taken to keep a layer of toluol over the solution, the stock material will be per- manent for a long period. Should this solution become turbid at any time, the material should be thrown away and a new stock solution prepared as above. The optical activity of this stock solution must be tested before each test. Obtaining The Serum. The serum of the patient is obtained as described under the complement- fixation tests. The venous puncture method is far preferable. Place the blood, as drawn, directly into a sterilized centrifuge tube, so that all cellular elements may be completely separated. For a successful test the serum must show no sign of the presence of cells. A further precaution to be taken is that no sign of hemolysis must be present in the serum. For this reason, the cells should be separated rapidly. It is wise to make the test on the same day on which the blood is taken, although, if the precautions above mentioned are observed, a delay of 24 to 48 hours does not materially affect the activity of the serum. Technic. Having prepared the io per cent, solution of placental peptone and having proven that it answers all the requirements above mentioned, place i c.c. of this clear solution (withdrawn by a pipet) in a small, clean, sterile test-tube. Add 2 c.c. of the clear suspected serum and shake the tube several times. Ex- amine the mixture carefully for any turbidity or precipitation. If any be ob- served, the test cannot be carried out. Add sufficient physiologic salt solu- tion to the mixture to fill the i decimeter polarimetric tube (see below). This mixture with salt solution is preferably made in this way, rather than to add the salt solution after the peptone solution and serum have been placed in the polarizing tube. Any turbidity may be much more easily detected. Pour the abpve mixture into the i decimeter tube, whose mantle has been filled with water at 370 C. Carefully determine the initial rotatory power of the mixture, checking the readings several times and controlling them by subsequent ones after 5 or 10 minutes. No change should be observed in these readings. Place the tube and THE BLOOD. 617 its contents in the incubator at 370 C. and repeat the readings every hour for a few periods and then continue every 6 or 8 hours. Do not extend the inves- tigation over more than 48 hours. Record all readings and interpret them as given below. Control tubes must be arranged as follows: (1) the peptone solution alone; (2) the suspected serum alone; (3) peptone solution plus normal serum; (4) peptone solution plus known positive serum; (5) peptone solution plus inacti- vated (heated to 6o° C.) suspected serum. In all of these controls the same conditions must be maintained and the same length of polarizing tube must be used as in the test itself. If any turbidity occurs in any of the control mix- tures or tubes, these must be disregarded in interpreting the test as turbid solutions give variable results with the polarimeter. It goes without saying that this test requires the very best equipment possible. The cheap polariscopes are absolutely useless as they are not deli- cate enough to detect the fine variations given. The three-shadow instrument of Landolt-Lippich, made by Schmidt and Haensch, is especially to be recom- mended. The polarizing tubes used are, preferably, the 1 decimeter tubes which are furnished with a mantle which may be filled with water at any desired temperature. If different length tubes are used in any of the tests, a correction must be made in order that comparative figures may be obtained. In performing this test, even to a greater degree than when the instrument is used in other work, much depends upon the ability of the worker to detect slight variations in the degree of the rotatory powers of the mixtures under investigations. The method is easy to learn, but the special sensibility toward such changes cannot be taught. Abderhalden cautions any one, who shows a working error of as much as 0.040 in his observations, against attempting to interpret the test. In reporting the result of this test, Abderhalden employes the following method. Deviations within 0.040 Negative Deviations between 0.05 and 0..10 Positive (+) Deviations between o.n and 0.20 Positive(++) Deviations over 0.2 degree Positive (+ + +) 2. The Dialyzation Method. This method is much more simple than the optic method, both as regards technic and apparatus. It must not be thought, however, that any less care is necessary in carrying out the details of the test. In fact, erroneous results are, perhaps, more easily obtained by careless manipulation when this method is employed. The basis of this method is the conversion of the colloidal non-dialyzable placental protein into dialyzable products through the activity of the ferments above mentioned. These products are, then, detected by simple color reactions in the dialysate. 618 DIAGNOSTIC METHODS. Preparation of Placental Albumin. Although Abderhalden does not regard the proteolytic ferments of the serum in pregnancy as absolutely specific for a given species of animal, he, nevertheless, insists that the protein preparations used in either of his methods be prepared from the placenta of the same species as that of the animal whose serum is to be tested. As autolysis proceeds fairly rapidly in placental tissue, the albumin should be prepared from placentas which are as fresh as possible. Remove the ex- ternal portions of the placenta, such as the membranes, and wipe away as much blood as possible. Cut the material into small pieces and wash for a short time in running water. While this is being done, boil about 2 liters of water to which are added 2 drops of glacial acetic acid. Throw the washed bits of placenta into this boiling water and boil for 5 to 15 minutes. Pour the mixture upon a loose quick-acting filter and boil the pieces again with a second portion of acidulated water for 5 to 15 minutes. Pour off this water and test with the triketohydrinden hydrate reaction given below. If a positive reaction obtains, the placental tissue must be again boiled with acidulated water until a negative reaction occurs. The essential points in this process are rapid and complete coagulation of the placental albumin and the removal of all soluble dialyzable material which may react with the reagent mentioned above. As soon as a negative result is obtained with the extractive water, pour the mixture into a wide-mouth flask, add some chloroform, overlay the fluid with toluol and stopper the flask; or place the material in several smaller glass jars and overlay with toluol. This placental albumin keeps almost in- definitely and may be removed from the containers as desired. It should be tested, from time to time, to show that it contains, in itself, nothing which may react with the reagents used in the later test.1 Obtaining The Serum. The serum is obtained by venous puncture as previously described, the blood (about io c.c.) being drawn directly into a sterilized centrifuge tube. The cellular elements are separated as quickly as possible and the serum drawn off into a clean sterile tube. It is of especial importance in this test that the serum show no sign of hemolysis. As it has been shown that amino- acids are present in the blood during digestion and may, therefore, give a posi- tive reaction with triketohydrinden hydrate, it is wise to take the blood in the morning before breakfast in all cases. If such be not done, a less amount of serum must be used in the later test to compensate for this possible error. Selecting The Dialyzing Tubes. It is evident that this part of the preparation for the test is of extreme im- portance. The dialyzing thimbles must be permeable for peptone but not for albumin. Unless these conditions obtain, the test is valueless. Not all of the thimbles, purchasable upon the market, are by any means available. Abder- 1 Abderhalden reccommends that this placental albumin be tested with ninhydrin before being used in any test. This is a vital point. THE BLOOD. 619 halden advises the use of the diffusion shells No. 579 of Schleicher and Schiill. Not all of these will answer the purpose. It is necessary, therefore, that all the dialyzing thimbles used in the test should have been previously tested and known to answer the above require- ments. As the thimbles are usually dry and hard when obtained, soak them in cold water for a few hours, place them in boiling water for a few minutes and keep them in water covered with toluol. To test these thimbles for their permeability for albumin, proceed as follows. Remove the thimble from the water and place in it 5 c.c. of serum or of a solution of egg albumin. Add a few drops of toluol to prevent bacterial action. Place 20 c.c. of distilled water in the dialyzing vessel and overlay this with toluol. This dialyzing vessel should be quite narrow, the distance be- tween the wall and the thimble (when in place) being about 1/4 cm. These vessels are kept plugged with cotton and are sterilized before use. Now sus- pend the thimble with its albuminous contents by means of a loop of thread in such a way that the fluid outside is as high or, preferably, a little higher than that within the thimble. Plug the vessel with cotton to hold the thread in place and put the apparatus in the incubator at 370 C. for 18 to 24 hours. At the end of this time, test the dialysate (outside fluid) for albumin by the biuret or triketohydrinden hydrate reactions given below. Those thimbles giving negative results are retained to be tested for their permeability to peptone. The shells permitting the passage of albumin cannot be used in the test. Select those thimbles showing impermeability to albumin and wash them thoroughly in water. Place in them 5 c.c. of a 1 to 1000 solution of Witte's or, preferably, peptone from silk (peptone La Roche) and add sufficient toluol to cover the solution. Dialyze as above against 20 c.c. of distilled water, placing the apparatus in the incubator for 18 hours at 370 C. Those thimbles which permit the passage of peptone, as shown by the triketohydrinden hydrate test, are then kept for use in these tests and the non-permeable ones are laid aside. The properly tested and selected thimbles are then preserved in water overlaid with toluol. Technic. Remove a few pieces of the coagulated placental albumin from the con- tainer, wash in distilled water and dry between filter paper.1 Break this up into very small bits or grind up in a mortar. Weigh out three portions of i gram each. Place i gram in each of three tested dialyzing thimbles in such a way that none of the material touches or remains upon the top or outside of the shells. Carefully wash off the outside of the thimble by means of a stream of distilled water or hold it under running water. This is done to remove any possible adhering placental albumin, which would vitiate the test later made. Now add to tube No. i, 2 or 3 c.c. of clear hemoglobin-free serum to be tested. This serum is withdrawn from its container by means of a sterile graduated 1 Test the material before use with the triketohydrinden hydrate reaction. Absolutely no trace of a blue coloration should obtain. 620 DIAGNOSTIC METHODS. pipet. Overlay the surface of the mixture in the thimble with toluol. This thimble is then placed in a sterile dialyzing tube, as described above, containing 20 c.c. of distilled water, which should stand slightly higher than the fluid in the thimble. Overlay the external fluid with toluol and plug the dialyzing vessel with cotton to hold in place the thread which suspends the shell. Place the apparatus in the incubator at 370 C. for 18 hours and then test the dialysate for peptone as outlined below. Controls. Charge thimble 2 with 1 gram of placental albumin and 2 c.c. of serum of a known positive control. Overlay with toluol and arrange as above. Charge thimble 3 with 1 gram of placental peptone and 2 c.c. of a known negative serum or with the nactivated (heated to 6o° C.) serum used in the test. A further control should be run, using 2 c.c. of the serum alone, without the addition of peptone, to prove that it does not contain any dialyzable sub- stances which will give the later reactions. The tests with all of these controls are carried out exactly as the test it- self, every precaution being taken to prevent the introduction of errors. The tests for cleavage products of albumin are made with one of the following tests, the latter being in some respects preferable. This was the test formerly employed by Abderhalden and has some ad- vantages in that it does not react with certain dialyzable products not infre- quently present in the serum of normal subjects. It requires considerable care and skill in manipulation as well as in interpretation. Doubtful results are very frequent unless every precaution be taken. Remove about io c.c. of the dialysate by means of a pipet dipping below the toluol. Place this in a test-tube and add 5 c.c. of a 33 per cent, sodium hydrate solution. Mix by careful shaking and add very carefully, drop by drop, from a buret a very dilute (0.25 per cent.) solution of copper sulphate in such a way that a distinct contact ring is formed. If peptone is present, a violet-red to a pure red contact ring will be observed, sharply differentiated from the lower colorless and upper blue solutions. This is a positive reaction. A negative result is shown by the appearance of a distinctly blue ring. It is not the simplest matter to distinguish between the various shadings which occur, so that one must not make his decision without having had some experience in differentiating the colorations obtained with pure albumin and peptone solutions. The Biuret Test. The Triketohydrinden Hydrate Reaction. This reagent occurs in colorless crystals readily soluble in water. It may be obtained from the Hochst-Farbwerke (Lucius & Bruning) Company under the trade name of "ninhydrin." Its formula is C6H4<^^^^/C(OH)2. It is of especial importance that every precaution be taken to prevent error when this reagent is used as it reacts with all substances having an amino THE BLOOD. 621 group in the a position to a carboxyl group. It is, therefore, not a specific reagent for peptone, as albumin and amino acids will also react. If any of the original placental albumin be left on the outside of the dialyzing thimble, a very obvious error will arise. So, also, if the serum used contains amino acids, as it may during digestion, a further error is introduced as these acids are dialyzable. Hence, he importance of taking the specimen of blood as far as possible (before breakfast) from the periods of digestion. A further precaution may be taken to use not more than 2 c.c. of serum to be tested. It will be seen, therefore, that this reagent is considerably more delicate than is the biuret test. One should be certain that the final extractive water, as well as the placental albumin itself, shows no trace of reaction with this reagent before using it in the test. A still further precaution to be taken in working with this reagent is to avoid the presence of acid or alkaline fumes in the lab- oratory, as these interfere markedly with the reaction. Moreover, all vessels and pipets used in the test must be absolutely clean. Remove 10 c.c. of the dialysate by means of a pipet dipping below the toluol and place this in a large test-tube. Add 0.2 c.c. of a 1 per cent, aqueous solution of triketohydrinden hydrate. Heat rapidly to the boiling point and keep the mixture boiling for one minute. If the reaction be negative, the solution remains colorless or becomes, at most, light yellow. If the reaction be positive, a deep blue color will appear either immediately or on allowing the tube to stand for a short time. The reaction is carried out in the same way with the control tubes. Tube 2 should show a distinct positive reaction, while tube 3 should give a negative result. The tube with serum alone should show a negative reaction, but oc- casionally it is positive owing to the presence of a large amount of amino acids in the serum as drawn. If the controls are all positive, the test is, of course, valuless, as some factor has been imperfectly controlled. Value of The Test. This test, applied by either one of the above methods, is extremely valu- able. Abderhalden's results, which have been definitely confirmed (see bibli- ography at beginning of this discussion), show that in every case in which the reaction was positive the patient has been proven to be pregnant; while nega- tive reactions have been given only in non-pregnant cases. In other words, a negative result is definite, a point of distinction from a negative result with the complement-fixation test; while a positive result is evidence of the presence of placental elements within the body (pregnancy), either at the time of the test or at least within a period which does not exceed two weeks (Schwarz). In the absence of a true pregnancy, hydatidiform mole and other pathologic conditions of the chorionic villi may be accountable for a positive reaction. As this reaction is evidence of pregnancy, in practically all cases giving positive results, it is especially interesting to find that it is positive from the middle of the second month of pregnancy on and disappears within io to 15 days after the termination of the pregnancy. It is, therefore, of great value in 622 DIAGNOSTIC METHODS. early diagnosis as well as in the diagnosis of a second pregnancy following so closely upon a previous one that other signs are not present. Differential diagnosis should be greatly improved by the use of this test and fewer laparot- omies performed under the mistaken diagnosis of tumor. A further value of the test is found in the fact that medico-legal cases maybe certainly established on a firm basis of absolute knowledge of the conditions present. IX. Medicolegal Aspects. In medicolegal examinations one is frequently called upon to determine the identity of various suspected stains. Such examinations may tax the entire re- sources of the worker and have, in many cases, led to absolutely negative results. It is not sufficient at the present day that the examiner state that a certain stain is blood, but he must be prepared to say what kind of blood it is. This latter point has been made possible by recent work so that one is fairly sure of lim- iting his statement at least to the blood of very closely associated animals. (1). Red Cells. It is not always a simple matter to determine that a stain on cloth, wood, iron, etc., is really due to blood. The color of the stain may be of any hue from dull red to a dirty gray, depending upon its exposure to various elemental conditions. The action of various substances, such as mortar, brick, or lime in any form, strong acids and alkalies, leather, chemicals in wall-paper, starched clothing, etc., may so change the blood or its reactivity to certain tests that no definite conclusions can be obtained. It is self-evident that the demonstration under the microscope of blood- cells is the surest proof of the presence of blood, the size and shape of the corpuscles frequently giving a clew as to the source of the blood. It becomes, therefore, necessary to make a suspension of the stained material preferably in isotonic (0.9 per cent.) sodium chlorid solution. The degree of solubility of the stain will depend upon the age of the stain, the heat to which it has been subjected, the amount of sunlight or of moisture to which it has been exposed, and the material upon which the stain is formed. Various fluids have been advanced, from time to time, as dissolving agents for the stains, many of these are active laking or hemolytic agents, especially for dried red cells, so that one sees in such preparations only the shadow of the red cell. Such laking agents are distilled water, 0.85 per cent, ammonium chlorid solution, 50 per cent, glycerin solution and many others. Frequently a strong 30 per cent, potassium hydrate solution will enable the worker to obtain a very good idea of the presence or absence of red cells. Marx's fluid (made as follows, hydrochlorate of quinin 10 c.c. of a 1 to 1,000 solution added to 10 c.c. of 33 per cent, potassium hydrate solution, and tinted with eosin which will stain the erythrocytes a characteristic reddish hue) is a very excellent examining fluid. As the presence of the red cells is so hard to demonstrate, search for them THE BLOOD. 623 is frequently omitted and reliance placed more upon the demonstration of the hemoglobin in such stains. If the stain is upon a hard surface it may be scraped off with a clean piece of glass. If it be upon cloth, a portion of the stained part and also of the unstained part is removed and cut into small pieces with scissors which are absolutely clean. Such small pieces not broader than 1 mm. are very easily handled and can be readily teased out if desired. (2). Guaiac Test. This test originally devised by van Deen is one of our oldest for the detection of blood coloring matter, but seems to be much more reliable in its negative phase than when positive. The principle of the test is as follows: To a watery solution of the suspected stain is added an equal portion of fresh tincture of guaiac. This tincture is best made by dissolving the guaiac resin in alcohol when needed. The addition of the guaiac tincture to the suspected solution causes a milky turbidity. On now adding ozonized oil of turpentine, peroxid of hydrogen, or oil of eucalyptus, so that these latter substances float upon the guaiac and blood solution, a distinct blue color will be manifested at the point of contact, and, on shaking the tube, the coloration will spread throughout the mixture. This test seems to be very delicate, demonstrating the presence of blood in a dilution of several thousand. The blue coloration observed in this test is due to the oxidation of the guaiac to guaiaconic acid, which is then further oxidized into guaiac blue by the catalytic action of the oil of turpentine. It has been shown by Taylor that this reaction is also given by many substances among which we find manganate and permanganate of potassium, peroxid of manganese, peroxid of lead, chlorin, bromin, iodin, nitric and chromic acids, ferric chlorid, salts of copper, ferro- and ferricyanid of potassium, gum acacia, gluten, unboiled milk, raw potato pulp, pus, and any living cell or its intracellular enzymes. Various enzymes such as the oxidases will give this reaction. Buckmaster advises the use of the pure guaiaconic acid along with peroxid of hydrogen in the place of the guaiac resin and believes that the test increases in delicacy and accuracy under these conditions. He states that fluids containing hemoglobin or most of its derivatives give this test when it is impossible to detect pigments by any other methods and rightly adds that boiling the fluid suspected of containing blood does not interfere with the reac- tion but, on the contrary, throws out of consideration the action of milk, pus, fibrin, or of any enzyme. For the proper performance of this test the fluid to be tested, should not be alkaline and only very slightly, preferably not at all, acid. (3). Schaer's Test. This test is similar in many ways to the van Deen test, but employs, instead of guaiac, a i to 4 per cent, solution of aloin in alcohol. On adding this tincture to the suspected solution a red color soon becoming distinctly cherry-red appears upon the addition of ozonized oil of turpentine. Many other substances give a pink color, but only after the lapse of one or two hours, while the color 624 DIAGNOSTIC METHODS. with blood appears in a very short time. The tincture of aloin should always be freshly prepared, as of itself it undergoes this color change after standing. U'). Phenolnhthalin Test. This test,1 originally employed by Meyer2 and, later, by Utz,3 has been so improved by Kastle and Amoss4 and Deldarde and Benoit5 that it is, probably, our most delicate one for the detection of blood, especially in stains, its limit of delicacy being about i part of blood in 8 million of water. As in the previous tests, the hemoglobin acts as an oxygen carrier, the active oxidizing agents being ozonized turpentine or hydrogen peroxid. The oxidases, peroxidases and catalases play small roles in this process in the case of blood-stains, but, in the various secretions and excretions of the body, their influence is very great. Phenolphthalin is a product of the reduction of phenolphthalein by zinc in alkaline solution. When oxidized in alkaline solution, it is converted into phenolphthalein with production of an intense red color. The reagent may be obtained in the market or, preferably, prepared as follows. Phenolphthalein is dissolved in a considerable excess of 30 per cent, sodium hydrate solution and boiled with an excess of zinc dust until a few drops of the strongly alkaline liquid no longer give a red color after neutralization with HC1 and sufficient alkali to alkalinize the solution. The solution is then decanted from the excess of zinc dust and the phenolphthalin is precipitated by acidifying with HC1. Collect the precipitate on a filter and purify by repeated crystallization from water and alcohol. This purification is continued until a white crystalline compound is obtained free from every trace of phenolphthalein (as shown by absence of red coloration on addition of alkali). Dry at room temperature or in the oven at 500 to 8o° C., care being taken to avoid contact with metallic surfaces. This compound should be kept in tightly stoppered bottles in a dark place, as oxida- tion gradually occurs. Rarely do we find that the coloration is sufficient to interfere with the tests, but in the delicate forensic tests it may be necessary to repurify a product which has stood for some time. The solution, as used in the test, is as follows: Mix a slight excess of phenol- phthalin, prepared as above, with 1 c.c. of N/10 sodium hydrate solution and a few c.c. of redistilled (from glass) wTater, shake thoroughly and filter. To the filtrate add 20 c.c. of N/10 NaOH solution, 0.1 c.c. of 3 per cent, hydrogen peroxid solution and make the mixture up to 100 c.c. This solution shows no trace of pink coloration when fresh, but gradually acquires a color, which may become so intense that the reagent can not be employed. In forensic work, use only freshly prepared solutions. Toi part of the aqueous solution of the stain or of the secretion or excretion 1 See Kastle, Bull. 51, Hyg. Lab. U. S. P. H. & M. H. S. 1909. 2 Munch, med Wochs., Bd. 50, 1903, S. 1489. 3 Chem. Ztg., Bd. 27, 1903, S. 1151. 4 Bull. 31, Hyg. Lab. U. S. P. H. & M. H. S., 1906. 3 Comp. Rend. Soc. Biol., Tm. 64, 1908, p. 990. THE BLOOD. 625 to be tested add 2 parts of the reagent and allow to stand for a few minutes. In the presence of blood a pink to red color appears, the intensity depending on the amount of blood present. This reaction is retarded by the extracts of various animal tissues or various secretions of the body. For this reason we are never able to detect as small amounts of blood in the secretions as in watery solutions of pure blood, the limit of delicacy of all reactions being far less in the former than in the latter case. Boiling the solutions before applying the test removes most of the interfering factors. If the secretion be treated with a thick cream of aluminium hydrate suspension, the precipitate will carry down the blood pigment and thus concentrate it. A small amount of this precipitate, Fig. 158.-Haemin Crystals from Human Blood. (Hawk.) Reproduced from a micro-photograph furnished by Prof. E. T. Reichert, of the University of Pennsylvania. whether derived from saliva, urine, feces suspension, milk, or exudates, will show a decided red color when added to 2 c.c. of the reagent. Of course, in applying the test to the aqueous solution of the blood stain, no such treatment need be employed. This test is especially recommended for the detection of blood in all secre- tions and excretions (boiling them before applying the test) and, particularly, in the examination of suspected stains for forensic purposes. All possible fallacies must be guarded against and confirmatory tests used if necessary.1 (5). Teichmann's Test. This is one of the most important tests for the presence of blood and when positive is conclusive proof of the kind of stain with which one is working. 1 Vas (Deutsch, med Wochs., Bd. 38, 1912, S. 1412) calls attention to the possibility of the presence of phenolphthalein as a purgative (if feces be tested). The addition of alkali alone will give a red color under this condition. 626 DIAGNOSTIC METHODS. A drop of blood or a portion of the suspected stain is spread upon a glass slide and covered with a drop of a very dilute solution (0.85 per cent.) of common salt.1 The salt solution is then evaporated at a low temperature. A few drops of glacial acetic acid are then placed upon the salted stain and the preparation is covered with a cover-glass. The acid is now slowly evaporated by holding the slide over a flame in such a way that the fluid steams but does not boil. As the acid evaporates, more is allowed to run under the cover-glass, this addi- tion and evaporation being done thrice. The specimen is cooled and mounted in glycerin or distilled water, after which it may be examined with the dry lens. A successful specimen, which is not always obtained, will show the presence of numerous brownish rhomboid crystals, which are separate or grouped in sheafs or rosettes. These are the hemin crystals or hydrochlorate of hematin, called, after their discoverer, Teichmann's crystals. It very frequently happens that the specimens are not successful, so that several slides should be made before a negative report is given. Reasons for failure may be found in the excessive heat applied to evaporate the acid, in the fact that the salt solution may have been too concentrated, or that the stain may have been very old. The addition of salt seems to be unnecessary except in the case of old blood stains or when the blood is poor in salts. It should even then be added in only very small amounts. As pointed out by Rose, when blood is mixed with iron rust the hemin test is usually negative. Certain characteristic appearances of the blood are noted on spectroscopic examination. The spectrum of the various blood pigments has been given in previous sections so that only a few additional remarks are necessary in this place. In spectroscopic examinations, the use of the ordinary hand spectro- scope is all that is necessary, the more complicated ones adding little to the differentiation of the various spectra. If the blood be fresh and the hemoglobin unaltered the spectrum is, of course, that of oxyhemoglobin; but as the usual stains submitted for examination are frequently very old ones, such a spectrum is practically never seen. The end-product of the alteration of hemoglobin as found in old blood stains is hematoporphyrin, which is iron-free and can be identified easiest by the spectroscope. The suspected stain is dissolved in concentrated sulphuric acid yielding a reddish-violet fluid, which is then examined spectroscopically. The spectrum of hematoporphyrin has been given previously. Many blood stains can be recognized only by this test, so that it is always advisable to make a spectroscopic examination, especially if other tests for hemoglobin have failed. (6). Spectroscopic Examination of Blood. (7). Precipitin Test. The precipitin test has been discussed in great detail previously (page 590). This is the absolute identification test for a stain, providing the stain has been 1 Nippe (Deutsch, med. Wochs., Bd. 38, 1912, S. 2222) advises one to mix the specimen with a few drops of a solution of 0.1 gram each of potassium bromid, iodid and chlorid in 100 grams of glacial acetic acid. This modification shows up the crystals even when blood is mixed with rust. THE BLOOD. 627 shown to be a blood stain by the tests given above. The precipitin test is specific, within very narrow limits, and is recognized in law as reliable in differentiating human blood from that of the domestic animals. Its one fallacy is that it reacts with the blood of anthropoid apes just as it does with that of man, but in most cases this element may be easily excluded. X. Value and Limitations of Blood Examinations. It has been truly said that the value of a blood examination "is measured by the practical use which may be made of it, and not by any interesting yet useless information it may throw on the case." In clinical work we are prone to follow well-established lines of procedure and to forget, when engaged in counting the cells in an obscure case, that "the meaning and aim of the clinical study of the blood covers a much wider field than is embraced in the mere investigation of the histological elements." Yet our knowledge regarding the variations in the plasma is so meager that the busy practitioner may be pardoned for such neglect. There can be little question that the so-called "blood diseases" have the most characteristic variation from the normal, yet even here it is not so easy, as one might assume, to make an indisputable diagnosis in all cases. Thus, aside from the pure parasitic diseases, leukemia and advanced pernicious anemia seem to be the only ones capable of a certain diagnosis from the examina- tion of the blood. True the various anemias may be detected, but the etiology is not always clear nor the differentiation of a primary from a secondary form invariably possible. It is, of course, evident that bacteriologic examination of the blood and the application of the agglutination reaction may clear up many an obscure case, but such work requires much more time and much more attention to detail than is at the disposal of the busy practitioner. It is, however, in just this class of infectious fevers that much benefit, in a confirmatory way, is forth- coming from a careful study of the blood. Thus, a leucocytosis or a leucopenia in any suspected case might lead to a differentiation of central pneumonia from typhoid fever or a scarlet fever from a case of measles. Moreover, a leucocytosis following a leucopenia in typhoid would strongly indicate perfora- tion and consequent surgical interference. Suppurative processes anywhere in the system are usually associated, unless very limited in extent, with a polynuclear neutrophile leucocytosis; so that the leucocyte count may be of great importance in judging of the extension of a pus infection. Immediate operation is occasionally decided upon after a leucocyte count, but this should not be made the only basis for such intervention unless frequent counts have given the surgeon the definite knowledge that a leucocytosis has suddenly occurred. It should be stated as an axiom that blood findings may never be interpreted except in the light of the clinical findings. To the clinician it should be said, never trust the laboratory report implicitly unless it agrees with the clinical 628 DIAGNOSTIC METHODS. manifestations; to the laboratory worker we would say, never report a blood finding as diagnostic without knowing something of the clinical history of the case. A single blood examination will rarely be of any greater value than will a single temperature determination. It is the series that enables one to decide either as to diagnosis or as to operative or therapeutic procedures. What we need most is not so much blood work, but much better work when done. Too much reliance is often placed on blood reports, so that a word of caution does not seem ill advised. Never rely solely on the blood findings, but use them merely as one of many links in your chain of clinical evidence. In this way mistaken diagnoses will occur less frequently. BIBLIOGRAPHY. i. Arneth. Die neutrophilen weissen Blutkorperchen. Jena, 1904. Diagnose und Therapie der Anamien. Wurzburg, 1907. 2. Besancon et Labbe. Traitd d'h^matologie. Paris, 1904. 3. Bramwell. Anemia and Diseases of the Ductless Glands. Edinburgh, 1899. 4. Buckmaster. The Morphology of Normal and Pathological Blood. Phila- delphia, 1906. 5. Cabot. Clinical Examination of the Blood. New York, 1904. 6. Canon. Bakteriologie des Blutes. Jena, 1905. 7. Celli. Malaria According to the New Researches. London, 1900. 8. Charles. Fonction des leucocytes. Paris, 1904. 9. Coles. The Blood: How to Examine It. London, 1898. 10. Craig. Estivo-autumnal Malarial Fevers. New York, 1901. The Malarial Fevers. New York, 1909. 11. DaCosta. Clinical Hematology. Philadelphia, 1905. 12. Ehrlich, Lazarus, Pinkus, und von Noorden. Blut-Krankheiten. Wien, 1898. Translation Philadelphia, 1905. 13. Ewing. Clinical Pathology of the Blood. Philadelphia, 1903. 14. Grassi. Die Malaria. Jena, 1901. 15. Grawitz. Klinische Pathologic des Blutes. Leipzig, 1911. 16. Hayem. Du sang. Paris, 1889. 17. Helly. Die hasmopoetischen Organe. Wien, 1906. 18. Janeway. The Clinical Study of Blood-pressure. New York, 1904. 19. Kahane. Die Chlorose. Berlin, 1901. 20. Laveran. Du paludisme et de son hematozoaire. Paris, 1891. Traits du Paludisme. Paris, 1907. 21. Laveran et Mesnil. Trypanosomes and Trypanosomiases. Chicago, 1907. (Translation by Nabarro.) 22. Lenhardt. L'Anomie £ type chlorotique. Paris, 1906. 23. von Limbeck. Klinische Pathologic des Blutes. Jena, 1896. 24. Mannaberg. Die Malariakrankheit. Wien, 1899. 25. Naegeli. Blutkrankheiten und Blutdiagnostik. Leipzig, 19 n. 26. Oliver. Studies in Blood-pressure. London, 1908. 27. Pappenheim. Atlas der menschlichen Blutzellen. Jena, 1905 and 1912. 28. Patella. I leucociti non granulosi del sangue. Torino, 1906. La genesi endoteliale dei leucociti mononucleali del sangue. Siena, 1907. 29. Potain. La Pression art€ri€lle de 1'homme. Paris, 1902. THE BLOOD. 629 30. Reinert. Die Zahlung der Blutkorperchen. Leipzig, 1891. 31. Ruge. Malaria Krankheit. Jena, 1906. 32. Schleip. Atlas der Blutkrankheiten. Berlin, 1908. Translation, New York, 1908. 33. Schoenborn. Gefrierpunkts- und Leitfahigkeitsbestimmungen. Wies- baden, 1904. 34. Stephens and Christophers. The Study of Malaria. London, 1905. 35. Sternberg. Primarerkrankungen. Wiesbaden, 1905. 36. Strauss und Rohnstein. Anamien. Berlin, 1901. 37. Sutherland. Blood Stains. New York, 1907. 38. Tallqvist. Ueber experimentelle Blutgiftanamien. Helsingfors, 1899. 39. Thayer. Lectures on the Malarial Fevers. New York, 1897. 40. Thayer and Hewetson. Malarial Fevers of Baltimore. Baltimore, 1895. 41. Turk. Vorlesungen ueber klinische Haematologie. Wien, 1904 and 1912. 42. Uhlenhuth. Das biologische Verfahren zur Erkennung und Unterscheid- ung von Menschen und Tierblut. Jena, 1905. 43. Watkins. Diagnosis by means of the Blood. New York, 1902. 44. Weiss. Haematologische Untersuchungen. Wien, 1896. 45. Wile. Blood Examinations in Surgical Diagnosis. New York, 1908. 46. Wolff. Die Kernzahl der Neutrophilen. Heidelberg. 1006. CHAPTER IX. TRANSUDATES AND EXUDATES. I. General Considerations. The serous membranes are normally kept moistened by liquids whose quantity is only sufficient in a few instances, as in the pericardial cavity and the subarachnoidal space, for a complete chemical analysis to be made of them. Under pathological conditions an abundant transudation may take place from the blood into the serous cavities, into the subcutaneous tissues or under the epidermis. If such conditions be the result of circulatory disturbance, the kidneys are usually unable to eliminate the normal amount of flu'd from the system and, as a result, the retained fluid collects both in the serous cavities and in the areolar tissue. Such accumulations of fluid, known as transudates, are similar to the lymph, being, as a rule, poor in cellular elements and yielding little or no fibrin. These transudates must be sharply differentiated from the accumulations of fluid, which are the result of direct inflammatory processes in the membranes lining the serous cavities and are known as exudates. These latter fluids are generally rich in cellular elements and yield relatively more fibrin. As a rule, the richer a transudation is in leucocytes, the closer it stands to pus, while the poorer it is in these elements the closer it resembles true lymph. The formation of true transudates is largely a question of filtration under the influence of the rate of blood flow, the blood pressure, the irritation of the capillary endothelium, and the variable permeability of the endothelial cells. We should expect, therefore, that the passage of dissolved substances from the blood would be regulated by the same laws that control the secretion of physiologic fluids, namely the laws of passage of fluids through semipermeable membranes. The crystalloids would be, therefore, in approximately the same concentration as in the blood plasma, while the colloids must be in far less concentration, the actual values being influenced, of course, by the special membrane through which the fluid passes. The condition of the blood would hence affect the chemical composition of such transudates, hydremic plasma yielding a fluid poorer in solids, while anhydremic blood is associated with a transudate of higher specific gravity. From a clinical standpoint a differentiation between transudates and exudates is not infrequently impossible, so that it is advisable to resort to aspiration of the fluid and to chemical and microscopical examination of the material withdrawn. The chief phases of such examinations are: (i) the chemical and physical properties of the fluid; (2) the bacteriological aspect of the fluids, and (3) the morphological characteristics of the cellular elements. 630 TRANSUDATES AND EXUDATES. 631 Obtaining the Specimen. Whenever fluid is to be withdrawn, either for diagnostic or therapeutic purposes, it is necessary to resort to puncture of the cavity containing the fluid. In all cases the site of puncture must be as carefully prepared as in any surgical procedure. Puncture may be performed with the ordinary trocar or, preferably, with a large needle which has a rather large lumen. The instruments must be carefully sterilized before use. If the skin is especially tough, it is advisable to make a small incision through the skin and insert the needle through the incision. Very little pain is felt by the patient during this procedure, but the writer is accustomed invariably to resort to the use of ethyl chlorid to anesthetize the part. In the withdrawal of pleuritic effusions the spot selected should be neither too high nor too low. It may be in the seventh intercostal space in the axillary line or in the eighth intercostal space at the outer angle of the scapula. The arm of the patient should be brought forward with the hand resting on the opposite shoulder, in order to widen the intercostal spaces. In inserting the needle it is wise to make the thrust close to the upper margin of the rib so as to avoid the intercostal artery. In all cases the fluid should be withdrawn slowly and the excess above that required for examination allowed to drain until the desired amount is obtained. If the puncture is for diagnostic purposes, io to 20 c.c. are sufficient, while the therapeutic withdrawal of fluid will vary with the amount present and with the clinical symptoms of the case. Should the patient show signs of shock or of depression during the operation, the procedure should be interrupted as quickly as possible. Aspiration is rarely necessary or advisable. In the withdrawal of fluids from the abdominal cavity, the needle or trocar is thrust through the lower abdominal wall and the fluid collected. In this procedure more or less danger of puncturing the bowel exists if the effusion be small, so that the needle should not be carelessly inserted lest this complication arise. Naturally, in well-marked ascites no such danger is present. II. Physical and Chemical Properties. Transudates are, as a rule, serous in character, usually transparent, colorless, or light yellow in color, but at times showing a milky, reddish, or a greenish tinge, the latter practically always being observed after the fluid has stood exposed to the air. Such solutions are, as a rule, dichroic, yellow by transmitted light and green by reflected light. They are alkaline in reaction and show a specific gravity, which varies, according to the origin of the fluid, from 1006 to 1018, while serous exudates from the same cavities show a much higher specific gravity.1 The variations in specific gravity depend largely upon the amount of albumin present in the transudate, this practically never being over 3 per cent, and usually much lower. The chief proteins present are albumin and globulin, these being related to one another in the transudates as one and one-half to one, while in the exudates the globulin is relatively much increased. The determina- tion of the total protein may be made by methods of fractional precipitation as 1 See Trevisan, Zeitsch. f. Exp. Path. u. Then Bd. io, 1912, S. 141. 632 DIAGNOSTIC METHODS. previously discussed. The transudates from the pleura contain the largest percentage of albumin, while edematous fluids rarely show over i per cent. Transudates do not coagulate spontaneously. Glucose is present both in transudates and exudates in amounts varying between 0.04 and 0.1 per cent. The mineral constituents of transudates are somewhat higher than in the exudates, the former averaging 0.96, the latter 0.89 per cent. Under patho- logical influences fat, blood, uric acid, and biliary pigment may find their way into both types of fluid. In diabetes an excess of sugar and the presence of acetone bodies may be detected. The exudates are usually straw- or lemon-yellow in color depending on the degree of inflammation, or they may assume colorations ranging from a deep red (hemorrhagic) to a milky (purulent) shade. Biliary pigments may cause a bright green, while various medicaments, such as methylene blue, may produce a greenish-blue coloration. The specific gravity is almost invariably above 1018, the reaction is alkaline, the albumin content is usually above 3 per cent., reaching as high as 7 per cent., while the globulin is relatively much increased in comparison with the albumin. This globulin increase is largely due to paraeuglobulin. Nucleoprotein is especially abundant in purulent exudates in which the autolytic processes are more or less marked. The total nitrogen of the various fluids varies from 0.22 to 1.38 per cent. The nitrogen partition may be seen from the following table of Gerhartz.1 Exudates coagulate spontaneously on standing.2 Figures in terms of % (grams per 100 cc.) Total N Precipi- table N Protein N Ammonia N Purin N Urea N Amino-acid N Transudate Serous exudate. ... Purulent exudate.. 0.22-0.58 0.43-1.09 1.11-1.33 0.18-0.53 0.39-1.07 0.94-x.22 0.17-0.52 0.37-x.03 X.X4 O.OO7-O.OI 0.01 -0.03 o.ox 0.002-0.007 O.OI 0.007 0.01-0.05 0.01-0.06 0.02-0.1 0.002-0.005 0.007 0.004-0.28 The exudates, which, accurately speaking, are always of inflammatory origin, may be serous, serofibrinous, seropurulent, hemorrhagic, purulent, putrid, chylous, and chyloid. Serous Exudates. These are clear, of a light straw color, and show a specific gravity above 1018. There is a large amount of fibrin, as shown by the dense network microscopically, containing a few red cells which may be derived from the bleeding at the point of puncture, a few leucocytes which may vary in type according to the kind of bacteria causing the infection, and large endothelial 1 Chemie der Transudate und Exsudate, Jena, 1908. See, also, Wiener (Biochem. Zeitsch., Bd. 41, 1912, S. 149) regarding the presence of proteolytic ferments and amino-acids in exudates and Lenk and Pollak, Deutsch. Arch. f. klin. Med., Bd. 109. 1913, S. 351. 2 Rivalta advocates the following test to differentiate an exudate from a transudate. In a conical glass place about 200 c.c. of water and add 2 drops of glacial acetic acid. Mix thoroughly and allow one or two drops of the puncture-fluid to fall into this weak acid solu- tion. If the fluid be an exudate, a distinct cloud will be observed in the wake of the falling drop, while if it be a transudate little or no turbidity will be noticed. This reaction is due to the large globulin content of the exudate. TRANSUDATES AND EXUDATES. 633 cells from the serous membrane lining the cavity. If the blood-cells be present in sufficient numbers to give a distinct red color to the fluid it is termed a hemor- rhagic exudate, while if a few pus-cells are found it may be called a seropurulent type. The gradations between the true serous and serofibrinous types of exudate are exceedingly varied, so that it is difficult to tell which type is really present. Even from a purely serous exudate a certain amount of coagulation, with formation of a distinct fibrin network, may be obtained, so that the only criterion would be one of degree. The type of leucocyte present is usually of the polymorphonuclear variety, although other forms may be present. It is, therefore, an important part of the examination of an exudate to determine the percentage relations of these various cellular elements. This will be discussed in the section on Cytology. Chylous Exudates. On account of the close relationship between the abdominal, thoracic, and pericardial cavities, on the one hand, and the large lymphatic trunks, on the other, it is possible for lymph to pass directly into these cavities in case rupture of these vessels occurs. Such exudates show all the properties of chyle. The fluid is white and milky, contains between 1.5 and 2.5 per cent, of protein, and a considerable amount of fat, which may be demonstrated by staining with osmic acid or Sudan-Ill or by alkalinizing with sodium hydrate and shaking out with ether. Chyloid Exudates. It has been found that carcinoma, tuberculosis, extreme cardiovascular changes, hepatic disturbances, puerperal sepsis, and infection with the filaria may give rise to a chyloid type of ascites. The fluid in these cases is less milky than that of the true chylous type, contains less fat, and is not so completely cleared by shaking with ether. It contains serum-albumin, mucin and, es- pecially, rather large amounts of a complex of pseudoglobulin and lecithin, to which the opacity is due.1 Traces of sugar are found but not as large in amount as in the true chylous types. These fluids resist putrefaction for long periods owing to their lecithin content. The accumulation of fluid after tapping is much more rapid in the chylous than in the pseudo-chylous types. Hemorrhagic Exudates. This type is, in reality, a serofibrinous form containing large numbers of blood-cells. It is observed in patients with hemorrhagic diathesis, in connection with active tuberculosis, with neoplasms of the serous cavities, and following injuries to the chest or abdomen. In this form of exudate, the exciting bacterial agent, usually the tubercle bacillus, may occasionally be found, but not always. The type of leucocyte (mononuclear) would be very strong presumptive evidence in favor of tuberculosis, even though no bacilli were found. If the exudate be due to a malignant growth, it is sometimes possible to obtain shreds of the tumor tissue and thus make a probable diagnosis. 1 See Wallis and Scholberg, Quart. Jour, of Med., vol. 3, 1910, p. 301; Ibid., vol. 4, 1911, p. 153. Also Henry, New York Med. Jour., vol. 94, 1911, p. 1. 634 DIAGNOSTIC METHODS. In judging of the malignancy of the cells, it is sometimes difficult to differentiate the abnormal from the normal, especially in the affections of the pleura. These malignant cells are usually extremely large and are characterized by their vacuolation and fatty degeneration (see cut). It is not infrequent to find in hemorrhagic exudates, which have remained in the body cavity for some time, a large number of cholesterin crystals and occasionally small masses of hemosiderin. Purulent Exudates. These are composed either of true pus or of seropus. They are more or less yellow in color, thick and occasionally tenacious, and separate on standing or centrifuging into a cellular deposit and a pus serum. The cells forming the pus are not infrequently in a condition of advanced fatty degeneration and may contain numerous bacteria. The addition of acetic acid will usually clear up the cells so that the nuclei become recognizable. Fat is present in amounts as high as 7 per cent., while a relatively large amount of various pro- teins and extractives is shown on chemical examination. Fatty acid crystals and cholesterin may be abundant in the sediment, especially in old abscesses. Fresh pus is usually alkaline, but may become distinctly acid, owing to the development of lactic acid in the process of autolysis. Purulent exudates are investigated with special regard to the type of cell present and to the organism associated with the pus formation. In ordinary pus, the cell is of the polymorphonuclear type, although occasionally the mononuclear form may be predominant. The bacteria are very numerous and include many of the most important types. While the pyogenic bacteria are more frequently the various types of staphylococci and streptococci, it is to be remembered that other organisms may produce pus under certain condi- tions. Thus the typhoid bacilli, colon bacilli, pneumococci, Friedlander's bacilli, gonococci, diphtheria bacilli, Morax-Axenfeld and Koch-Weeks bacilli, and influenza bacilli, among others, may be the causative factor in the formation of a purulent exudate. It will be seen, therefore, that the bacterial examination of a purulent exudate may require a very extensive research. As it is frequently impossible to decide, from a stained specimen, as to the special organism causing the infection, cultures should be made in every doubtful case. The peculiarities of such cultures may be found in any work on bacteriology. Putrid Exudates. These may be observed in various cavities of the body or in the substance of various organs, especially the liver and lungs. They arise from the entrance of pus into these cavities from perforation of a gangrenous area, gastric or intestinal ulcer, malignant growths, etc. The material obtained by puncture is usually brownish or greenish in color, has a very offensive odor, and is usually alkaline, but may be acid. Microscopically degenerated cells, numerous bacteria, cholesterin, fatty acid, and hematoidin crystals are observed. In some cases various portions of an echinococcus cyst may be found in the exu- TRANSUDATES AND EXUDATES. 635 dates. Bilirubin crystals and various amino acids may be found in rupture of an hepatic abscess. III. Bacteriology. It is always advisable to make cultures of the material obtained by puncture in order to discover the organism which is acting as the exciting cause of the condition under investigation. Usually a few c.c. of the fluid are allowed to drop into a flask containing 50 c.c. of sterile nutrient bouillon and the mixture incubated for 24 to 48 hours. From the growth, obtained in this preliminary work, subcultures are made on various media and microscopical examination employed to identify the organism. It is always advisable to make at least two microscopical specimens, staining one with the ordinary methylene blue stain and the second with Gram's stain. Tubercle Bacilli. If these organisms are suspected, the technic of staining is the same as outlined under Sputum. It frequently happens, in the examination of suspected tubercular exudates, that the presence of a large amount of fibrin, either with or without spontaneous coagulation of the specimen, makes it exceedingly difficult to find the organism, even though it be present. In such cases, advantage is taken of a procedure, recommended by Jousset,1 known as inoscopy. If the fluid has not coagulated, it is allowed to do so in order that the coagulation may enclose the bacilli within the fibrinous network. The coagulum is separated as completely as possible from the supernatant fluid, is washed with distilled water, and is treated with 30 to 40 c.c. of the following mixture, which will digest the fibrin. Pepsin, 2 grams. Sodium fluorid, 3 grams. Glycerin, 10 c.c. Hydrochloric acid (cone.), 10 c.c. Distilled water, q.s., ad., 1000 c.c. This mixture is placed in the incubator for 24 hours, when the fluid becomes homogeneous. The digested fluid is centrifuged and smears are made from the sediment as previously described. It is always advisable to treat such smears with a small quantity of albumin fixative, so that the organisms may not be washed from the slide. The smear is fixed in the flame and stained in the usual way with carbol fuschin. The tubercle bacilli may not appear as deeply stained after such treatment. This method has been replaced, to a large extent, by the Antiformin method (see p. 18). Gonococcus. The gonococcus appears in stained specimens as small biscuit-shaped or coffee-berry-shaped cocci, which are arranged in pairs separated by a narrow unstained portion. Occasionally two of these pairs of hemispheres are joined ^em. M&L, Tm. 23, 1903, p. 22. 636 DIAGNOSTIC METHODS. together, forming tetrads. They may be stained by any of the anilin dyes, but are recognized especially by their reaction toward Gram's method. In this con- nection it must be said that other organisms, among which we find the diplococcus intracellularis meningitidis of Weichselbaum and the micrococcus catarrhalis, resemble the gonococcus both in morphological and staining characteristics, so that a differentiation by this method is not always possible. Fortunately, how- ever, the cultural peculiarities of these organisms absolutely differentiate them.1 Gram's Stain. The principle of the staining of various organisms by Gram's method is that certain organisms retain the primary color after treatment with decolor- izing agents, while others lose this primary stain and must be treated with a contrast stain for their recognition. If the organism retains the primary blue color it is called Gram-positive, while if it lose the primary color and take on the contrast stain it is called Gram-negative. Among the organisms which are Gram-negative we find the gonococcus, meningococcus, micrococcus catarrhalis, influenza bacillus, typhoid bacillus, colon bacillus, Koch-Weeks bacillus, and the Morax-Axenfeld bacillus ;|while the tubercle bacillus, smegma bacillus, diphtheria bacillus, pneumococcus, streptococcus, staphylococcus, and various saprophytic cocci found in the smears both of the male and female urethra are Gram-positive. The solutions required in Gram's method are: (i) an anilin oil-gentian violet mixture, consisting of 84 c.c. of anilin water (water saturated with anilin and filtered), and 16 c.c. of a saturated alcoholic solution of gentian violet; (2) a solution of iodin consisting of one gram of iodin, 2 grams of potassium iodid, dissolved in 300 c.c. of water; (3) a dilute solution of carbol fuchsin, a dilute aqueous solution of safranin, or a 1 per cent, aqueous solution of Bismarck brown as a contrast stain. Technic. Smears of the exudate are made, in the manner previously described for making blood smears, by receiving a drop of the purulent material upon one end of a glass slide and spreading it in a thin even layer by means of a second slide. These smears are then fixed by passing several times through the flame. When the slides have cooled, they are covered with the gentian- violet solution which is allowed to act for one to three minutes. The solution is then poured off and the excess removed by washing in water. Without drying, the iodin solution is placed on the slide and allowed to act for one-half to one minute. It is then washed in water and the preparation treated with 95 per cent, alcohol until no more color is removed by it. The alcohol is then removed by washing with water and the smear is covered with one of the contrast stains above mentioned, in the writer's laboratory the safranin solu- tion, which is allowed to act for only a few seconds. The stain is then washed off with water and the slide dried between folds of filter-paper. 'See Warden (Jour, of Inf. Dis., vol. 12, 1913, p. 93), who emphasizes the importance of a slightly acid reaction in the media containing preferably human body fluids, such as ascitic or hydrocele fluid. PLATE XXXIII. Gonococci in Urethral Discharge. (Gram's Stain.) TRANSUDATES AND EXUDATES. 637 The specimen under the microscope shows the Gram-positive organisms stained deep blue, while the Gram-negative bacteria and the bodies of the pus- cells take the red safranin stain. In such specimens the gonococcus, if pres- ent, will be observed both intracellularly and extracellularly, the former being the more characteristic. The ordinary pus cocci may likewise be intracellular, but these are distinctly Gram-positive, while the gonococcus is Gram-negative. In the purulent exudate from the urethra, large masses of mucoid material may be present, which are known as gonorrheal threads. These may be found in the urinary sediment and are usually easily recognized. They contain masses of pus cells, within which may be found numerous gonococci. These shreds may persist for years in anyone with a history of a previous gonorrhea, but they may then contain no organisms. The cytology of gonorrheal pus presents nothing characteristic beyond the presence of numerous eosinophiles with large numbers of polynuclear neutrophiles and an occasional lymphocyte. Smegma Bacilli. In the exudate of the preputial follicles, known as smegma preputii, are found fat globules, ammonium soaps, cholesterin crystals, and a few epithelial cells. In this smegma are found many bacilli, smegma bacilli, which show the same morphological and practically the same staining characteristics as the tubercle bacillus. This is differentiated, as described under Sputum. If the urine is to be examined for tubercle bacilli, it is much better technic to use a catheterized specimen than it is to resort to methods of questionable differentiation. If this is done any acid and alcohol-fast bacilli may be regarded as the tubercle bacilli, but in all cases the smegma bacilli must be excluded. Ducrey's Bacillus. This bacillus is the organism causing soft chancre and is known as the bacillus ulceris cancrosi. It is found in the purulent discharge from the chancre, but rarely, if ever, in pure culture. The sections of tissue may, how- ever, show none but this organism. In making preparations for study of this organism, the ulcerated surface is scraped with a platinum loop, the pus spread upon a slide, dried in the air, and fixed with alcohol-ether or over the flame. It stains readily with all bacterial stains, but decolorizes with Gram's stain. In stained specimens the organism appears as a short, thick, oval bacillus with rounded ends and two lateral indentations, which occasionally give it the appearance of a figure 8. Ordinarily the extremities are more deeply stained than the central portion, this appearing almost clear. The organism has a tendency to form chains or groups, which are rarely found in the pus-cells but are frequently seen within the epithelial cells. Spirochaeta Pallida. The Treponema Pallidum, discovered by Schaudinn and Hoffmann,1 is now definitely established as the causative factor of syphilis. It may be 1 Arbeiten aus der kais. Gesundheitsamte, Bd. 22, 1905, S. 527. 638 DIAGNOSTIC METHODS. obtained from the primary chancre, the incised papules, condylomata, mucous patches, inguinal glands, as well as from the internal organs in tertiary and congenital syphilis. Its size, form, type of motility and method of cultivation have been discussed under blood. To obtain this organism from the primary lesions, the chancre is thoroughly cleansed either with normal salt solution or with soap and water, rinsed with salt solution and dried. If the sore is healed, the epithelial covering is removed. The chancre is lightly curetted or the edges scraped, the blood, which should be avoided, is wiped away and the serum which exudes used for the later work. A drop of this serum is placed upon a slide, covered with a cover-glass and ex- amined at once under the dark-field illuminator, the spirochete appearing as colorless, glistening, rapidly motile spirals. If this method, which is to be prefer- red to any staining method, is not possible, a thin smear is made from the serum, transferring a drop by means of a platinum loop to a slide and using a rapid circu- lar motion in spreading. The specimen is dried in the air and stained as follows. 1 unnicliff (Jour. A. M. A., vol. 58, 1912, p. 1682) calls attention to a very simple and reliable method formerly used by Oppenheimer and Sachs as well as by Ploeger. Make smears as usual, fix in the flame and cover for a few seconds with a 10 per cent, mixture of a saturated alcoholic gentian-violet solution in 5 per cent, phenol. Wash in water and dry. Giemsa's Stain. The air-dried film is fixed with chemically pure methyl alcohol and stained with Giemsa's stain (p. 460) for 20 hours, using 1 drop of the stain to 1 c.c. of water. The spirochete are stained a violet-red color, the refringens being blue. Ghoreyeb's Method.1 The smears are made as above and air-dried. Cover the smear with 1 per cent, aqueous solution of osmic acid for 30 seconds. Wash thoroughly in running water. Cover with solution of lead subacetate (liquor plumbi subace- tatis diluted 100 times) for 10 seconds. Wash in water. Cover with 10 per cent, aqueous solution of sodium sulphid for 10 seconds. Wash in water. The above process is repeated three times. Following this the osmic acid solution is applied for 30 seconds, the specimen is washed in water, dried and mounted in balsam. The spirochete, bacteria and cellular detritus are stained black. This method is quick, reliable, and is to be recommended. India Ink Method. Burri2 has advanced this method, which is simple and reliable, although some specimens of ink may show confusing artefacts according to Barach,8 who used an inferior ink. A drop of the fresh serum from the lesion is placed at one end of a glass slide and immediately mixed with a small drop of Gunther- Wagner India ink (Chin-Chin liquid pearl ink). The mixture is then spread and allowed to dry. The specimen is studied with the immersion lens. The 1 Jour. A. M. A., vol. 54, 1910, p. 1498. 2 Das Tuschverfahren, Jena, 1909. 3 Jour. A. M. A., vol. 55, 1910, p. 1892. PLATE XXXIV. Spirochete Pallide in Tissue. (Levaditi's Stain.) TRANSUDATES AND EXUDATES. 639 whole field is a homogeneous brown or black color, the treponema, blood cells, etc., appearing as colorless highly refractile bodies. The morphology, length and numbers of spirals should distinguish the pallida from the other types.1 The cytology2 of transudates and exudates has reference to the study of the various types of cells found in such fluids. As a rule, such investigations, known as cytodiagnosis, are carried out more frequently on the nonpurulent types of fluid, as the examinations of the purulent fluids are more especially concerned with the bacteria present. IV. Cytology. Technic. The technic of obtaining the cellular components of the puncture fluid will vary according as the effusion does or does not contain fibrin. If this be present in fairly large amount, the fluid must first be defibrinated before the next steps are possible. Here, again, the procedure may be complicated by the presence or absence of coagula. If the fluid be not coagulated, it is placed in a large sterile flask which contains a few sterile glass beads, the mixture is actively shaken until a firm clot is obtained, in the meshes of which a few of the cells will necessarily be enclosed, but the majority will remain sus- pended in the liquid. The fluid is then separated from the clot and placed in centrifugal tubes which are drawn out to a rather fine point. If the fluid be coagulated when collected, it should be shaken with glass beads to break up the clot and liberate the cellular elements. The fluid is then separated and treated as follows. The remaining portion of the technic is the same for fluids which contain or do not contain fibrinous material. The centrifuge is rotated rather rapidly for about ten minutes in order to collect the cells as a sediment. In most of the exudates the number of these cells is very small so that a concen- tration is absolutely essential. As Widal and Ravaut have shown, the polynu- clear cells seem to be somewhat more affected by the defibrinization than the other types, so that these cells may show a relative diminution. After the cells have collected, the fluid is removed by rapidly inverting the tube in such a way that the sediment does not follow the liquid. Some workers advise the withdrawal of the sediment by a pipet with a long fine tip, but the writer has not found this method any more advantageous than the one spoken of above. The sediment in the tube is shaken so as to mix it thoroughly and a drop placed upon the glass slide and spread as described under Blood. Where very few cells are present it is wise to allow the drop to dry on the slide without spreading, in order to concentrate the cells in a small area. The specimen is then treated with Wright's stain and examined under the immersion lens. If stains are to be used which contain no fixative, such as methyl alcohol, it is 1 Goosmann (Jour, of Cut. Dis., vol. 29, 1911, p. 628) has advised the use of nigrosin as a substitute for India Ink. 2See Labbe, Cytodiagnostic, Paris, 1903; Ravaut, Cytodiagnostic, Paris, 1901; Brion, Centralbl. f. allg. Path., Bd. 14, 1903, S. 609. 640 DIAGNOSTIC METHODS. necessary that the specimen be fixed by methods previously discussed in the section on Blood. The stain to be used will depend largely upon the points to be studied, the eosin-hematoxylin stain being especially serviceable in differentiating the nuclear structures of the cell. As the cells in the various pathologic fluids frequently show more or less degeneration, the nuclear portion is more suitable for study than is the cell protoplasm. For this reason the writer prefers the use of the eosin-hematoxylin method along with the Wright stain or the triacid stain for the granules of the cells. These specimens may also be used for the study of the bacteria present. When this object is to be subserved, it is advisable to make two specimens, staining one with the ordinary methylene blue stain and a second by Gram's stain. If the material be very limited in amount, it is possible to combine the above staining methods by treating first with the eosin-hematoxylin method, washing in water and then following the ordinary procedure of the Gram method of staining. Such specimens are extremely panoptic and are especially to be recommended. Cytology of Normal Fluids. The number of cellular elements in fluids from the various serous cavities of the body may vary from a very few to a large number. The cells observed are the red and white corpuscles of the blood, the latter of which are usually relatively more numerous than in the circulating blood and are usually largely of the polynuclear type, although mononuclear forms are frequently present. Neutrophiles and eosinophiles are present under normal conditions, the latter being relatively more abundant than in the blood. If a large number of red cells are found, an injury of the small vessels during puncture usually explains their presence. Besides these types of cell, which are exactly similar to those of the blood, a few endothelial cells are practically always seen. These cells are observed of different shapes, may be single or grouped in sheets, and may be very much degenerated. They are larger than the other cellular elements, their contour is usually circular, but may be irregular; they are mononuclear, and frequently contain round vacuoles. In examining normal as well as pathologic fluids for their cellular content, xoo cells should be counted if possible and the percentage of each type thus determined. This constitutes the cytologic formula of the exudate. Cytology of Pathologic Fluids. According to the theory of Metschnikoff, the presence of a bacterial in- fection is associated with attraction of the leucocytes to the infected area. These cells then enter into combat with the bacteria and either destroy the organism or are destroyed by them. As has been shown, certain organisms, especially the tubercle bacillus and probably the typhoid organism, attract the lymphocytes, while most of the other organisms attract the polymorphonu- clear neutrophiles. Theoretically, therefore, it should be possible to decide as to a tubercular or nontubercular condition by the presence or absence PLATE XXXV. Exudate from Tubercular Pleurisy. (Eosin-hematoxylin Stain.) PLATE XXXVI. Exudate in Pneumonic Pleurisy. (Eosin-hematoxylin Stain.) TRANSUDATES AND EXUDATES. 641 of an increased number of the mononuclear types of leucocytes. This is the basis of the attempt at differential diagnosis by means of cytodiagnosis. Not infrequently one finds in malignant conditions the so-called specific cells which are either sarcomatous or carcinomatous. These, although specific, are not easy absolutely to identify. These cells are very large, frequently showing fatty degeneration, extensively vacuolated, and showing a mitotic mulberry-like nucleus. Although differing from the endothelial cell, confusion is very apt to arise, so that it is difficult to make a diagnosis in all cases from the appearance of such cells. Pleural Exudates. Primary Tubercular Pleurisy. This is characterized especially by an increase, both relative and absolute, in the number of lymphocytes. A pleural lymphocytosis exists when there is an excess of mononuclear cells, with abundant protoplasm, a large nucleus, and smaller than the endothelial cells. In the very early days of the infection a neutrophilia may exist, but this is rarely seen, as attention may not be drawn to the condition sufficiently early. Associated with these polynuclear cells in the early stage we may find an increase in the number of endothelial cells. Eosinophile cells are frequently observed, but do not have any definite relation to tuberculosis as an infection. The red cells may be occasionally numerous, but are usually small in number. As a rule, tubercular pleurisy secondary to a pulmonary tuberculosis yields a liquid which is poor in cells, practically all of which are very much altered and in some cases very difficultly recognizable. The polynuclear types may predominate to such an extent that a distinct polynucleosis exists as evidence of a septic rather than a true tubercular pleurisy. The polynuclear cells are usually old, much deformed, and their nature recognizable only by staining their neutrophile granules. Where the infection is directly tubercular and not mixed (the latter, however, usually being the case), an approximately equal division of the polynuclear and mononuclear forms may obtain. In this type of pleurisy the eosinophile cells may be very numerous, in one case of Widal and Ravaut constituting 54 per cent, of the cellular elements. The endothelial cells may be, as in the primary tubercular pleurisy, sufficiently numerous to constitute a distinct endotheliosis. They are, however, single and very rarely grouped in masses. Secondary Tubercular Pleurisy. Pneumococcus Pleurisy. This is a truly septic type of pleurisy and is characterized by a distinct polynucleosis. In the early stages of this type of pleurisy, the endothelial cells may be very numerous, while in the later stages they may be much dimin- ished. In this, as in all types of pus accumulations, marked autolysis is present so that the cells may show extreme degeneration. As this condition progresses toward recovery, some of the polynuclears may be replaced by the mononu- clear lymphocytes; while if suppuration becomes extensive the polynuclears 642 DIAGNOSTIC METHODS. increase and autolysis becomes extremely marked. In such exudates the pneumococcus may be demonstrated by staining methods. Streptococcus Pleurisy. This type is especially associated with a polynucleosis. These cells are frequently observed undergoing karyolysis, the cell body usually being markedly degenerated. In its early stage it may be accompanied by an endotheliosis, but when well developed is usually associated with the presence of only a few isolated endothelial cells. Stained smears show streptococci in large numbers. Typhoid Pleurisy- In this type a lymphocytosis is usually observed along with an endothe- liosis, these latter cells being in large masses instead of in single isolated forms as observed in the secondary tubercular pleurisy. This point may be valuable in differentiating these two conditions, which are associated with an increase in the number of lymphocytes. The eosinophiles may be increased and red cells may be present in large numbers. The identification of the specific organism will serve as a positive differentiation from tubercular conditions. Malignant Pleurisy. This is a type of the aseptic pleurisies and may accompany malignant growths of the lung or pleura. Nothing characteristic is found in the cytologic formula of such exudates, but occasionally portions of the tumor mass may be obtained or certain cells may be found which are more or less distinctive, although not absolutely pathognomonic. These cells are frequently confused with the larger endothelial cells of the pleura, but are characterized, according to Deguy and Guillaumin, as follows: Malignant pleural cells contain glycogen which is recognized by the brown coloration shown on treatment with dilute iodin solutions, they contain large amounts of fatty material, they are ex- tensively vacuolated, these vacuoles may be large or so numerous and small that the cell resembles a sponge, the cells are extremely large, and the nucleus usually shows mitotic figures.1 Nephritic and Cardiac Pleurisy. The secondary exudate observed in cardiac and renal conditions is charac- terized especially by the marked endotheliosis. These endothelial cells are grouped in masses of 5 to 10 cells, which show more or less degeneration, their contours and their large nuclei being distinct. This endotheliosis is not com- plicated by the presence of many other cells in the pure nephritic pleurisy, but in the cardiac type we usually find a polynucleosis at the same time. This polynucleosis is much more marked where infarcts or emboli have complicated the condition than when it is due to a pure congestion. Numerous red cells are especially observed in association with a congestive pleurisy. Peritoneal Exudates. The cytological examination of peritoneal exudates has, as yet, yielded fewer diagnostic points than has that of pleural exudates.2 It is, however, 1 See Warren, Arch, of Int. Med., vol. 8, ign, p. 648. 2 See Szecsi, Folia Haemat., Bd. 13, 1912, S. 1. PLATE XXXVII. Exudate in Malignant Pleurisy. (Eosin-hematoxylin Stain. After Deguy and Guillaumin.) TRANSUDATES AND EXUDATES. 643 sometimes possible to differentiate a tubercular peritonitis from an ascites or an ovarian cyst by means of such examinations. Tubercular peritonitis usually shows a lymphocytosis and also a relative polynucleosis. A few endothelial cells may occasionally be found, but these do not yield much information. In ascites of hepatic origin few cellular elements are observed beyond the peritoneal endothelial cells. In ovarian cysts there are fewer cellular elements as a rule, but those present are usually large, round, or oval, and filled with vacuoles. Moreover, cylindrical ciliated epithelial cells as well as goblet cells or red cells are frequently relatively numerous. V. Cyst Fluids. (1). Ovarian Cysts. Such cysts are true retention cysts formed by dilatation of the Graafian follicles and retention of the ovarian secretion and are known as Hydrops folliculorum Graafii. They contain a clear, watery, serous liquid, which has an amber color, a specific gravity ranging between 1005 and 1022, and a chemical composition practically identical with that of other serous fluids. (a). Serous Cysts. (6). Myxoid or Colloid Cysts. These are proliferating cysts developed from the epithelial tubules. "We sometimes find in small cysts a semisolid, transparent, or somewhat cloudy or opalescent mass which appears like solidified glue or quivering jelly, and which has been called colloid because of its physical properties. In other cases the cysts contain a thick, tough mass which can be drawn out into long threads, and, as this mass in the different cysts is more or less diluted with serous liquids, their contents may have a variable consistency. In other cases the small cysts may also contain a thin watery fluid. The color of the contents is also variable. In certain cases it is bluish-white, opalescent, and in others yellow, yellowish-brown, or yellowish with a shade of green. They are often colored more or less chocolate-brown or reddish-brown, due to the decomposed blood pigment. The reaction is alkaline or nearly neutral. The specific gravity, which may vary considerably, is generally 1015 to 1030, but may be, in a few cases, 1005 to 1010 or 1050 to 1055. Though the contents of the proliferating cyst may have a variable composition, still it may be charac- terized, in typical cases, by its slimy or ropy consistency; by its grayish-yellow, chocolate-brown, sometimes whitish-gray color; and by its relatively high specific gravity. Such a liquid does not ordinarily show a spontaneous coagula- tion" (Hammarsten). Microscopical examination of the sediment shows red and white blood- cells, large epithelial cells, which may be filled with vacuoles, cylindrical or goblet cells, granular cells showing more or less fatty degeneration, fatty granules, cholesterin crystals, and colloid corpuscles in the form of large, circular, highly refractile bodies. 644 DIAGNOSTIC METHODS. Chemically, these cysts are characterized by the presence of colloid, which is not a distinct chemical entity. It is a gelatinous substance, insoluble in water and acetic acid, soluble in alkalies, and yields a reducing body on boiling with acids. Not infrequently pseudomucin (metalbumin) is found, es- pecially in the extremely viscid fluids. For its detection the serum albumin must be previously removed by the additipn of acetic acid, boiling, and filtering. The filtrate is treated with alcohol when a thready precipitate forms. If this precipitate be boiled with HC1, a substance is formed which reduces copper solutions quite markedly. Pseudomucin is distinguished from true mucin by the fact that it is not precipitated by acetic acid. Mitjukoff has isolated a further colloid body from certain ovarian cysts, to which he gives the name of paramucin. It is precipitated by acetic acid and is soluble in an excess. If treated with alkali it first swells up and then dissolves in an excess of the reagent. It differs from mucin and pseudomucin in the fact that it reduces copper solution without previous boiling with acids. McConnell1 has recently reported the finding of a multi-locular cyst of the ovary, which contained true mucin. These are intraligamentary types and contain a yellow, yellowish-green, or brownish-green fluid which contains only traces of pseudomucin. The specific gravity ranges between 1032 and 1036. (c). Papillary Cysts. These may be seen in the shape of small cysts not larger than a pea, but usually they are much larger, in some cases reaching the size of a man's head. The cyst usually contains a fatty unctuous material, which is derived from the epidermal lining of the cyst, and associated with it fat, desquamated epithelial scales, hair, teeth, bone, cartilage, etc. (d). Dermoid Cysts. (2). Parovarian Cysts. These cysts of the organ of Rosenmuller contain a clear, pale yellow, or col- orless, limpid fluid or occasionally one showing slight opalescence. The specific gravity ranges from 1002 to 1010 and differs from that of the ovarian cyst by its usual limitation to these lower figures. Albumin may be present in small amounts or be entirely absent, while pseudomucin is rarely, if ever, present. (3). Hydrocele. The contents of such a cyst are usually clear, show a color which may range from yellow to green, have a specific gravity of 1015 to 1030, and usually coagulate spontaneously. They contain a relatively large percentage of albumin, of which about 50 per cent, is globulin. In ordinary hydrocele many large oval cells may be seen, which have an eccentric nucleus and may be grouped in masses, although more frequently they appear as isolated cells; in some cases many eosinophiles, but this is rare in uncomplicated hydroceles. If the hydrocele be of tubercular origin a marked lymphocytosis usually exists. 1 Jour, of Med. Res., vol. 20, 1909, p. 105. TRANSUDATES AND EXUDATES. 645 (4). Spermatocele. Fluids from such cysts are usually thin, colorless, and cloudy like thin milk. They may have an acid reaction, but are ordinarily alkaline. The specific gravity ranges between 1006 and 1010. Such fluids do not coagulate either spontaneously or after the addition of blood. Microscopically, one observes cell detritus, fat granules, leucocytes, and spermatozoa. (5). Hydronephrosis. This is a true retention cyst of the kidney, due primarily to obstruction of the ureter, which may be either congenital or acquired. Material aspirated from a renal cyst is usually clear, but may be yellowish or reddish and distinctly turbid. Its specific gravity varies from 1010 to 1015, while the chemical composition is usually suggestive of urine. For some time the presence of urea and uric acid in hydronephrotic cysts was supposed to be pathognomonic, but it has been shown that these substances may be present both in ovarian and pancreatic cysts and may even be lacking in old renal cysts. Occasionally epithelial cells, derived from the uriniferous tubules may be found, but such cells are not always present nor are they sufficiently characteristic to be of great importance from the diagnostic standpoint. (6). Hydatid Cysts. The fluid obtained by puncture of an echinococcus cyst is usually clear and of an alkaline reaction, has a specific gravity varying between 1005 and 1015, is practically free from albumin, and contains a large amount of inorganic salts, especially of sodium chlorid. The characteristic findings of such a cyst are the booklets, scolices, and shreds of faintly laminated membrane. In some cases no trace of any morphological elements can be found, but usually the diagnosis is rendered certain, especially if careful search be made, by the presence of some portion of the parasite or the cyst membrane. (7). Pancreatic Cysts. The puncture fluid from a pancreatic cyst varies in its physical properties, depending upon the nature of the cyst as well as the length of time the fluid has remained in the cyst. It is usually bloody in character, has a specific gravity ranging from 1010 to 1030 and may contain methemoglobin, hematin, and cholesterin. As characteristic constituents of such a cyst one finds fer- ments, which will digest all types of food material.1 Such tests may be per- formed as outlined in the sections on Gastric Contents and Feces. VI. Cerebrospinal Fluid. Since the introduction of lumbar puncture by Quincke, the cerebrospinal fluid has gained more or less importance from the diagnostic point of view. So much may be learned, either from the standpoint of direct or differential diagnosis, that every practitioner should be able to perform a lumbar puncture and to examine the fluid obtained. There is little danger in the procedure 1 See Christian (Arch, of Int. Med., vol. 9, 1912, p. 143) who reports peculiar findings in the fluid of an epigastric cyst. 646 DIAGNOSTIC METHODS. as the spinal cord does not reach to the point of puncture and the fibers of the cauda equina are sufficiently movable to escape the needle. While few bad effects are observed in the ordinary run of cases, a few have been reported in which symptoms of collapse were evident. It should be a rule, therefore, to stop proceedings if such symptoms arise and also to keep the patient quiet in bed for at least 24 hours following the puncture, so that the pressure in the cerebrospinal cavity may become equalized. Lumbar Puncture. The patient is placed upon his left side near the edge of the bed, the knees should be flexed upon the abdomen, and the site of puncture prepared as for any surgical procedure. The needle used for puncture should be from 5 to 10 cm. long and have a lumen from 1 to 2 mm. in diameter. It is always wise to provide the larger needles with a stylet, so that tissue fragments or blood may not gain entrance accidentally to the tube and thus lead to possible diagnostic errors. This stylet may be removed the moment the needle pene- trates the dura mater. The site of puncture should be on a level with the junction of the third and fourth lumbar vertebrae at a point about 1 cm. to the side (preferably the upper) of the median line. The needle is directed slightly upward and inward, the depth to which the puncture should be made varying with the age of the patient, the younger the child the less the depth. This puncture should be made carefully and yet with sufficient force to penetrate easily the muscula- ture. If any marked resistance arises, it is probable that the needle has struck the vertebra, in which case the pressure must be reduced or the needle may break. This is not an infrequent oc- currence with those not used to the technic, so that it may be advisable for the student to practice the procedure upon the cadaver. As soon as the dural sac is reached, the cerebrospinal fluid will flow from the canula, the rate of flow indicating in a general way the pressure of the fluid. No aspiration should be used at any time, as this procedure is extremely dangerous. It is frequently advisable to know exactly what this pressure is, so that one may resort to the following method as used by Sahli. As soon as the needle nenetrates the dura, a connection is made with a mercurv manometer Fig. 159.-Lumbar puncture: A, Quincke's site; B, Maran's site; c, Chipault's site. {Tyson ) TRANSUDATES AND EXUDATES. 647 by means of a rubber tube filled with a x per cent, solution of carbolic acid. The portion of the manometer above the level of the mercury, forming the con- nection between it and the carbolic acid tube, must also be filled with the fluid. The manometer is filled with mercury to the zero point and held in such a manner that this point is on a level with the point of the aspirating needle, which is possible with ordinary manometers only when the connecting tube is of considerable length. Under normal conditions the dural pressure, in the dorsal position, ranges between 5 and 7.5 mm. of mercury, or 60 to 100 mm. of water if a water manometer be used. In pathologic conditions, such as meningitis or brain tumor, it ranges between 15 and 60 mm. of mercury or 200 to 800 mm. of water. Normal cerebrospinal fluid is colorless, limpid, and free from morphologi- cal elements. Its specific gravity ranges between 1002 and 1010. It is alkaline in reaction, the degree of alkalinity varying between 15 and 20. It contains a trace of protein and about 0.1 per cent, of glucose. The salt content of this fluid is, according to Zdarek, 0.836 gram, of which 0.429 gram is referable to sodium oxid and 0.017 to potassium oxid. This would rather militate against the older statements that a large amount of potassium salts as com- pared with sodium salts was present. The relation of KC1 to NaCl is, accord- ing to Nawratzki, 1 to 18, while Zdarek gives this ratio as 1 to 40. The former figure agrees closely wfith that obtained by various workers with pathological cerebrospinal fluids. The amount of fluid obtained by lumbar puncture is extremely variable. In normal individuals this amount is unknown as puncture is rarely performed upon normal cases. Pathologically, the amount obtained varies between 10 and 100 c.c. Naturally, if the communication between the subarachnoid spaces of the brain and of the spinal cord is blocked by a tumor or inflammatory adhesions, or if the aqueduct of Sylvius or the foramen of Magendie be obliterated, little fluid may be obtained by puncture, although large amounts may be present above the obstruction. The largest amounts are seen in cases of serous or tubercular meningitis, so that such conditions may usually be ruled out if a small amount of fluid is obtainable. Pathologically, we may observe a very cloudy fluid, due to the presence of leucocytes, erythrocytes, and endothelial cells. This cellular admixture may be so extensive that the fluid resembles pure pus. In cases of cerebral hemorrhage from the ventricles, hemorrhagic pachymeningitis, or traumatic lesions of the spinal cord, so much blood may be present as to give the appear- ance of practically pure blood, the color varying from a bright red to a brownish or greenish-red, depending upon the length of time it has remained in contact with the remaining portion of the fluid. This admixture with blood may lead to the spontaneous coagulation of the fluid. This may serve as a differen- tiating point between inflammatory and noninflammatory lesions. Thus, in tubercular meningitis very slight coagulation may be observed, while in the epidemic cerebrospinal meningitis the coagulum may be very firm. The chemical examination of the cerebrospinal fluid offers some points of 648 DIAGNOSTIC METHODS. clinical value. While the albumin content normally is much less than o.i per cent., it may vary under pathologic conditions to as high as 0.8 per cent. The total protein, especially the euglobulin portion, is increased in all cases of acute exudative inflammations of the meninges, in hydrocephalus as Polanyi1 has shown and, also, in syphilitic and parasyphilitic diseases of the cerebrospinal tract. Glucose is usually present but may entirely disappear under pathologic influences due to the autolysis controlled by the leucocytic ferments, the glucose being converted into lactic acid. Cholin is present normally in traces, while pathologically it may vary, according to Donath, between 0.021 per cent, and 0.046 per cent. Koch has introduced the following method for examination of this fluid, which should extend our knowledge. The fluid is centrifuged and the sediment used for the bacteriological and cytological examinations. Ten c.c. of this clear fluid are placed in a beaker, 3 c.c. of a saturated solution of ammonium nitrate and 5 drops of nitric acid are added, and the mixture heated on a water- bath until complete coagulation has occurred. This precipitate of protein2 is filtered through a perforated Gooch crucible, washed with 0.1 per cent, ammo- nium nitrate solution and then with alcohol, dried at ioo° C. and weighed. The filtrate is taken before the washing with alcohol is begun, concentrated to 10 to 15 c.c. and 2 c.c. of 1 per cent, barium nitrate solution added while boiling. The barium sulphate precipitate is allowed to settle over night, is filtered off, dried and weighed. With the filtrate a phosphorus estimation is made by the molybdic method, with subsequent precipitation as magnesium ammonium phosphate in the usual way. Increase in the phosphorus and sulphur content is associated with degenerative changes in the nerve tissue. A test for reducing sugar is made with 1 c.c. of the original fluid. Microscopic Examination. By far the most important part of the clinical examination of cerebro- spinal fluid is a study of the bacteriology and cytology of the fluid. Normally the fluid contains practically no morphologic elements, while under pathologic conditions large numbers of various types of cells may be present. The material for examination is obtained as previously described in the section on cytology. In true tubercular meningitis a lymphocytosis is almost invar- iably observed, while in the epidemic type, due to the meningococcus of Weich- selbaum, the cells are of the polynuclear type. In the chronic cases of epidemic meningitis as well as during convalescence from this disease, the lymphocytes may be present to such an extent that a slight degree of lymphocytosis is present, but never to the same extent as observed in the tubercular type. A lymphocy- tosis is also observed in syphilitic lesions of the central nervous system. This is important from the standpoint of differential diagnosis. In the meningitis due to the pneumococcus, a polynucleosis is the rule, although occasional 1 Biochem. Zeitsch., Bd. 34, 1911, S. 205. 2 Mott advises the following method for protein: 10 c.c. of the centrifuged fluid are faintly acidified with acetic acid and 20 c.c. of absolute alcohol are added. The mixture is boiled and filtered, the precipitate of protein being dried and weighed. TRANSUDATES AND EXUDATES. 649 cases are seen in which a lymphocytosis obtains. In the prodromal stages and early days of acute anterior poliomyelitis a marked increase of fluid is accom- panied by an increase in number of cells, which are largely of the mononuclear type. These are replaced by the polymorphonuclear forms as the number diminishes. In the microscopic examination of the cerebrospinal fluid it is sometimes of interest to know the number of cells per cu. mm., of the fluid. The technic is as follows: The following mixture is employed for staining the white and dissolving the red cells. Methyl violet 0.2 gram, glacial acetic acid 5 c.c., water to 100. Fill the pipet with stain to the mark 1 and then fill to point 11 with uncentrifuged cerebrospinal fluid. Shake thoroughly, let stand for five minutes and count, using a Fuchs-Rosenthal counting chamber. The ruled surface of the cell contains 3.2 cu. mm. of fluid. Count all of the white cells in the entire ruled area, multiply by 11 and divide by 32. The result is the number of cells per c.c., which varies from 1 to 10 normally. Epidemic Cerebrospinal Meningitis. In these cases the fluid may be transparent, but is usually somewhat opalescent and may be thick and purulent. The cellular elements are usually polynuclear in type and red cells may be more or less numerous. Smears made from the sediment show the presence of numerous diplococci, which closely resemble the gonococcus in morphological and staining character- istics. This organism, the diplococcus meningitidis intracellularis of Weichsel- baum, appears as a diplococcus, each element forming a hemisphere with its parallel side contiguous to that of its mate.1 It is sometimes seen in the form of tetrads or as isolated cocci, which appear as true spheres of variable size and showing a clear space in their interior. It is stained with the ordinary dyes and is negative to Gram's stain. For its cultural peculiarities the reader should consult works on bacteriology. Not infrequently one observes specimens of the meningococcus which show a Gram positive reaction, so that it is difficult to distinguish them, especially when they are in the form of isolated cocci or in groups of two or four from the pneumococcus. This type has been described as the meningococcus of Bonome, while Jaeger and Heubner describe a diplococcus which is Gram-positive and may be confused with the unusual types of Weichselbaum's meningococcus. Recently Flexner has succeeded in pre- paring an anti meningococcic serum which appears to have remarkable results in controlling this hitherto unmanageable disease. In using this serum, the injection must be made directly into the spinal cavity. More or less frequent injections of the serum and examina- tions of the lumbar fluid are made and the influence of the serum estimated Fig. 160.-Diplococcus intracellularis meningitidis. (Councilman.) 1 See Frost, Pub. Health Rep., vol. 27, 1912, p. 97. 650 DIAGNOSTIC METHODS. by bacteriological and cytological examinations. In the cases of mixed cerebral infection, in which the meningococcus is associated with the pneumo- coccus, streptococcus, typhoid bacillus, staphylococcus, and other organisms, this serum does not seem to have as much influence as in the pure meningo- coccus infections.1 It is to be remembered that a purulent meningitis may be secondary to infection with practically all of the pus-forming organisms found within the system. It is, therefore, essential that any infection showing meningeal symptoms should be investigated by an examination of the cerebrospinal fluid. In sleeping sickness, a study of the cerebrospinal fluid frequently reveals the presence of the trypanosoma Gambiense. These parasites are not found in all cases, but when present usually furnish a grave prognosis. Recent investigation of the cerebrospinal fluid, applying Wassermann's serum reaction for syphilis, has shown that in the large majority of nervous cases of syphilitic origin a positive reaction is obtainable. Tubercular Meningitis. The fluid in such cases is usually clear, but may be slightly opalescent. The cellular elements are largely mononuclear while a few red cells may be present. If tubercle bacilli cannot be found after careful search, the presence of a lymphocytosis along with an increased protein-content will be at least sug- gestive, while animal inoculation or the tuberculin test will clear up the diag- nosis in most cases. (See Lucas2). Acute Anterior Poliomyelitis. This disease has been definitely classified as infectious through the work of Flexner and his associates. The causative factor has not, as yet, been isolated, but it is known to belong to the very minute ultramicroscopic types of organism, of which the point of ingress and egress is the nasal and pharyngeal mucosa.3 The changes in the spinal fluid, while not absolutely diagnostic are sugges- tive. The amount of fluid usually increases early in the disease, this increase being associated with a relative increase in the mononuclear types. These cells may first be of the small lymphocytic variety but they soon change to the large mononuclear form. As the disease progresses, the cells diminish in num- ber, the polymorphonuclear forms assuming the ascendency. The total protein is increased in amount (as detected by the Noguchi or Ross-Jones method), while fibrin formation appears to be an early sign, although this may soon disappear.4 1 Collignon and Pilod (Presse med., Tm. 19, 1911, p. 732) advocate the following test to differentiate meningococcic from other types of meningitis. To a few c.c. (50 drops) of centrifuged spinal fluid add one to four drops of antimeningoccus serum and incubate the mixture (along with a control tube containing only spinal fluid) at 56° C. for 12 hours. If the serum and spinal fluid be homologous, a definite precipitation will be observed. This test appears to be specific. 2 Amer. Jour, of Dis. of Chil., vol. 1, 1911, p. 230. 3 Rosenau has demonstrated that the virus of anterior poliomyelitis is transmitted by means of the biting stable fly (stomoxys calcitrans). See Mitzmain (Pub. Health Rep., vol. 28, 1913, p. 345) for a discussion of the life history of this fly. Anderson and Frost (Pub. Health Rep., vol. 27, 1912, p. 1733) have confirmed these findings. Howard and Clark (Jour, of Exp. Med., vol. 16, 1912, p. 850) show its transmission by the common fly and the bed bug. 4 See "A Clinical Study of Acute Poliomyelitis" by Peabody, Draper and Dochez, pub- lished by the Rockfeller Institute in 1912. TRANSUDATES AND EXUDATES. 651 Cerebrospinal Syphilis. Although such cases do not show pathognomonic findings in the cerebro- spinal fluid, yet certain factors are of importance in diagnosis. A lymphocyto- sis is usually observed together with a marked increase in the globulin content, which is estimated as follows:1 Noguchi's Butyric Acid Test. One part (0.1 or 0.2 c.c.) of spinal fluid is mixed with 5 parts (0.5 c.c.) ot a 10 per cent, butyric acid solution in physiologic salt solution. This mixture is heated to boiling and immediately 1 part (0.1 c.c.) of normal (4 per cent.) sodium hydrate solution is added and the mixture again boiled for a few seconds. The presence of an increased content of protein in the fluid is indicated by the appearance of a granular or flocculent precipitate which gradually settles under a clear supernatant liquid. This precipitate appears within a few minutes in a specimen containing a considerable increase in protein, while two hours may be required to obtain a distinct reaction in specimens weaker in protein. May2 has called attention to the advisability of using larger portions of the fluid, preserving the same proportions of reagents as used by Noguchi. This reaction appears regularly in the cerebrospinal fluid of patients with syphilitic and parasyphilitic affections and also in all cases of inflammation of the meninges caused by such organisms as the meningococcus, pneumococcus, influenza bacillus, tubercle bacillus, etc. These latter cases are, however, easily differentiated from the syphilitic affections. Normal fluid gives a tur- bidity but the granular precipitate does not occur at all or only after many hours. Although this reaction is given in many non-syphilitic conditions, it has a certain value. It is not specific and, when present, does not necessarily indicate a syphilitic infection. On the other hand, it can be employed to establish or confirm a deduction based upon the clinical history and the results of the .Wassermann reaction and cytodiagnosis, thus becoming of great indirect diagnostic value. A syphilitic infection is, however, practically excluded by a negative reaction. In this respect this test has advantages over the Wassermann reaction, in which a negative result is not always reliable as indicating absence of syphilitic infection.3 BIBLIOGRAPHY. 1. Adami. Principles of Pathology. Philadelphia, 1908. 2. Kraus und Levaditi. Handbuch der Immunitatsforschung. Jena, 1908. 3. Jordan. General Bacteriology. Philadelphia, 1908. 4. Mestrezat. Le Liquide Cephalo-Rachidien. Paris, 1912. 5. Oppenheimer. Handbuch der Biochemie. Jena, 1908. 1 See Kaplan and Casamajor, Arch, of Int. Med., vol. 9, 1912, p. 262; also, Ball, Jour. A. M. A., vol. 59, 1912, p. 1272. 2 Arch, of Int. Med., vol. 8, 1911, p. 183. 3 Nichols and Hough, Jour. A. M. A., vol. 60, 1913, p. 108, report the demonstration of spinchseta pallida in the cerebrospinal fluid and the successful inoculation of a rabbit therefrom. CHAPTER X. SECRETION OF THE MAMMARY GLANDS. I. General Considerations. The normal secretion of milk takes place in the mammary glands of the female after delivery. It is true that a small quantity of milk may be secreted by the new-born of both sexes for a few days after birth. Moreover, cases have been reported in which the adult male secreted sufficient milk to act as a wet-nurse, but these must be regarded as cases of extreme rarity. During the course of a normal pregnancy a small amount of a thin, yellowish fluid may be expressed from the mammary glands, but as a rule the first real secretion is observed following delivery of the child. This secretion is thin and watery, more or less translucent, and shows a distinct yellowish color. Fig. 161.-Normal Milk and Colostrum. (Hawk.) a. Normal milk; b, colostrum. Microscopic examination shows the presence of rather large cells in which are many fat granules and occasionally a distinct nucleus. This secretion, which is called colostrum, continues for three to four days and is distinguishable from the later secretion by the fact that it contains relatively more salts than normal milk and, according to the usual statements, more sugar. In a series of determinations of breast milk collected during the fourth day after delivery, the writer could not show any marked increase in the sugar in all cases, but in three out of eight cases examined a percentage higher than eight of lactose was obtained. 652 SECRETION OF THE MAMMARY GLANDS. 653 It is probable that the function of the colostrum is gradually to accustom the child to the taking of food by diminishing to a slight extent the elements, protein and fat, which are more apt to cause digestive disturbances than is the lactose. The secretion of true milk begins about the fourth day and continues for a variable length of time. Marked variations are observed both in the quantity and the quality of this secretion in various women so that no hard and fast rule can be given as to the composition of normal mother's milk. While it is probably true that the normal woman should nurse her child during a large part of the first year, it is rare to find, especially in private practice, many such cases? Either the milk becomes scanty and loses in nutritive power or becomes excessive and consequently diluted. In either case the child is not receiving the most suitable nourishment, so that breast-feeding is abandoned under these circumstances. It seems to be a general rule, which is difficultly explicable, that the more highly socially developed the woman the less apt she is to nurse her child with any success. It is possible, and this seems to the writer the most probable explanation, that the child usually receives more or less constant attention from the physician, and variations in the breast milk are more frequently noticed than in the case of the poorer child who rarely has the advantage of medical attention unless more or less serious illness occurs. II. Physical and Chemical Properties. Various figures have been given for the composition of human milk so that it is difficult to strike an average. So much depends upon the nourishment of the mother, upon the amount of exercise taken by her, and upon the general condition of the system that marked variations exist in the proportion of the chemical constituents. The following table, taking averages of examinations reported by various writers, may serve as one representing more nearly the normal condition.1 Water, 87.24 Salts, 0.26 Protein, 1.50 Fat, 4.00 Lactose, 7.00 These figures are not as high as regards protein as those usually given, but are higher as far as the lactose is concerned. Although many writers have given the percentage of protein much higher than 1.5 per cent., the writer has never been able, in several hundred examinations of milk in his laboratory, to find many showing a protein percentage of two or more. As it is frequently necessary to modify cow's milk so that it may more 1 See Schloss, Monatsschr. f. Kinderhkde., Bd. io, 1912, S. 499. 654 DIAGNOSTIC METHODS. nearly approach mother's milk in composition, the writer inserts the following table for comparison. Water, 87.25 Salts, 0.75 Protein, 3.50 Fat, 4.00 Lactose, 4.50 It will be seen that cow's milk shows a higher per cent, of protein and a lower per cent, of carbohydrate. It is necessary, therefore, that this be modified by diluting the milk so as to diminish the protein and by adding lactose to make up for the deficiency of carbohydrate. In the dilution the fat content will necessarily be lowered so that this may be remedied, as sug- gested by Backhaus, by the addition of cream. The writer must refer to works on pediatrics for the various methods of modifying cow's milk. It is to be remembered that no modification is equal to mother's milk. The casein of human milk forms a much finer clot with the gastric juice than does that of cow's milk so that the latter may not be well tolerated by the child. More- over, some unknown principle present in human milk is accountable for a distinct biologic difference in these two types of nutritive material. (1). Appearance and Color. Normal human milk is a white fluid which usually has a slight bluish tinge except immediately after birth when the color may be distinctly yellowish from the presence of colostrum. If the percentage of fat be relatively high the color will be more nearly a pure white with little of the blue tone. (2). Specific Gravity. The specific gravity of human milk vanes between 1028 and 1034. An increase in the percentage of fat will usually lower the specific gravity while the protein and carbohydrates will increase it. Cow's milk shows approximately the same specific gravity. If a low specific gravity is obtained it is evidence either of an increased percentage of fat or of dilution with water. If the percentage of fat be low the milk is unquestionably a watered one. If the milk tested shows a high specific gravity the chances are that most of the fat has been removed either by skim- ming or by centrifugation. The determination of the specific gravity may be made by a special in- strument known as Quevenne's lactodensimeter or by the use of the ordinary hydrometer used in urine work. As it is never of any clinical importance or even of any marked practical importance that the specific gravity should be absolutely accurately determined, the writer is accustomed to use the urinometer for such determinations. Corrections for variations in tempera- ture are necessary only when great differences exist between the tempera- ture of the room and the temperature at which the instrument is calibrated. SECRETION OF THE MAMMARY GLANDS. 655 (3). Reaction. Normal human milk as well as cow's milk shows an amphoteric reaction to litmus-paper and an acid reaction to phenolphthalein, cow's milk being somewhat more acid than mother's milk toward the latter indicator. (4). Coagulation. If milk be allowed to stand, the reaction gradually becomes more and more acid owing to the development of bacteria, especially of the bacillus acidi lactici. When the degree of acidity reaches a certain point, casein sepa- rates first in the form of flocculi and later the entire fluid may coagulate to a jelly-like mass. This mass soon contracts and settles out leaving a slightly turbid fluid known as milk plasma or acid whey. In order to inhibit the development of bacteria and prevent this coagulation, known as souring, certain preservatives are frequently added to market milk and should be capable of detection by the practitioner. These will be discussed in a later section. Besides this type of coagulation of cow's milk a second form is observed which takes place under the influence of chymo in without any change in the reaction of the fluid. In this case the whey is sweet and contains practically all of the lactose originally present in the milk. (5). Total Solids. Five to ten c.c. of the well-mixed milk are placed in a weighed platinum dish, evaporated to dryness on a water-bath, and dried to constant weight in the oven at 1050. The difference in weight between the platinum dish and its contents, on the one hand, and the platinum dish, on the other, gives the amount of total solids in the milk taken. A simple calculation will yield the percentage of total solids. Under normal conditions the total solids of both human and cow's milk should average between 12 and 13 per cent. Variations in this figure are, of course, due to fluctuations in the various constituent elements. (6). Ash. The platinum dish containing the dried residue of the milk is heated over a direct flame until the residue is completely incinerated. This is then placed in the desiccator and dried to constant weight. The difference in weight between the di^h and contents and the dish alone represents the amount of salts present in the milk originally taken. (7). Protein. The methods for the determination of protein are divided into those for estimation of the total protein material present and into those in which separate determinations are made of the casein, on the one hand, and albumin and globulin, on the other. 656 DIAGNOSTIC METHODS (a). Total Protein. Ten c.c. of milk are diluted with 90 c.c. of water, 5 c.c. of a saturated so- dium chlorid solution, and 15 c.c. of Almon's tannic acid solution are added. The mixture is thoroughly stirred and the dense precipitate which forms is allowed to settle. The composition of Almen's tannic acid solution is as follows: Tannic acid, 4 grams. Acetic acid (25 per cent.), 8 c.c. Alcohol (50 per cent.), qs., ad., 200 c.c. The precipitate, which consists of the total protein of the milk and a large portion of the fat carried down by the precipitate, is then filtered off through a fine filter and is washed with cold water. The filter-paper and its contents are then placed in a Kjeldahl flask and a nitrogen determination made as described under Urine. It is advisable to use 20 c.c. of sulphuric acid instead of the ten employed in the case of urine, as the mixture oxidizes much more rapidly under these conditions. If-the nitrogen obtained in this determination be multiplied by 6.37, the result will be the protein in 10 c.c. of milk. Method of Sebelien. For routine work this method is perhaps more advisable for the general prac- titioner than is the preceding, but it does not always give confirmatory results. It is based upon the fact that the total protein of milk is precipitated by phospho- tungstic acid in hydrochloric acid solution, the amount of precipitate being measured in an Esbach tube. The reagent used has the following composition: Phosphotungstic acid, 25 grams. Concentrated hydrochloric acid, 25 c.c. Distilled water, q. s., ad., 250 c.c. It has been found that the milk should be diluted with water before adding the reagent if the results are to be accurate. As a rule, a dilution of 1 to 10 for human milk and 1 to 20 for cow's milk suffices. The diluted milk is poured into the Esbach tube1 to the mark U and the reagent added to the mark R. The tube is then closed with a stopper and inverted several times thoroughly to mix the contents. It is then set aside for 24 hours when the percentage of protein in the milk is read off directly from the calibrations on the tube in case the dilution was 1 to 10, while with a dilution of 1 to 20 the figures are multiplied by 2. This method, while convenient, is open to the objection that many factors may influence the depth to which a precipitate settles. Moreover, the Esbach tubes reading as high as 12 parts do not give as satisfactory results as those with readings from one to seven. Such being the case the method must be used more for clinical purposes than for scientific estimations. Method of Boggs. 1 If a Purdy centrifuge tube be filled with the diluted milk to the io c.c. mark and the reagent added to 15 c.c., the amount of protein may be determined quickly by centrifuging for 3 minutes and reading the percentage directly. The figures obtained agree quite closely with those of the gravimetric method. SECRETION OF THE MAMMARY GLANDS. 657 (ft). Casein. Twenty c.c. of well-mixed milk are measured into a beaker and approxi- mately 380 c.c. of water are added. The mixture is thoroughly stirred and very dilute acetic acid added drop by drop with constant stirring until a flocculent precipitate is observed. When this point is reached a stream of carbon dioxid is passed through the mixture for one-half hour, after which the vessel is allowed to stand until the next day. The above part of the technic is directly applicable to cow's milk. If human milk is being examined it is necessary to heat the vessel to 400 C. during the passage of the carbon dioxid. After the mixture has stood overnight, it is filtered through a nitrogen- free filter and washed with water. The residue on the filter contains casein which is mixed with a portion of the fat present. The filter-paper and contents are then placed in a Kjeldahl flask and a nitrogen determination is made as previously described. Multiplication of the nitrogen value by 6.37 yields the amount of casein in the 20 c.c. of milk originally taken. (c). Albumin and Globulin. The filtrate from the above precipitation of casein contains the remainder of the protein material and the carbohydrate of the milk. This filtrate is placed in a porcelain dish and heated for a few minutes to the boiling-point. The protein material is coagulated and may be filtered through a nitrogen-free filter and washed several times with cold water. A nitrogen determination is then made and the value multiplied by 6.37 to obtain the amount of albumin and globulin present in the 20 c.c. of milk.. The filtrate from this latter precipitation contains the lactose, which may then be determined by titration with Fehling's solution, as described under Urine. It is to be remembered that 10 c.c. of Fehling's solution are reduced by 0.0678 gram of lactose, and not by 0.05 as in the case of glucose. (8). Fat. It is important in the determination of the fat content of milk that a thoroughly mixed specimen be examined. As the fat tends to rise to the surface of the milk, the fluid should be poured from one vessel into another several times to insure thorough mixing and an immediate measurement made of the portion to be tested. For clinical purposes as well as for examination of market milk the method of Babcock is to be recommended. For accurate results, however, this method is not to be advised. Babcock's Method. This method consists in the destruction of the organic matter, except the fat, by means of sulphuric acid. The fat is then separated by centrifuging and determined by reading off the percentage from the calibrations in the neck of the bottle used. In the case of cow's milk or where sufficient human milk may be obtained, 658 DIAGNOSTIC METHODS. 17.6 c.c. of milk are measured into the bottle and 17.5 c.c. of sulphuric acid added. These fluids are then mixed by shaking and rotating the bottle in such a way that no curds pass into the neck of the bottle. As soon as the mixture becomes homogeneous and dark brown or even black in color, the bottles are placed in a special centrifugal machine and whirled for five minutes. If the room be very cold it is advisable to fill the holders of the centrifuge with boiling water in order to keep the fat melted while it is being centrifuged. At the end of five minutes, centrifugation is discontinued and the neck of the bottle filled with boiling water. The melted fat will rise in the neck of the flask and may be read off from the cali- bration. In order to facilitate this the bottles are again centrifuged for one minute. If a small amount of milk only is available the smaller tubes shown in cut may be used. Milk is added to the mark five and sulphuric acid poured in so as to fill the body of the tube. It is usually necessary to add the milk and acid by means of thin narrow pipets, as the neck of the tube is too small to permit of easy entrance of the fluids otherwise. The milk and sul- phuric acid are mixed by rotation of the tube until a homogeneous fluid results. The mixture is centrifuged for a few minutes, the neck of the tube being filled with a mixture consisting of equal parts of concentrated hydrochloric acid and amyl alcohol. The percentage of fat is then read off from the calibra- tions on the tube. Extraction Method. Fig. 162.-Babcock Bottles. a, Milk bottle; b, cream bottle. A few grams of dried washed sand are placed in the extraction shell of a Soxhlet apparatus and io c.c. of well-mixed milk are allowed to fall upon it drop by drop. This is dried at a temperature of ioo° C. for one to two hours and is placed in the tube of the extraction apparatus. The fat is extracted in the usual way by the use of gasoline or anhydrous ether, complete extraction usually requiring from two to three hours. The apparatus is disconnected, the ether evap- orated from the distilling flask, the residue in the flask dried at ioo° C., and the flask and contents dried to constant weight in the desiccator. The differ- ence between the original weight of the flask and its weight including the ex- tracted residue, yields the amount of fat in the io c.c. of milk. This method Fig. 163.-Bottle for human milk. SECRETION OF THE MAMMARY GLANDS. 659 is the most accurate one, but is not as convenient as the preceding for the general practitioner. (9). Lactose. In general routine analyses of milk the lactose may be determined by difference. By this is meant that subtraction of the values for the sum of water, ash, protein, and fat from 100 will yield the percentage of lactose. For clinical purposes this is usually sufficient, but for the more ac- curate work it does not give exact figures, as slight amounts of other undetermined sub- stances are present. For any direct determination of lactose it is necessary that the larger portion of the protein material be removed previously. This may be done by the method outlined under Determination of Casein, Albumin, and Globulin; or for clinical purposes sufficiently accurate results may be reached by acidifying with acetic acid, boiling, and filtering. It is advisable always to take the time to saturate the mixture with carbon dioxid after the casein has been precipitated with acetic acid, as the results are more satisfactory. The fluid is then titrated by the use of Fehling's or Purdy's solution, using all the precautions mentioned under these tests in the section on Urine. Ten c.c. of Fehling's solu- tion are reduced by 0.0678 gram of lactose, while the 35 c.c. of Purdy's solution are re- duced by 0.02712 gram of lactose. If the polarimeter is to be used for the estimation of lactose in milk the following procedure may be used. Fifty c.c. of the well- mixed milk are placed in a flask, 25 c.c. of a solution of neutral lead acetate are added, the flask is closed with a stopper through which passes a glass tube approximately 30 cm. in length. The mixture is then heated over a small flame to boiling. After the mixture has cooled it is filtered through a dry filter into a dry vessel and polarized. Fig. 164.-Soxhlet apparatus. (Hawk.) (10). Preservatives in Cow's Milk. (a). Sodium Carbonate. To hide the acid reaction of a spoiled sample of milk sodium carbonate is often added, and may be detected as follows: Ten c.c. of milk are mixed with ten c.c. of 96 per cent, alcohol and a drop of rosolic acidluti soon. Pure 660 DIAGNOSTIC METHODS. unadulterated milk produces a brownish-yellow color, but, in the presence of sodium carbonate or bicarbonate, a rose color is obtained. For greater precision in doubtful cases the questionable sample should be compared with known unadultered milk. Phenol-phthalein solution may be used as an in- dicator in place of rosolic acid. By this method 0.05 per cent, of carbonates may easily be detected. (6). Salicylic Acid. Twenty c.c. of milk are treated with two or three drops of sulphuric acid and shaken with an equal amount of ether. The greatest possible part of the ethereal solution is drawn off and evaporated, the residue extracted with 40 per cent, alcohol, filtered, and 5 c.c. of the filtrate treated with a few drops of ferric chlorid solution. A violet color shows the presence of salicylic acid or some other hydroxy derivative of benzol. (r). Formaldehyd. Two c.c. of concentrated sulphuric acid are placed in a test-tube and a drop of ferric chlorid solution added. A few c.c. of milk are allowed to run from a pipet upon the surface of the mixture in such a way that a distinct line of contact forms. A violet color at the point of contact of the two liquids is characteristic of formaldehyd in the presence of casein. (d}. Boric Acid and Borax. Fifty c.c. of milk are alkalinized with milk of lime, evaporated to dryness and incinerated. The resulting white crystalline residue is treated with a few drops of tincture of turmeric and very dilute hydrochloric acid and is then dried on the water bath. The presence of the slightest trace of boric acid gives to the dry residue a beautiful vermilion or cherry red color. It is possible by this method to detect o.ooi per cent, of boric acid in milk. Only very dilute hydrochloric acid must be used in testing for boric acid, since the concentrated acid itself gives with tincture of turmeric a red color. The coloration produced by boric acid is distinguished from that produced by hydrochloric acid by the fact that it does not disappear by treatment with water in the cold, but only after long boiling, while the color caused by hydro- chloric acid disappears as soon as it is diluted with water. If the crystalline residue obtained as above described be treated with alcohol and the alcohol ignited, a flame tinged with a beautiful emerald-green color is obtained in the presence of boric acid. III. Bacteriological Examination of Milk. The bacteriological examination of human milk is frequently desirable from a clinical standpoint as milk may become contaminated as it passes along the lacteal ducts. In pathologic conditions many types of organisms, such as typhoid bacilli, pneumococci, and tubercle bacilli may be obtained, although the latter are extremely rare. SECRETION OF THE MAMMARY GLANDS. 661 With cow's milk the chief question at issue is whether the milk contains sufficient bacteria to be harmful to the child. Aside from the presence of the tubercle bacillus in the milk of infected animals, large numbers of saprophytic organisms must find their way into this fluid and will, if in large numbers, influence the intestinal activity of the child. The methods of examining milk for bacteriological differentiation are the same as for any other fluid and will be found in works on bacteriology. It is, however, a matter of some interest to the practitioner to determine the number of bacteria present, as a control of the sanitary conditions or as a check on the efficacy of pasteurization, without regard to the various types present. It is to be remembered that bacteria develop rapidly so that a specimen should be examined as fresh as possible in order to determine with more certainty the degree of original contamination. The method of perform- ing this enumeration may be found in general works on bacteriology. The number of bacteria in milk may vary from a few thousand to many millions. It is an impossibility to obtain a specimen which' is sterile so that the presence of as high as 100,000 organisms should not be regarded as danger- ous unless pathogenic types are present. If the freshly examined specimen contains between 10,000 and 50,000 bacteria per c.c., it is probable that all possible precautions have been taken to prevent contamination, although the writer has found some specimens showing a count of only 100 per c. c. In the examination of milk it is customary also to make a cytological examination. Normally milk contains only a few scattered leucocytes and epithelial cells. If examination reveals numerous leucocytes, absolute evidence is present of infection of the animal and consequent unfitness of the milk for use. The methods used in this cytological examination are the same as out- lined in previous sections. BIBLIOGRAPHY. 1. Czerny und Keller. Die Nahrungs-Pathologie. Berlin, 1907. 2. Holt. Diseases of Infancy and Childhood. New York, 1906. General Clinical Diagnosis. i. Boston. A Text-book of Clinical Diagnosis. Philadelphia, 1905. 2. Brown and Ritchie. Medical Diagnosis. Edinburgh, 1906. 3. Brugsch und Schittenhelm. Lehrbuch klinischer Untersuchungsme- thoden fiir Studierende und Aerzte. Berlin, 1911. 4. Butler. Diagnostics of Internal Medicine. New York, 1905. 5. Da Costa. Medical Diagnosis. Philadelphia, 1890. 6. Debove et Achard. Manuel de diagnostic medical. Paris, 1900. 7. Deguy et Guillaumin. Microscopic clinique. Paris, 1912. 8. Emerson. Clinical Diagnosis. Philadelphia, 1911. 9. French. Medical Laboratory Methods and Tests. Chicago, 1912. 10. Guiart et Grimbert. Precis de diagnostic. Paris, 1912. 11. Greene. Medical Diagnosis. Philadelphia, 1907. 12. Hare. Practical Diagnosis. Philadelphia, 1908. 13. Hutchinson and Rainy. Clinical Methods. Chicago, 1908. 662 DIAGNOSTIC METHODS. i4- Koranyi und Richter. Physikalische Chemie und Medizin. Leipzig, 1908. 15. Jagic. Klinische Mikroskopie. Wien, 1908. 16. von Jaksch. Klinische Diagnostik. Berlin, 1910. 17. Klemperer. Grundriss der klinischen Diagnostik. Berlin, 1906. 18. Lenhartz. Mikroskopie und Chemie am Krankenbett. Berlin, 1910. 19. Leube. Specielie Diagnose der inneren Krankheiten. Leipzig, 1908. 20. Musser. Medical Diagnosis. Philadelphia, 1904. 21. von Noorden. Handbuch der Pathologic des Stoffwechsels. Berlin, 1906. 22. Sahli. Lehrbuch der klinischen Untersuchungsmethoden. Wien, 1910. 23. Schilling. Aerztliche Technik. Wurzburg, 1906. 24. Simon. A Manual of Clinical Diagnosis. Philadelphia, 1911. 25. Todd. Manual of Clinical Diagnosis. Philadelphia, 1912. 26. Wesener. Medizinisch-klinisch Diagnostik. Berlin, 1907. 27. Wood. Chemical and Microscopical Diagnosis. New York, 1911. 28. Zuelzer. Chemische und Mikroskopische Diagnostik. Leipzig, 1906. INDEX. Abderhalden's sero-diagnosis of preg- nancy, 613 Abortion, 374 Abscess, blood in, 524 indican in, 243 of liver, sputum in, 33 of lung, sputum in, 33 Absorptive power of stomach, 82 Acanthia lectularia, 155 Acarus scabiei, 153 Accidental albuminuria, 247 Acetic acid in gastric contents, 59, 68, 72 Aceto-acetic acid, 180, 312 Acetone bodies, 180, 305 in blood, 427 in urine, 180, 305 determination of, 309 significance of, 305 tests for, 308 Acetonemia, 427 Acetonuria, 304 Acholic stools, 101, no Achorion Schonleinii, 156 Achroiocythemia, 418 Achromatophilia, 473 Achroddextrin, 35, 78 Achylia gastrica, 73, 85 Acid, acetic, 59, 68, 72, 235 aceto-acetic, 180, 312 alloxyproteic, 197, 235 amino, 78, 180, 233 bile, 56, 109, 180, 320 butyric, 59, 68, 72, 235 cholalic, in, 180 chondroitin-sulphuric, 180, 197, 239 diacetic, 180, 312 diamino, 78, 87, 233 fatty, 13, 108, 117, 180, 235, 423 formic, 235 glycocholic, no, 180 glycosuric, 321 glycuronic, 180, 302 hippuric, 180, 234, 335 homogentisic, 180, 321 hydrochloric, 46, 60 hydroquinone-acetic, 321 lactic, 59, 68, 69, 238 nucleinic, 222, 240 oxalic, 180, 236 oxaluric, 180, 236 /?-oxybutyric, 180, 314 oxymandelic, 180 oxyproteic, 180, 197, 235 phosphoric, 180, 187 Acid, picric, 259 propionic, 235 rosacic, 241 rosolic, 21, 207 succinic, 258 sulphuric, 195 taurocholic, no, 180, 239 uric, 180, 219, 423 uroferric, 197 uroleucic, 180, 321 Acid-fast organisms, 20 Acidity of gastric juice, 59 of urine, 172, 175 Acidophilic cells, 473 granules, 476, 483 Acidosis, 215, 305 Acid stains, 454 unit, 175 Actinomyces in sputum, 27 Actinomycosis, 27 Acute anterior poliomyelitis, 650 bronchitis, sputum in, 31 gastritis, gastric juice in, 86 hemorrhage, anemia due to, 513 infectious conjunctivitis, 43 diseases, blood in, 527 leukemia, 521 nephritis, 166, 168, 178, 183, 189, 206, 252 rheumatism, blood in, 534 yellow atrophy of the liver, urine in, 210, 233, 332 Addison's disease, blood in, 526 Adenin, 228 Adler's test for blood, 103 Adolescent albuminuria, 250 ^Estivo-autumnal malaria, 544 Age, effect of, on red blood cells, 470 on white blood cells, 499 Agglutination, 575, 583 Agglutinins, 575 Agglutinophore, 575 Agonal leucocytosis, 494 Air in sputum, 4 Albumin, determination of, 258 in exudates, 622 in feces, 116 in milk, 655 in sputum, 5 in transudates, 631 in urine, 247 of blood, 420 quotient, 261, 421 removal of, 261 serum, 247 significance of, in urine, 248 663 664 INDEX. Albumin, tests for, 253 Albuminuria, accidental, 247 adolescent, 250 after baths, 248 alimentary, 248 colliquative, 251 constitutional, 249 cyclic, 249 false, 247 febrile, 250 functional, 248 hematogenous, 252 hypostatic, 250 intermittent, 249 lordotic, 249 mixed, 267 neurotic, 252 of the new-born, 248 of pregnancy, 249 orthostatic, 249 orthotic, 249 physiologic, 247 post-infectious, 251 postural, 249 renal, 252 structural, 252 thermolytic, 264 toxic, 252 traumatic, 251 true, 247 with definite renal lesions, 252 Albumon, 420 Albumose, Bence-Jones, 263 in blood, 422 in feces, 116 in gastric contents, 78 in urine, 263 Albumosuria, 263 alimentary, 267 digestive, 267 enterogenous, 267 febrile, 267 hematogenous, 267 hepatogenous, 267 myelopathic, 264 pyogenic, 267 significance of, 267 tests for, 266 Alcohol-fast organisms, 20 Alimentary albuminuria, 248 albumosuria, 267 chloruria, 181, 362 galactosuria, 293, 300 glycosuria, 271 levulosuria, 293 lipuria, 335 pentosuria, 294 Alkaline phosphates, 188 tide of urine, 175 Alkalinity of blood, 386 Alkalinuria, 189 Alkali therapy, 424 Alkaptonuria, 321 Allantoin, 180, 235 Alloxur bases, 116, 180, 228, 330, 424 Alloxyproteic acid, 180, 197, 235 Almen-Nylander's test for glucose, 279 Almon's tannic acid solution, 656 Aloin test for blood, 102, 614 Altitude, effect of, on red cells, 470 Alveolar epithelial cells in sputum, 11 Amblyochromatic erythroblasts, 468 Amboceptors, 595, 600, 601, 606 Amboceptor paper, 607 Ameba coli in feces, 126 in sputum, 27 in urine, 356 pulmonalis, 27 Amebic dysentery, 127 American hook-worm, 144 Amino-acetic acid, 234 Amino acids in urine, 180, 233 a-aminoisobutvl-acetic acid, 332 Ammonia in blood, 424 in urine, 215 Ammoniemia, 424 Ammonium magnesium phosphate in feces, 114 in sputum, 14 in urine, 337 urate calculi, 359 sediment, 336 Amniotic fluid, 373 Amount of blood, 378 of cerebrospinal fluid, 647 of feces, 97 of gastric juice, 52 of sputum, 2 of urine, 165 Amyloid kidney, globulin in urine of, 261 4 urine in, 166, 170, 240, 252, 261 Amylopsin, 95 Amylosis, 33 Anachlorhydria, 66 Anchylostoma duodenale, 143 Anchylostomiasis, 143 Anemia, 506 aplastic, 512 Biermer's, 508 chlorotic, 506 definition of, 506 Ehrlich's, 512 due to acute hemorrhage, 513 to acute infections, 515 to bad air, 514 to blood poisons, 516 to chronic diseases, 535 to chronic hemorrhage, 514 to inanition, 514 to intestinal parasites, 143, 515 febrile, 515 hemolytic, 515, 516 infantum pseudoleukemica, 511 von Jaksch's 511 leukanemia, 511 lymphatic, 522 of the South, 144 of the tropics, 470 primary pernicious, 508 progressive pernicious, 508 secondary, 512 INDEX 665 Anemia, simple primary, 506 splenic, 510 Anemic degeneration, 474 Anesthesia, changes in urine after, 307 effects of, on blood, 524 Angioneurotic hematuria, 344 Anguillula aceti in urine, 356 intestinalis et stercoralis, 139 Anhydremia, 380 Animal gum in urine, 301 Animal parasites in blood, 539 in ear, 42 in feces, 125 in gastric contents, 58 in sputum, 27 in urine, 356 Ankylostomum duodenale, 143 Anopheles maculipennis, 539 Anterior urethritis, 342 Anthracosis, 4, 33 Antiamboceptors, 575 Antibodies, 573, 599 Anticomplement, 575 Antiformin method for tubercle bacilli, 18 Antigens, 573, 596, 597, 598, 602 Antihemolysins, 575 Antihuman amboceptor, 606 Antimeningococcic serum, 649 Antisheep amboceptor, 600 Antitoxins, 572 Antityphoid vaccination, 578, 590 Anuria, 168 Aplastic anemia, 512 Appearance of blood, 383 of exudates, 632 of feces, 98 of gastric contents, 52 of leucocytes, 479 of milk, 654 of red cells, 462 of semen, 365 of spinal fluid, 647 of sputum, 4 of transudates, 631 of urine, 168 Appetite juice, 46 Arabinose, 296 Arginin, 234 Arneth's classification of neutrophiles, 482 Arnold's test for diacetic acid, 313 Arnold- V olhard m ethod for chlorids, 184 Arterial blood, 384 Arthropoda, 152 Ascaridae in feces, 137 Ascaris alata, 138 caniculae, 138 canis, 138 canis et martis, 151 cati, 138 felis, 138 graecorum, 138 lumbricoides, 137 lumbricus canis, 138 marginata, 138 Ascaris mystax, 138 teres, 138 trichiura, 141 tricuspidata, 138 vermicularis, 138 visceralis et renalis, 151 wemeri, 138 Ascitic fluid, cytology of, 642 Asexual cycle of malarial parasite, 542 Ash-free diet of Taylor, 184 Asiatic cholera, feces in, 99, 121 organism of, 121 Aspergillus flavus, 16 fumigatus, 16 in aural secretion, 42 in sputum, 16 niger, 16 subfuscus, 16 Assimilation limit, 271 Asthma, bronchial, 32 eosinophilia in, 10, 32, 497 sputum in, 32 Atrophic gastritis, 85 Aural secretion, 42 bacteria in, 42 larvae in, 42 molds in, 42 Autotoxic enterogenous cyanosis, 473 Autovaccines, 577 Azoospermatism, 368 Azotorrhea, 108 Babcock's method for fat in milk, 657 Bacillary dysentery, 124 index, 578 Bacilluria, 356 Bacillus anthracis, 26 coli communis, 42, 120, 356, 566 comma, 56, 121 Ducrey's, 637 icteroides, 560 leprae, 23 mallei, 26 mucosus capsulatus, 25, 41 of Boas-Oppler, 57 of Bordet-Gengou, 25 of bubonic plague, 26 of diphtheria, 37 of dysentery, 124 of Finkler-Prior, 56, 121 of Friedlander, 25, 41 of glanders, 26 of Hansen, 23 of influenza, 25 of Kitasato and Yersin, 26 of Klebs-Loffler, 37 of Koch, 18, 124, 354, 564, 635 of Koch-Weeks, 43 of Morax-Axenfeld, 43 of Pfeiffer, 25, 32 of Sanarelli, 560 of Shiga, 124 of soft chancre, 587 of tuberculosis, 18, 124, 354, 564, 635 of Vincent, 39 666 INDEX, Bacillus of whooping-cough, 25 paratyphoid, 123, 566 pertussis, 25 pestis, 26 pseudo-diphtheria, 38 pyocyaneus, 4 smegma, 23, 354, 637 timothy, 24 typhosus, 26, 121, 356, 565, 642 ulceris cancrosi, 637 xerosis, 38 X of Sternberg, 560 Bacteria in blood, 563 in conjunctiva, 43 in ear, 42 in exudates, 635 in feces, 119 in gastric contents, 57 in milk, 660 in mouth, 35 in nasal secretions, 40 in sputum, 15 in urine, 353 Bacterial flora of feces, 120 of vagina, 370 Bacterial vaccines, 576 Bacteriemia, 563 Bacteriology of blood, 563 of cerebrospinal fluid, 649 of exudates, 635 of feces, 119 of milk, 660 of sputum, 15 of urine, 353 Bacteriolysins, 573 Bacteriuria, 353 Balantidium coli, 130 Baldwin's method for oxalic acid, 237 Bang's test for albumose, 266 for sugar, 284 Banti's disease, blood in, 510 Barberio's test for semen, 370 Barber's itch, organism of, 157 Bases, alloxur, 116, 180, 228, 330, 424 hexone, 78, 87, 234 nuclein, 116, 180, 228, 330, 424 purin, 116, 180, 228, 330, 424 xanthin, 116, 180, 228, 330, 424 Basic stains, 454 Basket cells, 486 Basophile leucocytes, 484 Basophiles, 484 Basophilia, 475, 484 Basophilic degeneration of red cells, 475 stippling of reds, 475 Bass and Watkin's macroscopic agglu- tination test, 588 Baths, albuminuria following, 248 effect of, on red cells, 471 Beckmann apparatus, 397 Bed bug, 155 Beef tape-worm, 132 Bence-Jones, body, 263 protein, 263 amount of, 263 Bence-Jones protein, significance of, 263 tests for, 264 Benedict's test for sugar, 277 Benzidin test for blood, 104 Benzoic acid, 234 Bial's test for pentose, 296 Biermer's anemia, 508 Bile acids in blood, 427 in feces, no in gastric contents, 56 in urine, 180, 320 pigments in blood, 427 in exudates, 632 in feces, 101, no in gastric contents, 54, 56 in sputum, 3 in urine, 180, 317 significance of, 317 tests for, 318 Bilharzia hematobia, 29, 562 Bilharziasis, 562 Bilicyanin, in Bilifuscin, in, 317 Bilihumin, in Biliprasin, 100, 317 Bilirubin, 54, 101, in, 317, 334 Biliverdin, 54, 101, in, 317 Biologic test for blood, 590 Bismuth oxid in stools, 100 test for glucose, 279 Biuret test for protein, 265 Black's method for /3-oxybutyric acid 3i5 Black sputum, 4 urind, 170 -water fever, 268 Bladder, inflammation of, 175, 342 tuberculosis of, 355 Blastomycetes in skin, 160 in sputum, 18 Blastomycosis, 18, 160 Blennorrhea, 371 Blood, 374 acetone in, 427 after anesthesia, 524 after splenectomy, 538 after surgical intervention, 523 albumin in, 419 alkalinity of, 386 ammonia in, 424 bacteriology of, 563 biliary constituents in, 427 carbohydrates in, 425 casts, 341 cells in, 462 chemical properties of, 399 chemical tests for, 101, 622 coagulation of, 393 color-index of, 418 color of, 385 constituents of, 385 counting of cells of, 431, 441, 487 crises, 468 cryoscopy of, 396 cultures, 563 dust, 502 INDEX. 667 Blood, electric conductivity of, 398 enumeration of cells of, 431, 437, .44i, 483,487 fat in, 426 ferments of, 430 fixation of smears of, 450 formation of, 377 fresh, 446 gases in, 430 hemoglobin of, 401 in abscess formation, 524 in acute infections, 527 in acute rheumatism, 534 in Addison's disease, 526 in aplastic anemia, 512 in bilharziasis, 562 in carcinoma, 537 in chlorosis, 506 in chronic diseases, 535 in chronic tuberculosis, 535 in diabetes mellitus, 525 in diphtheria, 533 in distomiasis, 562 in filariasis, 554 in gout, 526 in kala-azar, 554 in leprosy, 537 in leukanemia, 511 in leukemia, 517 in malaria, 539 in measles, 532 in myxedema, 527 inorganic constituents of, 428 in pernicious anemia, 508 in pertussis, 534 in pneumonia, 528 in primary anemia, 506 in pseudoleukemia, 522 in relapsing fever, 552 in rickets, 526 in Rocky Mountain spotted fever, 56i in scarlet fever, 531 in secondary anemia, 512 in sleeping sickness, 553 in splenic anemia, 510 in syphilis, 536, 556 in typhoid fever, 530, 565 in varicella, 533 in variola, 532 in whooping-cough, 534 in yellow fever, 560 limitations of examinations of, 627 medico-legal aspects of, 622 tests for, 622 morphology of, 445 needle, 383 nitrogen of, 422 obtaining of, 383 odor of, 386 osmotic pressure of, 396 parasitology of, 539 pathology of, 505 general, 523 special, 505 physiology of, 377 Blood, pigments of, 401 plates, 445, 501, 503 properties of, 384 proteins of, 419 reaction of, 386 red cells of, 462 serum reactions of, 583 smears of, 447 solids of, 401 specific gravity of, 390 spectroscopic tests for, 401, 626 staining of smears of, 453 tests for, 101, 622 total solids of, 401 urea in, 422 uric acid in, 423 value of examinations of, 627 viscosity of, 392 vital staining of, 461 volume of, 378 volume relations of elements of, 381 white cells of, 479 xanthin bases in, 424 Blood casts, 351 cells in exudates, 583 in feces, 113 in gastric contents, 57 in sputum, 3, 5, 11 in urine, 316, 344 plates, 501 appearance of, 501 counting of, 445 function of, 502 number of, 501 size of, 501 staining properties of, 502 poisons, 516 smears, fixation of, 450 preparation of, 447 staining of, 453 staining, 453 Bloody sputum, 3, 5, 11 Boas' method for lactic acid, 71 for estimating gastric motility, 81 test for free hydrochloric acid, 62 test-meal, 50 Boas-Oppler bacillus, 57 Body louse, 154 Boggs' coagulometer, 395 method for protein in milk, 656 Bone-marrow, 377, 503 function of, 377 morphology of, 503 Borax as preservative in milk, 660 Bcric acid as preservative, 660 Bothriocephaloidea in feces, 135 bothriocephalus latissimus, 135 bothriocephalus latus, 135 bothriocephalus sp. Ijima et Kuri- moto, 136 dibothriocephalus cordatus, 136 Bottcher's crystals, 14, 365 Bradshaw's myelopathic albumosuria, 263 Breakfast, test, 49 668 INDEX. Bremer's blood-test in diabetes, 525 Bricklayers' anemia, 146 Brodie-Russell coagulometer, 395 Bronchial asthma, 32 stones, 8 Bronchioliths, 8 Bronchiolitis exudativa, 7 Bronchitis, acute, 31 chronic, 31 eosinophilic, 11 fetid, 31 fibrinous, 32 putrid, 31 Bunge-Trantenroth's method for tuber- cle bacilli, 23 Broncho-pneumonia, sputum of, 30 Burri's method for spirochaete, 638 Busk's intestinal fluke, 150 Butyric acid in gastric contents, 59, 68, 72 Cabot's ring bodies, 475 Cachexial fever, 554 Cadaverin, 117 Caffein, 228 Calcium carbonate calculi, 359 sediment, 338 of urine, 201 oxalate calculi, 359 crystals in sputum, 14 in urine, 236, 330 phosphate calculi, 359 sediment, 335, 337 soaps in feces, 114 sulphate sediment, 334 Calculi, ammonium urate, 359 biliary, 111 bronchial, 8 calcium carbonate, 359 oxalate, 359 classification of, 357 cystin, 360 examination of, 358 formation of, 357 hepatic, in intestinal, 112 nasal, 41 phosphatic, 359 pulmonary, 9 renal, 357 table for examination of, 358 ureteral, 357 urethral, 357 uric acid, 359 urostealith, 360 vesical, 357 xanthin, 360 Calmette's tuberculin reaction, 581 Cammidge's reaction, 299 Cancer (see Carcinoma) Cane sugar, absorption of, 117 digestion of, 59 in urine, 301 Carbohydrates, digestion of, 35, 59, 78 in blood, 425 in exudates, 632 Carbohydrates in feces, no, 117 in milk, 659 in urine, 271 Carbol-fuchsin solution, 19 Carbonates in urine, 200 Carbon dioxid hemoglobin, 404 in blood, 430 monoxid hemoglobin, 404 poisoning, 404 Carcinoma, blood in, 537 cells in exudates, 638 fragments in feces, 112 in gastric contents, 58, 89 in urine, 353 of cervix, 375 of kidney, 344, 353 of pleura, 642 of rectum, 112 of stomach, 87 of uterus, 375 Cardiac albuminuria, 252 edema, 167 pleurisy, 642 Carnin, 228 Casein in feces, 108 appearance of, 108 Leiner's test for, 109 in milk, 657 Casts, 345 blood, 350 chemistry of, 346 colloid, 350 epithelial, 349 fatty, 349 fibrinous, 349 in sputum, 8 granular, 348 hyaline, 345 mixed, 346 origin of, 345 prostatic, 367 pseudo, 351 pus, 350 significance of, 347, 352 size of, 346 staining of, 346 testicular, 367 true, 345 waxy, 348 Catalase, 500 Catarrhal stomatitis, 39 Cellulose in feces, no Centrifugation, 326 Cercomonads in feces, 129 in sputum, 28 in urine, 356 in vaginal secretions, 372 Cercomonas coli hominis, 129 hominis, 129 intestinalis, 129 seu Bodo urinarius, 129 Cerebrospinal fluid, 645 bacteriology of, 648 chemistry of, 647 cytology of, 648 in nasal secretion, 41 INDEX. 669 Cerebrospinal fluid, microscopy of, 648 obtaining of, 646 pressure of, 646 properties of, 647 Cerebrospinal syphilis, 650 Cerumen, 42 Cestodes in feces, 130 in sputum, 28 Chalicosis, 4, 33 Chancre, organism of, 637 Character of blood, 384 of exudates, 632 of feces, 97 of sputum, 4 of urine, 165 Charcoal in feces, 94 Charcot-Leyden crystals in feces, 114 in sputum, 14, 30, 32 Cheesy masses in sputum, 6 Chemical fixation of smears, 451 Chemotaxis, 576 Childhood, red cells in, 478 white cells in, 499 Chinese liver-fluke, 150 Chinovose, 296 Chloremia, 418 Chlorid excretion in urine, 181, 362 retention, 181 Chlorids of the blood, 428 of the urine, 181 amount of, 181 estimation of, 184 variations of, 182 Chloromata, 4 Chlorosis, 506 Chlorotic anemia, 506 Chloruria, 181, 362 Cholecyanin in feces, 100 in urine, 317 Cholelithiasis, in Cholemia, 427 Cholera spirillum in feces, 121 Cholesterin crystals in feces, 114 in sputum, 13 in urine, 335 Choletelin in urine, 317 Choluria*, 317 Chondroitin-sulphuric acid, 180,197,239 Chromogenic bacteria in sputum, 4 Chromogens in urine, 240 Chronic bronchitis, sputum in, 31 diseases, blood in, 535 gastritis, gastric juice in, 86 nephritis, urine in, 166, 170, 178, 183, 189, 252 Chyloid exudates, 633 Chylous exudates, 633 Chyluria, 170, 336 Chymosin, 59, 76 Cimasnomonas hominis, 129 Cimex lectularius, 155 Cladocoelium hepaticum, 148 Clark's method for elastic tissue, 12 Clay-colored stools, 100, no Cleaning glass-ware, 446 Clearing of urine, 284, 287 Coagulation of blood, 393 of exudates, 632 of milk, 655 of urine, 270 time of blood, 395 Coagulometer of Boggs, 395 of Russell-Brodie, 395 of Wright, 394 Coal pigment in sputum, 4, 33 Coarsely granular cells of Schultze, 483 Coating of the tongue, 37 Coccidium hominis, 128 perf orans, 128 Coefficient, creatinin, 230 of Haeser, 177 of Haines, 178 of Long, 178 refraction, 380, 420 Colitis, catarrhal, no malignant, 112 mucous, 105 Collection of feces, 92 of gastric contents, 47 of puncture fluids, 631 of sputum, 2 of urine, 164 Collignon and Pilod's test, 650 Colloid casts, 350 cysts, 643, Colloidal nitrogen, 235 Colon bacillus, 42, 120, 356, 566 Color-index of blood cells, 418 of blood, 385 of exudates, 632 of feces, 98 of gastric contents, 54 of sputum, 3 of urine, 169 Colorimeter, 231 Colostrum, 652 Coma diabeticum, 305 Combined hydrochloric acid, 66 Comma bacillus of Koch, 56, 121 Common flea, 156 liver-fluke, 148 Complement, 574, 601, 602, 607 fixation test, 594, 612 Complementophile, 574 Composition of blood, 399 of milk, 653 of urine, 179 Concretions, biliary, in in sputum, 8 in bronchioliths, 8 in pneumoliths, 9 intestinal, 112 coproliths, 112 enteroliths, 112 nasal, 41 renal, 357 vesical, 357 Conductivity, electric, of blood, 398 of urine, 361 Congo-red test, 61 Conjugated glycuronic acids, 119, 302 Conjunctival secretions, 42 670 INDEX. Conjunctivitis, diphtheritic, 43 gonorrheal, 43 infectious, 43 vernal, 44 Consistency of blood, 392 of feces, 98 of gastric contents, 54 of milk, 653 of spinal fluid, 647 of sputum, 2 of urine, 169 Constipation, 98 Coproliths, 112 Corpora amylacea, 366 Cough, whooping, blood in, 534 organism of, 25 Counting of blood plates, 445 of pus cells, 343 of red cells, 431 of white cells, 441 Crab louse, 154 Creatin in urine, 229, 233 Creatinin coefficient, 230 estimation of, 231 metabolism of, 229 tests for, 231 variations of, 229 Crenation, 464 Crescents in blood, 545 Crises, blood, 468 Cryoscopy of blood, 396 of urine, 361 Crystals in feces, 114 in gastric contents, 59 in semen, 365 in sputum, 13 in urine, 327 Culex mosquito, 561 Cultures, blood, 563 throat, 38 urine, 353 Curds in feces, 108 Curschmann's spirals, 7, 30, 32 Cutaneous reactions, 579, 581 Cyclic albuminuria, 249 Cylindroids, 350 Cylindruria, 352 Cyst, colloid, 643 dermoid, 644 fluids, 643 hydatid, 645 hydrocele, 644 hydronephrotic, 645 myxoid, 643 ovarian, 643 pancreatic, 645 papillary, 644 parovarian, 644 serous, 643 spei matocele, 644 Cy stein, 197 Cysticercus cellulosae, 132 Cystin, 198 calculi, 361 sediment, 331 Cystinuria, 331 Cystitis, 175, 342, 355 Cystospermium hominis, 128 Cystotaenia solium, 132 Cytology in cardiac pleurisy, 642 in malignant pleurisy, 642 in nephritic pleurisy, 642 in pneumococcus pleurisy, 641 in primary tubercular pleurisy, 641 in secondary tubercular pleurisy. 641 in streptococcus pleurisy, 641 in typhoid pleurisy, 642 of ascitic fluid, 642 of cerebrospinal fluid, 648 of exudates, 648 of peritoneum, 642 of pleura, 641 bf normal-fluids, 640 of sputum, 11 technic of, 639 Cytophile, 573 Dahlia stain, 484 Daland's hematocrit, 381 Dare's hemoalkalimeter, 387 hemoglobinometer, 412 method for alkalinity of blood, 387 Darling's method of staining amebae, 127 Day urine, 165 Dechloridization. 183 Deficit of hydrochloric acid, 68 Definitive host, 132 Degenerated forms of red cells, 475 of white cells, 486 Degeneration, anemic, 474 hemoglobinemic, 476 Degree of tolerance, 271 Delayed chloroform poisoning, 307 Demodex folliculorum, 154 Dermacentor Andersoni, 562 venustus, 562 Dermoid cysts, 644 Desmoid bag, 83 Deutero-albumose, 78, 262 Dextrin in urine, 301 Diabetes alternans, 222 insipidus, 167 mellitus, 167, 170 blood in, 525 Bremer's test in, 525 lipemia in, 525 urine in, 167, 170 Williamson's test in, 526 phosphatic, 190 renal, 271 Diabetic coma, 305 Diacanthos polycephalus, 146 Diacetic acid in urine, 312 Diagnosis, functional,, 360 Diagnostic value of agglutination test, 589 of complement fixation test, 610 of precipitin test, 593 Diamines in urine, 117, 331 Diamino acids, 78, 87, 233 Diaminuria, 331 Diarrhea, 98 INDEX 671 Diazo reaction, 322 Dibothriocephalus cordatus, 136 latus, 135 Dibothrium latum, 135 Diet of Folin, 94 of Schmidt and Strasburger, 93 of Taylor, 184 Diffeiential counting, 487 Diffusible alkalinity of blood, 386 Digestion, gastric, 78 intestinal, 95, 114 leucocytosis of, 490 products of, 78, 95 Digestive insufficiency, 107 Dilatation of stomach, 79, 80 Diluting fluids for blood, 435, 442 Dimethylaminoazobenzol test, 6t Dimethylaminobenzaldehyd reaction, 325 Dimorphous muris, 129 Diphtheria, bacillus of, 37 blood in, 533 taking smear in, 38 Diphtheritic conjunctivitis, 43 laryngitis, 38 Diplacanthus nana, 133 Diplococcus intracellularis meningiti- dis, 649 lanceolatus, 24, 30 of Bonome, 649 of Fraenkel, 24, 30 of Jaeger and Heubner, 649 of Neisser, 635 of Weichselbaum, 649 pneumoniae, 24, 30 Diplogonoporus grandis, 136 Diplomellituria, 274 Dipylidium caninum, 133 cucumerinum, 134 Distoma capense, 563 pulmonale, 29 Ringeri, 29 Westermanii, 29 Distomiasis, 563 Distomum buski, 149 caviae, 148 conus, 150 crassum, 149 hematobium, 29, 563 hepaticum, 148 hepatis endemicum seu pernicio- sum, 150 hepatis innocuum, 150 japonicum, 150 lanceolatum, 150 sibiricum, 150 sinense, 150 spathulatum, 150 tenuicolle, 150 Ditrachyceros rudis, 146 Dittrich's plugs, 6 Dochmius anchylostomum, 143 duodenalis, 143 Dohle's inclusion bodies, 532 Donne's test for pus, 343 Donogany's test for hemoglobin, 270 Doremus ureometer, 212 Drigalski and Conradi's media, 122 Drop method for Wassermann test, 605 Dropsical cells, 466, 507 Dropsy of chorionic villi, 374 Drugs, effects of, on blood, 471, 494 reactions of, in urine, 171, 325 Ducrey's bacillus, 637 Durham's hemocytometer, 443 Dum-dum fever, 554 Dwarf tape-worm, 133 Dysentery, amebic, 127 bacillary, 124 Dysmenorrhea, 373 Dyspepsia, 86 Earthy phosphates, 188, 190 Eberth's bacillus, 26, 121, 356, 565, 642 Echinococcus in feces, 134 in sputum, 9 in urine, 356 Ectasis, gastric, 81 Eel, vinegar, 356 Effusions, pleuritic, 641 Egg-yellow reaction, 323 Egyptian chlorosis, 146 Ehrlich's anemia, 512 anemic degeneration, 474 classification of leucocytes, 488 dahlia stain, 484 diazo reaction, 322 dimethylaminobenzaldehyd reac- tion, 324 egg-yellow reaction, 323 hemoglobinemic degeneration, 476 side-chain theory, 571 tri-acid stain, 457 triple stain. 457 Einhorn's method for total acidity, 68 saccharometer, 290 Elastic tissue in feces, 108 in sputum, 12 Electric conductivity of blood, 398 of urine, 361 Ellermann and Erlandsen's method for tubercle bacilli, 19 Empyema, perforating, 33 Endotheliosis, 641 Entamoeba coli, 128 histolytica, 126 tetragena, 126 Enteritis catarrhal, no malignant, 112 membranous, 105 mucous, 105 Enterokinase, 95 Enteroliths, 112 Enthelmintha, 130 Entozoa in feces, 130 Enumeration of blood cells, 431,44 j ,445 of pus cells, 343 Eosin-hematoxylin stain, 456 methylene-blue stain, 455 Eosinophiles, 483 Eosinophilia, 496 Eosinophilic bronchitis, 11 conus, 150 672 INDEX. Epicritic elimination of nitrogen, 205 polyuria, 167 Epidemic cerebrospinal meningitis, 649 Epiguanin, 228 Episarkin, 228 Epistaxis, Gull's renal, 344 Epithelial casts, 349 cells in feces, 113 in gastric contents, 57 in semen, 366 in sputum, 11 in urine, 339 Erepsin, 94 Error in cell counting, 443 Erythrasma, 160 Erythroblasts, 466 Erythrocytes, 462 appearance of, 462 color-index of, 418 counting of, 431 crenation of, 464 degenerations of, 475 formation of, 376 functions of, 478 isotonicity of, 476 nucleation of, 466 number of, 469 pathological types of, 465 recognition of, in stains, 577 resistance of, 476 rouleaux formation of, 463 Erythrocytes, shape of, 464 size of, 464 staining properties of, 473 structure of, 462 variations of, 470 Erythrocytometer, 432 Erythrocytosis, 472 Erythrodextrin, 35, 78 Esbach's method for albumin, 259 Essential albuminuria, 248 pentosuria, 296 renal hematuria, 344 Esterification method of Fischer, 234 Estivo-autumnal malaria, 544, 547 Ethereal sulphates, 196, 199 Euchlorhydria, 65 Euglobulin, 239 European cat-fluke, 150 hook-worm, 143 Eustrongylus gigas, 151 visceralis, 151 Ewald test-meal, 49 Ewald and Siever's method for gastric motility, 81 Exercise, effect of, on red cells, 471 leucocytosis due to, 495 Extraction method for fat, 648 Extraneous material in sputum, 9 Extruded intracellulars, 542 Exudates, 630 bacteriology of, 635 chyloid, 633 chylous, 633 conjunctival, 42 cytology of, 638 Exudates,.formation of, 630 hemorrhagic, 633 obtaining of, 631 peritoneal, 642 pleural, 641 properties of, 632 purulent, 634 putrid, 634 serofibrinous, 632 serous, 632 urethral, 635 False albuminuria, 247 Famine fever, 552 Fasciola hepatica, 148 humana, 148 Fasciolopsis buski, 149 Fasting stomach, contents of, 55 Fat in blood, 426 in exudates, 633 in feces, 109, 117 in milk, 657 in urine, 335 Fatty acids in blood, 426 in exudates, 633 in feces, 109, 117 in sputum, 13 in urine, 236 casts, 349 granules in leucocytes, 486 stools, 100 Favus, 156 Febrile albuminuria, 250 albumosuria, 267 anemia, 515 diseases, blood in, 527 urine, 168 Fecal vomitus, 56 Feces, 92 amount of, 97 bacteriology of, 119 bile acids in, no biliary pigments in, 101, 110 blood in, 101 carbohydrates in, no, 117 chemical examination of, 114 color of, 99 concretions in, 112 consistency of, 98 crystals in, 114 fat m, 109, 117 food remnants in, 106 formed, 98 macroscopic examination of, 97 marking of, 94 microscopic examination of, 112 morphological elements in, 113 mucus in, 104 normal, 92 odor of, 99 parasitology of, 125 protein in, 108 pus in, 106 reaction of, 115 tissue fragments in, 112 total nitrogen of, 116 INDEX. 673 Feces, total solids of, 115 unformed, 98 Fehling's test for glucose, qualitative, 278 quantitative, 283 Female secretions, 370 Fermentation method of Schmidt, 118 test for diphtheria bacillus, 38 for glucose, 280, 291 Fermentative dyspepsia, no Ferments in blood, 430 in feces, 95 in gastric juice, 59, 73 in leucocytes, 430 in sputum, 6 intestinal, 95 in urine, 238 pancreatic, 95 Ferrocyanide test for albumin, 258 Ferrometer of Jolies, 429 Fibers, elastic, 12, 108 muscle, in feces, 108 in gastric contents, 55 Fibrin ferment, 393 in blood, 393 in urine, 270 network, 396 significance of, 396 tests for, 270 Fibrinogen, 393 Fibrinous casts in sputum, 8 in urine, 349 Fibrinuria, 269 Filaria Bancrofti, 555 in blood, 554 in urine, 336, 356 nocturna, 555 sanguinis hominis, 554 Filariasis, 554 Finely granular cells of Schultze, 481 Fischer's esterification method, 234 test-meal, 50 Fish tape-worm, 135 Fittipaldi's method for albumose, 266 Fixation of complement, 594 of smears, by chemicals, 451 by heat, 450 Fixed alkalinity, 175 Flagellata in feces, 129 in sputum, 28 in urine, 356 Flat worms, 130 Fleischl-Miescher hemometer, 409 Flexner's serum, 649 Florence's test for seminal fluid, 368 Fluids, diluent for blood, 435, 441, 577 Fluke-worms, 136, 147 Folin's method for acetone, 312 for acidity of urine, 173 for ammonia, 218 for creatinin, 231 for free mineral acidity, 174 for indican, 246 for sulphates, 199 for urea, 213 for uric acid, 223 Folin's standard diet, 94 Foreign bodies in sputum, 9 Form of stools, 98 Formaldehyd as preservative, 165, 660 Formalin method for ammonia, 219 Formation of blood, 377 of casts, 345 of exudates, 630 Fraenkel's diplococcus, 24, 30 Fragments of tissue in feces, 112 in gastric contents, 58 in sputum, 12 in urine, 353 Free hydrochloric acid, 60 amount of, 64 detection of, 61 determination of, 62 formation of, 46 significance of, 65 variations of, 65 Freezing point of blood, 396 of urine, 361 Fresh blood, 446 Friedlander's bacillus, 25, 41 Frommer's test for acetone, 309 Fucose, 296 Functional albuminuria, 248 diagnosis, 360 hematuria, 344 Functions of gastric ferments, 79 of intestinal ferments, 95 of leucocytes, 499 of red cells, 478 Fusaria mystax, 138 vermicularis, 138 Fusiform bacillus of Vincent, 39 Futcher and Lazear's fixation method, 452 malarial stain, 461 Gabbet's staining method, 20 Gabritschewsky's polychromatophilia, 474 Gaffky's table, 22 Galactosuria, 293, 300 Gall stones in feces, 111 appearance of, 111 composition of, in Gamete, 542 Gametocyte, 542 Gametoschizonts, 550 Gases in blood, 430 in feces, 119 in gastric contents, 78 Gastric carcinoma, 87 contents, 45 acetone in, 79 after test meals, 57 amino acids in, 78 bacteria in, 57 blood in, 54, 56, 78 crystals in, 59 digestion products in, 78 epithelial cells in, 57 food remnants in, 57 from fasting stomach, 55 674 INDEX. Gastric contents from vomitus, 55 gases in, 78 indirect examination of, 83 macroscopic examination of, 52 microscopic examination of, 57 mucus in, 55, 57, 86 obtaining of, 47 protozoa in, 58 pus in, 57 tissue fragments in, 58 crises, 85 juice, 52 acetic acid in, 59, 68, 72 acidity of, 59 amount of, 52 butyric acid in, 59, 68, 72 combined hydrochloric acid in, 66 composition of, 59 deficit of hydrochloric acid in, 68 ferments of, 59, 73 free hydrochloric acid in, 60, 87 hyperacidity of, 60, 66, 84 hypersecretion of, 85 hypoacidity of, 60, 65 in disease, 84 lactic acid in, 69, 88 organic acids in, 68 Pawlow's work on, 46 properties of, 52 secretion of, 46 motility, 80 ulcer, 87 Gastritis acute, 86 atrophic, 85 chronic, 86 Gastrosuccorrhea, 85 Genital organs, secretions of, 365 Genito-urinary tuberculosis, 342, 355 Gerhardt's test for diacetic acid, 313 Ghoreyeb's stain for spirochaete, 638 Giemsa's stain, 460, 638 Gigantoblasts, 468 Gigantocytes, 464 Glanders, bacillus of, 26 Globular decolorization, 474 Globulin-albumin ratio, 261 in blood, 261, 421 in exudates, 622 in milk, 657 in urine, 261 significance of, 261 tests for, 262 Glomerular insufficiency, 360 Glossina palpalis, 553 Glucose in the blood, 425 in the urine, 271 determination of, 283 significance of, 272 tests for, 275 Glutoid capsules, 96 Glycemia, 425 Glycocholic acid, no, 180 Glycocoll, 234 Glycogen in the blood, 426 Glycosuria, 271 Glycosuria, alimentary, 272 after poisoning, 274 after use of drugs, 274 diabetic, 274 e saccharo, 272 ex amylo, 272 neuro-hepatogenous, 273 physiologic, 272 transitory, 272 Glycosuric acid, 321 Glycuronic acid, 302 Glycyl-tryptophan test, 89 Gmelin's reaction for biliary pigments, 318 Goldhorn s stain, 558 Goldschmiedt's test for glycuronic acid, 304 Gonococcus, 635 Gonorrheal conjunctivitis, 43 stomatitis, 39 threads, 342, 587 urethritis, 636 Goodman and Stern's method for albu- min, 259 Gout, blood in, 526 perinuclear granules in, 482 urine in, 222 Gowers' hemoglobinometer, 413 Gram-negative organisms, 636 Gram-positive organisms, 636 Gram's stain, 636 Granular casts, 348 cells in blood, 481, 483, 485 in prostatic fluid, 367 in sputum, 10, 11 degeneration, 475 Granules in blood, acidophile, 483 basophile, 474, 484 of Grawitz, 475 Ehrlich's a, 483 0, 483 r, 484 J, 484 «, 481 eosinophile, 483 fatty, 486 glycogen, 486 Grawitz, 475 hemoconien, 502 in malaria, 541, 543, 545 mast cell, 484 melanin, 541, 543, 545 Neusser's, 482 neutrophile, 481 oxyphilic, 483 perinuclear, 482 sudanophile, 486 in sputum, 11 Grape-sugar in urine, 270 Gravel in urine, 357 Grawitz' basophilia, 475 Green sputum, 3 vomitus, 56 Griess-Ilosvay reagent, 35 Grinders' rot, 4 Gross' method for trypsin, 95 INDEX, 675 Ground itch, 144 Gruber-Widal reaction, 584 Guaiac test for blood, ioi, 102, 613 Guanin, 228 Gull's renal-epistaxis, 344 Gum, animal in urine, 302 Gummatous lymphoma, 523 Gunning's mixture, 207 test for acetone, 309 Gunzburg's package, 83 reagent, 61 test for free hydrochloric acid, 61 Gynecophorus haematobius, 356, 562 Haeser's coefficient, 177 Haines' coefficient, 178 test for glucose, qualitative, 279 quantitative, 286 Haldane and Smith's method for volume of blood, 379 Halitus sanguinis, 386 Hammarsten's test for biliary pig- ments, 319 Hammerschlag's method for specific gravity, 391 for pepsin, 75 Haptines, 572 Haptophore, 572 Hard chancre, organism of, 637 Hart's method for /3-oxybutyric acid, 3J4 Harvest Dug, 154 Hayem's solution, 435 Hay fever, 41 Hay's test for bile acids, 320 Head louse, 154 Heart disease cells, 12 pleurisy of, 632 Heat fixation of smears, 450 Heat test for albumin, 254 Hehner-Maly method for organic acids, 68 Heller's table for examination of calculi, 358 test for albumin, 255 for hemoglobin, 269 Hemameba malarias, 543 vivax, 541 Hemamebiasis, 539 Hematemesis, 3, 56 Hematin, 405 hydrochlorate, 406, 626 Hematoblasts, 501 Hematochyluria, 556 Hematocrit, 381 Hematogenous albuminuria, 251 albumosuria, 266 urobilinuria, 241 Hematoglobulin, 403 Hematoidin in the blood, 406 in sputum, 12, 14 in urine, 317 Hematopoietic organs, 377, 503 Hematoporphyrin in blood, 406 in feces, 101 in stains, 627 Hematoporphyrin in urine, 316 Hematoporphyrinuria, 316 Hematozoon falciparium, 544 Hematuria, 344 angioneurotic, 344 constitutional, 344 essential, 344 extra-renal, 345 functional, 344 idiopathic, 344 renal, 344 Hemin, 406, 626 Hemoalkalimeter of Dare, 387 Hemochromogen, 402 Hemoconien, 502 Hemocytometer of Durham, 443 of Oliver, 444 of Thoma-Zeiss, 432 Hemoglobin in blood, 401 amount of, 407 derivatives of, 405 estimation of, 407 properties of, 401 variations of, 418 in sputum, 3, 5, n, 12 in urine, 268, 316 tests for, 269 quotient, 419 value, 419 Hemoglobinemia, 267 Hemoglobinemic degeneration, 476 Hemoglobinometer of Dare, 412 of Oliver, 414 of Tallqvist, 416 Hemoglobinuria, 267, 316 paroxysmal, 268 significance of, 267 tests for, 269 Hemolysins, 573 Hemolysis, 573 Hemolytic anemia, 515 Hemometer of Fleischl-Miescher, 409 of Sahli, 413 Hemophilia, renal, 344 Hemoptysis, 3 Hemorrhage, anemia due to, 513 occult, 102 Hemorrhagic exudates, 633 nephritis, 344 Hemosiderin, 12, 407 Hepatic insufficiency, 210, 272, 293 Hepatogenous albumosuria, 266 urobilinuria, 241 Herpes tonsurans, 157 Heteroalbumosuria, 263 Heterochylia, 85 Heteroxanthin, 228 Hexamitus duodenalis, 129 Hexone bases, 78, 87, 234 Hippuric acid in urine, 234, 335 Histidin, 234 Histon in urine, 270 Histoplasma capsulatum, 554 Histoplasmosis, 554 Hodgkin's disease, 522 Homogentisic acid, 321 676 INDEX Hopkin's method for uric acid, 223 Hoppe-Seyler's colorimetric pipet, 408 Howell's immature nucleated reds, 467 mature nucleated reds, 467 Huppert-Messinger method for acetone, 310 Hyaline casts, 345 Hydatid cysts, 134, 645 Hydatidiform degeneration, 374 Hydremia, 380 Hydrobilirubin in stools, 100 Hydrocele fluid, 644 Hydrochloric acid in gastric juice, 46 amount of, 64 combined, 66 deficit, 68 estimation of, 62 free, 60 physiologically active, 68 test for, 61 Hydrogen sulphide in gastric contents, 79 in urine, 197 Hydronephrosis, 645 Hydrops folliculorum Graafii, 643 Hydroquinone-acetic acid, 321 Hydruria, 167 Hymenolepis diminuta, 134 flavopunctata, 134 murina, 133 nana, 133 Hypalbuminosis, 420 Hyperacidity of gastric juice, 60, 66 Hyperalbuminosis, 420 Hyperchlorhydria, 66, 84 Hyperglycemia, 425 Hyperinosis, 396 Hypermotility of stomach, 80 Hypersecretion of gastric juice, 85 Hypertonic solutions, 477 Hyphogenous sycosis, 157 Hypinosis, 396 Hypochlorhydria, 65 Hypostatic albuminuria, 250 Hypotonic solutions, 477 Hypoxanthin, 228 Idiopathic enterogenous cyanosis, 404 pentosuria, 296 Ilosvay's reagent, 35 Immature nucleated reds of Howell, 467 Immunity, 571 Inactivation of serum, 573 Inanition, anemia due to, 514 Index, bacillary, 578 color, 418 hemoglobin, 418 opsonic, 576 phagocytic, ^78 volume, 382 India ink method for spirochaete, 638 Indican, 243 Indicanuria, 243 Indigo blue in urine, 243 red in urine, 244 Indirect examination of gastric con- tents, 83 Indol-acetic acid, 246 Indoxyl-potassium sulphate in urine, 242 Infectious diseases, blood in, 527 Influenza, bacillus of, 25 sputum in, 32 Infusoria, in feces, 130 in sputum, 28 in urine, 356 Inorganic constituents of blood, 428 of urine, 181 Inoscopy, 635 Inosite in urine, 301 Insecta, 154 Insufficiency, digestive, 107 glomerular, 360 hepatic, 210, 233, 272 motor, 80 renal, 360 tubular, 360 Intermittent albuminuria, 249 Intestinal concretions, 112 digestion, 95 juices, 95 obstruction, 107 parasites, 125 sand, 112 Iodide of potassium test, 82 Iodoform test for acetone, 309 for lactic acid, 71 lodophilia, 487 Iron in the blood, 429 in urine, 202 Irritation forms of leucocytes, 486 Isomaltose in urine, 302 Isotonicity of red cells, 476 Isotonic solutions, 476 Itch parasite, 153 Jaeger and Heubner's diplococcus, 649 Jaffa's test for creatinin, 231 for indican, 244 von Jaksch's anemia, 511 Japanese liver-fluke, 150 Jaundice, blood in, 428 sputum in, 4 urine in, 170, 317 Jecorin, 425 Jenner's stain, 458 Jigger, 156 Jolies' ferrometer, 429 Jousset's fluid, 635 Juice, gastric, 52 intestinal, 95 Justus' test for syphilitic blood, 536 Kahler's disease, 263 Kala-azar, blood in, 554 parasite of, 554 Karyomorphism of neutrophiles, 482 Kastle and Loevenhart's method for lipase, 238 Kathrein's test for bile pigments, 318 Kelling's test for lactic acid, 70 INDEX. 677 Kendall and Day's method for typhoid bacilli, 123 Kidney, abscess of, 341 acute inflammation of, urine in, 166, 168, 178, 183, 189, 206, 252 amyloid disease of, urine in, 166, 170, 240, 252, 261 cancer of, 344, 353 chronic inflammation of, urine in, 166, 167, 170, 178, 183, 189, 252 echinococcus cysts of, 356 hemorrhagic lesions of, 344 hydronephrotic cysts of, 645 malignant disease of, 344, 353 stones, 357 suppurative lesions of, 341 syphilitic disease of, 252 tubercular, 342 Kjeldahl's method for nitrogen, 206 Klebs-Ldffler bacillus, 37 Knop-Hiifner method for urea, 211 Koch's bacillus, 18, 124, 354, 564, 635 comma bacillus, 56, 121 tuberculin, 18, 579 Koch-Weeks bacillus, 43 Kohlrausch's method for electric con- ductivity, 361, 398 Krabbea grandis, 136 Kreatin (see Creatin), 229 Kreatinin (see Creatinin), 229 Lab, 59, 76 Labor, albuminuria following, 249 Lactic acid in blood, 427 in gastric contents, 69 in carcinoma, 88 significance of, 69 tests for, 70 in urine, 237 Lactose in milk, 649 in urine, 300 Lactosuria, 300 Laiose, 294 Laking of blood, 386 Lamblia intestinalis, 129 Large lymphocytes, 480 mononuclear leucocytes, 480 Larvae in aural secretions, 42 in feces, 139, 141, 145 Laveran's malarial organism, 539 Layers of sputum, 5 Lead, anemia due to, 516 basophilia in poisoning, 475 Lecithin globules in semen, 366 Legal's test for acetone, 308 Leiner's test for casein, 109 Leishman-Donovan bodies, 554 Le Nobel's test for acetone, 308 Lee's method for chymosin, 77 Leprosy, bacillus of, in sputum, 23 blood in, 537 Leptodera intestinalis et stercoralis, i39 Leptothrix buccalis, 36 Leptothrix in sputum, 15 Leptus autumnalis, 154 Leube's test of gastric motility, 81 Leucin in sputum, 14 in urine, 233, 332 Leucocytes, 479 appearance of, 479 basophilic, 484 counting of, 441 degenerated forms of, 486 differential counting of, 487 eosinophilic, 483 ferments of, 430 formation of, 378 functions of, 499 granules in, 474, 483, 484 in blood, 479 in exudates, 638 in feces, 113 in gastric contents, 57 in milk, 661 in sputum, 10 in urine, 341 irritation forms, 486 karymorphism of, 482 large mononuclear, 480 lymphocytes, 479 mast-cell, 484 myelocytes, 485 neutrophilic, 481 number of, 488 oxyphilic, 483 pigmented, 545 polymorphonuclear, 481 small mononuclear, 480 splenocytes, 480 transition forms of, 481 types of, 479 variations in number of, 489, 498 Leucocytic crystals in sputum, 14 Leucocytometer, 433 Leucocytosis, 489 agonal, 494 antemortem, 494 cachectic, 493 eosinophilic, 496 infectious, 492 inflammatory, 492 mast-cell, 498 mixed, 495 of digestion, 490 of pregnancy, 491 of the new-born, 492 polymorphonuclear, 489 post-hemorrhagic, 493 therapeutic, 494 Leucohydrobilirubin in feces, 100 Leucopenia, 489, 498 Leucorrhea, 371 Leucourobilin, 100 Leukanemia, 511 Leukemia, 517 acute, 521 lymphatic, 520 mixed, 521 splenomyelogenous, 517 678 INDEX. Levulose in urine, 293 determination of, 294 recognition of, 294 significance of, 292 tests for, 293 Levulosuria, 294 Lieben's test for acetone, 309 Lientery, 106 Limitations of blood examinations, 617 Limnaea truncatula, 149 Lipacidemia, 427 Lipaciduria, 235 Lipase in gastric juice, 59, 77 in pancreatic juice, 95 in urine, 238 Lipemia, 427, 524 Lipliawsky's test for diacetic acid, 313 Lipuria, 335 Liquor sanguinis, 384 Lithemic diathesis, 221 Liver, abscess of, sputum in, 33 insufficiency of, 210, 233, 272 Lobar pneumonia, blood in, 528 chlorids in urine of, 182 organism of, 24, 30 sputum in, 30 Lochia alba, 369, 373 cruenta, 373 rubra, 373 serosa, 373 Loffler's antiformin method, 18 methylene blue, 20 Lohnstein's saccharometer, 291 Long's coefficient, 178 Lordotic albuminuria, 249 Luetin reaction, 581 Lumbar puncture, 646 Lung, abscess of, sputum in, 33 fluke in sputum, 29 inflammation of, 30,182, 528 stones, 9 Lymphatic leukemia, 520 pseudoleukemia, 52^ Lymphemia, 520 Lymphocytes, 479 Lymphocytosis, 496 Lymphopenia, 496 Lymphosarcoma, 522 Lysins, 234, 574 Lytic action, 574 Macrocytes, 465 Macrocythemia, 466 Macrocytosis, 466, 509 Macrogamete, 542 Magnesium ammonium phosphate in feces, 114 in sputum, 14 in urine, 337 phosphate in urine, 188, 337 salts in urine, 201 soaps in feces, 109 Malaria, blood in, 550 fresh blood in, 540 mosquito theory of, 539 Malaria, parasites of, 541 estivo-autumnal, 544, 547 quartan, 543, 547 tertian, 541, 546 stained smears in, 546 Malarial pigment, 407 Male secretions, 365 Malfatti's method for ammonia, 219 Malignant disease, blood in, 537 gastric juice in, 87 urine in, 344, 353 lymphoma, 522 pleurisy, 642 Maltose in urine, 301 Mammary secretions, 652 Maragliano's endoglobular degenera- tion, 474 Marechalt's test for bile pigments, 318 Martius and Liittke's method for HC1, 66 Marx's fluid, 577 Mast-cell granules, 484 leucocytosis, 498 Masturbators, albuminuria of, 250 Mature nucleated reds of Howell, 467 May-Grunwald stain, 458 McCrudden's method for calcium, 201 Meals, test, 49 Measles, blood in, 532 Medicinal leucocytosis, 494 Medico-legal aspects of blood, 622 of semen, 369 Megaloblasts, 468 Megalocytes, 464 Megalogastria, 80 Megastoma entericum, 129 intestinale, 129 Melanin, 171, 321, 407 Melanogen, 171 Melanuria, 320 Membranous dysmenorrhea, 373 enteritis, 105 ureteritis, 238 Meningeal fluid, examination of, 645 Meningitis, epidemic cerebrospinal, 649 tubercular, 648 Meningococcus of Bonome, 649 of Weichselbaum, 649 Menstruation, 373 Messinger method for acetone, 310 Metalbumin in ovarian cysts, 643 Metamyelocytes, 482 Methemoglobin, 403 Methylene azure stains, 460, 638 blue in urine, 171, 324 stains, 20, 455 test for functional activity, 362 Methylphenylosazon, 295 Methylxanthin, 228 Mett's method for pepsin, 75 Meyer's test for blood, 624 Microblasts, 467 Micrococcus, catarrhalis in sputum, 17 tetragenus in sputum, 16 Microcytes, 465 Microgametes, 542 INDEX. 679 Microgametocytes, 542 Microscopy of blood, 445 of exudates, 638 of feces, 112 of gastric contents, 57 of milk, 660 of semen, 366 of sputum, 9 of urine, 325 Microsporon Audouini, 158 furfur, 160 minutissimum,' 160 Miescher's hemoglobinometer, 409 Milk, 642 appearance of, 653 ash of, 655 bacteriology of, 660 coagulation of, 655 composition of, 653 cow's, 654 curds in stools, 108 fat of, 657 human, 653 lactose of, 659 microscopy of, 660 preservatives in, 659 properties of, 653 protein of, 655 reaction of, 655 specific gravity of, 654 sugar of, 659 total solids, 655 Milk-curdling ferment, 59, 76 Milky zone, 541 Mineral acidity of urine, 174 Mintz's method for free HC1, 63 Mixed infection in tuberculosis, 23 leucocytosis, 495 Molds in aural secretions, 42 in buccal secretions, 35 in sputum, 15 Monocalcium phosphate in urine, 335 Monocercomonas hominis, 129 Monochromatophilia, 473 Mononuclears, basophile, 486 eosinophile, 485 large, 480 neutrophile, 485 small, 480 Moore and Wilson's test for alkalinity, 387 Morax-Axenfeld diplobacillus, 43 Morner-Sjdqvist method for urea, 214 Morner's mucin-like bodies in urine, 239 test for tyrosin, 333 Morning sputum, 1 Moro's tuberculin reaction, 580 Morphology of blood, 445 of blood-forming organs, 502 Mosquito, anopheles, 539 culex, 561 cycle oi malarial parasites, 548 stegomyia, 561 theory of malaria, 539 of yellow fever, 561 Motility of intestine, 97 of stomach, 80 detection of, 81 types of, 80 Motor insufficiency, 80 Moults, 153 Mouth, inflammation of, 39 catarrhal, 39 gonorrheal, 39 mycotic, 40 ulcerative, 39 ulceromembranous, 39 secretions of, 34 Much's method for tubercle bacilli, 21 Mucin in ovarian cysts, 644 in sputum, 5 in urine, 239 Mucinophiles, 484 Mucoid material in urine, 339 sputum, 5 Mucopurulent sputum, 5 Mucor in sputum, 15 Mucous corpuscles 339 threads in urine, 342, 637 Mucus in feces, 104 appearance of, 104 detection of, 104 significance of, 105 in gastric contents, 55, 57, 86 in sputum, 5 in urine, 239 Miillern's blood stain, 455 Murexid test, 328 Muscle fibers in feces, 108 in gastric contents, 55 in sputum, 12 Mycelial casts in sputum, 8 Myelemia, 517 Myelin granules in sputum, 11 Myeloblasts, 480 Myelocytes, 485 basophile, 486 eosinophile, 485 neutrophile, 485 Cornil's, 485 Ehrlich's, 485 Myelocytosis, 495 Myelogenous leukemia, 517 Myeloid leukemia, 517 Myelopathic albumosuria, 263 Myxedema, blood in, 527 Myxococcidium stegomyiae, 560 Myxoid cyst of ovary, 643 Myxoma of the placenta, 374 Nakayama's test for bile pigments, 319 Naphthoresorcin test, 304 Nasal secretion, 40 bacteria in, 40 spinal fluid in, 41 composition of, 40 concretions in, 41 pathology of, 41 Neisser's diplococcus, 635 stain for diphtheria bacillus, 38 Nematodes in feces, 137 in urine, 356 680 INDEX. Nephritic albuminuria, 252 hematuria, 344 oliguria, 168 pleurisy, 642 Nephritis, acute, 166, 168, 178, 183, 189, 206, 252 albuminuria of, 252 chronic diffuse, 178, 183, 189, 252 interstitial, 166, 170, 252 parenchymatous, 166, 168, 252 hemorrhagic, 344 suppurative, 341 syphilitic, 252 unilateral, 360 Nervous dyspepsia, 86 type of albuminuria, 251 of polyuria, 164 Neubauer and Fischer's test for gastric carcinoma, 89 Neuberg and Wohlgemuth's method for pentose, 298 Neuberg's test for glycuronic acid, 304 for levulose, 294 Neusser's granules, 482 Neutral calcium phosphate in urine, 335 dyes, 454 stains, 454 sulphur in urine, 197 Neutrophile cells, 481, 485 granules, 481 Neutrophilic karyolobism, 482 New-born, albuminuria of, 248 leucocytosis of, 492 Night urine, 165 Nikiforoft's method of fixation of smears, 452 Nitric acid test for albumin, 255 Nitrites in saliva, 35 Nitrogen of sputum, 2 of urine, 203 partition of blood, 422 of urine, 204 Nitrogenous balance, 203 bodies in blood, 422 in exudates, 631 in feces, 116 in gastric contents, 89 in milk, 655 in sputum, 2, 5 in transudates, 631 in urine, 203 allantoin, 235 bodies in urine, alloxyproteic acid, 197, 235 amino-acids, 233 ammonia, 215 creatinin, 229 hippuric acid, 234, 335 oxyproteic acid, 235 purin bases, 228 undetermined, 233 urea, 209 uric acid, 219 total, 203 equilibrium, 203 Nitroprussid test for acetone, 308 Nocht's malarial stain, 461 Noguchi's antigen, 598 butyric acid test, 650 luetin reaction, 581 method for spirochete, 559 modification of Wassermann test, 605 Nondiffusible alkalinity of blood, 386 Normal feces, 91 salt solution, 477 Normoblasts, 477 Normocytes, 462 Nose, secretions of, 40 Nubecula, 168 Nubecular threads, 168, 339 Nucleated red cells, 466 Howell's immature, 467 mature, 467 Nuclein bases, 116, 180, 228, 330, . 424 Nucleinic acid, 222, 239 Nucleo-albumin in urine, 239 histon in urine, 270 Number of blood plates, 501 of leucocytes, 488 of red cells, 469 of stools, 97 of tubercle bacilli in sputum, 22 Nummular sputum, 5, 6 Nutrition, effect of, on blood, 470, 5i4 Nycturia, 166 Nylander's test for glucose, 281 Obermayer's test for indican, 244 Obermeier's spirillum, 552 Obtaining blood, 382, 562 exudates, 631 gastric contents, 47 Occult blood in feces, 102 Ochronosis, 321 Odor of blood, 385 of exudates, 632 of feces, 99 of gastric contents, 54 of sputum, 4 of urine, 172 Oidium albicans, 16 Oil test-breakfast, 95 Oligemia, 380 Oligochromemia, 418 Oligocythemia, 471 Oligoplasmia, 481 Oliguria, 167 Oliver's hemocytometer, 444 hemoglobinometer, 414 test for bile acids, 320 Oocyst, 550 Ookinet, 548 Operation, blood after, 523 Opisthorchis felineus, 150 sinensis, 150 Oppier-Boas bacillus, 57 Opsonic index, 576 Opsonins, 576 INDEX. 681 Optical activity of carbohydrates, 288 of conjugated glycuronates, 289 of glycuronic acid, 289 of urine, 179 Oral secretions, 34 Orcein stain, 13 Orchiococcus of Eraud and Hugounenq, 37i Orcin test for pentose, 298 Organic acidity of urine, 174 acids in gastric contents, 68 in urine, 235 Organized sediments in urine, 339 bacteria, 353 blood cells, 344 casts, 345 epithelial cells, 339 mucoid material, 339 parasites, 336 pus cells, 341 spermatozoa, 353 tissue fragments, 353 Origin of casts, 345 of leucocytes, 378 of red cells, 377 Orthostatic albuminuria, 249 Orthotic albuminuria, 249 Osler's disease, 473 Osmotic pressure of blood, 396 of urine, 362 Otomycosis, 42 Ova in feces, 145 in sputum, 29 in urine, 357 of anopheles, 539 of intestinal parasites, 145 Ovarian cysts, 643, colloid, 643 dermoid, 644 myxoid, 643 papillary, 644 serous, 643 Ovoids in malarial blood, 545 Oxalate of calcium in sputum, 14 in urine, 236, 330 calculi, 359 Oxalic acid in urine, 236 amount of, 236 determination of, 237 origin of, 236 variations of, 236 Oxaluria, 237 Oxaluric acid, 236 Oxid of bismuth in feces, 99 /?-oxybutyric acid in urine, 314 determination of, 315 significance of, 315 tests for, 315 Oxyhemoglobin, 403 p-oxyphenyl- a-amino-propionic acid, 333 Oxyphilic cells, 473 granules, 476, 483 Oxyproteic acid, 180, 235 Oxyuris vermicularis, 138 Ozena, 41 Palpation, albuminuria due to, 251 Paludism, 539 Pancreatic cysts, 645 disease, feces in, 107 fluid, 645 juice, 95 composition of, 95 ferments of, 95 insufficiency of, 107 Panoptic staining, 455, 458 Papillary cysts of the ovary, 644 Pappenheim's amblyochromatic ery- throblasts, 468 heteroplastic promyelocytes, 485 method for tubercle bacillus, 21 stain for blood smears, 458, 460 trachyochromatic erythroblasts, 467 Paracresol, 321 Paragonimus westermanii, 29 Paramecium coli, 130 Paramucin, 644 Parasites, anemia due to, 514 eosinophilia due to, 498 in blood, 539 in feces, 125 in sputum, 27 intestinal, 125 in urine, 356 malarial, 539 of the skin, 152 Parasitology of the blood, 539 of the feces, 125 of the skin, 152 Paratyphoid bacillus, 123, 566 Paraxanthin, 228 Parenchymatous nephritis acute, 166, 168, 178, 183, 189, 206, 252 chronic, 166, 167, 252 Parhemoglobin, 401 Parovarian cysts, 644 Paroxysmal hemoglobinuria, 268 polyuria, 167 Pathogenic bacteria in blood, 563 in exudates, 635 in feces, 119 in gastric contents, 57 in milk, 660 in sputum, 18 in urine, 353 Pea-soup stools, 122 Pedicuius capitis, 154 pubis, 154 vestimenti, 154 Penicillium glaucum, 16 Pentose in urine, 296 determination of, 297 significance of, 296 tests for, 297 Pentosuria, alimentary, 295 essential, 295 idiopathic, 295 intrinsic, 295 Penzoldt and Faber's test, 82 Pepsin in gastric juice, 59, 73 activity of, 73 682 INDEX. Pepsin in gastric juice, detection of, 73 determination of, 74 significance of, 73 in urine, 238 Pepsinogen, 73 Peptic glands, 45 Peptone in the blood, 420 in gastric contents, 78 in urine, 268 Peptonuria, 268 Pref orating empyema, sputum in, 33 Pericardial fluid, 630 Perinuclear granules of Neusser, 482 Periodic albuminuria, 249 polyuria, 167 Peritoneal exudates, 596 composition of, 596 cytology of, 596 Permeability of red cells, 396, 476 renal, 360 Pernicious anemia, 508 Pertussis, blood in, 534 organism of, 25 Pessary forms of red cells, 462 Pettenkofer's test, 320 Pfeiffer's bacillus, 25, 32 Phagocytic cells, 4, 576 index, 578 Phagocytosis, 576 Pharyngomycosis leptothrica, 37 Phenol in feces, 92 in urine, 321 Phenolphthalin test, 624 Phenolsulphuric acid, 321 Phenolsulphonephthalein test, 363 Phenylglucosazon, 281 Phenylhydrazine test for glucose, 281 Phloridzin test, 362 Phloroglucin test for pentose, 297 vanillin test for HC1, 61 Phosphates, calcium, 187, 201 in blood, 429 in sputum, 14 in urine, 187 magnesium, 187, 201, 337 magnesium-ammonium, 14, 114, 337 triple, 14, 114, 337 Phosphatic calculi, 359 diabetes, 190 sediments in urine, 335, 337 Phosphaturia, 189, 337 Phosphorus containing proteins, 239 poisoning, blood in, 516 urine in, 210 Phthirius inguinalis, 154 Phthisis, blood in, 535 hemoptysis in, 3 melanotica, n sputum in, 29 stone-cutters', 4 Physiological albuminuria, 247 glycosuria, 271 salt solution, 477 variations in blood cells, 469, 490 Physiologically active HC1, 68 Physis intestinalis, 146 Pigment, bile, in blood, 427 in feces, 101, no in gastric contents, 54, 56 in sputum, 3 in urine, 317 blood, in feces, 101 . in gastric contents, 54, 56, 78 in sputum, 3, 5, 11, 12, 14 in urine, 317 coal, in sputum, 4, 33 in leucocytes, 4, 545 in red cells, 475, 541 of blood, 401 of urine, 241, 317 Pin worm, 138 Pineapple test, 72 Piroplasma hominis, 561 Piroplasmosis, 561 von Pirquet's tuberculin reaction, 580 Placenta cells, 374 Plague bacillus, 26 Plasma, 384 Plasmodium malariae, 539 precox, 544 vivax, 541 Platelets, blood, 500 Platodes in feces, 130 Plehn's karyochromatophilic granules, 547 Plethora, cellular, 380 serous, 380 true, 380 • ver a, 380 Pleuritic effusions, 641 cytology of, 641 withdrawal of, 631 Plugs, Dittrich's, 6 prostatic, 367 Pneumococcus of Fraenkel, 24, 30, 567 pleurisy, 641 Pneumonoconioses, 33 Pneumoliths, 9 Pneumonia, blood in, 528 chlorids in urine in, 182 organism of, 24, 30 sputum in, 30 urine in, 182 Pneumonomycosis aspergillina, 16 Poikilocytes, 466 Poikilocytosis, 466 Poisons, blood, 516 Polariscope, 287 Polariscopic method for glucose, 288 Poliomyelitis, acute anterior, 650 Polychromasia, 473 Polychromatophilia of Gabritschewsky, 474 of Maragliano, 474 Polychrome dyes, 458 Polychromemia, 472 Polyglobulia, 472 Polymorphonuclear basophiles, 484 eosinophiles, 483 neutrophiles, 481 Polymorphonuclear neutrophiliosis, 489 INDEX. 683 Polynucleosis, 489, 641 Polyplasmia, 380 Polyuria, 166 epicritic, 167 paroxysmal, 167 periadic, 167 Poor, anemia of the, 514 Pork tape-worm, 132 Posterior urethritis, 342 Post hemorrhagic anemia, 513 infectious albuminuria, 251 Postural albuminuria, 249 Potassium acid urate sediment, 329 ferrocyanide test for albumin, 258 iodide test, 82 of blood, 400 of urine, 200 sulphocyanate in saliva, 34 Precipitinophore, 576 Precipitins, 576 Precipitin test for blood, 590 Pregnancy, albuminuria of, 249 ammonia in urine of, 216 anemia of, 514 blood in, 491 leucocytosis of, 491 sero-diagnosis of, 613 Preparation of blood smears, 447 Preservation of urine, 164 Primary anemia, 506 pernicious anemia, 508 proteoses, 263 tubercular pleurisy, 641 Products of gastric digestion, 7 8 of intestinal digestion, 95 Progressive pernicious anemia, 508 Promyelocytes, 485 Propepsin, 46, 73 Prostatic casts, 367 fluid, 365 plugs, 367 secretion, 365 Prostatitis, 342 Prostatorrhea, 367 Protalbumose, 262 Protamine in urine, 270 Protein in blood, 419 in exudates, 632 in feces, 108 in gastric contents, 78, 89 in milk, 656 in sputum, 5 in urine, 247 quotient of serum, 421 Proteoses in urine, 263 Prothrombase, 393 Protoryxomyces coprinarius, 129 Protozoa in blood, 539 in feces, 125 in gastric contents, 58 in sputum, 27 in urine, 356 Prowazek-Green trachoma bodies, 43 Prune-juice sputum, 3, 30 Pseudo casts, 351 diphtheria bacillus, 38 Pseudo elastic tissue, 12 gall stones, 111 globulin, 261 hemoglobin, 403 leukemia, 522 infantum, 511 mucin, 597 nucleation, 474 parasites, 146 rhabditis stercoralis, 139 Ptomaines in feces, 117 in urine, 331 Ptyalin, 34, 35, 78 Ptyalism, 36 Puberty, albuminuria of, 250 Puerperal infection, 373 Pul ex irritans, 156 penetrans, 156 Pulmonary actinomycosis, 27 gangrene, 32 hemorrhage, 3 tuberculosis, 29 Punctate basophilia of Grawitz, 475 Purdy's method for albumin, 261 for chlorids, 187 for glucose, 286 for phosphates, 194 for sulphates, 200 Purin bases in urine, 180, 228, 330 Purpura hemorrhagica, blood in, 509 Purpurin, 241 Purulent exudates, 634 sputum, 5 urine, 341 Pus casts, 350 cells in feces, 106 in gastric contents, 57 in sputum, 10 in urine, 341 enumeration of, 343 significance of, 342 tests for, 343 Putrescin, 117 Putrid bronchitis, 31 exudates, 634 Pycnometer, 176, 390 Pycnotic nucleus, 467 Pyelitis, urine in, 341 productiva, 238 Pyelonephritis, 341 Pyloric glands, 45 stenosis, 81, 88 Pyogenic albumosuria, 266 Pyonephrosis, 341 Pyrocatechin, 321 Pyuria, 341 Quantity of blood, 377 of gastric juice, 52 of urine, 165 Quartan malarial parasite, 543 asexual cycle of, 543 sexual cycle of, 544 Quotient, albumin, 261, 421 protein, 261, 421 volume, 382 684 INDEX, Ratio of N to Cl output, 182 to P2O6 output, 189 to SO3 output, 195 Ray fungus, 27 Reaction of blood, 386 of feces, 115 of gastric contents, 52 of milk, 655 of spinal fluid, 647 of sputum, 3 of urine, 172 Reactivity of blood, 387 Receptors, 572 Rectum, blood in cancer of, 101 Red cells (see Erythrocytes), 462 in exudates, 633 in feces, 102, 113 in gastric contents, 54, 56, 78 in sputum, 3, 5, 11 in suspected stains, 577 in urine, 344 Red, indigo in urine, 244 sputum, 3 Refraction coefficient of serum, 380, 420 Reichmann's disease, 85 Relapsing fever, 552 Relative value of phosphoric acid, 190 Removal of albumin, 261 of glucose, 279, 291 of turbidity, 253 Renal abscess, 341 aneurism, 344 calculus, 357 concretions, 357 diabetes, mellitus, 27;, 363 diagnosis, 360 epistaxis, 344 epithelial cells in urine, 339 hematuria, 344 hemophilia, 344 insufficiency, 360 Rennin, 59, 76 Resistance of red cells, 476 Resorcin test for free HC1, 62 Rhabdonema intestinalis, 13-9 strongyloides, 139 Rhamnose in urine, 296 Rheumatism, blood in, 534 Rhizopoda in feces, 126 Rice-water stools, 99, 121 vomitus, 56 Rickets, blood in, 526 Ricketts' organism of spotted fever, 562 Riegel's method for chymosin, 77 test meal, 50 Ring bodies in red cells, 475 of Cabot, 475 worm of the beard, 157 of the body, 157 of the scalp, 158 Rivalta's test for exudates, 632 Roberts' method for glucose, 292 Rocky Mountain spotted fever, blood in, 561 Rocky Mountain spotted fever, blood in organism of, 562 Romanowsky's stain, 458 Ronchese's method for ammonia, 219 Rosacic acid, 241 Rosenbach's method for bile pigments, 319 for skatoxyl, 245 Rosenberger's method for tubercle bacilli, 570 Rosenow's capsule stain, 568 Rose's method for creatin, 233 Rosin's test for bile pigments, 318 Rot, grinders', 4 Rouleaux formation, 463 Round worms in feces, 137 Rowntree and Geraghty's test, 363 Rubner's test for lactose, 301 Rudisch and Kleeberg's method for uric acid, 225 Rudolf's method for coagulation time, 394 Ruhemann's uricometer, 227 Russell and Brodie's coagulometer, 395 Russo's test for typhoid, 324 Rusty sputum, 3 Saccharometer of Einhorn, 291 of Lohnstein, 291 Saccharomyces cerevisiae, 15, 277 Saccharose in urine, 301 Sagitula hominis, 146 Sago-like granules in sputum, 1, 12 Sahli's desmoid reaction, 83 hemometer, 413 test-meal, 51 Salicylic acid as preservative, 660 test for gastric motility, 81 Saliva, 34 amount of, 34 bacteria in, 35 cells in, 35 chemistry of, 34 ferments in, 34 microscopic examination of, 35 nitrites in, 35 obtaining of, 36 pathologic changes in, 36 potassium sulphocyanate in, 34 ptyalin in, 34, 35 Salivary corpuscles, 35 Salivation, 36 Salkowski method for alkalinity of blood, 389 Salkowski-Ludwig method for uric acid, 224 Salol test of Ewald and Sievers, 81 Salomon's test for gastric carcinoma, 89 Salzer's test-meal, 51 Sand flea, 156 intestinal, 112 in urine, 357 renal, 357 Sanguinous exudates, 633 sputum, 5 INDEX. 685 Saprophytes in feces, 119 in sputum, 15 in urine, 353 Sarcinas in feces, 120 in gastric contents, 58 in sputum, 17 in urine, 353 ventriculi, 58 Sarcoma, blood in, 425, 522 Sarcoptes scabiei, 153 Saturation deficit, 68 Scarlet fever, blood in, 531 Schaer's test for blood, 102, 574 Scherer's method for albumin, 259 test for leucin, 332 Schistocytes, 465 Schistosomum hematobium in blood, 562 in urine, 356 Schizogone, 541 Schizont, 550 Schldsing's method for ammonia, 216 Schmaltz' specific gravity tubes, 391 Schmidt and Strasburger's standard diet, 93 Schmidt's fermentation method for feces, 118 Schiigner's granules, 547 Schultze's granular cells, 481, 483 Sclerostoma duodenale, 143 Scybala, 98 Seat worms, 138 Sebelien' s method for protein in milk, 6 5 6 Secondary anemia, 512 proteoses, 262, 265 tubercular pleurisy, 641 Secretin, 95 Secretion of gastric juice, 46 of genital organs, 364 of mammary glands, 652 of urine, 164 Sedimentation, 325 method of Spengler, 21 Sediments in urine, 325 bacteria, 353 bilirubin, 334 blood cells, 344 calcium carbonate, 338 oxalate, 330 phosphate, 335 sulphate, 334 casts, 345 cholesterin, 335 cystin, 337 epithelial cells, 339 fat, 335 hematoidin, 334 hippuric acid, 335 indigo, 243 leucin, 332 magnesium ammonium phosphate, 337 phosphate, 337 mucoid material, 339 mucous threads, 342 nubecular threads, 168, 339 Sediments, organized, 339 parasites, 356 phosphates, 335, 337 preservation of, 164 pus cells, 341 spermatozoa, 353 tissue fragments, 353 tyrosin, 333 unorganized, 328 urates, 328 uric acid, 328 xanthin, 330 Sedimentum lateritium, 328 Seliwanoff's test for levulose, 294 Semen, 366 chemistry of, 366 medico-legal aspects of, 369 microscopic examination of, 365 pathology of, 368 recognition of stains of, 369 spermatic crystals in, 365 spermatozoa in, 366 Seminal stains, 368 medico-legal aspects of, 368 Septic pleurisy, 641 Sero diagnosis, 583 of pregnancy,613 of syphilis, 594 of typhoid, 584 Serous cysts of the ovary, 643 exudates, 632 plethora, 380 pleurisy, 633 sputum, 5 Serum albumin in blood, 420 in urine, 247 diagnosis of syphilis, 559 of typhoid, 565 globulin, 261 determination of, 262 significance of, 261 test for, 262 variations of, 261 pathology, 571 reactions, 564 refraction coefficient of, 380, 420 special properties of, 569 Sex, variations of blood cells due to, 469 Sexual cycle of malarial parasite, 548 secretions, 365 Shadows of leucocytes, 486 red cells, 462 Shaffer's method for acetone bodies,316 Shiga's bacillus, 124 Showers of casts, 348 Side-chain theory of Ehrlich, 572 Siderosis, 33 Signet rings in malarial blood, 541, 544 Significance of acetonuria, 306 of albuminuria, 248 of albumosuria, 267 of Bence Jones proteinuria, 264 of hematuria, 317 of cylindruria, 347, 352 of free HC1 in gastric contents, 65 686 INDEX. Significance of globulinuria, 261 of glycosuria, 271 of lactic acid in stomach, 69 of leucocytosis, 489 of levulosuria, 293 of mucous threads in urine, 342 of mucus in feces, 105 of nitrogen-partit: on of urine, 202 of ^-oxybutyric acid in urine, 314 of pentosuria, 296 of pepsin in gastric juice, 73 of pyuria, 342 Sjoqvist's method for urea, 214 Skatoxyl-potassium sulphate in urine, 245 Skin, blood in diseases of, 497 parasites of, 152 Sleeping sickness, blood in, 553 organism of, 553 Small-pox, blood in, 532 Smears, preparation of, 447 of blood, 447 of exudates, 636 of feces, 113 of pus, 636 of sputum, 10 of syphilitic material, 638 Smegma bacillus in buccal secretions, 36 in exudates, 637 in sputum, 23 in urine, 354 preputii, 637 Smith's test for bile pigments, 318 Soaps in feces, 114 Sodium acid urate, 329 carbonate as preservative of milk, 649 chlorid retention, 181 in nephritis, 183, 362 in pneumonia, 182 in blood, 400 in urine, 200 Soft chancre, organism of, 637 Soluble starch, 35, 78 Solvents for blood stains, 577 Specific gravity of blood, 390 of cerebrospinal fluid, 647 of exudates, 632 of milk, 654 of serum, 391 of transudates, 631 of urine, 175 Specificity of agglutination test, 589 of complement-fixation test, 610 of precipitin test, 593 Spectrophotometer of Hiifner, 408 Spectroscopic examination, 576, 627 tests for blood, 401, 627 Spengler's sedimentation method, 19 Spermatic crystals, 365 Spermatocele, 644 Spermatorrhea, 353, 367 Spermatozoa, 353, 365 Spermin crystals, 14, 365 Spiegler's test for albumin, 258 Spirals of Curschmann, 7, 30, 32 Spirillum of Asiatic cholera, 56, 121 of Obermeier, 552 of relapsing fever, 552 of Vincent, 39 Spirocheta, buccalis, 35 pallida, 556 characteristics of, ^7, ccg cultivation of, 559 in blood, 556 in exudates, 637 in tissues, 637 staining of, 558, 638 refringens, 558 Spit cups, 2 Spleen, diseases of, blood in, 510 removal of, 538 Splenectomy, blood after, 538 Splenic anemia, 510 Splenocytes, 480 Splenomegaly, blood in, 510 tropical, 554 Splenomyelogenous leukemia, 517 Spore cyst, 550 Sporoblast, 550 Sporogone, 539 Sporogony of malarial parasite, 548 Sporothrix Schenckii, 162 Sporotrichosis, 162 Sporozoa in feces, 128 Sporozoits, 550 Spotted fever, organism of, 562 Sputum, 1 air in, 4 albumin in, 5 amount of, 2 bacteria in, 15 biliary pigments in, 3 blood in, 3, 5, 11 character of, 4 cheesy particles in, 6 chemistry of, 5 chromogenic bacteria in, 4 coal pigment in, 4, 33 coctum, 31 collection of, 2 color of, 3 concretions in, 8 consistency of, 2 cotton fibers in, 4 crudum, 31 crystals in, 13 Curschmann's spirals, in, 7, 30, 32 cytology of, 11 deportment on standing of, 5 Dittrich's plugs in, 7 echinococcus membranes in, 9 elastic tissue in, 12 epithelial cells in, n extraneous matter in, 9 fatty acids in, 13 ferments in, 6 ferric oxid in, 4 fibrinous casts in, 8 flour in, 4 foreign bodies in, 9 INDEX, 687 Sputum, fundum petens, 5 heart disease, cells in, 12 hemoglobin derivatives in, 12, 14 in abscess of the lung, 33 in actinomycosis, 27 in acute bronchitis, 31 in bronchial asthma, 32 in broncho-pneumonia, 30 in chronic bronchitis, 31 in croupous pneumonia, 30 in fibrinous bronchitis, 32 in gangrene of the lung, 32 in influenza, 32 in jaundice, 4 in perforating empyema, 33 in pneumonoconioses, 33 in pulmonary tuberculosis, 29 in putrid bronchitis, 31 leucocytes in, 10 macroscopic examination of, 6 microscopic examination of, 9 morning, 1 mucin in, 5 mucoid, 5 mucopurulent, 5 myelin granules in, 11 nitrogen of, 2 nummular, 5, 6 odor of, 4 origin of, 1 parasites in, 27 prune-juice, 3, 30 purulent, 5 pus cells in, 10 reaction of, 3 red blood cells in, 11 sanguinous, 5 serous, 5 spit-cups for, 2 stone dust in, 4 tenacity of, 3 types of, 5 Staining characteristics of tubercle bacillus, 19 methods, principles of, 453 of bacteria, 625 of blood smears, 455 of casts, 346 of elastic tissue, 13 properties of cells, 473 vital, 461 Stains, blood, 577 Bunge and Trantenroth's, 23 Burri's, 628 dahlia, 484 Ehrlich's tri-acid, 457 triple, 457 eosin-hematoxylin, 456 methylene blue, 455 Gabbet's, 20 Ghoreyeb's, 638 Giemsa's, 460, 638 Goldhorn's, 558 Gram's, 636 india-ink, 638 iodine, 484 Stains, Jenner's, 458 Leiner's, 109 Loffler's methylene blue, 20 May-Grunwald's, 458 Miillern's, 455 Neisser's, 38 Nocht's, 461 orcein, 13 osmic acid, 109 Pappenheim's blood, 458, 460 for tubercle bacillus, 21 polychrome, 458 - Romanowsky's, 458 safranin, 626 scharlach R, 109 seminal, 368 sudan III, 109 thionin, 105, 461 Tunnicliff's, 638 Turk's iodine, 484 Unna-Tanzer's, 13 Weigert's fibrin, 8 Wright's, 459 Zenoni's, 5 Ziehl-Neelsen, 19 Staphylococcus pyogenes in sputum, 26 Starch in feces, 118 detection of, 118 digestion of, 35 estimation of, 118 Steatorrhea, 108, 117 Stegomyia fasciata, 561 Stercobilin, 100, 241 Sterility, 367 Stippled cells, 475, 546 Stomach, absorptive power of, 82 carcinoma of, 87 contents, 45 dilatation of, 79, 80 diseases of, 84 ectasia of, 80 fasting, 55 function of, 79 histology of, 45 • inflammation of, 86 motility of, 80 tube, 47 ulcer of, 87 washing, 48 Stomatitis, catarrhal, 39 gonorrheal, 39 mycotic, 40 ulcerative, 39 ulceromembranous, 39 Stone cutters' phthisis, 4 Stones, bronchial, 8 gall, in in bladder, 357 in kidney, 357 in lung, 9 intestinal, 112 in ureter, 357 in urine, 357 nasal, 41 renal, 357 ureteral, 357 688 INDEX Stone's vesical, 357 Stools (see Feces), 92 acholic, 101 clay-colored, 100 curds in, 108 frequency of, 97 pea-soup, 122 rice-water, 99, 121 Strasburger's method for bacteria in feces, 119 Strauss' test for lactic acid, 71 Streptococcic sore throat, 39 Streptococcus pleurisy, 641 pyogenes in sputum, 26 in blood, 569 viridans in blood, 569 Streptothricosis, 27 Streptothrix eppingeri, 15 Striatula, 146 Strongyloides intestinalis, 139 Strongylus duodenalis, 143 gigas, 151 quadridentatus, 143 renalis, 151 Structural albuminuria, 252 Sudanophiles, 486 Sulphates of urine, 196 easily split, 195 ethereal, 196 preformed, 195 total, 196 unoxidized, 196 Sulph-hemoglobin, 404 Sulphhemoglobinemia, 404 Sulphocyanates in saliva, 34 Sulpho-salicylic acid test for albumin, 258 Sulphur compounds in urine, 195 amount of, 195 determination of, 198 neutral, 197 types of, 195 variations of, 196 Sulphur test for bile in urine, 320 Surgical interference, blood after, 523 Syphilis, albuminuria of, 252 blood in, 536 organism of, 556 hemoglobin test of Justus in, 536 serum test of Wassermann in, 596 Table for examination of calculi, 358 Gaffky's, 22 Taeniidae in feces, 132 taenia aegyptica, 133 canin a, 133 cucumerina, 133 cucurbitina, 132 dentata, 132 diminuta, 134 echinococcus, 134 elliptica, 133 flavopunctata, 134 inermis, 132 lata, 135 leptocephala, 134 Taeniidae taenia mediocanellata, 132 minima, 134 moniliformis, 133 nana, 133 saginata, 132 solium, 132 varerina, 134 Tallqvist's hemoglobinometer, 416 Tape-worms in feces, 130 Tartar of the teeth, 37 Taurocholic acid, no, 180, 239 Taylor's ash-free diet, 184 Teichmann's crystals, 625 test for blood, 625 Tenacity of sputum, 3 Tertian malarial organism, 541 asexual cycle of, 541 sexual cycle of, 542, 548 Test meal of Boas, 50 of Ewald, 49 of Fischer, 50 of Riegel, 50 of Sahli, 51 of Salzer, 51 oil, 95 Testicular casts, 367 Thecosoma hematobium, 562 Theobromin, 228 Theophyllin, 228 Therapeutic measures, effect of, on blood, 471, 494 Thermolabile substances, 574 Thermostable substances, 574 Thionin stain, 106, 457 Thiosulphuric acid in urine, 180 Third corpuscles of blood, 500 Thomas and Weber's method for pepsin, 76 Thoma-Zeiss hemocytometer, 432 Thorn-apple crystals, 336 Threads, mucus in urine, 168, 339, 342 Thread worm, 138 Throat cultures, 38 Thrombase, 393 Thrush, 40 Tick fever, 561 Tide, alkaline, of urine, 175 Timothy bacillus in Sputum, 24 Tinea barbas, 157 circinata, 157 favosa, 156 sycosis, 157 tonsurans, 158 versicolor, 160 Tissue, elastic in sputum, 12 fragments in feces, 112 in gastric contents, 58 in sputum, 9 in urine, 353 Toisson's fluid, 436 Tollen's orcin test for pentose, 297 Naphthoresorcin test for gly- curonic acid, 304 phloroglucin test, 297 test for pentose, 297 Tongue, coating of, 37 INDEX 689 Tonsillitis, leucocytosis in, 492 Topfer's method for combined HC1, 67 for free HC1, 63 test for free HC1, 6t Total acidity of gastric juice, 59 components of, 59 determination of, 59 limits of, 60 acidity of urine, 173 nitrogen of blood, 422 of feces, 116 of gastric juice, 78 in carcinoma, 89 of urine, 203 amount of, 203 determination of, 207 variations of, 204 solids of blood, 401 of feces, 115 of milk, 655 of urine, 177, 180 volume of blood, 378 Towel test for hemoglobin, 408 Toxemia, hepatic, 210, 233, 272 intestinal, 120 renal, 360 Toxogenic protein decomposition, 205 Toxoids, 572 - Toxones, 573 Toxophore, 572 Trachoma bodies, 43 Trachyochromatic erythroblasts, 467 Transitional leucocytes, 481 Transudates, 630 coagulation of, 632 obtaining of, 631 properties of, 631 Traumatic albuminuria, 251 Treatment, effect of on Wassermann's test, 611 Trematodes, 147 in feces, 136 in sputum, 29 Treponema pallidum (see Spirocheta pallida), 556, 559 Triacid stain of Ehrlich, 457 of Pappenheim, 458 Trichina spiralis, 141 Trichinella spiralis, 141 Trichinosis, 141, 498 Trichiuris trichiura, 141 Trichocephalus, dispar, 141 hominis, 141 mastigodes, 141 trichiuris, 141 Trichomonas hominis, 129 intestinalis, 129 vaginalis, 372 in urine, 357 Trichophyton megalosporon endothrix, i57 microsporon, 158 Trichotrachelidse, 141 Triple phosphates as calculi, 359 in sputum, 14 in urine, 337 Tripperfaden in urine, 342, 637 Trommer's test for glucose, 276 Tropeolin test for HC1, 62 Tropical splenomegaly, 554 Tropics, anemia of, 470 Trousseau's test for bile pigments, 3l8 True albuminuria, 247 Trypanosoma Brucei, 554 equiperdum, 554 Evansi, 554 Gambiense, 553 in the blood, 553 in spinal fluid, 554, 649 Trypanosomiasis, 553 Trypsin in feces, 95 in pancreatic cysts, 645 in urine, 238 Tryptophan test, 90 Tsetse flies, 553 Tsuchiya's method for albumin, 260 Tube casts in urine, 345 Tubercle bacilli in the blood, 570 Rosenberger's method, 570 in exudates, 635 inoscopy, 635 in feces, 124 in sputum, 18 morphology of, 20 number of, 22 in sputum, staining of, 19 value of examinations for, 21 in urine, 354 Tubercular meningitis, 648 pleurisy, 641 Tuberculin, 18 reactions, 579 Tuberculosis, blood in, 535 of bladder, 355 of intestine, 124 of kidneys, 342 of lymph glands, 523 of meninges, 648 of peritoneum, 642 of pleura, 641 pulmonary, blood in, 535 sputum in, 29 Tuberculous cystitis, 355 Tubular insufficiency, 360 Tumor shreds in feces, 112 in gastric contents, 58 in urine, 353 Tunnel workers' anemia, 146 Tunnicliff's stain for spirochaete, 638 Turk's iodin stain, 484 counting chamber, 434 Two-glass test, 342 Typhoid bacillus in the blood, 565 in feces, 121 Drigalski and Conradi's media, 122 Kendall and Day's media, 122 in urine, 356 fever, blood in, 530, 565 feces in. 121 Widal reaction in, 584 690 INDEX. Typhoid pleurisy, 643 Tyrosin in sputum, 14 in urine, 233, 333 Uffelmann's test for lactic acid, 70 Uhlenhuth's antiformin method for tubercle bacilli, 18 Ulcer of the stomach, 87 Ulceromembranous angina of Vincent, T 39 Unaltered bile in feces, 101 Uncinaria Americana, 144 duodenalis, 143 Uncinariasis, 143 Undetermined nitrogen of urine, 233 Unilateral nephritis, 360 Unit of counting chamber, 439 Unna-Tanzer's stain, 13 Unorganized sediments in urine, 327 Unoxidized sulphur of the urine, 197 Uranium method for phosphates, 191 Urates in urine, 328 Urea in blood, 422 in urine, 209 amount of, 210 determination of, 211 variations of, 210 Uremia, blood in, 512 urine in, 360 Ureometer of Doremus, 212 of Hinds, 213 Ureteral calculi, 357 Ureteritis membranacea, 238 Urethritis, anterior, 342 posterior, 342 Uric acid, 219 calculi, 359 diathesis, 221 in blood, 423 in the urine, 219 determination of, 223 metabolism of, 219 variations of, 221 sediment, 327 Uricacidemia, 422 Urine, 163 acetone in, 180, 305 acidity of, 172 albumin in, 247 albumoses in, 263 alkaline tide of, 175 alkapton bodies in, 171, 321 alloxur bodies in, 180, 228, 330 amino-acids in, 233 ammonia in, 215 amount of, 165 animal gum in, 301 animal parasites in, 356 appearance of, 168 ash of, 180 bacteria in, 353 Bence-Jones protein in, 263 bile acids in, 320 biliary pigments in, 317 black, 171 blood cells in, 344 Urine, pigment in, 316 blue, 171, 243 calcium in, 201 calculi in, 357 carbohydrates in, 270 carbonates in, 200, 270, 338 casts in, 345 changes on standing of, 168 chemistry of, 179 chlorids in, 181 cholesterin in, 335 chromogens in, 240 chyle in, 170, 336 clearing of, 283, 285 collection of, 164 color of, 169 composition of, 179 consistence of, 169 creatin in, 229 creatinin in, 229 cryoscopy of, 361 cultures of, 353 cystin in, 198, 331, 360 dextrin in, 302 dextrose in, 270 diacetic acid in, 312 drug reactions in, 171, 324 Ehrlich's benzaldehyde reaction in, 324 diazo reaction in, 322 egg-yellow reaction in, 323 electric conductivity of, 361 epithelial cells in, 339 fat in, 335 fatty acids in, 235 ferments in, 238 fibrin in, 269 foreign bodies in, 356 free mineral acidity of, 174 organic acidity of, 174 functional diagnosis from, 360 glucose in, 271 glycosuric acid in, 321 glycuronic acid in, 302 green, 171 hematoporphyrin in, 317 hemoglobin in, 268, 316 hippuric acid in, 234, 335 histon in, 270 homogentisic acid in, 171, 321 indican in, 243 indigo in, 243 inosite in, 302 iron in, 202 lactic acid in, 237 lactose in, 300 laiose in, 293 leucin in, 233, 332 leucocytes in, 341 levulose in, 293 magnesium in, 201 maltose in, 301 melanin in, 171, 321 microscopy of, 325 mucin-like substances in, 239 mucoid material in, 339 INDEX. 691 Urine, neutral sulphur in, 197 nitrogen in, 203 nitrogenous bodies in, 203 nubecula in, 168, 339 nuclein bodies in, 228 nucleo-albuminin, 239 nucleo-histon in, 270 odor of, 172 optical activity of, 179 organized sediments of, 339 oxalic acid in, 236 oxaluric acid in, 236 /?-oxybutyric acid in, 314 parasites in, 356 pentoses in, 296 peptone in, 268 phosphates in, 187, 335, 337 physical properties of, 165 pigments in, 239, 317 potassium in, 200 preservation of, 164 protein of, 247 proteoses in, 263 ptomaines in, 331 purin bases in, 228 pus in, 341 quantity of, 165 reaction of, 172 Russo's reaction in, 324 sediments of, 327 serum-albumin in, 247 globulin in, 261 skatoxyl in, 244 sodium in, 200 solids of, 177, 180 specific gravity of, 175 spermatozoa in, 353 sugar in, 270 sulphates in, 196 sulphur compounds in, 195 tissue fragments in, 353 total solids of, 177, 180 tyrosin in, 233 urates in, 328 urea in, 209 uric acid in, 219, 328 urobilin in, 241 urochrome in, 241 uroerythrin in, 241 urohematin in, 244 urorosein in, 247 xanthin bases in, 228, 330 Urinometer, 176 Urinous odor, 172 Urobilin, 241 Urobilinuria, 241 Urochrome, 241 Uroerythrin, 241 Uroferric acid, 197 Uroleucic acid, 171, 321 Urophain, 246 Urorhodin, 244 Uroroseinogen, 247 Urorubin, 244 Urostealith calculi, 360 Uterine secretions, 372 Vaccination, antityphoid, 578, 500 Vaccines, 577 diagnostic use of, 579 Vaccine therapy, 577 Vacuolization, 474 Vaginal secretions, 370 Vaginitis, catarrhal, 370 gonorrheal, 370 Value of blood examinations, 627 of functional renal diagnosis, 360 of search for tubercle bacilli, 21 Van Deen's test for blood, 102, 623 Vaquez' disease, 472 Variations in number of leucocytes, 489, 498 of red cells, 469, 478 Variola, blood in, 532 Venous blood, 385 puncture, 384, 563 Vermiculus, 552 Vernal conjunctivitis, 44 Vesicular mole, 374 Vincent's angina, 39 bacillus, 39 spirillum, 39 Vinegar eel in urine, 356 Viscosity of blood, 392 Vitali's test for pus, 343 Vital staining of blood cells, 461 Volatile alkalinity of urine, 175 Volhard's method for chlorids, 184 Volume index of blood, 382 of blood, 378 quotient, 382 value, 382 Vomitus, 55 bile in, 56 blood in, 56 fecal, 56 green, 56 mucus in, 56 odor of, 56 pancreatic fluid in, 56 parasites in, 56 pus in, 56 rice water, 56 Wang's method for indican, 245 Wassermann's serum reaction for syphilis, 596 Waxy casts, 348 Weber's test for blood, 103 Weidel's test for xanthin, 330 Weinstein's test for gastric carcinoma 9° Weyl's test for creatinin, 231 Whetstone crystals of uric acid, 327 of xanthin, 330 Whip worm, 141 White blood cells (see Leucocytes), 479 Whooping cough, bacillus of, 25 blood in, 534 Widal reaction, 584 Williamson's blood test in diabetes,526 Winternitz method for gastric motil- ity, 82 692 INDEX. Wintemitz', Henry and McPhedrans' test for catalase, 500 Wright's coagulometer, 394 method for coagulation time, 394 Wright's opsonic method, 576 vaccine therapy, 577 stain for blood smears, 459 Wright and Kinnicutt's method for blood plates, 445 Xanthin bases in blood, 424 in feces, 116 in urine, 180, 228, 330 calculi, 360 Xerosis bacillus, 38 Xylose in urine, 296 Yeast cells in feces, 119 in gastric contents, 56, 57 in sputum, 15 in urine, 275 Yellow fever, blood in, 560 mosquito theory of, 560 Ziehl-Neelsen method for tubercle bacilli, 19 Zygotes, 550 Zymogens in gastric juice, 73, 76 Zymophore, 575