F723< 1888 ^pi^-sS'£■■ '■'- ■ ■'■ '■■ v. .' STIONAL lIBRARV of medicine NLfl DD12Bbm Q \2L*&K&*w£rrs*-, !*!*&£*£ SURGEON GENERAL'S OFFICE Section, JVO./Z3/67. rr j»^ Jk-. ' m j» as_as_»*_ V*@ NLM001226140 v= / Xk (JOMPUMENTS OF THE AUTHOR. COMPARATIVE STUDIES --- OF --- IganiiQaliait Blood, With Special Reference to the Micro- scopical Diagnosis of Blood Stains in Criminal Cases. HENRY F\ FORMAD, B.M., IVL.D hi Lecturer on Experimental Pathology and Demonstrator of Morbid Anatomy in the University of Pennsylvania; Fellow of the College of Physicians of Philadelphia; Member of the Association of American Physicians and of the Franklin Institute; Vice-President of the Path- ological Society of Philadelphia; Pathologist to the Philadelphia (Blockley) and the University Hospitals ; Coroner's Physician of Philadelphia; Etc. With Sixteen Illustrations from Photo-Micrographs and Drawings. JJL3/J.7- Philadelphia : A. L. HUMMEL, M. D., Publisher, 224 South Sixteenth St. ■» 1888 -K< n v-\ v> <-j'/' vVGr PREFACE. '"PHROUGH the liberality of the Editors and the Publisher of the Journal of Comparative Medicine and Surgery, I am enabled to publish in the present form, for distribution among some of those interested in the subject of blood stains, this reseafch, which has emanated from the Pathological Laboratory of the University of Pennsylvania. The substance of this contribution to Medico-legal Science formed a paper read before the College of Physicians of Phila- delphia, May 2, 1888 ; published by permission of the College. Department of Pathology, University of Pennsylvania, July, 1888. TABLE OF CONTENTS. Page. Introductory Remarks, .... Physical Properties of Blood, Blood Under Changed Conditions, Chemical Tests, ...... Diagnosis Between Human and Animal Blood, Methods of Measurement, .... Micro-Photography, ..... A New Method of Measuring and Diagnosing Red Blood Corpuscles, .... Tables of Sizes of Corpuscles as Given by Various Observers, Biological Review of Measurements, Conclusions Regarding Diagnosis of Fresh Blood, Diagnosis of Blood in Criminal Cases, Sources of Blood, ..... Age of Blood Stains, ..... Diagnosis of Human Blood, Extraneous Matters in Blood Stains, Methods of Preparation and Examination of Blood Stains, Table of Experiments, Conclusions Regarding Diagnosis of Blood Stains, Sources of Error, Expert Testimony ; Legal Aspect, Requirements to be Fulfilled, Precautions, .... Authorities " Pro " and " Contra," List of Text-Books, Errata, ..... LIST OF ILLUSTRATIONS. rage. Figs. I, 2 and j. Blood Under Changed Conditions, . 4, 7. 9 Plate I, Fig. 4. Micrometry of Ox and Human Blood, . 13 Plate I, Fig. 5. Red Blood Corpuscles, Fresh Human Blood, Magnified 2,250 Diam., . • 13 Plate II, Fig. 6. Blood Corpuscles of Man ; Magn. 1,450 Diam., . . . . . . • 17 Plate II, Fig. 7. Blood Corpuscles of Guinea Pig ; Magn. 1,450 Diam., . . . 17 Plate III, Figs. 8, p, 10, 11 ana 12. Single Blood Corpuscles of Man, Dog, Ox, Sheep and Goat ; Magn. 10,000 Diam., . . . .21 Plate IV. Gulliver's Diagram of the Comparative Sizes of Red Blood Corpuscles in the Various Classes of Animals, . . . -24 Plate \\ Fig. ij Red Blood Corpuscles Recovered from Dried Blood of Man; Magn. 2,250 Diam.. . 32 Plate V, Fig. 14.. Red Blood Corpuscles of Ox Prepared and Magnified Similarly, . . . -32 Plate VI. Extraneous Substances Met in Blood Stains, . 35 Fig. 15. Blood Corpuscles Restored from Blood Stains, Caustic Potash Preparation, . 39 Reprint from The Journal op Comparative M edicine and Suroery July, 1888. Art. XIX.—COMPARATIVE STUDIES OF MAMMALIAN BLOOD. WITH SPECIAL REFERENCE TO THE MICROSCOPICAL DIAGNOSIS OF BLOOD STAINS IN CRIMINAL CASES.* BY HENRY F. FORMAD, B.M.,M.D., Lecturer on Experimental Pathology and Demonstrator of Morbid Anatomy, in the University of Pennsylvania—Coroner's Physician of Philadelphia, etc. Introductory Remarks—I. Physical properties of Blood. General considerations. Morphology of the blood corpuscles. Oviparous and mammalian blood. Effects of poisons and of disease upon blood ; blood outside of the body ; effects of dessication and moisture; chemical and spectroscopic tests.—II. Diagnosis between fresh human blood and that of animals. General distinctions. Uniformity of corpuscles in the individual. Measurements ; methods of preparation ; difference between dry and moist preparation regards diameter of corpuscles. Modes of measurements. Eyepiece micrometry; stage micrometry and photography; photo-micrographs of corpuscles of various ani- mals. Re-photographing of the same for diagnosis and gross measurements. Table of comparative measurements by the various observers. Gulliver's plate of comparative diagrams of the sizes of corpuscles; his classification and measurements. Critical review of the various measurements. General conclusions regards diagnosis of fresh blood.—III. Diagnosis of human blood in criminal cases. Liquid blood ; dried blood ; blood-stains ; old and new methods; liquids employed, methods of preparation ; ex- periments ; measurements, difference between re-moistened and fresh blood. Conclu- sions.—IV- Expert testimony upon blood in criminal cases. Abuse of science, perver- sion of facts. Peculiar cases from personal experience. Suggestions and precautions—V. Bibliography,''pro " and "contra," classified. Introductory Remarks.—The present article has been written to record personal observations in the domain of comparative histology of mammalian blood, together with a brief account of what is generally known about the micro- scopy of blood and blood examinations, in order to elucidate the subject more fully, and to make it intelligible to those not familiar with haematology. I made careful inquiry and experiments regarding points disputed in medical legal science, and I present the plain facts and conclusions as obtained from personal studies which extend over a num- ber of years. I submit, however, at the same time the evidence of others upon the same points, such as it is, •Read before the College of Physicians of Philadelphia. 2 Comparative Studies of Mammalian Blood. "pro " and "contra." The literature in this line of obser- vations is not extensive, as far as original microscopical studies are concerned. True, there are numerous works and papers (See "Index Medicus ") containing quotations from this or that author- ity on blood, such as the writers—writers often quite distin- guished, but who are not microscopists themselves—saw fit to refer to, or had access to. I sympathize with the author of a text book on Medical Jurisprudence who is not a practical microscopist himself, if he is undecided whose observations on blood he is to accept as correct, and consequently expresses an adverse or guarded opinion ; and I fully sympathize with the lawyer or physician who, as a rule, has access only to such books upon this particular subject that have outlived their use- fulness, and which only interfere with the cause of science and justice. The character and nature of expert testimony as regards blood stains in criminal cases will be fully considered and some peculiar incidents from my own practice will be cited. I do not enter here into the full details of chemical exam - ination of blood, and do not consider at all the spectro- scopic analysis, beyond a mere definition of the method, as both are beyond the scope of the present paper. Moreover, chemical and spectroscopic investigations are of subordi- nate value in the comparative diagnosis of blood ; the microscope only can decide the origin and source of any given specimen of blood, while the former can only estab- lish the mere presence of blood. I append in the bibliography (as complete as was possi- ble) those authors upon blood stains, and upon comparative studies of blood, which might be found useful for refer- ences ; as will be also the copy of Gulliver's diagram upon the comparative sizes of blood corpuscles which I introduce. (See Plate IV.) The rest of the illustrations in this article are all original, with the exception of one micro-photograph, by Seiler and two by Sternberg. My own photo-micrographs lack in artistic execution. Yet they are quite useful for the study Comparative Studies of Mammalian Blood. 3 of the comparative sizes of corpuscles, and have been de- clared satisfactory by expert photographers. At this place I wish to acknowledge the valuable assist- ance in measuring daily, for many months, blood corpus- cles in several of my legal cases, and making innumerable calculations, to Drs. A. J. Plumer and J. Leffingwell Hatch, and further I wish to thank Drs William Gray, A. J. Plumer, I. W. Blackburn and Mr. French, for assistance in photography and for making drawings. The sixteen micro-photographs were accurately reproduced by the Levytype Autoglyphic Process, through the liberality of the editors of this journal. But particularly do I wish to thank Mr. S. H. Ashbridge, Coroner of Philadelphia, Mr. Thomas J. Powers, his predecessor, as well as the District Attorneys throughout this State, for their kindness in sup- plying me with much material for study. I. Physical Properties of Blood.—The blood corpuscles, which form the subject of this paper, constitute, as is well known, the solid portion of the blood, and represent about one-half of the total volume of blood. The blood is an apparently red, viscid liquid of an alkaline reaction, and a specific gravity averaging 1055 in all animals. In its fresh state it has a salty taste, and an odor more or less peculiar to the animal, which is intensified on the addition of sulphuric acid and the application of heat. The blood cor- puscles are suspended in a colorless, clear, albuminous liquid, the liquor sanguinis or blood serum, and can be seen only by the aid of the microscope. The specific gravity of the corpuscles alone is said to be 1088, while that of the serum is 1028. The blood corpuscles, which have been known since Mal- pighi described them in 1661, are of three kinds : the red and the white corpuscles and the so-called blood plates; the latter having quite lately been discovered (in 1878 and 1882) by Hay em,51, and Bizzozero,52,* and proven to be the essen- *ln order to avoid frequent foot notes, the reference to authorities will be found in the Bibliography, at the end of this article. The number attached to the names of the authors refers to the number of reference in the Bibliog- raphy. 4 Comparative Studies of Mammalian Blood. tial element in the clotting of blood, which clotting, under normal conditions, takes place on its immediate removal from the body. For studies upon these blood-plates see Osier, 53, Welch,54, etc These blood plates or third blood corpuscles are very numerous (about 1 to every 20 of the blood corpuscles), and are said to differ in quantity and appearance in the various animals, and undoubtedly, on further study, may serve as a diagnostic factor. The white blood corpuscles exist in a proportion of about 1 to 500 of the red ; under normal conditions they have a pale, milky, granular appearance, are spherical in shape when at rest, but are endowed with amoeboid motion, by means of which they may assume almost any shape, and are capable of creeping through the tissues into any recess that offers itself. They have nuclei which become more prominent on the addition of acetic acid, or even water. In size the white blood corpuscles have an average diame- ter of 1-2700 to 1-3000 of an inch, the size being the same in all vertebrates. White blood corpuscles outside of the blood ves- sels in masses, are called pus corpuscles. The white blood corpuscles Fig. 1. Red and white blood corpuscles- are the progenitors of Normal. Action of Acetic acid. the red blood corpuscleS) and if the transformation to red blood corpuseles is re- tarded, then there are seen numerous nucleated red blood corpuscles, which are larger than the rest, and are com- mon in certain blood diseases, together with a large increase in the number of white blood corpuscles. In foetal life these intermediate nucleated red blood corpuscles are quite numerous, paler and more spherical, and may be diagnostic. Genererally, in intra-uterine life the red blood corpuscles are nucleated larger, and show great variation in size. In the normal child or animal, however young, the blood corpuscles are of the same size as in the adult. The red blood corpuscle?, which, however, in reality are not red, but yellow (of a lighter or darker hue in the Comparative Studies of Mammalian Blood. 5 various animals), are in man and mammals circular, bicon- cave, non-nucleated discs, being thinner at the centre than at the rounded edge. The only exception to this is found in the red blood corpuscles of the llamas and camels, which are oval, but, contrary to oviparous red blood corpuscles, which are oval and nucleated, are without that feature. As early as 1681 Leewenhoek observed this fact almost simultaneously with his discovery of the compound micro- scope. The red color of blood is merely an optical delu- sion, due to refraction of the haemoglobin of the red blood corpuscles, which, as stated, have really a yellow color, which is readily observed under the microscope. In size, the mammalian red blood corpuscles vary from 1-2745 of an in h in the elephant to the 1-12325 of an inch in the musk deer (Gulliver), the average diameter of the corpuscles of most animals fluctuating between 1-3000 and 1-5000 of an inch, the thickness of the discs being from i to i of its diameter. The largest corpuscles are those of the Amphiuma, a Louisiana reptile, measuring 3I5 of an inch in diameter. (Vide details in next section and Plate iv.) The number of red blood corpuscles, according to Malessez, is 4,000,000 to the c;bic millimetre in normal human blood. Vierordt and Welcker give 5,000,000 to the c. mm. Some clinicians regard 6,000,000 as the normal number. The goat, which has much smaller corpuscles (half as large as those of man), has 18,000,000 to the c. mm. (Toldt Gewebelehre.) So it appears that the smaller the corpuscles the greater the quantity. The quantity may, however, vary in the blood of different animals as well as in man. In the rabbit (according to Wormley) there are about 3,500,000 perc. mm. According to Dana, the ratio of the weight of the total bulk of blood to the weight of the body is considered for man to be 1-13, likewise in the dog ; but in the majority of domestic animals it is less ; for instance, the cat, 1-14 ; horse, 1-15 ; rabbit, 1-18 ; guinea pig, 1-19 ; calf, 1-21 ; sheep, 1-24 ; pig, 1-26 ; ox, 1-29. Morphology of red blood corpuscles.—The fact that mam- malian blood has no nuclei is no more disputed, the appar. 6 Comparative Studies of Mammalian Blood. ent nucleation being due to its biconcavity, on account of which the centre of the corpuscle appears dark in one focus with a light periphery, while in another focus the reverse occurs. Studies of the minute structure of the red blood corpuscles show (Rollet and Kollmann) that they are made up of a protoplasmic stroma which presents itself in the form of a colorless network, the fibres of which are albuminous and easily coagulable ; in the meshes of this network is a fluid, which contains the haemaglobin. The network of the stroma is attached to an investing mem- brane or cell wall of the blood corpuscle, if such exist. Although often rejected, the idea of a cell membrane in red blood corpuscles appears from time to time advocated by various observers. If there is a cell wall, it surely must be one possessing great elasticity, as every microscopist knows how soft, slippery and extremely flexible red blood corpuscles are. I do not think, however, that the red blood corpuscles have a cell wall or membrane, but agree with those who regard it to be only the outer hardened layer of the proto- plasm of the corpuscle. In my observations upon the microscopic changes produced by the venom of serpents upon blood corpuscles, which I made at the request of Dr. S. Weir Mitchell,* I have often seen, under the micro- scope, red blood corpuscles fuse into a colloid mass, which can be stretched and drawn out like molasses candy. I described these changes as follows : (See Fig. 2.) "The blood discs lose their bi-concavity and assume a spherical form, but without parting with their coloring matter. They exhibit also great adhesiveness, arranging themselves into various sized and shaped aggregations. The corpuscles comprising these groups sometimes appear to fuse so that their outlines cannot be deter- mined under the microscope, even by the highest amplification In addition, the corpuscles seem to soften and acquire a peculiar duc- tility and capability to be stretched into threads without fracture. By inclining the stage of the microscope, or making gentle pres- sure upon the cover glass, allowing, thereby, the liquid to flow, * " Researches upon the Venom of Poisonous Serpent■<," by S. Weir Mitchell and Edward T. Reichert. Published by Smithsonian Institute, Washington D. C, 1886. Chap, on Pathology, by H. F. Formad, p. 133. Comparative Studies of Mammalian Blood. the red blood corpuscles may be seen to elongate themselves into spindle-shaped or even into fine thread-like bodies. (See Fig. 2.) Such masses of corpuscles appear to act like colloid material. "This remarkable condition I found, in all experiments with venom, to be of only temporary duration. After a short time, which, in about a hundred observations, was found to vary from a few seconds to a quarter of an hour, the apparently homogeneous cell-mass breaks up anew into individual corpuscles of smaller but uniform size, which then continue to be isolated, or in bead-like rows, but remain spheroidal, i.e., do not regain their disc-like, bi- concave shape." Figure 2. A, normal red blood corpuscles lying flat and in rouleaux. B, the same, having oecume spherical upon application of crotalus venom, and at C, fused together partly and drawn out into soft string-like bodies ; D, five or more corpuscles fused and stretched out into a thin beaded thread. Magni- fied 400 diameters. The above experiments having been made on various ani- mals, prove that the red blood corpuscles have no cell wall, as the venom has no such effect upon cells that are known to have a cell wall. Under the influence of any liquid reagent of low specific gravity, e.g., water, the flat, bi-concave shape of the cor- puscles is lost, from the imbibition of the reagent, and it becomes spherical, and, consequently, reduced in diameter. Even the oval corpuscles of oviparous blood swell up to a spherical form under this treatment, and may be mis- 8 Comparative Studies of Mammalian Blood. taken for mammalian blood. (Vide Gulliver's plate (Plate IV.), 3d figure and xii., 2d figure, and figure 3 in text). I have, however, never seen any normal red blood cor- puscle increased in size by the action of reagents. Disease is liable to alter the size of corpuscles. The effect of high fever and of exhaustion in diphtheria, tends to diminish the corpuscles, by transforming them into more or less spheroidal bodies with reduction of size. I had the opportunity to observe this repeatedly with Dr. Horatio C. Wood, in our observations on Diphtheria, for the National Board of Health, during an epidemic of this disease in Michigan and this city.* The corpuscles fail to form rouleaux in fatal cases of Diphtheria, in consequence of the loss of their bi-concavity. According to Manassein, who made extensive observations at the Military Medical Academy, at St. Petersburg (Cen- tral Blatt fur Med. Wis., 1871), the blood corpuscles dimin- ish in size from the effect of high temperature and car- bonic acid gas, and he also found that this occurs in Septicaemia. On the other hand, blood diseases proper are capable of giving rise to an increase in the size of the red blood cor- puscles. In pernicious anaemia and chlorosis, this has been observed by Eichhorst. The corpuscles become of uneven size, some are enlarged, while others are diminished below average size. I can corroborate this observation. Further, it is said (Manassein and others), that the red blood corpus- cles enlarge from the effects of agents lowering the tem- perature of the body, such as Alcohol and Quinine. In my own observations in the pathology of Alcoholism, made upon many hundreds of fatal cases, I have found that alco- hol has no such effect. Blood, outside of the body, when slowly evaporating or drying, "en masse," shows a diminution in the diameter of the corpuscles ; it never, however, shows an increase, except when a thin layer of fresh blood is dried upon a glass slide, and then this increase is hardly appreciable. I made *Vide Memoirs on the Nature of Diphtheria. By Drs. H. C. Wood and H. F. Formad. Report of the National Board of Health for 1882. Comparative Studies of Mammalian Blood. 0 a series of experiments with blood, regarding the behavior of the corpuscles, as affected by time, under the most vary- ing conditions, to which I will refer at another place. In general, it may be said that rapid desiccation stops all changes in the blood. Blood stains or clots of several years' standing, which had been thoroughly dry from the begin- ning, showed under proper treatment as perfect corpuscles as dried clots several days old. Blood which dries slowly does not give good results, and moisture may lead to such degree of decomposition within several days or weeks that no method will ena- ble us to establish the kind of blood; the shape and size of the corpuscles are much altered, and some- times it is impossible to make out even the outlines of the corpuscles. Under such circumstances, and when the question relates only to establishing the presence of blood, without regard to its source, the chemical and spectroscopical tests do good service, and can be relied upon. Hcemin crystals, which represent a product of decom- position of the coloring matter of the blood, may be prepared by the addi- tion of glacial acetic acid and sodium chloride to dried blood. A few granules of dried blood are pulverized on a glass slide together with a few granules of salt; having covered it with a glass circle, a drop of the acid is allowed to flow under; the slide is then submitted to heat, when the peculiar crystals appear. (Vide Fig. 3.) The crystals are also known as Teichmann's crystals, after their discoverer, who attributed to them diagnostic properties as regards the blood of different animals. This, however, has not been substantiated by later observations. Hsemin crystals can Figure 3. Normal white and red blood coi'puscles; a few are crenated from effect of drying. Hcemin crystals from human blood and a few corpuscles swol- len from action of water. 10 Comparative Studies of Mammalian Blood. be relied upon as indicating the presence of blood, but cannot be relied upon with certainty as indicating the kind of blood. The Guaiacum test is also an old and a good chemical test. Wormley describes it as follows : '' On treating a solution of the coloring matter of blood with an alcoholic tincture of guaiacum and an ethereal solution of hydro- gen peroxide, a deep blue coloration is produced, due to the oxida- tion of the guaiacum resin. The alcoholic solution should be freshly prepared from inner portions of the resin. The ethereal solution of peroxide of hydrogen, known in the shops as ozonic ether, may be prepared by suspending some pure barium dioxide in water, adding an equivalent quantity of dilute sulphuric acid, and extracting the liberated hydrogen peroxide by ether. A portion of the ether extract, if fit for use, will strike a beautiful blue or violet colora- tion on the addition of a fragment of chromic acid. " In applying this test, a drop of the blood solution, placed over a white surface or in a porcelain dish, is first treated with a drop of the guaiacum tincture, and then a drop of the ether reagent added, when, even if only a trace of the coloring matter of blood be present, a blue color will immediately or very quickly appear. A drop of a l-1000th solution of blood will thus immedi- ately yield a decided blue coloration ; and a l-5000th solution a quite distinct reaction. '' The test may be applied directly to the stain, if on a wh ite fabric, by moistening it with a drop of water, and then adding the guai- acum and ethereal solutions. Even the minutest shred of a blood- stained fabric may show this coloration. When the stain is on colored material, it may be as advised by Dr. Taylor, thoroughly soaked with a drop of water, and the liquid absorbed by slips of white bibulous paper; these, while still moist or after they have dried, are submitted to the action of the reagents. " This test will react even with very old stains, provided they are first well moistened with water ; and even when the stains have been washed, evidence of their nature may be obtained." The spectroscopic test for blood consists in passing light through a suspected liquid, and thence through a prism, and if the liquid contains the least trace of blood^ certain rays of light will be absorbed, which will cause cor- responding dark bands in the spectrum. This is an infal- lible test for blood, but is also useless when a differential diagnosis as to the kind of blood'is to be made, and is Comparative Studies of Mammalian Blood. 11 superfluous when a single red blood corpuscle can be seen by the microscope ; yet Sorby claims that the delicacy of this test is such that a faint spectrum can be obtained from a single red blood corpuscle. It was introduced by Hoppe- Seyler in 1862. I cannot spare it further attention in this article. II. Diagnosis Between Fresh Human Blood and that of Animals.—There is nothing by which human blood, "en masse," can be distinguished from that of animals, by the naked eye or by any chemical tests. The differentiation rests entirely upon the microscopical appearance, and chiefly upon the size of the corpuscles. The shape of the corpuscles is diagnostic in distinguishing oviparous blood (uon-placental animals) from mammalian (placental ani- mals). The corpuscles of the former have a nucleus, and are invariably oval and more or less convex, whereas the corpuscles of the mammals are devoid of a nucleus and pre- sent themselves as round, bi-concave discs ; the only excep- tion from this is the camel species of the ruminantia. (Vide Plate IV., Figs, i, k, 1, m, n.) The red corpuscles of all the mammals (with the excep- tions already stated) have absolutely the same appearance morphologically, but differ merely and solely in size. The distinction, then, whenever it can be made, is only to be established by the measurement of the diameter of the corpuscles. True, there are certain peculiarities in the hue of the cor- puscles. The highest developed animals have corpuscles which are somewhat deeper colored (yellowish red) than those of animals of lower type, and as we descend in the scale of animal life, the corpuscles become of paler yellowish hue, until they are quite colorless as in the amphyoxis. All observers seem to have noticed that the red blood corpuscles of the normal man are, as a rule, the deepest in color, and, as some microscopists express it, "have the stamp of indi- viduality." * * The corpuscles of the dog appear to have a color next deepest in intensity to man, and it seems that those animals who have corpuscles approaching in diameter those of man. are likewise of a deeper hue ; such is the case in the kangaroo, opossum and others (Gulliver). The palest in color, among mam- mals within our reach, are those of the rabbit. 12 Comparative Studies of Mammalian Blood. I fully agree with this, after looking nearly every day at numerous specimens of human blood, often side by side with other kinds of blood in the laboratory instruction of my students, for a good many years. Yet, " impressions " do not come into consideration, and should not on the wit- ness stand. I will return now to the question of distinguishing human blood from that of other animals, by means of micrometry of their corpuscles. I will limit myself exclusively to mam- malian blood, as the oval, nucleated blood of the ovipara cannot be confounded with human blood. (See Plate IV.) Measurement of Blood Corpuscles.—Although the aver- age size of corpuscles in the same individual is quite uni- form, there are often seen corpuscles which deviate from the average, being either somewhat larger or smaller. The number of such corpuscles is, however, by no means as large as generally held, and their presence mostly due to changes subsequent to the removal of the blood from the body. Prominent physiologists assert that there are hardly any variations in size among the corpuscles of the same indi- vidual while circulating through the living blood vessels. This is in accord with my own observations,and I have found that in microscopic blood preparations, when successful, or at least in one out of ten of them, there will not be more than five to ten corpuscles among one hundred that deviate from the average in any perceptible manner. The distin- guished haematologist, Hay em, does not admit that there are, in the blood of an individual, more than twelve corpuscles larger, and twelve smaller than the average, among one hundred corpuscles measured. There may be differences in the size of the corpuscles in different indi- viduals of the same species, but these variations are, in my experience, very insignificant, and do not require further consideration unless abnormal conditijns come in question. Of some interest in this connection is the paper of Dr. J. G. Richardson,39, on the identity of the red blood in the different races of mankind. The average diameter of the human red blood corpuscles is generally given as 1.3250 on an English inch, or 0.0078 mm, PLATE I. OX AND HUMAN Fig. 4. Blood Corpuscles Side by Side, Magnified 500 Diameters. Micrometry Illustrated. Photo-Micrograph by Dr. Seiler. Fig. 5. FRESH HUMAN BLOOD. Red Blood Corpuscles, Magnified 2.>o0 Di- ameters. Photo-Micrograph. 1-18 Zeiss Horn Oil Immersion and Projection Kye-Piecc. Comparative Studies of Mammalian Blood. 13 Gulliver,26, and others, give it as 1-3200, which figures I have adopted as the average, because it better agrees with all the measurements I ever made, or any one of my assistants made. The variations in the size of individual normal human corpuscles ranges between 1-2900 and 1-3800 of an inch, with but few corpuscles of either extremes, the bulk (90 per cent.) measuring between 1-3100 and 1-3300 in. The means for establishing the average diameter of the corpuscle in any individual are as follows : 1. By micrometry directly. (See Fig. 4.) 2. By measuring photographic negatives of blood cor- puscles mounted upon a stage micrometer. 3. By re photographing micro-photographs of corpuscles and comparative gross measurements of the amplified photographs. (S^e Figs. 8, 9, 10, 11 and 12.) The method generally adopted for preparing blood for micrometric purposes, as well as for photographing, is, to spread it in a thin layer, single layer if possible, upon a glass slide, and dry it rapidly. This is best done by putting a small fresh drawn drop upon a slide and quickly drawing the edge of another slide across the field in such a manner that the corpuscles become evenly dis- tributed ; they may also be spread upon a cover glass. Only those preparations are successful which dry rap- idly. Only when immersion lenses are used is it neces- sary to cover the slide by means of a cover glass. This method has been attributed to Dr. Christopher Johnson, of Baltimore, although it appears that Gulliver and others employed the same method in their earliest observations. Blood may be examined fresh in its liquid state by put- ting a minute drop upon a glass slide, covering it with a glass circle, and, in order to prevent evaporation, ringing it with oil or melted paraffme. From repeated observation I have found that the blood corpuscles prepared by the dry method give diameters slightly larger than those of blood mounted in its liquid state. The reason is that the corpuscles are often more flattened from collapse of the stroma and loss of the bi- 14 Comparative Studies of Mammalian Blood, concave form, which results in a slight increase in their diameter. Yet this increase is scarcely appreciable even by micrometry, though it may account for the variations in the results of measurements. This is also in accord with Gulliver's,26* Eichardson's,2. DOG. (l-:(500.) 2 4-5 inches. Fig. 12. GOAT. (1-0100.) 1 :'.-5 inches. Single red blood corpuscles of man, dog, ox, sheep and goat, magnified all to the same scale, 10,000 diameters. Average-sized corpuscles, selected from micro-photographs made separately of each, under a uniform amplification of 2250 diameters, with 1-18 Zeiss Horn Immersion, and then re-photographed all alike to amplification of 10,000 diameters in each ease, "'he gross measurement reached by each is indicated upon the plate in inches. Comparative Studies of Mammalian Blood. 21 Such corpuscles of each animal which represent the aver- age size having been selected, the photographic amplifi- cation gives such results as are seen in Figs. 8, 9, 10, 11, upon Plate III. (page 273). Whereas, the difference in the sizes of blood corpuscles is often not quite obvious under low amplifications, it becomes very apparent and evident when enlarged to such magnitude by means of simple photography. The difference between 1-3.00 in. (man) and 1-3400 (guinea-pig) is only 1-54,400 inch. When magnified 10,000 diameter, this difference is equal to nearly 1-6 of an inch. It is out of place to speak here on the tech- nique of photography. I always prefer to have the assist- ance of an expert photographer ; this leads to good results and saves much time. When thus magnified to 10,000 diameters, we find that the corpuscles measure as follows : Human 1.3200 of an inch actual micrometry)=3i inches. Guinea pig (1-3400) =3 Dog (1-3500) =254 Ox (1-4200) =2* " Sheep (1-5000) =2 Goat (1-6100) =1§ " The first column of figures indicates the actual size of the selected corpuscles; the second gives the gross meas- urement of the same corpuscles when amplified 10,000 diameters and the photographic image measured (upon the negative) with an ordinary gross carpenter's tape-measure.* Such illustrations of the comparative sizes of the red blood corpuscles I found to be very convenient for use in lectures and for general demonstration of the subject o.i the witness stand. I do not entirely rely upon photo micrography for the direct and absolute measurements of corpuscles, but I do consider photography applied by my own method as the best and surest means for establishing the relative or comparative diameter of corpuscles when the question arises to decide between two or more kinds of blood. * Measurements are to be made upon the negatives. Printed photographs are not reliable for measurements on account of the stretching of the paper. COMPARATIVE TABLE OF THE AVERAGE RESULTS OF MEASUREMENTS OF RED BLOOD CORPUSCLES OF MAMMALS. Each column giving the average size (diametei) of the Corpuscles as obtained by various observers, expressed in fractions of the English inch, side by side with the common expression (roughly) in millimeters. Gulliver. 1845 and 1875. WoRMLLT, 1S85. O. Schmidt, 1848. Mallinin, 1875. 1 IFrencii Medico [ Legal Socie-TT, 1873, AND Wplker. Masson, 1885. Hans Schmid, 1878. Woodward, 1875. Personal Ob-servations. In. M.M. In. M.M. In. M. M. In. M. m. In. M. M. In. M. M. In. M.M. In. M.M. Great Anteater. 1.2745 1.2769 1.2769 1.2865 1.3000 1.3099 1.3557 1.3190 1.3200 1.3281 1 .3325 1.3550 1.3369 1.3413 1.3440 1.3538 1.3600 1.3532 1.3007 1.4000 1,3754 1.3693 1.3814 0.0092 0.0088 0.0088, 0.0086 0.0081 0.0080 0.0071 0.0080 0.0079 0.0078 0.0070 0.0072 0.0075 0.0074 0.0074 0.0071 0.0070 0.0071 0.0070 0.0003 0.0008 0.0070 0.0007 1.2738 .... Ornithorynchus Capybara .. .. 1.3145 1.3104 1.3250 o.ooso 0.00S0 0 0078 i I "i.3257 "6 0078 ."i'3257 ' "6'66r8 "i!3412 "6!6o74 1.3092 1.3300 0.0077 0 0082 1.3200 0.0079 Seal............ Beaver.......... Monkey......... Guinea Pig-. .. Wolf-........... Do?............ Rabbit-........ 1.3382 1.3410 1,3223 1.3422 1.3501 1.3053 1.3620 1.3052 1.3050 1.3743 1.3700 1.4140 1.4219 1.4208 1.4243 1.4372 1.4912 1 0189 0.0074 0.0079 0.0071 0.007] 0.CO7O 0,007(1 0.007'0 o.oooo 0.0067 0.0067 9.0001 0.0000 0.0059 0.0059 0.0058 0.0951 0.0041 ... 1.3213 'T3246 0.0079 "6.0078 1.3400 1.3450 1.3580 1.3662 1.3300 ( 0 0075 1.3030 1.3908 6.0070 0.00O4 "i.'3485 1.3053 "6.0073 0.0069 1.3577 1.3030 ....... 0.0071 0.0070 1.3840 6.0000 ........ 0.0074 0.0071 0.0069 1.390!-1.5000 0.0004 0 0051 1.5000 0.00511 '. . ..... ...... Squirrel........ Ox............. Pig............. Horse.......... Cat............. Sheep........... i.4o6o 1.4207 1.4230 1.4600 1.4404 0.0004 0.0000 0.0000 0.0055 0.0058 "1.4354 1.4098 1.4404 1.4545 1.5049 1.0309 "67)058 0.0002 0.0057 0.0051; 0.0045 0.0040 "i,4545 0.0050; 1.4098 0.0002 1.4545 0.0050: 1.3922 0.0005 1.5070 0.0050 1.5525 0.0040: 1.4237 1.4098 1.4440 1.4404 6.0000 0.0002 "6.0057 0.0057 '1.4695 1.4098 "i.0000 "6.0054 0.0002 "6.'6042 11 "i*4266 1.4250 1.4310 " i!5666 1.6100 0.0060 0.0060 0.0059 6!0051 1.6366' 0.0040 0.0042 I a 8 a a be © a. Comparative Studies of Mammalian Blood. 23 This table is an accurate compilation of the average measurements of the red blood corpuscles of mammals, as given by the various original observers, each being repre- sented by a separate column. All other tables of measure- ments, as given in books, are according to some one of the authorities quoted in this table, some of whom are not accessible to the casual reader. For convenience, I have put the English fraction of an inch and the French millimetre, side by side; the former being transformed into the latter, or the latter into the former, as the case required, by Drs. J. L. Hatch and A. J. Plumer, and my brother, Dr. R. Formad, Jr., Veterinarian. A study of this table will show its usefulness ; it illus- trates and elucidates many interesting points. It shows that the results obtained by the various observers,as regards the micrometry of the blood corpuscles of the majority of animals, is remarkably uniform, and that some of the measurements made by Gulliver,26 with imperfect instru- ments, nearly fifty years ago, are in accord with those made with the more perfect instruments of the present day. (In fact, Gulliver states that Jurin, 150 years ago, estimated the human blood corpuscles as 1-3240 of an inch in diameter.) On the other hand, the table shows there is quite a dls crepancy as regards the diameters of the corpuscles of some animals, so that it entirely depends upon whose figures we accept, whether we can or cannot discriminate between the human blood and the blood of certain animals. The most extensive measurements are those of Gulliver. I have given in this table only a small part of his work, viz.: that relating to measurements of the corpuscles of those ani- mals which have a more or le ss peculiar interest. The total number of his measurements embraces nearly 800 animals, and extended over thirty-five years. He made them solely from a biological standpoint, claiming that the blood cor- puscles are one of the prominent means of the classification of animals into species. Gulliver is the pioneer in haematology, and it may be interesting to note his own opinion regarding micrometry. He says : '' My tables cannot pretend to absolute exact 24 Comparative Studies of Mammalian Blood. ness, and are only offered for what they may be worth ; and in the estimation of their value, allowance should be made for errors, whether instrumental or personal, more or less inevitable, notwithstandii:g the greatest care, in observations so extensive." " Nevertheless," he adds, "the relative value of the measurements, though probably not unexceptionable, may be entitled to more confidence as fair approximation to the truth." No doubt, the comparative relations of the sizes of the corpuscles are given correctly by Gulliver; accurately enough for scientific as well as all practical purposes. They are nearly uniformly in accord with all later measurements. Gulliver made his measurements both from fresh blood and from blood thinly smeared and dried upon a glass slide, pre- cisely by the same method as is generally used at present. A most convenient and beautiful illustration of the differ ence in the sizes of red blood corpuscles of the various ani- mals, is Professor Gulliver's plate, which I here reproduce. The diagrams upon this (Gulliver's) plate give a most ex- cellent idea of the comparative diameters of blood corpus- cles in fresh blood. I reproduce also Gulliver's own explanation of the plate and the classification of the animals, which is quite inter- esting from a biological standpoint, and it explains itself. It gives the measurements of the diameters of corpus cles of a number of animals that are not incorporated in my table of comparative observations. Both plate and explanation of plate (page 278) are accurate copies from Gulliver's famous article; I only added the English names to the Latin denominations. (Some oversights of the engraver as regards omission of figures on the plate are explained in the text.) Explanation of the figures upon Gulliver's plate—All the ob- jects are red blood corpuscles done to one and the same scale, which is at the foot of the drawing- The whole length of the scale represents 1-1000 of an English inch, and each one of the ten divis- ions 1-10,000 of an inch. Only corpuscles of the average sizes and quite regular shapes are given; and they are all magnified to the same, to wit, about 800 diameters. For details see description be- low.* ♦It seems to be 900.—H. F. F. PLATE IV. Gullivers micrometry of red blood corpuscles, all to a uniform scale. I. MAN. II. QUADRUMANA. III. CHEIROPTERA. I OOo | ooo |ooo IV. PER^, looooOOoooo Pq r 3 >t , u wxyz V. CETACEA. VI. PACHYDERMATA. IOOO|Oooooo VII. RUMINANTIA. |ooo oooOOOOOOOo^Own 'abc de f g h , klmn *-------.-------< t o VIII. RODENTIA. IX. EDENTATA. X. MARSUP. XI. MONOTR |OOoo|OOo|oo|o XII. AVES. 3 4 5 6 7 8 9 10 XIII. REPTII.IA ET BATRACHIA. 11 12 13 14 Lepidosuen. XIX. PISCES Perca. Tinea. Esox. Salmo. Gymnotus. SquaTus. Ammocoutta. ntaith of an inch ,.......... x 900 6. Gulliver ad. nat. del. I Comparative Studies of Mammalian Blood. 25 A.—YERTEBRATA APYRENJBMATA. ( See Plate IV.) I. Homo (Man)........................................... 1-3200 *1. Corpuscles lying flat. 2. The same on edge. 3. Membraneous base of same after removal by water of coloring matter ; it shows diminution in diameter on account of acquired spherical shape. II. QUADRUMANA (MONKEYS.) 4. Simia troglodytes (Chimpanzee)................... 1-3412 5. Ateles ater. (Black-faced spider monkey)......... 1-3602 6. Lemur anguanensis............................... 1-4003 III. Cheiropetera (Bats.) 7. Cynonycteris collaris (fruit bat).................. 1-3880 8. Vespertilio noctula (large bat).................... 1-4404 9. Vespertilio pipistrellus (common bat)............. 1-4324 IV. Ferje (beasts of prey. ) (p) 10. Sorex tetragonurus (shrew)..................... 1-4571 (i) 11. Ursus labiatus (lipped bear).................... 1-3728 (r) 12. Bassaris astuta civet .cat).................... 1-4033 (s) 13. Cercoleptes caudivolvulus (kinkajou).......... 1-4573 (l) 14. Trichechus rosmarus (walrus)................. 1-2769 (u) 15. Canis dingo (dog, Australian).................. 1-3395 (w) 16. Mustella zorilla (weasel)...................... 1-4270 (a) 16. Felis leo (lion).....-............................ 1-4322 (b) 16. Felis leopardus (leopard)....................... 1-4319 (x) 17. Felis tigris (tiger).............................. 1-4206 (y) 18. Paradoxurus pallasii (Pallas paradoxure)...... 1-5485 (z) 19. Paradoxurus bondar (Bondar paradoxure).... 1-5693 (a) 19. Hyena striata (striped hyena).................. 1-3735 V. Cetacea. (Whales.) 20. Balaena (boops—whale).......................... 1-3099 21. Delphinus globiceps (caing—whale).............. 1-3200 22. Delphinus phocaena (porpoise).................... 1-3829 VI. Pachyderm at a. 23. Elephas indicus (elephant)...................... 1-2745 24. Rhinoceros indicus (rhinoceros).................. 1-3765 25. Tapirus indicus (tapir)........................... 1-4000 26. Equus caballus (horse)........................... 1-4600 27. Dicotyles torquatus (peccary)................... 1-4490 28. Hyrax capensis (Cape hyrax)................... 1-3308 VII. Ruminantia (Ruminants.) (») 29. Tragulus javanicus, (Javan chevrotain, musk deer)........................................ 1-12325 ♦Through an oversight, some of the figures are not marked upon the plate. 26 Comparative Studies of Mammalian Blood. )30. (-) 31. ( ) 32. (e) 33. (f)34. (8) 35. (>') 36. 0)37. (k) 38. 0)39. (m) 40. (n) 41. (") 42. 1-3361 1-6229 L. D. 1-3361 1-6229 Tragulus meminna (Indian chevrotain)........ 1-12325 Tragulus Stanleyanus (Stanleyan chevrotain).. 1-10825 Cervus nemorivagus (deer).................... 1-7060 Capra Caucasica (Caucasian ibex)............. 1-7045 Capra hircus (domestic goat)................... 1-6366 Bos urus (represented by Chillingham cattle).. 1-4267 Camelopardalis giraffa (giraffe)............... 1-4571 A , - . , . . j L. D. 1-3555 Auchema vicugna (vicuna)............i g^ ^ i_«587 Auchenia paca (alpaca)...............j oV j^ Auchenia glama (llama)............... „; jy Camelus dromedarius (single hump) L. D. 1-3254 camel................•............... / Sh. D. 1-6931 Camelus bactrianus (double humpjL. D. 1-3123 camel)..............................\ Sh. D. 1-5876 Cervus Mexicanus* (deer—Mexican)........... 1-5175 VIII. Rodentia (Rodents). 43. Hydrochcerus capybara (capybara)............... 1-3190 44. Castor fiber (beaver)............................. 1-3325 45. Sciurus cinereus (squirrel)....................... 1-4000 46. Mus messorius (harvest mouse).................. 1-4268 IX. Edentata. 47. Myrmecophaba jubata (ant eater)................ 1-2769 48. Bradypus didactylus (sloth)..................... 1-2865 49. Dasypus villa (armadillo)....................... 1-3315 X. Marsupialia. 50. Phascolomys (wombat).......................... 1-3456 51. Hypsiprymnus setosus (kangaroo rat).......... 1-4000 XI. Monotremata. 52. Echidna histrix (echidna)....................... 1-3840 B.-VERTEBRATA PYRENJSMATA. XII. Aves (Birds) L D. Sh. D. 1. Struthio camelus (ostrich)................ 1-1649—1-3000 2. The same made round and deprived of color by water. 3. Vanga destructor (East India shrike)..... 1-2019—1-3892 4. Lanius excubitor (great grey shrike)...... 1-1989—1-5325 5. Bubo virginianus (horned owl)............ 1-1837—1-4000 6. Syrnea nyctea (snowy owl).............. 1-1555—1-4042 7. Columba rufina (rufous pigeon)........... 1-2314—1-3329 8. Columba migratoria (wild pigeon)......... 1-1909—1-4626 9. Dolichonyx oryzivorus (rice bird)......... 1-2400—1-4167 * The only animal in which the red blood corpuscles present a variety of shapes in thf snme individual .—Gulliver. Comparative Studies of Mammalian Blood. 27 10- Buceros rhinoceros (rhinoceros hornbill)... 1-1690—1-3230 11. Psittacus augustus (August amazon)..... 1-2085—1-3606 12. Phasianus superbus (barrel-tailed pheasant) 1-2128—1-3587 13. Pelecanus onocrotalus (white pelican)..... 1-1777—1-3369 14. Trochilus sp. (humming bird)............. 1-2560—1-4000 Figures XII., XIV., XVI., XVII. andXVIII. represent red blood corpuscles of Reptilia and Bactrach'a; while under figure XIX. those of the fishes are given. In all these figures the names of the animals are inserted upon the plate, and they do not require any comment at this place. It is evident that the blood corpuscles of the Amphiuma are so large that they can be perceived by the naked eye. Carl Schmidt,7 the Russian pioneer haematologist (Dor- pat, 1848), although making his measurements of blood cor1 puscles with an amplification of only 500 diameters and drawing his averages from only forty measurements of the corpuscles of each individual, furnished accurate micro- metric figures of diameters of blood corpuscles. In fact, they are, with but few exceptions, in accord with Gulliver's (1S42) and with measurements made of late years. Carl Schmidt is the father of micrometry as applied to blood stains in criminal cases (1840). It is peculiar that he made most of his measurements of corpuscles from dried blood ; tcuting thin sections, by means of a razor, from blood clots and examining them in oil. But he, as well as Gulliver, also used our "modern" method of spreading single layers of blood corpuscles on glass slides and drying them rapidly. From measurements of his own he showed that there was a reliable mean average of the diameter of the blood corpuscles of the different mammals, which could be used for' diagnosis in criminal cases. He showed also, that if there was a shrinkage in the size of the corpuscles from drying, this should not debar us from an accurate diagnosis, because he established that the shrinkage was proportion- ately uniform in the blood corpuscles of all animals. His figures, quoted in my table, show that the latter proposition is correct, since all his measurements are.somewhat less than those of other observers, but in a constant, uniform ratio, making them correspond proportionately with the others. Wormley,46 Professor of Chemistry in the University of 28 Comparative Studies of Mammalian Blood. Pennsylvania, has furnished a good article on the examin- ation of blood, in the appendix of his work on the Micro- Chemistry of Poisons. I have quoted Prof. Wormley's fig- ures of average measurements, which extended over forty- six different animals (thirty-eight mammals, four birds and four reptiles), in my table of comparative measurements. See page 275. On the whole, they are practically in accord with Gulliver's and Carl Schmidt's measurements, and absolutely correct as far as they went. Wormley dis- agrees essentially with Gulliver only in regard to the measurements of the opossum and guinea pig (see table); bur, as he explains, the species of the animals examined in the latter case was not the same. Ti e late J. G. Richardson,19 w whose researches on blood, first published in 1869, have been followed with interest by medical jurists and biologists the world over, was unques- tionably the most reliable and most prominent American haematologist. He was the first to employ and to advocate such high microscopic objections as 1-25 and 1-50 in the diagnosis of blood corpuscles. Under an amplification of 3,700 diameters obtained with a 1-50 immersion ob- jective, Dr. Richardson found an average-sized human corpuscle to measure \ of an inch in diameter, while that of a sheep was only f of an inch across ; an ox blood corpuscle measured -J of an inch. Thus magnified, the comparative difference was quite apparent. I have often assisted Dr. Richardson in measuring blood corpuscles in his medico-legal cases, and profited much by learning his methods of measuring with high power. He was one of the most prominent advocates of a positive diagnosis of human blood from that of all domestic animals by meaDS of micrometry under high amplification. His observations in this field of study are extensive and well known. They are in accord with nearly all later observations, and will be referred to later on. Masson47 is the most recent of observers on the measure- ment of blood corpuscles. He thinks it difficult if not impossible to distinguish the blood corpuscles of man i rom thos° of the guinea-pig. From the results of his measure- Comparative Studies of Mammalian Blood. 29 ments he found the average diameter of the corpuscles of the guinea-pig to be 1-3300 inches (.0077 mm.). He meas- ured also the blood coi puscles of the rabbit, dog, ox, pig and sheep, with results quoted in my general table. With re- , gards to all these animals, however, he expresses the positive opinion that the blood of neither of those animals can be mistaken for that of man when careful measurements of the corpuscles are made. He says : " One can distinguish with certainty the blood of man and guinea-pig from that of the dog and rabbit, and the blood of the last two named animals from that of the pig, ox and cat." Lacour48 published his observations last year ; his results were identical with those of Masson. He asserts that if the average diameter of the blood corpuscles exceeds 1-127 of a mm. (1-3225 of an inch), then the blood is either human or that of the guinea-pig ; but if the diameter is less than 1-127 of a mm., then the blood may be that of the dog or rabbit. If less than 1-135 of a mm., then the blood is not human, nor that of the guinea-pig, rabbit or dog, and if less than 1-400 mm. it may pertain to the ox, pig, sheep, etc. General Resume.—From all the studies referred to so far, it can be regarded as established, that the microscopist has ample and sure means to diagnose fresh or well-preserved human blood from that of certain animals, provided he has the proper experience and employs rightful and honest means. Surely, human blood can be told from that of all the ordinary domestic animals, not counting the guinea-pig as a domestic animal. It depends, however, whose figures are accepted for the mean diameter of this animal's cor- puscles, whether guinea-pig's blood may be mistaken for human. All the animals, whose blood corpuscles closely approach in diameter those of man, are wild, or menagerie animals, and the micrometry of their blood corpuscles has no other but a purely biological interest, unless when improper use is made of it in the defense of criminals. Strictly speaking, only the following animals have cor- puscles larger than man, i.e., larger than 1-3200 of a inch: Elephant, great ant-eater, walrus, sloth, platypus, whale, capybara, and (according to Wormley) opossum. Animals, the corpuscles of which are slightly below man in size, i.e., 30 Comparative Studies of Mammalian Blood. having corpuscles from 1-3500 to 1-3200 of an inch average diameter, are the seal, beaver, musk-rat, porcupine, mon- key, kangaroo, wolf and guinea-pig. (See table page 275.) None of these are domestic animals. . All other ani- mals, including all domestic animals, have blood cor- puscles of a mean diameter, less than 1-3500 of an inch, and, in fact, those animals which, as a rule, are blamed for blood stains found on the clothing and apparel of criminals (ox, pig, horse, sheep and goat), have corpuscles with an average diameter less than 1-4000 of an inch (while all birds and fishes have oval corpuscles); and for a microscopist to say that such blood might be con- founded with human, is preposterous and ridiculous under the present state of knowledge, especially if the question relates to fresh or unaltered blood. It must also be remembered that with but a few excep- tions, all the animals whose red-blood corpuscles approach in size those of man are inhabitants of either the tropics (South America, Australia or Africa) or the Arctic regions, unless found caged in a menagerie. The exceptions are such animals as the guinea-pig and opossum, the geo- graphical distribution of which is also quite limited. There- fore a suggestion of the blood-expert that any one of those animals was to be blamed for the blood stains upon a person accused of murder, would be met with ridicule by the court, jury and public ; in fact, it is out of place for the expert to make any suggestions at all of that kind. (See section on expert testimony.) The suggestion that human blood may be mistaken for ox blood, or the blood of any domestic animal, on account of variations in blood corpuscles, is also out of place, since such variations rarely amount to more than one to three per cent., and conclusions as to the kind of blood are not drawn from the measurements of a few of the largest or smallest corpuscles. Conclusions are deducted from the mean of hundreds of measurements of average sized, round, well-shaped corpuscles, of which there are at least ninety per cent, in good preparation and under favorable condi- tions. Comparative Studies of Mammalian Blood. 31 It is a difficult task sometimes to diagnose rabbit's and dog's blood from human blood; the average diameter of the corpuscles of these animals being about 1-3600 in.; but only under unfavorable conditions. Fresh or well-preserved blood of these animals can be easily distinguished from human blood, by the quite appreciable smaller diameter plainly seen under high amplification. When it comes to diagnose guinea-pig's blood from that of man, then, how- ever, I would hesitate to make a positive distinction, since the difference in diameter between the two is too insignifi- cant. Conclusions Regarding Examination of Fresh Blood. 1. The blood corpuscles of birds, fishes and reptiles being oval and nucleated can never be mistaken for human blood. 2. Fresh human blood cannot be mistaken, under the microscope, for the blood of any animal, the corpuscles of which have a mean diameter of less than 1-4000 or even 1-3600 of an inch. 3. (a.)—If the average diameter of blood corpuscles in fresh blood is less than 1-4000, then it cannot possibly be human blood. (b.) If the diameter is more than 1-3500, then it may be human blood. (c.) If the blood corpuscles, after exhaustive measure- ment, give a mean diameter of more than 1-3300, then it is human blood (provided it is not the blood of one of the wild beasts referred to). So far we have considered exclusively fresh blood, and the conclusions stated above referred to the examination of fresh blood, or blood well preserved. We will now enter upon the consideration of a subject more difficult, viz.: the diagnosis of human blood, in a dried state, in criminal cases. III. Diagnosis of Human Blood in a Dried State and Blood Stains in Criminal Cases.—On rare occasions, liquid, or freshly clotted blood at a scene of murder, is to be examined, and its source to be established by the medical expert. The procedure of examination is the same as that described in the former section for fresh blood. j* - 32 Comparative Studies of Mammalian Blood. It is impossible to distinguish arterial from venous blood, yet, if the blood is sprinkled over a considerable area, then it is likely to be the former, whereas venous blood is likely to be in larger quantity in a mass and covering less space. Venous blood is just as red as arterial, when exposed to the air. Sometimes the question arises, whether blood is men- strual, and very frequently, whether its source was from the bleeding of the nose. Occasionally, blood specks upon linen garments are attributed to the biting of insects. Micro- scopical examination shows menstrual blood to contain a great deal of mucus and vaginal (large, elongated, flat) and uterine (columnar, ciliated) epithelial cells; the red blood corpuscles do not form rouleaux, and when fresh the blood has an acid reaction and is not coagulable. Blood from the nose (epistaxis) also contains a large amount of mucus, may contain large, columnar, ciliated epithelium from the snyderian membrane. The blood of epistaxis may contain much fibrine and may be coagulable ; reaction neutral. The elongated shape of the blood stains, and the location, direction and relation of the stains to each other, may be diagnostic. Blood stains, due to insect bites, are more or less peculiar in their location and distribution, and it is well to trace their source to the wearer of the garments. The blood stains from insect bites, etc., coming from the body of the person are more prominent on the inside of the fabric of the garment he wears. It is important in all cases of blood stains upon garments to examine the person of the accused. The age of dried blood stains is usually impossible to establish, because blood, once well dried, does not undergo any alterations ; yet a freshly dried clot will dissolve much more readily in water than an old one, whereas, dried blood, a day or two old, will disintegrate and will show liberated blood corpuscles often immediately from the effect of proper reagents. The facility of the disintegration is more difficult in direct proportion to the age of the clot (see table of experiments). This, however, depends, much upon the PLATE V. Fig. IS. HUMAN BLOOD CORPUSCLES RECOVERED FROM DRIED CLOT, TWO DAYS OLD. Magnified 2250 Diameters. Fig. 14. OX BLOOD. Fresh Dry Preparation. Magnified 2250 Diameters Both Photographed I'nder Same Amplification, IIS Zeiss Horn Immersion and Projection Eye-Piece. Comparative Studies of Mammalian Bload. 33 conditions under which the blood was kept, and I know it to be impossible to tell whether a blood stain is ten days or five years old. There is said to be a possibility of fixing the age of blood stains from the condition of the coloring matter. In one instance a hardened drop of blood upon a garment of a supposed murderer proved to be more or less liquid in its interior after the lapse of two days. This disproved his claim that the blood stain was three weeks old. Well dried and preserved blood and blood stains and like- wise mounted specimens of blood for microscopic demon- stration keep indefinitely. Is it Human Blood ? In some cases the expert is required merely to establish the presence of blood without regard to its source, this being often established by witness testimony; usually, however, he is required to tell whether it is human, orblood from some other animal. In former days great stress was laid upon the smell of the blood as developed upon the addition of sulphuric acid (1 part of blood to 1^ parts of sulphuric acid). and applica- tion of heat. This test was introduced by Burruel1 in 1829, when a case of murder was decided upon the evidence derived from this test, the celebrated Orfila coinciding with him. Subsequently the sulphuric acid test was adopted all over Europe, and many cases of murder were decided by it. Burruel claimed that an odor was developed by this test peculiar to the animal from which the blood was taken (blood of horse—stable smell; cow—cow stable smell; dog—dog smell; but particularly, Burruel claimed, do peculiar odors develop in the case of cat, sheep and goat). Ritter6 (1816), in an elaborate essay, for which he received a prize from the German government, proved that blood can be diag- nosed in criminal cases with great certainty by combining the sulphuric acid test with the measurement of red cor- puscles. It appears that he was the first to claim reliable results from the micrometry of blood in criminal cases. For the physical properties of dried blood, as well as for the consideration of the conditions which produce various changes, see I. and II. Sections. 34 Comparative Studies of Mammalian Blood. Dried blood presents itself to the examiner in various forms, either in dried masses embedded into substances or in the form of stains upon various substances, such as wood, stone, glass, instruments, and missiles of various kinds, and various kinds of apparel, such as cloths, linen, leather, etc. Blood stains are best seen by artificial light. The examiner must note with precision the general appearance (whether in spots, smears or drops), the exact size, shape, number, location and distribution of blood stains upon the articles sub- mitted to him. He must determine whether water was applied, as in washing. He must also carefully note any extraneous sub- stances associated with the blood, such as fragments of mineral substances, wood, wool and other fabric, bone, hairs, spermatozoa, epithelial, muscular and any other animal and vegetable cells. (See Plate VI.) The presence of any such substances in the blood may sometimes be indicative of its source, and, as well as the sur- roundings of the case, should be taken into consideration and may aid in the decision, whether the blood is human or not. The court always sustains any point of the expert that is properly demon- strated. The examination of dried blood en masse gives better results than mere stains, the shape of the corpuscles being better pre- served in the former than in the latter. As to stains proper, those upon any substance which does not absorb the blood give better results; viz., stains upon glass, stone, metal, etc., are better for examination than upon soft wood, and better upon cloth than upon linen. For examination a granule of dried blood, no matter how small, is preferable to any even larger diffused stain on a fabric. In order to examine blood in this form, the corpuscles which are glued together in a mass or are adherent to the substance upon which found, have to be freed from it and each other by macerating in certain liquids, and brought back to their natural shape. The liquids employed for remoistening and disintegrating dried clots must be of such nature as to produce the desired effect without doing harm to the corpuscles. If the ques- tion is simply to determine the presence of blood corpuscles, then almost any liquid may be employed, such as water, alcohol, oil, glycerine ; but in order to preserve or restore Comparative Studies of Mammalian Blood. 35 the shape of the corpuscles, various liquids have been sug- gested, which will be enumerated further on. PLATE VI. Fragments of various substances that occasionally may be seen In microscopical preparations and in blood stains : (a, and b.), wool; (c), cotton ; (d.), linen ; («.), silk ; (/".), feather; (g- >, mycelium and spores of a mould fungus ; (A.), mucus; (i. and ;'.), rolled up masses of bacteria and granular debris ; (£.), human hair: (/.), cork ; (m.), vaginal epithelial cell : (».), starch ; (o.), wood splinter. (Magnified 300 diameters.) 36 Comparative Studies of Mammalian Blood. In nature, these liquids are of two kinds : First, such as have the property of dissolving the fibrine which glues the corpuscles togel her ; and, secondly, such liquids which will restore and preserve the shape of the corpuscles. The best liquid for this purpose is a strong (30 per cent.) watery solution of caustic potash, introduced for examina- tion of blood stains by Briicke and Virchow, in 1854, which usually fulfills both requirements, though not always. Another good liquid for remoistering blood is Miiller's fluid, first suggested by Prof. Rudnew, of St. Petersburg. This liquid, I prefer it when mixed with a little (5 per cent.) of glycerine and then diluted by water to the same specific gravity as blood serum (1028), often gives remarkable results. The composition of these liquids, as well as that of some others that may be applied in the remoistering of blood, are as follows : 1. Virchow or MoleschoWs Liquid. Caustic Potash............................. 30 to 33 parts. Water...................................... 70 2. Mutter's Fluid. Bi-chromate of Potassium.................. 2 Sulphate of Sodium........................ 1 Water......................................100 3. Wilberfs Fluid. Bichloride of Mercury...................... 0.5 Chloride of Sodium........................ 2.0 Water......................................100 4. Pacini's Liquid. Water...................................... 300 Glycerine.................................. 100 Chloride of Sodium........................ 2 Bichloride of Mercury..................... 1 5. Ranvier's Liquid. (Iodized Serum.) Potassium Iodide......................... 2 Iodine, sufficient for saturation............ Water...................................... 100 6. Malassez Artificial Serum. Solution of Gum Arabic, sp., gr. 1020. Solution of Chloride of Sodium, sp., gr. 1020. Solution of Sulphate of Sodium, sp., gr. 1020. Of each equal parts. 7. Roussirts Liquid. Glycerine.................................. 3 Sulphuric Acid............................ 1 Water sufficient to make the liquid of spe- cific gravity.............................1028. 8. Robin's Solution is a saturated solution of sulphate of sodium. Comparative Studies of Mammalian Blood. 37 9. Richardson's Salt Solution. Chloride of Sodium................................... 0.75 Water................................................ 100. Having the corpuscles isolated by this liquid, he stains them with a little aniline or iodine. 10. Welcker's Fluid. Glycerine............................................. 1 Water............................................... 7 11. He also used the following solution (artificial serum): Chloride of Sodium.................................... 1 Egg Albumen......................................... 300 Water................................................ 2700 12. Malinirfs Solution. Saturated alcoholic solution of caustic potash (90 per cent, alcohol). Either of these solutions may be used, the author of each claims for his own the best results. It might be well in an important investigation to experiment with several solu- tions, because sometimes one, sometimes another, yields better results. Proper manipulation and some experience in this kind of work are, however, more essential factors in successful preparations than any particular liquid em- ployed. I stated above that Muller's fluid and very strong solu- tions of caustic potash are the two reagents which in my hands gave the best results. In order to obtain the largest possible quantity of unaltered measurable corpuscles from old dried clots and blood stains, I found that the applica- tion of slight heat for several days and of moisture (to pre- vent evaporation) to be of advantage. The procedure I adopted is as follows :—A small granule of the suspected blood or a fibre from the blood-stained fabric is placed on a glass slide in a drop of a 30 to 35 per cent, solution of caustic' potash and covered with a cover glass. If the blood stain was recent, the disintegration of the clot commences at once, and the isolated corpuscles separate and swim swiftly through the liquid if the stage of the microscope is slightly inclined. It is quite interesting to observe how perfectly well-shaped blood discswill tear themselves away from the original formless brown mass. In direct proportion to the age of the stain, from one to within ten days, the softening of the microscopic blood mass and the isolation of the corpuscles is protracted. In 38 Comparative Studies of Mammalian Blood. dried blood older than ten days the ratio of softening or disintegration cannot be well observed, and a stain of two years old behaved like one of ten days. The examination can be made under comparatively low amplification, such as 300 to 500 diameters:but when meas- urements are necessary, then an immersion lense, giving a magnifying power of about 1,000 diameters, better be sub- stituted. Sometimes but a few well-shaped measurable corpuscles are seen, but quite often, in successful preparations from recent blood stains, nine-tenths of the corpuscles in a cer- tain microscopical field will appear quite perfect and fit for measurement. If the blood specimen is slow in disintegrating and the corpuscles imperfect in appearance, then I adopt the fol- lowing procedure :—The glass cover beneath which the blood fragment is mascerating on the reagent is ringed with a little oil, or, still better, with some cement, in order to fasten it and to prevent evaporation, and placed in a moist chamber (a glass vessel, lined with moist paper and cov- ered). The chamber itself put in a water-oven (in *ubator such as used in Bacteria investigation) and subjected to uniform slight heating, not exceeding 100° F., and kept there from one to three days, or as long as is necessary to obtain the desired result, the specimen being examined from time to time. Care must be taken not to over-heat the preparation and guard against evaporation of contents, i.e., of the liquid between the glass slide and glass cover in which the blood specimen mascerates. A number of experiments may be made simultaneously, some of the blood specimens being treated with a strong solution of caustic potash, others by Miiller's fluid, the latter often succeeding in very old blood clots to restore shrivelled and to isolate perfect corpuscles when the former fails. Whereas, the Miiller's fluid with glycerine must be diluted with water in order to obtain the desired specific gravity, a peculiarity in the action of the caustic potash solution must be borne in mind, viz. : that the stronger the solution the better its effect, whereas, weak solutions Comparative Studies of Mammalian Blood. 39 of this reagent (caustic potash) destroy the blood corpuscles or reduce them in size by making them spherical. A strong solution (30 to 35 per cent.)gives most beautiful results : The red blood corpuscles have an absolutely natural appearance ; retain their perfect color and shape or sometimes resume it, if previously lost, form rouleaux, show the normal bi-concavity of the discs, and even show normal diameters on measurement; in short, behave like normal blood. Such is the case when the blood stain was a recent one, and, in fact, the rapid appearance of such good and perfect pictures under the microscope are indi- cations of the recency of the blood stain. Suspected blood in a criminal case, from minute blood clot two days old, caustic pot- ash preparation X 830 diam. Micrometry showed 1-3185 in. as the average diameter of the corpuscles. All blood stains only a few days old that I ever exam- ined in criminal cases, as well as those produced experi- mentally, behaved in the manner described. (See Fig. 15.) When the dried blood or blood stain is more than a week or ten days old, then the reaction is less prompt, 40 Comparative Studies of Mammalian Blood. much fewer corpuscles are perfect and measurable, but still enough to make the result of examination quite satisfactory. The copy of my photo-micrograph, Plate V, Fig. 13, represents red blood corpuscles that I restored from a minute blood clot taken from within the seam of a handker- chief which was claimed by the defendant was stained by ox blood. Measurements showed the average diameter of one hundred of the largest corpuscles to be 1-3185 of an inch, which proved the range of human blood corpuscles. A similar preparation of ox blood corpuscles (Fig. 14) was simultaneously and under similar amplification photo- graphed, and gave a micrometry of 1-4168 of an inch. The difference in size is also quite obvious in the photo- micrographs 13 and 14. In blood stains we meet quite often with a few or with the majority of the red-blood corpuscles shrivelled or con- tracted (crenated) from the effects of drying. (See also phys- ical properties of blood, section I. of this article.) When such corpuscles are remoistened with liquids of less or of the same density as the blood serum (spec, gravity 1028) they only partially and very slowly regain their normal shape. More often, however, they become spherical, and conse- quently diminished in diameter. This is even the case with the oval oviparous blood which we then are able to tell from mammalian only by the size and by the presence of the nuclei. But we further observe that under equal conditions there is a certain definite ratio in the diminution in size of these artificially spherical corpuscles which is the same in all the various animals. It can be easily observed that human corpuscles thus altered appear of the size of the corpuscles of the ox, and that similarly spherical ox blood corpuscles appear reduced to about the diameter of sheep's corpuscles. It was thus necessary to establish what the average sizes of these spherical corpuscles were, whether the ratio of diminution was really uniform in all animals, whether it was constant, and, per consequence, whether it could be applied and relied upon in the diagno- sis of any blood thus altered. Comparative Studies of Mammalian Blood. 41 It was evident that in the micrometry of blood corpuscles in blood stains at least two scales of measurements* for each animal must be established. Further, that the distinction between normal disc-like, bi-concave co: puscles (the larger ones) and corpuscles that had become artificially spherical (and hence the smaller ones) must be rigorously observed in the micrometry of prepared blood of every animal examined. Finally, it was obvious that only strictly disc- like and fully spherical corpuscles should be submitted to measurements, and that any transitional stages in the cor- puscles should be carefully avoided. Although the best occasions for experiments were fur- nished me in the ample material from actual criminal cases (the source of the blood being subsequently con- firmed by either confessions of the criminals or by witness testimony), I made invariably in connection with every case some control experiments and measurements upon blood prepared under known conditions. The combined average results of some of these measurements made by myself, and which bear directly upon these questions, I present in tabular form.f The following table of experiments and measurements explains itself : (See page 295.) Conclusions Regarding Diagnosis of Blood in Its Dried State and in Blood Stains.—We have seen that blood can be diagnosed in its dried state and in blood stains with the same certainty as fresh blood, provided the dry- ing of the blood was rapid and perfect. The blood cor- puscles preserve fully their color, size, shape (bi-concavity) and even their arrangement into rouleaux (only occasionally are such corpuscles a trifle smaller than in fresh blood). But no diagnosis should be made unless the shape of the corpuscle is well taken into consideration, all abnormally small and disfigured corpuscles excluded. At least 500 *Masson, however, thinks that the ''crenated "corpuscles can be more relied upon for uniformity in size than the spherical ones, a proposition with which I do not agree. f Being limited in space for this article, I must, for the present, omit all lesi essential details and a plainer classification of the experiment, COLLECTIVE RESULTS OF SOME OF THE SERIES OF MEASUREMENTS OF RED BLOOD CORPUSCLES IN BLOOD STAINS AND IN EXPERIMENTALLY DRIED BLOOD Normally shaped (bi-concave, disc-like) corpuscles only being measured. I a a OK a. ^ 02 Man..... Man..... Man..... Man..... Man..... Man...... Guinea-pi Wolf .... Dog...... Babbit .., Ox....... Sheep. . . Goat...... P02 & Knife and Glass. Cloth ........ Wood and Linen Paper ............. Knife......... Stone............ Glass............ Glass............ Cloth............ Knife........... Cloth............. Glass............. Knife............ 2 days. 7 days. 10 days. 14 days. 2 years. 6 years. 7 days. days. Condition, or how Prepared. 7 days. 7 days. 7 days. 7 days. Rapidly dried............... Slowly dried............... Slowly dried................ Decomposed from moisture. Well dry preserved.......... Well preserved.............. Rapidly dried stains......... Rapidly dried stains........ Rap'dly dried stains......... Rapidly dried stains......... Rapidly dried stains......... Rapidly dried stains......... Rapidly dried stains......... 3 S O M 03 <£ g^ ( 10 2 4 I 1 1 6 1 4 10 10 3 3 ^ fi a * es 30 10 20 10 10 30 18 50 12 30 30 m g g« . set. Sao *K. O. H. K. O. H. *M. F. M. P. K.O.H.&M.F, K.O.H. &M.F, K.O.H. &M.F. K.O.H.&M.F. K.O.H.&M.F. K.O.H.&M.F. K.O.H.&M.P. K.O.H.&M.F. K.O.H.&M.F. o cS g« 5 to 30 min'ts. 14 hour to 2 dys 2 hrs to 2 dys. 3 days. 2 days. 3 days. 1 to 2 days. 1 to 2 days. 1 to 2 days. 1 to 2 days. 1 to 2 days. 1 to 2 days. 1 to 2 days. Percent, of Meas-urable Corpus-cles in each preparation. <4-l CO <=03 1*2 ■ i NLM001226140