G 000 005 402 3 H-.:'''Uhi \n.:::\]yv.i: ■!..; ■ i J i '■ V 1 ii . ■ -^ ■ ;!-i ; > <; THE LIBRARY OF THE UNIVERSITY OF CALIFORNIA LOS ANGELES GIFT Dr. Oustav F. Ruediger CONTRIBUTIONS TO MEDICAL SCIENCE THE UNIVERSITY OF CHICAGO PRESS CHICAGO. ILLINOIS Bgents THE BAKER & TAYLOR COMPANY NEW TOBK CAMBRIDGE UNIVERSITY PRESS LONDON AND EDINBURGH '/^ .~r^ Therapy (Chicago, 1906). Two pathological reports on cases -!-^ V WJt'OSO vi Introductory Note of blastomycosis and certain notes of a strictly preliminary nature have been omitted also, but references to them are inserted in the proper places. LuDviG Hektoen Preston Kyes E. R. Le Count George H. Weaver H. Gideon Wells Committee TABLE OF CONTENTS PAGE Introductory Note v Howard Taylor Ricketts 3 LuDviG Hektoen. A New Mould-Fungus as the Cause of Three Cases of So- Called Blastomycosis or Oidiomycosis of the Skin . ii H. T. Ricketts. Oidiomycosis (Blastomycosis) of the Skin and Its Fungi i8 H. T. Ricketts. An Organism from Cutaneous Oidiomycosis (Blastomycosis). Remarks on Classification 176 H. T. Ricketts. The Immunological Reactions of Oidiomycosis (Blastomy- cosis) IN THE GuiNEA-PiG 1 79 Benj. F. Davis. Lymphatotoxic Serum: Notes on Its Constitution; Pre- liminary Experiments Bearing on Its Influence on Experimental Infections 231 H. T. Ricketts. Preliminary Report on the Action of Neurotoxic Serum 240 H. T. Ricketts and T. Rothstein. The Reduction of Methylene Blue by Nervous Tissue 245 H. T. Ricketts. Concerning the Possibility of an Antibody for the Tetan- ophile Receptor of Erythrocytes: A Receptor Study 254 H. T. Ricketts. The Adjuvant Action of Serum, Egg-Albumin, and Broth on Tetanus Intoxication 266 H. T. Ricketts A>rD E. J. Kirk. The Study of "Rocky Mountain Spotted Fever" (Tick Fever ?) by Means of Animal Inoculations. A Prelimi- nary Communication 278 H. T. Ricketts. viii Table of Contents PAGE The Transmission of Rocky Mountain Spotted Fever by THE Bite of the Wood-Tick (Dermacentor occidentalis) 288 H. T. RiCKETTS. Further Observations on Rocky Mountain Spotted Fever and Dermacentor Occidentalis 291 H. T. RiCKETTS. Observations on the Virus and Means of Transmission of Rocky Mountain Spotted Fever 299 H. T. RiCKETTS. The R6le of the Wood-Tick (Dermacentor occidentalis) IN Rocky Mountain Spotted Fever, and the Suscepti- bility OF Local Animals to This Disease. A Prelimi- nary Report 312 H. T. RiCKETTS. Further Experiments with the Wood-Tick in Relation to Rocky Mountain Spotted Fever 324 H. T. RiCKETTS. A Summary of Investigations of the Nature and Means of Transmission of Rocky Mountain Spotted Fever . . 333 H. T. RiCKETTS. Studies on Immunity in Rocky Mountain Spotted Fever 343 H. T. RiCKETTS and L. Gomez. A Micro-organism Which Apparently Has a Specific Rela- tionship to Rocky Mountain Spotted Fever. A Prelimi- nary Report 368 H. T. RiCKETTS. Some Aspects of Rocky Mountain Spotted Fever as Shown BY Recent Investigations (The Wesley M. Carpenter Lecture of the New York Academy of Medicine, 1909) . 373 H. T. RiCKETTS. Unfinished Experiments of Dr. Howard T. Ricketts on Rocky Mountain Spotted Fever 409 Benj. F. Davis. Complement Deviation in Rocky Mountain Spotted Fever 419 Benj, F. Davis and William F. Petersen. Table of Contents ix PAGE Time Relationships of the Wood-Tick in the Transmission OF Rocky Mountain Spotted Fever 428 JosiAH J. Moore. Transmission of Spotted Fever by the Tick in Nature 437 Maria B. Maver. Transmission of Spotted Fever by Other than Montana and Idaho Ticks 44° Maria B. Maver. A Contribution to the Pathological Anatomy of Rocky Mountain Spotted Fever 445 E. R. Le Count. The Typhus Fever of Mexico (Tabardillo). Preliminary Observations 451 H. T. Ricketts and Russell M. Wilder. The Transmission of the Typhus Fever of Mexico (Tabar- dillo) BY Means of the Louse (Pediculus vestimenti) . 463 H. T. Ricketts and Russell M. Wilder. The Etiology of the Typhus Fever (Tabardillo) of Mexico City. A Further Preliminary Report .... 473 H. T. Ricketts and Russell M. Wilder. The Relation of Typhus Fever (Tabardillo) to Rocky Moun- tain Spotted Fever 479 H. T. Ricketts and Russell M. Wilder. Further Investigations Regarding the Etiology of Tabar- dillo, Mexican Typhus Fever 491 H. T. Ricketts and Russell M. Wilder. HOWARD TAYLOR RICKETTS Vhrau KiC 7 /111 HOWARD TAYLOR RICKETTS.^ LuDviG Hektoen. Dr. Ricketts was born in Findlay, Ohio, February g, 187 1, passed his boyhood in Nebraska, and received the Bachelor's degree from the university of that state in 1894. He took his medical course at the Northwestern University Medical School, graduating in 1897, and then served as interne in the Cook County Hospital in Chicago. In 1899 he became fellow in cutaneous pathology in Rush Medical College and continued as such for two years. During part of this time he worked also in the dermato- logical clinic. In 1900 he married Myra Tubbs who supported him in his work with rare devotion, keen interest, and steady encouragement. There are two children, son and daughter. In 1902, returning from a year's visit to European laboratories, he came to the University of Chicago as instructor in the newly founded Department of Pathology and Bacteriology. Later he became assistant professor and in the early part of 19 10 he accepted the chair of pathology in the University of Pennsylvania, the duties of which he fully expected to assume in the fall. He died from typhus fever in Mexico City, May 3, 1910. Dr. Ricketts was a modest and unassuming man, of great determination and of the highest character, loyal and generous, earnest and genuine in all his doings — a personahty of unusual and winning charm. His associates of the hospital and fellowship days, who knew him well, knew his abihty and energy, his distinct fondness for the day's work, all looked to him for the more than ordinary achievement. He deUberately turned away from the allurements of active medical practice and decided to devote himself to teaching and investigation in pathology. He had early become possessed of noble ideals and had a pure love for the search after truth in his chosen field, which abided with him and gave him a high concep- tion of all his duties and relations and placed a special stamp on > An address delivered at a memorial service in the University of Chicago, May 15, 1910, and now expanded to include a few more details. 3 4 Contributions to Medical Science his work. His instinct for research at no time was permitted to lie dormant and unused, but growing stronger it carried him on farther and farther, and in due time the University freely and in special ways promoted the work in which he was to accomplish such large results. The torch was placed within the grasp of hands fit to carry it forward, and during the few short years given him he advanced it farther than we may realize at this moment, because he broke open paths for future progress. His earher researches on blastomycosis and immunological problems are all marked by thoroughness and directness, by clear and forceful reasoning; it is in the brilliant work on Rocky Moun- tain fever, however, that Dr. Ricketts fully reveals himself as an investigator of the first rank. He took up the study of this fever in the spring of 1906 as a sort of pastime during an enforced hoHday on account of overwork. The disease is a remarkable one; it occurs in well-defined areas in the Mountains, is sharply limited to the spring months, varies greatly in severity, the mortality in one place being about 5, in another between 80 and 90, per cent. For some time it had been regarded as caused in some way by the bite of a tick. Dr. Ricketts promptly found that the disease is communicable to lower animals and that a certain tick, which occurs naturally on a large number of animals in those regions, by its bite can transmit the disease from the sick to the healthy animal. These observations opened a new field, and henceforth he devoted himself untiringly to the investigation of the many prob- lems that arose one after another as the work went on, both in the laboratory here and in the field. As we follow the various stages in the progress of this intensely active work it becomes very clear that Dr. Ricketts not only was gifted with imaginative power so that he could see and trace the various fines along which the solution of a problem might be sought, but that he also possessed in a full measure the capacity for that hard, accurate, patient work necessary for the more dijficult task of finding the one, true solution. This combination of speculative abihty and the power to do steady toil and even drudgery, often under great difficulties, made him a great investigator and brought him success. Howard Taylor Ricketts S Some of the experiments devised to lay bare the secrets of the different orders of living things concerned in spotted fever are masterful in their ingenuity and comprehensiveness, notably those bearing on the hereditary transmission of spotted fever virus in ticks, on the occurrence of mfected ticks in nature, and on the part played by small wild animals like the squirrel as source for the virus. Having solved many hard questions he came to the conclusion that in man spotted fever depends simply on the accidental bite by an adult tick carrying active virus. As only adult ticks find their way to man and as they occur only in the spring, the pecuHar seasonal prevalence of the disease is nicely explained. It is almost unnecessary to point out that the work furnishes clear and direct indications as to what to do in order to prevent the disease. Finally last year, he discovered what seems to be the immediate cause of spotted fever, namely a small bacillus, which he found in the blood of patients and in ticks and their eggs. Strains of this bacillus present in ticks from different places vary greatly in virulence, and this fact may explain why spotted fever varies so greatly in severity. Many of the observations and discoveries in connection with this work have a much wider significance, and will surely prove of value and service on the ever-shifting battleground with infectious diseases. Rocky Mountain spotted fever in many ways resembles typhus fever. As he was completing his three years' study of the Rocky Mountain disease, having determined its mode of transmission, its cause, and a rational method for its prevention. Dr. Ricketts became more and more strongly impressed with the thought, which he had had for some time, that the special knowledge and training thus acquired would prove of great value in the study of typhus fever and thereby perhaps be put to the best use. This idea met with encouragement, and in July of last year it was definitely decided to take up the study of typhus fever in the City of Mexico, that being the nearest place, so far as known, where any such work could be done. I speak of this date because I wish to make it clear that Dr. Ricketts reached his decision before, and entirely independently of, the establishment by the Mexican government 6 Contributions to Medical Science of certain prizes for successful investigation of the typhus fever of Mexico (Tabardillo). Typhus fever has been one of the great epidemic diseases of the world. Its devastations are recorded ''on the dark pages of history, the pages that tell of war, overcrowding, want, and misery." Until the middle of the last century it prevailed in practically all large European cities; now it has largely disappeared, owing, it is believed, to better sanitary conditions; but it is still smoldering in many centers, and in some places, as in Mexico, typhus in one of its forms now claims hundreds of victims each year. When it assumes its most virulent forms typhus fever may become one of the most contagious of diseases, and there is no disease that has claimed so many victims among physicians and nurses. It is stated that in a period of twenty-five years, of 1,230 physicians attached to institutions in Ireland 550 succumbed to typhus. Of the six American scientists who have studied the typhus fever of Mexico since December last, three have been stricken and two have died — Conneff of the Ohio State University expedition and our Ricketts. It is when the sick are aggregated in hospital wards that the danger of infection is especially great. Until very recently nothing was known as to the cause of typhus fever and the exact mode of its transmission. Fully acquainted as a matter of course with all the character- istics of the disease, Dr. Ricketts and his volunteer assistant, Mr. Russell M. Wilder, began their work in December last. Before many weeks had passed results of great importance were secured; it was found that tophus is different from the Rocky Mountain fever, although they have many points in common; that the Mexican t>'phus is communicable to the monkey; and that it may be transmitted by an insect (Pediculus vestamenti). Some of these results are confirmatory of very recent results obtained by others, but on April 23 they were able to announce the discovery of a micro-organism, apparently a bacillus, in the blood of typhus patients and in the insect. It seems reasonable to believe that this organism may be the cause of typhus fever. While courageously and devotedly pushing this and other work on to completion Dr. Ricketts was stricken with typhus, and the Howard Taylor Ricketts 7 unfinished investigations of such fundamental importance must be taken up by others. Thus a young and noble career of great achievement and of large service to humanity came to a sudden and heroic end, and a new name was placed on the martyr roll of science. Those near to him know that he fully understood the dangers to which he would be exposed and the risks he would run. He decided he would take those risks, meet the dangers with all pos- sible means of prevention, and do the work that would come to his hands. And so he made the great sacrifice and gave all that a man can give for his fellow-men. CONTRIBUTIONS TO MEDICAL SCIENCE BY HOWARD TAYLOR RICKETTS A NEW MOULD-FUNGUS AS THE CAUSE OF THREE CASES OF SO-CALLED BLASTOMYCOSIS OR OIDIOMYCOSIS OF THE SKIN/ H. T. RiCKETTS, Fellow in Cutaneous Pathology in Rush Medical College. (From the Pathological Laboratory of Rush Medical College.) The following cases of oidiomycosis (blastomycetic dermatitis, Gilchrist) of the skin have been studied recently, and a mould- fungus isolated from each. Case I. — A German, 73 years old, a laborer, and of negative family history, came to my dispensary clinic at Rush Medical College in October, 1900. In February, 1898, a pimple appeared on the scalp above the lobe of the left ear. It became a pustule which crusted, increased in size, and gradually presented a verrucous surface. Extension occurred with some rapidity, the oldest portions heahng. When he was first seen nearly the whole left side of the scalp and temporal region, including the adjacent skin of the ear, had been covered by the process. Six months before I saw him another pimple appeared in the left malar skin, which now measured two inches in diameter. The center appeared flat, depressed, crusted, and bled easily. The peripheral portion was elevated from a half to one centimeter above the skin level, and was covered by very large papillary growths. The surrounding skin of both lesions was infiltrated, reddened, and beset with many minute abscesses. Healing had occurred over a large part of the left scalp, leaving an active verrucous periphery. There is no history of venereal disease. Verrucous tissue, and pus from the small abscesses, mounted in potassium hydrate solution, uniformly show an organism existing singly, in pairs, or in small groups, the members of which are capsulated, possess a granular protoplasm which often contains many refractive spherules and occasionally vacuoles; they pro- liferate by budding. An adventitious capsule is found surrounding 'A preliminary report, Jour. Boston Soc. Med. Sci., igoi, 5, p. 453- II 12 Contributions to Medical Science many cells. This examination repeated many times gives similar findings. Tubes of ordinary media were inoculated several times from the verrucous tissue and from the abscesses with no growth except that of cocci and bacilli. Eventually, however, a maltose agar, prepared after the formula of Sabouraud for the ringworm fungi, was inoculated with pus from the abscesses in the reddened areola, the skin being first cleansed with alcohol. After six days small hyphal colonies were noted on all of the tubes. They increased in size slowly, the hyphae reached the surface of the tubes, and the substratum presented a moulded appearance. Eventually aerial hyphae (Plate i, Fig. i) covered the whole inner surface of the tube. Under the microscope this aerial structure consists of fine branching threads beset with lateral pedunculated conidia which multiply in situ by a budding process. That portion of the growth which infiltrates the substratum is composed of similar hyphae, which, however, are larger and plainly segmented and produce lateral, sessile, or pedunculated conidia and unicellular offshoots. The reverse surface of agar slants exhibits a rich, even, golden- brown color, hke that of a well-colored meerschaum. Growth occurs slowly but steadily on potato, blood-serum, gelatin, and in various bouillons. In the last-mentioned medium a coherent, fluffy mass forms, the supernatant fluid remaining clear. The organism does not ferment saccharine solutions. ProHferation occurs by the segmentation of hyphae and the ab junction of the resulting oval cells; by the formation of terminal and lateral spore groups or conidia, and by the budding of ascus- like cells. A nvunber of subcutaneous and intraperitoneal inoculations into animals were without result. However, a dog which received an intravenous inoculation died in one month from a mycosis of the entire lung tissue. Cultures of the organism were obtained from the nodules comprising the consohdation. Microscopically, the organisms as they existed in the fresh lung consisted of spherical and budding cells which usually were more or less filled with small structureless refractive spherules. Similar bodies were found free, and it is thought that transition forms have been observed A New Mould-Fungus Causing Oidiomycosis of Skin 13 between these and the adult capsulated organism. There were no hyphae in the lung tissue. A portion of the nodule containing spherical organisms was inoculated into a hanging drop of bouillon; in a few hours the spherical cells sprouted hyphae, and within twenty-four hours only a few spherical cells were found, the remainder having devel- oped one or more hyphal elements (Plate i, Fig. 2), In cultures, also, similar spherical cells filled with the refractive bodies spoken of have been found repeatedly. Histologically, the human tissue presents two or three interest- ing features. The verrucous papillae are remarkably large, and many give off secondary branches. There is an unusual amount of abnormal cornification throughout the epidermis. The number of miliary abscesses is large, and it has been possible to trace the formation of multinuclear giant cells from the cells of the stratum mucosum. A unique feature in the cell infiltrate of the corium is the strikingly large number of eosinophiles. These are found also in the intra-epithelial abscesses and in interepithehal spaces. Tubercle bacilli could not be found. Case II. — A PoHsh woman, of good parentage, and with no tuberculous or venereal history. The disease began about eighteen months ago as a pustule, and has covered the right side of the face, the bridge of the nose, and has extended moderately on to the supra-orbital skin. When first seen by Dr. Montgomery, to whom I am indebted for the privilege of studying the case, healing had occurred for the most part, except at the periphery, where there was still a verrucous structure. Located here and there in the periphery and in the central scar tissue were a number of small projecting tubercles, which at first were considered pustules, but on being incised were found to consist of a cellular or mucoid substance. Cultures were at first taken from the verrucous tissue, but the contamination was so great that the tubes were abandoned. At this time an examination of tissue in potassium hydrate solutions showed the presence of numerous ''blastomycetes." Subse- quently, several of the myxoma-like nodules which were covered with smooth skin were cleansed with green soap and alcohol, the 14 Contributions to Medical Science overlying epidermis incised, the soft contents scraped out and inoculated on various media. It was demonstrated at this time that these nodules contained organisms. Two weeks later the development of a mould-fungus was noted in some of the tubes. Growth proceeded slowly, and eventually the organism in its gross and microscopic aspects appears identical with the organism of the preceding case. Animal experiments have not yet been made. The tissue removed for histological study included a myxoma- like nodule and a small amount of verrucous tissue. The organisms are found in small groups in the corium. There are a number of tubercles resembling closely those of tuberculosis. Tubercle bacilli cannot be demonstrated. As in the preceding case, eosinophiles are very conspicuous as a part of the cell infiltrate. Case III. — (From Professor Senn's surgical clinic. I wish to thank Professor Senn for permission to use this material.) A German woman, 33 years old, giving no personal or family history of tuberculosis, venereal disease, or carcinoma. Two years ago three pustules appeared simultaneously; one on the left cheek, the second on the left wrist, and the third on the right buttock. They increased in size, became verrucous, and the last two disappeared spontaneously after about three months. That on the face progressed, and eventually measured about two inches in diameter; presented a depressed center, and an elevated periphery, composed of very coarse papillary processes; the surrounding areola contained many minute abscesses. The diseased tissue was removed by Dr. Graham, and through accident was immersed in Zenker's fluid, for thirty or forty seconds. It was immediately put into running water, and washed for two hours. It was feared that cultures might not be obtained because of this treatment. However, a portion of the verrucous tissue was thoroughly disintegrated in a sterile mortar, the resulting pulp suspended in bouillon, and inoculations made on all available culture media and agar plates. After four days rosettes of hvphae were seen to have grown out from a number of small fragments of tissue which had been transplanted. Growth proceeded as in A New Mould-Fungus Causing Oidiomycosis of Skin 15 the two preceding cases, and eventually the same mouldy surface- appearance resulted. Microscopically, this organism differs in no way from the two preceding. A portion of the tissue was inoculated subcutaneously in the inguinal region of a guinea-pig. After a week a nodule developed, which doubled in size at the end of the second week, and was found to contain cheesy pus which on cultures yielded the organism isolated from the human tissue. Spherical capsulated fungus cells were found microscopically; no tubercle bacilli could be demonstrated. A portion of this pus mounted in hanging drop exhibited a growth of hyphae from the spherical cells, many of which contained the refractive bodies alluded to above. Aside from the usual features noted in " blastomycetic derma- titis," the presence of large numbers of infiltrating eosinophils characterizes this case as it does the two preceding. No tubercle bacilH could be demonstrated in the human tissue. The organisms so far described in connection with this disease have been of the blastomyces and oidium types. Hence it is somewhat startling that from each of three successive cases mould- fungi were cultivated which appear identical. The question must soon be decided as to whether these varying organisms obtained from a constant cHnical entity are to be con- sidered of common generic or specific dignity. In considering this subject recourse must be had to the observations constantly made by botanists that the members of a particular genus or even of a particular species are subject to variations in morphology and proHferation which may often be remarkable. It is to be noted, also, that many botanists consider blastomyces-like forms and oidium-like organisms as the conidial stages of somewhat higher fungi. In view of these facts and in view of the fact that the various organisms exist as the cause of a constant clinical entity, the conclusion is reasonable that the fungi are very closely related organisms. This is substantiated by the observation that the higher types (that is, the mould-fungi) possess all the morpho- logical and proUferative possibihties of the oidium-like and blasto- myces-like forms. The conclusion, then, seems reasonable that the 1 6 Contributions to Medical Science latter are adaptation forms or conidial stages of the mould-fungi which have been considered in this communication, aerial fructi- fication having been suppressed in the simpler forms. It is important to note that Ophuls and Moffitt, in June, 1900, pubHshed an account of a mould-fungus cultivated from the so- called protozoic disease first described by Wernicke and studied in detail later by Rixford and Gilchrist. All previous writers had considered the organism a protozoon. The mould-fungus of Ophuls and Moffitt, however, was found in all accessible lesions and a disease was produced in animals similar to that in the patient, and the study of the organism from the lesions in hanging drop cultures proved their identity with the mould-fungus obtained on solid agar. In tissues the organism existed as a spherical capsulated cell, filled with refractive globules Hke those existing in the organ- isms isolated from the three cases presented herewith. A study of the facts at hand points toward the identity of the mould-fungus of Ophuls and Moffitt with those cultivated from the three cases of oidiomycosis of the skin under consideration. conclusions. 1. A group of closely related organisms, which, however, show rather constant differences, exists as the cause of blastomycetic dermatitis (Gilchrist). 2. The simpler organisms probably are adaptation forms or conidial stages of the more complex mould-fungus, possessing, however, fixed specific characteristics. 3. The protozoic (?) disease of Posadas, Wernicke, Rixford, and Gilchrist, and others may be a general disease of which the so-called blastomycetic dermatitis is a local manifestation. 4. Those cases of oidiomycosis of the skin in which a mould- fungus exists as the cause seem to be characterized by an eosinoph- ilous cell-infiltrate. (Since the above was written, cultures have been made from two additional cases of "blastomycetic" dermatitis, and a mould- fungus, appearing identical with the three referred to, isolated from each.) PLATE 1 Fig. 1. ^rm^is^mm^ ■y^-^_ -^"X. / '^ Fig. 2. A New Mould-Fungus Causing Oidiomycosis of Skin 17 EXPLANATION OF PLATE i. Fig. I. — Showing aerial hyphae with lateral conidia, and terminal pseudo- ascospores (?). (Sketched from a culture tube under a low power.) Fig. 2. — The development of hyphae from spherical capsulated spore-containing (?) cells. Fluid from a mycotic pulmonary nodule (dog) containing the organisms was mounted in a bouillon hanging drop and the growth observed. The nature of the enclosed spherules is uncertain. A, B, C, and D illustrate successive stages of hyphal growth; E and F, the later formation of pedunculated conidia. The larger spherical cells were 10-13 niicrons in diameter; the conidia 3-4 microns. OIDIOMYCOSIS (BLASTOMYCOSIS) OF THE SKIN AND ITS FUNGI/ Howard T. Ricketts, Fellow in Cutaneous Pathology in Rush Medical College. From the Dermatological Clinic and the Pathological Laboratory of Rush Medical College in affiliation with the University of Chicago. ) CONTENTS. Introduction I. History and Pathogenic Properties of Oidia Blastomycetes and Saccharomycetes in Human and Animal Lesions Other than Blastomycetic Dermatitis .... Experimental Blastomycelial and Similar Infections in Animals Blastomycetic Theory of Tumors ...... Sporothrix Schenckii ........ II. Summary of So-called Blastomycetic Infections in Man 1. Saccharomycosis hominis, Busse 2. Saccharomycosis humaine, Curtis 3. Blastomycetic dermatitis, Gilchrist 4. The Protozoic (?) Infection of Wernicke and Others 5. Summary ....... III. Clinical, Mycological, and Histological Study of Twelve New Cases of Oidiomycosis of the Skin Case I Case II Case III Case IV Case V Case VI Case VII Case VIII Case IX Case X Case XI Case XII IV. Analysis of Blastomycetic Infection of the Sb:in Clinical Data ....... Diagnosis ....... Prognosis ........ Treatment ....... PAGE 19 19 23 26 29 30 30 30 35 36 SI S6 56 S6 72 77 87 95 96 lOI no 112 116 116 1x8 119 119 126 128 129 • From Jour. Med. Research, 1901, 6, p. 374. Path. Soc, 1899-1901, 4, p. 348. A preliminary summary is published in Trans. Chic. Oidiomycosis of the Skin and Its Fungi 19 Pathogenesis Pathological Anatomy Serum Reactions . Special Histological Features Comparative Study of Organisms, Classification and Nomenclature General Summary Conclusions V. Bibliography .... VI. Explanation of Plates . with Table PAGE 130 132 134 ^35 144 159 162 167 168 173 INTRODUCTION. Since attention was called to the existence of blastomycetic dermatitis of Gilchrist in Chicago and vicinity the number of cases observed has rapidly multiplied. I am greatly indebted to Pro- fessor Hyde and Professor F. H. Montgomery for the opportunity to study the clinical features of a number of cases, to secure material for investigation of the fungi concerned and of the histological changes, and for their many courtesies. For co-operation and advice I wish to thank Professor Hektoen, in whose laboratory the work has been done. The organisms, which I have isolated from cases of blasto- mycetic dermatitis, seem to fall into three groups which have so many biological and morphological features in common that they probably represent different species of a common genus. They all belong to the type of fungi that Link designated as Oidium. This being the case, I have considered it advisable to briefly review the Hterature in regard to oidia, more particularly the pathogenic forms, before taking up the study of the so-called blastomycetic infections proper. I. HISTORY AND PATHOGENIC PROPERTIES OF OIDIA. Link gave the name Oidium to a genus of the fungi, having an elementary organization, many species living as moulds. They have more or less ramifying and segmented filaments, which produce at their extremities short cells, very often rounded. These cells, which are united end to end, rosary-like, detach themselves when mature. These are the spores, which germinate and produce filaments like those from which they came. 20 Contributions to Medical Science Of the oidia, the Oidium albicans was the first of medical im- portance, and began to occupy the attention of writers in 1840. We learn from Robin that Berg first saw the filaments and spores of this oidium under the microscope and recognized the disease it produced in the mouth, pharynx, and esophagus, the stomach and rectum. Gruby, in 1842 and 1844, published reports of his investi- gations, in which he considered the parasite of thrush analogous to Sporotrichum. He used the name ''Species Sporotrichum afl&nis, Gruby." There followed a discussion between Gruby and Berg as to the nature of the organism. Oesterlen's findings were similar to those of Gruby. Vogel's work was in the nature of a review; he offered the opinion that the fungus did not grow upon the intact mucous membrane, but upon the exudation into which the latter is converted when it begins to decompose. Between the years 1841 and 1847 other writers, including Eschericht, Hannover, Hoenerkopf, Baum, Slayter, Gubler, and Remak, described thrush clinically or from a mycological standpoint. Robin followed with his classical works in 1847 ^-^^ i^SS- He observed that the spores again grow into the tubular form of the organism, a matter not understood previously and in reahty not thoroughly established until Grawitz in 1880 pubHshed an account of his investigations. Since Robin's time much has been written concerning the botanical position of this fungus, and the non-pathogenic Oidium lactis. Instances of morbid processes caused by the former, aside from thrush, have been reported from time to time. In 1848 Bernard expressed the conviction that yeast-like organisms found their way to the gall-bladder. Reubold in 1854 confirmed what had been observed in regard to the effects of Oidium albicans upon the epithelium of the mouth, pharynx, esophagus, and vocal chords. He also found the fungus in the stomach, intestines, and respiratory passages, where it had not penetrated the epithelium. Virchow found the submucous tissue of the esophagus invaded by the oidium. In i860 v. Zenker found multiple brain abscesses in a man affected with hemiplegia, and a coexistent "Soor" of the throat. " Soor-Massen " were found in the abscesses. (Baumgarten holds that the cerebral abscesses must have been due to a mixed infection with pyogenic bacteria.) Oidiomycosis of the Skin and Its Fungi 21 In 1869 Parrott described a nodule of Oidium albicans in the lungs and a peritoneal localization. E. Wagner (1868) found oidium threads in the connective tissue of the esophagus and in neighboring blood-vessels. A. Vogel found similar conditions. Grohe, in 1870, observed mycelial abscesses in the Hver. Slawjansky found Oidium albicans in a spongy, pea-sized nodule in the lung of a patient who had died of pneumonia. Ribbert (1879) described multiple brain abscesses in a twelve-day-old child dying with thrush of the tonsils, pharynx, esophagus, and vocal chords. Mycelium was found in one abscess. There were no micrococci. Parrott found oidia penetrating the tissue of the esophagus and stomach. Birch-Hirschfeld found Oidium albicans in the lungs of a four-day-old boy, and Rosenheim in a case of fetid bronchitis. Heller, in 1888, found oidium threads penetrating the connective tissue in an esophageal ulcer. Similar conditions existed in a case of laryngeal thrush. Schmorl, in 1890, obtained an oidium from multiple abscesses of the kidneys. The patient had died of typhoid fever and had extensive thrush. In five infants dying of gastro- enteritis, Schmidt found thrush deposits in the walls of the esopha- gus, larynx, trachea, and larger bronchi, and once in the lung tissue. The conidial form predominated. They penetrated the submucosa, entering the blood-vessels, and in two cases the mus- cular layer was reached. Ross (1891) found Oidium albicans in cyst-hke lesions of the lungs. Giulini reported a case of ''Soor der Vulva" in a twenty-four-year-old woman, three months pregnant. A few easily detached islands first appeared, and in a few days a uniform membrane covered the vulva. Guide described a lung abscess in a boy three years old, caused by Oidium albicans. Sud- den hemorrhage resulted. Preyhan (1891) discovered spores of the aphtha fungus in the sputum of a patient suffering from pneu- monia and hemorrhagic pleuritis. Brandenberg (1893) found Oidium albicans in two cases of angina, and Charrin in a submaxil- lary abscess. Langerhans (1894), at the autopsy of an eighty- three-year-old woman, observed round, whitish masses filling the uterus. They contained mycelium resembling that of Oidium albicans. No cultures were made. Grassett (1893) found gingival abscess, supposedly caused by the thrush parasite, but staphylo- 22 Contributions to Medical Science cocci were present in large numbers. Heller (1895) examined post mortem the mucous membranes of twenty-five cases affected with thrush. In some cases only the epithelium, in others the underlying tissue was invaded, thromboses being found in six cases. In nine cases ulcers were present. In one case large masses were found in the bronchi, penetrating the blood-vessels. He proved experimentally that these processes were not due to post- mortem invasions. V. Herflf (1895) ^^ twenty-six mycotic infec- tions of the vagina found Oidium albicans in sixteen, Monilia Candida in four, Leptothrix vaginalis in one, and another oidium- like fungus in one. Galli-Valerio discovered a variety of Oidium albicans in the discharges of a child dying from gastro-enteritis. It formed threads and chlamydospores, and produced fermentation with an intense odor. He concluded that oidia may be a cause of gastro-enteritis in children. The organism differed in cultural properties and virulence from t}^ical Oidium albicans and Oidium lactis. Giuseppe Cao, considering many of the blastomycetes so far described to be oidia, experimented with forty-one varieties isolated from sputum, feces, fruit-juice, etc., and found many of them pathogenic to animals in varying degrees. He divided them into four groups, depending on their cultural and pathogenic properties. The cultural and pathogenic properties of the Oidium albicans and other closely allied oidia have been studied by Grawitz, Klem- perer, and many others. Grawitz contended that it was not an oidium, but a budding fungus and nearly related or identical with Mycoderma vini. Plaut considered it identical with MoniKa Candida (Bonorden), a widespread lower organism belonging to the Torulaceae. Others have considered it a "blastomyces," or saccharomyces (Hansen, Jorgensen). It has been found to produce general mycosis in rabbits, guinea-pigs, and mice, but is less pathogenic to dogs. Roger and Charrin, and Otrowsky were able to produce a high degree of immunity in animals. The cultural properties have been found to be fairly uniform, but certain varia- tions in morphology and biologic properties lead to the conclusion that there are different varieties or adaptation forms. Oidiomycosis of the Skin and Its Fungi 23 blastomycetes and saccharomycetes in human and animal lesions other than blastomycetic dermatitis. Organisms somewhat resembling oidium have been found in various conditions in man, other than " blastomycetic " dermatitis and tumors; and called Blastomycetes or Saccharomycetes. Oudemans and Pekelharing obtained from Pityriasis capitis an organism which they called Saccharomyces capilitii. They claimed to have reproduced the disease in rabbits. Buist cultivated a blastomyces from the lesions of variola and vaccinia. Babes found blastomycetes in certain ulcers of the skin. Colpe described an oval yeast as the only organism present in a chronic catarrh of the endometrium. Busse found yeast-like bodies in a long-standing pyosalpinx; also in the products of a three-months' abortion. Cultures were sterile in both cases. De Simoni reports that in five cases of hypertrophic tonsils he found blastomyces-like bodies in fresh preparations and stained tissue. They were not present in normal tonsils, but cultures were not made. It is not clear that these bodies were not the result of degenerative tissue changes. Corselli and Frisco's case of supposed sarcoma of the mesenteric lymph glands and peritoneum, with metastases in the thorax, seems rather to have been an inflammatory process, caused by a blastomyces. The organism produced similar conditions in animals. Secchi in the stained sections of two cases of keloid acne found blastomyces-like bodies. De Stoecklin examined five hundred cultures from suspected diphtheritic throats. In thirty-seven were thrush-like fungi, which he considered blastomycetes. Nineteen of these cases also yielded LoefB.er's bacillus. He concluded that the fungus intensified the pathogenic effect of the diphtheria bacillus, and that the fungus alone could cause a severe angina. Troisier and Achalme found in a case of thrush an organism which they considered a yeast (a saccharomyces). It differed culturally from the Oidium albicans. Demme (1891) describes a red budding fungus (Saccharomyces ruber) as a contamination of milk. Seven children who drank infected milk suffered from gastritis. The organism was not pathogenic for guinea-pigs and dogs when injected subcutaneously, but when infected milk was fed, gastro- 24 Contributions to Medical Science enteritis resulted. In a pseudo-membrane of secondary syphilis of the throat Teissier found a pure culture of a yeast resembling the "levure du muguet." He concluded that the syphilis favored the localization of the yeast infection. Hannover (1842) found certain torulae in fermenting diabetic urine, and De Gaetano the Saccharomyces septicus in urinary sediment. Busse obtained cultures of yeast fungi from nasal polypi and found similar bodies in the tissues. Owing to contaminations, he was unwilKng to consider the organisms the cause of the polypi. Buschke (1898) obtained from a vaginal secretion a yeast which produced a cutaneous blastomycosis in animals. It differed from the organism described by Curtis and from that isolated by Maffucci from a skin disease in a guinea-pig. Brazzola (1896) described a case of general infection of a child by a saccharomyces. The patient had died supposedly of diphtheria, but the author found a pure culture of a saccharomyces in the throat. The organism resembled that cultivated by Busse from a case of pyemia, and caused general infection in animals. Buschke also isolated blastomycetes from five out of six cases of seborrhea. One organism was pathogenic to animals. Behla (1897) cultivated blastomycetes from the so-called Miescher's tubules. Previous attempts had failed. Alexander G. R. Foulerton inoculated animals with various yeasts, including one — Saccharomyces tumefaciens albus — which he isolated from two cases of pharyngitis. Death often resulted in from a few days to three or more weeks, and granulomatous swelhngs often developed at the site of inoculation, from which the yeasts were again cultivated. Stoewer obtained from three cases of keratitis pure cultures of a red yeast, which proved pathogenic to rabbits' eyes when injected. He also produced an inflammatory condition in rabbits' eyes with Curtis' organism. Lundsgaard obtained a pure culture of a blastomyces from a case of hypopyon. Inocula- tions caused haziness in the eyes of rabbits. Busse 's case of blastomycetic septicemia and Curtis' of myxoma-like tumors caused by a blastomyces are considered later. Organisms similar to the above have been found in diseases of animals. Oidiomycosis of the Skin and Its Fungi 25 Monospora bicuspidata, a blastomyces, was recognized by Metschnikoff in a disease of the daphnia. Its morphology and life history were followed out histologically, cultures not being obtained. It was oblong, often tube-like in shape, and possessed a needle-shaped spore lying in the long axis of the cell. The organ- isms are ingested, the capsule dissolved, the spores penetrate the intestinal wall and enter the circulation. Lateral conidia sprout, and these by a budding process produce adult cells, which again form spores. Pfeifer found torula forms in bovine vaccine lymph. Lymphangi- tis epizootica (Lymfangite epizootica, Linfangite farcinoide, in Italy; Farcin de riviere, Farcin d'Afrique, in France), affecting horses especially, was long supposed to be of parasitic origin. Rivolta called certain highly refractive bodies, which were present constantly in the pus, " Cryptococcus farcinimosus Rivoltae." CanaHs considered them coccidia, and Plana and Galli-Valerio, sporozoa. The disease affects the skin and upper respiratory passages especially. In the skin, nodules and ulcers form; glandular metastases occur, which suppurate, and sinuses form in the subcutaneous and deeper muscular tissue. Recovery usually follows in a few months, although in some cases general infection, with death, occurs. Earlier attempts at cultures failed, although inoculations with the secretions have been carried out successfully by Rivolta and Micellone, Bassi, Chernier, Chravat, Delamotte, Penpion, and Boinet. Aruch, in 1892, was not able to obtain cultures, but in 1895 Ferni and Aruch cultivated a blastomyces from the abscesses. In unstained preparations of the pus, budding forms and empty capsules were found. After three days colonies grew on potato. The cells were rounded, oval, or apiculate, and buds formed at the ends. Hyphae are not mentioned. Suppurating nodules were produced in the testicles of rabbits by injecting pure cultures, and the organism was reclaimed from the resulting pus. Tokishige in Japan and Tartakowsky in Russia obtained similar results. Tokishige's organism, however, differed markedly from that of Fermi and Aruch. From thirty to fifty days were required for development on artificial media, and in time the surface became folded like coils of intestines. Microscopically, hyphae and yeast- 26 Contributions to Medical Science like cells occurred together. Animal experiments gave positive results. Similar epidemics have been observed in Guadaloupe, France, Algiers, and Norway. Ferran, in 1889, isolated from the brain of mad dogs a blastomy- ces which he supposed was the cause of the mania. Memmo also found an organism of this type in mad animals. Maffucci and Sirleo (1894) gave the name of Saccharomyces niger to a blastomy- ces cultivated from the tissues of a guinea-pig which died in a marasmic condition. The animal had been inoculated with a portion of the Uver of an embryo guinea-pig, the mother of which had died of tuberculosis. The lungs and l>Tnph glands contained myxoma-like nodules, and similar nodules were found microscopi- cally in the kidneys. They contained large numbers of the organism. There was great hyperplasia of the pulmonary epithehum, which was at first considered cancerous. This position was, however, abandoned later. Corselh and Frisco (1895) found tumors in the lymph glands of a cachectic guinea-pig, with metastases in the inner organs, and isolated a blastomyces. SanfeHce (1895) designated as ''Saccharomyces Hthogenes" a blastomyces which he cultivated from a carcinomatous metastasis in an ox, the primary tumor occurring in the liver. It degenerated into calcareous-Kke masses in animal tissue. Also, in 1897, SanfeUce pubHshed experi- ments to show that the so-called " Taubenpocken " is caused by a blastomyces. In 1898 Maffucci and Sirleo again cultivated from a guinea-pig a blastomyces which corresponded to their first one from a similar source. There were the same miliary nodes. San- feHce (1898) called an organism ''Saccharomyces granulomatosus," which he obtained from a granulomatous nodule of a pig. Inocu- lation in swine produced similar lesions, but the parasite was not pathogenic to other animals. EXPERIMENTAL BLASTOMYCELIAL AND SIMILAR INFECTIONS IN ANIMALS. Since 1848 infections of animals with yeasts have been produced experimentally, various manifestations resulting. In that year Claude Bernard inoculated dogs intravenously with beer yeasts to determine the disposal of sugar. The animals died. Grohe Oidiomycosis of the Skin and Its Fungi 27 (1869) inoculated yeasts into the blood and peritoneal cavity. Rapidly fatal mycosis resulted, nodules containing the blastomy- cetes being found in various organs. Popoflf (1872) inoculated dogs with impure yeasts and obtained septicemic and typhoid conditions; in the viscera were miliary nodules, in which, however, blastomycetes or myceHum were not demonstrated. Falk (1896) failed in similar experiments. He considered Popofif's results due to other and bacterial organisms. However, Falk may have used non-pathogenic yeasts. Neumayer (1891) found that animals could be fed Hberally with yeasts if no fermentable substance were introduced at the same time; otherwise gastro-enteritis developed. No pathological changes resulted after subcutaneous inoculations. This means that either non-pathogenic yeasts or non-susceptible animals were used. Mosler fed animals beer yeasts; dogs were unaffected, but rabbits developed diarrhea. Buist produced pock-like lesions in monkeys using beer yeasts. Roussy produced fever and collapse by injecting beer yeasts into dogs and rabbits. He isolated a ferment from cultures, calling it pyretogenin. A commission later decided that this substance was invertin. Raum (1891) inoculated rabbits intravenously with various yeasts, obtaining dyspnea, fever, prostration, and death, the organisms often being recovered from the blood or viscera. A thrush culture also proved fatal, the fungus being found in liver and kidneys. Hueppe (1892) injected wine yeasts into guinea-pigs intraperito- neally, a disease resembHng guinea-pig cholera resulting. Sanfehce (1895) cultivated his Saccharomyces neoformans from fermenting fruit-juice. In guinea-pigs small nodules, consisting largely of the fungous cells, were formed in all organs except the brain, heart, and suprarenals, death occurring twenty to thirty days after inoculation. The nodules to him suggested neoplastic growths. An inoculation into the breast of a dog resulted in a condition which he called carcinomatous. There were widespread metastases. He emphasized the similarity of the organisms in tissue to the so-called coccidia in cancers. Sanfelice's Saccharomyces Htho- genes killed white mice in eight days after subcutaneous inocula- tion. Degenerative masses of the organism, especially in sheep, took up calcium salts. General infection occurred in guinea-pigs, 28 Contributions to Medical Science and the blastomyces Sanfelice isolated from " Gefliigelpocken " seems also to have been pathogenic to animals. The Saccharomy- ces niger of Maffucci and Sirleo called forth enlargement of lymph glands and suppuration at the point of inoculation in guinea-pigs, rabbits, chickens, and dogs. In some cases mihary and submiliary nodules occurred in the lungs, liver, kidneys, and peritoneum, consisting largely of epithelioid cells and the parasite. Dogs and guinea-pigs were about equally susceptible; rabbits were less so. Proliferation of the epithelium of the skin, trachea, and pulmonary alveoli, and of the endothelium of the lymph channels resulted. In the trachea, in one case, a growth developed resembling a true papilloma. However, there seems to be no reason for considering any of these changes other than inflammatory. Of great impor- tance is the work of Lydia Rabino witch, who among forty yeasts obtained from various sources found only eight which were patho- genic to animals in different degrees. Hence, from the study of a single yeast, or perhaps more, one cannot form sweeping con- clusions about the pathogenic properties of yeasts in general. Casagrandi (1897) found that Saccharomyces ruber (Demme) was pathogenic for guinea-pigs, dogs, and mice when inoculated subcutaneously or into the abdomen. The Saccharomyces septicus which De Gaetano obtained from urinary sediment seems to have been exceptionally virulent, producing fatal fibrinous peritonitis and septicemia in guinea-pigs in twelve hours. Noisette, using a commercial yeast, produced a mycosis in animals similar to that which is obtained with the Oidium albicans. The effect of certain yeasts in rabbits' eyes observed by Stoewer has been mentioned. Nesczadimenko, using two spherical and one oval yeast, inoculated mice, rats, rabbits, guinea-pigs, and dogs intraperitoneally, fibri- nous and purulent peritonitis resulting, with death in eight to twelve days. Subcutaneous injections produced inflammatory nodules, which suppurated in seven to nine days. Klein isolated a blastomyces from milk; subcutaneous inoculations produced gelatinous tumors and glandular swelHngs in guinea-pigs. Intra- peritoneal inoculations caused nodules in liver, lungs, pancreas, ovaries or testes, and intestines. The nodules consisted largely of the organism, which sometimes produced hyphal forms not Oidiomycosis of the Skin and Its Fungi 29 noted in the cultures. The lesions are similar to those caused by Curtis' organism. Finally, all those workers who claim to have cultivated blastomy- cetes from mahgnant tumors have demonstrated the pathogenicity of their organisms for various animals in greater or less degree ; and similar organisms concerned in blastomycetic dermatitis have been proved to be more or less pathogenic to animals. BLASTOMYCETIC THEORY OF TUMORS. Since Virchow described certain conditions in cancer-cells as ''endogenous cell- formation," some observers have held that these bodies were protozoa (coccidia), and others that they were fungous bodies (blastomycetes) ; the advocates of each theory claiming that they were dealing with the specific cause of cancer. Both theories have been combated by many pathologists, who hold that the bodies in question are a form of degeneration of tissue cells. This matter concerns us because blastomycetes are held to cause malignant tumors; even Sanfelice affirms that hyaline bodies and blastomycetes are identical; because we find in "blasto- mycetic" dermatitis both blastomycetes and hyaline bodies, the two being distinct morphologically in staining properties. Secchi and Anna Stecksen have given excellent summaries of the Hterature bearing on this matter. Leopold also has recently added original observations. Pellagatti concluded that the blastomyces-like bodies which observers have noted in carcinoma are the result of degeneration of plasma cells into hyaUn spherules. As noted later, the bodies which Russell, Ruffer and Walker, and Plimmer consider the parasites of malignant tumors are present abundantly in oidiomycosis of the skin. This much only is certain: that blastomycetes which are able to cause septicemia, pyemia, and local morbid conditions in animals, with connective-tissue, endothehal, and epithelial hyperplasia, have been isolated from malignant tumors. The evidence is not sufl&cient to prove that the changes produced in animals are other than inflammatory. 30 Contributions to Medical Science sporothrix schenckii. The sporotrichal infection of Schenck, also studied by Hektoen, is of interest here, because of certain similarities between its organ- ism and some of those causing blastomycetic dermatitis. A com- parison of Fig. 7, Plate 3, of Hektoen's article {Jour. Exp. Med., 1900, 5), with our Fig. 13, Plate 7, showing a form of the organ- ism from Case I, shows considerable similarity between the two organisms. The surface appearance of the sporotrichum on agar also is simi- lar to that of the organism from Case III. The pigment production of the former and the nature of the morbid process caused by the fungus seem to indicate that Sporothrix Schenckii occupies a somewhat remote position in relation to the organisms of oidiomy- cosis of the skin. The trichophyton fungi, likewise, in their morphology and methods of reproduction bear considerable resemblance to some of the oidia under consideration (i.e., the mould-fungi). The determination of the exact relationships of these various organisms will need further study. For the sake of clearness, and to group the facts, we shall consider in some detail the authentic cases of infections in man by oidium-hke and similar organisms aside from those mentioned in the foregoing; these have been pubhshed since 1894. II. SUMMARY OF THE SO-CALLED BLASTOMYCETIC INFECTIONS IN MAN.' I. SACCHAROMYCOSIS HOMINIS, BUSSE. Busse in 1894 pubUshed his first account of a case of pyemia caused by a pathogenic yeast. Three further communications appeared during the three following years. Also Buschke has furnished contributions based on this case and its organism, with particular reference to the skin lesions. ' The terminology used by the various authors is retained in the following summaries. Only peculiar cultural features are included in the abstracts; details are to be found in the table on p. 148. Oidiomycosis of the Skin and Its Fungi 31 The patient was a woman, thirty-one years old, and the wife of a shoemaker. Busse and Buschke differ as to the time of appear- ance of the skin lesions, the former stating that while a patient in the clinic a number of small round ulcers appeared on the face; while Buschke insists that previous to the trouble in the leg small ulcers would appear on the neck and forehead, which sometimes healed spontaneously. Thus the latter would make the skin the infection atrium. Besides the tibial lesion, all accessible lymphatic glands were found to be enlarged. CHnically, the tumor below the knee resembled a gumma or a softened sarcoma. From it forty or fifty cubic centimeters of glairy, reddish fluid, containing macro- scopic granules, were evacuated. From this pus Busse obtained impure cultures of a supposed yeast fungus. Later the disease extended to the knee-joint, the surrounding tissues became swollen, the joint less and less movable, and the leg flexed. Busse describes the ulcers on the face as small and round, with elevated borders, and secreting glairy, sticky pus. The influence of neighboring foci caused larger irregular ulcers. According to Buschke, the primary facial efflorescences resembled acne infiltrations, the apices of which underwent necrosis. Growth and coalescence of neighbor- ing lesions eventually formed sensitive, crater-Hke ulcers, which were rather broad, but also involved the tela subcutanea. The borders were sharp, irregular, somewhat undermined or precipitous, moderately infiltrated, and surrounded by a livid zone. The bases were of soft granulation tissue and discharged a tenacious, trans- lucent, grayish or brownish-red secretion, which contained coarse granules. Subcutaneous foci, which ruptured through the skin, caused similar ulcers. They occasionally began in hair follicles. From these skin lesions both Busse and Buschke isolated the organism found in the tibial abscess. A beginning ulcer covered by the primary crust always gave a pure culture of the fungus. In a few months there formed an abscess of the right ulna, near the elbow, and another of the left sixth rib. From the former, pure cultures of the fungus were again obtained. The latter abscess was not incised. The fungus was not found in the urine, but eight weeks before the patient's death Buschke obtained a pure culture from the left median vein. 32 Contributions to Medical Science The patient died thirteen months after the appearance of the tibial abscess and, according to Buschke, a year and a half after the development of the first skin lesions. Preceding death there were extreme marasmus, an irregular pulse, and a variable temperature. During the patient's stay in the clinic the tibial abscess was dissected out and sent to Busse for examination. The pus consisted of ordinary pus cells, large numbers of giant cells, connective- tissue cells, and the organism. The giant cells were so numerous that a softened giant cell sarcoma was suggested. The organisms were present in enormous numbers, both extracellular and intracellular, in the latter case occurring both in giant cells and mononuclear smaller ceUs. The abscess wall consisted of an outer osseous struc- ture, a middle zone of fibrous tissue, a small cell infiltrate, and an inner necrotic zone. The organisms were found in large numbers in the inner zone and to a less degree in the middle zone. As stated above, the fungus was obtained in a pure state from the pus. A noteworthy characteristic of the organism was the formation of an adventitious or accidental capsule, also described by Buschke, and later by Curtis in connection with his organism. Internal to the adventitious capsule was the cell membrane proper, which was a translucent, homogeneous, doubly contoured structure. The proto- plasm contained many moving granules and some more highly refractive bodies, which were considered fat. Prohferation in the tissue and pus was by budding, no mycelium being found. The cells were either spherical or oval. Some biscuit-shaped figures and sickle-like bodies were also found. These were considered degenerating forms. Buschke studied the skin lesions in detail. The tissue of the large ulcer on the forehead and the t>'pical focus induced by experi- mental inoculation of the patient were used. In the necrotic base of the ulcers there was no epithelium. The surface of the tissue immediately beneath it contained large numbers of parasites, some bearing the accidental membrane, and others being sickle- shaped. Here also were polynuclear leukocytes, red blood cells, and fibrin. The number of parasites decreased toward the under- lying tissue. In what he termed the middle zone were many giant cells, containing from six to twenty peripherally disposed epitheHoid Oidiomycosis of the Skin and Its Fungi 33 nuclei, leukocytes, and the fungous cells often in the budding state and surrounded by the accidental capsule. Abscesses were found in the papillary layer and deeper cutis, containing many poly- nuclear and a few mononuclear leukocytes and epithelioid cells, red blood cells, fibrin, nuclear detritus, and occasional giant cells. Parasites occurred freely and diffusely in this zone. The elastic fibers had disappeared, although in the papillae and deepest inflam- matory zone traces were found. Plasma cells occurred here and there without any special grouping. The deeply lying giant cells had fewer nuclei than the superficial, and cells of this type contained no organisms. The epithelium was affected in two ways; either it was totally destroyed, as in the base of the ulcer, or it underwent a carcinoma-like proKferation. The latter occurred particularly at the margin of the ulcer. Fungous cells were found in the horny layer and in larger numbers in the deeper rete and the enlarged pegs. They occurred within epithehal cells, the nuclei of which were flattened against the cell periphery. In all cases where the yeast cells came in relation to the epithelium the latter was destroyed. No mention is made of intra-epithelial abscesses. He considered that the parasite gained entrance into the epithelium by the agency of leukocytes. The number of organisms in the epidermis was small compared to those seen in the cutis. Buschke considered the cutis to be involved primarily, the epithelium secondarily. The necropsy disclosed, besides the tibial lesion, similar abscesses in the right ulna and the left sixth rib, and granulation foci or abscesses containing slightly cheesy pus in the lung, left kidney, and spleen. The same yeast was demonstrated in all these lesions. Many microscopic tubercles occurred in the lungs without the presence of tubercle bacilH. The kidney lesions, varying greatly in size, consisted almost entirely of yeast cells, which furthermore infiltrated the kidney in a diffuse manner. The masses in the spleen were similar to those in the kidney. The cultures, which were obtained in a pure state from the abscesses in the ulna and the rib and from certain skin lesions, developed freely in forty-eight hours. The points of peculiar interest are that it proliferates in cultures chiefly by budding; however, elongated cells were common, and Buschke speaks posi- 34 Contributions to Medical Science tively of mycelial formation with attached conidia. The organism produced alcohol and carbon dioxide in grape sugar solutions, and gelatin was not Hquefied. The formation of acid was not mentioned. The cells on culture media were of the naked type usually, but eventually Busse discovered on old cultures cells with the accidental capsule. Malt extract, bouillon, gelatin, plum-decoction gelatin, and glycerin agar with malt extract constituted the most favorable media. No top growth occurred in liquid media, and in old cultures a brown pigmentation was acquired. The formation of endospores could not be induced by the use of high temperatures (41° C.) on a moist gj-psum block or in distilled water. The use of H. Moeller's method led to demonstration of a nucleus-Hke structure. On account of a reaction obtained with osmic acid, Busse considered certain refractive globules to be fat. Pus from the tissues of the patient and pure cultures of the organism were inoculated into the tibia and beneath the periosteum in dogs. Chronic inflammation resulted, in which the fungus multipHed greatly, but spontaneous recovery occurred. An intraperitoneal inoculation of pus pro- duced a fibrinous peritonitis, in which the organism was found in large numbers. Subcutaneous inoculation with bouillon cultures produced inflammatory thickenings, which sometimes suppurated, but eventually healed spontaneously. Intravenous inoculation in dogs and rabbits produced no general disease. White mice inocu- lated in the muscles died at first in four to ten days. At a later period, presumably on account of weakened \drulence of the organ- ism, the mice lived from two weeks to thirty-three days. There were local inflammatory changes and numerous yeast cells in the blood and vessels of the lungs and kidneys, the glomeruli of the latter being chiefly involved. There was little visceral inflam- matory reaction. On the whole, dogs were not very susceptible. Rabbits, guinea-pigs, and cats reacted similarly. It was empha- sized that fatty degeneration occurred in the muscles and other organs near the points of inoculation. MyceHum did not develop in the animal tissues, proUferation occurring by the usual budding process. Busse attempted to produce epithelial proliferation by inoculation into the mammary glands and testes, but purely inflam- matory thickenings resulted. He considered the disease purely Oidiomycosis of the Skin and Its Fungi 35 of inflammatory nature, without relation to sarcoma, as held by Sanfelice. 2, SACCHAROMYCOSIS HUMAINE, CURTIS. Curtis pubHshed in 1896 a complete study of a case in which myxoma-like tumors were caused by a yeast-like organism. A preliminary report had already been published. In July, 1895, ^ tumor of this sort came into. Curtis' hands in a fresh condition. It had been situated on the right thigh, near Scarpa's triangle, was as large as two fists, and clinically seemed a deep abscess or a softened sarcoma. Similar lesions were situated in the lumbar region, and near the right ribs. The abscesses contained thick and flaky pus. The patient had been in good general health, and gave no history of hereditary disease. Later he died of meningitis of undetermined nature. Histologically the tumor consisted of a mucous or gelati- nous substance, suggestive of myxo-sarcoma. This was caused by the large numbers of the organism, which in places constituted almost the entire tumor substance. In the center was a network of connective-tissue fibers, infiltrated with leukocytes, the yeast cells lying in the fiber meshes. A zone of cell infiltration, and another of fibrous lamellar tissue constituted the abscess wall. The parasites were intracellular or extracellular. The latter often possessed the adventitious capsule described by Busse, but intracellular organisms would lose this. It grew best on acid or neutral media, development occurring in forty-eight hours to three days as small, opaque, white colonies, which fused slowly. In about eight days cultures acquired a fleshy or paste-like consistence. Gelatin was not liquefied. Saccharose solutions were fermented, with production of alcohol and acetic acid. An extraneous capsule also formed on cultures, and a brownish pigmentation was acquired eventually. Guinea-pigs were not susceptible, according to Curtis. Anna Stecksen, however, produced abscesses and septicemia in these animals with a culture which Curtis sent her. Subcutaneous inocu- lations into rabbits caused abscesses, which healed spontaneously. A general disease could not be produced by intravenous inoculations. In white rats tumors resulted, which resembled those seen in the 36 Contributions to Medical Science patient. One rat died after three months, with soUdified lungs and miUary tumors of the pleurae, spleen, and kidneys. Pure cultures were obtained from these lesions. Gray mice died in about a month. White mice developed tumors like white rats. Dogs presented variable susceptibility, small nodules or ulcerating infil- trations resulting. Since in animals the tumors exactly resembled those of the man, the conclusion was positive that the yeast caused the human disease. Spores and nuclei were not demonstrated clearly in this organism. The fungus received the name of Sac- charomyces subcutaneus tumefaciens. 3. BLASTOMYCETIC DERMATITIS, GILCHRIST. To Gilchrist belongs the credit of having first observed and described a pure skin disease caused by a yeast-like fungus. The above name, which he applied, has been used by all subsequent writers in describing this condition. So far it is exclusively an American disease, the skin lesions described by Buschke being merely concomitant foci in the pyemia of Busse's case. I. In May, 1894, at a meeting of the American Dermatological Association, six months previous to the appearance of Busse's first article, Gilchrist demonstrated sections of a pecuHar skin disease. Dr. Duhring had described the case as a typical scrofulo- derma on the back of a man's hand. Tissue was excised and sent to Gilchrist for examination. Under surgical treatment the dis- ease was cured and cultural experiments became impossible. The surface of the sections was irregularly papillomatous and covered with granular and cellular detritus. The homy layer was pro- nounced, and infiltrated with collections of leukocytes. "The epithelium was enormously hyper trophied, and consisted of irregu- lar prolongations of various shapes and sizes, extending downward into the corium. The most noticeable feature in the rete was the presence of numerous well-defined miliary abscesses of different sizes. The smallest of these were about equal in area to that of three adjoining epithehal cells, while the largest were just visible to the naked eye. They contained polynuclear leukocytes, nuclear detritus, and a few detached epithehal cells, now and then a giant cell, a number of pecuhar nuclear cells, and the organism which Oidiomycosis of the Skin and Its Fungi 37 characterized the disease." The abscess walls were made up of flattened epithehal cells. Even small aggregations of polynuclear leukocytes contained one or more organisms. There was also a diffuse infiltration of the rete, with leukocytes, and in places "the line of demarkation between the epithelium and corium was not easily recognized." There were numerous examples of premature cornification. Pigment was absent entirely from the epithelium. The papillae were greatly deformed and nearly reached the surface in places. Miliary abscesses, a few giant cells, occasional mast-cells, a large amount of granulation tissue and new-formed blood-vessels, surrounded by leukocytic and endothelial infiltration, small hemor- rhages, and fibrin constituted the chief subcutaneous lesions. Occasionally a myxomatous appearance occurred near the surface. From four to fifteen organisms were found in the miliary abscesses, very few in giant cells, and occasional examples in the granulation tissue. They measured from ten to twenty-three microns in diame- ter, and were spherical or oblong. Structure: First, a doubly contoured capsule; second, a light peripheral zone; third, a central protoplasm; and fourth, a vacuole. The capsule measured from 0.6 to 0.8 micromilHmeters in thickness. The peripheral zone was prominent, of varying thickness, and stained a homogeneous reddish- white with carbol-fuchsin and safranin. The protoplasm was seven or eight microns in diameter, reticular, and finely granu- lar, and possessed in addition numbers of larger granules of vary- ing sizes. The vacuole appeared structureless. Multiplication took place by budding. The capsule became thin at one point and the bud or protoplasm pushed out. The capsule of the bud gradually became thicker, its protoplasm increased in size, a clear space soon separated it from the mother protoplasm, and eventually the capsule united between the two portions. The parent cell would sometimes produce more than one bud at the same time. Degeneration forms were represented by sickle-shaped bodies and capsules, with no central staining substance. "The opinion was expressed by Gilchrist before the society that these organisms should be classed as belonging more to plant than to animal fife. It was not until Busse's report appeared that these organisms were posi- tively identified as blastomycetes." 38 Contributions to Medical Science 2. A preliminary report of this case by Gilchrist and Stokes appeared in July, 1896, and an extended study was published in 1898. The patient was a man thirty-three years old, in good general health, and with a negative family history. The disease began eleven years pre\"iously as a pimple behind the left mastoid process. This ulcerated, and the lesion extended forward, encircling the ear and spreading to the left side of the face and chin, affecting the palpebral and supra-orbital skin, and eventually crossed the nose to the right side of the face, forming lesions almost symmetrical with those on the left side. The disease was transplanted to various parts of the body during the eleven years. Histologically the epithelial hyperplasia, intra-epithehal abscess formations, generaHzed infiltration of the epithehum, with poly- nuclear leukocytes, abscess formation in and infiltration of the cutis, together with the organisms, showed the case to be identical with that first reported by Gilchrist. There was much granulation tissue and a few giant cells in the corium. The organism was found in all miliary abscesses, occasionally in the granulation tissue of the cutis, and to a small degree in the giant cells of the cutis. They were described as "doubly contoured, refractive, round and oval bodies, varying in size from ten to twenty microns .... presenting buds of various sizes. A vacuole was often, although not always, discernible. The bodies were usually found singly or in pairs, but they were sometimes arranged in groups." Parasites occurred occasionally in giant cells, but not in mononuclear cells. "Pseudo- tubercles" in the corium had almost the typical structure of those seen in tuberculosis. It is stated that ''pseudo-tubercles were more numerous in the papillomatous or chronic variety of the lesions; whereas, in the acute or ulcerative form, miliary abscesses, and consequently more parasites, were present." The structure of the organism was identical with that of Gilchrist's first case. Occasionally the endosporium was extruded through a break in the capsule; a second body would sometimes form before the first was detached; occasionally an organism possessed a "fibrous-like" structure external to the capsule; a half-moon shaped contraction of the protoplasm was seen occasionally. With Unna's polychrome Oidiomycosis of the Skin and Its Fungi 39 methylene blue method certain coarse granules in the fungus protoplasm stained red. Cultures from the tissues were greatly contaminated with pyogenic cocci. From the pus, however, pure cultures of the fungus appeared in one week. The length of time required for development is noteworthy. The organism was capable of pro- liferation either by a budding process or by the intervention of a mycelium, which formed lateral and terminal conidia. The extraneous capsule did not appear in cultures. Certain proto- plasmic granules exhibited Brownian movement, stained deeply in aniUne dyes, and by Millon's reagent appeared to be albuminous. The protoplasm was otherwise finely granular and occasionally vacuolated. A nucleus and endospores were not demonstrated. The capsule did not give the cellulose reaction. From the round or oval bodies segmented mycelial threads developed, on which in turn new spherical organisms or conidia formed. This observation estabhshed the unity of the budding cells and the mycelial threads. Through several generations the organism varied in its power to form mycelium, at times growing largely by budding process. Buschke, to whom Gilchrist sent a culture, found that upon potato there formed an aerial mycelium, which developed short lateral branches, without, however, forming organs of fructifica- tion, as in hyphomycetes. Inoculation of the tissue into mice and guinea-pigs produced no results, serving at least to eliminate tuberculosis. Three dogs, a sheep, and a horse were inoculated intravenously with bouillon suspensions of the organism. It was common for them to develop nasal discharges; also sinuses at the point of inoculation. They were killed in from one to two months, and in all were found vari- ously sized tubercle-like nodules in the lungs and mediastinal l3^mph glands. Cultures of the blastomyces were obtained from the lung lesions except in one dog and the horse. One intra-abdominal inoculation into a guinea-pig proved negative; another developed testicular abscesses, which discharged and from which the organism was obtained in pure culture. Organisms reclaimed from inocu- lated animals were inoculated into a pig, guinea-pigs, and mice, successes and failures of infection being about equally divided. 40 Contributions to Medical Science Attempts to produce lesions in the skin of animals, resembling those of the patient, failed. In speaking of the lung lesions in animals, the authors said that "the general appearances were those of a chronic inflammation, and one would not be hkely to regard the lesions as those of a neoplasm." In some instances they were "typical pseudo-tubercles with caseation." MyceKum was not produced in any of the lesions in animals nor in the disease of the patient. In their second report, in speaking of the classification of their organism, they say: "In our preliminary communications we classed our organism as an oidium, for the following reasons: the organism did not ferment glucose, saccharose, or lactose, and, although it only developed by gemmation in all the tissues, human and animal, yet upon artificial culture media it developed mycelia with the formation of conidia. After consulting Brefield's writings, and also after a personal conference with Dr. Irwin Smith, of Washington, an acknowledged authority in mycology, we were inclined to consider the parasite as an oidium. On further investi- gation, we find that mycologists are not at all certain about the classification of either the yeast fungi or the oidia. We find that there are a number of wild yeasts which do not ferment sugar and which form myceUum on the usual bacteriological media. In accordance with the prevailing custom in the nomenclature of organisms presenting characters similar to ours, we now prefer to designate the parasite as a blastomyces rather than an oidium." Accordingly they applied the name "Blastomyces derma titidis " to the organism. 3. This case was described by H. G. Wells in March, 1898. The patient was a healthy farmer, forty years old, and a native of Iowa. The family history was negative, and the general physical condition of the patient good. The disease began eleven years pre\iously as a pimple on the dorsal skin of the left little finger. Tliis ulcerated and extended gradually until the whole dorsum of the hand was covered, the palm being avoided. The whole lesion was removed and the defect remedied by a plastic procedure. Cultures yielded saprophytic bacilli and pyogenic cocci. A piece Oidiomycosis of the Skin and Its Fungi 41 of fresh tissue placed subcutaneously in a rabbit caused an abscess the size of a hickory nut, which discharged spontaneously on the ninth day. From the resulting sinus white pus could be squeezed. A second rabbit was inoculated with this pus. An indurated node formed, which subsided without suppuration. Repeated examinations of the pus failed to reveal yeast cells, and cultures yielded Staphylococcus pyogenes aureus. Microscopically, the tissue corresponded closely to the cases of Gilchrist, and Gilchrist and Stokes. Epithelial hyperplasia, intra-epithelial abscesses, dense infiltration of the cutis, giant cell formation, and the presence of a capsulated budding fungus were characteristic. The abscesses contained one or more parasites. Dr. Wells notes that in some of the intra-epithelial abscesses, "having no communication with corium, tubercle-like giant cells are found, which must have been formed either from the epithelium or from the cells that have wandered in." Numerous giant cells of the tuberculous type, often vacuolated, were found in the corium. However, "none of the characteristic zones of tubercular foci are found about the giant cells." Tubercle bacilli could not be demonstrated. The number of single and budding organisms found within giant cells was remarkable. The structure of the organism was given as follows, from without inward: First, a capsule; second, a trans- parent zone; third, a central protoplasm; fourth, a vacuole within the protoplasm. The vacuole was present inconstantly. Typical budding occurred. Plasma cells have been demonstrated recently in large numbers in the sections. It is to be noted that the subcutaneous implantation of the human tissue into a rabbit produced lesions identical with those observed in animal experiments by other writers using pure cul- tures of "blastomycetes." This justifies the suspicion that a fungus of this sort caused the swellings and abscesses. 4. Reported by Hessler in August, 1898. In October, 1897, the patient was cut below the chin while being shaved by a barber. The wound healed in a few days, but a red papule soon developed. It was about the size of a half-grain of wheat, hard, elevated, and remained stationary for almost three 42 Contributions to Medical Science months. It increased to pea-size and suppurated. No pus organ- isms were found in smear preparations, but there were large budding cells, which, it is stated, were largely within leukocytes. Agar inoculations gave pure cultures of a large budding cell "in a few days." On the surface of glycerin agar stabs a dirty white mass developed, ''consisting of a dense network of fibers, with cells of various sizes." The organism grew poorly in plain glucose solutions, "but abundantly on the addition of a little extract of beef and without the formation of carbonic acid." It was stated that only on solid media did fungus threads appear "after days or weeks." The organism was slightly oval, with a clear outer envelope, and somewhat smaller than the yeast described by Gilchrist. Welch considered it a blastomyces. 5. Reported by Hyde, Hektoen, and Bevan in 1898; Hektoen in 1899 gave an additional detailed report on the organism. The patient was a Hollander, fifty-six years old; his general health was poor. The disease began four years previously as a reddish spot on the right leg, nearly the whole anterior surface being finally covered. The case was regarded as one of tuberculosis until Hektoen demon- strated budding organisms in frozen sections. The histology was that of previous cases; epithehal hyper- plasia, intra-epithelial and subcutaneous miliary abscesses, dense infiltration of the cutis, and the presence of a characteristic organism being the principal features. There were isolated epithehal whorls, partially cornified. The abscesses were not different from those described by Gilchrist. Plasma cells are mentioned for the first time in connection with this disease. Mulberry-Hke masses of hyalin bodies are described, one occurring in a cell with a plasma-cell-like nucleus. The organism measured from ten to twelve microns in diameter and presented the same structural features found in those of previous cases. It was not found in the cutis, nor within giant or epithelial cells, but frequently in the intra-epitheUal abscesses. Budding forms were numerous. Methyl- ene blue was considered the best stain for the parasite, the capsule staining deeply, and the protoplasm a fighter blue. Oidiomycosis of the Skin and Its Fungi 43 In cultures a yeast fungus occurred constantly, but intimately associated with a pseudo-diphtheria bacillus. After much diffi- culty the fungus was separated. However, further confusion resulted by the development of a small budding organism, which was a Httle larger than a large coccus. This was proved satis- factorily to be a form of the larger yeast plant. On agar and gelatin plates, distinct, coarsely granular colonies appeared in twenty-four hours. Individual cells could be recognized with low magnification. Threads did not shoot out, as happened in similar cultures of Gilchrist's organism. Multiplication by budding occurred in all media. On translucent solid media, abundant, feathery, cone-shaped down-growths, resembling those seen in some varieties of ray fungi, appeared in from eight to fourteen days. This also occurred in stab cultures. Yellowish-brown pigment formed around the spherical cells on plain agar. No acid was produced, and when the organism was first described no fermentation occurred in lactose, dextrose, and saccharose solu- tions. (It has been demonstrated lately that the ability to ferment maltose and dextrose solutions has been acquired.) No indol was formed. It was perhaps a little smaller in cultures than in tissue. The capsule, clear zone, and central protoplasm were recognized in the cultures. Gradually, on a part of the culture tube, the organisms became larger and more vacuolated. In liquid media a small form, one to five microns in diameter, frequently formed. They were observed to give rise to a large type of the organism. Myce- lium formed frequently in liquid media. Buds would arise from any point of the myceHum, and were sessile or pedunculated. A guinea-pig inoculated with tissue died a month later, the pseudo-diphtheria bacillus being recovered. The liver was cir- rhotic and necrotic, these changes being ascribed to the bacillus. Two white rats received subcutaneous inoculations of a bouillon suspension. Abscesses resulted from which the fungus was recov- ered in pure culture. Two gray mice were inoculated. One died after five days. Pure cultures of the blastomyces were obtained from the local abscess, but cultures from the internal organs were sterile. Staphy- 44 Contributions to Medical Science lococcus albus was found in the abscess and in all the internal organs of the second animal. Two inoculations in the anterior chambers of the eyes of rabbits resulted in h>popyon, and on their death, forms resembling the blastomyces were found in the pus. Inoculations into guinea-pigs gave rather negative results. A white rat died from an intra-abdominal inoculation. There were several foci in the limgs, in which "large, round, doubly contoured bodies" were found. A rabbit inoculated intravenously was killed after a month. In the lungs were found mihary foci of granulation tissue, but no blastomycetes. An intraperitoneal inoculation of a dog proved negative. A rabbit received an intravenous inoculation of four cubic centimeters of a bouillon suspension. Death occurred the next night. Smears and cultures from all organs yielded the blastomy- ces. On section, the lungs were congested, and blastomycetes were found in necrotic areas. The liver and spleen were normal. In the kidneys the glomerular and tubular epithehum were degen- erated, and in the former, bodies resembhng blastomycetes were found. A later intravenous inoculation into a dog resulted in its death in about a month. There were "minute foci of granulation tissue throughout the lungs, and softened cellular areas with yellowish contents in the medullary pyramids of the kidneys. The blastomy- cetes were recovered in pure growth from the lungs and kidneys." Hektoen comments: "Our organism grows much more rapidly than the one of Gilchrist and Stokes. The formation of mycelium is not nearly so marked as in their cultures, in which were not seen the pecuhar down and out growths and lateral branchings, nor the pigment formation (on agar-agar) characteristic of the blasto- myces now described. Both organisms correspond, however, in their action on gelatin, which is not liquefied, in the non-production of indol, and in the complete absence of fermentation of various sugars. Morphologically they are also quite or nearly alike. Gilchrist and Stokes make no mention of such great variability in the size as observed in our organism." Oidiomycosis of the Skin and Its Fungi 45 6. Reported in November, 1899, by Owens, Eisendrath, and Ready. The patient was a widow, thirty-eight years old. Four years previously a swelling had appeared on the antero-internal aspect of the lower third of the left thigh. On April 29, 1900, the whole lesion was removed. The subcu- taneous tissue was not involved. The wound was successfully closed with skin grafts. "Under the low power, the resemblance of the section to an epithelioma is most striking." Epithelial down-growths were pro- nounced, and many of them contained miliary abscesses. The cutis was densely infiltrated. The organism was found in the epithelial prolongations, either between cells or within giant cells; to a less degree in miliary abscesses, and most of all in the cutis, either surrounded by infiltration cells or within giant cells. The organisms were capsulated, contained a clear zone and a central protoplasm, the latter sometimes showing vacuoles, and sometimes possessing ''small bodies like spores." Budding was observed. The best stain was methylene blue. A guinea-pig was inoculated subcutaneously with a piece of fresh tissue. An abscess appeared in eight days, and on the twelfth day discharged a "creamy, viscid, foul-smelling material." Blas- tomycetes were not demonstrated in smear preparations of this pus. Cultures, however, yielded blastomycetes, which were much contaminated by bacilli and cocci. The organism was obtained in a pure state by growing the impure cultures in potassium iodide solutions. Contaminating organisms were thus eliminated. The blastomyces suffered a reduction in size in this process, the former size being regained by successive inoculations on favorable media. As far as described, this organism is identical with that of the Hyde, Hektoen, and Bevan case. However, the character- istics were not fully worked out. It is unfortunate that the pure culture was obtained through the medium of an animal inoculation rather than directly from the human tissue. The chance for error was thus greatly increased. 7. By Ludvig Hektoen in December, 1899. "Mrs. W., aged sixty-four, married, with an uneventful family 46 Contributions to Medical Science and personal history, was admitted into Professor Murphy's clinic November lo, 1897, on account of a lesion of the right leg. When she appeared in the clinic there was a circular, elevated, cauhflower- like mass, as large as a silver dollar, with an irregularly ulcerated and red surface. The growth was regarded as probably car- cinomatous, and removed." The tissue was preserved as a museum specimen and later used for histological study. It was described as an ovoid, raised area, four by five centimeters in its principal diameters. The verrucous character of the surface gave it a cauliflower appearance. The color was yellowish and showed superficial erosions. On section, the yellowish tissue of the lesion was sharply defined from the underlying fatty tissue. Histologically the elevation was made up largely of hyperplastic epithehum, which formed irregular and branching bands, masses, and nests, presenting a "marked resemblance to flat-celled carcinoma." The rete exhibited the usual leukocytic infiltration and intra-epithelial abscesses. Miliary abscesses also were found in the cutis. Plasma cells, lymphocytes, eosinophiles, giant cells of the tuberculous type, and new-formed vessels containing lymphocytes, plasma cells, and polymorpho- nuclear leukocytes were present in the cutis. Smaller giant cells were found in the miliary abscesses in the corium and epithelium. The organisms, "typical, doubly contoured blastomycetes," occur in the miHary abscesses, in the connective tissue of the corium, and occasionally in giant cells. "The organisms are round, about twelve microns in diameter; distinct budding forms are present; some of the bodies contain one or more vacuoles of irregular size and outHne Some organisms are stained very deeply in the methylene blue specimens." The author emphasizes the clinical resemblance of the growth to carcinoma, and its histological resemblance on the one hand to carcinoma and on the other to tuberculosis; true carcinoma may some day be found in connection with blastomycetic dermatitis as it has been found in lupus. He considers the disease well estabhshed, and points to the probability of more than one variety of fungus being the etiological factor in different cases. Oidiomycosis of the Skin and Its Fungi 47 8. Anthony and Herzog in 1900 reported a case in a man who gave an uneventful personal and family history. Of the histology of this case, it is sufficient to say that it corre- sponds with that typical of " blastomycetic " dermatitis. Cultural and inoculation experiments are not recorded. 9. By W. E. Coates, February, 1900. The patient was a married man, thirty-eight years old, giving no personal or family history of carcinoma, tuberculosis, or syphilis. His general health was good. In January, 1898, the knees and elbows were involved in acute, painful rheumatism, with little swelling. Two weeks later severe pains began in the chest, and he was confined to bed for three months by what was called a pleuro-pneumonia. There was a cough and bloody expectoration. While recovering from this illness, a small nodule appeared on the lower right eyelid, the lesion reaching the size of a bean in three weeks, when it was removed. At different times other small nodules appeared elsewhere. The diagnosis of multiple epithelioma was made by a previous examiner, based upon microscopic examination of the tumor of the eyelid. Dr. Coates, however, discovered blastomycetes in the tumor of the lip, which histologically resembled an epithelioma. The epithelium had proliferated greatly and formed large masses and pegs, penetrating the cutis. There were many intra-epithelial abscesses which contained giant cells and a budding, capsulated cell about twelve microns in diameter. Smaller spherical bodies, taking up nuclear stains, the author thinks are small forms of the organism, although the difficulty of distinguishing them from tissue detritus is admitted. Some, he thinks, are small spores. The large spherical cells are considered spore-bearing bodies, although the evidence for this belief is not presented. There are also miliary abscesses in the cutis, containing fungus cells and a few giant cells. The cutis is densely infiltrated with leukocytes, plasma- and mast-cells. The organism does not occur in giant cells. Tubercle bacilli could not be demonstrated in the sections. The author, on the basis of a chain of four or five loosely united 48 Contributions to Medical Science fungus cells, theorizes quite positively as to hyphal elements, mycelium, sporing cells, and secondary spores. He considers that the intra-epithelial abscess is "lined by deHcate fungus filaments." This theory, however, is not satisfactorily demonstrated. 10. At the meeting of the Chicago Academy of Medicine, on January 12, 1900, Dr. Baldwin presented a case under the tentative diagnosis of epithelioma. The disease occurred in a Polish woman, forty years old. As far as described, the beginning of a lesion was either a pimple or a small swelling, "having a covering like tissue paper," the rupture of which resulted in a discharge of pus. The more recent lesions measured from three to eight centimeters in diameter, were roughly circular or oval, elevated from two to five milHmeters, and had a fungoid or cauliflower-like surface. The borders were usually more elevated than the centers (which were sometimes decidedly depressed), and surrounding was a dark-red halo of elevated skin. There was a little oozing, some fetor, and a varying amount of pain and tenderness. The older lesions were flatter than the recent ones. There was no glandular enlarge- ment. Recently some anesthesia had developed in the right arm and hand, and there were "rheumatic pains over the entire left tibia." No history of syphilis could be obtained. A brief histological report was offered: "Histologically, an epithelioma; a typical epithelial cell proliferation into the deeper tissue. Sections were stained for the tubercle bacilli, but none were found, though stained by thionin, methylene blue, and the Klebs and triple stain for protozoa; some deeply staining bodies, with hyalin rings in epithelial masses, were probably protozoa. A few organisms were found with double contour." In the discussion which followed, the case was called variously epithelioma, lupus, and Mycosis fungoides. One critic observed that "there is no such thing as blastomycetic dermatitis per se, but that blastomycosis is found secondarily to previous lesions." It is unfortunate that attempts were not made to cultivate an organism from this case, and that a more detailed histological report was not offered. 11. Dr. A. W. Brayton, in April, 1900, reported a case in a livery-stable employee, fifty years old, married. Oidiomycosis of the Skin and Its Fungi 49 The description of the lesions is brief: "There were three open and undermining ulcers the size of a bean. The swollen margins were very painful." The article reports that Professor Catherine Golden found "Blastomyces dermatitidis " in the sections, and that Dr. W. T. S; Dodds cultivated the yeast plant on wort-gelatin, subsequently inoculating a guinea-pig, which two weeks later "showed the recognized symptoms of blastomycetic infection." The yeast plant was found in tissue macerated in potassium hydrate solutions. The communication does not give cultural or histo-pathological data. 12. Reported to the International Congress of Dermatology and Syphilography in Paris, August 2-9, 1900: The patient came to the hospital dispensary of Jefferson Medical College in April, 1899. He had been scratched or bitten by a cat on the back of the right hand. This wound healed. A few weeks later a pustule appeared at the point of traumatism. Gradual extension occurred, eventually covering the dorsum of the hand. Doctor Rosenberger gives the following histological and cultural report : The pus both from the arm and the hand lesions contained pus cells, bacilli, cocci, and spherical cells possessing buds. In sec- tions similar spherical budding cells were found deep in the tissue, arranged mostly in groups. After macerating tissue in bouillon it was inoculated on glycerin agar and a liberal culture of the large spherical budding cell obtained, which stained homogene- ously. Other cultural and histological data are lacking. It was con- sidered an undoubted case of blastomycosis of the skin. Complete cure had not been obtained. The description given of the lesions on the hand answers for that of typical "blastomycetic" dermatitis, except for miliary abscesses. No mention is made of this feature. The lesions on the arm, however, are a deviation from the type and have not been observed in other cases. The formation of an abscess which is at first entirely subcutaneous, rupturing later at several points, causing a cribriform appearance, suggests to a degree the sporothrix infections of Schenck and Hektoen. 50 Contributions to Medical Science 13. By Isadore Dyer, January, 1901. The patient was an Englishwoman, fifty-five years old. In 1898 a growth called an epithehoma was removed from the forehead by a caustic paste. In February, 1899, "granulation sores" appeared on the backs of both hands, and a warty growth on the face. The author emphasizes the clinical resemblance of this case to "yaws." Gilchrist studied the tissue and found all the characteristic features of "blastomycetic" dermatitis. The organisms were not numerous. Dr. P. E. Archinard after many experiments obtained an impure culture of the fungus, which was sent to Gilchrist. So far, the report on the organism is incomplete. Gilchrist found spherical and oval forms and a branched jointed mycelium, "which frequently contained spores." It is remarked in a general way that the growth on media is not at all like that of the Gilchrist-Stokes organism. The photographs and description of the case are typical of "blastomycetic" dermatitis, and speak for the oidium-like nature of the organism. 14. Presented by Frederick G. Harris at the February, 1901, meeting of the Chicago Pathological Society. The patient was a German woman, seventy-eight years old. Four years ago a pimple formed on her left hip. This became roughened and denuded ; it extended slowly, and was accompanied by moderate itching. On admission to the Cook County Hospital of Chicago, the clinical diagnosis of syphilitic ulcer was made. The diagnosis was changed to epithehoma after a small piece had been removed and examined microscopically. Acting on this, the area was excised and the defect covered with flaps. Microscopically it was found that the epitheHum formed down-growths some distance from the border, these increasing in depth and complexity as the border was approached. The papillae were formed by extensions of the corium, which were covered with the proliferating epithelium. MiHary abscesses Oidiomycosis of the Skin and Its Fungi 51 occurred in the hyperplastic epidermis, and there were many epithelial whorls. The structure and contents of the abscesses are similar to those in other cases. Giant cells occurred in the intra-epithehal abscesses and in the cutis. The cutis was densely infiltrated with polymorphonuclear leukocytes and plasma cells, and contained small abscesses. The organism was found in intra-epithehal and subcutaneous abscesses and occasionally in the granulation tissue of the cutis and in giant cells. It formed budding pairs and groups of three or four, being "often surrounded by a granular material, which took the eosin stain." They were not found free in the rete nor in the surface discharge. A small tumor which had existed on the nose for twenty years was excised and foimd to be a benign epithelioma, of the tj^De which originates from the gland ducts. Through Dr. Harris' kindness I was able to examine his speci- mens and I can fully corroborate his statements. 4. THE PROTOZOIC (?) INFECTION OF WERNICKE AND OTHERS, ITS RELATION TO SACCHAROMYCOSIS HOMINIS AND OIDIOMYCOSIS OF THE SKIN. A consideration of this question is justified here because of the similarity of the so-called protozoic infections described by Posadas, Wernicke, Rixford and Gilchrist, and others, with Busse's Sac- charomycosis hominis, because of the resemblance of the skin lesions in these two diseases to oidiomycetic dermatitis, and because some of the organisms concerned are virtually identical. Wernicke in 1892 reported briefly the first infection of this type, accompanied by the doubtful diagnosis of Mycosis fungoides. Rixford and Gilchrist, however, in 1896, pubhshed a thorough description of two cases, estabhshing a rather characteristic histo- pathology, and concluding that the organisms were true sporozoa. They gave the names " Coccidioides immitis" and " Coccidioides pyogenes" to the organisms in the two cases, because of their close resemblance to coccidia and the active pyogenic properties of one of them. Rixford and Thorne in 1894, and Posadas, Ophiils and Moffitt, and D. W. Montgomery in 1900 have also reported cases. 52 Contributions to Medical Science Tropical and semi-tropical countries favor the infection, judging from the histories of the few cases reported. Three of the cases developed the disease while living in the San Joaquin Valley, CaHfornia. The patients have been men in early or middle life. Heredity seems to play no part, and the occupations have been various. It seems probable that the infection atrium is in the lungs in most cases, although in Rixford and Gilchrist's first case it is fairly positive that the primary seat was in the skin. The disease may be very acute, beginning with a chill and fol- lowed by rapid lung and skin involvement (Ophiils and Mofiitt). A cough may be the first symptom noticed, serious lung and skin involvement soon following (D. W. Montgomery); or skin lesions may appear before visceral involvement is observed (Rixford and Gilchrist's first case and Posadas' case). In Rixford and Gil- christ's second case the general infection was fulminant, following violent hematemesis. After these symptoms of general invasion, emaciation and asthenia follow quite rapidly; the temperature may rise to 101-103° F. in the evening and fall to almost normal in the morning, death occurring in a few months. Rixford and Gilchrist's first case lived nine or ten years after the appearance of the first skin lesion, and about a year after the first hnrnphatic involvement. Their second case lived four months; Ophiils and Moffitt's case, only three months; and D. W. Montgomery's case, about a year after the appearance of the first symptoms, which were visceral in these cases. The skin lesions in all cases have certain common character- istics. They may begin as subcutaneous, or more superficial abscesses; or granulomatous growths which may be nodular (Wernicke, Posadas, D. W. Montgomery), suggesting Mycosis fungoides. Eventually, however, in all cases the surfaces are covered with thickly set large papillae, between and covering which is an offensive purulent secretion. The surface is fungating or cauliflower-like, borders usually elevated and surrounded by a livid and moderately indurated narrow zone, and the centers of larger lesions depressed and granulating. Histologically there are great epithelial hyperplasia, epidermal and subepidermal abscesses, and a granulomatous condition of the corium and subcutaneous Oidiomycosis of the Skin and Its Fungi 53 tissue, marked by dense infiltration, giant cells, and tuberculoid nodules. The organisms are present in vast numbers in the surface pus, in the epidermal and subepidermal abscesses, and in the granulation tissue of the corium. Lungs, pleurae, Kver, kidneys, spleen, bones, suprarenals, testicles, and the lymph glands adjacent to these organs have been found invaded. In the lungs small and large abscesses, miliary nodules, consohdation, and bronchopneu- monia occur. The lung involvement is easily recognizable clini- cally. In liver, spleen, kidneys, testicles, and lymph glands there may be small nodules, or abscesses containing cheesy or necrotic material. In Rixford and Gilchrist's first case there was a focus of osteomyelitis in the left tibia and another in a metacarpal bone and articulation, the joint being disorganized. The suppurating sinuses led to sequestra. Ophiils and Mofhtt found ostitis and periostitis of the frontal bone and of the upper part of the left tibia, with suppuration of the joint. The abscesses and nodules in all organs consist largely of the protozoa ( ?) , to the exclusion of all other organisms. The adult parasite in tissues is unicellular, varies in diameter in different cases from sixteen to thirty-five microns, possesses a doubly contoured capsule and a non-nucleated protoplasm, which gradually is transformed into a hundred or more structureless sporozoites ( ?) . These escape through a rupture in the maternal capsule, and eventually acquire the capsules and the structure of the adult cell. No budding forms have been mentioned. One of the organisms described by Rixford and Gilchrist possessed vacuoles, in the other none were found. This is the life history in tissues as detailed by Rixford and Gilchrist, and by Ophiils and Mofhtt. The latter, however, cultivated a mould-fungus from all diseased organs, the growths being unmixed for the most part. Pus containing organisms was mounted in a hanging drop and the shooting out of mycelium from the spherical cells noted. Tissue and pure cultures inoculated into animals produced conditions analogous to those in the man, the spherical, sporing form always appearing in tissues and the mould-fungus in cultures. These results certainly seem to establish the identity of the forms found in the tissues and the culture. 54 Contributions to Medical Science There seems to be little room to doubt that the case of Ophiils and Moffitt is identical with the so-called "protozoic" disease of Wernicke and others. Then, it must be asked, why did Rixford and Gilchrist fail to obtain cultures in their cases? In answer it may be pointed out that the identical state of affairs is seen in "blastomycetic" dermatitis. The organisms in two cases of "blastomycetic" dermatitis may be quite similar, yet the cultural pecuHarities of the two may differ widely, one growing readily on all ordinary media, and another being at first cultivated only on a special medium. After a period of artificial nourishment, its adaptabiUty may make it possible for growth to occur on media which at first were unfavorable. In cases where an organism could not be cultivated, the nutrient requirements have probably not been met. Although it is anticipating somewhat, it seems proper to consider briefly in this place the interrelationship of the three diseases which have been summarized. Comparison of Busse's case with the so-called protozoic infec- tions shows a close clinical and anatomical analogy. In both are found a chronic pyemia, eventually ending fatally, characterized by multiple abscesses and miliary or larger nodules in nearly all viscera, and by a clinical course similar to general miUary tuber- culosis. In both the lesions contain vast numbers of parasites, and the purulent secretions Ukewise are heavily laden with organ- isms. There are slight differences in the histology of the skin lesions, but the visceral conditions are virtually identical. The bone involvement is similar. The difference noted in the organisms concerned is not sufficiently great, moreover, to warrant a wide separation of the two conditions, as illustrated again by referring to "blastomycetic" dermatitis. In two of the cases of the latter disease, identical in their clinical and pathological findings, I found two different organisms in the cultures, one closely related to that in Busse's case (appearing largely as a budding fungus), and the other closely related to that in Ophiils and Mofl&tt's case (growing as a mould-fungus). In other words, there are no greater differ- ences between the two cases of fungus septico-pyemia reported by Busse, and Ophiils and Moffitt, respectively, than between Oidiomycosis of the Skin and Its Fungi 55 Cases I and VII of " blastomycetic " dermatitis, in respect either to clinical, pathological, or cultural findings. That oidial or blastomycetic dermatitis is related to the so- called protozoic infections and to Busse's Saccharomycosis hominis seems extremely probable, particularly when the conditions existing in Case V of this series are studied. It seems probable that the first-named disease seldom becomes a generahzed in- fection. The difference between the skin lesions of the "protozoic" disease and of ''blastomycetic" dermatitis may be considered as one of virulence purely. In the former more pus is formed, and the organisms are present in much larger numbers. There are the same papilliform lesions, the same minute changes in epidermis and corium, and the same relation of the organisms to the lesions. The essential difference in the organisms is that those of the so-called "protozoic" infections have not been observed to proliferate by budding, while in blastomycetic dermatitis this is the only proved method of proliferation in tissues. The former appear to produce endogenous spores; several organisms of the latter disease (Cases III, VI, VII, VIII, IX, and X) produce similar endocellular bodies. The organism described by Ophiils and Moffitt is a mould-fungus; five of ours are mould-fungi. Apparently, then, the differences may be those of adaptation, or, at the most, specific variations. Between blastomycetic dermatitis and Busse's Saccharomycosis hominis there are certain differences. The former so far is a local disease; Busse's was a general infection. In the skin lesions of the latter the papillomatous surface did not develop and necrosis was extensive. According to Buschke the parasite destroyed the epithelium rather than caused hyperplasia. We infer that intra-epitheHal abscesses did not form, as the authors do not speak of them. If that is true, it constitutes an essential histological difference between the two processes in the skin. As shown in the table (p. 148), Busse's organisms differ only in a few details from a whole group isolated from cases of blastomycetic dermatitis. 56 Contributions to Medical Science SUMMARY. The disease called a protozoic infection by Wernicke, Rixford and Gilchrist, and others seems identical with the case reported by Ophiils and Moffitt, from which the latter cultivated a mould- fungus. The disease described by Busse as Saccharomycosis hominis is identical clinically and anatomically with the Ophiils and Mofl&tt case, the difference in the skin lesions not being essential, and the two organisms being separated by specific rather than generic differences. Oidial dermatitis, called by Gilchrist *'blas- tomycetic dermatitis," differs from both the above in being purely a local disease, and in having a relatively small number of organisms in the purulent secretions and skin lesions. It resembles the skin lesions of the so-called protozoic infection both grossly and histologi- cally. It differs from the skin lesions of Saccharomycosis hominis in that the latter exhibits an excess of destruction over hyperplasia; and its various organisms differ from those of the protozoic disease and Saccharomycosis hominis in no greater degree than they differ among themselves; hence the differences between these diseases may all be explainable on the score of specific variations of the fungi concerned. III. NEW CASES OF OIDIOMYCOSIS OF THE SKIN. Case I.^ — Clinical history (Dr. Montgomery). — J. H. C. pre- sented himself August i8, 1899, with a distinct tumor-Uke swelling of the lower lip. He was forty-five years old. Examination showed an irregularly rounded tumor of the lip, situated a little to the left of the median line, averaging about two centimeters in diameter. On his second \isit the man stated that the corn crop of the previous year, in his locality, had been unusually subject to a form of "dry rot," producing a fine, brown, very light powder between the husk or kernel and the cob. He had handled some of this grain, and said that many cattle and some horses had died after eating corn so diseased that had been left in the fields. He further stated that he was in the habit of chewing and biting grain. » A brief report of Cases I and II has been made (Montgomery and Ricketts), Jour. Cutaneous and Genito-Urinary Diseases, igoi. Oidiomycosis of the Skin and Its Fungi 57 The improvement under iodide of potassium on the whole was so slight that an operation was decided upon, and on November 4 the tumor was removed. Cultures. — Deeply situated fragments of tissue were excised aseptically, teased, and inoculated on ordinary culture media. At the end of twenty-four hours small, elevated colonies, with a moist, whitish, and gHstening surface, were observed on agar and blood-serum slants; and on potato small, dry, white elevations, which in a few days grew directly into the air in the shape of fine granular shafts. Two other varieties of colonies appeared, which proved to be staphylococci and streptococci. Smears from the colonies first mentioned, stained with LoefHer's methylene blue, disclosed a large micro-organism of varied mor- phology. The predominating form was oval, and to many of these were attached smaller, apparently budding daughter cells. Short segmented threads were common. Both the oval forms and threads had a delicate, translucent capsule and a central substance, which stained irregularly. The single forms averaged seven-eighths micron in diameter, though there was great variation in size. The colonies of the fungus were greatly in excess of those of all other organisms combined. Its biological characters were studied on various culture media. Ox-blood serum. — A broad, grayish-white line develops within twenty-four hours, which grows slowly at the room temperature, and more rapidly when in the brood-oven. Within five days the growth is elevated, substantial, moist, one and one-half millimeters broad, and has a smooth, glistening, white surface, marked by scattered pin-point elevations and depressions. The edge is irregular and sends out hnear projections, straight or curved, which may possess small, knob-like extremities. The growth, under the microscope, presents spherical and oval adults, budding forms, and occasional elongated cells (abortive myceUum). There are some particularly large cells (involution forms), of irregular contour, much vacuolated, and with very little chromophilic substance. Agar slants. — On the agars (glycerin, glucose, maltose, plain, and beer-wort agar) the growth is more rapid than on blood-senmi. 58 Contributions to Medical Science It grows least readily on plain agar. There are the same surface appearances as on blood-serum, but the edge is flat and smooth. After three or four weeks the surface becomes more uneven, appear- ing as if bubbles of gas had burst, making circular rents. This is most noticeable on glucose and beer-wort agar. Fine rays penetrate the medium vertically. Microscopically the agar growths dififer from the blood-serum only in forming more myce- lium. Specimens from a maltose agar tube show polychromatism, certain cells exhibiting red granules with alkaline methylene blue. Young cells stain deeply and uniformly; older cells have one or more deeply staining granules; the oldest and largest may either stain deeply or not at all. On -f lo oxalic acid maltose agar, development is slow, and the surface has a moist, glassy, yellowish- white appearance. No mycelium, not even elongated cylindrical cells. After one hour in a hanging drop of tap water mycehal segments began to shoot out and budding was stimulated. The central protoplasm is sometimes differentiated into a spherical, slightly tinted, nucleus- like substance, which may or may not contain one or two fat-hke globules. On a glycerin agar slant, four months and ten days old, the surface is Kght brown and shows circular rents. Mounted in a KOH solution, the spherical cells are from 2 to i6 microns in diam- eter. The protoplasm is granular, and contains many fat-Hke globules. In the largest cells the capsule is from 1.5 to 2 microns thick, but the accidental capsule seen by Busse and Curtis in their organisms does not occur. There are mycehal fragments of similar structure. The reverse surface of agar slants does not show a golden-brown color. Glucose agar stab. — No growth occurs in the anaerobic stab. In the aerobic, the line is gradually marked by a whitish growth, which soon sends deHcate horizontal rays into the medium. This is seen more clearly in the gelatin stab. Glucose agar plate. — Within twenty-four hours small white colonies develop, which have characteristics identical with those seen on the original agar slants. Under the low power they appear Oidiomycosis of the Skin and Its Fungi 59 granular, and around the border minute globular forms are seen, closely packed, but sharply defined. No mycelium. The buried colonies are smaller than those on the surface. Bouillon gelatin stab. — ^A white, finely granular growth occurs along the line within twenty-four hours. Gradually deHcate threads penetrate the medium transversely, and eventually become densely packed. The medium is not liquefied. A plum-decoction gelatin, however, was liquefied slowly. Potato. — At the end of twenty-four hours the growth is con- spicuous, and proceeds rapidly; is coarsely granular, elevated, and tends to heap up rather than extend over the surface of the medium. When the potato is fresh, the growth has a moist, grayish-white color, but later a dry, white, powdery appearance is acquired. Many large vacuolated and granular degeneration forms occur. Bouillon. — Growth is somewhat more rapid in two per cent glucose than in plain bouillon. In two to five days, at brood- oven temperature, a grayish- white, fluffy cloud lies at the bottom of the tube, the overlying fluid remaining clear. The sediment is easily disseminated by shaking. Microscopic examination shows mycelium and spherical forms promiscuously mixed. These are best studied in the hanging drop. The findings are similar in maltose bouillon. Hanging-drop culture. — Growth occurs with or without myce- lium. In the latter event, it consists of ascus-like cells, which multiply by budding. The mycelium is segmented and branching. The oldest segments are from four to six microns in diameter, taper toward the growing end of the thread, and are more or less irregular in outline, offering a multitude of small projections, from which buds or conidia have recently separated. The end of a thread sometimes, but not habitually, develops a small cluster of conidia (Fig. 13, Plate 7). The segments are from two to five times as long as broad. The capsule is more delicate than that of the adult round organism. Within the capsule lies a relatively narrow clear zone, and centrally the finely granular protoplasm, which may be vacuolated. Whole threads are sometimes made up of short, incomplete segments. Portions of threads may be made up of globules of granular protoplasm in chains, surrounded 6o Contributions to Medical Science and separated from each other by a clear substance, the whole chain being enveloped in a cyUndrical membrane. Only septa are necessary to the production of chains of spherical forms, and this is accompUshed gradually. The round or oval adult organism (the ascus-like body) sometimes forms chains, giving off here a bud and there a link of myceUum. It has a capsule, a clear space internal to this, and a granular central protoplasm, which is some- times vacuolated. The budding process occurs as described by Gilchrist and Hektoen. The daughter cells have a thin membrane and do not possess a clear space or vacuoles. Occasionally a large form is seen with several daughter cells attached. When separated, they each leave a small projection at the point of attach- ment to the mother cell. Daughter cells may multiply before separating, and with further multipHcation a cluster of from four to twelve organisms may form. On old agar tubes the cells are eight to ten microns in diameter, sUghtly shrivelled, and much vacuolated (involution forms). The new cells, however, growing in a hanging drop, are apiculate and do not form mycelium. After a few months' Hfe on artificial media, myceUum almost entirely ceased to form, prohferation taking place by budding of the spherical or oval cells. After inoculating into two mice successively and reclaiijiing the organism after death of the hosts, myceHum again grew freely. A minute, coccus-like form is sometimes seen singly or in groups of from two to eight, and exhibits Brownian movement when free. They are frequently adherent to the larger forms. Clusters of small forms are bound together by a gha-Uke substance. (In succeeding cultures this type failed to develop.) Neither endospores nor nuclei have been demonstrated. Litmus milk. — A Uberal bottom and a surface growth occur. The reaction is not changed and the milk is not coagulated. Fermentation tests. — Glucose and maltose solutions ferment readily, with the production of alcohol and COj; lactose and saccharose solutions are not fermented. Potassium iodide cultures. — The organism grows liberally in a one or a five per cent solution. An amorphous brown pigment is formed in the bottom of the tube. Budding is quite general, Oidiomycosis of the Skin and Its Fungi 6i but there is no mycelium, A small amount of proliferation took place in a hanging drop of a 50 per cent solution. At the end of two weeks, however, these organisms had died, as shown by a refusal to grow on favorable media. Indol test. — In sugar-free bouillon prepared by the Theobald Smith method, a faint indol reaction was first obtained after ten days. During the next ten days it more than doubled in intensity. The ordinary forms of the organism grew in this solution. Thermal death point. — 55° C. for four minutes kills the organism. 55° C. for two minutes retards growth twenty-four hours. HiSTOPATHOLOGY. — Alcohol, Zenker's fluid, and saturated solu- tion of bichloride of mercury were used as fixatives; celloidin and parafl&n imbedding; hematoxyhn and eosin, eosin and methylene blue, polychrome methylene blue, the Gram-Weigert method, the tubercle bacillus method, acid fuchsin, and other stains were used. Under the low power the surface contour is uneven, with occa- sional cross-sections of horizontally placed villiform processes. A varying amount of debris is found on the surface. The horny layer is either entirely lost, or composed largely of flattened nucle- ated cells, and not sharply defined from the underlying rete. Infil- trating leukocytes are so numerous as to make the layer almost indistinguishable under the low power. Disintegrating red blood cells He in small masses and rows between the layers of epithelial cells. The nuclei of the infiltrating cells are much distorted in accommodating themselves to the spaces they occupy. At times a filamentous or branching form is assumed, which gives a close resemblance to fungous hyphae. The prickle-cell layer has no constant thickness or configu- ration. In places it is exposed, and often only one or two layers of rete cells protect the cutis. However, throughout the diseased area there is an enormous proliferation, which has led to the forma- tion of grotesque epitheUal processes, which extend into the cutis to varying depths, some being delicate, others constituting large masses. Counter-formations of papillary tissue are closely mingled with the epithelial processes, and deformed papillae are often cut transversely, appearing as islands of cutis tissue. None of the 62 Contributions to Medical Science epithelium is in isolated masses. The cells constituting the down-growths are often stretched, elongated, spindle-form. Throughout, the individual prickle-cells are large, commonly vesiculated (edematous), and the prickles conspicuous. The nuclei likewise appear swollen and edematous. Mitoses are fre- quent in the deeper portions. The granoplasm in certain areas is unevenly distributed, being massed in the deeper portions of the cell. It stains better with basic than with acid dyes. Such an area gives a peculiar terraced appearance to the field, the "cobblestone" appearance. EpitheHal cells with two, three, or even four vesicular nuclei are found, not always at points of marked infiltration. Mitoses are seen occasionally in such cells. Inclusion of one epithelial cell by another as cited by Gilchrist is observed frequently. Occasion- ally an epithehal cell has included, or has been invaded by, a polymorphonuclear leukocyte; the epithehal nucleus may retract or shrink to make room for the leukocyte in the perinuclear space. Polymorphonuclear and mononuclear leukocytes, eosinophiles, plasma- and mast-cells are found in the rete, the last three occur- ring rarely. The polynuclears and mononuclears are more numer- ous. In certain places the polynuclears occur in distinct groups, almost to the exclusion of all other wandering cells, constituting intra-epithelial abscesses. The distinguishing feature of the abscesses in this case is their minute size. The origin may be within an epithelial cell, which eventually is destroyed as the leukocytes multiply. Included thus no compression effect is made on surrounding epithelial cells until the invaded cell is destroyed. Originating in intra-epitheHal spaces, adjacent prickle-cells are flattened to varying degrees, and the prickles are lost while the abscess is still minute. The flattening process continues as the abscess increases in size, so that eventually the surrounding cells appear as fibers in which nuclei may or may not be found. Thus a Hmiting capsule is formed of several concentrically disposed layers of flattened epithelial cells. The more remote cells are less and less flattened. Between the fiber-like cells nearest the abscess leukocytes accumulate. As they increase in numbers the flattened and elongated rete cells are loosened and become a part of the Oidiomycosis of the Skin and Its Fungi 63 abscess contents. As such they are relatively long threads, with tapering extremities, having elongated flattened nuclei, and some- times presenting the appearance of branching because of close union with other cells. They are sometimes not unlike translucent fungous filaments. Instead of flat cells the wall is at times made up of cuboidal, normally shaped rete cells which may become detached and form part of the abscess contents. Such cells become spherical, their protoplasm granular, and strongly acidophilic, their nuclei swollen and possessing faint staining properties. Either before they become separated from the wall or later they may enclose leukocytes and other epitheUal cells. When a poly- morphonuclear leukocyte is first enveloped its protoplasm is prominent as a clear hyaHn mass, imbedded in the acidophilic protoplasm of the epithelial cell. The leukocytic nucleus stains deeply. MultipUcation takes place or new cells are added until the epithehal cell forms a mere shell around the included cells. The nucleus is, of course, much flattened out in the surrounding shell, but still surrounded by body protoplasm. One epithelial cell may include another within the abscess, and either the including or the included cell, or both, may have enclosed leukocytes or nuclear detritus. Epithehal cells sometimes enclose a large number of small spherules of uniform size which stain indistinctly. Their identity is uncertain. As mentioned, the dense infiltration of the cutis and rete in many places makes it difficult to separate the layers. This infiltra- tion extends into the fatty and muscular tissue. There is no distinct abscess formation in the cutis, no collection of polymor- phonuclear leukocytes to the exclusion of other cells, notably plasma cells; and in no place are the minute abscesses of the epidermis seen communicating with the cutis. Plasma cells, lym- phocytes, polymorphonuclear leukocytes, mast-cells, and eosino- philes constitute the mass of infiltration. In addition are newly formed blood and lymph channels, with marked congestion of the former. New fibrous tissue is forming at the periphery of the lesion. Elastic tissue has disappeared from the areas of most severe inflam- mation, but is still present rather freely in the peripheral portions. In certain places the polymorphonuclears predominate; healthy 64 Contributions to Medical Science connective-tissue fibers and cells, plasma cells, and granular ma- terial are intermingled. Mononuclears are scattered more diffusely than the polymorphonuclears and are not collected into decided groups. They are commonly more numerous in the midst of plasma cells. Eosinophiles are insignificant in number. The striking feature of the subcutaneous tissue is the number of mast- and plasma-cells. The latter occur in groups and columns and usually in close proximity to blood-vessels. This relationship is not constant, however. They occupy broad areas at times, associated with other cells, particularly mononuclear leukocytes, mast- and connective- tissue ceils. The plasma cells remain at points relatively remote from the foci of most intense inflammation. They are present everywhere in the periphery of the lesion, extend- ing at times in small groups into the muscular and fatty framework. Occasionally they are seen within blood-vessels where the plasma cell morphology is not always complete in all details. For example, it is not uncommon to find in blood-vessels cells with the typical plasma cell nucleus, but with a protoplasm which stains very faintly, or not at all, with alkaline methylene blue. One typical plasma cell was seen emerging through the vessel wall. In intimate association with plasma cells everywhere is newly forming fibrous tissue. Another conspicuous property of the plasma cell is the so-called hyalin degeneration. Mast-cells are thickly and widely distributed. Like the plasma cell, they are absent from intensely inflamed areas. A certain tj^e (leukocytic?) is found in close proximity to blood-vessels. Another variety (the connective-tissue cell type) occurs in the periglandular and perifollicular tissue, but is found also in the stroma of adipose and muscle tissue. Furthermore, a form inter- mediate between plasma- and mast-cells seems probable. This cell has a distribution similar to the type just mentioned. That the mast-cell may be endowed with motile properties is indicated by its occasional presence in the rete. Giant cells of the tubercular type are seen commonly in groups in the areas of most dense infiltration. They often contain from one to six or eight vacuoles, usually oval in form. It has not been possible to demonstrate positively that these spaces were formerly Oidiomycosis of the Skin and Its Fungi 65 occupied by organisms. From their almost constant oval form, however, this seems probable. Hundreds of sections have been searched in the vain hope of finding in the hardened tissue the organism which was cultivated from the fresh so readily. Single and combined stains, with diverse methods of fixation and differentiation, have been used. One would expect to find the causal agent, of course, where the fray is fiercest. The miliary abscesses may be taken as the cul- minative effort of resistance on the part of the tissues against the invading organism. Many extra- and intra-cellular forms have been found which might represent the organism. The most common is a spherical body, from one to three microns in diameter, which occurs in groups in phagocytic cells or free within abscesses. They stain with hematoxyhn, by Gram's method, with acid fuchsin, and a blue violet with polychrome methylene blue. When it is remembered that a small form of the organism was developed in cultures, and furthermore that the morphology of these organisms is variable, the temptation is strong to rest with the finding of such a form as mentioned above, particularly since it occurs where one expects to find the organism. However, the morphological evidence is not sufficient, as there is no distinct capsule, no evident budding, no mycelium. Moreover, the staining results apparently indicate that they are a degeneration product of nuclear substance. It was hoped that the organism as it appeared in animal would afford a clue to its appearance in human tissues. The kidney lesions in mice are very clear. Virtually they consist of a mass of delicate, short or long, segmented fungous filaments, and small round forms, which stain well with methylene blue. Similar forms could not be found in the human tissue. The idea occurred that by inoculating the organism subcuta- neously into a guinea-pig at several points and resecting specimens from all stages of the process — incipient induration, abscess for- mation and resolution — perhaps a means would be suggested of recognizing the organism in the lip ''tumor." Details are cited imder animal experiments. The experiment did not yield the needed help. The first piece of tissue resected, however, afforded a suggestion. It appeared grossly as a solid cellular mass, resem- 66 Contributions to Medical Science bling closely a cross-section of the lip tissue. It required the closest search to find a few scattered spherical organisms with typical structure, capsule, and all. Yet in a few days this same area broke down into a thick, yellowish pus, which was loaded with mycelium and spherical forms. This points very strongly to an intermediate unrecognized form, which may exist in tissue. Sections of the human tissue were treated with potassium hydrate solutions and digested with pepsin, but organisms were not revealed. Animal Experiments. — Guinea-pigs. i. An intraperitoneal inoculation with teased tissue. Death two months later, when a number of animals died from some obscure cause. Postmortem findings and cultures negative. 2. Teased tissue was placed subcutaneously in the right inguinal region; also the abdominal skin was scarified and portions of tissue were rubbed into the lesion. Five days later a sinus formed at the point of the subcutaneous inoculation, which discharged cheesy pus for a week and then healed. Neighboring lymph glands enlarged, and a temperature of 104° F. developed, which soon disappeared. The pus, mounted in KOH solution, showed both the spherical and myceUal forms of the organism. Cultures yielded besides the fungus a streptococcus, staphylococci, and a long baciUus. The scarification wound healed promptly. 3. One c.c. of a ten-day-old bouillon culture injected subcutaneously in the interscapular region. Temperature of 101° F. resulted. An induration formed, which subsided after five days without suppurating. The animal died from unknown causes, in company with guinea-pig i. The postmortem findings and cultures from organs were negative. 4. Twenty minims of a si.x-day glucose bouillon culture inoculated subcutaneously in the interscapular region. A moderate fever resulted, which subsided in three or four days. Pus formed in two days. It contained grotesque mycehal fragments and single adult and budding forms. Cultures yielded the fungus. Later three sub- cutaneous inoculations from glucose bouillon cultures were made in this animal simultaneously. The resulting lesions were resected at different stages. First. A deep, indurated mass, which developed in five daj^s after inoculation- On section it appeared homogeneous and grayish-white, rather anemic. Pus had not formed. Sections showed a central mass of leukocytes, erythrocytes, and connective- tissue cells, with occasional fragments of elastic tissue. Curious forms of phagocytosis or cell invasion were conspicuous. After much search, spherical and budding forms of the fungus were found in small numbers, but there was no mycelium. The presence of peripherally disposed granules, which stained well, had not been seen previously in the organism. Otherwise the structure corresponded to previous descriptions. Second. A suppurating lesion ten days old. The thick, whitish pus contained, in addition to spherical and budding forms, grotesque and imperfect mycelial frag- ments. In stained sections it was impossible to recognize anything but leukocytes, connective-tissue cells, fibrin, and debris in the abscess wall. Oidiomycosis of the Skin and Its Fungi 67 The explanation is not clear as to why the organism should be found in abundance in the pus and yet be unrecognizable in the stained section of the abscess wall. Third. The third focus discharged pus, and when this had nearly ceased the lesion was excised. Buried in the subcutaneous tissue was a mass composed of poly- morphonuclear leukocytes, endothelioid cells, phagocytic endothelioid and giant cells, degenerating red blood cells, nuclear and granular detritus, and fibrin. The phagocytic endothelioid cells were sometimes multinuclear, without presenting the appearance of giant cells. There was no characteristic arrangement of the nuclei, nor was the protoplasm granular. They contained polymorphonuclear leukocytes, other endothelioid cells, relatively minute spherical bodies staining intensely with hematoxylin and vacuoles. The minute spherical bodies are considered remnants of nuclear tissue, though the possibility is in mind that they may be small forms of the organism inoculated. Many are found extracellular and imbedded in a granular matrix. A capsule could not be demonstrated, and there was nothing fungoid in their arrangement. Rabbits, i. Subcutaneous inoculation of fresh tissue. Death after four days. Cultures yielded only Staphylococcus pyogenes albus. 2. One c.c, of a ten-day bouillon culture inoculated into vein of the right ear. The culture had been kept at the room temperature and contained abundant mycehum, budding cells, and a few degenerate forms. The temperature rose to 105.5° F. on the second day, returning to normal after four or five days. On the second day an induration 3 cm. long and i cm. broad formed in the course of the vein, beginning at the point of inoculation and extending centrally. After twelve days pus was discharged, in which degenerate forms of the organism were found. Cultures were negative. Mice.^ I. One-half c.c. of a three-day-old glucose bouillon culture inoculated subcutaneously. Died in thirty hours. Cultures of the fungus obtained from the kidneys. 2. A portion of a glycerin agar culture was suspended in bouillon and twenty minims of the mixture inoculated subcutaneously. The animal appeared sick for a few days, but recovered. Later, ten minims of a six-day-old glucose bouillon culture were inoculated. On the second day the animal died. Pure cultures of the organism were obtained from the blood and kidneys. 3. Portion of culture from a solid medium was rubbed into a scarification of the skin over the buttock. The animal died in two days. Postmortem, microscopic, and bacteriological examination negative. 4. Twelve minims of a twelve-day-old glucose bouillon culture inoculated subcu- taneously. Died on the second day. Organism recovered in pure culture from kidneys, blood, and spleen. Tissue changes in mice. — Except an unusual vascularity of the kidneys, gross changes were not noted. All organs were fixed in alcohol and imbedded in paraffin. Kidneys. — Acute granular degeneration of the tubular epithe- lium is present in all cases, but in varying degrees of intensity. In ' Ordinary house mice were used. 68 Contributions to Medical Science mouse 2 well-staining nuclei are rare. In mouse i this degeneration is much less marked. The chief interest centers around tubules and groups of tubules, which are unstained or faintly stained with hematoxyhn, but take up methylene blue deeply. They are granular masses and contain small spherules and short threads, which may extend into the interstitial tissue. Often what appear to be spherules are really cross-sections of threads. With a methylene blue stain, and an oil-immersion objective, the spherules and threads are recognized as part of a micro-organism. Larger spherical capsulated organisms are sometimes seen in the central part of an area, which correspond to the adult spherical cells s.een in cultures. The protoplasm of such cells often stains dimly. The predominating form is a segmented thread, not unlike bacilH in chain form, except for the large size. The segments are from one and one-half to four microns wide, and the length from two to four times greater than the breadth. The ends are rounded bluntly. Not more than six segments have been seen in a thread, and the longer chains taper to a narrower extremity. Often the terminal, and occasionally an intermediate, segment is oval in form. Isolated oval forms are common, but the spherical capsu- lated organism is seen less frequently. The latter are constantly larger than the former. Branching threads are not observed. With polychrome methylene blue, the periphery of both mycelium and oval forms stains deeply, leaving an unstained or faintly stained center. The large spherical forms present a thin, sharply stained periphery, which apparently represents the capsule. The earhest condition found is that of a few organisms in the lumen of a tubule, the epitheUal cells being very granular. In the larger foci no traces of kidney cells are apparent, the areas consisting of dense accumulations of the organism. There is virtually no inter- stitial proliferation, and a marked absence of leukocytes. No plasma- nor mast-cells are found ; only a thin shell of cells surrounds a focus, and these are of the endothehoid and lymphoid type. Heart. — Circumscribed areas of infiltration, separating and occasionally destroying the muscle fibers, are widespread. In mouse I no organism was demonstrated positively in these cell masses. In the remaining mice, adult and budding spherical Oidiomycosis of the Skin and Its Fungi 69 forms were numerous. Single adult forms measured from six to ten microns in diameter. No mycelium was found, though elongated single cells were fairly numerous. Lungs. — Only extreme congestion. The liver and spleen normal. White rat. Twenty minims of a four-day-old glucose bouillon culture inoculated intraperitoneally on November 27, 1900. No symptoms followed, and the animal is still living. Dogs. I. On October 27, 1900, four cubic centimeters of a four-day glucose bouillon culture were injected into the left saphenous vein. Prostration, fever, and emaciation followed. After a week gradual recovery set in. On November 28, 1900, four cubic centimeters of a one-month-old glucose bouillon culture were injected into the right axillary vein. Symptoms like those following the first inoculation developed, but of less severity. On December 3, 1900, while still sick, the animal was chloroformed. The axillary glands on the right side were very large, and there was mild general lymphadenopathy. They presented no soften- ing. No gross changes in heart, liver, and lungs. Spleen very hard and malpighian corpuscles are large and are shelled out easily. Kidneys are hard, and there are many subcapsular, white, scar-like points. Smears from the wound, which had been torn open by the animal, showed many budding fungus cells. None were found in smears from the viscera and the heart's blood. Inoculations from the heart's blood, lungs, liver, kidneys, and spleen produced no growths on suitable media. Microscopic examination of an axillary lymph gland showed new-formed connect- ive tissue and large masses of plasma cells. There were intracellular hyalin bodies, but no fungous cells were recognizable. The liver contained small areas of round cell infiltration, but the organism could not be detected. In the lung were small masses of new fibrous tissue, but no fungi. The white subcapsular areas in the kidneys were composed of masses of plasma cells, a few endothelioid cells, and portions of kidney tubules. Other plasma cell foci were scattered throughout the kidney, but no organism could be recognized. The heart and spleen appeared normal. 2. Subcutaneous inoculation of two cubic centimeters of a three-day-old plain bouillon culture. An induration developed, which subsided in five days. The autopsy, ten days later, showed no changes. Cultures from the viscera were sterile, and microscopically the organs were normal. Human. An area on the calf one-half inch square was scarified superficially and a portion of a growth on an agar slant thoroughly rubbed in. Soreness for three or four days resulted, when a healthy crust formed and perfect healing occurred. Morphology, etc. — The organism may proliferate by (i) budding of spherical or oval cells; (2) the formation of lateral and terminal conidia; and (3) by the splitting-up of the segmented mycelium into chains of "spores," yo Contributions to Medical Science The formation of endospores and the presence of a nucleus have been sought under many conditions. The protoplasm possesses inconstant morphological elements. It may be absolutely clear, structureless, or may undergo differentiation into apparently complex elements. In small buds and young isolated cells it is usually clear, and with all dyes stains evenly and deeply. In active full-grown cells it may possess one or more of the following differentiated elements: (i) granules, (2) vacuoles, and (3) a nucleus-like or spore-like (?) structure; or it may be absolutely clear and stain like the younger cells. There may be only one granule or several. In living cells some of them move actively in a vibrating manner through the protoplasm. Others are quies- cent. In stained mounts perhaps only one or two granules will take up the dye; or a small granular stained mass may lie in the center or at one side of the cell. It is common to find a deeply stained, irregular, semi-lunar or horse-shoe shaped mass, closely appHed to the cell wall, occupying one-half to three-fourths of the periphery. As these granules are not found in the buds, their relation to proliferation, if they have any, must be an indirect one. The nucleus- or spore-like structure does not occur in all active adult cells. It is often, but not uniformly, present in the parent budding cell. When the cell is not budding, this structure usually lies in the center of the protoplasm. In oval cells it Hes in the smaller pole. As budding begins, and during the process, it approximates the cell wall where the bud is to arise, and a similar body may or may not appear in the daughter cell. It may contain one or two highly refractive granules. The conditions of its appearance and disappearance have not been made out. The relation it bears to reproduction is not clear, as cells bud in which this substance is not seen. That it is a nucleus in the ordinary sense of the word cannot be accepted, as it is inconstant, pos- sesses no nuclear membrane, does not participate directly in cell division, and is not stained constantly with nuclear dyes. In using hematoxylin, very dilute carbol-fuchsin, and neutral red as vital stains, it colors more quickly and more densely than other structures; but in fixed films it could not be clearly differentiated Oidiomycosis of the Skin and Its Fungi 71 by staining methods. That it is a spore is also uncertain. It has not been observed to be hberated and to develop new cells; and it does not take part in a "free cell-formation," and lacks the structure demanded of spores. It is often, but not uniformly, present in mycelium. It occurs both in the segmented and the unsegmented threads, but is more common in the former, where they may have a chain-like disposition, each element being separated from its neighbor by the surrounding clear substance; or they may be irregularly distributed and assume different sizes. In general, this structure is present in actively proliferating cells (mycelium or budding cells), and absent in old, much vacuolated, involution forms. Vacuoles are found most of all in the large old involution forms, less often in the active adult cells, and almost never in young cells or buds. In involution cells they are often multiple, and occupy nearly all the intracapsular space, being surrounded only by a thin zone of granular protoplasm. Other old forms show absolutely no internal structure, only a hollow capsule being visible either in living or stained specimens. It was endeavored to produce endogenous spores by growth in distilled water, at low temperatures, high temperatures (41° C), on highly acid media, in potassium iodide solutions, and on a moist gypsum block. Very highly refractive globules, one to four in a cell, which sometimes resulted,^ were considered plasmo- lytic effects. In a hanging drop of dog's blood-serum, an immediate formation of mycelium begins. A spherical cell sends out a small filamentous shoot, which increases in length rapidly, becomes segmented, and acquires in irregular points a mildly refractive spherical or oval body, tinted a faint pink or orange. This is the nucleus-like structure spoken of, and is also formed in the spherical cells giving rise to the threads (Fig. 14, Plate 7). The mixture of spherical cells and mycelium is similar to that formed by the microsporon furfur. Resume. — The reasons for considering the fungus described the cause of the disease in this case are as follows: I . The chnical history suggests an unusual etiology. 72 Contributions to Medical Science 2. Voluminous cultures of the fungus were obtained from the deep portion of the "tumor" under aseptic precautions. 3. The histopathology of the tissue — intra-epitheHal abscesses, epithehal hyperplasia, and the character of infiltrating cells — places it with a group of diseases which have been proven to be caused by fungi similar to the one under consideration. 4. Portions of the tumor, introduced subcutaneously into guinea-pigs, produced an abscess in which was found the yeast obtained in cultures from the tissues. 5. Its variable morphology in animal tissues gives ground for the belief that in human tissues it may have assumed a form different from that observed in animals or cultures, and hence Ukely to escape detection by the means used. As an example of this variability : In mice myceUal forms were found in the kidneys and an oval and spherical form in the heart. In subcutaneous inoculation in guinea-pigs it was almost impossible to find the fungus in the densely cellular masses which immediately preceded pus formation, although quantities had been injected; yet a day or two later, when hquef action had occurred, there was no difficulty in demonstrating large numbers of the organism, principally as hyphae. Previous to liquefaction, the forms found were of the oval type. Case //.—Clinical history (Dr. Montgomery).— T. R., aged thirty-eight, married, by occupation a carpenter. On June 15, 1894, while tearing down an old kitchen, he injured the back of his hand. The place became covered with a crust, on the removal of which, several days later, he saw an elevated, angry-looking spot. Before the hand was quite well, a barber cut a wart from his neck, just under the chin. In a few days there appeared at this point a little elevation, covered by a crust. The clinical picture was largely that of a verrucous tuberculosis, though in places it suggested more a papillary epithelioma. The border was elevated from a sixteenth to a quarter of an inch, and was sharply defined, except for a narrow zone of bluish-red congestion, which for the greater part was inconspicuous. The growth for the most part was verrucous, situated on a very slightly Oidiomycosis of the Skin and Its Fungi 73 infiltrated base, though in places the latter was quite firmly indu- rated. Here and there the base was quite soft, and pus could be pressed out between the projecting papillae. The pus was examined under the microscope and a nmnber of culture media were inoculated. Tissue was removed from one border. The major portion of it was hardened in alcohol, but a small piece was teased in distilled water and used for the purpose of inoculating more media. Bits of this tissue were introduced into the subcutaneous tissue of two guinea-pigs and a rabbit. On the culture media no growths were obtained except those with pus organisms, while the results of the animal inoculations were abso- lutely negative. Microscopical examinations of the sections showed an unusual appearance, which will be described later. The diagnosis made at this time was that of a probable cutaneous tuberculosis, though the possibility of blastomycetic infection was considered. DiUgent examination of the stained specimens failed to show blastomycetes. At the meeting of the American Dermatological Association in 1898, I mentioned this case during the discussion of Dr. Shepherd's paper on "A Strange Case of Granuloma of the Face and Extremi- ties," the photograph of his case and his description bearing a striking resemblance to the clinical features presented by this case a few weeks previously. The man did not present himself again until August 26, 1899, eighteen months after his first visit. The disease had extended over the left cheek to the nose, had surrounded the left orbit and involved both upper and lower eyelids; the cicatricial tissue producing aversion of the lower lid. In places there are raised, smoother areas; in others the base beneath the papillae is soft, and pus can be expressed. Various culture media were inoculated with the pus and with tissue taken from an advancing border. No cultures of blastomyces were obtained, though the organisms were demonstrated later in typical forms in pus and tissue which had been subjected to the action of a weak solution of potassium hydrate. The man again disappeared from view for four months, at the end of which time he applied for admission to the Illinois Eye and 74 Contributions to Medical Science Ear Infirmary. To Dr. Dodd of that institution, who came to consult me regarding the man, I am indebted for further oppor- tunity to study the case, and for carrying out treatment with the iodide of potassium, imder the influence of which, in large doses, the progress of the disease has been arrested, and the verrucous growth is being replaced by the characteristic smooth red scar. Fresh Tissue. — Tissue was taken for cultures February 8, 1900. A portion of this tissue, teased and mounted in 25 per cent potassium hydroxide solution, shows typical blastomycetes proliferating by budding. These forms measure about twelve microns in diameter. Besides the large cells, another form is seen, which averages 3.75 microns in diameter. They are recognized as blastomycetes or as a similar organism because of a distinct capsule, a central protoplasm, and proliferation by budding. The capsule is deUcate and structureless. Very fine granules are seen in the protoplasm, and no nuclei or vacuoles are observable. The budding process is identical with that seen in the larger forms. They exhibit a marked tendency to form in chains, some of the segments being oblong or cylindrical, imitating the segments of mycelial threads. Chains containing from four to six segments were observed com- monly, none, however, possessing more than eight. These forms were observed in a subsequent examination of teased tissue, at no time being seen in the pus of the miliary abscesses. Cultures. — On August 26, 1899, culture media of potato, blood- serum, and glucose agar were inoculated both from teased tissue and from the pus obtained from miliary abscesses. The results were negative. The diflferent staphylococci only were obtained. On February 8, 1900, four plain agar, four glycerin agar, and two plain bouillon tubes were inoculated with verrucous tissue, which had previously been teased finely in bouillon. One half of these tubes were kept in the incubator and the other half at room tem- perature. Very profuse growths of staphylococci, a few colonies of streptococci and saprophytic bacilli were the result of these inoculations. The tubes were kept under observation for two months and no growth of blastomycetes developed. Animal Experiments. — On August 26, 1899, a guinea-pig was inoculated intra- peritoneally with a portion of the verrucous tissue, and at the same time the abdomen Oidiomycosis of the Skin and Its Fungi 75 at one point was scarified and bits of verrucous tissue rubbed thoroughly into the lesion. This animal died in two days. The autopsy showed peritonitis, and cultures from the peritoneum, viscera, and the blood resulted in a growth of staphylococci. A rabbit on the same day was inoculated subcutaneously with a portion of the teased tissue. A mild reactive inflammation was the result at the seat of inoculation. This induration subsided gradually without the formation of pus. Cultures of the indurated nodule gave no results. On the same day a white mouse was inoculated with a piece of teased tissue beneath the skin of the back. The mouse died in five days of staphylococcus septicemia. HiSTOPATHOLOGY. — For histological purposes, two portions of tissue were removed from the periphery of the diseased area, each containing a portion of adjacent healthy skin. The tissues were hardened and fixed in absolute alcohol, and imbedded in celloidin. The microscopic pictures of the two pieces of tissue are practically identical. Sections were stained with hematoxylin and eosin, with polychrome methylene blue, Gram-Weigert method, acid orcein, and with carbol-fuchsin for tubercle bacilli. The surfaces of the sections were covered with accumulations of loosened horny material and red blood corpuscles. Gram's method showed many cocci among the loosened homy lamellae, and some buried more deeply in the horny layer. The hair follicles were accentuated in their width. There was a marked irregularity of the surface, the homy layer in places dipping down deeply, and in these depres- sions occurred masses of homy lamellae. In the diseased portion, the horny cells showed the remains of nuclei. The deeper horny cells had nuclei uniformly. The stratum lucidum was not demon- strable in the diseased portion. The stratum granulosum, however, was very conspicuous, except where abscesses actually communi- cated with the surface. It averaged from three to six cells deep. The rete showed extensive proliferation, long, coarse, finger-like projections dipping down deeply into the cutis, almost to the subcutaneous tissue. These processes were greatly deformed, at one place being very delicate, constricted, and almost separated from the surface rete, and immediately below perhaps enormously enlarged from accumulations of leukocytes. A columnar basal layer was quite well maintained throughout all the gyrations. The individual rete cells possessed deeply staining nuclei and granular protoplasm, which took up eosin rather intensely. The prickles stood out prominently. Many of the nuclei were shrunken, allow- 76 Contributions to Medical Science ing the formation of a clear space between the nucleus and the cell protoplasm. In certain portions small foci of rete cells appeared necrotic. The sizes of the individual cells were subject to consider- able variation. The deeper cells were quite large, conspicuously granular, and the intercellular spaces widely dilated. Many partial whorl formations were seen. These whorls usually did not contain a demonstrable hornified center. Infiltrating cells, largely polymorphonuclear leukocytes, were very abundant in the inter- epithehal spaces, and in addition there were a relatively small number of mononuclear leukocytes, and occasional eosinophiles. In some places the leukocytes accumulated to form small abscesses. The rete cells immediately surrounding the abscesses have been compressed and flattened, and partially hornified. This went so far that the surrounding cells gave the appearance of a limiting abscess membrane, composed of fiber-like lamellae, the cells of which largely retained their nuclei. Cutis. — The papillary layer had lost its typical structure and form. There were dense accumulations of small cells, and the corium was exceedingly vascular from dilated and newly formed blood-vessels, lymph channels, and spaces. The fibrous and elastic tissues were broken up and separated by infiltrating cells. In many places the accumulation of leukocytes was so dense that all normal tissue was either replaced or obscured. The infiltration extended as far as the subcutaneous fibrous and fatty tissue; and even into these, columns of infiltrating cells follow the blood-vessels. The stroma of the fatty tissue was infiltrated in places, and the fixed cells were in a state of prohferation. The infiltrating cells were polymorpho- and mono-nuclear leukocytes, plasma cells, and very few eosinophiles; also a relatively large number of mast- cells and of newly formed connective-tissue cells. Of these the polymorphonuclear leukocytes formed by far the greatest number. They were somewhat irregularly distributed, in one place being so densely crowded as to constitute small abscesses, in others less densely and evenly distributed. The plasma cells were unevenly distributed, their tendency being to shun the foci of most acute inflammation. They were collected in masses at a short distance from the abscesses, however; in general, where the process was Oidiomycosis of the Skin and Its Fungi 77 subacute. There was nothing peculiar about the plasma cells in this case to distinguish them from the classical cell of that name. They were not gathered in masses to the exclusion of other cells, leukocytes, mast-cells, or connective-tissue cells, but they did occur in loosely constructed masses, mostly near the blood-vessels; hence certain sections showed these cells in the formation of columns. Giant cells were numerous and occurred in noticeable groups. The miliary abscesses. — In the intra-epitheHal abscesses were, besides the leukocytes, loosened epithehal cells, the forms of which vary greatly; a certain amount of nondescript debris, fibrin, and the organism. There were occasional plasma cells. For the most part the nuclei of all the cells stained distinctly, the number of broken-up cells being small. The parasites were circular or sHghtly oblong, existing singly, in pairs, or in clumps of three or four. Typical budding was common. The individual spherical form consisted of a doubly contoured capsule, which stained inconstantly with methylene blue, and a central protoplasm staining a lighter blue and separated from the capsule by a con- centric relatively clear zone. When they occurred in pairs, there were certain differences between the elements making up the pair. One of the cells may have a thicker, more deeply staining capsule than the other, and a larger area of protoplasm. Distinctly vacuolated forms were seen very rarely in this tissue. A condition of interest existing in many cells is that of large granules arranged peripherally in the protoplasm of the fungous cell. The number of these in different cells was subject to considerable variation. As many as twenty-four were seen in a single cell, the usual num- ber being eight or ten. The budding seen in this case differed in no way from that described by previous observers. The number of organisms was relatively few and they were seen only occasionally outside of true abscesses. They were not seen within tissue cells. A few were found in the granulation tissue of the cuts. Case 1 11.^ — CHnical history. — Mr. F. S. S., age 33 years, by occupation a farmer, resident of Elburn, 111., applied for treatment » A brief report of Cases III and IV has been made by Hyde and Ricketts, Jour. Cutaneous and Genito-Urinary Diseases, 1901. 78 Contributions to Medical Science December 15, 1899. His family history was satisfactory. For three years previous to the present date he had been engaged every autumn in handHng a threshing-machine. His general appearance was that of a man in sound health and with good habits. The present cutaneous affection began about two years ago. When examined, the region involved was recognized to be the left lower lid and the parts adjacent in both cheek and temple. Cultures. — Mounted in 30 per cent KOH solution, pus from the miliary abscesses shows a multitude of ordinary pus cells, which soon disintegrate, and in addition cells of another nature, which resist the action of the alkaH. The latter are spherical in form and arranged singly, in pairs, or, more rarely, in groups of three or four, the individual cells being united. Fully developed cells have an average diameter of twelve microns. The structural features are: an external, highly refractive, homogeneous, doubly contoured capsule, having a thickness of from one-fourth to one micron; a "clear zone" internal to the capsule, presenting no structure and not constant; and, finally, a central protoplasmic substance, which is sharply separated from the clear space by a delicate membrane. In the protoplasm is a varying amount of rather fine granules, fat droplets, and occasionally vacuoles, from one to three in number. No nucleus is seen. Budding forms are numerous. The bud is a projection from the mother cell and has the same three principal structural elements, each being continuous with the corresponding parental layer. Both capsule and clear space become thinner as they pass into the bud. The daughter cell reaches approximately the size of the parent before it separates completely, the capsule gradually becoming heavier. Isolation of the cells is accomplished by the capsule uniting through the point of constriction between the cells. No vacuoles are seen within daughter cells so long as they retain connection with the mother cell. The same organism is found in teased verrucous tissue mounted in KOH solution. After two fruitless attempts the organism was obtained on ordinary media inoculated with pus from the small abscesses and Oidiomycosis of the Skin and Its Fungi 79 with teased verrucous tissue. About half the tubes were con- taminated with a few colonies of a white staphylococcus. Agar-agar. — The growths on plain-, glycerin-, and glucose-agar are identical in appearance and develop more slowly on the first than on the last two. Two entirely different appearances may result, depending, so far as observed, on whether the culture is kept in the brood-oven or at room temperature. Cultures in the brood-oven: at the end of twenty-four hours a broad, thin, coarsely granular, whitish growth occurs along the stroke. In twenty-four hours more it doubles in proportions, and the granular condition develops into relatively large, prominent irregularities. The color is grayish- white and the surface moist. At the same time a light, hazy down-growth of delicate fibers penetrates the depth of the medium. The heaping-up on the surface progresses rapidly. In the center the growth, instead of being purely granular, comes to resemble a heap of tangled earth- worms, the separate coils standing out distinctly and separated from adjacent coils by irregular clefts. As one passes from the lower moist surface of the medium to the upper, somewhat dried portion, the vermicular appearance gradually passes to that of the coarsely granular. The lateral growth is made up of delicate fibers like those penetrating the medium vertically. It does not rise above the level of the medium, but is intimately associated with it. The surface presents a dull, opaque, frosted white appear- ance, finely granular and marked by extremely fine rays, originating in the central portion. By reflected light the peripheral limitation seems quite abrupt, but by transmitted light an extension beyond this border is seen in the shape of branching, mossy, tree-like rays of great delicacy. The growth into the depth of the medium progresses steadily until the transparency of the latter is destroyed. At room temperature: the growth does not rise above the surface of the medium conspicuously, but becomes intimately incorporated with it. At first the surface appears merely wrinkled. The folds increase in size rapidly, and there are corresponding depressions, and a positive consumption of the medium takes place. The surface finally resembles a piece of crumpled cloth. The medium thus impregnated with the fungous growth may be broken 8o Contributions to Medical Science into soft, brittle fragments, which, when smeared over the surface of a fresh tube, give a uniform growth of the organism. The color is grayish-white. Maltose agar. — The growth here rises above the surface in rounded prominences, and the centers of colonies are depressed. Threads penetrate the medium deeply. The surface is covered with short upright spikes, which do not appear like true aerial h}q)hae. Microscopically, they are composed of loosely segmented threads, the segments being elongated or spherical. Often long segments lie closely applied side by side, phalanx-hke. Pink nucleus-like bodies are seen in many. In addition, numerous cells are more or less filled with clear, highly refractive globules (Fig. i8a, Plate 9). There are in addition finer non-segmented threads, which usually contain no nucleus-like structure. The threads penetrating the medium have a similar structure, and give off a few conidia, which produce buds. +70 Oxalic acid maltose agar. — The growth is less pronounced than on other agars, and is less intimately incorporated with the medium. The microscopic findings are the same as on maltose agar. (Other organisms studied do not grow so well on this medium.) The reverse surface of agar slants shows no pigmentation. Glycerin agar plate. — In twenty-four hours minute white dots appear, which grow to a millimeter in diameter in twenty-four hours more. The naked-eye appearance is that of dehcately chased, almost circular crystals. With low magnification, the colonies are seen to consist of a dense central area, from which threads shoot out radially, growing less numerous as the periphery is approached; are loosely segmented, and produce false, loosely attached branches, coming off at nearly right angles. There are no lateral or terminal conidia, and the threads are slightly tortuous. The deep colonies show the same structure as those on the surface, and are of equal size. There is no aerial growth. The appearances were similar on glucose agar plates. Glucose agar stab. — A finely granular development occurs along the stab, from which delicate rays extend laterally into the medium. Deeply along the line of puncture the growth is Oidiomycosis of the Skin and Its Fungi 8i less pronounced than near the surface. The surface of the medium is soon covered with a thick growth, presenting characteristic irregularities. Glucose agar anaerobic stab. Gelatin anaerobic culture. \ Bouillon gelatin stab (aerobic). — Same appearance as in agar stab. The lateral rays are seen more distinctly on account of the transparency of the medium. The surface growth is much wrinkled, and gradual liquefaction takes place. Plum-decoction gelatin is Hquefied readily. Ox-blood serum. — Development proceeds more slowly than on the agars. The surface growth is not so vermicular, though this tendency is apparent. The lateral portion is present also, but the fine structure cannot be seen. Potato. — Development is rapid. In a week's time the whole surface is covered with a light fawn-colored growth, about one- sixteenth of an inch deep. The surface presents a densely packed vermicular structure. At the upper portion, where the potato has dried, the growth is powdery white. Bouillon. — Identical appearances were produced on plain and on a 2 per cent glucose bouillon, but development is more rapid on the glucose. A sediment is seen at the end of twenty-four hours. In another day particles appear on the surface and adhering to the tube. The surface growth increases until it is a soHd floating layer, which, if broken up by shaking, sinks in large flakes. If the tube is undisturbed, the growth at the bottom also becomes dense, and when agitated breaks up as the surface growth does. The microscope shows a great deal of granular material and broken-up cells, besides chains, threads, and isolated spherical forms. The top growth consists of densely interwoven mycelial threads, which bear conidia. There are a few spherical budding cells, and a gelatinous and granular ground-substance. In maltose bouillon the mycelial growth predominated. There were a few spherical budding cells. The pink nucleus-like structure occurred often. In sugar-free bouillon the threads showed little tendency to be segmented. 82 Contributions to Medical Science From agar slants which grew according to the second method, fragments of the impregnated medium were introduced into glucose bouillon tubes. Around each fragment a light fluffy tuft grew to considerable dimensions. Each consisted of a densely woven mycelium. Slowly the large flakes formed as described. Litmus milk. — Growth is slower than in bouillon. A large mass appears on the surface. In three or four days a slight reddish tint appears in the medium, and after a week pronounced acid reaction is noted. Hanging-drop culture in glucose bouillon . — Growth occurs readily at room temperature. A characteristic feature is the formation of isolated clumps of a tortuous mycelium, with almost no tend- ency to shoot out into the surrounding medium. The borders are fringed with crooked, tortuous threads, projecting for a short distance, the segments being of an irregular form. The impression is that of a tangled mass of earth-worms, the microscopic appear- ance here suggesting the gross appearance on solid media. Myce- lium originates from individual cells or as a prolongation from already existing threads, and is of a branching nature. A long, homogeneous, translucent, tapering, encapsulated thread is sent out from a single cell or chain of cells. This thread may be suffi- ciently long to accommodate from ten to twenty segments, which may form successively, or one or two well-formed segments may appear at the middle point of a homogeneous thread, the remaining portion showing no differentiation at the time (Figs. 9 and 10, Plate 6). Gradually more septa are formed. The cells grow or swell so that constrictions occur at the point of division and the protoplasm becomes granular. Eventually the segments of a thread separate. Many are dormant and when watched for days show no change. Others start new threads. Many of the granules may be spoken of as ''protoplasmic," and some seem fatty; but certain others appear of another nature. Cells containing these granules increase in size to from fifteen to twenty-five microns, are globular or oblong, and acquire highly refractive globules, the largest of which are a fraction of a micron in diameter. These cells disintegrate. In the immediate vicinity there develop in a Oidiomycosis of the Skin and Its Fungi 83 few days large numbers — from fifty to a hundred — of young cells. They are oblong, lanceolate, homogeneous, and soon reach a length of from two to five microns, with a diameter of one-half of that extent. They gradually become separated, increase in size, assume adult characteristics, and some were seen to sprout new threads. These are the same refractive bodies seen in mounts from maltose agar tubes. In a hanging drop of 50 per cent iodide of potassium solution, many cells become filled with similar bodies. Cultures from the drop were sterile after two weeks. The organism was studied in a hanging drop of dog's serum. Proliferation occurred solely by the segmentation and breaking- up of mycelium into elongated, oval, or spherical cells, which elongated to form new threads. Growth was very rapid. The nucleus-like structure and many highly refractive spherules ap- peared in numerous cells. Fermentation tests. — Glucose solutions are fermented readily; lactose, maltose, and saccharose solutions are not affected. Alco- hol, carbon dioxide, and acetic acid result in the process of fermen- tation. Potassium iodide cultures. — Vigorous development occurs in one and five per cent solutions in bouillon, with no alterations in the morphology of the organism. A considerable amount of brownish-black pigment forms in the bottom of the tube. Indol. — None is produced in the Theobald Smith sugar-free bouillon. HiSTOPATHOLOGY.— On two separate occasions, under freezing anesthesia, verrucous tissue, containing miliary abscesses, was removed, fixed in bichloride of mercury, and imbedded in paraffin. The two specimens show the same changes. Horny layer. — Absent in small foci, exposing the rete. In other places it is greatly heaped up, forming plugs which fit into depres- sions of the rete and rise above the surface level. Composed largely of flattened nucleated cells, in the midst of which are polymorphonuclear leukocytes, bacteria, and debris. The stratum lucidum is not observed in the diseased area. 84 Contributions to Medical Science Granular layer. — Present except where desquamation is extreme. Where the rete is thick the granular layer acquires proportionate thickness. Rete. — The enormous proHferation and down-dipping charac- teristic of the diseases are conspicuous. Conditions of the rete cells differ Uttle from those in the preceding case. Edema is more marked. Individual cells are large, their nuclei stain well, the prickles are distinct and stain faintly with protoplasmic dyes. There are isolated examples of epithelial cells including polymor- phonuclear leukocytes. These latter cells are not so uniformly and densely distributed as in the first case. No plasma- or mast- cells are found in the rete. The intra-epithelial abscesses are larger than those of the first case, reaching at times the diameter of a milhmeter, and they frequently connect with abscesses on the cutis. They contain polymorphonuclear leukocytes, epithelial cells, granular detritus, and the organism peculiar to the case. From two to four giant cells may be found in the section of an abscess, commonly near the periphery. The nuclei are arranged peripherally. Free epithelial cells are found less frequently than in the previous case, and they do not present so many examples of phagocytosis or cell invasion. They degenerate much more readily, and usually are seen as acidophilic granular masses. From one to several organ- isms may be seen in a single section of an abscess; on the other hand several sections may be examined without encountering any organism. Cutis. — Normal structures are obhterated by a mass of infiltrat- ing cells, among which polymorphonuclear leukocytes and plasma cells predominate, the latter being present in smaller numbers than in the first case. They have the same relation to the polymor- phonuclears and to new-forming connective tissue. Hyalin bodies are met with less frequently. The polymorphonuclears form distinct subcutaneous abscesses, and they are also diffusely dis- tributed in smaller numbers. In the abscesses are, besides the leukocytes, portions of connective-tissue fibers, an occasional elastic fiber fragment, proliferating connective-tissue cells, granular Oidiomycosis of the Skin and Its Fungi 85 material, giant cells, and the organism. Mast- and plasma-cells are found rarely in the periphery of the abscesses. Mast-cells of the types already mentioned, and in like dis- tribution, are numerous. Giant cells of the tuberculous type are relatively numerous, and are found not only in the abscesses, but also in connective tissue less markedly infiltrated. None are found containing the organism. The tubercle bacillus could not be found in properly stained sections. The organism. — It is found both in the epithelium and cutis. In the former it is surrounded by varying numbers of leukocytes, which often constitute small abscesses. In the cutis it is found both in distinct abscesses and where there are fewer inflammatory cells. The characters are the same, whether found in the epithe- lial layer or the cutis. The typical form is spherical. Occasionally an elongated cylindrical form is seen, either alone or with a bud or large spherical form attached to one end. The average diameter is twelve microns. Budding as described by previous writers is common. A limiting, homogeneous, translucent, doubly con- toured capsule, a clear space internal to this, inconstantly seen, and a central protoplasm are the principal structural details. In methylene blue preparations the capsule usually stains deeply, and often its outer surface is irregular, giving the impression of adhering foreign matter. The protoplasm is granular, vacuoles occurring in occasional cells. In a pair, the protoplasm of one cell may stain deeply with methylene blue, and that of the other not at all. Other nuclear stains are treated with like indifference. With all stains one finds inconstantly in the periphery of the central protoplasm a number of minute spherical bodies. Bodies similar to these are frequently seen free, but their identity is uncer- tain. Old cells in which there is no clearly recognizable central protoplasm do not stain. With acid orcein, the capsule and a limiting protoplasmic membrane stain deeply. The protoplasm has but slight affinity for hematoxylin. Fuchsin is taken up moderately, but not in a distinctive manner. With carbol-toluidin blue the same results are obtained as with methylene blue. No constant nor distinctive staining affinities have been found. 86 Contributions to Medical Science Animal Experiments. — Guinea-pigs. i. Intraperitoneal inoculation of teased verrucous tissue in normal salt solution. The animal died after two days, of staphylo- coccus septicemia. 2. Intraperitoneal inoculation of twenty minims of glucose bouillon culture grown in the incubator for one week. The inoculation was without apparent results. 3. Twenty minims of the culture used in guinea-pig 2 were inoculated beneath the abdominal skin. An abscess developed after five days. It ruptured sponta- neously. The organism was found in the pus mounted in KOH solution, and cultures on agar yielded the fungus inoculated. 4. Four centimeters of a ten-day-old glucose bouillon culture inoculated beneath the skin of the back. The animal ran a temperature for three days, appeared sick, and refused food. However, abscess formation did not occur, and complete recovery took place. 5. Subcutaneous inoculation of two centimeters of a sugar-free bouillon culture. Four days later was found dead. The autopsy showed pulmonary tuberculosis. Microscopic examination corroborated the gross findings. Cultures were sterile. Mice. I. Twelve minims of a glucose bouillon culture injected subcutaneously. Recovery apparently complete. 2. Two centimeters of a glucose bouillon culture inoculated subcutaneously. Died after two days. Cultures yielded pus cocci. The organism was not discovered in sections. Dogs. I. November 28, 1900. Three centimeters of a five-day-old glucose bouillon culture inoculated intravenously. Killed one month later. Cultures were sterile. No gross pathological conditions except a few small subcapsular, scar-like lesions of the kidneys. Under the microscope these were miliary nodules, consisting of new connective-tissue- and plasma-cells. No fungus cells recognizable. No tubercle bacilli present. Other organs were normal. 2. December 3, 1900. Four centimeters of a five-day-old glucose bouillon culture injected subcutaneously. No effect. Animal living (March 2, 1901). Rabbits, i. February 13, 1901. Died five minutes after an intravenous inocula- tion. Pulmonary embolism. 2. February 14, 1 90 1. Two centimeters of bouillon suspension of a culture on -|-io oxalic acid maltose agar, intravenously. Died two days later. Enlarged, pale kidneys. Cultures sterile. In the lungs were a few small foci of epithelial prolifera- tion, and infiltration with lymphocytes and polymorphonuclear leukocytes. The liver and kidneys were fatty. Other organs normal. White rats. i. November 11, 1900. Intraperitoneal inoculation of one and one-half cubic centimeters of a four-day-old glucose bouillon culture. No symptoms followed. The animal is still living (March 3, 1901). 2. February 14, 1901. Subcutaneous inoculation of two centimeters of sugar- free bouillon culture. No effect. Animal living (March 3, 1901). Summary of Biologic and Pathogenic Properties. — Young colonies are grayish, round, elevated, granular, and early become incorporated with agar and blood-serum, and gelatin media. Colonies fuse readily, and eventually form an elevated, deeply and coarsely wrinkled surface, of a dirty gray color. The growth Oidiomycosis of the Skin and Its Fungi 87 is teased apart with difficulty. The posterior surface of the slant does not have the rich golden-brown color often seen in cultures of fungi. In gelatin stabs lateral rays are formed; no liquefaction. In liquid media a membranous top growth forms and coherent flakes and tufts develop as a sediment and adhering to the tube surface. Milk is coagulated and acidified; no indol-formation. Mycelium always predominates, but it is accompanied by a varying amount of budding on the part of spherical or oval cells. No demonstrable nucleus. The formation of endospores in one observation seemed fairly positive; but further study of this change in the protoplasm indicates rather a plasmolysis, which does not necessarily involve the death of the cell. A study of Figs. i8a and iSb, Plate 9, shows the method of development of these bodies. The objections to their being considered as ascospores are (i) the large number which may be found; and (2) they have never been observed to grow into spherical cap- sulated cells or fibers. The organism reproduces, then, by: 1. A segmented branching mycelium. 2. The division and separation of mycelial segments, producing spherical cells or "spores." 3. Lateral conidia (for a time). 4. Budding of spherical and oval cells. 5. Endogenous spore-formation (?). Subcutaneous abscesses may form in guinea-pigs and rabbits. Dogs and mice are not susceptible. Its pathogenic effect on animals is weak. An inoculated scarification in man healed promptly. Case IV. — Clinical history. — Mr. F. M., a resident of Chicago, presented himself at the clinic December 23, 1899. He was forty-seven years of age, single, with good general history. His present disorder dated from November, 1897, when the skin of the right forearm, three inches below the elbow-joint, was scraped when in contact with a wooden truck. The resulting wound covered itself with a crust, which later fell and was suc- ceeded by another and yet another, no healing resulting. The patch, when examined, was found to be in the form of a 88 Contributions to Medical Science distinctly defined band, about six centimeters in greatest width, completely encircling the arm from in front backward and meeting posteriorly by a narrow line of contact between the two rounded borders of the fillet, as if the process had advanced from the anterior to the posterior surface. The affected area was of a dull, reddish hue, moist throughout, secreting a sero-purulent fluid, granulating, and the source of a disgusting fetor. Its surface was irregular, and there were islets of pinkish hue between the elevations and depres- sions of the florid granulations visible. Here and there thin patches of false membrane had formed. At the present date the arm of the patient is better, and practi- cally healed after treatment with iodide of potassium. Examination of the Fresh Tissue and Pus. — On December 23, 1899, large villiform processes were teased and examined in 30 per cent KOH solution. The patient had received no potassium iodide. The specimen was kept under observation for three days, at the end of which time the forms to be described were relatively intact. Bodies possessing the characteristics ascribed to blastomycetes were abundant. The typical solitary form was spherical, and had a doubly contoured, structureless, clear, highly refractive capsule, and a central protoplasm which was finely or coarsely granular, and sharply differentiated from the capsule. Often certain clear, structureless, highly refractive bodies were suspended in the protoplasm. These were considered vacuoles. The number in a single ceU varied from one to several, but they were often absent; fat globules common. Certain other granules of nearly constant size were observed frequently, usually arranged around the periph- ery of the cell. They may have been spores. Nothing was seen of a distinct nucleus. One chain of four organisms was observed, each united to its neighbor by a portion of capsular substance, the protoplasm of each being enclosed in its own cap- sule. Organisms with two buds were seen. A sort of abortive budding occurred occasionaUy, the large, cylinder-like body having a slight mesial constriction. Some of the forms exhibited a cup-Uke depression of the capsule, indenting the protoplasm. A "clear zone" was ijot always demonstrable. Many of the Oidiomycosis of the Skin and Its Fungi 89 organisms were degenerating, even in perfectly fresh preparations. At first the capsule becomes marked with meridian-like lines, which increase in number and prominence, and are seen to involve the whole thickness of the capsule, eventually appearing as lines of fracture. The fragments finally become separated, forming a layer of debris around the protoplasm. The latter substance in the beginning of the process acquires coarse granules, which merge, and as this takes place the volume of the mass is contracted. The protoplasm finally is converted into a perfectly clear sub- stance. The final dissolution stage could not be found. The diameter of the solitary adult form averaged from twelve to fourteen microns; none exceeded sixteen. Pus taken from small abscesses showed organisms of the same character. On January 3, 18, and February 10, further examinations of the pus and fresh tissue in potassium hydrate solutions were made. The findings recorded were duplicated, and, moreover, what is considered as a small form of the organism was seen. It may be that this was an involution form induced by the potassium iodide administered. The identity of this form rests on a purely mor- phological basis, because of the failure to obtain cultures. They measured from four to six microns in diameter, had delicate but distinct capsules, multiplied by budding, and were prone to form short threads of four to six or eight segments. The protoplasm was finely granular and showed no vacuoles or nucleus. Forms intermediate between this and the larger were not seen. This small form was found only in teased preparations of the tissue, not in pus from miliary abscesses. On February 27 were found, besides the typical adult form, a few large individuals of twenty to thirty microns in diameter. The capsule was only sUghtly thicker than that of the average form. The center was made up of granular protoplasm, in which were suspended bodies of varying degrees of refraction, measuring from one to four microns in diameter. They show no definite structure. A further examination was made on May 2, 1900, when he had been under potassium iodide for several months. Extensive 90 Contributions to Medical Science healing had taken place, leavmg a thick, smooth scar, which still contained a few mihary abscesses. These yielded the organism in its ordinary spherical form, exhibiting no marked degeneration, as one might expect to find in such a rapid subsidence of the process. However, they were much fewer. Cultures. — Culture tubes of the ordinary media were inocu- lated many times, both from the teased verrucous tissue and from the pus of the miliary abscesses. Numerous ordinary bacteria grew on the tubes, but an organism resembling the usual "blas- tomyces" or the form seen in the tissues never was obtained. Control examinations were always made at the same time and from the same material used in inoculations. It was found invari- ably that the pus or the teased tissue experimented with contained numerous organisms. Each time inoculations were made, how- ever, there grew on one or two tubes one or more colonies of a thread-like fungus, which penetrated the medium, the center of the colony showing a small conical elevation, but producing no true surface growth. Film preparations showed threads with small lateral pedunculated conidia. Transplantation on other media produced no growth, and further study could not be made. Maltose-agar was not used at that time. The early appearance of this fungus was exactly like that in a more recent case in the light of which it seems probable that we were dealing with the specific organism. Animal Experiments. — Fresh tissue teased in bouillon was used. Guinea-pigs. i. Intra-abdominal inoculation. Death in three days. Peri- tonitis. Microscopically, mild fatty degeneration of the liver and kidneys. 2. Subcutaneous inoculation and scarification. The latter healed promptly; the former was swollen for a few days, but healed without suppuration. Death a month later, preceded by emaciation. Postmortem findings normal. Macro- scopically, cloudy swelling of the organs. 3. Subcutaneous inoculation. Induration slight, which disappeared slowly. Death three weeks later. There were no gross findings. Microscopically, only cloudy swelling of liver and kidneys. Owing to the absence of special tissue changes, and to the fact that several other animals died about this time from obscure causes, it seems improbable that the inoculations were responsible for the death of the last two guinea-pigs. Histopathology. — Four specimens were removed at different dates, representing various conditions. Oidiomycosis of the Skin and Its Fungi 91 The low-power view is that of a promiscuous mixture of epithe- lium and altered papillary tissue. The latter is in the form of tortuous processes and islands, always sheathed by epithelium. The horny layer is excessive. It follows the irregular pro- jections and depressions of the papillae, and furthermore in the depressions forms loose masses and whorls of flattened cells, many of which are nucleated. Among them are occasional polymor- phonuclear leukocytes and a few red blood corpuscles. Large masses and layers of cocci, which stain by Gram's method, are present. Acid fuchsin shows many variously sized hyalin bodies both within and between the cells. Rete.— It follows the meanderings of the papillary tissue as a thick sheath. Its cells are large, vesicular, their nuclei stain well, and the prickles are conspicuous and often widely separated. There is no stratum lucidum, and only traces of a granular layer are found. Whorl formation, with partial comification, is seen often in the depth of the rete. These whorls contain leukocytes in the center and between the cells, but none have been found containing organisms. Mitoses are common. The basal layer retains distinct columnar formation, but it is often subject to distortions. Large circular spaces between epithelial cells speak for a marked edema of the rete. The number of infiltrating leu- kocytes in the rete is small compared with other cases. The polymorphonuclear cells predominate, but mononuclears and eosinophiles are also seen, the latter in conspicuous numbers. Red blood cells are relatively numerous between the rete cells. The leukocytes are gathered into small masses in many places to form intra-epithelial abscesses, which are smaller than in the second case, but usually larger than those in the first. They contain, besides the leukocytes, loosened epithelial cells of types already mentioned, occasional eosinophiles and red blood cells, the organisms, and the usual granular material. No plasma- or mast-cells are seen within the abscesses or free in the rete. Often there is a group of organisms with only a few surrounding leukocytes. The conditions of the epithelium within and surround- ing the abscesses are like those described in connection with the two preceding cases. 92 Contributions to Medical Science Cutis. — In the active verrucous condition, the epithelium penetrates the cutis promiscuously, but there are no isolated masses or whorls. No papillary layer as such is to be distinguished. The blood-vessels are widely dilated, containing, besides the red and polymorphonuclear cells, a relatively large number of eosino- philes. No plasma cells are seen in the vessels. In the superficial cutis are many small hemorrhages. These are so numerous that in a good eosin stain the field appears quite red. The distribution of infiltrating cells and of particular varieties is irregular. Polymor- phonuclear leukocytes, lymphocytes, plasma- and mast-cells, eosinophiles, young connective-tissue and giant cells, red blood cells, and the organism constitute the mass of cells. The polymorphonuclear leukocytes here, as in other cases, collect in masses to form small subcutaneous abscesses, and are moreover distributed diffusely in small numbers. Plasma cells are present in characteristic masses and colimms, but in much less numbers than in the preceding cases. In the verrucous tissue they exist only in small collections. Where scar tissue is forming actively, they are distributed in accordance with the amount of scar tissue present. Where the latter is marked, plasma cells are less numerous than in places where such tissue is beginning to form. This is in harmony with the conception that plasma cells are sacrificed in the formation of inflammatory connective tissue. Examples of cells which have lost part or nearly all of their granular protoplasm are common. It is noted also in this tissue that the last plasma cells to leave the field are those situated around the blood-vessels. Hyalin degeneration of the protoplasm occurs, but is a less conspicuous feature than in the first case. Mast-cells in small numbers are diffusely distributed, being absent, however, in dense collections of polymorphonuclear leu- kocytes. In the active verrucous process eosinophiles are common, occur- ring singly and in small groups. Their granules stain equally well with eosin or acid fuchsin. Giant cells have a promiscuous distribution, being found in mihary abscesses, in neighboring acutely involved tissue, and to a less degree in the more deeply lying scar tissue, where ordinary Oidiomycosis of the Skin and Its Fungi 93 infiltration is not marked. They are of the tuberculous type, and commonly contain, besides the intrinsic nuclei, polymorphonuclear and mononuclear leukocytes, and some larger (connective-tissue ?) cells. None contained the fungus, so far as observed. Tubercle bacilli could not be demonstrated by the usual staining methods. Abscesses.— -Those seen in the subcutaneous tissue may be as large as a millimeter in diameter. The intra-epithelial are usually smaller. Giant cells have not been found in the latter, but degener- ating epithelial cells, phagocytosis, and cell invasion are conspicu- ous. The including cells are epithelial usually, and the included bodies are leukocytes and small globules, considered as free and altered nuclear material. Besides the above cells, only polymor- phonuclears, and in small abscesses an occasional eosinophile, are found. The subcutaneous abscesses are made up of polymorphonuclear and mononuclear leukocytes, connective-tissue and giant cells, portions of adult fibrous tissue, fragments of elastic tissue, occa- sional mast- and plasma-cells, and the organism pecuKar to the case. The polymorphonuclear leukocyte is the prevailing cell. The organism is found in the subcutaneous and intra-epithelial abscesses and frequently in connective tissue without true abscess formation. It is always surrounded by infiltrating cells, often by a very small number. It occurs characteristically in clusters of from two or three to six or eight. Single forms are rare. Iso- lated pairs are seen occasionally. In a group the sizes of the indi- viduals vary, the adults or larger forms measuring about fourteen microns in diameter. Budding forms are numerous and may be seen in all stages. An adult cell may send off two or three buds, which are of different sizes, indicating separate development of the different buds. Individual organisms have the structure typical of this variety of fungus, i.e., capsule, clear space, and central protoplasm. The capsule has a peculiar substance depos- ited on its surface, which gives it a rough, somewhat ragged, contour. This may be a secretion of the organism or a reactionary deposit from the tissue fluids. It appears, however, to be an integral part of the capsule, and in a group of organisms seemingly acts as an agglutinating substance. It stains deeply with eosin 94 Contributions to Medical Science in Zenker-hardened specimens, and a bluish-green with polychrome methylene blue. There is nothing pecuhar about the clear space. It is usually present in the hardened tissue. In the central proto- plasm of young cells no structure can be made out. As they grow larger, however, a granular condition is apparent. In Zenker specimens stained deeply with eosin a spongy structure is brought out well. Neutral orcein also stains this substance. Moreover, in the older cells it is usual to see from ten to twenty or thirty minute spherical bodies arranged peripherally in the protoplasmic substance, having a diameter of one-half to one micron. They stain well with hematoxyUn and with methylene blue. A ruptured cell was found, in which these bodies were lying partly outside of the cell. One may suppose that they are endospores. Vacuoles are often seen in the larger cells. They are structureless and their definition smooth and sharp. The organism has no demonstrable nucleus. Separating the protoplasm from the clear space is a well-defined membrane. The efect of potassium iodide on the lesion. — Gradually the verrucous condition subsided, and a smooth, ghstening, scaling, thick, leathery, red scar resulted. Many minute abscesses were visible just beneath the epithehum. A portion of this tissue was removed for examination, fixed in absolute alcohol, and imbedded in parafl&n. It cut with uniform fibrous resistance, the cut surface being pale and relatively bloodless. The cicatricial portion was one centimeter deep. It showed a thickened homy layer and stratima granulosum. The rete is moderately thickened, but presents no positive down-growths, the deeper surface being com- paratively smooth. Correspondingly, the normal rete pegs are lacking, and the papillary layer is, of course, absent. The sub- epithelial tissue consists almost entirely of young connective- tissue fibers arranged in planes parallel to the surface. There are occasional groups of plasma cells in close proximity to blood cells. Mast-cells are numerous, whereas they were scarce in the verrucous condition. They are distributed with no reference to any special structure or condition. They are distinctly of the connective- tissue cell type. There are no eosinophiles. Only traces of elastic tissue are present. While intra-epithelial abscesses are Oidiomycosis of the Skin and Its Fungi 95 not present, several abscesses are encountered in the subcutaneous tissue. They contain a few organisms, which often are in a state of disintegration. Case F.* — Chnical history. — C. W. T., aged thirty-three, carpenter by occupation. Had always been unusually strong and vigorous. Seven years previous to his coming to the hospital he noticed a pimple on the right shoulder, which became covered with a crust which gradually increased in size. The man's general health had been unaffected, the disorder causing him no pain, though at times it was sensitive to the rubbing of his clothing or other irritation. Attempts at cultures and animal inoculations both with pus and fresh tissue gave no positive results beyond cultures of pus organisms. The large number of giant cells found in the corium led to a diagnosis of probable cutaneous tuberculosis, with a possible epitheliomatous compUcation, as there was marked hyperplasia of the rete. He was again in the Cook County Hospital, under the care of the late Dr. D. D. Bishop, from March 8 to April 19, 1895. Dur- ing this time the cutaneous disorder continued to progress, and the man gradually lost strength and became decidedly emaciated. He died soon after, of what his physician called acute miliary tuberculosis, the postmortem showing mihary tubercles in the lungs and in other internal organs. It was reported that in these lesions tubercle bacilli were discovered, and that guinea-pigs inoculated with tissue developed generalized tuberculosis. During the eight months in which he was under observation, numerous attempts were made to obtain cultures, but nothing was ever found except pus organisms. A number of guinea-pigs were inoculated at different times with bits of tissue both from older and from recent lesions. The majority of animal inoculations gave entirely negative results, no evidence of tuberculosis being obtained in any of them. Two guinea-pigs died of septic infection a few days after inoculation. - Tissue from the older lesions showed a markedly infiltrated cutis, with many giant cells, some of which were unusually large. ■ The following report is, in brief, as given by Professor Montgomery, who discovered the nature of the case. 96 Contributions to Medical Science In some places as many as forty giant cells could be distinguished in a single field under the microscope. Hyperplasia of the rete was very pronounced, the process extending deep into the cutis, with irregular shape and branches. In these epitheliomatous processes were miliary abscesses of various sizes, some of them bounded by markedly flattened and shghtly hornified epithelium. In other places cornified masses or whorls were apparent. At this time we did not consider blastomycetic dermatitis, and did not look for the organisms found in this disorder, though it is strange they should have escaped observation, as they are numerous in most of the sections that have been preserved. The organisms in their morphological features correspond closely with those described by Gilchrist and others. The organisms are found usually in pairs, both in giant cells in the intra-epitheHal abscesses and in the corium, but not within other cells. Although these bodies are present in large numbers, budding forms are difficult to demonstrate, partially because the sections have faded some- what. Distinct budding forms are seen, however. Tissue was removed from a very recent lesion on the face, soon after it had assumed the verrucous character, and used for culture and inoculation experiments with negative results. A piece of this tissue was also hardened in alcohol. Something over two hundred sections were stained and examined for bacilli. In three of these sections (supposed') tubercle bacilH were found. In one section there was but one, in another there were two, while in a third there were three fairly distinct baciUi. These were all found, however, in a small abscess communicating with the surface of the skin. Case VI. — Clinical history. — Mrs. ]\I. applied to Dr. Hyde for treatment on February 12, 1900. Patient is a German, fifty-one years old. Previous history good. On the dorsal skin of the first metacarpal are two crusted ( (' Note. H. T. R.) Hence if there were tuberck bacilli, it seems probable that their presence was accidental. Also it is possible that they may have been saprophytic acid-proof bacilli. Dr. Walker, who performed the autopsy, and made the diagnosis of pulmonary miliary tuberculosis, recently re-examined sections of the lung tissue, and found the nodules to be composed almost entirely of "blastomycetes," and unassociated w^ith tubercle bacilli. He reported his findings at the recent meet- ing of the American Dermatological Society in Chicago (igoi). The importance of this finding will be referred to later.] Oidiomycosis of the Skin and Its Fungi 97 ulcers, one near the proximal, the other over the distal extremity of this bone. A third lesion lies near the distal end of the fifth metacarpal. The three lesions now present are each roughly circular and covered with a thick, dense, dirty-yellow, closely adherent crust, situated on a prominently elevated base. On removing the crusts, bleeding occurs and small portions of tissue are torn away. The exposed surface is verrucous or papilHform. Pus lies in the clefts separating the papillae. Cultures. — Pus from the base of the ulcer and from the minute abscesses of the areolar skin were studied in a potassium hydrate mount. Verrucous tissue was teased and mounted in a similar solution. A study of the crusts in such mounts proved unsatis- factory. In the pus and teased tissue were found resistant spherical cells, which possessed a thick capsule and a central protoplasm. A few vacuolated and numerous budding forms were found. The protoplasm was finely granular, and usually there were several larger, highly refractive bodies. Several cells measured had diameters of from 7.5 to 9.4 microns. A few were oblong. February 15, 1900, ordinary media were inoculated with crusts, with pus from small abscesses, and with a piece of diseased tissue, which had been teased in sterile bouillon. The crusts yielded a prolific growth of ordinary fungi, forming greenish and brownish pigment and having spore-bearing aerial hyphae. Staphylococcus aureus and albus occurred Hberally. The pus yielded almost a pure growth of a fungus in isolated colonies and on all the tubes inoculated. The cultures from the tissue showed this fungus in only one tube. A small number of colonies of staphylococci developed. The staphylococci were not followed out pathogenetically. It required ten days for the fungus to develop macroscopically. As the appearance of the cultures was an unexpected one, it is possible they may have been discoverable earlier. The colonies were intimately incorporated with the media (agars and blood-serum). The central point was sKghtly elevated, opaque, and grayish in color, and myceHal rays extended peripher- gS Contributions to Medical Science ally, growing less numerous as the border was approached. There was no growth above the surface. By using a low magnification, individual threads were visible. They shot out straight and gave off branches at an acute or right angle. Segmentation of the mycehum could not be determined. Attempts to grow succeeding generations on all available soUd and liquid media failed, both at brood-oven and room tem- peratures. A successful hanging-drop culture was obtained finally by planting a portion of glucose agar culture (consisting of the soHd medium impregnated with the growth) in a drop of glucose bouillon. Growth proceeded slowly. Straight, translucent, capsulated threads extended from the agar fragment into the bouillon, reached the border, and, growing further, carried with them sheaths of the medium. Branching at an acute angle occurred. Both the main threads and the branches divided into segments, which showed little tendency to separate. There formed both on the threads and their branches spherical or oblong ascus-like cells. These commonly contained about eight highly refractive spherules, which were structureless. The spherical cells eventually separated, sometimes before the refractive spherules were conspicuous. Such cells were seen to bud like blastomycetes. Inoculations from this colony on other media gave no cultures. A smear preparation was made from an agar culture, broken up by teasing, and stained with Loefiier's methylene blue. It showed a close mixture of mycehal threads, with oval, spherical, cyKndrical, and budding cells, the two varieties forming about equal parts of the mass. The threads varied from two to four microns in thickness, presented deUcate capsules, which stained faintly, and a central hyalin-Hke substance, which contained granules, spherules, and irregular masses, staining deeply with methylene blue. They branched freely, and possessed numerous lateral conidia of various sizes. Often a thread terminated as a large spherical or ovoid segment. Frequently sessile buds had sprouted a mycelial thread, which might be delicate or coarse. The threads segmented freely, and in this preparation the segments were less coherent than in the hanging-drop cultures. Oidiomycosis of the Skin and Its Fungi 99 Portions of threads and a few spherical cells were completely filled with a mass of deeply staining granules and spherules. The spherical and oval forms were doubly contoured and possessed a central structure similar to the mycelium. Many appeared to be degenerating, only the capsules staining. They were from six to ten microns in diameter. A few were seen to bud (April 22, 1 901). After some months it was possible again to inoculate culture media with pus from the miliary abscesses. I was successful this time, by using a maltose agar, in cultivating a mould-fungus which morphologically is identical with the fungus obtained from Case VI. Animal inoculations will be carried out and reported in the future. HiSTOPATHOLOGY. — A small amount of blood lies on the horny layer. The contour of the surface is most grotesque, forming many depressions and elevations of various sizes and shapes. A few of these depressions lead to ruptured abscesses. Others are small and cup-shaped and filled with distorted masses of horny material, retaining a rough lamellar structure and showing numer- ous nuclei. The horny layer varies in thickness and structure. Everywhere there is a degree of loosening and desquamation into lamellae ol various thickness. Nucleation of the deeper layers is present irregularly. In places the horny layer is absent, the rete being exposed. No stratum lucidum is evident in the diseased area. The granular layer shows great variability in occurrence and depth. It may extend to a depth of a dozen or fifteen cells where there is epithehal prohferation, forming a cone-shaped, granular appearance. The rete presents the carcinomatoid proliferation which is a classical feature of the infection. There are large blocks of epithelial cells connected with the surface rete, and also columns of varying thickness extending toward the subcutaneous tissue from these masses. There is an intercellular infiltration of wander- ing cells, intensified in places to small abscess formation. Among these cells the polymorphonuclear leukocytes predominate. No plasma cells, eosinophiles, or mast-cells are seen in the rete. The rete cells are swollen, granular, and the prickles are con- loo Contributions to Medical Science spicuous except in areas of partial cornification. Abnormal comification occurs in the deeper part of the rete and in the down- growths, in isolated cells, and in groups of cells. There are many- incipient whorls, but more are found isolated in the cutis. Isolated examples of epitheUal giant cell formation are found. Epithehal cells, which are otherwise apparently normal, are found including other epitheUal cells and polymorphonuclear leukocytes. The basal columnar layer is well formed. The intra-epitheHal abscesses are characteristic. They consist largely of polymorphonuclear leukocytes, granular material, and nuclear detritus. In addition, many loosened epitheUal ceUs, giant cells, and fungus cells are found. The epitheUal ceUs are spherical, irregular, or filamentous. The first type has usuaUy a dense, highly acidophiUc protoplasm and a nucleus which may stain deeply or faintly. They sometimes include other epitheUal ceUs or leukocytes. The filamentous epitheUal cells often resemble fungus threads, but in clearly stained specimens a flattened nucleus is usuaUy demonstrated. The giant cells within abscesses are small and not numerous. The abscess wall is composed of flattened epitheUal ceUs, which are being loosened continuaUy by the ceUular infiltrate. OccasionaUy larger abscesses communicate with the corium. The cutis is occupied by well-vascularized granulation tissue, ceUular infiltrate, and small abscesses. Many vessels have rather thick walls and a proUferating endotheUum. A few scattered, altered elastic fibers are found in the inflammatory area. Occa- sional bundles of white fibrous tissue are found. With a poly- chrome methylene blue and eosin stain, this tissue stains a Ught blue-green, while the deeper, healthy fibers stain with eosin. Coil glands found in the diseased area show epitheUal proUferation. The abscesses are not different from those of other cases. They contain occasional giant cells and the organism. The infiltrating cells are polymorphonuclear and mononuclear leukocytes and plasma cells. The last-named have the customary distribution, and a few contain hyalin spherules. Giant cells are smaU. A few are found in the granulation tissue. The organisms are found in the intra-epitheUal and subcu- Oidiomycosis of the Skin and Its Fungi ioi taneous abscesses and in the surrounding granulation tissue. They are spherical, sometimes in pairs, or presenting an oblong cylindrical form, and are frequently seen in process of budding. They have not been found in giant cells. In the granulation tissue the number of surrounding leukocytes is sometimes small. Structurally they possess a thick, hyalin-like capsule and a central protoplasm, which is granular. Certain large granules stain well with the polychrome methylene blue. Vacuoles, which may be central or eccentric, single or multiple, are found in occasional cells. A "clear space" is seen inconstantly. Several degeneration forms have been encountered (containing no central protoplasm). The capsules are frequently fissured. Careful searching of appropriately stained sections failed to reveal tubercle bacilH. No animal experiments were made. Case VII} — CHnical history. — Patient is seventy-three years old; bom in West Prussia, where he was a farmer; general history fair. In February, 1898, while engaged in general labor, a pimple appeared on the scalp, about one and one-half inches above the left ear. Pus formed and discharged. The lesion crusted and extended gradually, causing no pain, but was tender on pressure. Slowly the disease reached its present proportions, tending to heal in the center. Six months ago a pimple appeared on the left malar skin. This suppurated, extended rapidly, and in two months was an inch in diameter. Since then it has progressed continu- ously. Both foci have discharged freely. The patient's general health has been unaffected. Cultures. — Teased tissue and pus from the small abscesses have been examined repeatedly with uniform findings. A capsu- lated budding organism is always present. It is from eight to sixteen microns in diameter; has no demonstrable "clear space"; the protoplasm is finely granular, and there are in addition larger granules of nearly uniform size, which move rapidly in the pro- toplasm. Vacuoles exist in many cells. After standing for two days in a strong potassium hydrate solution, the only change • Cases VII, VUI, and IX are reported in Jour. Boston Soc. Med. Sciences, igoi, s. P- 4S3- I02 Contributions to Medical Science noted was indentation of the capsule in some of the cells (Fig. 12, Plate 7). In the pus it is common to find the organism in groups of four to eight. Many single and paired forms have been found filled to bursting with refractive spherical bodies, less than one-half micron in diameter. Some of these cells were budding. Exactly similar bodies occurred free in groups of twenty to fifty. They were not seen in actual egression from the parasite. In mor- phology it would be impossible to distinguish them from degenera- tion products of tissue cells. A platinum loop of the pus was mixed with three or four times its volume of nutrient bouillon. Many of these small bodies were found, but during a week's obser- vation none were found to develop into the fungus cell. On October 16, 1900, tubes of ordinary media were inoculated; some with pus from abscesses and others with teased tissue. Some of the tubes inoculated with pus remained sterile ; others had a few colonies of cocci. The tubes inoculated with teased verrucous tissue grew numerous colonies of cocci and bacilli. No fungus grew. This was repeated twice with similar results. Six plate cultures of the pus were made also, but the expected fungus did not grow. December i, 1900, tubes of maltose agar, prepared according to Sabouraud's method for the ring-worm fungi, were inoculated with pus from the abscesses visible in the reddened areola. The skin was first cleansed with alcohol. On December 7 small colonies were noted for the first time, i.e., six days after inoculation. With low magnification they were seen to consist of radiating hyphae. Examination of a colony in water showed hyphae branching at an acute angle, and possessing lateral pedunculated conidia — the "Aaron's rod" appearance. There were no isolated spherical cells. In four days more the colonies had grown rapidly, rising above the surface and forming an irregular, grayish mass, which was closely adherent to the medium and could be broken up only with considerable force. They rapidly became white and powdery from the formation of an aerial growth. Within three weeks hyphae reached the tube and grew up along the glass surface. In two months nearly the whole inner surface of the tube was covered Oidiomycosis of the Skin and Its Fungi 103 with a downy hyphal growth, which could be removed in soft coherent flakes with the needle. Under the microscope this structure is seen to consist of fine branching mycelial threads, which have lateral conidia and asco- spores. The branching occurs at, or nearly at, right angles and is monopodial. Near the ends of threads, branching is often at acute angles. The main hyphae can be followed to extreme lengths. Neighboring hyphae and their branches unite, and the older portions of threads are divided into segments, which show no tendency to separate. Conidia develop principally on the older portions of threads and on the lower branches, the growing ends being almost devoid of them. They may be sessile or pedunculated, usually having a delicate pedicle, the protoplasm of which at first com- municates with both that of the conidium and the hypha. Later a capsular partition separates each from the other. The conidia multiply in situ by budding, so that it is common to find a group of four to six at the end of a pedicle. The commonest picture, however, is that of a thread giving off single pedunculated conidia, arranged on alternating sides along the course of the hypha. No structure can be seen in the conidial forms. Ascospores ( ?) form at the ends of hyphae and the higher branches. Occasionally the end of a thread is dilated into a spherical form, with no visible structure. The largest are twelve to fourteen microns in diameter, and the protoplasm is divided in a radiating manner, forming regular cone-shaped segments. The number could not be exactly ascertained. Young ascal cells are seen free and attached, con- taining from one to several spores ( ?) . Exactly similar growths occur when the organism is transferred to glycerin, glucose, plain agar, and blood-serum, the first visible colonies appearing in four to six days. On none of these media is the growth so pronounced and rapid as on maltose agar. On plain agar it is rather meager, but there are always aerial hyphae, which have the structure described above. Viewing the reversed surface, the agars exhibit a rich, even, golden-brown color, like that of a well-colored meerschaum, becoming rather yellowish near the margins. The aerial growth on ox-blood serum is sometimes very marked. I04 Contributions to Medical Science No growth formed on highly acid media, as +io oxalic acid agar. Likewise, prune-decoction gelatin constitutes a poor mediiun. Glycerin agar plate cultures. — Some difficulty was experienced in getting disseminated colonies, owing to the nature of the growth. A portion of a bouillon culture, however, was broken up in a sterile mortar and utilized for plate cultures. After three days colonies are barely visible, and are not conspicuous until five days. At the end of a week they are one or two millimeters in diameter, roughly circular, star-shaped, feathery, sending out rays of varying lengths. The borders are frayed and uneven to the unaided eye. With low magnification the colonies appear entirely hyphal, threads radiating from a dense central network, and becoming sparse as the periphery is approached. They branch at an acute angle. No spherical conidia are seen, but there are many lateral, elongated, unicellular shoots. Segmentation of threads is indis- tinct (Figs. 7, 8). The deep colonies grow as rapidly as the super- ficial. Eventually aerial hyphae develop. Potato. — Small portions of infiltrated agar cultures were smeared over the surface of sterilized potato. A beginning growth is noted in three days, and in two days more a white, downy surface has formed. Under the microscope the aerial growth shows branching hyphae, with lateral sessile and pedunculated conidia and elongated unicellular offshoots. New colonies very early produce an aerial growth. After the medium has dried, the surface of the growth is white and powdery and rubs off easily; but beneath this it is hard and some force is required to break it up. The hard portion under the microscope shows faintly segmented threads, with occasional pedunculated conidia, mingled with large numbers of spherical cells of various sizes. Many of the latter are small enough to be considered detached conidia, while there are other larger spherical cells (10-12 microns), many of which are budding typically. The ends of some threads consist of a spherical segment. The spherical cells are capsulated and granular, but none are seen forming the characteristically shaped spores observed on the aerial h^-phae of maltose agar cultures. Bouillon. — The growth in plain bouillon made with meat extract Oidiomycosis of the Skin and Its Fungi 105 is more pronounced than in glycerin bouillon with meat extract. In sugar-free bouillon prepared by the Theobald Smith method it is still more marked. Tubes were inoculated either with a mass of the aerial growth or with fragments of infiltrated solid medium. The first appearance of growth is a hazy, coherent zone surrounding the transposed fragment. This increases in size, and in about ten days other hazy flakes appear, which have, no doubt, arisen from cells loosened from the primary mass. Eventually the medium is largely occupied by such material, the overlying fluid, however, being perfectly clear. When the fluid is completely filled or nearly evaporated, aerial hyphae shoot up from the growth as it lies adherent to the tube. Microscopically one sees a tangled mass of long, straight, or slightly curved mycelial threads, which interbranch, and a small number of spherical cells, which multiply by budding. The hyphae vary in structure and size. Some are only one to two microns thick, shghtly tortuous, branch freely, segment very indistinctly or not at all, and possess a granular protoplasm, which in addition is differentiated into oval or spherical, slightly tinted bodies, nearly the width of the thread. Other hyphae are six to eight microns in thickness, clearly segmented, give off short branches and conidia, and possess a protoplasm Hke that of the finer threads. There are threads of intermediate sizes. At the end of a larger segmented hypha a cluster of spherical cells may form. The terminal segment of a thread becomes spherical, and multiplies by budding, the process continuing until from ten to thirty or forty cells accumulate. The cluster eventually breaks up. These spherical cells in small part again produce buds, but largely develop into new threads. Each contains from two to four or five spherical, highly refractive granules. All these forms possess the customary capsule. There are a few short chains of spherical cells (Figs. 7, Plate 4; 8, Plate 5; 11, 12, Plate 6; 17, 1 8a, Plate 9). On litmus milk growth is very slow; the medium becomes slightly alkaline without being coagulated. The growth hes on the surface, where it has an oily appearance. Owing to the agglutination of the growth in all media, successful io6 Contributions to Medical Science stab-cultures in agar and gelatin have not been made. A hyphal growth occurs in gelatin, however, without liquefaction. In hanging drops growth was very slow. Proliferation took place by loosely segmented hyphae and by budding. Fine threads did not form. Saccharine solutions were not fermented. Using several maltose agar tubes at each inoculation, the same organism has been cultivated four times in a pure state from the mihary abscesses. HiSTOPATHOLOGY. — Vcrrucous tissue from the border was removed, the section including adjacent skin. One portion was fixed in ten per cent formalin, and a second in a mixture of Muel- ler's fluid and formalin. ParafiBm imbedding. Over the surface hes a large amount of homy debris, bacteria (largely cocci), many red blood cells, and multitudes of leukocytes. The homy material often has a lamellar arrangement, the plates being separated by leukocytes, red blood cells, and bacteria. A large proportion of the horny cells are nucleated, although horny masses are found which have undergone almost complete granular degeneration. Rarely a " blastomyces " is found in the horny debris. That part of the homy layer which is still attached is relatively thin. Masses of concentrically arranged cells frequently lie in depressions of the rete. Cornification is incomplete, as shown by the presence of nuclei, a granular eosinophiUc protoplasm, and the breadth of the cells. Flattened polymorphonuclear (a few mononuclear) leukocytes lie in numbers between the lamellae. In one place a mihary abscess, containing a pair of organisms, is found within the horny layer. Stratum lucidum: Absent. Stratum granulosum: Present in some places, absent in others. When present to a prominent or normal degree the overlying horny cells approach the normal, being usually non-nucleated and having a lamellar formation. Where the granular layer is entirely absent there is the imperfect cornification mentioned — nucleated, granular, partially flattened cells, with thick walls. In the depths of the rete, keratohyalin granules are found in the outer cells of epitheUal pearls; and occasionally an isolated cell, or a group of three or four, contains such granules. Stratum mucosum: The enormous Oidiomycosis of the Skin and Its Fungi 107 hyperplasia characteristic of the affection is accentuated. This would be inferred from the usual size of the papillae as seen grossly. In the amount of hyperplasia of the rete this case dupHcates Case IV. Cross-sections of deformed and enlarged papillae and of intra-epithelial abscesses are numerous. The cells adjacent to the cutis have a protoplasm which is more granular, more chromo- philic, and of smaller dimensions than superficial cells; the nuclei stain more densely and exhibit mitoses. The rete cells are imi- formly large, granular, finely vesiculated, and have distinct prickles. The interepithelial spaces are often large. Except where there are small foci of necrosis the nuclei stain well. There are various sized epithelial whorls and examples of premature and partial comifica- tion. In the interepithehal spaces are polymorphonuclear leu- kocytes, a few eosinophiles and mast-cells, and red blood cells. Rarely a parasite is foimd in this position (Fig. 3, Plate 3). A few large hemorrhages are seen in the rete, the surrounding cells being flattened. Intra-epithelial abscesses are numerous and variable in size. They have the usual contents, including giant cells and the organ- isms. There are many examples of epithelial giant cell formation in the body of the rete apart from those in abscesses. They may contain leukocytes or other organisms. The cutis is marked by greatly deformed papillae, dilated blood- and lymph- vessels, and numerous small hemorrhages. A mucoid appearance in the superficial cutis seems to have been caused by the marked edema. The fixed tissue cells have pro- liferated excessively. Plasma cells are less crowded than in some previous cases. They occur in groups and in columns, surrounding blood-vessels, and occasionally contain hyalin bodies. Mast-cells are seen singly and in groups. Eosinophiles are found in large numbers, often in groups, constituting another point of resemblance to Case IV. There are tubercles resembhng those of tuberculosis. They differ from the latter in being well vascularized. Giant cells are exceedingly numerous, but no tubercle bacilli can be found. The subepithelial abscesses are like those of other cases. Poly- io8 Contributions to Medical Science morphonuclear and mononuclear leukocytes infiltrate the cutis diffusely. The organism in tissue. — In the cutis it is seen in abscesses, among connective tissue apart from abscesses, and in giant cells. In the epithehum it occurs in the abscesses and between epithelial cells, often with httle or no surrounding zone of leukocytes. It is found within giant epitheHal cells, but not within mononuclear cells. In all tissues it multiplies by the characteristic budding process. Groups of several organisms are seen occasionally, as in Case IV. Such groups commonly have a small amount of inter- cellular cement substance. The usual structure is found, a doubly contoured capsule, a clear zone, and a central protoplasm. Polychromatic methylene blue stains the organism characteristically. The capsule appears deep blue, the clear zone has a reflected blue tint, and the central protoplasm stains reddish-blue. The same results are obtained with carbol-toluidin blue and carbol-thionin. These stains make the organism stand out clearly, while hematoxylin, with eosin and carbol-fuchsin as counter-stain, is less satisfactory. The central protoplasm is granular, sometimes containing the spherical granules of uniform size mentioned under preceding cases. They are rather difficult to see in the methylene blue stains, as the whole protoplasm may be deeply stained. A hematoxylin stain brings them out more clearly. Vacuoles are common. Degeneration forms are found in numbers, particularly within giant cells. Organ- isms with fractured and others with indented capsules are met with occasionally. The average diameter is ten to twelve microns. Animal Experiments. — Guinea-pigs. i. Inoculation of tissue beneath skin over the right shoulder. The wound became swollen, and on the third day death resulted. Staphylococci cultivated from the wound and the viscera. 2. The coherent fungus mass from a bouillon tube was placed subcutaneously. After three days a nodule formed, which discharged in a week. In the pus short mycelial forms were found. The neighboring lymphatics continue enlarged, but the wound healed completely. White Rats. i. A block of infiltrated agar was inserted subcutaneously with no result. Dogs. I. Blocks of infiltrated solid media were placed within the abdomen. No observable effect after two months. Oidiomycosis of the Skin and Its Fungi 109 2. Intravenous inoculation of a bouillon culture broken up in a mortar. After ten days became somewhat emaciated, and had a purulent conjunctivitis, con- taining only cocci. One month later the right cornea was involved in a scar and slightly deformed. As emaciation and weakness were extreme, and the animal was dyspneic, chloroform was administered. At the autopsy, all organs are found to be pale and otherwise unchanged, except the lungs. Both parietal and visceral pleurae are smooth, but the latter is mottled gray and red, and presents rather large grayish elevations. Only portions of the upper lobes crepitate, the remaining lung tissue being involved in a rather soft nodular consolidation. Section shows numerous grayish, confluent, and ill-defined nodules, from which fluid containing minute granules can be squeezed. A similar fluid escapes from the bronchi. No scars or well-marked abscesses are noted. A portion of a nodule teased and examined in water shows almost a solid mass of spherical, capsulated, and budding organisms, which are often filled with refractive, spore-like bodies; similar bodies are seen singly and in groups; extracellular organ- isms may have a marked adventitious capsule, are often agglutinated, and one or more may exist in a mono- or multi-nuclear tissue cell. Apparently gradation forms exist between the extracellular spore-like bodies and distinctly capsulated forms of the organism, but as these bodies mounted in a hanging drop do not develop further, their position as spores remains uncertain. The only clear method of proliferation in the lung tissue is by budding. Vibrating granules are often found in the fungus cells. The organisms in the bronchial fluid have a similar morphology. Some fluid from a nodule, mounted in a hanging drop of bouillon, showed in a few hours the sprouting of hyphae from the spherical cells, both those containing spore- like bodies and those with relatively clear protoplasm. Within twenty-four hours nearly all cells had started similar processes, which eventually reached great lengths, became segmented, and gave off lateral conidia and unicellular offshoots. The spore-like bodies took no evident part in this development, and they disappeared largely after a few days. Pure growths of the fungus inoculated were obtained from lung nodules, but inoculations from other viscera and the blood remained sterile. A lung nodule sectioned and stained consists almost entirely of fungous cells, with a doubly contoured capsule, clear space, and protoplasm, which often contains stained spore-like bodies and vacuoles. They are often in groups, surrounded and held together by a granular, cement substance. Mono- and multi-nuclear endothelial cells contain one or more organisms. Remnants of alveolar structure are seen, but the cells are very granular. Vessels are moderately congested; no hemorrhages. There are small masses of fibrin and a few mast-cells. The reaction to Gram's stain is peculiar, as only one out of many dozen cells is stained, and very deeply; the capsule is partly decolorized. Extracellular spore-like bodies stain with nuclear dyes, but they present no differentiated structure. Organisms are not found in peribronchial lymph glands. Kidneys: Diffuse granular degeneration of the epithelium; ,a few foci of perivascular plasma cell infiltration and endothelial proliferation; no fungous cells. Liver: Fatty degeneration. Brain, spleen, heart, and suprarenals normal. Rabbit. Intravenous inoculations of bouillon culture. No effect. no Contributions to Medical Science Summary of the Biological Features of the Organism. — In tissues it proliferates only by budding, so far as can be observed, although certain granules in the cells and the location of large forms between epithelial cells suggest that spores may form and be transported by lymph currents. From three to eight days are required for development on artificial media. The medium becomes infiltrated with h^-phae and an aerial growth occurs, on the threads of which pedunculated lateral conidia and terminal ascus cells form, which contain spore- like bodies. Aerial threads are divided into closely cohering segments. Conidia may multiply by budding. The buried hyphae are more loosely segmented and have few lateral conidia, but many unicellular shoots; some budding cells are seen also. In Hquid media the growth is mycelial, branching, segmented, few lateral conidia, but terminal spore groups, the elements of which multiply by budding. Milk is coagulated; produces a slight alkahne reaction. Gelatin is not liquefied. On plates feathery, irregular, hyphal colonies develop, the submerged growing as well as the surface colonies. On all media the protoplasm is clear or finely granular, capsulated, and may contain spherical or oval, sHghtly tinted bodies of uncertain function. Few vacuolated bodies are seen. ProHferation then takes place by: 1. Lateral aerial conidial formation. 2. Terminal ascospores (?). 3. Separation of mycelial segments. 4. Budding of terminal spore groups. Case VIII .^ — Mrs. A. M., age twenty-eight; Bohemian; occu- pation, housework. Has had the usual diseases of childhood. No specific history. On examination the disease, which first appeared about twenty months ago, was seen to involve the entire region surrounding the right orbit, extending well over the cheek and up on to the ' Occurred in the service of Dr. W. L. Noble at the Illinois Eye and Ear Infirmary; it was referred to Professor Montgomery for diagnosis. I am indebted to them for the privilege of reporting the case. Professor Montgomery made a rejxirt of the case before the American Dermatological Association, May 30, 1901. See also note to Case \^II. Oidiomycosis of the Skin and Its Fungi • iii forehead. The process had travelled over the bridge of the nose and had involved the left upper eyelid and a small region imme- diately below the eye. The clinical features were identical with those of several cases of blastomyces previously described. Cultures. — ^At first no fungus grew on media inoculated with tissue and surface pus ; ordinary bacteria appeared freely. Later the skin overlying several myxomatous tubercles was steriUzed with alcohol, incised, and the soft contents scraped out with a sterile curette. Examination of the fresh scrapings showed groups, pairs, and single forms of spherical, capsulated, budding cells. Inoculations of this tissue were made on ordinary media and maltose agar. A growth was not noted for five or six days. Then it was seen that hyphae had sprouted from fragments of tissue placed on maltose agar, but no growth occurred on other media. In about two weeks hyphae rose above the surface, and in a week more the mould-fungus nature of the organism was clear. In morphology and methods of proliferation this fungus appears identical with those cultivated from Cases VI, VII, IX, and X. Animal inoculations are in progress. HiSTOPATHOLOGY. — The tissue removed included a myxomatous nodule and a small amount of the verrucous border. So much healing had occurred under the administration of the iodide of potassium that the typical findings could not be anticipated. In KOH mounts of fresh tissue, groups of blastomycetoid bodies were found. EpitheHal hyperplasia is marked only in the verru- cous tissue, the epidermis covering the nodule being only moder- ately thickened, and not sending prolongations into the underlying tissue. Only a few intra-epithelial abscesses are encountered in the verrucous tissue and none in the covering of the nodule; three are found in the corium. There is moderate diffuse infiltration of the rete in the verrucous tissue, with leukocytes, but none in that of the nodule. The infiltrating cells of the corium, as in other cases, are poly- morpho- and mono-nuclear leukocytes, and plasma cells; in addi- tion, there are numerous eosinophiles, as in Cases VII and IX, There are many giant cells of the tuberculous type, and a number 112 Contributions to Medical Science of cell nodules are present. One of these is very similar to the t}T)ical tubercle; however, there is no central necrosis, and lym- phocytes infiltrated the structure diffusely. There is a large amount of cicatricial tissue, corresponding to the degree of healing noted cHnically. Few fungous cells are found. They lie in two groups, each surrounded by leukocytes. The individuals measure about ten microns in diameter, proHferate by budding, contain spore-Uke bodies, possess a doubly contoured and a thin adventitious capsule. Case /X.'— Clinical history.— Mrs. J. K., age thirty-three years, born in Germany. She had poor health prior to puberty, and passed through the usual diseases of childhood. Close questioning reveals no history of syphihs. A Httle more than two years ago three pustules appeared at about the same time, one in the skin overlying the left lower jaw, another on the dorsum of the left wrist, and the third on the right buttock. The one on the face continued to spread at all points, became more and more warty, and discharged pus freely. The diseased area was excised by Dr. Graham, a zone of healthy skin one and a half centimeters wide, and the underlying fatty tissue, being included. Gross Appearances. — The principal diameters of the area are 5.5 by 4.5 centimeters, the form being roughly ovoid, and the contour a little irregular. The lesion presents a central and a peripheral portion. The latter is from 0.5 to 2.0 centimeters wide, elevated from 0.2 to i.o centimeters above the surrounding skin, and has a characteristic coarse verrucous or papillomatous appear- ance. It meets the surrounding skin abruptly, and where it is thickest and highest becomes rounded and overhangs and conceals the healthy border, cauliflower-hke. The papillae are of various sizes and grotesque shapes. Some are millet-seed size and thickly set, each being separated from its neighbor by a deep, narrow cleft; this type occurs nearest the central portion. Others are large fungous masses, from 0.2 to 0.5 centimeters in thickness, with a relatively slender base, and having a surface set with second- » See note to Case VII. Oidiomycosis of the Skin and Its Fungi 113 ary millet-seed size or larger papillae; these occur principally in the outer part of the peripheral zone. In general, the papillae are shorter and simpler as the center of the lesion is approached (Fig. lb, Plate i). The central portion is from 0.5 to 2.0 centimeters wide and 3.0 centimeters long. It is depressed about 0.5 centi- meter below the level of the highest peripheral portion, but is elevated sUghtly above that of the surrounding skin. Its surface is relatively smooth, and at a few points somewhat glistening. An outer zone, however, is coarsely granular, this appearance being caused by inconspicuous, blunt, short papillae. The papillomatous tissue is exceedingly friable and was torn shghtly in manipulation. The central area is a little more dense, and suggests new cicatricial tissue. The whole lesion is more or less covered with blood, secondary to the operation. The tissue was bisected to expose the deeper structure. The subcutaneous fat is sharply and rather evenly separated from the overlying diseased tissue, and presents a few moderately enlarged vessels. Between the subcutaneous fat and the papillary layer is a homogeneous, whitish zone, from 0.2 to 0.5 cm. deep, which in places exhibits hemorrhages, varying from pin-point size to several milHmeters in extent. Some have an oblong or irregular shape. This zone, no doubt, represents the densely infiltrated corium. The papillary layer consists of two portions: a surface zone, represented by shaft-Hke papillae, separated from each other by clefts; and a buried zone, in which the papillae are united side by side. Those in the latter zone are sharply contoured by white lines of horny tissue, which caps the surface, sheaths the deeper portion, and cements each papilla to those adjacent for the depth of 0.2 to 0.4 cm. The extension of these papillae above the surface of course gives rise to the isolated surface papillae. A central "core" of vascular tissue occupies the center of each papilla. It is not possible in any portion of the tissue to detect microscopic abscesses ; this probably is due to an unfortunate accident in tech- nique. (However, after the tissue had been passed through the KaiserHng solutions for preservation as museum specimens, minute abscesses were seen in cross-sections.) HiSTOPATHOLOGY. — Two or three papillae were clipped off 114 Contributions to Medical Science and examined in a KOH solution. Portions of elastic fibers, epithelial, blood, and other tissue cells were recognized readily. No difficulty was encountered in finding numbers of a blastomyces- like organism. Eight occurred in one field; many other fields, however, contained none. As to form, they were, as usual, spherical, budding, and occa- sionally elongated; a few cup-shaped and sickle-forms were seen. Several groups of three to five organisms each were found. The capsule appeared, as usual, with a clear double contour and of uniform thickness. The protoplasm in certain cells was almost clear, in others contained a few granules, and in many was com- pletely occupied by minute spherical refractive bodies, smaller than those ordinarily seen in Cases VI, VII, and X. Transition forms between the minute bodies and the adult capsulated cell are present. They are intermediate in size, and present no differen- tiation except a very delicate capsule. Two cells were found enclosing pigment granules. The average diameter of twenty cells was 9.5 microns. Four portions of tissue have been used for microscopic study: (i) The surrounding relatively healthy skin; (2) the outer densely verrucous zone; (3) the central smooth, cicatrized portion; and (4) a point intermediate between 2 and 3. All tissues had similar treatment, being in Zenker's fluid for a few seconds by accident, washed in water for two hours, fixed in formalin and water, dehy- drated, cleared in cedar oil, and imbedded in paraffin. The description of the histological pathology of Case VII applies in all essential details to that of this one. More underlying tissue is included in this case, which merely shows that the leu- kocytic cell infiltrate extends superficially into the subcutaneous fatty and muscular tissue. The number of intra-epithelial and subcutaneous abscesses is large. The organisms occur singly, in pairs, and in groups, are occasionally surrounded by an adventitious capsule, may contain staining spore-like bodies and vacuoles, and possess a doubly contoured membrana propria; they are found in all abscesses, between relatively healthy epithelial cells, and in the granulation tissue of the corium. As in Case VII, eosinophiles are a conspicuous feature of the cell infiltrate, extending markedly Oidiomycosis of the Skin and Its Fungi 115 into the rete and intra-epithelial abscesses. The rete shows the types of atypical cornification, and other epithehal changes noted in other cases. Plasma cells are numerous, but polymorphonuclear leukocytes are equally so. The areolar skin shows a small amount of cell infiltrate and a few minute abscesses in the corium. In the central depressed portion of the lesion the surface papillae have disappeared almost entirely, but the subcutaneous changes are as pronounced as in any portion of the tissue. There are remnants of elastic tissue in the infiltrated corium. Many groups of hyalin bodies are found in plasma cells. Giant cells in the rete and corium are numerous and of the tuberculous type. Many contain the fungus. Cultures. — A portion of verrucous tissue and infiltrated corium, with a small amount of bouillon, was placed in a sterile mortar and thoroughly disintegrated. This suspension was used in inoculating various media, including maltose agar; plate cultures also were made in plain-, glycerin-, and glucose-agar. The tubes and plates were examined daily under a low objective. On the fourth day hyphae began to grow out from several tissue fragments. Growth proceeded slowly, the threads at first growing on the surface or immediately beneath it. In eight or ten days an aerial growth was perceptible ; this increased and soon appeared identical with the organism cultivated from Case VII. Study of its mor- phology and methods of proliferation also shows the apparent identity of this fungus with that from Case VII. That is, the aerial hyphae bear lateral pedunculated conidia, which may multi- ply by budding, and terminal bodies, which seem to contain spores; there are large cells filled with refractive, structureless, spore-like bodies; submerged spherical cells multiply by budding, and the submerged mycelium is segmented, contains nucleus-like bodies, and gives off conidia and unicellular shoots. Both aerial and submerged hyphae present monopodial branching. Animal Experiment. — In order to eliminate the possibility of tuberculosis, a portion of tissue was placed subcutaneously over the inguinal lymph glands of a guinea-pig. Swelling was noted in a few days, and a hard nodule soon formed, from which thick, creamy pus could be pressed. The latter contained spherical forms, the ii6 Contributions to Medical Science protoplasm of which was often differentiated into small, spherical, refractive bodies. Inoculations on media resulted in growths of a fungus which morphologically is iden- tical with that cultivated from the tissue. In the pus the organism prohferated by budding. The inoculation wound healed gradually, no general infection resulting. Case X. — This case was brought to my notice through the kindness of Dr. W. E. Marquardt, who has furnished the following clinical history. Mr. R., fifty-eight years old, a carpenter, was bom in Germany; general history good. His affection began in December, 1895, as a papule half an inch above the styloid process of the left radius. On examination it presents a reddish granular surface, excavated in places and surrounded by an elevated, reddish, puffy areola. Throughout the lesion are pin-head-sized abscesses, which are more densely distributed at the margins. As the patient would not consent to the removal of tissue for examination, a report on the histopathology cannot be given. Cultures. — When first seen with Dr. Marquardt, the lesion had been macerated in moist dressings, and no miliary abscesses were visible. Papillae were cKpped off, however, and inoculated on ordinary media and maltose agar. At the same time a few papillae teased and mounted in a KOH solution showed capsulated, budding cells, existing singly, in pairs, and in groups. Many were filled with structureless, spherical, refractive bodies, Uke those seen in some previous organisms. The growth of ordinary bacteria on all tubes was so abundant that new inoculations were tried. The patient's hand had been dressed dry for a number of days, and many miliary abscesses had formed. Dr. Marquardt kindly inocu- lated tubes of maltose agar with pus from the abscesses, after cleansing the skin with alcohol, and sent them to the laboratory. A week later the development of a mould-fungus had begun. It is necessary to add only that this fungus in its morphology and reproduction appears identical with those cultivated from Cases VI, VII, VIII, and IX. The organism was obtained so recently that it has not been possible to terminate animal experiments. Case XI. — This was a bottle of celloidin sections, with an absolutely imknown history, bearing the label "Tuberculosis Oidiomycosis of the Skin and Its Fungi 117 verrucosa cutis." It was one of a number of cases of supposed tuberculosis of the skin kindly placed at my disposal by Dr. Le Count. The sections are all quite thick. The method of fixation is not known. The horny layer is thickened, desquamating, imperfectly corni- fied, and infiltrated moderately with polymorphonuclear leukocytes. In places the granular layer is thickened, and it is absent in others. The rete is exposed occasionally, and a sinus is seen communicat- ing with an abscess in the corium. There is dense infiltration of the rete with polymorphonuclear leukocytes. This layer also presents the great hyperplasia and deformity characteristic of the disease, and contains many miliary abscesses of various sizes. The cells are large, vesicular; the nuclei usually stain well, and the prickles are distinct except around abscesses and areas of abnormal cornification. There are occasional examples of epithelial giant cell formation, and of the inclusion of leukocytes by epithelial cells otherwise apparently normal. The basal columnar formation is maintained with some deformities, the protoplasm staining deeply. No mitoses are found (perhaps because of the method of fixation). Large knob- like masses of rete cells are quite sure to contain one or more minute abscesses. The cutis is largely a mass of granulation tissue, densely infil- trated with wandering cells, of which polymorphonuclear leukocytes form the preponderating number. Lymphocytes also occur in great numbers, having a fairly uniform distribution. Frequently the polymorphonuclears form distinct abscesses. Plasma cells are numerous and have the characteristic distribution for this cell. HyaHn bodies have not been found in them nor in other portions of this tissue. Mast-cells cannot be demonstrated. This again may be due to fixation methods or the age or method of preservation of the tissue. There are only fragments of poorly staining elastic tissue in the parts severely involved. Hair foHicles, sebaceous and sudoriparous glands are included in the sections. The hair and sebaceous glands are normal, but surrounded by an excess of fibrous tissue. The sweat glands also are surrounded by an excess ii8 Contributions to Medical Science of connective tissue, many of the tubules are almost obliterated, and others are completely filled with epithelial cells. The number of giant cells in the corium is strikingly small. In the intra-epithe- lial abscesses there is a large number of detached and alterated epithehal cells. They possess the characters described in connec- tion with other cases of this series. The organism is found in the abscesses of the epithelium and cutis, and in the granulation tissue of the latter. The large number in the subepithelial tissue is a distinguishing feature of the case. The diameter is from ten to sixteen microns. There are many budding and some large cylinder-like forms. As in other cases, the organism possesses a doubly contoured capsule and a central protoplasm, which is both finely and coarsely granular. A number of coarse spherical granules are sometimes arranged peripherally in the protoplasm. Tubercle bacilli could not be found. Case XII. — This patient, Mr. R., was admitted into the Presby- terian Hospital in the service of Professor Senn. The history is, unfortunately, meager. He was forty years old and married. His personal history was marked "good." He noticed a soreness of the right hamstring muscles in January, 1897. ^^ September 12, 1899, a piece of tissue was removed without anesthesia. The clinical diagnosis was "tuberculosis of knee-joint." Subsequent attempts to locate the patient were futile. The tissue in the possession of the laboratory is supposedly that removed at the last operation recorded. Professor Hektoen discovered the nature of the tissue, and kindly submitted sections for the purpose of this work. The characteristic changes of blastomycetic dermatitis are found throughout the sections both in the epithelium and sub- epithelial tissue. To enumerate: There are the great epithelial hyperplasia, leukocytic infiltration of all layers of the epithelium, intra-epithelial abscesses, the usual dense infiltration, congestion, and edema of the cutis. Abscesses are found in the cutis, and plasma cells are numerous. The organisms are found in the intra- epithelial and subepithelial abscesses and in the granulation tissue Oidiomycosis of the Skin and Its Fungi 119 of the cutis. They are spherical, and budding forms are encoun- tered. Structurally there are a thick capsule, a clear one, and a central staining protoplasm, which is granular and sometimes has from one to three vacuoles. The usual diameter is ten microns. There are a few giant cells in the cutis, which are immediately surrounded by leukocytes and plasma cells, a tubercle-like zone of endotheloid cells being absent. As the tissue was exhausted, further staining could not be done. Nothing can be said relative to the presence of tubercle bacilli. A satisfactory discussion of the clinical diagnosis cannot be given. As stated, the diagnosis of tuberculosis of the knee-joint was made. According to the tissue findings, it is probable that the chnicians dealt with a case of oidiomycosis rather than tubercu- losis. IV. ANALYSIS OF BLASTOMYCETIC INFECTION OF THE SKIN. CLINICAL DATA. Although the facts seem to indicate that the so-called pro- tozoic disease, Busse's Saccharomycosis hominis, and Gilchrist's blastomycetic dermatitis are related processes, the infection as it occurs in the skin obviously claims our chief attention at this time. In all, twenty-six cases of so-called "blastomycetic dermatitis" have been collected, the clinical histories of which (except Case XI) have been given very briefly here by the original observers. A large proportion of the patients have been laboring men in various capacities, and perhaps more susceptible to infection because of the greater likelihood of traumatism. Eight were women, all engaged in housekeeping. About half of the cases were of rather low station, but only a few lived under conditions where filth diseases are ordinarily found. Many had suffered from the usual diseases of childhood, or from some later febrile process (typhoid fever, smallpox, rheumatism), which appeared to be in no way associated with the development of the skin disease. It is important to note that, though always diUgently searched for, a history of syphilitic infection, or of any of the usual manifesta- I20 Contributions to Medical Science tions of syphilis, has in no case been obtained; this includes, in many cases at least, inquiry as to the condition of the wife and children of the patient; similarly no evidence of hereditary syphilis was discovered in any case. Two (Hyde, Hektoen, and Bevan, and Busse) might have been tuberculous, while others were entirely free from any suspicion of tuberculosis. The family history has been uneventful or entirely negative in most instances. In one, however, five members of the family had died of tuberculosis (Hyde, Hektoen, and Bevan), and the mother of another succumbed to the same disease (F. G. Harris). There was a doubtful family history of carcinoma in one case. No patient knew of a similar disease in his immediate family or ante- cedents. When such a small number of individuals is infected with a disease, the specific cause of which is probably widespread, we must suppose that some unusual conditions determine the infection; conditions relating either to the patient or the micro-organism or perhaps both. It seems probable that the oidia are subject to great variations in their pathogenicity, just as individuals are in susceptibiHty, and that the mutual adaptabihty requisite for infection is not often realized. Among the nationalities represented by the patients are Ameri- can, German, Irish, and PoHsh. Infection in the foreign-born occurred in all cases after immigration to America. Previous or present residence in America seems to have had no relation to the development of the disease. The oldest patient was a German of seventy-three years (Case VIII), and the youngest a little more than thirty years; the average age at inception being about forty. Physical examination has, almost without exception, shown the patients to be in excellent general health. Of the cases I have examined personally none exhibited general or visceral diseases; Case VII showed moderate arteriosclerosis, incident to his years. The following list of areas which have been attacked in the different cases indicates that probably no part of the skin surface is immune: scalp, face, including ears, eyelids, nose, cheek, chin, and lip; neck, chest, back, buttocks, forearm, and arm; scrotum, thigh, and leg; backs of hand and foot; 9,nd in one case the sole Oidiomycosis of the Skin and Its Fungi 121 of the foot. In none of those cases where the palpebrae have been involved has the conjunctiva or the orbit been invaded, though the former has suffered severely from exposure and secondary infec- tions. [In Busse's case of Saccharomycosis hominis, however, the eye was invaded; the same has occurred in the protozoic disease.] For unknown reasons the scalp also usually resists invasion. The first finding on the part of the patient, in all cases, is a sluggish red papule or pustule. If there has been a preceding traumatism he may observe that either the wound does not heal (Case IV) or that apparent heahng occurs, the papule or pustule not appearing for several days (Hessler), or weeks later (Stelwagon). The primary papule ordinarily becomes pustular in a short time, and the apex of the pustule transformed into a crust, which when removed is found to cover an irregular, elevated, reddened base which secretes a small amount of glairy muco-pus; a narrow red areola appears early. Extension takes place steadily so that in from two to four or six months the lesion has a diameter of one to two inches. In the meanwhile the rough granular condition of the base becomes greatly exaggerated and gradually a coarse papillomatous or villiform surface results, the processes of which are separated by deep, irregular clefts which are filled with pus. The appearance sometimes is decidedly cauliflower-like, the verrucous tissue rising precipitously above and overhanging the surrounding skin. In other cases the papillomatous tissue is more flattened and shows little of the cauliflower appearance; this is always the case in the process of healing. The skin immediately surrounding the verrucous tissue forms a characteristic areola. In the most extreme condition it is red or bluish-red, tumid, rises somewhat to meet the base of the ulcer, possesses an unbroken horny layer, and is studded with numerous sub- or intra-epithelial abscesses, which appear as minute yellowish points, I to 1.5 millimeters in diameter. On being pricked they yield a droplet of sticky, glairy pus in which the organisms are easily demonstrated. The degree in which these findings are present varies in different cases and in the same case at different times. When the lesion has covered an area of two inches, or perhaps 122 Contributions to Medical Science less, the center loses in part, or even completely, its papillomatous surface, becomes depressed below the level of the surrounding tissue, and either appears red, moist, and granulating, or it may- be dry, and show cicatricial healing. In most cases the disease exhibits periods of rapid progression interrupted by periods of relative quiet. During rapid extension the whole surface is more swollen, tumid, and red, secretes an offensive pus more freely, the areola becomes broader, more turges- cent, and contains a larger number of abscesses, the pus of which is decidedly bloody. In an interim of quietude the surface flattens out noticeably, is dryer, the areola narrower and flatter with fewer abscesses, which now contain little or no blood. Case VII illustrates strikingly the possibihty of rapid pro- gression. During the first two or three months that the patient was under observation, the process remained almost stationary under large doses of iodide of potassium. However, even while he was taking three hundred grains of the drug daily, rapid exten- sion set in and the borders extended from one-half to three-fourths of an inch in a month's time. It then quieted down for about two weeks, when a still more rapid progression developed which sub- sided only after continued ingestion of increased doses of the iodide. The periods of rest and progression were characterized by the features mentioned above. The disposition of the disease to creep gradually from the point of origin to rather distant areas is well illustrated in the case of Gilchrist and Stokes, where it began as a papule over the left mastoid process, whence it encircled the left ear, invaded the left cheek, chin, palpebral, and supraorbital skin, successively, then the bridge of the nose, and finally similar points on the right side of the face; the older areas cicatrizing as the borders advanced. The discomfort attending the process varies from a sensation of soreness to severe pain, depending apparently on the location of the lesion and the degree of temporary virulence and mixed infection. More pain is experienced when it involves a dependent area, or skin which is necessarily subject to frequent tension, as that over the dorsum of the hand (Hyde, Hektoen, and Bevan) ; Oidiomycosis of the Skin and Its Fungi 123 the patient of Gilchrist and Stokes, whose face was chiefly involved, suffered little. That a particular focus may heal completely with no treatment is shown in the cases of Gilchrist and Stokes, Anthony and Herzog, Coates, Stelwagon, and in Cases II and IX of this series. The extension is accomplished by a direct invasion of the contiguous skin, in which it differs from the sporotrichal infection described by Schenck and by Hektoen and from the " blastomycetic " disease in animals described by Fermi and Aruch, and by Tokishige. As stated, heahng begins by the gradual disappearance of the verrucous tissue from the central or oldest part of the lesion. It is difficult to know what becomes of the papillae in all cases; some melt away obscurely, others remain as prominent shafts which become desiccated, horny, or papyraceous, and eventually fall off, leaving an absolutely smooth base. Concomitant with or preceding this, the number of surface abscesses becomes smaller and the discharge of pus consequently less. Eventually the result is cicatricial epidermization. If, however, the process has been so intense as to destroy the epidermis over a considerable area, a granulating surface supervenes, which may continue raw, oozing, and exuberant for months after the remainder of the surface has healed. The same steps in heahng, of course, occur whether the result is accompHshed spontaneously or by medication. As healing progresses, the central cicatricial area becomes larger and larger, the peripheral fringe of papillae more and more narrow and flat, and the surrounding areola gradually disappears. The last vestiges of the disease are usually seen at the periphery of the lesion, in the form of a slightly reddened zone and a few scattered stumpy processes which either are absorbed or dry up and desquamate. In the early stage of cicatrization it is not uncommon to find scattered miliary abscesses, containing a few fungous cells, in the cicatrix itself even when the verrucous tissue has largely disap- peared (Cases IV and VII) . The study of sections shows that the organisms may remain imbedded in the deeper granulation tissue long after heahng has been well started (Case IV). This fact seems to explain the recurrence of the disease when, supposing cure had been established, the patient ceases medication. 124 Contributions to Medical Science The resultant scar is at first thick and leathery, but in the course of a few months or a year becomes exceedingly thin and delicate, freely movable over the underlying tissue, and having, in some instances, a pink color not far removed from that of normal skin. The surface, however, remains somewhat glistening, perhaps slightly wrinkled, and is often marked by telangiectases. The line of demarkation from the surrounding skin is quite sharp. Hypertrophic scars have formed in none of the cases of this series, although Gilchrist noted such an event. The amount of resulting deformity depends naturally upon the area involved. In Case VI the whole dorsum of the hand had been affected, yet the newly formed epidermis, although cicatriform, was perfectly lax and caused no deformity. In Case X also nearly the whole anterior surface of the fore-arm had been traversed, yet the scar is now barely perceptible. The amount of contraction in Case IV is not known, as the patient has been lost sight of. It is chiefly in those cases where the palpebral tissues have been involved that harmful contraction has occurred. A high grade of ectropion results invariably, leading to more or less severe conjunctivitis or keratitis, and even complete loss of vision as in Case VIII; these conditions being the result of exposure and bacterial infections rather than invasion by the oidium. The "protozoic" disease has most frequently begun as a visceral infection (i.e., in the lungs). Rixford and Gilchrist's first case, however, shows that it may originate in the skin. Apparently metastases occur much more readily from the lungs to the skin than in the opposite direction. Whether the disease is primarily or secondarily located in the lungs the subsequent course is very similar to that of pulmonary tuberculosis, where abscesses and miliary nodules are formed. Fever, rising at night to ioi°-io3° F., and falling in the morning, is present until death approaches, when the temperature may be subnormal. Prominent symptoms are cough, profuse expectora- tion, pleuritic pains, and the physical signs common to pulmonary tuberculosis; severe sweating has also been noted. Extreme emaciation and asthenia supervene, and vomiting and stupor may be terminal symptoms. The urine shows nothing characteristic Oidiomycosis of the Skin and Its Fungi 125 and may be quite normal. After general or extensive pulmonary involvement has begun life may be prolonged for four months to a year. Metastatic foci in the skin may appear as subcutaneous nodules and abscesses, which may persist as such for some time before involving the epidermis; or the epidermis may be involved almost from the start. Metastatic skin lesions appear to be more destruc- tive than the lesions of "blastomycetic dermatitis" and cause a freer formation of pus, in which the organisms are said to be present in astounding numbers. Clinically the foci in the skin appear much like those of ''blastomycetic dermatitis." Rixford and Gilchrist describe the same papillomatous surface and cauli- flower appearance. Some of the lesions appear tumor-like, how- ever, resembUng the tumor formation in Mycosis fungoides (Wer- nicke, Posadas, D. W. Montgomery). The cHnical history, physical findings, and the course of Busse's case of Saccharomy- cosis hominis are similar to those of the "protozoic" disease. In both were found metastatic foci in various internal organs and in certain bones (tibia, ulna, rib, and frontal). It seems, however, that the skin lesions in Busse's case did not develop a verrucous surface; that the infection was so virulent, or the patient's tissues so weakened, that the changes were almost entirely destructive. In none of the cases of so-called blastomycetic dermatitis, have there been general disturbances, such as fever, headache, and nausea, and the urine has been persistently normal. Exception must be made in Case V, since, as stated, it is now known that the disease eventually became pulmonary. It appears striking that out of twenty-six cases of "blasto- mycetic dermatitis" only one resulted in general invasion and death, while all cases described as "protozoic" have been fatal. The condition would seem to militate against the conception that the two processes represent a disease unit. This difference in termination seems explainable, however, by the fact that in the "protozoic" cases the infection was usually primary in the lungs, where extension is much easier than in the skin, metastases occur with greater ease, and general toxic disturbances are more easily produced. In one of Rixford and Gilchrist's cases of ''protozoic" 126 Contributions to Medical Science infection, where the primary infection was in the skin, many years passed before general invasion occurred; in which respect it is quite comparable to Case V of this series. The observations are yet too few and cover too short a time to be able to say in what proportion of cases a primary cutaneous infection would lead to general involvement. diagnosis. The process in the skin has now been studied clinically and histologically in so many cases, and with such uniform findings, that a characteristic clinical appearance and course, as well as a characteristic histopathology, may be considered well estabhshed. These have been described in previous pages. It is to be emphasized that the most ready means of making an immediate diagnosis is to examine the contents of the mihary abscess, or a fragment of verrucous tissue, microscopically, for the fungous cells. The pus may be mounted either in a 30 per cent KOH or NaOH solution, or in water, and examined imme- diately. If the alkaline mounts are allowed to stand for an hour or more, the field will be partially cleared of tissue cells. Probably the disease has been most often confounded with verrucous tuberculosis, both cHnically and histologically. Case XI, on the strength of microscopic examination, had been previously called tuberculosis, and Cases IX and XII were considered tuber- culous, cHnically. Furthermore, the enormous hyperplasia of the epithelium suggests carcinoma so strikingly that many cases doubtless have been passed over as carcinoma of the skin. In spite of certain similarities, there are clinical and histo- logical features which separate the disease from both tuberculosis and carcinoma of the skin. It is, first of all, probable that the current conception of verrucous tuberculosis has been colored somewhat by the inclusion of cases under that title which really were examples of oidiomycosis of the skin. Without entering into a detailed discussion of the characteristics of verrucous tuber- culosis it will be sufficient to point out the essential differences between the two diseases. The course of oidiomycosis of the skin Oidiomycosis of the Skin and Its Fungi 127 is very rapid compared to that of tuberculosis, and in only two or three months may cause an ulcer an inch or more in diameter. As extension is more rapid, so is the healing process. In four to six months one may find a lesion two inches in diameter, the center of which has healed largely, while the periphery continues to advance. Such rapid strides of invasion and healing are quite foreign to verrucous tuberculosis. The verrucous processes in oidiomycosis of the skin are soft, friable, succulent, and attain large dimensions; while in tuberculosis they are often smaller, drier, hard, and somewhat horny. The purulent secretion of the former exceeds that usually seen in the latter. Finally, the areola, marked by many minute abscesses, which is so characteristic of oidiomycosis, is entirely absent in verrucous tuberculosis. Guinea-pigs inocu- lated with oidiomycotic skin have never developed tuberculosis; an abscess may result, however, from which may be cultivated the fungus peculiar to the case. The epithelial hyperplasia and cell infiltrate may be very similar in the two diseases; but the amount of cell infiltrate in oidiomycosis greatly exceeds that in tuberculosis. The microscopic intra-epithelial abscesses always found in the former are absent in the latter. In addition, the micro-organisms found in the two diseases are characteristic and differential. The giant cells and tubercles in the two diseases are similar in many respects. (See "Special Histological Features.") In separating oidiomycosis from carcinoma of the skin, a fact of prime importance is that metastases of neoplastic epithelial cells do not occur in the former; where metastases have occurred they appear to consist solely of the fungous cells. The rarity with which oidiomycosis of the skin invades even neighboring lymph glands has already been referred to. Carcinomata of the skin do not become papillomatous ordinarily; certain mushroom-like or cauliflower-Uke carcinomata of the skin, which are spoken of, are conspicuously neoplastic, even clinically. The presence of miliary abscesses in the areola differentiates carcinoma, as it does tubercu- losis, from oidiomycosis of the skin. As to histology, carcinomata do not have histologic miliary abscesses, a budding fungus has never been satisfactorily demonstrated in them. Isolated nests of epithelial cells penetrate the subcutaneous, fatty, and muscular 128 Contributions to Medical Science tissue, and the amount of surrounding inflammatory cell infiltrate is, ordinarily, moderate ; the contrary conditions prevail in oidiomy- cosis of the skin. In regard to the opinion held by some, that the disease may be a form of syphilis, it seems almost sufficient to say that in no reported case of " blastomycetic dermatitis" has a history of syphilis in any form been obtained. The fact that improvement or cure follows the administration of the iodide of potassium seems to be the only justification for this belief. For three reasons this position cannot be held : (i) The fact that certain types of actinomy- cosis improve or heal under the ingestion of the iodide destroys the unconditional specificity of the drug as a differential test for sj^hilis. (2) Oidiomycosis demands for cure a much larger dosage and a longer term of treatment than any case of cutaneous syphilis can withstand. (3) Mercurials are not effective in reducing oidiomycosis of the skin. There seems to be no reason for confounding Kaposi's Syphihs cutanea vegetans with this condition. The histology of syphilis is so clearly different that detailed comparisons seem unnecessary. The diagnosis of the condition of the lungs may be dismissed with the statement that its marked similarity to pulmonary tuberculosis has given rise to confusion. It seems probable that further observation will show that a clinical differentiation can be made by microscopic examination of the sputum. prognosis. That the disease, when primarily in the skin, is not often of great malignancy, is indicated by its tendency to remain localized even for ten or more years. It is not certain, in any case, that generalization ever would result. However, Rixford and Gil- christ's first case of the "protozoic" disease and Case V of this series furnish convincing proof that a fatal generahzation may occur. When confined to the skin, the disease yields readily to proper treatment, with a varying amount of cicatricial deformity. Spontaneous healing is possible, at least in part. In several cases where there were multiple foci, certain of the latter healed entirely with no treatment. As stated, in progressive lesions, spontaneous Oidiomycosis of the Skin and Its Fungi 129 healing occurs in the oldest parts as the border advances. Although it has not been observed, it seems probable that the disease may at times make an absolute spontaneous recovery. TREATMENT. The observation made by Professor A. D. Bevan that the disease improved under the internal administration of the iodide of potassium has been substantiated by the experience of others, particularly Professors Hyde and Montgomery. Three cases have recovered completely, and four more have improved greatly under this treatment, combined with the apphcation of antiseptics to the ulcers. It is, as usual, advisable to begin with about ten grains three times a day, and increase the dose steadily. Improve- ment may begin when twenty grains is reached; this improvement usually has been found to be only temporary, however, making it necessary to increase the dosage to fifty, seventy-five, one hundred, or even to one hundred and fifty grains three times daily. In the treatment of Case VII Httle improvement resulted until one hundred grains was reached, then healing progressed rapidly for a time; it again became stationary and it was necessary to go first to one hundred and twenty, then to one hundred and fifty grains per dose, when rapid improvement again set in. Locally the use of hot antiseptics is of great value in disposing of secondary infection and reheving the soreness. Hot boracic acid and corrosive sublimate solutions have been used principally. It was thought that the latter, when applied continuously, would penetrate the tissues sufficiently to check the infection; this result, however, was not realized in the treatment of Case VII, and the boracic acid applications were resumed. Also no good effect was noted from the use of the tincture of iodine over the surface, or the subcutaneous infiltration of an iodo-potassic-iodide solution. No marked improvement has occurred in any case under merely local antiseptic treatment. Total excision of limited areas has given perfect results (Cases I and IX). This is easily possible, owing to the superficial location of the infection. Thorough curetting followed by cauterization would seem to offer good results, a general anesthetic, of course, I^O Contributions to Medical Science being necessary. When cicatricial ectropion of the eyelids has resulted, care of the eye is quite necessary to prevent conjunctivitis and keratitis. Dr. Dodds, of the Illinois Eye and Ear Infirmary, has corrected the marked ectropion resulting in Case II by plastic operations. The continuous administration of the iodide of potassium, then, in large doses, offers the best hope of cure, where excision cannot be practiced. The necessary duration of treatment varies from three to four months to a year or perhaps longer. With this, however, the use of local antiseptics should be combined. pathogenesis. It has been stated that no part of the skin surface is proof against invasion. The scalp, palmar, and plantar tissues, however, seem provided with considerable resistance to the infection, and the mucous surfaces, in particular, usually escape. (The transition surface of the lower Hp, in Case I, was invaded.) From the statements made in literature it may be gathered, also, that many of the internal organs are equally susceptible, when an internal focus is once estabhshed, as in cases of the "pro- tozoic" disease, Busse's Saccharomycosis hominis, and Case V of this series. The indications are, as previously suggested, that the original point of infection may lie either in the skin or in one of the internal organs, particularly the lungs. In the condition described as blastomycetic dermatitis it seems clear in most cases that the skin was the part primarily invaded. In several cases of the *' pro to- zoic" disease it seems equally clear that the lungs were first invaded. Busse and Buschke do not agree concerning the infection atrium in the former's case, Busse considering it visceral, and Buschke that it was first situated in the skin. No sweeping statement can be made concerning the mode of primary invasion of the skin. Quite surely, however, three of the infections followed traumatisms; in one case a razor-cut (Hessler), in the second a scratch or bite by a cat (Stelwagon), and in the third an abrasion made by the handle of a wooden truck (Case IV). It cannot, of course, be known whether the Oidiomycosis of the Skin and Its Fungi 131 organisms had previously dwelt on the intact skin, whether they were carried there by the instrument of traumatism, or whether they were transferred to the wound after it was made. In Case I it seems probable that there was a pre-existing abrasion of the lower lip, and that the infection was derived from blighted cereal grains, with which the patient was frequently in intimate contact. How the organisms penetrate the skin in the absence of traumatism is not clear. They may have been placed superficially in the horny layer, or deposited in a hair follicle or gland duct ; in such a position the toxins secreted by the organisms would perhaps be sufficient to cause small areas of necrosis, into and through which the organism and its spores might extend. That is to say, an organism may, itself, be able to cause all the traumatism necessary for primary invasion. Unquestionably the disease, when once established in the skin, is easily transplanted to other portions; this is illustrated in the cases of Gilchrist and Stokes, Hyde and Hektoen, Anthony and Herzog, Stelwagon, and by Cases II and VII of this series. In the case described by Coates the patient's condition pre- ceding the development of the foci in the skin affords some ground for the behef that the disease was primarily localized in the chest, and that the numerous points which suddenly appeared in the skin were of metastatic origin. As the visceral ailment finally healed, the conclusion would naturally be that it healed after causing cutaneous metastases. Although this may not be possible, it seems unlikely, in view of the fate of cases which are known to have had internal foci. The three lesions which originally developed in Case IX appeared "at the same time." It seems most probable, of course, that the patient came into external contact with the infection and that the three implantations occurred "at the same time." How the organisms migrate in the epidermis and corium is not clear. There is no evidence that they are motile, and the adult cell is not found in wandering tissue cells. Still they are often seen isolated in the epidermis and corium. As stated elsewhere, it may be that minute spores are produced which can easily be carried in wandering cells, or in the lymph currents. 132 Contributions to Medical Science pathological anatomy. As the gross pathology of the lesions in the skin involvement has been given in detail in the descriptions of the various cases and in preceding pages, it will sufl&ce here to enumerate categori- cally the principal points. Primary lesion. — A reddish papule which is soon transformed into a crusted pustule. Intermediate appearance. — A prominently elevated ulcer, pos- sessing a granular base, secreting glairy muco-pus, which dries to form crusts. Fully developed lesion. — An elevated, coarsely papillomatous, sharply circumscribed ulcer, which secretes pus freely, contains many minute abscesses, and which, cauhflower-like, may overhang the surrounding reddened areola, which also contains numerous miHary abscesses. Later, the central portion is depressed, and almost entirely free from papillae and abscesses, concomitant with heahng and cicatrization. Cross-section. — Two zones are visible, a papillary and a deeper homogeneous zone containing minute macroscopic abscesses and small hemorrhages. The papillary zone is made up of two portions, the deep, which is a palisade-Hke layer of closely united papillae, and the superficial, composed of a smaller number of papillae, which rise above the general level of the ulcer and are separated by clefts. The study of sections shows the following points in relation to the gross appearances: That portion of the surface pus which adheres to the tissue contains many polymorphonuclear, a few mononuclear leukocytes, few or no eosinophiles, no plasma- or mast-cells, masses and layers of desquamated horny material, many bacteria, and very few of the specific fungous cells. The papillae are shafts of the rete mucosum, which are sheathed by a thin, imperfectly cornified layer, and each is suppUed with a central vascularized cone of mesoblastic tissue. Each of the processes constituting the deeper portion of the papillary zone is closely cemented to those contiguous by septa of horny material. The deep surface of the epidermis is very irregular because of the processes which penetrate the homogeneous zone. The homo- geneous zone is largely a mass of infiltrating leukocytes, new Oidiomycosis of the Skin and Its Fungi 133 fixed tissue cells, minute abscesses and vessels. It has supplanted the corium proper, and the superficial portions of the subcutaneous tissue, which have lost all or nearly all their elastic fibers; the remaining collagenous tissue is very edematous. The line of demarkation from the underlying fatty tissue is quite sharp. The miliary abscesses are located chiefly in the deeper epitheHal tissue, and in the homogeneous zone. Many smaller ones, however, are found in the surface papillae. The fungous cells are distributed numerically as follows: (i) the largest number is found in the abscesses of the corium and deep epithelial tissue; (2) moderate numbers are found in more super- ficial epidermal abscesses; (3) a few are encountered between the cells of the rete mucosum, and in the granulation tissue of the corium, unassociated with large accumulations of leukocytes. As stated previously, the areolar abscesses contain ordinarily a droplet of yellowish viscid pus, which, however, during rapid progression of the disease is strongly tinged with blood. It is here that the organisms can be most easily demonstrated, while in the surface pus they are difficult to find. [On the contrary, the organisms were found in large numbers in the surface secretions in both the "protozoic" disease, and Busse's case of Saccharomy- cosis hominis.] It has been noted that the central portion of the lesion may heal readily and permanently, while the periphery is still quite active; and that the scar which eventually forms is thin, pliable, little or not at all adherent, and may have almost the appearance of normal skin. The perfect epidermization, no doubt, depends on the fact that the surface at no time has lost its epithelial cover- ing entirely, the latter being only stimulated to an abnormal and excessive proliferation. Consequently when subsidence occurs, a massive amount of epidermis remains to be utilized. In those few cases, however, where the epidermis has been entirely destroyed over small areas, the result is seen in exuberant, granulating sores, which may not heal even before all traces of the active disease have disappeared. Sections show that the active disease involves only a moderate degree of necrosis; that the changes are superficial, the cell infiltrate not entering the subcutaneous tissue to any 134 Contributions to Medical Science degree; and that the elevation and tumor-like appearance of the lesion is due more to an enormous cell infiltrate, to edema and h>^eremia, rather than to excessive reproduction of fixed tissue elements. These conditions would seem to account for the mild nature of the scar in comparison with the appearance of the earlier lesions. From what is given in literature concerning the protozoic infection, there appears to be no essential difference between the anatomy of its cutaneous lesions and that of so-called blastomy- cetic dermatitis. But in Busse's case of Saccharomycosis hominis the process in the skin appeared more destructive than in any of the cases of this series. The changes produced in animal tissues by inoculation of the organisms depend on the particular organism and the animal employed, and the character of the inoculation. The reader is referred for details to the abstracts of the cases in literature and to the descriptions of the cases of this series. The different organisms vary greatly in virulence. Subcutaneous inoculations of pure cultures commonly cause abscesses in guinea-pigs and rabbits, and may be fatal to mice. Intraperitoneal injections are relatively harmless. When placed intravenously, transient septicemia or fatal pulmonary mycosis may result. The pathogenicity of the various organisms is indicated roughly in the table (pp. 148 f.); the one from Busse's case was highly virulent to animals. The subcutaneous implantation of infected human tissue into animals may produce abscesses from which the organism pecuhar to the case may be cultivated, but tuberculosis never has followed such an inoculation. It seems probable that artificial immunity is not readily produced in animals. SERUM reactions. Professor Hektoen has found that the undiluted serum of a dog which had received successive inoculations of the organism from Case I causes gradual clumping of the organism diffused in bouillon. Several hours elapse before the fullest extent of clump- ing possible is reached. Organisms from other cases show only a sHght degree of clumping with the same serum. All organisms Oidiomycosis of the Skin and Its Fungi 135 grew well in the serum and the production of mycelium was espe- cially noticeable. Normal dog's serum causes no clumping of any of the organisms; abundant mycelium is produced in all cases. The organism from the case reported by Hyde, Hektoen, and Bevan was repeatedly inoculated into the abdominal cavity of a rabbit. It was found that the animal's serum would cause fairly distinct agglutination of the organism inoculated. Further studies are indicated in regard to immunization, agglutination, and allied problems. SPECIAL histological FEATURES. Cornification. — Abnormal or incomplete comification is found in the horny and granular layer, in the depth of the rete, in single cells or groups of cells, surrounding abscesses and giant cells, and in the desquamated rete cells in the intra-epithelial abscesses. On the surface this is manifested by the retention of cell nuclei and keratohyalin and the lack of fibrillation, or by the entire absence of keratohyalin in the cornification process. The proto- plasm of such cells is moderately acidophilic. Occasionally a single cell is seen in the depth of the rete with a shrunken nucleus and strongly acidophilic protoplasm and a few keratohyahn granules. Such a cell may be surrounded by other cells, which are more or less flattened and may or may not contain kerato- hyalin, but are strongly acidophilic; or a single cell may contain another epitheUal cell and one or both be partly cornified. Typical comified whorls occur with or without the intervention of ker- atohyalin; a mass of leukocytes, an organism, or a giant cell is commonly found in the center of a whorl. The epithelial cells constituting the wall of an abscess may or may not be cornified, and if cornified with or without keratohyahn formation. In case cornification occurs the cells are flattened, somewhat fibrillated, and take acid stains deeply. Surrounding centers of irritation such as accumulations of leukocytes, organisms, or giant cells, the prickles are prone to disappear, giving place to a dense, acido- philic, homogeneous fiber. Remote from centers of irritation the prickles are present quite constantly. In the intra-epithelial abscesses are numerous desquamated epithelial cells, which show 136 Contributions to Medical Science atypical cornification. Some are perfectly spherical, but appear dense, take up eosin strongly, and have a shrunken nucleus; such cells may contain other epithelial cells or leukocytes. The epithe- lial cells within abscesses often become fibrillated, some with needle- shaped nuclei, and others in which nuclei are no longer discernible. Atj^ical cornification in blastomycetic dermatitis appears to be due to an abnormal stimulation to growth, brought about by the presence of foreign and toxic elements; such elements appear to be leukocytes, giant cells, and the parasites. The effect is a hastened and imperfect maturation of the surrounding epithehal cells, partial or complete cornification, with or without the inter- vention of keratohyahn. As intimated above, this stimulation may result in epithehal giant cell formation instead of cornifica- tion. In certain fields the granoplasm of epithelial cells may be cen- tered in a particular portion of the cell. This results in the so-called ''cobblestone" appearance. Epithelial giant cell formation. — The occurrence of numerous giant cells in intra-epithelial miUary abscesses suggested to Wells that they might be formed from epithelial cells. The tissue in certain cases of blastomycetic dermatitis shows numerous giant cells in the depth of the rete, which may or may not be surrounded by a few wandering cells. In one case where such giant cells are numerous their origin from epithelial cells seems clear. The first step consists in the presence of some foreign element such as leukocytes or of one or more organisms. Following this, the nuclei of adjacent epithehal cells become smaller and stain more densely, the prickles disappear, the cell body becomes smaller and fuses with that of neighboring cells, constituting a multi- nuclear giant cell with peripheral disposition of the nuclei. Al- though mitoses may occur in this process, it has not been possible to find them. A further accumulation of leukocytes leads gradu- ally to the formation of a small abscess; and so it happens that in most intra-epithelial abscesses there are one or more giant cells possessing epithelioid nuclei. These giant cells contain one or more leukocytes or organisms, or both. Phagocytic epithelial cells. — ^A conspicuous feature is the number Oidiomycosis of the Skin and Its Fungi 137 and the character of epithelial cells, which enclose either leukocytes or other epithelial cells. A single rete cell, which is normal in every other respect, retaining its prickles and customary staining properties, may present a nucleus which is retracted at one point, leaving a space which is occupied by an invading leukocyte. It is in the abscesses, however, that cell invasion or phagocytosis occurs to a pronounced degree. Case I shows many small intra- epithelial abscesses, which contain from ten to thirty or forty such cells in a single section. Presumably the condition begins by an epithelial cell taking up or being invaded by a leukocyte; this occurs either while the former constitutes a part of the abscess wall or after its desquamation. At this point the cell may be partly cornified; and if partial cornification occurs, further phagocytosis or cell invasion is precluded. In the event that cornification does not take place, the invasion continues until the cell is filled with from three to six or eight leukocytes, the protoplasm forming a thin shell and retaining its nucleus in a flattened condition. The protoplasm of the epithelial cell continues granular until the distension is extreme, when it exists as a transparent hyalin film around the enclosed cells. At any time in this process another epithelial cell may be enclosed, and this cell in its turn may take up leukocytes. It has been somewhat difficult to determine whether this was purely cell invasion or purely phagocytosis, or whether both processes took place. In the struggle for existence between the including cell and the enclosed cells the former often loses its vitality and further infiltration occurs. It is not known positively whether or not the leukocytes multiply after their envelopment. Many forms of degenerating and phagocytic epithelial cells appear to be identical with certain so-called para- sites of cancer, described by Piimmer and others. Desquamated epithelial cells may undergo granular degen- eration with virtually no change of form. Occasionally one finds an area of from two to four or five epithe- lial cells which have undergone coagulation necrosis. There may be no surrounding round cell infiltration. The cutis. — The dense infiltration of the cutis is emphasized in all cases. Polymorphonuclear and mononuclear leukocytes, eosino- 138 Contributions to Medical Science philes, plasma- and mast-cells participate in this infiltration. There is in addition a vast increase of the fixed tissue element, including blood and lymph channels. Polymorphonuclear leu- kocytes are often concentrated to form subepithehal abscesses. Mononuclear leukocytes occur in connection with the polymor- phonuclears, but, together with the plasma cells, avoid the areas of most intense reaction. Eosinophiles occur alike in remote areas or abscesses, and may wander into the epithehum. Mast-cells of different types occur near blood-vessels and gland structure especially, but have, moreover, a somewhat general distribution. They are also absent from areas of intense reaction. Eosinophiles. — In a certain type of cases, polymorphonuclear leukocytes are present in overwhelming numbers, and it is noted also that in the same class of cases eosinophiles are present con- stantly in large numbers. These cases are distinguished chnically by the presence of many visible miliary abscesses and by the excessively verrucous appearance of the diseased surface. Such cases have a correspondingly smaller number of plasma cells. A second type of the disease, which is characterized by its some- what milder clinical appearance, possesses fewer polymorphonu- clear leukocytes, and few eosinophiles. It is of great interest to note in this connection that in those cases where eosinophiles and polymorphonuclear leukocytes are present in such excessive numbers, the organisms so far isolated have been of a mould- fungus type. Three of the cases and the one described by Harris are of the extremely verrucous type and presented many miliary abscesses in the surrounding areola. From two of our cases a mould-fungus was cultivated. Culture experiments from another failed. In Harris' case no cultures were attempted. In the case, however, where I failed to obtain satisfactory cultures, there developed on two occasions an organism which in its early condi- tion resembled the two mould-fungi isolated from similar cases. It seems quite possible, then, that the presence of such a large number of eosinophiles may be associated with a special clinical type, and possibly with the presence of the mould-fungus type of organism. At any rate, there is clearly an eosinophilous form of the disease, histologically. Oidiomycosis of the Skin and Its Fungi 139 Plasma cells and fibrous tissue. — It has been emphasized in the descriptions of the histopathology in these cases that the plasma cell occurs chiefly in the subacute areas of the inflammatory process, particularly in the peripheral portion^ of the lesion, where new fibrous tissue is forming rapidly and in large quantities. The fact that plasma cells are abundant in the neighborhood of such tissue suggests a possible relationship between the two. In studying this relationship, it has been of service to use a water- blue and carbol-fuchsin stain. Plasma cells are not stained specifically by this method, but the chromatin takes up the fuchsin somewhat specifically, and the protoplasm stains a sky-blue with the water-blue, leaving certain unstained spaces, which doubtless represent the protoplasmic granules of the cell. In addition, the so-called hyalin bodies are brought out well by the fuchsin, and the finest connective-tissue fibrillae by the water-blue. The sections were stained in a j to ^ of i per cent aqueous solution of water-blue for twenty seconds, treated with water, then in Ziehl's carbol-fuchsin for ten or fifteen minutes, decolorized in alcohol, to which had been added a small amount of the tincture of iodine, and then passed through absolute alcohol and cedar oil into balsam. The polychrome methylene blue has been used as a supplementary and corroborative stain. It is to be noted in the first place that there are relatively few plasma cells visible in blood-vessels. They exist in fairly large numbers immediately surrounding the vessels, and occasion- ally one is seen in the process of exit from a vessel. The occurrence of the cells immediately surrounding blood-vessels is responsible for a rough, columnar appearance of plasma cell groups under a low power. As they exist here, there is virtually no intercellular substance. In a field of plasma cells more remote from blood- vessels, where the cells are supposedly older, it is usual to see a columnar arrangement of a different sort. There are straight or tortuous rows of cells end to end, the columns being separated by connective-tissue fibers, giving off delicate branches, which pass between and surround the plasma cells. In the intercolumnar fibrous tissue are found occasional elongated cylindrical or oval nuclei, characteristic of new connective- tissue cells. They have I40 Contributions to Medical Science a small amount of chromatin, the arrangement of which is in no way suggestive of the plasma cell nucleus. There is a distinct nuclear membrane, and the nucleus itself may shrink from the fiber-Hke cell body, forming a clear perinuclear space. A strictly columnar arrangement of plasma cells is not at all necessary for this relationship to fibrous tissue. Groups of cells everywhere show the same conditions except where they are newly formed. Concomitant changes occur within the plasma cells themselves. Polychrome methylene blue gives a granular protoplasm, which stains deeply, and an eccentric nucleus, which possesses about eight chromophilic masses, arranged peripherally and connected by means of dehcate fibers. Among the older cells this type is departed from. Those having the most intimate relation to new- forming fibrous tissue gradually lose their protoplasmic granules and the volume of protoplasm decreases. The extreme change is seen in a cell whose protoplasm is clear or finely granular, almost entirely unstained, and possessing an eccentric nucleus character- istic of the plasma cell. The cell body disappears by a gradual peripheral disintegration, and the outline becomes uneven and ill-defined. One often sees a plasma cell nucleus, with the merest fringe of protoplasm, exhibiting no afiinity for the alkaline methyl- ene blue. It is interesting to note that where the plasma cells are relatively intact, the surrounding fibrous tissue is correspond- ingly young; and where the plasma cells show the disintegration spoken of, the fibrous tissue is more adult in character. A close study has been made with a view of determining the transition of the plasma cells into fibroblasts, as suggested by H. F. Harris and others, but no facts have been found to substantiate this theory. There is always a clear difference between the most changed plasma cell and the nuclei of surrounding connective tissue; that is to say, no transition forms have been found. That, however, the plasma cell is sacrificed in an indirect way for the benefit of new-forming scar tissue seems very probable. Hyalin bodies in plasma cells. — A change of a different character is seen quite constantly in large numbers of plasma cells, namely, the so-called hyahn degeneration. In unstained sections this substance is seen as glassy, highly refractive spherules, varying Oidiomycosis of the Skin and Its Fungi 141 in size from a small granule to one-third or one-fourth of the volume of the cell body. They are resistant to strong acids and alkalies. They stain well by Gram's method or acid fuchsin, or in Van Gieson's mixture, carbol-fuchsin, and certain of them in alkaline or polychrome methylene blue. All, however, have stronger acidophilic than basophilic properties. In carbol-fuchsin preparations the staining affinities are not uniform. The small bodies commonly stain a deep red, while the larger ones have a lighter but more brilHant color. Occasionally all those in a cell have taken up so little fuchsin that they appear pale yellow. The affinities are equally inconstant for the other dyes mentioned. These bodies are found customarily in intracellular groups. They also occur in groups in lymph spaces and channels and between connective-tissue cells. Cells containing such bodies are altered in shape to accommodate the contents, and in properly prepared specimens a cell nucleus is usually demonstrable, which almost invariably possesses the characteristics of the plasma cell nucleus. Where there are only a few hyalin spherules in the cell, the proto- plasm stains characteristically, but never deeply, with polychrome methylene blue. As they increase in number, the basophilic granules of the cell body decrease. This suggests that the grano- plasm may be transformed into these so-called hyalin bodies. Eventually the cell is consumed or broken through, and the hyalin bodies lie free in the space formerly occupied by the cell, or they find their way into lymph spaces. The nucleus is more resistant than the cell body and keeps its form and staining properties while the cell is still filled to bursting with hyahn bodies. The dissolution stage of the plasma cell nucleus has not been observed. Mast-cells. — In the study of mast-cells, Unna's polychrome methylene blue and carbol-toluidin blue have been used to stain the specific granules. The question arises as to whether all cells whose protoplasmic granules have a specific affinity for the red in these solutions should be considered mast-cells. I have not arrived at a satisfactory conclusion on this point, but for the sake of con- venience it seems justifiable to consider them in one group. As seen in the tissues under discussion, they fall into the following classes: 142 Contributions to Medical Science 1. The leukocytic mast-cell. — These occur especially in the immediate vicinity of blood-vessels, about which they may be closely packed. They measure from five or six to ten or twelve microns in diameter; are spherical or cuboidal in shape; have a relatively large, intensely staining nucleus, from three to six microns in diameter, and a number of small protoplasmic granules, which are stained bright red. A single cell may have only from six to twelve granules, or it may be densely packed with them. In one place two of these cells were seen within a blood-vessel. Just outside were several which had two small nuclei each; these apparently were dividing cells. They were so numerous in certain fields that they gave a reddish tint to the field with a low-power magnification. From their size, their relation to blood-vessels, and their actual presence within blood-vessels, the inference follows that they are of leukocytic origin. We have not been able to determine their fate. 2. The mast-cell of connective-tissue type. — This is the classical mast-cell, spherical, oblong, cuboidal, spindle, or grotesquely shaped. Apparently its shape is determined by its surroundings. Probably the most characteristic feature is that several processes extend in three or more directions. The spindle form is seen very often. A number of these cells have been found whose protoplasm at each pole dwindled away into long clear fibers, the number of specific granules being very few. The nucleus of this variety is oval or cylindrical, and stains a pale blue. Two or three granules of chromatin stain deeply, and have no characteristic arrangement. It is very similar to the nucleus of the young connective-tissue cell. This type is found in the papillary tissue, and they may enter the prickle-cell layer; they are also found in definite masses around hair follicles and sebaceous glands, sweat glands and their ducts, and in the connective-tissue framework of muscles and adipose tissue, and imbedded in the walls of larger blood-vessels. Their origin and fate are alike obscure. The idea that they are mucinoblasts (H. F. Harris) I am not now able to affirm or refute. On a morphological basis, some appear to be transformed into connective-tissue fibers. That they also give off some substance which may be of use to surrounding cells is shown by the next type. Oidiomycosis of the Skin and Its Fungi 143 3. The mast-cell which possesses a halo. — Certain mast-cells are surrounded by an ill-defined, structureless zone, which is tinted a homogeneous faint red by the polychrome methylene blue. It seems probable that the same substance which in the cells attracts the methyl red has become diffused into the surrounding tissues, and retains the original tinctorial affinity. This accounts for the diffuse red color seen in some fields where plasma cells are numerous. 4. The plasma-mast cell. — This name is applied to certain cells which possess characteristics of both the plasma- and mast-cells. They are spherical, ovoid, or cuboid; the protoplasm densely granular, and taking up both the red and the blue of the poly- chrome. The blue commonly predominates, giving a violet, in which a distinct trace of red is visible. However, the proportions of red and blue taken up are inconstant. They occur where other mast-cells are found, mixed with the dense cell infiltrate, and in the stroma of muscle and adipose tissue. The nucleus in its morphology may or may not resemble that of the plasma cell. The clear homogeneous nucleus seen in certain mast-cells occurs here at times. Often it is entirely obscured by the deeply staining protoplasm. There seems to be no clear relationship between plasma cells and mast-cells. It is certain that it is not the common fate of either to be transformed into the other. The cellular nodules. — The tuberculoid nodules and multinu- clear giant cells of cutaneous oidiomycosis possess few features to distinguish them from those of tuberculosis. The nuclei of giant cells have the same disposition as in the latter disease. In- stead of tubercle bacilli, the fungus cells are enclosed; and it is more common to find vacuoles and leukocytes in the protoplasm. It seems probable that the vacuoles are spaces formerly occupied by organisms or leukocytes. The nodules may resemble closely the tubercles of tuberculosis. Central necrosis, however, is very unusual, there is not ordinarily a well-defined zone of endothelial cells, and the whole mass is infiltrated with leukocytes; they are more vascular than those of tuberculosis. Tubercle bacilli have not been satisfactorily demonstrated in 144 Contributions to Medical Science any reported case, and nothing similar to them has been found in the cases I have studied. A COMPARATIVE STUDY OF THE ORGANISMS. For consideration we have seventeen organisms cultivated from cases of oidiomycosis or blastomycosis of the skin, including the cases in literature, the one from Busse's case of Saccharomy- cosis hominis, the one cultivated by Curtis from myxoma-like tumors, and the one by Ophiils and Moffitt from the so-called protozoic disease. A mould-fungus noted by W. D. Montgomery on tubes inoculated from lesions of the protozoic disease was thrown away without being studied. Aside from this, there are several cases of skin oidiomycosis and five of the protozoic disease from which cultures were not obtained, but in which the character- istics of the organisms in the human tissue are described. Since the various organisms have not been studied on uniform lines by their discoverers, the desired thoroughness in comparison cannot be realized. For example, the facts as to indol formation, acid production, endogenous spore-formation according to pre- scribed conditions, details as to fermentation and temperature limits, have not been accurately described in reference to this or that organism, and often points in morphology and modes of reproduction are not always clear. However, differentiation and comparison may be made with reference to certain points which are given with fair uniformity. General cultural characteristics and reactions, the morphology of the organisms in tissues and cultures, their methods of repro- duction, and their pathogenicity constitute points which may be utilized in making comparisons. I shall refer to the organisms by the numbers used in the chart, where cultural characteristics are indicated briefly; supplementary statements, however, are neces- sary for the sake of clearness. Most of the organisms develop on artificial media in from one to two days after inoculation. Certain others require from two to seven days. It is of interest that the latter organisms are those which are particularly prone to form myceHum. Those organisms which form a smooth, paste-like surface develop very Oidiomycosis of the Skin and Its Fungi 145 readily; those with irregular surfaces and aerial hyphae develop more slowly. As noted, most organisms grow readily on ordinary media, either at the room or brood-oven temperature. Faintly acid, saccharin, or glycerin media, or a wort preparation and potato constitute a favorable media. In our experience the maltose agar of Sabouraud, made sHghtly acid, is the safest and surest single preparation. The observation that mycelium forms abundantly in dog's blood-serum suggests that a preparation of this material may prove a suitable medium. As to the surface appearance on solid media the organisms fall into three classes, of which organism 5 is the type of one, 11 of another, and 12 of the third. The first group presents in young cultures elevated, moist, soft, white colonies which soon coalesce to form a fleshy growth of paste-like consistence. The second type at first presents a granular surface, sHghtly elevated, and as the colonies coalesce the elevation becomes more pronounced, the growth incorporates itself with the medium, and eventually the surface appearance is that of a piece of crumpled cloth or a niass of coiled earth-worms. The third type is represented by those fungi which produce aerial hyphae, the surface being white, dry, and flour-like, or exhibiting hyphae, which eventually cover the inner surface of the tube. Those following the first type are easily broken up and smeared over the surface. Those included in the second and third types are firmly incorporated with the medium and can be broken up only by loosening fragments of impregnated sub- stratum. The surface of these organisms in the first type becomes dirty or slightly brown with age. This also shows through the reverse surface of the tube. The reverse surfaces of the second and third also show color changes, the second becoming a dirty brown, and the third exhibiting a rich golden-brown color, like that of a well-colored meerschaum. The organisms of the first type primarily send segmented hyphae into the substratum; this property is lost eventually during the artificial life of the organism, only a surface growth appearing. The second and third groups persistently infiltrate the substratum. In liquid media, also, the same groups produce three corresponding appearances. 146 Contributions to Medical Science The first occurs in the bottom as a flocculent sediment, the super- natant fluid remaining clear; shaking disseminates the growth and gives the medium a milky appearance. The second type, which is strongly mycelial, produces a membranous top growth, and in addition coherent masses or tufts of mixed hyphal and spherical elements which lie in the bottom of the tube and adhere to the sides. As the top membrane becomes more dense it sinks to the bottom and a new membrane develops. The organisms with a pronounced aerial growth form at first a single coherent mycehal tuft. After two or three weeks, because of the separation of spores and segments, other small tufts appear. A top growth is not formed, and the intermediate fluid is unclouded. Likewise on gelatin or agar plates the three groups remain distinct. The first forms closely granular colonies, the individual cells of which can be seen with a low-power objective. Films from such colonies show only a mixture of spherical and elongated cells. To the naked eye they appear as round, white, moist, elevated points, and the deeper colonies are much smaller than those on the surface. The second type produces colonies which to the naked eye are somewhat glistening. Under a low power they are seen to consist of radiating, segmented, and branched hyphae, the contour of the colony being almost circular. The third type produces colonies of a feathery naked-eye appearance, the processes being of un- usual length. Under the low power this is seen to be made up of branching hyphae which are closely segmented and beset with numerous lateral, unicellular offshoots. As the hyphae reach the surface a moulded appearance develops. Organisms 11 and 13 (see comparative table) produce acetic acid. Organisms 3, 4, 6, and 12 possess no fermenting powers. A large percentage ferment glucose and maltose solutions. Organism 2 is unique in being able to invert and ferment saccharose and to ferment lactose solutions. In all cases where fermentation results CO2 and alcohol are formed; two organisms only produce acetic acid in the process. Organism 10 is unique in its ability to produce indol, using Theobald Smith's method. Oidiomycosis of the Skin and Its Fungi 147 All are obligate-aerobic in their growth, but No. 2 existed in a latent state for a number of days without the presence of oxygen. Freezing has not killed those organisms with which the experi- ment has been tried; the thermal death point is in the neighbor- hood of 45° to 58° C, for from two to five minutes. They resist desiccation. Explanations. — Cases are referred to by the names of the observers; and those composing the present series, by the Roman numerals used in preceding pages. The numbers indicate chronological order (except 17), and correspond to numbers used in ''Comparative Analysis of the Organisms," p. 144. Letters refer to notes on following page. The signs for the most part are self-explanatory. + : affirmative. — : negative. H — , or — f- : inconstant, or in moderate degree. ? : not clear, or satisfactory. Blank squares: no information at hand. la. Buschke observed that short threads sometimes formed. ih. Demonstrated by Busse, using H. Moeller's method. 2a. Does not grow on blood-serum. 2h. Not freely. 2c. Noted by Anna Stecksen. 3a. When myceUum ceased to form, the surface appeared gelatinous. 36. This is not mentioned by Gilchrist, but Buschke observed it in cultures sent to him by Gilchrist, who observed only a prickly surface. 3c. The maximum Hmit for growth is 40° C. 4a. In a few days, as stated by Hessler. In this laboratory its development is noted after twenty-four hours. 46. A "little smaller than Gilchrist's" (Hessler). 5a. Observed recently. 6a. Through the medium of a guinea-pig which had been inoculated with fresh tissue. Is Si a 3 □ 3 3 O " tumefaciens Oidium dermatitidis Oidium Hek- 'c a = .2-" 5^ S2 Oldium Hek- toenii II(io,o) Oidium aceticum Oidium hyphomycoide Oidium hyphomycoide Oidium hyphomycoide Oidium hyphomycoide Oidium hyphomycoide Terms Which Have Been Applied TO THE Various Organisms S3 11 « 2 K E2§ m .2 toT3 n 5 STVHINV 01 AiDINaOOHXVJ +T + + + A, + ; + + o § ei B< H Pi "3 u 3 1 saiodsoDsy 1 1 1 1 1 -' ' 1 + + + + ^ 1 1 1 1 Btprao3 jEiajE-j 1 1 1 1 1 1 -x. 1 1 + + + + + ^ cd to II 3 3 sajodsopng | JO sajodsoDsy | 1 1 ^1 (V. (V. + ' + + ru ru ru fu + + + + ■Eiprao3 iBurtn -J3X JO IBJ3JB7 + + + - -+I + + + + + nouBjuara -Sa's I^qdXH 1 + + + + - ++ + + + + + + aaippng | ++ + + + + + +++ + + + + + + 3 saiodsopnj 1"-- (-c f\- - ru '^"' "" + ^ ^ A- ft. , + + + + + noijB^uaniSas q^iM SBqdXjj II III l'^ 1 1 1 1 1 1 Stnppng | ++ + + + + ++^ + + + + + + >> o s o n o 1 §■ 2 (I )saiodsopn3-opn3sj + + *l + + ++ + + + + + + + + + + snapn^ + 1 1 /%- '^ '^ 1 1 ri. ri. A. n. 1 + 1 1 saionD^A | ++ + + + + + + + + + + + M iBfnu'BiQ Xiasj'BOQ | +-|- + + + + + + + + + + + h4 jBinireio /{pnij | ++ + + 1 + + + + + + + + ^ "3 a c3 51 S3jniin3 1 ++ 1 1 1 1 1 1 1 1 1 1 snssix ++ + 1 + 1 ^ 1 + li — ■& 2 Oc S3inipi3 1 1 ++ + + + + +++ + + + + + + anssix 1 ++ + + + + ++<^. + + + + + + + acqdXH pajmof paqDUBjg s3Jnipio ni 5J + + + ++ + + + + + + + anssjx HI 1 1 1 1 1 1 r^ 1 1 1 1 1 1 1 o o ?r to ro o I HI .O 1 O >o O t<5 O O o o -o o o MM M PAO S3IIHP13 ni 1 ++ + + + + ++ + + + + + 4- snsstx ni | | 1 1 A. 1 1 1 1 1 1 fBDuaqds saiTupio ni 1 ++ + + + +++ + + + + + + snssjx 11 1 ++ + + + + ++-- + + + + + + + ONiAHa xa aaruji | 1 1 1 III | SXD aHX ;moj-qjB3a « Snizaai J Xq paiirj | 1 1 1 oiaoHSV aivonao | ++ + + + + + + + a: o 1 1 2 Oh PPV apaay | + 1 1 1 + 1 joqoDiv 1 ++ + + + + 1 'OD 1 ++ + + + + 1 9sopbt| :^ I I I I III asojiBj^ 1 + + + + + 1 asoanio | ++ 1 + ^ 1 + + 1 asoi'BqDD'BS JO nopBinanuaj pnB uoisjaAnj + 111 III |.« to ■* u> vO f^ 00 O* O M C» W M M to •* V) \o t^ nopBinaBoo I I I I I I P-iinsN I + +++ ann^^nv I I PPV I I III l + l + 1 I I 1 + 1 + NOixo0aoHriO «J a 0*0 r^ * 'r h ** fft O S^ „s>>^>^>« t/i k- o cA^ S^ rt-— ^ v\ 'pe; 3 probably should be placed here, and also 9, so far as Gilchrist's description indicates; the last, however, has not been completely worked out. Third, these are the mould-fungi. They have all the prohf- erative possibilities of the first and second groups, and produce in addition aerial hyphae, which bear lateral conidia capable of genmiation, and occasionally form terminal ascospores, or naked spore groups. It seems quite certain that the cell proto- plasm may be broken up into large numbers of spores, but com- plete proof has not been obtained. Organisms 12, 13, 14, 15, and 16 constitute the group, and in addition 17, so far as described. Buschke, in studying Gilchrist's organism No. 3, observed that aerial hyphae formed on certain media, assigning to it character- istics belonging variously to hyphomycetes, oidia, and blastomy- cetes. To sum up the question of reproduction: All organisms repro- duce in tissues by gemmation, and possibly by endogenous spore formation, the latter not being capable of proof at present; in Oidiomycosis of the Skin and Its Fungi 159 cultures, by gemmation, lateral conidium formation, and mycelial spore formation following the oidium t>'pe; those most highly differentiated by the production of conidium-bearing aerial hyphae, in addition to all the other methods. There is thus an ascending scale as to complexity, the lowest represented by sac-like budding cells, the highest by the mould-fungi. Endogenous spore forma- tion in cultures is probable, but not proved. The various organisms produce the same type of lesions in animals, but there are clearly different degrees of virulence among them. (See the summary of literature and description of our cases.) classification and nomenclature. It is well known that the micro-organisms of the simplest con- struction, as staphylococci, are capable of considerable variations. Diphtheria and tubercle bacilli show greater variations in accord- ance with their more complex structure. Brefield, Hansen, Jor- gensen, and others cite many examples of variation in frutification and morphology among the genera and species of hyphomycetes. "Brefield showed that many Ustilagineae (smut-fungi), basidio- mycetes, and other fungi may enter upon a budding fungus stage." Sabouraud fell into great confusion by the almost numberless varieties of trichophytons he obtained from various cases of ring- worm, which were identical clinically. Hansen has showed how easy it is to produce new species of yeasts. From an isolated cell he succeeded in obtaining three distinct species: one which always, one which rarely, and one which never formed endogenous spores. Jorgensen states that a certain variety of Saccharomyces cerevisiae grows with the usual ascus-like cells at 27° C, but at 75° C. grouped colonies with mycelium-like branches are obtained. What is considered endogenous spore formation in the saccharomy- cetes is retarded or accentuated by slight variations in external conditions (temperature, age, the chemical qualities of the sur- rounding medium, etc.). "The majority of saccharomycetes are only known as budding fungi with endogenous spore formation. In the minority a mould i6o Contributions to Medical Science stage is also known, which bears a certain resemblance to dema- trium, oidium, or moniha" (Jorgensen). Among the lower fungi it seems that the higher the state of organization the greater the possibilities of variations in mor- phology in a particular genus or even species. Continued study of disease processes is leading to the conclusion that a particular morbid condition is associated invariably with a particular cause or a group of causes. We are drawing closer constantly to the ideal of specific etiology. In actinomycosis and trichophytosis, however, we have two examples of morbid conditions caused by groups of organisms, the individuals of which may show wide morphological variations. Those who have studied them, however, have concluded that the members of each group are very closely related, and consider them as members of a common genus or as specific adaptation forms. It is of the greatest importance to recognize that in the blasto- mycetic dermatitis of Gilchrist, twenty- three cases having been given careful clinical and histological study, we have a disease entity which is uniform clinically and histologically, except for unimportant details. It has been separated absolutely from some other diseases which have one or more points of similarity to it. Of equal importance is the fact that three different types of organisms, as described above, appear to have etiological rela- tionship to blastyomcetic dermatitis. A priori, one would conclude that the various organisms should be closely related, having in mind the similar conditions in actinomycosis and trichophytosis. The study of cultures shows some constant differences among the organisms, and, on the other hand, many more common points. All may proliferate by budding, by an oidium-like segmentation of hyphae, and by lateral conidium formation. One type is unique in producing aerial hyphae, another is conspicuously oidium-like, and the third blastomyces-like. The important characteristics not common to all, so far as we know, are aerial fructification and fermenting powers. The latter is absent from the mould-fungi, and from four of the blastomyces-like organisms; hence this property is not an adequate differential point. It follows, then, that aerial fructification is the most striking differential feature. Oidiomycosis of the Skin and Its Fungi i6i Is this, then, sufficient to disallow classification of the organisms under a common genus ? Having consulted the works of De Bary, Curtis, Cooke, Under- wood, and others, I find that there are not greater differences between the various organisms under consideration than are commonly allowed to certain genera of the hyphomycetes. With this as a basis, and considering the common source of the organisms, the conclusion is reached that those cultivated from the various cases of blastomycetic dermatitis should be placed in a common genus. It may be assumed that the ability to form aerial hyphae has in some cases been suppressed, and other differences are explain- able along the same line. Difficulty arises when the attempt is made to locate this genus in a botanical system, and pathologists cannot hope to succeed where botanists have failed. In concluding that they are hypho- mycetes, we have in mind the statement of botanists that many hyphomycetes are merely conidial stages of certain ascomycetes; yet considering the most highly differentiated (the mould-fungi) as the appropriate type of the group, the character of the fructifi- cation and the abundant formation of hyphae indicate their posi- tion under hyphomycetes. Further classification can be accomplished only by making comparisons with some pre-existing and accepted form. In concluding that these organisms belong to the oidia, it is with the knowledge that the position of the genus is uncertain, and that in it are included some forms which are said to represent conidial stages of higher organisms. However, among the organisms studied, the morphologic characteristics of oidium are quite con- stant, and constitute the most distinguishing feature of the group. The fact that budding takes place in all is not sufficient justifica- tion for considering them blastomycetes, or saccharomycetes, as this method of proliferation is likewise found in the conidia of many fungi, particularly the oidium forms; it seems clear that the term " blastomyces " is not sufficiently broad. The similarity of the organisms described by Busse, Curtis, and by Ophiils and Moffitt to those studied in this article suffices to place them in the same group with the latter. 1 62 Contributions to Medical Science The specific names applied (see Table i, pp. 148 f.) represent peculiarities of the organisms as far as possible. In accordance with the name applied to the organisms, the infections in man should probably be spoken of as oidiomycosis, and the disease in the skin as oidiomycosis of the skin. GENERAL SUMMARY. It has been generally considered that thrush is the only example of a disease caused by an oidium, the process being limited largely to the mucous membranes of the mouth, pharynx, and esophagus. Observers have shown, however, that the parasite of thrush may penetrate to the submucosa of the pharynx, esophagus, and stomach, and even produce thromboses with consequent ulcera- tions. Of greater importance are the observations of abscesses in the brain, Uver, and kidney, and of abscesses and areas of consoli- dation in the lungs, the conditions being associated with the presence of organisms of the oidium type. In the most of such cases there was a concomitant mucosal thrush, and so far as stated there was no mixed infection in the abscesses. Also, organisms not differing essentially from Oidium albicans have been cultivated from cases of tonsilitis, thrush, and infections in other parts of the body, and have been termed saccharomycetes or blastomycetes by their observers. The natural pathogenicity of organisms of this type extends to certain animals, notably horses, which in Japan, Italy, Russia, and some other countries suffer epidemically from a stubborn burrowing infection of the skin, superficial lymphatics, and upper respiratory passages. Fermi and Aruch, and Tokishige have cultivated "blastomycetes" from the lesions of such cases, and proved the specificity of the organisms for the disease. Guinea- pigs also appear to suffer from invasion by blastomycetoid organ- isms. Aside from the theoretical relationship of yeast-fungi to malig- nant tumors, the current impetus to their study originated in Busse's discovery of a generalized fatal case of " Saccharomycosis hominis" in 1894. Curtis followed with a somewhat similar case. Gilchrist in 1896 greatly increased the general interest in the Oidiomycosis of the Skin and Its Fungi i6 o subject by the description of his second case of ''blastomycetic dermatitis" and the organism cultivated from it. Since that time the number of reported cases of the infection in the skin has increased to twenty-six. Of this number fifteen have been dis- covered in Rush Medical College, the present report including twelve cases. (As noted, some of these cases have appeared in the literature previously.) Clinical study was possible in ten of the cases, the other two being museum specimens which were origi- nally considered tuberculous. All were given thorough histological study. It was possible to attempt the cultivation of organisms from nine out of the twelve cases. Only seven permanent cultures were obtained, however, the eighth dying before it could be studied and another case yielding no oidium or yeast-like organism. There is, essentially, a uniform clinical history, which is briefly as follows: 1. An incipient papule which becomes pustular, yielding a glairy, somewhat tenacious pus. 2. Gradual extension of the ulcerating surface, which is soon covered with coarse, soft, and friable papillae, and is surrounded by a reddened areola in which many minute abscesses are visible. 3. Cicatricial healing in the oldest portions of the lesion as the border advances. 4. A variable amount of pain, depending on the site involved, upon a temporary increase or decrease of the virulence of the specific infection, and the amount of secondary infection present. 5. The absence of lymphatic glandular involvement in a great majority of cases. (It seems probable, however, when generalized infection occurs, that it may take place through the medium of the lymphatic system.) 6. The protracted and progressive course of the disease, which may extend over ten or twelve years, and in the face may cause great cicatricial deformity. 7. Periods of rapid extension, interrupted by periods of relative quietude. 8. The tendency of the scar to eventually approach the appear- ance of normal skin. 9. The upper extremities (hand and fore-arm) and the face are 164 Contributions to Medical Science most frequently attacked, although no portion of the skin is proof against invasion. 10. Extension to mucous surfaces does not occur readily. 11. The absence of general toxic disturbances, attributable to the local infection. The gross anatomy is characteristic and uniform in all essen- tials. In addition to the features mentioned in i, 2, and 3, above, naked-eye inspection of a cross-section shows, from without inward : 1. A papillary zone, composed of a superficial layer of isolated villiform processes, and a deeper layer of similar processes which are united side by side. 2. A homogeneous, vascularized, grayish-red, cellular zone, in which are formed minute abscesses. 3. An unaltered layer of subcutaneous fat, as the limit of deep extension. A positive diagnosis is readily made by microscopic demonstra- tion of the organisms in the pus of the miHary abscesses, or in the verrucous processes as described. Stained sections of the tissue exhibit the following histological features: (i) a vast amount of " carcinomatoid " epithehal hyperplasia; (2) minute intra-epithelial abscesses; (3) a granulo- matous condition in the corium, characterized by masses of plasma cells, minute abscesses, and tuberculoid nodules and giant cells; (4) the presence of the spherical, capsulated, budding organism, particularly in the epidermal and subepidermal abscesses, but also distributed unevenly and in small numbers in epithelial masses and granulation tissue. The study of plasma cells as seen in the skin infection indicates that they play some part in the production of new fibrous tissue; not, however, being transformed into connective-tissue cells, but probably providing some substance which is utilized by these cells in the process of fiber formation. The fate of the plasma cell in this relationship is gradual disintegration, and the result accom- plished is a peripheral zone of new fibrous tissue as an obstruction to extension of the infection. Mast-cells proliferate extensively, and morphologically and Oidiomycosis of the Skin and Its Fungi 165 tinctorially are of four types: (i) leukocytic; (2) connective- tissue cell type; (3) the plasma-mast cell; (4) the mast-cell with a halo. Their fate and function have not been made out, but it is not clear that they are " mucinoblasts " (Harris); their anatomical distribution speaks against this theory, and the tinc- torial test is not conclusive. It has been made out that in many cases the multinuclear giant cells seen in the epidermis and intra-epithelial abscesses are of epithelial origin. Those cases from which a mould-fungus has been isolated distinguish themselves from all others by the presence of great numbers of eosinophils in all parts of the diseased tissue. The organism in human tissue has an average size of about ten microns and never forms threads similar to those seen in cultures. Proliferation (in tissue) is by gemmation; it seems probable, also, that endogenous spores may form. In addition to blastomycetic dermatitis, a number of cases of a generalized infection with concomitant cutaneous lesions have been described (particularly by Rixford and Gilchrist), the disease having a close clinical similarity to diffuse, ulcerative, and miliary tuberculosis. Rixford and Gilchrist, and others, have considered the infection protozoic in nature, because of the structure of the large, spherical, capsulated organisms, and their inability to grow them on artificial media. From a case of this type, Ophiils and Mofl&tt recently cultivated a mould-fungus pathogenic for animals. This mould-fungus appears identical, morphologically and patho- genetically, with that cultivated from five cases of this series. The cutaneous lesions in the so-called protozoic disease, moreover, are identical in gross and histological appearances, with those of oidial dermatitis, all the evidence pointing to a close relationship between the two conditions. The organisms which have been cultivated by various observers from oidiomycosis of the skin fall into three groups: (i) a blasto- mycetoid; (2) an oidium-like; and (3) a hyphomycetoid group. All the organisms exist in tissue only in the blastomycetoid form. On culture media the first group exists chiefly as spherical or oval budding cells, but under suitable conditions may produce abundant 1 66 Contributions to Medical Science mycelium. The second forms submerged myceKum which breaks up into chains of "spores"; while proliferation by budding is not a prominent feature. The third produces fruit-bearing aerial h>'phae, submerged hyphae similar to those of the first two groups and is also capable of multiplying by gemmation. Hence from the first to the third group there is a gradual transit. The specific characteristics of individual organisms remain, for the most part, constant, so it seems that an organism belonging to one group is not readily induced to grow Uke those of the group higher or lower. Sufficient variations, however, have been noted, to point to the possibility of a lower organism acquiring some of the properties of a higher, or vice versa. (Quite recently the mould-fungus from Case VIII began to grow with the peculiar folded surface produced by the organism from Case III, losing its mouldy surface at such places.) It seems probable that those organisms which grow chiefly as budding fungi or in the form of segmented threads are examples of a reversion, on the part of a higher organism, to a more simple or continued conidial form. The organisms, further- more, fall into two physiological divisions: the fermenting, includ- ing the first and second groups; and the non-fermenting, the third group. The morphological and biological limitations assigned to oidium by botanists indicate the placing of all the organisms cultivated from cases of " blastomycetic dermatitis," Saccharomy- cosis hominis, and probably the so-called protozoic disease in this as a common genus, the different species being well differentiated. The pathogenicity of the different organisms for animals varies. Subcutaneous abscesses, septicemia, pulmonary consolidation, and pyemic foci have been induced by various methods of inocula- tion. Attempts to reproduce a typical lesion in man by inocula- tion of pure cultures have, so far, failed. Since susceptibility to the infection seems to be quite rare, these few failures to repro- duce the disease in man obviously cannot be regarded as definitely settling the question. On the other hand, the claim that these organisms are specific for the disease is not out of range of strict, technical criticism, until the lesions seen in man are reproduced by the inoculation of pure cultures into animals or man. Aside Oidiomycosis of the Skin and Its Fungi 167 from this crucial test, the etiological position of the organisms seems unquestionable. Until the distinguishing features of the disease are grasped, it is natural to expect adverse and biased criticisms, a noteworthy example of which appeared recently.^ conclusions. 1. The so-called protozoic disease of Posadas, Wernicke, and others; Busse's and Curtis' Saccharomycosis hominis; and Gil- christ's blastomycetic dermatitis are various manifestations of the same disease. 2. The condition in the skin possesses constant clinical and histological characteristics, which separate it positively from all other skin diseases, particularly verrucous tuberculosis, carcinoma, and syphilis. 3. The organisms isolated from various cases differ in minor respects among themselves, but are so closely related morphologi- cally and biologically as to justify their inclusion in a common genus, oidium; they are thus analogous in a pathogenetic sense to the fungi which cause actinomycosis, and to those causing tricho- phytosis. 4. The variations among the organisms allow the recognition of three morphological types: (i) blastomycetoid or yeast-like; (2) oidium-Hke; (3) hyphomycetoid. 5. There are two histological forms of the disease in the skin, the eosinophilous and the non-eosinophilous, the former being associated with the mould type of the organism. 6. In accordance with conclusion 3, oidiomycosis is an appro- priate term for the conditions caused by the organisms, and Oidiomycosis cutis for the disease as it occurs in the skin. 7. Aside from the infections considered in this communication, certain cases which have been described in the literature from time to time indicate that oidium-Hke organisms may cause other severe pathological conditions in man. ' Kromayer, Ernst, Centralbl. filr allg. Path, und path. Anat., igoo, ii, p. 777. 1 68 Contributions to Medical Science BIBLIOGRAPHY. ANTHO>fY, H. G., AND Herzog, M. "A Case of Blastomycetic Dermatitis Engrafted on Syphilitic Ulcers," Jour, of Cutan. and Gen.-Urin. Dis., 1900, 17, p. i. Baldwin, L. Blake. Jour. Am. Med. Assn., 1900, 34, p. 292. Behla, R. "Ueber die systematische Stellung der Parasiten der Miescher'schen Schlauche und deren Zuchtung," Berlin, ihierdrztl. Wchnschr., 1897. Abstr. in Baumgarten's Jahresbericht, 1897, 18, pp. 752, 787. Berg. Arch. f. Anal. u. Phys., 1842, 201. . Clinique des Hopitaux des Enfanls de Paris, 1842. Cited by Robin. . An. d'Anat. et physiol. Pathol., 1846. Cited by Robin. Bernard, Cl. Leqons de physiologie e.xperimentale, 1855, 239. Brandenberg. "Angina oidica bei Erwachsenen," Cor. Bl.f. Schweiz. Aerzte, 1893, 709. Brayton, a. W. Indiana Med. Jour., 1900, 18, p. 403. Brazzola, S. "Contributo alio Studio dei Saccaromiceti patogeni," Bull, delta scienze med. di Bologna, 1896. Abstr. in Baumgarten's Jahresbericht, 1896, 12, p. 660. BtnsT, J. Vaccinia and Variola. A Study of Their Life History. London, 1888. Abstr. in Baumgarten's Jahresbericht, 1888, 3, p. 48. Buschke, A. "Ueber Hefenmykosen bei Menschen und Thieren," Volkmann's Samml. Klin. Vortrdge, 1898, No. 218. . "Ueber Hautblastomykose," Verhandl. der Deutsch. Dermal. Gesellsch., 1899, 6, p. 181. Busse, Otto. "Ueber parasitare Zelleinschlusse und ihre Zuchtung," Centralbl.f. Bakt., 1894, 16, p. 175. . "Ueber Saccharomycosis hominis," Virchow's Archiv, 1895, 140, p. 23. . "Experim. Untersuch. iiber Saccharomycosis," Virchow's Archiv, 1896, 144, p. 360. Die Hefen als Krankheitserreger. Berlin, 1897. Cao, Giuseppo. "Oidien und Oidienmykose," Ztschr. f. Hyg. u. Infeklionskr., 1900, 34, P- 282. Casagrandi, O. " Saccharomyces ruber (Demme)," Ann. d'Ig. sper., 7. Ref. m Baumgarten's Jahresbericht, 1897, 13, p. 751. CoATES, W. E. "A Case of Blastomycetic Dermatitis — Clinically and Histologically, an Epithelioma," Medicine, 1900, 6, p. 100. CoLPE. Arch.f. Gynak., 1894, 47, p. 635. Cooke, M. C. Introduction to the Study of Fungi. London, 1895. CoRSELLi, G., e Frisco, B. " Blastomicete patogene nell' Uomo, Contributo all' Etiologia dei Tumori maligni," Ann. d' Ig. sper., 1895, 5, p. 433. Centralbl. f. Bakt., 1895, 18, p. 368. Curtis. "Contribution a I'etude de la saccharomycose humaine," Ann. de Vinst. Pasteur, 1896, 10, p. 449. Curtis, C. C. A Textbook of General Botany. 1897. Demme, R. Ueber das Vorkommen eines rothen Sprosspilzes in der Milch und im Kdse, und das Auftreten von Darmkatarrh bei Kindern friihesten Alters dutch Genuss derartig inficirter Milch. Berlin, 1890. Abstr. in Baumgarten's Jahresbericht, 1891, 7, p. 378. Oidiomycosis of the Skin and Its Fungi 169 Dyer, Isadore. "Blastomycetic Dermatitis and Its Relation to Yaws — A Case in Point," Jour. Cutan. and Gen.-Urin. Dis., igoi, 19, p. 14. Faxk, F. "Ueber Hefe-Einspritzung," Arch. f. Anal. u. Phys., 1886. Fermi, Claxjdio, and Aruch, E. "Ueber eine neue pathogene Hefenart und iiber die Natur des sogennanten Crj^jtococcus farciminosus Rivoltae," Centralbl. f. BakL, 189s, 17, p. 593. FouLERTON, A. G. R. Jour. Path, and Bad., 1899, 6, p. 37. DE Gaetano, L. "Di un Blastomiceti patogeno, dotato di rapido Potere setticemico per le Cavie," Rifornia med., 1897. Abstr. in Centralb.f. BakL, 1899, 25, p. 833, and in Baumgarten's Jahreshericht, 1897, 13, p. 751. Galli-Valerio, B. "Sur une variety d'oidium albicans isolee des selles d'un enfant atteint de gastro-enterite chronique," Arch, de parasiloL, 1898, i, p. 572. Gilchrist, T. C. "A Case of Blastomycetic Dermatitis in Man," Johns Hopkins Hospital Reports, 1896, i, p. 269. Gilchrist, T. C, and Stokes, W. R. "The Presence of an Oidium in the Tissues of a Case of Pseudo-Lupus Vulgaris," Bull. Johns Hopkins Hasp., 1896, 7, p. 129. . "A Case of Pseudo-Lupus Vulgaris Caused by a Blastomyces," Jour. Exper. Med., 1898, 3, p. 53. Grasset. "Etude de champignon pyogene parasite de I'homme," Arch. med. exper. et d'anat. path., 1893, $■ . Eludes sur les Muguet. These. Paris, 1894. Cited by Cao. Grawitz. "Ueber die Parasiten des Soors, des Favus und Herpes tonsurans," Virchow's Archiv, 1886, 103, p. 393. Grohe. Berl. Klin. Wchnschr., 1870, 7, p. 8. Gruby. "Recherches anatomique sur une plante cryptogene qui constitue le vrai muguet des enfants," Compt. rend. acad. d. sc, 1842, 14, p. 634, 635; 1844, 17, p. 585. Cited by Robin. . Annales de I'anaiomie et de la physiologic pathogenique, by J. B. Piquei, 1846, under title: "De la structure anatomico-pathologique du muguet"; letter by M. Gruby. Cited by Robin. GuiDi, G. "Ueber Soor, seine Mykologie und Metastasenbildung," Wien. med. Bl., 1895. . Mughelto; Micologia e Melastasi del Mughelto. Abstr. in Baumgarten's Jahreshericht, 1896, 12, p. 652. GiTJLiNi, P. "Soor der Vulva," Centralbl. f. Gyndk., 1891, p. 1049. Abstr. in Baum- garten's Jahreshericht, 1891, 17, p. 375. Hannover. Arch.f. Anal. u. Phys., 1842, 281-95. Harris, F. G. "Blastomycetic Dermatitis of the Gluteal Region," Amer. Jour. Med. Sc, 1901, 121, p. 561. Heller. Tageblatl der 62 Vers. Deutsch. Naturforsch. utid Aerzte. Heidelberg, 1889, 342. . "Beitrage zur Lehre von Soor," Deutsch. Arch. f. Klin. Med., 1895, 55, p. 123. V. Herff, Otto. "Ueber Scheidenmykosen (Colpitis mycotica acuta)," Samml. Klin. Vorlrdge, 1895, No. 137. Abstr. in Baumgarten's Jahreshericht, 1895, II, p. 462. Herzfeld. "Zwei Falle von Mykose durch Soor," Berl. Klin. Wchnschr., 1897, 34, p. 99. Cited by Cao. lyo Contributions to Medical Science HuEPPE. Cited by Anna Stecksen. Hektoen, Ludvig. "The Organism in a Case of Blastomycetic Dermatitis," Jour. Exper. Med., 1899, 4, p. 262. . "A Case of Blastomycetic Dermatitis of the Leg," Jour. Am. Med. Assn., 1899, 33, P- 1383- . "Refractory Subcutaneous Abscesses Caused by Sporothrix Schenckii,' Jour. Exper. Med., 1900, 5, p. 77. Hessler, Robert. "Blastomycetic Dermatitis; Report of a Case," Indiana Med. Jour., 1898, 17, p. 48. Hyde, Hektoen, and Bevan. "A Contribution to the Study of Blastomycetic Dermatitis," Brit. Jour. Dermatol., 1899, 11, p. 261. Hyde, James Nevins, and Ricketts, H. T. "A Report of Two Cases of Blastomy- cosis of the Skin in Man; with a Survey of the Literature of Human Blastomy- cosis," Jour. Culan. and Gen.-Urin. Dis., 1901, 19, p. 44. Jorgensen, Alfred. Micro-organisms and Fermentation. Translated by Miller and Lennholm. 3d ed. 1898. Macmillan & Co. KxEiN. Brit. Med. Jour., 1901, 2, p. i. Klemperer, G. "Ueber die Natur des Soorpilz," Centralbl. f. Klin. Med., 1885. . Ueber den Soorpilz. Inaugural Dissertation. Berlin, 1886. Abstr. in Baumgarten's Jahresbericht, 1886, 2, p. 7,2>3- Langerhans. "Ein Fall von Soor des Oesophagus mit eitriger Entziindimg der Schleimhaut," Virchow's Archiv, 1887, 109, p. 352. . "Mycosis uterina," Berl. Klin. Wchnschr., 1894, 611. Leopold, G. " Untersuchungen zur Aetiologie des Carcinoms und liber die patho- gene Blastomyceten," Arch.f. Gyndk., 1900, 61, p. 77. Link. Cited in Diclionnaire encyclo pedique des sciences medicates, 2 Series, 14, p. 667. Linosseer, G., and Rotjx, G. "Sur la nutrition du champignon du muguet," Comp. rerui. acad. sc, 1890, no, p. 355. . "Recherches morphologiques sur le champignon du muguet," Arch. med. exper. et d'anat. path., 1890, i, p. 62. "Sur la fermentation alcoolique et la transformation de I'alcool en aldehyde provoques par le champignon du muguet," Comp. rend. acad. sc, 1890, no, p. 868. Abstr. in Bau)ngarien's Jahresbericht, 1890, 6, p. 423. Lundsgaard, K. "Ein Fall von H>'popyon keratitis mit Reinkultur von Hefe," Klin. Monatsbl. f. Augenheilk., 1890, 38, p. 13. Maffucci, a., and Sirleo, L. " Osservazioni ed Esperimenti intorno ad \m Blas- tomiceta patogeno," Centralbl. f. allg. Path. u. path. Anat., 1895, 6, p. 305. . "Ueber die Blastomyceten als Infectionserreger bei bosartigen Tumoren," Ztschr.f. Hyg. u. Infektionskr., 1898, 27, p. i. . "Neuer Beitrag zur Pathologie eines Blastomyceten," Centralbl. f. allg. Path. u. path. Anat., 1895, 6, p. 438. . "Osservazioni ed Experiment! sur un Blastomicete patogeno con Inclusione dello Stesso nelle Cellule dei Tessuti patogeni," II Policlinico, 1895. Abstr. in Baumgarten's Jahresbericht, 1895, 9, p. 469. . "Weitere experimentelle Untersuchungen iiber einen Blastomyceten," Centralbl./. allg. Path. u. path. Anat., 1896, 7, p. 977. Memo. Cited by Cao. Oidiomycosis of the Skin and Its Fungi 171 Metschnikoff. Ueber eine Sprosspilzkrankheit der Daphnien," Virchow's Archiv, 1884, 96, p. 177. Montgomery, D. W. "A Disease Caused by a Fungus; the Protozoic Dermatitis of Rixford and Gilchrist," Brit. Jour. Dermatol., 1900, 12, p. 343. Montgomery, F. H., and Ricketts, H. T. "Three Cases of Blastomycetic Infection of the Skin; One Case Limited to a 'Tumor' of the Lower Lip," Jour. Ciitan. and Gen.-Urin. Dis., 1901, 19, p. 26. MouRRiER. Gaz. med. de Nantes, 1895. Abstr. in Baumgarten's Jahreshericht, 1896, 12, p. 652. Nesczadimenko, M. p. "Zur Pathogenese derBlastomyceten," Centralbl. f. Bakt., 1899, 25, p. 55. Nextmayer, F. " Untersuchungen iiber die Wirkungen der verschiedenen Hefearten, welche bei der Bereitung weingeistiger Getranke vorkommen, auf den thierschen und menschlichen Organismus," Arch.f. Hyg., 1891, 12, p. i. Noisette. Recherches sur le champignon du mugtiet. These, Paris, 1898. Oesterlen. Roser und Wundcrlich, Med. Vrtljhrschr., 1842, 470. Cited by Robin. Ophuls, W., and Moffitt, H. C. "A New Pathogenic Mould (Formerly Described as a Protozoon: Coccidioides immitis pyogenes)," PAi7. Med. Jour., 1900, 5, p. 1471. OuDEMANS, C, and Pekelharing, C. A. Arch. Neerland., 20. Abstr. in Baum- garten^s Jahreshericht, 1886, 2, p. 337. Owens, Eisendrath, and Ready. "Blastomycetic Dermatitis (Pseudo-Lupus vulgaris, Saccharomycosis hominis, or Pseudo-Epithelioma with Blastomyces)," Ann. Surg., 1899, 30, p. 545. Parrott. Clinique des nouveau-nes, Paris, 1877, 214. Pelagatti, M. "Ueber Blastomyceten und Hyalin Degeneration," Monatsh. prakt. Dermatol., 1897, 25, p. 157; Virchow's Archiv, 1897, 140, p. 248. Pfeiffer, L. "Ueber Sprosspilze in der Kalberlymph," Cor. Bl. d. allg. drztl. Ver. V. Thiiringen. 1885, 14, p. 86. PlANA AND Galli-Valerio. // modemo Zooiatro, 1894. Cited by Fermi and Aruch. Plaut. Beitrag zur systematischen Stellung des Soorpilzes i. d. Botanik, 1885, Voight. Cited by Baumgarten in Lehrbuch der Path. Mykologie, 1890. . Neue Beitrage zur systematischen Stellung des Soorpilzes i. d. Botanik. Leip- zig, 1887. Popoff, L. "Untersuchungen iiber die Wirkung der Bierhefe, etc.," Berl. Klin. Wchnschr., 1872, 9, p. 513. Rabinowitsch, Lydia. "Untersuchungen iiber pathogene Hefearten," Ztschr. Hyg. u. Infektionskr., 1895, 21. Raum, Joh. "Zur Morphologie und Biologie der Sprosspilze," Ztschr. Hyg. u. Infektionskr., 1891, 10, p. i. Reubold. Virchow's Archiv, 1854, 7, p. 76. Ribbert. Sitzungsher. d. niederrhein. Gesellschaft in Bonn, 1879, 86. . "Weitere Untersuchungen uber das Schicksal pathogener Pilze im Organis- mus," Deutsch. med. Wchnschr., 1885. Rivolta. Parassiti vegetali, 1873, PP- 246, 525- . Gior. di anat. e fisiol. di anim., 1880. Cited by Fermi and Aruch. Rixford, Emmet, and Gilchrist, T. C. "Two Cases of Protozoon (Coccidioidal) Infection of the Skin and Other Organs," Johns Hopkins Hospital Reports, 1896, I, p. 209. 172 Contributions to Medical Science Rdcford and Thorne. "A Case of Protozoic Skin Disease," Occidental Med. Times, 1894, 8. Robin, Ch. Des vigetaux qui croissent sur I'homme et sur les animaux. Paris, 1847. . Histoire naturelle des vigetaux parasites qui croissent sur I'homme et sur les animaux vivants. Paris, 1853. Roger, H. "Serologie de I'Oidium albicans," Bull, med., 1896, 12, p. 652. Abstr, in Baumgarten's Jahresbericht, 1896, 12, p. 652. Ross. "Vorlaufige Mittheilung iiber einige Falle von Mykose im Menschen," Centralbl. f. Bakt., 1891, 9, p. 504. RoussY. "Recherches exp6rimentales sur la pathogenic de la fievre," Arch, de physiol., 1890, 23, p. 354. Cited by Raum. Sanfelice, Francesco. "Ueber einen neuen pathogenen Blastomyceten, welcher innerhalb der Gewebe unter Bildung kalkartig aussenhender Massen degenerirt," Centralbl. f. Bakt., 1895, 18, p. 521. . "Ueber die pathogene Wirkung der Blastomyceten," IV. Abhand. Bei- trage zur Aetiologie der sogen. Pocken der Tauben (Gefliigelpocken), Ztschr. f. Hyg. u. Injektionskr., 1897, 26, 298. . "Ueber die pathogene Wirkung der Blastomj'ceten," V. Abhand. Ein Beitrag zur Aetiologie der bosartigen Geschwiilste, Ztschr. /. Eyg. u. Injek- tionskr., 1898, 29, p. 463. . "Ueber eine fiir Thiere pathogen Sprosspilzart und iiber die morphologische Uebereinstimmung, welche sie bei ihrem Vorkommen in den Geweben mit den vermeinthchen Krebscoccidien," Centralbl. f. Bakt., 1895, 17, p. 113. . "Ueber die pathogene Wirkung der Sprosspilze. Zugleich ein Beitrag zur Aetiologie der bosartigen Geschwiilste," ibid., p. 625. . "Suir Azione patogena dei Blastomiceti come Contributo alia Eziologia dei Tumori maligni," // Policlinico, 1895. . "SuU' Azione patogena dei Blastomiceti," Ztschr. f. Hyg. u. Injektionskr., 21, 399- . "Ueber die experimentelle Erzeugung der Russel'schen Fuchsin Korper- chen," Centralbl. f. Bakt., 1897, 23, p. 276. Schenck. "Refractory Subcutaneous Abscesses Caused by a Fungus Possibly Related to the Sporotricha," Bull. Johns Hopkins Hosp., 1898, 9, p. 286. Schmidt, Martin B. "Ueber die Localization des Soorpilzes in den Luftwegen und sein Eindringen in das Bindegewebe der Oesophagus-Schleimhaut," Zeigler's Beitrdge, 1890, 8, p. 173. ScHMORL. "Ein Fall von Soormetastase in der Niere," Centralbl. f. Bakt., 1890, 7, P- 329- Secchi, Th. "Das Vorkommen von Blastomyceten bei der Keloid-Akne," Monatsh. /. prakt. Dermatol., 1896, 23, p. 509. Abstr. in Baumgarten's Jahresbericht, 1897, 14, P- 748. . "Ueber die pathogene Wirkung der Blastomyceten und ihre Bedeutung in der Aetiologie der Neubildungen und anderer Krankheiten. Kritische Ueber- sicht der neuesten Italienischen Arbeiten," Monatsh. f. prakt. Dermatol., 1897, 24, P- 554- DE SmoNi, A. " Ueber das Vorkommen von Blastomyceten in der hypertrophischen Tonsille," Centralbl. f. Bakt., 1897, 22, p. 120. Slawjansky. Abstr. in Virchow-Hirsch's Jahresbericht, 1867, i, p. 307. Oidiomycosis of the Skin and Its Fungi 173 Stecks£n, Anna. Studier ofver Curtis' Blastomycet fran Svulst-etiologisk Synpunkt. 1900. Stockholm. Steiner, Max. "Zur Pathogenese des Soorpilzes," Ceniralbl. f. Baki., 1897, 22, p. 385. Stelwagon, Henry W. "Report of a Case of Blastomycetic Dermatitis," Ant. Jour. Med. Sc, 1901, 121, p. 176. DE Stoecklin, H. "Recherches clinique et experimentales sur le r61e des levure trouvees dans les angines suspectes de Diphtheric," Arch. med. exper. et d'anat. path., 1898, II, p. I. Stoewer. "Ueber die Wirkung pathogener Hefen am Kaninchenauge," Arch. f. Ophth., 1899, 48, p. 178. Tartakowsky, M. G. Der afrikanische Rotz der Pferde. St. Petersburg, 1897. Teissier, J. "Sur un cas d'angine pseudomembraneuse chez un syphilitique avec presence exclusive dans I'exsudat des formes de levure du muguet," Arch. mid. exper. e. d'anat. path., 1895, 7> 265. Abstr. in Baumgarten's Jakresbericht, 1895, 11. TOKISHIGE, H. "Ueber pathogene Blastomyceten," Centralbl. f. Bakt., 1896, 19, p. 105. Troisier, E., and AcHALME, P. " Sur une angine parasitaire causee par une levure et cliniquement semblable au Muguet," Arch, de med. expir. et d'anat, path., 1893, 5, p. 29. Abstr. in Baumgarten's Jahresbericht, 1893, 13, p. 461. Underwood, L. M. Moulds, Mildews, and Mushrooms. 1899. ViRCHOW. Handbuch d. spec. Pathol, u. Ther., 1854, i, p. 358. Vogel. Anatomie pathologique generale, p. 1847. . Allg. Zeit.f. Chir. inn. Heilk. u. i. Hiilfswissensch., 1842. . Gaz. med. de Paris, 1842, 234. . Traite d'anatomie pathol. generale. Paris, 1847. Cited by Robin. . Cited in v. Ziemssen's Handb. d. Spec. Pathol, und Ther., 2 Aufl., 1878 8, p. 64. Wagner, E. Jahrb.f. Kinderheilk., 1868. Cited by Cao. Wells, H. G. "Case of Blastomycetic Dermatitis," New York Med. Jour., 18 67, p. 427. Wernicke, R. "Les protozoaires pathogfenes" (translated from Ann. d. circ. mid Argent., 1890, 13, 391), /. demicrog., 1891, 15. . "Ueber einen Protozoenbefund bei Mycosis fungoides (?)," Centralbl. /, Bakt., 1892, 12, p. 859. . "Nueva Contribucion a la Parasitologia Argentina," Ann. d. circ. mid Argent., 1892, 15. Zenker, F. A. Jahresb. der Gesellsch. f. Natur u. Heilkunde in Dresden, 1860-61, 51 EXPLANATION OF PLATES. Plate i, Fig. la. — Photograph of lesion in Case VII. Fig. 16. — Photograph of lesion in Case IX. Plate 2, Fig. ic. — Case V. A budding " blastomyces " and a mature single form in an intra-epithelial abscess. Hematoxylin and eosin. Photomicrograph X 1 200. 174 Contributions to Medical Science Fig. 2. — Case III. A giant cell of probable epithelial origin, containing a pair of degenerating organisms. Hematoxylin and eosin. Photomicrograph X 1200. Plate 3, Fig. 3. — Case VII. A pair of organisms in an incipient intra-epithe- lial abscess; also a single mature form between epithelial cells. Hematoxylin and eosin. Photomicrograph X1200. Fig. 4. — Case HI. Three organisms in an atypical chain, lying in an abscess of corium. Polychrome methylene blue. Photomicrograph X1200. Plate 4, Fig. 5. — Case III. Hanging-drop culture. Tortuous, segmented hyphae; an oidial form. Photomicrograph g in. objective, ocular No. 4. Bausch and Lomb. Fig. 6. — Margin of colony in Fig. 5. X1200. Fig. 7. — Case VII. Organism in plate culture. The feathery nature of colonies is shown well. Photograph slightly reduced. Plate 5, Fig. 8. — Photomicrograph of colony shown in Fig. 7. Branching hyphae giving off imicellular shoots. Plate 6, Fig. 9. — Case HI. Single plate colony of organism. Photomicro- graph X60. Fig. 10. — Case HI. Higher magnification of colony. Hyphae are segmented, and older cells contain refractive spore-like bodies. Photomicrograph X120. Plate 7, Fig. ii. — Case \TI. Submerged hypha, segmented, and sprouting spherical and elongated conidia; spherical cells, 6-8 microns in diameter; elongated cells, about 4 microns thick. Each segment contains from two to four small refractive bodies. Fig. 12. — Case VII. Forms seen in a nodule in the lung of a dog which died from an intravenous inoculation. Drawn from a fresh mount in water. A, B, and D measured 8 10, and 12 microns respectively; C and E, 15 microns each, enlarged. The refractive bodies in A, B, and D are similar to those in C. F, spherical spore- like bodies l3dng free among and sometimes included in tissue cells; from i to 3 or 4 microns in diameter. E, a budding organism with an adventitious capsule. Fig. 13. — Case I. Drawing from a drop mount of a bouillon culture. Hyphae with terminal and lateral conidium groups. Cells contain the nucleus-like structures spoken of. X 1200 (about). Fig. 14. — Case I. Sprouting of hyphae in blood-serum of dog, showing nucleus- like bodies. X 1 200. Plate 8, Fig. 15. — Case I. Forms seen in the pus produced by a subcuta- neous inoculation in guinea-pig. Fig. 16. — Case I. Hanging-drop culture in which the oidium morphology is maintained somewhat imperfectly. Photomicrograph X 1 200. Plate 9, Fig. 17. — Case VII. Terminal spore-groups. Liquid culture. 10-12 microns. Fig. i8a. — Case HI. Plasmolysis, pseudo-spore formation or perhaps true spore formation, occurring in segmented hyphae fragments. Width about 8 microns; enlarged. Fig. 18&. — Case HI. The morphology suggests endospores, but the bodies appear structureless. The cell measured 30 microns in diameter. PLATE 1, Fig. la. Fig. lb. PLATE 2. Fk;. Ir, PLATE 3. Fig. 3. Fig. 4. PLATE 4. p Fig. o. Fig. (j. Fig. 7. PLATE 5. ■■■j ■ ^Hp^Si^l^^^^^l ■■ ^^ >. • B^ ;:^r ■^■^,.. ... ^^it 1 t Fig. «. PLATE 6. v^Q Fig. 9. Ik;. 1(1. PLATE 7. Ricketts, Del. Fig. 12. \ ^ ^'. - Fig. 13. ^W) CC ^ ^'^ o ©• ov \ cs» •^>'* >r— • •^» \ • • V Fig • • 4 " w Fig. 6. Fig. 7. • ^ J«^^ f Fig. 8. PLATE 12. lOjV ^P Fig. 9. Fig. 10. Fui. 11. Oidiomycosis of the Skin and Its Fungi 175 Plate 10,' Fig. i. — A typical low-power view of a skin lesion, showing the carcinomatoid epithelial hyperplasia and intra-epithelial abscesses. From Case III. Methylene blue. Bausch and Lomb \ in. objective, No. 2 eyepiece. Fig. 2. — An intra-epithelial abscess, containing a group of organisms. The green color reaction of the capsules is perhaps somewhat exceptional. From Case IV. Polychrome methylene blue. Bausch and Lomb j'^ in. oil objective, eyepiece No. 2 Fig. 3. — Illustrating the "cobble-stone appearance" of the rete mucosum, fre- quently seen. The chromophilic substance of the cell body is concentrated in a peripheral segment of the cells in considerable areas of tissue, evidently an example of chromatolysis. From a laboratory case. Polychrome methylene blue. Bausch and Lomb i^a in. oil objective, eyepiece No. 2. Fig. 4. — Mycotic focus in kidney of a mouse which died from subcutaneous injection of the organism from Case I. The organisms within tubules and in inter- stitial tissue. Small and streptothrix-like forms. Granular degeneration of tubular cells. Polychrome methylene blue. Same magnification as Fig. 3. Plate ii,' Fig. 5. — Showing a change that occurs in the prickles of the mucosal cells in the vicinity of an abscess. The processes apparently are transformed into a semi-comified capsule for the cell. Figs. 8 and 6 show, the former a very early, the latter a later, stage, in the process of epithelial multinuclear giant cell formation. Fig. 7. — A small intra-epithelial abscess, surrounded by a partially cornified capsule. A pair of organisms and two eosinophile leukocytes are seen in the abscesses. Figs. 5-8. — Hematoxylin and eosin. Bausch and Lomb iV in. oil objective, eyepiece No. 2. Plate 12,' Fig. 9. — Cells seen in the lung of a dog which died from an intra- venous inoculation of the organism cultivated from Case VIII, stained with hema- toxylin, a) A large organism, the protoplasm of which contains many deeply staining spherical bodies (spores?), b) An alveolar cell containing a single organism, c) A mass of organisms surrounded by a finely granular substance. They lay within an air sac, dilating it. d) A multinuclear giant cell (of epithelial origin ?) containing many organisms. Bausch and Lomb objective iV in., eyepiece No. 2. Fig. id. — The relation of plasma cells to new forming fibrous tissue, as shown by a carbol-fuchsin and water-blue stain. From Case I. This figure and the one fol- lowing much enlarged from observations with iV in. oil immersion objective. Fig. II. — The occurrence of hyalin spherules in plasma cells. Case I. Carbol- fuchsin and water-blue. ■ Colored plates in the original. AN ORGANISM FROM CUTANEOUS OIDIOMYCOSIS (BLASTOMYCOSIS).^ REMARKS ON CLASSIFICATION. H. T. RiCKETTS. {From the Pathological Laboratory of the University of Chicago.) I desire to record a hitherto unreported case of blastomycosis, or oidiomycosis of the skin, from which the fungus was obtained in pure culture. I saw the patient at the request of Dr. Porter of Champaign, 111., in his office in the fall of 1902. It was the plan to publish a full clinical and pathologic report of the case at the time, but the patient refused to be photographed or allow the removal of tissue for study. A report on the organism has been delayed with the expectation that other material might come to hand which could be incorporated. The disease was of several years' duration, had its site on the left of the face, had involved the lower eyelid, and presented the verrucous surface and miliary abscesses characteristic of the pro- cess. Fresh preparations of the contents of miliary abscesses revealed numerous spherical, capsulated, budding cells. It was possible to obtain a few plain agar culture tubes from the laboratory of the state university; these were inoculated with the pus of the small abscesses, which are almost invariably found in or beneath the unbroken epidermis which surrounds the ulcerated or verrucous area. No incubator was at hand, but the summer temperature was favorable to the development of micro-organisms. At the end of twenty-four hours a few minute colonies had developed on one tube; these proved later to be staphylococci. On the third day small grayish points, which were less conspicuous than the colonies first noted, were seen on all tubes. These developed slowly, and in a few days formed opaque circular areas on the surface of the medium, the center of each colony became marked by a blunt, elevated point, and a little later the growth extended into the substratum. In about ten days it showed a downy surface, " From Trans. Chic. Path. Soc, igo4, 6. p. 113. 176 An Organism from Cutaneous Oidiomycosis 177 a feature which became more pronounced during subsequent days, and eventually resulted in the marked mouldy appearance which is more or less characteristic for this type of the organism. The growth in the substratum of the medium is made up of hyphae, which are divided into segments of varying lengths, and which give oflf abundant branches. It is not unusual to find the segments of a single thread separated; this is interpreted as pro- ' liferation by segmentation. The aerial threads are finer, more compact in appearance, and are not so prominently segmented as those of the underlying growth. They have many long fine branches, and in addition shorter ones, the terminal extremities of which possess either a single small spherical cell, or a group of such cells. These are the conidia mentioned in previous descrip- tions. In bouillon, growth occurs in the form of coherent, delicate tufts, which may reach large proportions in the course of two or three weeks. The organism was not pathogenic for guinea-pigs in subcuta- neous inoculations. In the particulars enumerated and in all others, this culture appears to be identical with that of an organism which I previously cultivated from five cases of cutaneous blastomycosis, and first described in a paper entitled **A New Mould-Fungus as the Cause of Three Cases of So-called Blastomycosis or Oidiomycosis of the Skin."^ A consideration of the whole subject, including a descrip- tion of the organisms cultivated by various investigators, was given later in the Journal of Medical Research (1901, 6, p. 374)- The characteristics of the organisms described in literature and of those which came under my own observation were such that, in the latter paper, they were divided into three groups. This division was based on both the cultural and the microscopic appearances of the three types. In the first group are those organ- isms which, on agar slants, form a smooth pasty growth, without aerial hyphae, and which may or may not penetrate the sub- stratum. Proliferation is chiefly by a budding process, but often myceHal formation and myceHal segmentation are conspicuous. The second group is intended to include those organisms which ' Jour. Boston Sac. Med. Sc, 1909, s, p. 453. 178 Contributions to Medical Science grow with a granular surface, which later may present rugae and folds, but do not develop true aerial hyphae; minute short spikes may project slightly above the surface. The depth of the medium is invaded. Proliferation in this type is chiefly by the formation of a segmented mycelium, the elements of which eventually separate after the manner of oidium. The third group consists of the mould-fungi which have been cultivated from the disease. These prohferate by the methods enumerated for the first two groups, and in addition conidia are formed on the aerial hyphae. The appearances on other media, especially bouillon, also show features which are peculiar to each group. In spite of cultural, morphologic, and certain biologic differ- ences which are noted among the three types, other features which they possess in common, and the fact that the three are found in the same disease, indicate that they are closely related organisms. For this reason they were all considered to occupy the same generic position, which, as nearly as could be determined, was that of oidium. The fact that all the organisms may prohferate by a budding process, and in tissues do so constantly, is no argument against referring them to the genus oidium, inasmuch as the oidia and even other h^^^homycetes are known to lapse into a budding stage. Holding these views, it has been interesting to note certain cultural changes which indicate the close relationship of the three types. I refer especially to the fact that an organism belonging to the first group, one which habitually grows with a smooth pasty surface, has been seen to grow with the uneven and rugated surface characteristic of group 2 ; and that every one of the mould- fungi which I have cultivated has at one time or another grown without the moulded surface, also resulting in an appearance which approaches that of group 2. Nevertheless, certain constant differences and the tendency to return to the original form of growth preclude the possibility of considering them identical. It is of interest that nearly all the cultures which have been reported in the past three years are of the mould t>TDe. Hence it seems probable that the mould-fungus will prove to be the pre- dominating organism in oidiomycosis. THE IMMUNOLOGICAL REACTIONS OF OIDIOMYCOSIS (BLASTOMYCOSIS) IN THE GUINEA-PIG.' Benjamin Franklin Davis. {From the Department of Pathology and Bacteriology, the University of Chicago, Chicago, III.) The object of this work was to examine the mode of resistance of guinea-pigs to oidiomycetic infection; to determine whether it might be possible to increase this resistance by immunization, and, if so, to investigate the factors upon which this increase in resisting power depended. LITERATURE. For a full discussion of the relation of oidioid and blastomy- cetoid organisms to human and animal diseases and for literature as to their relative pathogenicity for various animals, the reader is referred to the work of Ricketts," Brown,^^ Hektoen,^^ Mont- gomery and Ormsby,^^ and Spiethoff.'' The papers considered here are those which have dealt somewhat specifically with the question of the mode of resistance of animals to infection with mould-fungi, and with the demonstration of antibodies in the infected animals. ^Metchnikoflf' (1884) demonstrated that phagocytosis is the chief means of defense of the daphnia against infection with a mould-fungus christened by him "Monospora bicuspidata." This was followed by a number of articles by Ribbert^ (1887) who was inclined to believe that mould-fungi (Schimmelpilze) and pathogenic schizomycetes injected into rabbits are ingested by leukocytes within which they undergo intra- cellular digestion. Charrin and Ostrowsky* (1896) observed that immunized animals were but moderately resistant to reinfection with Oidium albicans. Roger' (1896) obtained considerable resistance to Oidium albicans by vaccination. Rabbits given repeated intravenous injections of sublethal doses became able to resist infection by double the lethal dose. In normal serum the organisms grew quite readily; in immune serum they were first agglutinated, then became hyaline, the capsule disin- tegrated, and later attempts to obtain cultures showed that the oldia were dead. Schattenfroh' (1896), using a non-pathogenic yeast, found that the sera of animals were not bactericidal but that peritoneal exudates were markedly so. He concluded that the germicidal power of peritoneal exudates depends on phagocytosis. Gilkinet^ (1897) believed from the results of his experiments, (i) that beer yeasts (Saccharomyces cerevisiae) introduced into rabbits either intravenously or subcutaneously produce » From Jour. Inject. Dis., 1911, 8, p. 190. 179 i8o Contributions to Medical Science neither local nor general symptoms; (2) that such yeasts do not multiply in living tissues; (3) that they are destroyed within a very short time, and cannot be discovered in the bodies of the animals by any method; (4) that this destruction is brought about by the substances in the plasma and is not a function of the body temperature, of chemical reaction, or of the absence of nutritive substances, but is a specific, unknown property of the organic fluids; (5) that this destructive property is not dependent upon the formed elements of the blood, but is exercised in the same degree by all the body liquids; and, finally, (6) that this property is itself destroyed by heating at 55° C. He found that yeast mixed with ox serum or rabbit serum and kept at 36° C. for from two to four days lost its power of growth on suitable media, but that if such mixtures were kept at room temperature for corresponding periods, the yeasts were unharmed. He placed porous tubes containing blastomycetes in the peritoneal cavities of three rabbits. The first rabbit was examined after four days; the yeast cells were not much changed, and were not in contact with leukocytes; in the second rabbit, killed after nine days, and in the third rabbit, killed after twelve days, there were but few normal yeast forms and apparently all were dead, since cultures were negative. Leukocytes were not in contact with the organism. The author produced edema of the rabbit's leg by tight bandaging; such edematous fluid, free from leukocytes, killed yeasts in three days at 37° C. Jona* (1897) gave rabbits intraperitoneal, intravenous, and subcutaneous injections of a non-pathogenic yeast, Saccharomyces apiculatus. He concluded that in all cases the organisms were destroyed within a few hours through the influence of the body fluids. Organisms injected into the peritoneal cavity, or subcutaneously, did not make their way into the blood. Obici' (1898) found that after repeated injections of the toxin (filtered broth culture) and small quantities of the spores, an immunity was established in rabbits, against Aspergillus fumigatus. This immunity, however, was incomplete, and vanished in a short time. He did not believe, with Ribbert, that phagocytosis was of such paramount importance in the defense of the animal body against this infection, as he saw evidences of degeneration in organisms in the absence of phagocytic cells. Skchiwan'" (1899) experimented on guinea-pigs and rabbits with Saccharomyces tumefaciens of Curtis. He con- cluded that the body fluids are not bactericidal since the organisms protected from the leukocytes by celloidin sacs grew well on artificial media after being in the peri- toneal cavity of the animal for four days. On the other hand, yeasts injected directly into the peritoneal cavity were phagocyted by certain cells and digested. The pha- gocyted yeasts lost their staining power in from two to four days, as well as their power to grow when planted on suitable media. The leukocytes formed rosette-like masses about the organisms. At first polymorphonuclear leukocytes were found in such masses; later, the groups consisted entirely of macrophages. Malvoz'^ (1901) experimented with six strains of yeast, some pathogenic and some non-pathogenic. The pathogenic cultures he obtained from Sanfelice, Plimmer, Curtis, and one he isolated himself from an epithelioma, calling it Blastomyces E (BE) as a convenient laboratory name. His non-pathogenic cultures consisted of a strain of Saccharomyces ellipsoideus and a powerfully fermentative organism which he called the yeast of Huy Rabbits were given two weekly injections of the organism of Huy, BE, and Sanfelice over a period of months, and the agglutinative properties of the sera were then tested. The specific serum agglutinated the organism of Huy in a dilution of 1:50; organism BE, 1:90; organism Sanfelice, 1:5. The specific serum for the organism of Huy agglutinated S. ellipsoideus in the same titre as the organism of Huy itself; more Immunological Reactions of Oidiomycosis i8i weakly the organism of Plimmer and organism BE; and the organism of Curtis not at all. The specific serum for organism BE agglutinated organism of Huy very slightly. The organisms of Huy, Sanfelice, and organism BE grew well in normal and immune serum, and grew on artificial media when transplanted after i to 24 hours. The author concluded that in this case either bactericidal antibodies were absent or the organisms were protected by their capsule. He suggests that the Bordet-Gengou reaction might be of service in deciding the presence or absence of such antibodies. Ricketts" (1901) says: "Professor Hektoen has found that the undiluted serum of a dog which had received successive inoculations of the organism from Case I (cutaneous oidiomycosis) causes gradual clumping of the organism diffused in broth. Several hours elapse before the fullest extent of clumping possible is reached. Organ- isms from other cases show only a slight degree of clumping with the same serum. All organisms grew well in the serum and the production of mycelium was especially noticeable. Normal dog's serum caused no clumping of any of the organisms; abun- dant mycelium is produced in all cases. "The organism from the case reported by Hyde, Hektoen, and Bevan was repeat- edly inoculated into the abdominal cavity of a rabbit. It was found that the animal's serum would cause fairly distinct agglutination of the organism inoculated." Sanfelice^ (1896) discovered that heating for 30" at 60° C. destroyed the viru- lence of his Saccharomyces neoformans for guinea-pigs, but did not destroy its power to grow on artificial media. He endeavored to immunize guinea-pigs against this organism (i) by the injection of organisms whose virulence had been destroyed by heating to 60° C. for 30"; (2) by repeated injections of filtered broth cultures of various ages; and (3) by the injection of the serum of guinea-pigs and of dogs which had received repeated injections of either the heated organism or the filtered culture, or had recovered from the infection induced by a dose of virulent organisms. He was wholly unsuccessful. In 1902 Sanfelice's was able to immunize dogs, cats, and rabbits to Saccharomyces neoformans, Plimmer's yeast, and to a non-pathogenic yeast isolated from the air, so that the animals could withstand, without symptoms, an intravenous injection of quantities of the first two organisms which caused the death of normal, control ani- mals. He had no difliculty in demonstrating the presence of a specific amboceptor in the serum of such immune animals by the fixation test. He was unable to demon- strate the antibody in actively infected animals which subsequently died of the dis- ease. Unfortunately he never took the trouble to control his fixation experiments by testing the effects of his antigen — emulsion of yeast — or of his antibodies — inactivated immune and normal sera — upon the hemolytic system which he used — serum of rab- bits immune to fowl corpuscles — a fact which detracts considerably from the value of his results. He says that Malvoz" demonstrated antiblastomycetic amboceptors by the fixation reaction in 1901. Sanfelice believed that the immune sera caused a change in the yeast cells exposed to their action such that the organisms assumed an appearance identical with the so-called Russell's fuchsin bodies of malignant tumors. Wlaeff"^ (1902) immunized geese and donkeys to a yeastlike organism which he isolated from the ascitic fluid of a patient whose disease had been diagnosed as inop- erable abdominal cancer with ascites. A piece of the growth, removed at an explora- tory operation, was diagnosed by Cornil as a typical cyhndrical-celled carcinoma. The sera of the immunized animals agglutinated and dissolved the specific organisms. Wlaeff claimed that, following repeated injections of such sera, the ascitic fluid of this 1 82 Contributions to Medical Science cancer patient likewise developed the property of agglutinating and dissolving the blastomycetes and that the cancer improved as the result of such treatment. Fabozzi'* (1905) concluded from his experiments that Saccharomyces neoformans never causes the development of tvmiors, and is finally destroyed by phagocytes. Christensen and Hektoen" (1906) say: "The character of the lesions of blas- tomycosis, the accumulation of leukocytes, the formation of giant cells, and the pha- gocytosis of blastomycetes — indicates that this is an infection in which phagocytosis is an important means of defense and healing. Certain preliminary test-tube experi- ments showed that phagocytosis of blastomycetes is favored by the presence of normal serum, and the idea arose that it might be possible to stimulate the greater formation, in cases of blastomycosis, of the body that promotes phagocytosis (opsonin), as well as the other antibodies, by the injection of blastomycetic substances in a readily absorbable form. It was thought that the resistant character of the micro-organisms, coupled with their inclosure in cellular exudate and granulation tissue, possibly pre- vents the absorption in proper quantities of the substances necessary to call forth strong immunizing reactions. Hence, in order to hasten, if possible, the reactions that favor healing, we injected in each of our cases a sterile blastomycetic vaccine prepared by Dr. H. T. Ricketts of the corresponding organism. Unfortunately the patients left the hospital at a time when no conclusions of value could be drawn as to the results of the vaccine." Marco del Pont^^ (1907), using Endomyces albicans, grew broth cultures with increasing doses, first, of normal, and later, of immunized rabbit's serum. The devel- opment of a mycelium indicated that the medium was unfavorable. After the 60th transfer the yeast grew normally in undiluted immune serum. It retained this faculty. The immunity was specific; yeast habituated to rabbit serum did not grow in the serum of the dog, goat, or rat. The immunized fungus had lost its virulence. The author thought that the immunity to the serum might be due to the development of a muci- laginous capsule by the organism; he also thought that a substance anti-sensibilisatrice might be produced by the micro-organism. Widal, Abrami, Joltrain, Brissaud, and WeilP' (1910) have found that the blood serum from cases of sporotrichosis possesses the power of agglutinating the spores — but not the cell bodies — of the Sporotrichum schenckii in dilutions of from i : 300 to 1 : 800. The agglutinability of the spores varies with the age of the culture and with the kind of medium, but not with the strain of the organism (10 strains were com- pared). The Bordet-Gengou reaction was constant. Among 168 normal subjects and subjects suffering from other diseases, they found an occasional agglutination or fixation reaction, but both never occurred in the same individual. The two reactions, therefore, control each other. Experimentally, two dogs and four rabbits were used. In dogs it was easy to produce infection with the organisms, and the agglutination and fixation reactions were constant and marked. In rabbits, intravenous injections of large quantities failed to cause infection. The agglutinative power of the serum increased appreciably; it varied from 1:10 to 1:30 in the different rabbits before injection, and rose, following injection, as high as i : 3,000 in one case and to but i :300 in another. The reaction of fixation, which was tested parallel with the agglutina- tion tests, gave "very incongruous results." No further details of these results are stated. They found that the agglutination and fixation reactions disappeared in man after the infection was conquered. The fixation reaction disappeared more rapidly Immunological Reactions of Oidiomycosis 183 than the other. They concluded that, so long as both are found, an active lesion is indicated; the disease is not really cured. Co-reactions. — The organisms tested which did not give co-agglutination and co-fixation reactions with immune sporothrix serum and vice versa were the tricho- phytons, the organisms of Erythrasma, Favus, and of animal aspergillosis. The re- actions of co-fixation and co-agglutination were obtained with organisms isolated from cases of actinomycosis and thrush (Oidium albicans). The serum from the case of infection with Oidium albicans agglutinated the spores of the sporothrix 3 times as strongly as it did the oidium itself, which was agglutinated only in dilutions of 1:10 to 1:50. The co-fixation reaction was positive with the following cultures of yeast: the organism of Curtis, of Blanchard, of Plimmer, O. luteum, S. granulatus, S. litho- genes, and S. caprae, using sera from cases of sporotrichosis, actinomycosis, and thrush. The organism of actinomycosis itself was not agglutinated by the specific serum. Rothe^* (1909), noting the preliminary reports of Widal,^' tested the agglutinating power of the blood serum from two cases of actinomycosis upon the spores of the sporo- thrix and found them positive in dilutions of i : 160 and i : 200 respectively. MATERIALS AND TECHNIC, The organism used was obtained in 1905 by Dr. Rosenow from a case of oidiomycosis in the Presbyterian Hospital of Chicago. It had, therefore, been growing on artificial media for three years before the present work was begun. The stock culture was grown on a I per cent glucose, i per cent acid agar, and also in nutrient broth of similar sugar and acid content. On acid-glucose-agar, at room temperature, the organism grows in the form of long hyphae which burrow peripheralward through the superficial layers of the agar. No aerial hyphae are developed excepting in those cultures which have been started in the incubator and are later placed at room temperature. The characteristics of these hyphae will be discussed later. In acid-glucose-broth, at room temperature, the organism grows in the form of fluffy balls, the size which the latter may reach apparently being limited only by the volume of the broth, the capacity of the retaining vessels, and the number of colonies which develop. There seems to be no production of gas. In old cul- tures a heavy membrane forms on the surface of the broth, and the medium, though remaining clear, assumes a dark amber color. Under the low power of the microscope in a hanging-drop prepara- tion kept at room temperature and first observed October 20, 1908, one of these "fluffy balls" appeared as a mass of mycelial threads arranged in two zones, a central mass of tangled threads and granu- 184 Contributions to Medical Science lar material too thick for light to pierce and so not discernible in detail, and a peripheral zone of radiating threads. These threads presented a rather homogenous appearance, were quite refractive to light, and possessed a slightly greenish color by transmitted light. Their central ends were lost in the central mass; their peripheral ends were bluntly rounded. Branches were few and these few were found toward the base of the radiating threads; the branches were equal in diameter to the parent stem, were of uniform thickness throughout their length, for the most part, and were given off, sometimes at right angles, but usually at an angle of about 45°. The contour of the angle was not that of a sharp corner, but rounded. No segmentation of the h>"phae nor lateral conidia were observed. No distinct cell wall appeared. On October 22, two days later, the following notes were made: Hyphae have at least doubled in length; three zones can now be distin- guished in the colony: (i) a central, practically opaque zone; (2) a middle zone of branched, interlacing hyphae bearing lateral conidia; most of these hyphae show marked but somewhat irregu- lar segmentation; and (3) a peripheral zone similar to the one described on October 20. The lateral conidia arise by short stalks from alternate sides of the hyphae, one conidium to each segment. In some cases no structures are discernible within the conidia; at other times they seem to be filled with sharply defined spherules. The segments consist of oblong, clear areas separated from each other by discs, cubes, and cylinders of a homogenous, highly refract- ive, greenish substance very similar in appearance to the material in the young hyphae described two days before. The peripheral ends of the hyphae and those intermediate portions in which the segmentation is but slightly marked contain scattered small, spherical, highly refractive bodies which seem to be attached to the inner side of the double-contoured cell wall which has now become visible. The hyphae are reaching far out into the vase- line by which the cover glass is attached. November 29. — No change in method of development. In agar hanging-drop preparations kept at room temperature es- sentially the same appearances are obtained as in the broth. Some- times the lateral conidia appear to be filled with sharply out- Immunological Reactions of Oidiomycosis 185 lined, homogenous spherules. If such conidia are kept under observation for a few weeks, the spherules can be seen to gradually fuse into larger and larger masses until finally the conidial content consists of a single homogenous mass. A somewhat similar con- dition has been described by Bowen and Wolbach.^" In the case described by these authors, bodies were found on agar tubes, which were filled with refractive spherules. When these bodies were placed on fresh media the spherules fused before mycelial formation began. On acid-glucose-agar at 37° C, the organism grew exclusively in the budding form, that is, like yeast, for about fourteen months, when it rather suddenly — in the course of a month or six weeks — developed a propensity for growing in the hyphal form to which it has clung tenaciously since. While growing in the budding form the colonies which formed on the agar were soft, yellowish-white, raised, and button-shaped, consisting of a flat central area sur- rounded by a thickened, rounded border. They were circular in outhne and from 1-7 mm. in diameter. The individual cells were from 10-30 At in diameter. With the assumption of the hyphal form of growth, the organ- isms became prone to the development of aerial hyphae. This tendency was especially noticeable in cultures kept at about 34.5° C. ; 36-37° C. and room temperature (about 19-20° C.) appeared to be very much less favorable to their production, excepting in cultures, as mentioned above, which had been transferred from the incubator to room temperature. To the naked eye, the hyphae — developed at 34.5° C. — appear as very delicate stalks perhaps a milHmeter in height (estimated), white by transmitted light, and having a tendency to assume a silvery sheen by reflected light. In the mass, they present a dead white, somewhat fuzzy appear- ance, resembling a piece of high grade, heavy filter-paper. About the edges of the growth, where the hyphae are less numerous, the cultures have a ''woolly" appearance. In this condition the culture is less adherent to the medium than at earlier stages — the hyphae begin to appear about 10 days to two weeks after inoculation — and may be peeled off in a single layer taking with it pieces of agar. At a few scattered points hyphae may be seen which are three or 1 86 Contributions to Medical Science four times the ordinary length and project above the general level at odd angles, like the poles from the top of an Indian wigwam — probably the "porcupine" appearance mentioned by Hamburger." Under the microscope the h}phae appear as slender, hollow rods, I to 2 M in diameter, with a very delicate wall. They contain a very fine granular substance in addition to quite highly refractive, homogenous, greenish spherules which vary in diameter from a frac- tion of a micron to one micron. Sometimes the hyphae are seg- mented and bear lateral and terminal conidia, as well as occasional enlargements of the individual segments, in the course of the tube itself. The conidia and other swelHngs are from 5 to lo /"• in diame- ter and contain, usually, one, and occasionally 3 or 4 spherules, simi- lar in appearance to those noted in the bodies of the hyphae, but of considerably larger size, 4 to 7 /^ in diameter. In the larger conidia, the double-contoured membrane which forms their wall is very conspicuous. Some of the largest conidia are empty; their walls are nearly i At in thickness and present breaks in their continuity. Free in the suspending fluid, 0.85 per cent NaCl solution, about the hyphae under examination, are numbers of spherules exactly similar in appearance to those described within conidia. Their origin from the latter seems probable. They frequently present a marked Brownian movement. In acid-glucose-broth cultures a granular sediment forms in the course of a week or ten days, the overlying fluid remaining clear. On shaking, the tube becomes diffusely clouded but resumes its original appearance after standing a short time. Microscopi- cally the typical budding yeast forms appear. Occasionally elongated forms suggesting h>phae are found. The same change has overtaken the broth cultures that has been mentioned as occur- ring in the agar cultures, that is, the organisms no longer grow in the budding form but have reverted to the growth character- istic of room temperature, namely, the mycelial form. Whether or not it will be possible to induce the organism to resume the former mode of growth at incubator temperature (37° C.) is a question which must be left for further observation. No attempt has been made to determine the finer details of the biology and morphology of this organism, since it is believed Immunological Reactions of Oidiomycosis 187 that enough has been said to warrant the assumption that we are deahng with an organism which undoubtedly belongs to that genus of the pathogenic mould-fungi placed by Ricketts," provisionally, among the oidia, and which therefore may be considered a fair sample of the parasites responsible for oidiomycosis (blastomycosis) in man. Besides the organism itself, a so-called oidiomycetic "extract" was made use of in the following experiments. This "extract" was made according to a method which has been in use in this laboratory for a number of years, and which is as follows: The organism, in order to obtain considerable quantities, is planted on a large covered plate or in flat-sided, wide-mouthed bottles. When a good growth has appeared, which may take from three weeks to a month, it is scraped off by means of a glass or platinum hoe and placed in a desiccator where it is allowed to dry. The dried material is weighed, then placed in a porcelain-ball mill together with an equal volume of sterile sand and about lo c.c. of sterile 0.85 per cent salt solution and ground for a couple of hours. Salt solution in small quantities, 5 c.c, is then added at short intervals until the volume of fluid is equal to about 50 c.c. The liquid is drawn off, centrifugated, the supernated fluid poured into a sterile bottle or other suitable recep- tacle, and the solid portion, which still contains large numbers of yeast cells, is returned to the mill and reground. This process is repeated until microscopical examination fails to reveal unbroken cells in the centrifugated sediment. The various fluid portions are added together and the total volume made up to such a point that every 100 c.c. represents one gram of the dried organism. As a preservative, 0.5 per cent carbolic acid or 0.3 per cent chloroform is then added. As a further precaution the "extract" is kept in the ice-box. If one wishes a fresh extract the moist organisms may be used. It has been shown by experiment that the yeast loses about } ^ of its weight when dried. Therefore if one grinds 1 2 grams of fresh, moist organisms and wishes to make a I per cent solution comparable to the preserved extracts, he makes the total volume of extract 100 c.c. This, of course, is not a method of great accuracy, owing to the variations in the amount of water which the moist organisms contain, but it seems fairly satisfactory. Upon standing any length of time a white, amorphous precipitate separates out of the extract and settles to the bottom. It does not go into solution again when mixed with the overlying fluid. The extract itself has a yellowish, opalescent appear- ance, which tends to clear somewhat on shaking with ether. It possesses a markedly yeasty smell. The following rough qualitative analysis was made on "Ext. BIR 1/19/09." The results are typical of those observed in all cases examined. To 10 c.c. of the extract were added 30 c.c. of absolute alcohol, and the mixture was allowed to stand in the ice-box for 30 minutes. At the expiration of that time, a fairly heavy, whitish, flocculent precipitate had formed, which showed a tendency to remain suspended in the liquid in large and small loose clumps. The material was filtered. The filtrate was evaporated to dryness at a temperature not over 65° C. and yielded a yellow, gummy-looking material which possessed a marked sweetish, yeasty smell. This material was allowed to stand over night at room temperature. 1 88 Contributions to Medical Science The precipitate on the filter paper was washed with 30 c.c. of distilled water in which it seemed to be readily taken up, giving the water a whitish, cloudy appearance. It had no characteristic odor. Under the microscope, the aqueous solution showed a fine, granular debris, but no crystalline suspension; the residue from the evaporated alcoholic extract was amorphous and of a yellow tinge. The aqueous solution was tested for dextrin, starch, and reducing sugars with negative results. The protein color and coagulation tests resulted as follows: Heller's nitric acid test negative Boiling with glacial acetic " Millon's reagent " Xanthroproteic test " Adamkiewicz +very sharp Biuret ? very slight if any Liebermann negative Hopkins-Cole +very sharp H2SO4 (conc.)+sugar negative Similar tests applied to the whole extract gave similar results. The yellow residue from the alcoholic extract was soluble in ether and chloroform and was not precipitated from the chloroform solution by acetone. It gave a negative Salkowski's test for cholesterol. Further analysis was not attempted. No experiments were undertaken with killed organisms, the living organisms and the extract alone being used in the immuno- logical work. As a matter of routine, injections of living organisms were made into the peritoneal ca\dty; the pleural cavity was used a few times. Intravenous injections are inconvenient in the guinea-pig, while subcutaneous injections are open to the objection that the result- ing subcutaneous nodules early cause necrosis and sloughing of the overlying skin, thus not only giving splendid opportunities for secondary infections but tending to scatter virulent mould-fungi about the cages, which was regarded as dangerous. The extract was injected subcutaneously or intraperitoneally according to the seeming needs of the particular experiment in which it was being used. oidiomycosis (blastomycosis) in the gulnea-pig. Dosage. — In a series of experiments carried out about three years previously with this same organism. Dr. Ricketts had deter- mined roughly its virulence for guinea-pigs. He found that 0.3 of a gram of an agar plate culture, grown at incubator tempera- ture, injected into the peritoneal cavity of a 375-400 gm. guinea-pig Immunological Reactions of Oidiomycosis 189 would cause death in about 35 days, with generahzed oidiomycosis, especially in the abdominal and thoracic viscera. In view of the prolonged course of the disease, even following the injection of such large doses as that mentioned, it did not seem so essential for present purposes to determine a uniformly fatal dose as to find out what dosage would give a constant clini- cal picture, and, at the same time, would not require a month or six weeks for development. With this in view a series of 42 guinea- pigs were injected with quantities of moist organisms from agar culture, the dosage varying from 0.5 to o.oooi of a gram. Each pig was examined daily with reference to weight, temperature, and to changes which might be revealed by inspection or palpation. The following general conclusion was drawn: There is no con- stant symptom by means of which one may diagnose oidiomycosis in the guinea-pig during life. The weight is an unreliable guide, it varies with the abundance of food and the length of time which elapses between feeding and the taking of weights. Again, even if the animal develops an apparently fatal infection, the weight before death may approximately equal that before inocu- lation, or, if it does decrease, the major portion of the loss occurs during the last week or so of life. The temperature is also unsat- isfactory. In the majority of cases when the oidiomycosis cannot be doubted, the temperature runs an absolutely normal course. When it does follow an abnormally high curve, it is difficult to exclude secondary infection, or the possible efi"ect of occlusion of excretory ducts (that is, ureter, seminal vessels, urethra). If one compares the effects in males and females, however, he finds that there is, symptomatically at least, one marked differ- ence. In males with doses as low as o.ooi gm. one usually can detect small nodules (o . 5-2 mm. in diameter) in the testicles within seven to ten days after inoculation; nodules can practically always be found with doses of o.oi gm. within five to seven days. With doses of o.i gm. or over, nodules not only occur regularly but they are of good size and the infection may have a fatal out- come. In the female pig there is absolutely no sure way of diag- nosing oidiomycosis during life except to open the abdominal cavity and inspect the contents. One might think of deep pal- iQo Contributions to Medical Science pation, but how distinguish nodules from feces ? We have found it practically impossible in most cases. Of great assistance in determining the severity of the infection — and this may be applied to both sexes ahke — is the "look" and "feel" of the animals. By the term "feel" well is meant that when the animal is handled the muscles are found to have their normal firmness and tone. In fresh guinea-pigs which have not been handled, the tense, hard, wiry feeling of the muscles is very noticeable. After the animals have been manipulated daily for a week or so, they evidently become used to it and relax readily when they are picked up. In this state the muscles are soft and pliable but have not lost their tone. If such pigs become the victims of a chronic wasting dis- ease, this normal muscular tone is lost; the muscles become decidedly flabby to the touch so that the animal feels like nothing so much as the time-honored "dish-rag." This "feel" may be present before the animal's weight has fallen off appreciably. The animal usually appears thin and we say it looks sick, but this appearance is due more to the staring coat and the "hunched-up" attitude which the animal assumes than to a real emaciation. These symptoms, of course, are characteristic of cachectic condi- tions in general and are not specific. It should be mentioned that the development of nodules at the point of inoculation noted in some of the pigs was regarded as the result of faulty technic and could be avoided to a considerable extent by rinsing the needle in water before making the injection. For the remainder of this work, in view of the above facts, the taking of temperatures, the recording of weights, and the use of female guinea-pigs were eliminated excepting in special cases. Male pigs were used as a matter of routine and their condition was adjudged by the results of careful palpation of the testicles and by their general "look" and "feel." o.i gm. of moist agar cul- ture was adopted as the standard, surely infective dose. The difference in the results of intraperitoneal inoculation in male and female guinea-pigs is very marked from the symptomato- logical standpoint. Take guinea-pig 75 and guinea-pig 81, for example. Each received the same dose from the same culture on the same day. Guinea-pig 75 (male) became weak and flabby Immunological Reactions of Oidiomycosis 191 to the touch ^vith a staring coat and enormously swollen scrotum which finally ulcerated. Guinea-pig 81 grew fat and sleek and it was only at autopsy that signs of the infection could be demon- strated. Some of the possible reasons for this diflference will be discussed later as will also the question of localization of infection, and, in a general way, the course of the disease and possible treat- ment. active immunization. In the endeavor to establish an active immunity in guinea-pigs, use was made of the living organism and of the extract. In the case of the living organism, the method pursued for the greater part of the time was to reinoculate a pig as soon as he had made an apparently complete recovery from his previous infection, and then to study the development of the new infection as regards (i) the time of the appearance of nodules in the testicles, their size, consistency, and the rate at which they disappeared, (2) the development of palpable nodules in the abdominal cavity, and (3) the "look" and "feel" of the animal. The results of such observations had led to the belief that it made practically no differ- ence how many times an animal was subjected to oidiomycetic infection and recovered, he always retained his original suscepti- bility. In order to put this conclusion to a thorough test the following experiment was performed. All the living pigs which had recovered from previous injections were gathered together with ten normal pigs of about the same average size and all were given an equal dose of living oidiomycetes (approximately o.i gm.) and the results carefully watched. The previous history of the so-called "immune" animals is given in Table i. On comparing the results of the inoculation in the immune and the control guinea-pigs, some sHght, but quite well marked differences were noted. First, regarding the course of the infec- tion, it is a fact worthy of remark, that in so far as the male pigs of the two series are concerned, the immune animals made a decid- edly more rapid recovery than the controls, as judged by the rate of disappearance of the nodule from the testicles. The female pigs offer no accessible basis for comparison. If we examine the 192 Contributions to Medical Science temperature charts of the immune and normal animals before and after injection we again find differences. The average tempera- ture for the immunized animals for the three days prior to the inoculation on which temperatures were observed, that is, the 19th, 20th and 23d, is 103.29°+ F.; that for the control animals is 103.31°+ F. The average of the observed temperatures of the immunized animals on the days immediately following the injec- tion (the 24th, 25th, and 27th) is 104.27°+ F.; that of the control History of ' TABLE 'Immune' I. ' Guinea-Pigs. GUINEA- PlG No. No. Times Previous- ly Inocu- lated Average Interval BETWEEN Inocula- tions Time Elapsed Since Last Dosage (in gm.) Inocu- lated I St Dose 2d Dose 3d Dose 1 4th Dose 3 4 I 4 4 3 4 2 2 2 2 2 3 6 1 days 24 " 24 " 24 " 33 " 24 " 24 25 " 2S " 25 " 32 " 68 davs 0.1 0.1 0.1 0.1 0.1 0.1 0.1 72 68 ' 166 ' 68 ' 68 ' 68 ' 68 ' 68 ' 240 ' 50 ' 50 ' 50 68 ' , cx>5 S OOI 0001 S I 75 80 1 81 83 O.l 47 93 94 95 69 guinea-pigs for the same period is 103.49°+ F., or practically 104.3° F. in the one case and 103.5° '^^ the other — a difference of nearly 1° F. To be sure, this difference is small, but in view of the fact that the temperatures of the two groups of pigs were almost identical before inoculation, while following the inoculation the rise was confined almost entirely to one group, it would seem that one would be justified in considering such a rise a posi- tive reaction of the animal organism against the injected oidio- mycetes. Apparently, then, a low grade of immunity is developed in guinea-pigs by the intraperitoneal inoculation of living organ- isms. As symptomatological evidence of the immunity — which seems to be emphatically a relative immunity — we have the more rapid disappearance of the lesions following infective doses and a small but fairly decided rise in temperature on the days imme- diately following the inoculation. These reactions may be mani- fest as long as 240 days after the last inoculation (guinea-pig 49) Immunological Reactions of Oidiomycosis 193 or, allowing the animal 40 days in which to recover from the immunizing inoculation, a liberal margin, the acquired power to react against renewed infection in a guinea-pig which has recovered from two moderate injections may persist at least 200 days. A large series of guinea-pigs were given repeated injections of the extract. Of this series but four pigs were tested by the injec- tion of li\dng organisms. Their histories follow: TABLE 2. No. of Guinea-Pig No. of Injec- tions Interval be- tween First and Last In- jection Interval be- tween Last In- jection and Day Tested Average Vol- ume of Injec- tion Route 13 37 38 S6 178 days 10s " los " 91 " 63 days 77 " 77 " 63 " About I . s c.c. 2 c.c. Intraperitoneal Each of these four pigs, together with each of seven control animals, was given o.i gm. BIR, intraperitoneally. All of the control animals developed severe infections, large nodules in the testicles, weakness, staring coat, etc.; and one of them died, post- mortem examination revealing generaUzed oidiomycosis. Of the immune animals, but one developed testicular lesions — a small, solitary nodule; all presented rather large, fairly hard, intra- abdominal masses, but no other symptoms. The subsequent history of the immune pigs follows. Guinea-pig 13 was bled to death on December 3, 1909, 102 days after injection. At autopsy the seminal vessels and portions of the small intestines were found adherent to a fibrous nodule on the anterior abdominal wall, within which was fluid detritus containing oidiomycetic forms; cultures were negative. Microscopic sections show oidiomycetes. No other abnormalities. Guinea-pig 37 died January 19, 1910, 149 days after injection. At autopsy a mass of dense fibrous adhesions was found about the point of inoculation. In the center of this mass was a cavity i cm. in length and 0.5 cm. in width and about i to 2 mm. deep in which was fluid detritus. Testicles were normal. No oidiomycetes were found, microscopically, in the contents of the nodule. The cause of death was pneumonia and pericarditis. No oidiomycetes were recovered in cultures. 194 Contributions to Medical Science Guinea-pig 38 was found dead on March 14, 1910, 203 days after injection. Autopsy showed pneumonia and old fibrous adhesions about the point of injection. Testicles were normal. Guinea-pig 56 was killed March 14, 1910, 203 days after injec- tion. Autopsy disclosed an oval fibrous sac about i cm. in greatest length attached to the anterior abdominal wall. Adherent to the sac were the seminal vesicles and portions of the small intestines. Within the sac was fluid detritus; microscopically forms very similar to oidiomycetes were observed; unfortunately these bodies were soluble in ether. Cultures were negative as regards oidiomycetes. One cannot argue from these experiments that the prolonged immunization with the ''extract" resulted in the development of such a high grade of immunity as would lead to the absolute walling off of infective doses of organisms at or near the point of inoculation, for it is not at all certain that those inoculations may not have been made into previously existing adhesions, resulting from the numerous immunizing injections, which could not be detected by external examination. On the other hand, in view of the fact that most of the control animals developed marked infection, one of them dying, and that the nodules which appeared in the testicles of one of the immune animals vanished with con- siderable celerity, it seems justifiable to assume that a certain degree of immunity, at least, was manifested. Another point of interest which will be considered later was the finding of mor- phologically typical oidiomycetes in stained sections from guinea- pig 13. Apparently this pig's immunity, whatever it may have been, was not of a markedly lytic t3^e. EXPERIMENT TO TEST THE EFFECT OF THE INJECTION OF " EXTRACT" INTO INFECTED ANIMALS. Eggers has shown {vide infra) that repeated intraperitoneal injections of oidiomycetic extract seem to decrease the resistance of guinea-pigs to accidental infection. He has also mentioned the fact that the animals so treated suffer a considerable loss of weight. It has been noticed in the present experiments that this loss of weight in guinea-pigs receiving 2 c.c. intraperitoneally Immunological Reactions of Oidiomycosis 195 every fourth day may be extreme, the animals become flabby to the touch and appear markedly cachectic. They slowly recover when the injections are discontinued. Bearing in mind the above findings, it became of interest to test infected animals with gradu- ated doses of the extract. The results, as shown in the following experiments, appear to be of possible importance. A series of 10 guinea-pigs which had recovered from spotted fever eight months to a year before were given intraperitoneal injections of o.i gm. of oidiomycetes from an agar slant culture on January 21, 1909. As soon as nodules became palpable in the testicles, five of these pigs were given subcutaneous injections of extract ("BIR 1/19/09," CHCI3 preservative) as indicated below, the other five animals being used as controls. Guinea-pig 45: January 21, 1909, control pig; wt. = 68o gm. o.i gm. intraperito- neally; January 25, nodules palpable in testicles; February 8, nodules in testicles about I cm. in diameter; ]\Iarch i, testicles normal. Guinea-pig 46: January 21, 1909, control pig; wt. = 775 gm. 0.1 gm. intraperi- toneally. Result like guinea-pig 45. Guinea-pig 47: control pig; wt. = 5i5 gm.; January 25, slight induration of testicles; February 8, same; true nodules never developed; March i, testicles normal. Guinea-pig 48: control pig; wt. = 5i5 gm.; January 25, slight induration of testicles; February 8, testicles contain nodules, some of which are about i cm. in diameter; March i, testicles normal. Guinea-pig 49: control pig; wt. = 955 gm.; January 25, small nodules can be felt in the testicles; February 8, large nodules are palpable in the testicles; March i, testicles normal. Guinea-pig 50: January 21, 1909, wt. = 69o gm. 0.1 gm. oidiomycetes intraperi- toneally. Received subcutaneous injection of 0.5 c.c. extract January 25, 30, Februarj' 4, 9, 15, 23; January 25, nodules palpable in testicles. Small nodule at point of inoculation drained and sterilized with 95 per cent carbolic acid and absolute alcohol; wound healed rapidly. Pus contained large numbers of oidiomy- cetes. February 8, small nodules in testicles; February 28, animal fully recovered. Guinea-pig 51: January 21, 1909, wt. = 68o gm. 0.1 gm. oidiomycetes intraperi- toneally. Received 0.5 c.c. extract subcutaneously January 25, 30, February 4, 9, 15, and 23; January 25, nodules in testicles; February 8, nodules in tes- ticles nearly as large as those in control pigs; March i, animal fully recovered. Guinea-pig 52: January 21, 1909, wt. = 690 gm. o.i gm. oidiomycetes intraperi- toneally. Received 0.5 c.c. extract subcutaneously January 25, 27, 29, Feb- ruary 1,3, 5,8, lo, 12, 15, 17, 23, and 25; January 25, slight hardening of tip of right testicle noted; February 8, nodules in testicles fully as large as the largest among the control pigs; February 25, on this date the nodules had almost disappeared so that the injections were discontinued. Guinea-pig 53: January 21, 1909, wt. = 72S gm. 0.1 gm. oidiomycetes intraperi- toneally. Received 0.5 c.c. extract subcutaneously January 25, 27, and 29; 196 Contributions to Medical Science January 25, nodules palpable in testicles; February i, found dead. Autopsy: The abdominal parietes, testicles, seminal vesicles, and diaphragm are peppered with nodules varying in diameter from 0.5 mm. to 4 mm. A few similar nodules appear on the surfaces of the spleen and liver. In the great omentum along the greater curvature of the stomach is a thick fibrinous mass; a mass of cheesy pus I cm. in diameter and 0.5 cm. in thickness surrounded by a hyperaemic zone occurs on the anterior abdominal wall at the point of inoculation. A few grayish patches, i to 3 mm. in greatest width, appear beneath the epicardium and seem to involve the heart muscles. The lungs do not appear to be so collapsible as normal; they appear to be congested; no gross nodules. Cultures from the various nodules show oidiomycetes in pure culture. Histological examination of the lungs shows that there is no edema nor e.xudation. Guinea-pig 54: January 21, 1909, wt. = 6io gm. o.i gm. oidiomycetes intraperi- toneally. Received 0.5 c.c. extract subcutaneously February i, 3, 5, 8, 10, and 12; February i, nodules palpable in testicles; February 8, fairly large nodules in testicles; February 15, found dead. Autopsy: The right testicle is trans- formed into a sac of pus containing oidiomycetes in large numbers. Scattered nodules i to 5 mm. in diameter are found on liver, spleen, mesentery, left testicle, and abdominal wall. A nodule in the intermuscular fascia of the anterior abdom- inal wall is I cm. in diameter and 2 to 3 mm. in thickness, and contains an enor- mous number of budding oidiomycetes. The left lung is very much enlarged and occupies three-fourths of the thoracic cavity; it is hard, non-collapsible, dark in color, with hemorrhagic spots. Heart and right lung appear normal. Material from various nodules smeared on glucose agar slants yielded oidiomy- cetes in pure culture. Histologically the left lung appears to be in a state of pneumonic consolidation. This experiment suggests that in the extract we may have a means of modifying to a large extent the course of an oidiomycetic infection in guinea-pigs. Those animals which received 0.5 c.c. of the extract subcutaneously every fifth day seemed to suffer a milder infection than the control pigs and to make, possibly, a somewhat more speedy recovery ; of those animals receiving o . 5 c.c. extract subcutaneously every second day, one— guinea-pig 42 — paralleled very closely the control pigs in this infection; guinea-pig 53 died in 11 days of generalized oidiomycosis, and guinea-pig 54 died 25 days after injection, autopsy showing exten- sive oidiomycosis of the abdominal parietes and viscera with a terminal lobar pneumonia; 0.5 c.c. of the extract injected sub- cutaneously every second day into infected guinea-pigs may be toxic; 0.5 c.c. of the extract injected subcutaneously every fifth day into similarly infected guinea-pigs apparently is not toxic and seemingly increases the animal's resistance to the disease. Broad generalizations from a single experiment such as this are, of course, Immunological Reactions of Oidiomycosis 197 unwarranted, but it seems possible that further work along the lines indicated by these results may lead to the development of a technic which will enable us to treat effectively oidiomycetic infections. EXPERIMENTS IN ANAPHYLAXIS. In testing further the properties of the "extract," an attempt was made to elicit with it the phenomenon of anaphylaxis in guinea-pigs. It was with such an end in view that the following experiments were performed. The temperature of the animals was observed only in cases where specific mention of that fact is made. Experiment i. — Guinea-pig i received an intraperitoneal injection of i c.c. of "Extract BIH2," which had been preserved in 0.5 per cent phenol for about 13 months, on October 27, 1908. Sixteen days later 5 c.c. of the "extract" were injected intraperitoneally. Result: no convulsions nor tremors of any kind; pig alive and well 24 hours later. Experiment 2. — Guinea-pig 17 was treated with doses of "Extract BIH2," similar to those given guinea-pig i. An interval of 27 days separated the two injections. Results: Negative, similar to Experiment i. Experiment 3. — Guinea-pig 17 received an intraperitoneal injection of 0.05 gm. of organism BIR on November 12, 1908. Twelve days later the animal was found to have lost 90 gra. in weight; both testicles had become swollen and hard. Five c.c. of "Extract BIH2" were injected intraperitoneally. There ensued a temporary fall of about 8° F. in temperature which returned to normal within 24 hours. No further symptoms appeared. The animal thereafter gained weight. Experiment 4. — Guinea-pig 31 was given 5 c.c. of centrifugated "Extract BIR" (preservative 0.5 per cent phenol) intraperitoneally on January 4, 1909. The injection was repeated on January 15, without the production of symptoms. Experiment 5. — Guinea-pig 40 received o . 2 c.c. " Ext. BIR" (preservative o . 5 per cent phenol) intraperitoneally on January 15, 1909. Guinea-pig 41 received 0.5 c.c. "Ext. BIR" intraperitoneally on January 15. Guinea-pig 42 received i.o c.c. "Ext. BIR" intraperitoneally on January 15. Guinea-pig 43 received 1.5 c.c. "Ext. BIR" intraperitoneally on January 15. Guinea-pig 44 received 2.0 c.c. "Ext. BIR" intraperitoneally on January 15. On January 26 — interval of 1 1 days — guinea-pigs 40 and 43 received 5 c.c. of Ext. BIR intraperitoneally — no reaction. On January 28 guinea-pig 41 and guinea-pig 42 with a normal con- trol pig received 7 c.c. Ext. BIR intraperitoneally. Negative results. Guinea- pig 44 died of pneumonia before he was tested. Experiment 6. — Guinea-pig 55, wt. 660 gm., received o.i c.c. Ext. BIR (preserva- tive CHCLj 0.3 per cent) intraperitoneally January 22, 1909. Guinea-pig 56, wt. 795 gm., received 0.7 c.c. Ext. BIR intraperitoneally. Guinea-pig 57, wt. 830 gm., received i . 5 c.c. Ext. BIR intraperitoneally. Guinea-pig 58, wt. 680 gm., received 3.0 c.c. Ext. BIR intraperitoneally. After an interval of 14 days, each of these four animals and each of two normal control pigs were given an 198 Contributions to Medical Science intraperitoneal injection of 5 c.c. of the extract. Only one animal presented any untoward symptoms and one of the control animals died in about 12 hours. Postmortem examination revealed an acute fibrinous peritonitis. Experiment 7. — Each of three small guinea-pigs averaging about 250-300 gm. in weight was given 0.3 c.c. of Ext. BIR intraperitoneally on February 26, 1909. The chloroform used as a preservative had been allowed to evaporate before the extract was used. After an interval of 17 days each of the above animals and two normal pigs of similar size received 5 c.c. of chloroform free extract intraperi- toneally. All of the animals appeared to be sick immediately following the injection. The three sensitized pigs were found dead next morning. One of the control animals died a week later without apparent cause; the other remained in normal condition. Experiment 8. — A couple of grams of moist oidiomycetes were ground in the porcelain- ball mill for three hours, and then taken up with NaCl solution (0.85 per cent). Four c.c. of this fresh extract were injected into each of two medium sized guinea- pigs. Twenty-six days later, each of these two animals and two normal pigs were given 10 c.c. of a fresh extract intraperitoneally. The two sensitized animals were foimd dead the following morning; one control pig died two days later, the other a week later. This extract yielded the usual white, flocculent precipitate upon the addition of absolute alcohol. Experiment g. — Fresh moist organisms were ground in a mortar with sterile sand in 0.85 per cent NaCl solution containing 0.2 per cent of normal NaOH. The total volume of the ground material was made up to 25 c.c. by the addition of the salt solution containing o. 2 per cent of normal NaOH. Five c.c. of the total was sediment. The fluid, when injected, was yellowish white and opaque. Guinea-pig (a): wt. 140 gm., received o.i c.c. intraperitoneally. Guinea-pig (i): wt. 215 gm., received 0.3 c.c. intraperitoneally. Guinea-pig (c): wt. 225 gm., received 0.5 c.c. intraperitoneally. Eighteen days later guinea-pig (6) and a normal guinea-pig of similar size were given an intraperitoneal injection of the original fresh extract, which had been kept in the ice-box, warmed to 37° C. The extract at this time yielded the usual precipi- tate with absolute alcohol. Immime Pig (6): Temp, before injection, 10:30 a.m., 102° F. Injection 10:50 a.m. Normal Pig: " " " " " 100.6° F. " Immune Pig (6): Temp. 10:55, 100. 2° F. Normal Pig: " " 99.8°F. 12:1s 1:4s 2:45 3:4s 4:4s 5:43 10:30 A.M. 99.8° F. 98. 2° F. 97° F. 97.6° F. 98° F. 99. 2° F. 102. 6°F. 96.8° F. 97.6° F. 96. 8° F. 97-3°F. 98. 2° F. 98.i°F. 100. 6° F. Two days after this, an interval of 20 days after sensitization, guinea-pigs (a) and (c) and each of two control animals of a size approximating that of pigs (o) and (c) were given intraperitoneally 5 c.c. of a freshly prepared extract, warmed to 37° C, similar to that originally employed. Injection at 12:20 p.m. Temp, at 11:50 12:4s 1:50 2:30 Guinea-Pig (a) : 102° F. ioi.4°F. 98.8° F. 944° F. Control: 100.4° F. 94° F. 96.4° F. 95.4° F. Guinea-Pig (c): ioi.o°F. 100. 4°F. 97.8°F. 94.2°F. Control: 101° F. 100. 2° F. 96.8° F. 94.6° F. 4:3s 11:00 A.M. 96.4° F. 102° F. 96° F. 102° F. 94° F. 104° F. 94.4° F. 100.8° F. Immunological Reactions of Oidiomycosis 199 In addition to the fall in temperature, pigs (6) and (c), and, to a lesser extent, (a) , developed a tense, swollen, and apparently painful abdomen which subsided as the temperature rose to normal. There was no suggestion of such a condition among the control pigs. From the above experiments, the conclusions seem justified that: (i) The extract is toxic ahke for normal, sensitized, and infected animals as indicated by the sharp fall in temperature following intraperitoneal injection of considerable quantities. (2) It seems possible to demonstrate the phenomenon of anaphy- laxis by the use of freshly prepared extract, or of extract freed from its preservative (CHCI3) , the reaction manifesting itself either by the death of the animal within 1 2 to 20 hours, or in the develop- ment of a sharp intraperitoneal reaction within two or three hours following the intoxicating injection, which subsides within 24 hours. Large doses may kill the control animals; in such cases, the death of the sensitized animals usually precedes that of the normal animals. passive immunization. In view of the seemingly low grade of active immunity which guinea-pigs developed against oidiomycetes, it appeared that efforts to confer a passive immunity could be more profitably expended along other lines so that no such experiments were undertaken. In Dr. H. T. Ricketts' notes on his work on oidio- mycosis, which he kindly placed at my disposal, the following experiment is reported: Rabbit 44, was immunized against BIR as follows: November 19, 1905, i tube of glucose-agar culture (killed) subcutaneously. Local abscess produced. November 26, 1905, i tube of glucose-agar culture (killed) intraperitoneally. December 4, 1905, i§ December 12, 1905, 15 December 19, 1905, i^ December 26, 1905, 15 January 7, 1906, 2 January 17, 1906, 2 January 23, 1906, i gram living agar culture intraperitoneally. February 9, 1906, i " " " " " February 16, 1906, 12 c.c. of blood drawn. This blood was defibrinated, centrifu- gated, and the serum used in the following experiment: 2CX> Contributions to Medical Science Experiment lo. — February i6, 1906. Weighed amounts of moist agar culture were mixed with varying quantities of "immune" rabbit serum and injected into the peritoneal cavity of a series of guinea-pigs. Control experiments were carried out with normal rabbit serum and with the untreated organisms. TABLE 3. Date Wt. in Grains Result Guinea-pig i received 0.3 gm. Bl-l-2 c.c. im. serum, intraperi- toneally on 2/16. Guinea-pig 2 received 0.3 gm. Bl-t-i c.c. im. ser.-fi c.c. NaCl sol. (0.85 per cent) intraperitone- ally on 2/16. Guinea-pig 3 received 0.3 gm. Bl-(-o.s c.c. im. ser.-f-i.s c.c. NaCl sol. (0.8s per cent) intra- peritoneally on 2/16. Guinea-pig 4 received 0.3 gm. Bl-j-2 c.c. normal serum, intra- peritoneaJly on 2/16. Guinea-pig s received 0.3 gm. Bl-f I c.c. normal serum-f-i c.c. NaCl sol. (0.85 per cent) intra- peritoneally on 2/16. Guinea-pig 6 received 0.3 gm. Bl-l-0.5 c.c. normal serum -|- i.S c.c. NaCl sol. (0.85 per cent) intraperitoneally on 2/16. Guinea-pig 7 received 0.3 gm. BH-2 c.c. NaCl sol. (0.85 per cent) intraperitoneally 2/16. Guinea-pig 8 received dose similar to that of guinea-pig 7. 2/16/06 3/ 7/06 3/16/06 3/24/06 2/16/06 3/ 7/06 3/16/06 3/24/06 2/16/06 3/ 7/06 3/16/06 3/23/06 2/16/06 3/ 7/06 3/1 1/06 2/16/06 3/ 7/06 3/18/06 2/16/06 3/ 7/06 3/16/06 2/16/06 3/ 7/06 3/16/06 3/29/06 2/16/06 390 39S 360 270 37S 370 340 24s 378 3SO 28s ? 400 400 ? 400 375 270 42s 410 380 386 36s 32s 230 390 Result: Death on 3/24 from gen- eralized oidiomycosis; interval of 36 days. Total loss of wt. = 120 gm. Result: Death on 3/24 from gen- eralized oidiomycosis; interval of 36 days. Total loss of weight = 130 gm. Result: Death on 3/23, from gen- eralized oidiomycosis; interval of 35 days. Result: Death on 3/11, from gen- eralized oidiomycosis; interval of 23 days. Result: Death on 3/18 from gen- eralized oidiomycosis; interval of 30 days. Total loss of weight = 130 gm. Result: Animal lost. Result: Death from generalized oidiomycosis on 3/29. Interval of 41 days. Total loss of weight = 156 gm. Result: Accidentally killed 2/22/06, six days after injection. One cannot argue from these results that the "immune" serum conferred an immunity of any degree. It should be remembered in this connection, however, that rabbit serum is somewhat toxic for guinea-pig leukocytes, a property which might mask the pro- tective power of such a serum. There seems to be a suggestion of a reaction of this kind in the results of the above experiment, but more work must be done before conclusions may be drawn. immunological reactions. Some preliminary work along this Kne, carried out with the organism "BIR," was done in 1905 by Dr. H. T. Ricketts, and continued during the year 1906-7 by Dr. H. E. Eggers. Dr. Ricketts performed the following experiment: A guinea-pig was immunized against the oidiomycetes by the repeated injection of Immunological Reactions of Oidiomycosis 20I killed organisms and of extract as is shown in the accompanying table. Guinea-Pig i. Inoculated with o. I gm. killed agar culture intraperitoneally O.I " O.I " " " " " 2 c.c. of extract subcutaneously Killed on account of loss of weight, and serum used for precipitation tests. Precipitation Tests. April 4, 1906. Blastomyces R — Extract II. Serum from Immunized Guinea-Pig i. I. Original Extract II, after prolonged centrifugation to remove all sediment, was used as antigen. The fluid was whitish-opalescent. Date Weight February 7 February 16 February 27 March 9 March 16 390 gm March 30 31S " Extract Immune Serum Normal Serum NaCI (0.8s per cent) Sol. Result o.S c.c. o.S " "S ! o.S o.S " 0.5 ' 0.1 c.c. 0.2 " 0.3 " 0.0 " 0.0 " 0.0 " 0.0 c.c. 0.0 " 0.0 " 0.1 " 0.2 " 0.3 " 0.4 c.c. 0.3 " 0.2 0.4 " 0.3 " 0.2 Distinct precipitate Marked " Fairly heavy " No II. Absolute alcohol precipitate of Extract II, freed of alcohol and redissolved in NaCl (0.85 per cent) solution. Fluid was slightly opalescent but much less so than the centrifugated extract. Redissolved Precipitate Immune Serum Normal Serum NaCl Sol. (o.8s per cent) Result o.S c.c. o.S " o.S o.S " OS " o.S 0.1 c.c. 0.2 " 0.3 " 0.0 " 0.0 " 0.0 " 0.0 c.c. 0.0 " 0.0 " 0.1 " 0.2 " 0.3 " 0.4 c.c. 0.3 " 0.2 " 0.4 " 0-3 " 0.2 " Distinct precipitate Marked " Fairly heavy " No •; III. Clear, colorless, non-opalescent filtrate of Extract II used as precipitogen. Filtrate Immune Serum Normal Serum NaCl Sol. (0.8s per cent) Result o.S c.c. o.S " o.S 0.5 0.1 c 0.2 0.3 0.0 0.0 0.0 .c. 0.0 c.c. 0.0 " 0.0 " 0.1 " 0.2 " 0.3 " 0.4 c.c. 0.3 " 0.2 " 0.4 " 0.3 " 0.2 " No precipitate V'ery faint precipitate Small, but distinct precip. No precipitate Normal serum alone = no precipitate. Immune serum alone = no precipitate. Centrifugated extract alone = no precipitate. Redissolved alcoholic precipitate alone = no precipitate. Filtrate alone = no precipitate. 202 Contributions to Medical Science The records of work done by Dr. H. E. Eggers were lost before the present work was begun. Dr. Eggers has, however, written a summary of his work from memory which is introduced at this point with his permission : The following work on attempted immunization to blastomycosis was carried on, on guinea-pigs, rabbits, and goats. The plan was followed of injecting into the animals, at periods of from seven to eight days, an extract prepared from the dried blastomyces by grinding. Before injection care was taken to shake the material thoroughly to secure suspension of the solid fragments of the organisms. Using this suspension, work was begun on 12 guinea-pigs and 9 rabbits. 0.5 c.c. of the material was injected into the guinea-pigs, i c.c. into the rabbits. Injec- tions were for the most part intraperitoneal; they were repeated every seven to eight days. The condition of the animals was controlled by weighing. If any considerable loss of weight followed an injection, one, occasionally more, periods were passed over without injection. In the case of some of the animals a sharp reaction followed the first few injections; several died within two or three days. Postmortem examina- tion revealed acute parenchymatous changes, particularly in the liver and kidneys. Other animals were substituted for the ones so lost. Several of the rabbits and guinea- pigs became pregnant during the course of the work; such animals were injected subcutaneously while pregnant. These injections were carried on from early in October, 1906, until January, 1907. At this time serum from several of the rabbits was tested with the homologous extract for a precipitation reaction. A positive reaction, not given beyond a dilution of 1 : 20, was found at this time with one animal. The serum of another one, which had shortly previously given birth to young, as well as the serum of the young was tested, both being negative. With the animal showing the slight positive serum reaction the injections were kept up for a month; no increase in reaction resulted, and shortly after this the rabbit died of a meningitis. The periodic injection was carried on longer, the doses being increased until the guinea-pigs received i c.c. at a time and the rabbits two. Evidently their vitality was considerably lowered by the repeated injections, as a considerable number, of the rabbits especially, died of secondary infections — meningeal for the most part. Although the injections of the few remaining rabbits were continued until June, 1907, in no case was I successful in getting a precipitation reaction beyond a dilution of 1:20. The rabbits showing even this reaction were too few to enable any decisive test as to their degree of immunization to be made. Early in April, 1907, that is, six months after injection had been begun, six of the eight guinea-pigs that had been injected a sufiicient number of times to warrant e.xpectation of results were injected intraperitoneally with varying doses of a freshly prepared suspension of blastomyces, the doses being 0.25, 0.5, i, 2, 3, and 4 c.c. respectively. Six normal animals were injected with similar doses. In no case did any of the animals die with any of the findings of blastomycosis. The general lowered resistance of the previously injected animals was betrayed by the fact that for the most part they were survived by the normal animals. Late in October, 1906, weekly injections of a goat were begun, the animal receiving at the start 3 c.c. of the extract subcutaneously, later increased gradually to 5 c.c. Immunological Reactions of Oidiomycosis 203 The serum of this animal, tested late in January, was found to give a positive precipitin reaction with the homologous extract in a dilution of i : 25. With the extract of another strain of the organisms — "H II" — it gave a precipitate in a dilution of 1:12.5; with the third strain no reaction at all occurred. Injections were continued, until the animal was noticed to be becoming thin and feeble. It finally succumbed late in February; postmortem examination revealed a bronchopneumonia, other tissues negative. The serum at the time of its death reacted in the same manner and in similar dilutions to the results just given. Work was at once begun with another goat, the injections, of 3 c.c. each, being made subcutaneously at eight day intervals. This was kept up until the beginning of June, three months, at which time precipitin reactions similar in every way to those obtained with the preceding animal were found. As the formation of antibodies with this method of repeated small injections was evidently rather slight, a second goat was obtained, and injected with larger and ascending doses at somewhat greater intervals, 10 or 11 days. The first dose, of 20 c.c. of extract, given subcutaneously, produced a marked reaction, the animal being quite sick for three or four days afterward. Local reaction at the site of injec- tion was extremely slight. The next dose of 30 c.c. was again followed by a marked reaction, the third dose of 40 c.c. by considerably less. The same plan was followed on rabbits and guinea-pigs. Before this time the writer left Chicago for the summer, and the injections were made without the posses- sion of facilities for testing the serum reactions of the animals. The change was disastrous to the rabbits, as all of them died of an acute enteric affection. The quantity of extract on hand did not warrant work being begun on new rabbits. The guinea- pigs were injected successively with doses of i, 3, 5, and 7 c.c. After two doses of this last amount had been given, it was dropped back to 2 c.c. per dose; the dosage of the goat, after reaching 40 c.c, was dropped back to 10. In the case of the guinea- pigs, the reaction from the larger doses used here was no more marked than from all the first few smaller doses used previously. ^\^lile out of the city, the first of the two goats, which was still being injected with the repeated small doses of extract, died; postmortem examination showed a condition of apparently long-continued pylorospasm, caused by the lodging of a pin in the gastric mucosa. The remaining animals, the second goat and the guinea-pigs, were brought back to the city. At this time the writer was obliged to discontinue the work. In the fall and winter of 1908-9 the work on precipitins was continued, guinea-pigs being used as the experimental animals. The results are tabulated below. Some of the animals were immu- nized by injections of the extract of another organism than BIR, known in the laboratory as Bl H II. The experiment of Dr. Ricketts with the centrifugated, filtered, and redissolved alcoholic precipitate of the extract was repeated, using the serum of guinea-pig 5 as the immune serum, with positive results in all cases. Thereafter the centrifugated extract was used exclusively as the precipitogen in the precipitation experiments. 204 Contributions to Medical Science The technic used in the work was as follows: Blood was drawn, usually from the guinea-pig's heart, defibrinated, and centrifugated, and varying quantities of the clear serum added to a constant amount of centrifugated extract in small test-tubes, and the mix- tures made up to a uniform volume by the addition of NaCl solu- tion (0.85 per cent). The tubes were then incubated at 37° C. for two hours, placed in the ice-chest, and results observed after 24 hours. Control tubes of normal serum plus extract in dilutions similar to those of the immune serum, and of immune serum and NaCl solution, normal serum and NaCl solution, and extract plus NaCl solution, were always prepared as a part of each experiment. It was noticed that occasionally after 72 to 96 hours in the ice- chest precipitates formed in tubes, which at the end of 24 hours were negative. As the control tubes remained clear and without sediment, and as no proofs of contamination could be obtained by cultural methods or by direct microscopical examination, it was assumed that such precipitates probably represented a specific reaction and they were therefore noted in subsequent experiments. Once in a while, the serum of normal animals formed a precipitate with the extract. Of the 22 normal animals tested, 2 or approxi- mately 10 per cent gave positive reactions, one in a dilution of 1 13 after 24 hours and the other in a dilution of i : 10 on the 4th day. In Table 4 the figures 1:7, 1:10, etc., represent the highest dilutions at which precipitates appeared; a dash ( — ) indicates that no observations were made and "o" indicates absence of precipitate. Six animals were tested more than once. Their history appears in Table 5. Of the 32 immune sera tested, 22 reacted positively in from 24 to 96 hours. Nine were absolutely negative in so far as they were observed. Of the sera of those animals which were immunized by the injection of extract alone, 8 reacted positively in 24 hours, o in 48 hours, 7 in 72-96 hours in one or another of the tests, and 4 were negative; of the sera of those which received the organisms alone, 4 reacted positively in 24 hours, o in 48 hours, i in 72-96 hours, and 4 were negative ; of the sera of the animals which received both extract and organisms, 2 were positive in 24 hours and i was Immunological Reactions of Oidiomycosis 205 TABLE 4. Precipitins. GUINZA- PlG No. No. Injec- tions OF No. OF Days Inter- val Since Average Dose No. In- jection Living Organ- isms Dose Gm. No. OF Days In- terval Since Re- covery FROM Last Injection Results Ext. HII Ext. BIR Last c.c. Injec- ! tion 24 Hrs. 48 Hrs. 72-96 Hrs. I s* 8 10 I 17 23 31 32 36 39 43 49 50 &:::::: S6 S7 S8 7S 83 9a 9S II Ill 4 9 5 4 8 2 S 8 4 9 6 6 13 9 2 2 4 37 13 28 24 6 14 11 IS 119 122 '11 48 48 28 2 2 ( H II I i BIR 2 2 s 2 2 2 2 S subcu. S subcu. 5 subcu. 3 3 3 ist dose 20 c.c. ( 2d 10 c.c. 3 I I 2 I I I 2 S 3 3 O.I o.os 0.01 aver, o.i O.I O.I O.I ( 1st dose ( 2d O.I (2d O.s < Others O.I ( each aver. . i 28 about 26 20 about 19 119 122 about 133 about 60 about 60 1:20 1:20 1:20 1:7 i:S ti:io Ji:lo 1:10 1:10 1:20 1:30 t:30 1:20 1:20 ?(o) 1:7 1:10 1:10 i: 10 * Not tested in higher dilution. t Ext. Hi used as precipitogen. t Blood drawn July 13 but not tested until July 22. TABLE s. GtJINEA Pig No. Total No. Injections before No. OF Days Interval BETWEEN Time of RF-SITLTS ist Test 2d Test 3d Test Last Injection or Infection and Ext. Liv. Orgs. Ext. Liv. Orgs. Ext. Liv. Orgs. ISt Test 3d Test Test ist Test 2d Test ■fL 13*.... 3S 37 38 55 81 17 8 8 8 12 4 23 14 14 II 12 4 23 3 13 14 11 13 6 27 3 6 S 3 60 60 100 1:20 1:20 after 3 da. 1:20 1:20 1:20 1:30 I : 5 after 4 da. I : I only after 4 days 1:20 1:5 af- ter 48 hrs. 1:10 'At cavity. autopsy of guinea-pig 13 (date of 3d test) oldiomycetes were found in a nodule in the abdominal Cultures were negative. 2o6 Contributions to Medical Science negative. There does not seem to be any very constant relation between the number of injections, the interval between the time of the last injection and that of the test, and the results obtained; nor does the quantity of the living organisms introduced seem to be of paramount importance. About all that one seems justified in concluding from the above experiments is (i) that the serum of some normal guinea-pigs will cause a precipitate when mixed in proper proportions with oidiomycetic extract, and (2) that immunization with the extract, infection with living organisms, or combined infection with oidiomycetes and immunization with extracts results in the development of precipitating substances in the serum of about 70 per cent of the animals so treated (the percentages in the above experiments were: extract alone 79 per cent, organisms alone 50 per cent, organisms plus extract [small quantities were used] 67 per cent). These figures suggest also that Hving organisms plus small quantities of extract, too small to produce noticeable symptoms themselves, may be much more effective in calling forth antibodies than hving organisms alone and so rather tend to support the deductions made from the results of the experiment described on p. 204, according to which it was suggested that the extract may have some value as a curative agent. As a diagnostic procedure, with guinea-pigs at any rate, the precipitation reaction would have to be controlled by symp- tomatological findings; a negative reaction would mean nothing, a positive reaction would be suggestive but not conclusive because of the fact noted above that normal guinea-pig serum may pre- cipitate with the extract in dilution as high as i : 10, after 96 hours in the ice-chest, at least. According to Dr. Eggers' work, the results with rabbits are even more variable than with guinea-pigs. The goat seems to give a more constant reaction, but even in that case the highest dilution at which a precipitate formed was 1:25, which is cer- tainly not evidence of a very powerful serum. agglutinins. At the same time that the precipitating powers of the various sera were being tested, parallel experiments on their agglutinating Immunological Reactions of Oidiomycosis 207 Serum Emulsion 0.05 C.C. 0.5 C.C. O.I " O-S " O.IS " O-S " 0.2 " O.S " 0.25 " O-S " 0-3 " o-S " 0.3s " o-S " 0.4 " o-S " 0.4s " o-S " o-S " 0.5 " I.O " I . loop power were attempted. Emulsions of the whole organism in 0.85 per cent NaCl were used in these experiments. The serum dilutions were as follows: For methods (a) and (6) — see below — whole serum i part + emulsion i part ; 50 per cent serum i part + emulsion i part; 10 per cent serum i part + emulsion i part. For method (c) the dilutions were : NaCl 0.4s C.C. 0.4 " 0.35 " 0.3 " 0.2s " O. 2 " 0.15 " O.I " 0.05 " 0.0 " loop of moist organisms 0.0 " Three different methods were tried: (a) hanging drop in a hollow- ground slide, (b) quantities of i to 2 c.c. in watch glasses, which could be observed with ease both with the unaided eye and with the microscope, and (c) the materials were placed in narrow test- tubes with the idea that something might be learned from possible variations in the rate with which the oidiomycetes settled out of the suspension. In each experiment the following mixtures were made: (i) immune serum plus emulsion, (2) normal serum plus emulsion, (3) salt solution plus emulsion. These mixtures were incubated at 37° C. for from i to 12 hours and then allowed to stand at room temperature until observations were discontinued. The results were practically identical with the normal and immune sera in all the experiments. It was found that if the organisms were allowed to stand in contact with the serum, normal or immune, for a few hours, a beautiful agglutination reaction, as judged by the naked eye, occurred. This is especially noticeable in the watch-glass preparations. Under the microscope, however, the clumps which had attracted the attention resolved themselves not into masses of agglutinated oidiomycetes, but into groups of oidiomycetes which had developed h^phae, and it was to the appear- ance of this mycelial network that the apparent agglutination was due. 2o8 Contributions to Medical Science Regarding the rate at which the organisms settled out of sus- pension in the tubes, no differences between the normal and immune sera could be determined. The salt-solution tube sedi- mented first, the other tubes sedimented at about the same rate, and the organisms, like those used in the other methods, sprouted in a few hours. It is true that sometimes, especially with the hanging-drop and the watch-glass methods, there appeared to be a tendency for the organisms to gather into a few, large, loosely constructed groups, but this occurred with normal serum as well as immune, and almost as frequently with the salt solution alone, and the question arose whether it might not be the expression of an imper- fect trituration of the organism in the preparation of the emulsion. Certainly no specific agglutination of oidiomycetes by "immune" sera which did or did not yield a precipitate with the extract was observed. Possibly the experiment might be more successful if the organ- isms were first partially ground and if the fragments thus obtained were used; the spherules described within the conidia and the hyphae might also be used. LYTIC AND bactericidal SUBSTANCES. It had been observed in the agglutination experiments that oidiomycetes were able to develop hyphae when suspended in undiluted or diluted normal and immune serum, and that some- times organisms which had been subjected to the action of such fluids for three or four days grew when planted on agar. It was therefore argued that neither normal nor immune sera possessed lytic or bactericidal properties. To put the matter to a test a number of experiments, of the following nature, were performed. The sera of guinea-pigs ii and 55 (see Tables i and 2) were mixed with constant quantities of oidiomycetic emulsion in dilutions simi- lar to those used in method (c) of the agglutination tests at the same time that those tests were made. These mixtures were incubated for 30 minutes, placed in the ice-chest over night, and then allowed to stand at room temperature for three days. At the expiration of that time the entire contents of each tube was Immunological Reactions of Oidiomycosis 209 transferred to an agar slant, which was kept in a horizontal posi- tion for 30 minutes in order to allow the organisms to settle on the agar surface, and then placed in an upright position in the incubator with as little agitation as possible. Observation five days later (June 30, 1909) showed growth on the agar surface in all tubes, both above and beneath the surface of the hquid. The colonies growing beneath the surface were in general much smaller than those above the fluid, and gave the impression that they were growing much more slowly. On July 6 the following notes were made : Heavy growth on agar slope above fluid in all tubes except- ing tube containing o . 8 c.c. serum (guinea-pig 55) and tube contain- ing 0.2 c.c. serum (guinea-pig 11); no evidence of further growth beneath the surface of the fluid. In tubes containing o . 3 c.c, o . 6 c.c, 0.7 c.c, and i.i c.c. serum of guinea-pig 55 and that containing 0.5 c.c. serum of guinea-pig 11, practically all of the water of con- densation and introduced fluid has evaporated. In such cases a heavy growth extends to the bottom of the slope. July 8, no further growth beneath the surface of the fluid in any case; otherwise con- ditions are unchanged. Growth is abundant in all tubes excepting tubes 0.2 c.c. and 0.5 c.c. (guinea-pig 11) and tubes 0.4 c.c. and 0.8 c.c. (guinea-pig 55), in which the growths are but fair. The sera of guinea-pigs 50, 51, and 52, together with the sera of three normal pigs, were next tested in a manner similar to the above, with the difference that in these cases the tubes were not placed in the ice-chest, that control tubes of emulsion plus salt solution were run and the mixtures were incubated for 48 hours before being transferred to the agar slope. In these cases the results with normal and immune sera were about the same. In all tubes containing serum, growth was slight or did not occur at all; while in the control salt-solution tubes growth was abundant. It would seem from these experiments that normal and immune guinea-pig serum may have the power of impairing the vitality of oidiomycetes, as suggested by Gilkinet for beer yeasts when sub- jected to the action of rabbit serum. However, these experiments are regarded as merely of a preliminary nature, and it is hoped that it may be possible to test the above conclusion thoroughly at a later date. 2IO Contributions to Medical Science DEVIATION OF COMPLEMENT. It cannot as yet be stated positively whether or not normal or immune guinea-pig serum will deviate the complement accord- ing to the Bordet-Gengou technic. The few sera which have been tested, six recently immunized and one normal, have given absolutely negative results. The experiments have been carried on side by side with successful efforts to demonstrate a specific antibody in the serum of guinea-pigs immunized to a bacillus of the hog cholera group, so that in one way at least the work has been abundantly controlled. On the other hand, the positive results reported by other workers with various yeasts and molds make it still seem possible that the technic used and not the oidio- mycetes may be at fault. One thing which has been noticed is that O.I c.c. of a moderately turbid emulsion of oidiomycetes, or O.I c.c of a freshly made extract, or 0.15 c.c. of redissolved alcohohc precipitate of the extract, is usually sufficient to inactivate completely the complement in o.oi c.c. of guinea-pig serum which has stood in the ice-chest over night. This is, of course, but the expression of the property, common to all yeastlike forms, of absorbing or destroying complement. With this fact in mind, however, it is to be regretted that those who have reported posi- tive results with various mold fungi have neglected to give the full protocol of even one experiment. It might be of interest to compare the complement-binding power of various pathogenic and non-pathogenic yeast plants; possibly, by this means, some further idea of the factors which determine their power of resist- ance to destruction by the animal body might be gained. CUTANEOUS AND OPHTHALMIC REACTIONS. From the results of the skin and conjunctival tubercuhn tests as applied in man, it appeared of interest to experiment in an analogous way with the oidiomycetic extract on infected guinea-pigs. Cutaneous reaction. — The method of applying this test was as follows: A space on the abdomen about 3 cm. square was shaved and washed with HgClj (i : 1000) and alcohol (80 per cent). After drying, four small areas, forming the corners of a square of about I cm. in breadth, were scarified with the blade of a blunt knife Immunological Reactions of Oidiomycosis 211 which had been dipped into the following materials: for area No. I, the extract of organism BIH2; for area No. 2, the ex- tract of organism BIR; for area No. 3, 0.5 per cent carbolic acid. For area No. 4, the dry blade alone was used. 0.5 per cent car- bolic acid was applied because the extracts had been preserved in that strength of carbolic acid. In cases in which chloroform had been used as a preservative, a 0.3 per cent aqueous solution of chloroform was substituted for the phenol. In cases in which fresh extract alone was used, the carbolic acid and chloroform controls were omitted. Experiments on animals suffering from light, moderate, and severe infection resulted uniformly and negatively. It appeared to make no difference at what time with reference to the infection (early, middle, or late) the tests were appHed; the results never varied. Repeated scarification of the same areas after intervals of from 3 to 14 days called forth no reaction. Ophthalmic reaction. — The technic used in these experiments was as follows: To 10 c.c. of the extract was added an equal volume of absolute alcohol. The resulting precipitate was removed by filtration, washed in absolute alcohol, and dried in a partial vacuum over sulphuric acid. The dried material was redissolved in salt solution (o. 85 per cent NaCl). A drop of this solution was instilled into the. eye of the animal to be tested, the dose being repeated in the same and in the opposite eye after an interval of from 3 to 10 days. The results were always negative. The fresh extract and the extract preserved in phenol and in chloroform were also used, care being taken to control with corresponding aqueous solution of carbolic acid and chloroform, with similar negative results. Apparently then cutaneous and ophthalmic reactions against the extract do not occur in guinea-pigs infected with oidiomycosis. THE FATE OF THE OIDIOMYCETES WHEN INJECTED INTO THE PERITONEAL CAVITY OF THE GUINEA-PIG. Changes in the peritoneal fluids of recently injected animals. — The technic and results of this study are best illustrated by the report of the following typical experiment. 212 Contributions to Medical Science Guinea-pig 86 (normal) was given an intraperitoneal injection of 3 c.c. of a rather dense emulsion in salt solution (0.85 per cent) of a nine day agar culture of oidiomyces at 11:10 A.M., April 8, 1909. By means of a fine glass tube, specimens of peritoneal fluid were drawn off at approximately hourly inter\'als. Some of this material was placed upon a slide, a cover glass applied, and sealed with parafi&n. The specimen was then examined microscopically without further delay. Cover glass smears were made of other portions and stained with the Giemsa mixture. These preparations were examined at leisure. The guinea-pig was kept on his back for a few minutes before fluid was withdrawn, but was allowed to run free in his cage between times. 12:00 M.: Two specimens of clear, slightly viscid fluid were obtained. One, sUde I, was incubated for 30 minutes, the other, slide 2, was examined immediately. Shde 2: Quite a number of red cells; a very small number of leukocytes, and these mostl3' mononuclear; no oldiomycetes found; red cells probably due to trauma inci- dent to the insertion of the glass pipette, since this occasioned a slight hemorrhage. Slide I is similar to shde 2 in all respects. Stained smear. — Very few polymorphonuclear leukocytes, about in the same pro- portion to the number of red cells as in the blood; the vast majority of the leukocytes are mononuclear cells; some are clearly small lymphocytes; most of them are large mononuclear leukocytes, some approaching a distinctly endotheUal type as judged by the abundance of their cytoplasm. Many of the leukocytes appear to be disin- tegrating and losing their staining power. One organism is present. It is free. The cytoplasm takes an irregular bright blue stain; the capsule appears as a bright, sharply outlined, shiny rim, with merely the suggestion of a bluish tinge. 1 : 20 P.M. Slide 3: Still many red cells; many times more leukocytes, seemingly mostly mononuclears, a few organisms, some apparently free, most of them surrounded by from three to forty leukocj'tes; in some cases the organism has been ingested by a single leukocyte while the other leukocj'tes crowd around; in other cases ingestion has seemingly not occurred, though two or three layers of leukocytes inclose the organism. The latter appear to be unchanged. Stained smear. — Mononuclear cells still predominate, although the proportion of polymorphonuclear cells has increased. The leukocytes are frequently clumped, in some cases about organisms. The leukocytes in these clumps seem to be mostly mononuclears; a few polynuclears occur. 2:20 P.M. Slide 4: There are about the same number of red cells as in slide 3, but many more leukocytes. Leukocytes occur singly and in large and small groups, the tendency being toward the formation of grape-bunch-like masses of cells so large and compact that it is impossible to make out individual cells excepting at the peripher>'. Sometimes one can distinguish organisms down deep in these masses; frequently cannot make sure of the identity of anything there; no organisms found in very small masses or in single leukocytes. In one instance an oidiomycete can be seen within a single leukocyte, near the periphery of one of the large masses; other organisms are present within that mass, but one cannot tell whether they have actually been ingested or not. No organisms lying free. Exudate is more viscid than before, but does not seem to coagulate readily. The oidia are apparently unchanged. Stained smear. — The majority of the leukoc>'tes approach the transitional type; there are quite a number of polyiiuclears; cells grouped about organisms are mostly large mononuclear to transitional in type. 3:20 P.M. Slide 5: Exudate is very abundant and viscid; cloudy and granular. Immunological Reactions of Oidiomycosis 213 Microscopically very similar to slide 4, with the exception of the greatly increased number of leukocytes. Many of the organisms can be seen definitely phagocyted as described in 4. All organisms which can be well defined are found to be completely engulfed by one to three leukocytes, generally one, the rest of the leukocytes in the mass simply forming a many layered cp.psule about the organism and its phagocytes. Exudate does not clot in 30 minutes. Stained smear. — Large numbers of polynuclear cells, otherwise similar to 4. Cyto- plasm of cells grouped about oidiomycetes is fused into a single mass in some cases; in others the cells are simply connected by strands of cytoplasm, giving a vacuolated appearance to the region between the cells. 4:20 P.M. Slide 6: Exudate is abundant. It is still richer in leukocytes than before; there are quite a number of phagocyted organisms. The leukocytes do not seem to be bunched so strikingly as in previous slides; the oldia are inclosed within from I to 10 or a dozen leukocytes; the leukoc>'tes are more compact about the organisms than formerly, and instead of appearing like a bunch of grapes, that is, grouped rather loosely about phagocyted organisms, the leukocytic mass now has the appearance of segmenting frog eggs, all of the cells fitting closely to each other. None of the very large bunches of leukocytes are seen which were so striking in slides 3, 4, and 5. Exudate does not coagulate readily. Stained smear. — Polynuclears predominate largely, and seem to be the phago- cytes. 5:20 P.M. Slide 7: Large clumps of leukocytes about oidiomycetes do not appear. Frequently a single organism may be found within one leukocyte; almost never are more than five or six leukocytes concerned, excepting where a number of organisms occur together, in which case the whole mass may be fairly large, though the number of leukocytes per organism is apparently not what it was in earlier slides. The organisms appear to be unchanged. Stained smear. — Similar to 6. The nuclei of the leukocytes appear to be under- going a process of karyorrhexis. 6: 20 P.M. Slide 8: Similar to the above in the main. One important difference, i.e., leukocytes immediately surrounding organisms seem to be breaking up; they are losing their distinctness of outline and the granules are spreading out diffusely into the surrounding zones. Oidia are all budding, this condition seeming to be a little more marked than in earlier specimens of the exudate. Stained smear. — Similar to 7, more leukocytic changes. It is hard to determine whether the phagocytic cells are mononuclear or polymorphonuclear; it seems best to regard them as mononuclear, although doubtful. March 9, 1909. 9:30 a.m. SHde 9: Leukocytes grouped about organisms seem to be fragmenting. Leukocytes are scattered evenly over the field. Aggre- gations of 20-30 about an organism, especially a budding form, are still to be found; those leukocytes in the center of such masses seem to be more or less fused with each other. Oidiomycetes contain variable numbers, generally inconsiderable (10-20), of small, somewhat highly refractive, spherical bodies, arranged as a rule near the periphery of the cell within the capsule. Frequently they are scattered irregularly through the cell. Many of the leukocytes in diverse parts of the field contain very similar granules. Sometimes organisms appear to be full of such bodies. Stained smear. — In a number of instances it can be seen that the phago- cytosis is undoubtedly by large mononuclear cells. One group is especially noticeable; 214 Contributions to Medical Science it consists of a budding organism inclosed by four leukocytes. The leukocytes are arranged radially about the organism. The peripheral three-quarters of the cyto- plasm of each cell is independent and distinct, the inner fourth, however, has merged with a corresponding part of the other three to surround the oldium by a uniform cytoplasmic mass. The disposition of the cells reminds one a little of the structure of a four-leaf clover. The nuclei of the leukocytes stain purple; the cytoplasm, which is moderately vacuolated, a robin's-egg blue. The intracapsular portion of the organisms takes an irregularly distributed, deep to light blue stain; the capsule is clear, homogeneous, and of a very light blue color. 10:30 A.M. Slide 10: Cells, unstained, appear like large mononuclears for the most part. Groups of leukocytes are sometimes found arranged radially about an indistinct, somewhat homogeneous central mass in which it seems one can make out leukocytic shadows. The cells at the periphery of such masses are more or less cone- shaped, apex inward, nucleus at the base, peripheralward. The apex is drawn out into a narrow projection which loses itself in the central area. Budding oldia are found packed full of granules; then again non-budding forms occur which contain a very indistinct, faintly granular, slightly refractive material. Organisms are still engulfed by leukocytes, though it looks sometimes, especially in the case of the bud- ding, strongly granular forms, as though the leukocytes were beginning to break away from the organism and its immediate phagocyte. Again we find budding, markedly granular organisms in the center of such leukocj'tic masses as are described above. Granules may or may not be present in the buds of such organisms. Phagocyted oldia are observed in which the cytoplasm seems to have become homogeneous and is divided into two or three segments. Stained smear. — Sometimes the nucleus and cytoplasm of the leukoc>'tes are markedly vacuolated, and the cell appears to be disintegrating. In a few instances the intracapsular portion of the oidia stains much less deeply than usual while the immediate leukocytes are quite well preserved. Polymorphonuclear cells are much more numerous than the other varieties of leukocj'tes, but appear to take no part in the phagocytosis. Tjpical small Ij-mphocytes are very few in number. The large mononuclear cells with a great deal of cytoplasm form the great bulk of mononuclear leukocytes. 2:30 P.M. Slide 11: Essentially similar to 10. One organism is present which has an unusually thick capsule (2-3 times ordinary thickness). The outside of the capsule is smooth, the inside irregular. This organism has one small bud. Other oidiomycetes in the same mass appear normal. Stained smear. — Phagocytosis is not exclusively by mononuclear leukocytes; a few polymorphs may be included in the plasmodial masses. 5:00 P.M. SHde 12: Leukocytic groups occur about organisms as in the two preceding slides. Fusion of leukocytes and extrusion of granules — breaking up of leukocytes ? — is, if anything, more marked. As many as seven budding organisms are found in one leukocytic mass. Oidia seem to be budding and multiplying since we now find two, three, or four cells grouped together, sometimes in chains — mother cell, daughter cell ? — sometimes in pairs nearly always giving evidence of budding. One organism is found with two buds. Also find free organisms for the first time since the second hour. One of these is rather thick-walled and is packed with 25-30 spherical, homogeneous bodies. The spherules are not so numerous but that each can be dis- tinguished. Immunological Reactions of Oidiomycosis 215 Stained smear. — Similar to ii. Phagocytic cells seem to be breaking up. April 9, 1909. 8:20 A.M. Slide 13: Many free organisms. Also many large accumulations of leukocytes in the midst of which groups of organisms, two to six in a group, occur. Exudate is relatively scanty. Stained smear. — Similar to 12. Mononuclear cells seem to be relatively more numerous than in last few specimens. April 12, 1909. 11:30 A.M. Slide 14: (Pig seems somewhat asthenic. Tissues are swollen at abdominal wound. Wound is closed but bleeds readily. Small, pin- head nodules are palpable in right testicle. Left testicle is hardened; nodules similar to those in the right testicle are found.) Exudate is very scanty. One small drop obtained which shows very marked concentration of leukocytes about organ- isms, giving an appearance very similar to typical giant cells of the foreign body type. Stained smear. — Mononuclear cells only, big, endothelial-like cells for the most part. They occur in huge masses, surrounding numerous organisms. All leukocytes in these masses seem to be losing their staining power; cytoplasm and nucleus is vacuolated. This specimen was the last that could be obtained. Slide 14 was allowed to stand at room temperature. Examined the next day, it was found that long hyphae had grown out from many of the organisms which were inclosed in giant cells. The guinea-pig was etherized two days after the making of sHde 14. At autopsy the peritoneum, visceral and parietal, was dotted with grayish nodules from the size of minute points to nodules having a diameter of from 5 to 7 mm. The omentum was practically a mass of such nodules. The testicular surfaces also presented numerous grayish nodules, the largest being about i mm. in breadth. There was no excess of fluid in the peritoneal cavity. The nodules at first sight seemed to be retroperitoneal. Attempts to remove them showed that most could be peeled off without a great deal of difficulty, leaving a roughened opaque surface, while a few were considerably more adherent and, when removed, left a roughened opaque surface on which could be seen an occasional bleeding point. On cutting the nodules, some, es- pecially the larger ones, were found to have soft centers; smears of this material showed polymorphonuclear leukocytes and oidiomy- cetes. Microscopically the nodules fixed in Zenker's fluid, imbedded in celloidin and stained with hematoxylin and eosin, were found to consist of masses of fibrin and leukocytes. The leukocytes were mostly of the large mononuclear, endothelial-like type, although in some isolated spots polymorphonuclear leukocytes predominated. Oidiomycetes were present in large numbers, usually in the center 2i6 Contributions to Medical Science of the nodule, and inclosed usually within cells having from i to 20 oval, well stained, peripherally placed nuclei. As a rule the nuclei numbered from 3 to 8. The abundance of the cytoplasm of such cells was in direct proportion to the number of nuclei and took a marked uniform eosin stain with suggestion of a very finely granular structure. Sometimes the organisms were apparently free in the nodule; in such cases they formed the center of a small, circular, sharply circumscribed accumulation of polymorphonuclear leukocytes. There were no evidences of necrosis. The oidiomycetes them- selves presented a variety of appearances. The budding forms frequently occupied the center of the small abscesses just men- tioned; they presented a brightly shining, unstained, double- contoured external membrane, within which was an irregularly distributed, blue-stained, somewhat granular material which seemed, however, to form a lining layer just within the limiting membrane and not to be diffused to any great extent throughout the cell. In other organisms, the double-contoured membrane was unchanged and the material within the cell still took a blue stain; the latter, however, had seemingly shrunk away from the outer wall at all but one point, leaving a clear, unstained, new-moon shaped space. In still other organisms, this material had seem- ingly shrunk still more and was beginning to take an eosin stain; in others, it was beginning to lose its granular structure and to assume a homogeneous appearance, taking a sharp eosin stain. In such organisms as this, the central portion had sometimes become completely separated from the cell membrane so that the new-moon shaped space had grown into a complete circle; in other organisms, an equally pink-stained, homogeneous central portion was still in contact with the external membrane at one point, lead- ing one to imagine that possibly the complete separation just described might be more apparent than real owing to the plane in which the organism had been cut. Quite frequently no sign of "shrinking" was observed, the cells being a homogeneous pink without other modification. In those organisms which took the blue stain, it appeared as though the unstained portions within the capsule represented the homo- Immunological Reactions of Oidiomycosis 217 geneous spherules described in the fresh specimens while the stained parts represented what might be termed the interspheru- lar substance. A somewhat similar series of changes occurring in yeast intra vitam has been described by Potron.^^ The peritoneum extended intact beneath most of the nodules. The cells of the peritoneum, however, had assumed a marked cuboidal or rounded shape and seemed to have broken contact with each other. In a few instances a typical organization was proceeding from the sub-peritoneal tissues into the overlying oidiomycetes containing exudate. The kidneys were normal, as were the other organs. It will be remembered that some of the organisms in the peritoneal exudate of shde 14 grew. Oidiomycetes developed in pure cul- ture from material taken from nodules at autopsy. It does not appear to be necessary to report further experi- ments in detail. We shall summarize the course of events in the peritoneal cavity of a guinea-pig following the injection of an emulsion of living oidiomycetes in 0.85 per cent NaCl solution, basing our summary on numerous experiments such as the fore- going, on the autopsy findings of pigs dying or killed during the course of an experimental infection, and on observations made at exploratory operations, as follows: There is first a great transuda- tion of fluid poor in fibrinogen and leukocytes into the peritoneal cavity. The few leukocytes present are mostly large mononuclears, though lymphocytes and polymorphonuclear leukocytes occur. After one or two hours the leukocytes begin to accumulate rapidly, and at the end of three or four hours, the peritoneal fluid, at first thin and watery, becomes very cloudy and quite viscid. This cloudiness and viscidity gradually increase as the fluid becomes more and more scanty, until finally, after three or four days, no more fluid can be obtained. If one opens the abdom- inal cavity of a guinea-pig 24 hours after the injection of a large dose of oidiomycetes, he finds an acute, diffuse, suppurative, and fibrinous peritonitis. The peritoneum is covered with a thick layer of turbid, sticky fluid in which are occasional yellowish-gray clumps. These clumps are soft and may be easily lifted from the peritoneal surface; in fact, they can hardly be said to be adherent. 2i8 Contributions to Medical Science The microscopical appearance of this exudate has been described above. The clumps consist of accumulations of phagocytic cells and groups of cells — the "rosettes" of Skchi wan— supported by a scanty fibrinous mesh work. After four days very little fluid will be found ; the peritoneum may have a dull grayish tinge or may be about normal in appearance; it surely, however, will present numerous raised, convex, grayish nodules 0.5 to 5 mm. in diameter and sometimes whitish patches as much as i cm. in greatest extent, but these latter are not usual. As a rule, nodules and patches are easily detachable, leaving a dull, rough surface which sometimes may be slightly granular. They consist of masses of single large mononuclear leukocytes, some containing organisms, of polymorphonuclear leukocytes and masses of the phagocytic "rosettes," in a fibrinous meshwork. As the obser- vations are extended to include lesions at longer and longer intervals following the injection, it is found that the nodules, beginning about the fifth day, become more and more difficult to remove, and, when removed, become more and more prone to leave a bleeding surface. Some seem to sink down into the tissue upon which they at first rested; the peritoneum grows over them; they become less and less prominent, and finally disappear entirely in the course of 15 to 30 days. Other nodules — and this seems to be especially true of nodules in the testicles following the intra- peritoneal injection of o.i gm. of organism or more — commonly grow progressively larger for from 4 to 10 days, become somewhat soft and doughy, and then, in favorable cases, gradually harden, de- crease in size and disappear entirely within 40 days. In unfavorable cases the process of enlargement and softening continues until, as in guinea-pig 75 (see p. 205), the scrotum becomes greatly distended, the overlying skin becomes involved, and the lesion finds an external opening. Microscopically, in the first type, blood-vessels are ob- served to grow into the "rosette" masses from underlying tissues, connective tissue is laid down, the nodules become young connective- tissue growths, and presently, the "rosettes" formed from free, phagocytic macrophages are transformed into morphologically typical Langhans giant cells of inflammatory granulation tissue Immunological Reactions of Oidiomycosis 219 merely by this replacement of their supporting fibrinous reticulum by a meshwork of capillaries and their accompanying connective- tissue cells (see Fig. 5). In the case of very large nodules such as those which formed in the peritoneal cavities of guinea-pigs 13, 37, 38, and 56 (see p. 205) the connective-tissue growth may be so extensive as to cut off its own blood supply, thick-walled cysts being formed which contain a pus-Hke fluid. This material is not pus, however, but consists of cellular detritus, and, in the earlier stage, contains oidiomycetes showing various degrees of degeneration. In stained sections of this necrotic material from guinea-pig 13, autopsied 100 days after the last injection, one finds oidiomycetes which are practically normal in appearance, others have assumed appear- ances similar to those described in the nodules of guinea-pig 86, while others have gone still farther in that the cell membrane has either become thickened, homogeneous, eosin-stained, irregular in outline, or in other instances has disappeared leaving merely the inner homogeneous, eosin-stained spheres. Microscopical exami- nation of the unstained detritus reveals bodies very similar to organisms in appearance but which are soluble in ether. In specimens from other nodules of similar age, guinea-pig 56, for instance, autopsied 116 days after last injection, the ether-soluble bodies are the only things which occur that in any way resemble oidiomycetes. Whether or not they bear any direct relation to the organisms has not been determined. In the case of those nodules which become progressively larger and tend to soften, the bulk of the leukocytes consists of poly- morphonuclears. In such masses, organisms commonly occur free; frequently they remain inclosed in Langhans giant cells which thus appear in stained sections as eosin-stained islets in the mass of blue-lobed nuclei. Such nodules occur to some extent in the early stages of all cases, but in the later stages of fatal cases only. Mention has been made of nodules "in the testicles." This statement is not strictly correct since most nodules palpated ''in the testicles" are merely in the peritoneal and fibrous tunics. 220 Contributions to Medical Science Even in the most advanced cases the testicle itself may be and generally is unharmed, as in guinea-pig 75 (see p. 205) in which at autopsy the testicles were found normal. Besides the peritoneal lesions, macroscopic nodules have been found in the liver and in the lungs. In the liver they appear as white to grayish-red patches 1-5 mm. in cross-sectiorf, rather sharply demarkated. In the lungs they appear as translucent droplets 0.5-2 mm. in diameter, occurring both sub-pleurally and on the cut surface. The animals seem to have died before larger nodules had time to develop. Under the microscope in sections stained with hematoxylin and eosin, tj^pical liver nodules appear as follows: Section of the liver of guinea-pig 53 (for history, see p. 195; see also Plate i). Slight passive congestion; sUght central fatty infiltration; there is one area with a diameter of 4-6 liver lobules in which the parenchjTna cells seem to have been replaced by widely dilated, erythrocyte-filled sinuses, and oidiomjxetes inclosed in cells having from I to 20 nuclei. The multinucleated cells resemble closely foreign-body giant cells, excepting that the nuclei tend to be more rounded than those commonly seen in the latter cells, and, when oval in shape, their long axis tends to run parallel with the surface of the cell rather than centralward. This area is sharply marked off from the hver tissue by a narrow layer of parallel, wa\y, pink-stained fibers, peripheral to which is quite a marked infiltration of mononuclear cells, mostly of the hmphoid type, though many are cells with abundant cytoplasm. The neighboring parench>Tna cells are normal in most cases, although occasional islets of liver cells in which are found all stages of degeneration, even to that of complete necrosis, appear in the border tissue. In the lungs, the nodules occur principally in the neighborhood of the bronchi or larger blood-vessels. They consist of dense accumulations of round and epitheHoid cells with an organism at rare intervals; no necrosis; no edema; no exudation into alveoli. Aside from the nodules the lungs are normal. phagocytosis in vitro. The leukocytes. — Guinea-pigs were given intraperitoneal injec- tions of a thick suspension of sterile aleuronat in 0.85 per cent NaCl solution. From examination of the exudate at hourly intervals it was found that, after a lapse of from 12 to 15 hours, the exudate was very rich in leukocytes and was still sufficient in quantity to be easily obtained. Of the leukocytes, from 25-50 Immunological Reactions of Oidiomycosis 221 per cent were large mononuclears, the remainder being mostly polymorphonuclear neutrophiles and eosinophiles. Both the resh exudate and ''washed" leukocytes were used. The "washed" leukocytes were obtained by washing the fresh exudate first in sodium citrate solution and then in one or two changes of physio- logical salt solution according to the ordinary opsonic technic. The suspension was then made up to the original volume of the exudate by the addition of 0.85 per cent NaCl solution. The fresh exudate clotted readily, so that, in using it, rapid work was necessary. Exudate serum. — This was obtained by allowing the fresh exudate to clot spontaneously in test-tubes, centrifuging imme- diately, and removing the serum by means of a glass pipette. The o'idiomycetes were placed in a bottle of normal salt solution and shaken vigorously in a shaking machine for several hours before being used, in order to break up any clumps which might be present. An emulsion of the organisms in salt solution was then made which contained approximately 5,000 organisms per c.mm. In order to do this, counts were made of the organisms in the emulsion in the shaking bottle by means of the Thoma-Zeiss hemo- cytometer, and this emulsion was then diluted to the volume required as shown by the excess of organisms over the 5,000 per c.mm. which was desired. Method. — The observations were made on hanging-drop prep- arations. The various materials were mixed on clean cover glasses in tripHcate, in the dilutions called for by the protocol, mounted on hollow ground slides, and the cover-slip sealed to the sUdes with paraffin. The preparations were incubated at 37° C. and observed after 30 minutes, after 3 hours, and after 24 hours. No stain was employed. The first attempt was directed to determining (i) whether or not phagocytosis would occur in vitro, using {a) the fresh exudate and {b) washed leukocytes, and (2) the possible influence of guinea- pig serum upon the process. Several preliminary experiments of which the following is a typical example were performed : 222 Contributions to Medical Science TABLE 6. EXPERI»£ENX IN PHAGOCYTOSIS. SUde No. Organisms (Units of Emulsion) Fresh Exudate (Units) Washed Leu- kocytes (Units) Normal Blood Serum (Units) Exudate Serum (Units) NaCl (o.8s per cent) (Units) o I 2 3 4 s 6 7 8 9 10 I I o I o I o I I I I o o I o I o I o o o o 20 lO 8 o 6 I 4 2 o o o o o o I o o o o o o o I 2 O 4 O 6 8 lO The final examination of these slides after 24 hours at 37° C. disclosed the following conditions: Slide o: Well rtuxed; many red cells; leukocytes occur singly but also very frequently in groups of two to a dozen or more — most commonly three or four cells per group. Organisms may or may not, infrequently not, be found within these groups. The rtile appears to be for the leukocytes to pay no attention to the oidiomycetes. Very rarely a single leukocyte may be seen which has partially or totallj' engulfed an organism. Shde I : Organisms oc- curring singlj^ are seen only within leukocytes. Organ- isms occurring in clumps are always surrounded and, to be perceived by careful focusing, engulfed by leuko- cytes. Leukocytic clumps are found in the absence of organisms, but not to the extent observed in slides 2, 4, and 6 (see below). Many large leukocytes with abundant cytoplasm occur; they are the actively phagocytic cells. There are many red blood corpuscles, but ver>- few of them have been ingested. Slide 2: Leukocytes and organisms are quite numerous. Both types of cells, leukocytes and oidiomycetes, occur singly, by twos and threes, and in large groups, the organisms to a greater extent than the leukocytes. The cells are well intermixed and frequently approximate each other, but this is not marked enough to suggest phagocytosis or even positive chemiotaxis. Slide 3: Very similar to slide i, excepting that there is more extensive ingestion of red cells. In some cases leukocytes are merely grouped about organisms, in others there is not only grouping, but phagocytosis. Slide 4: Similar to slide 2, excepting that two leukocj^tes are present which appear to be phagocytic. Each has partly surrounded an elongated blastomycete. Slide 5 : Many red blood cells are present. A good many of them have been phagocj'ted. Organisms are present in fair numbers and occur for the most part Fig. I. — Types of phagocytic cells: a, erythrocytes; 6, organ- isms; c, cell nucleus. Immunological Reactions of Oidiomycosis 223 within single, or plasmodial masses of, leukocytes. It is noticeable that the leuko- cytes which take up the red blood corpuscles also take up the organisms. It is common to find a leukocyte which contains 2, 3, or even 4 red cells in addition to an or- ganism. Such large cells seem to be less numerous here than in shde 3. They assume bizarre shapes — greatly elongated, spindle forms, star-forms, and various other irregular outlines, with long, slender projections. This is especially noticeable about the edges where the drop is thin. Shde 6 : Same general appearance as slide 4. Shde 7: Fewer cells of all kinds; otherwise similar to slide 5. Slide 8: Cells of all kinds seem to be still fewer than in slide 7. Absolutely free organisms are quire rare; on the other hand, ingestion is not always com- plete and many of the groupings suggest that shown in Fig. 2. This appears to be due in some degree to a dearth of the big leukocytes, since where they occur ingestion is as complete, as a rule (there are ex- ceptions), as their bulk makes possible, as is suggested in the accompanying sketch. Red blood corpuscles have been phagocyted quite freely but by no means to the degree, relatively, that is evident in the case of the oldiomycetes. Slide 9 : Complete ingestion of single and budding organisms occurs with fair frequency; on the other hand, in Fig. 2. — Sketches of two strikingly similar oidiomycetic groups to illustrate the difference between (a) mere grouping of leukocytes, and (6) grouping with phagocytosis; (c) shows a leukocyte containing two organisms. These sketches also illustrate the type of ceU which is actively phagocytic. Fig. 6. many instances the leukocytes seem to be entirely indifferent to the oldia. In plas- modial masses which are in relation to organisms there is generally more or less com- 224 Contributions to Medical Science plete phagocytosis; contrarywise, places are not hard to find where mere grouping, with little or no phagocytosis, occurs. Slide id: Conditions here are very similar to those in shde o. If anything, phagocytosis may be slightly less marked, something of a hair-spitting distinction, however. From this experiment it appeared (i) that the leukocytes of the fresh exudate, either without the addition of normal serum or when mixed with excessive quantities of normal serum, do not phagocyte oidiomycetes as well as such leukocytes plus moderate quantities of serum; (2) that washed leukocytes do not phago- cyte oidiomycetes readily, when suspended in simple salt solu- tion, when suspended in salt solution plus normal serum, or salt solution plus exudate serum; (3) that phagocytosis is carried on in the fresh exudate in vitro by cells of the same type as those which are responsible for the major part of the work in the peritoneal cavity. An extensive series of further experiments was planned with the idea of testing thoroughly the above deductions. Unfortu- nately, the adoption of the mycelial mode of growth by the organ- ism interrupted the work while it was still far from being complete, so that the results reported below must be regarded as purely tentative. The summary of events follows: (i) It appears that the leuko- cytes of the undiluted fresh exudate are very actively phagocytic; complete phagocytosis is the rule within 30 minutes at 37° C; that is, as a rule, every organism in the preparation will have been ingested within that space of time. The leukocytes com- monly form a plasmodial mass about the oidiomycetes, the mass being surrounded by a narrow clear zone of fairly uniform width which separates it from the surrounding leukocytes. If the exudate is diluted by salt solution, normal serum, immune serum, or exudate serum, the degree of phagocytosis seems to decline, roughly, with the increase of the dilution. In other words, the more widely the organisms are separated from the leukocytes, the less apt is phagocytosis to occur. The organisms do not seem to be strongly chemiotactic in the sense that leukocytes will be attracted to them from a distance, the ingestion of the oidiomycetes by the leukocytes seems to be a sort of contact phagocytosis. The sera Immunological Reactions of Oidiomycosis 225 do not encourage phagocytosis to any greater extent than does physiological salt solution. (2) As was indicated in the earlier work, washed leukocytes are very inconstant with respect to their ability to phagocyte oidio- mycetes. The addition of the various sera mentioned above does not help matters. It was thought that perhaps the manipulation to which they were subjected injured the leukocytes in some way and an effort was made to obviate this as much as possible by decreasing the number of washings and the length of time in the centrifuge, by keeping the solutions at 37° C, etc., but without effect. Once in a while such leukocytes ingested organisms, but, in so far as could be determined, with neither rhyme nor reason. As these leukocytes were capable of ingesting carmine granules as well as an occasional organism it appeared that they had not lost their powers entirely; why, then, this failure to engulf oidiomy- cetes as actively as the leukocytes in the fresh exudate ? Several explanations were suggested. The technic may have been unsuited to the materials with which the work was being done. Savtchenko^'' found that the mononuclear cells of the peritoneal exudate of guinea-pigs have a tendency to clump and not to phagocyte after centrifugation in citrate solution. Brisco^^ has made a similar observation with respect to the alveolar cells of the lung. It was noted that the ingested carmine granules were much smaller than the oidiomycetes. Possibly the leukocytes had been injured sufficiently to deprive them of the ability to envelop such giants as the yeasts cells but still retained vitality enough to engulf smaller particles. Again, the fresh exudate was quite viscid and fibrin threads formed within it in a very short time after the mix- tures were made. Possibly the viscidity of the medium and the fibrin network were the factors which enabled the leukocytes to extend themselves over the organisms. It would be interesting, in this connection, to observe the effects of the addition of a solution of gelatin upon the phagocytic activity of the washed leukocytes. This might also be the key to the eft"ects of dilution on phagocytosis in the fresh exudate. (3) The proposition that phagocytosis is carried on in the fresh exudate in vitro by cells of the same type as those which are respon- 226 Contributions to Medical Science sible for the major part of the work in the peritoneal cav-ity is borne out by further observations. This does not mean that, either within the peritoneal cavity, or out of it, all the work is done by the macrophages. Such is not the case; polymorphonuclear leukocytes frequently take part in the work, and sometimes are the only leukocytes present in the plasmodial masses. They, of course, are too small to ingest any but the smallest oidiomycetes. Nevertheless, they are active in surrounding and ''hemming in" the organisms. On the other hand, there is no room for doubt that the bulk of the work is done by the macrophages. In three experiments, normal and immune serum and sodium citrate solution seemed actually to inhibit phagocytosis in fresh exudate, while the leukocytes in the salt-solution dilution of the same exudate were actively phagocytic. No explanations of these results have suggested themselves. Fresh normal rabbit serum was mixed with guinea-pig peritoneal exudate. The leukocytes in the exudate became spherical in shape, and were strongly agglutinated within 30 minutes at 37° C. The observations were not carried further. Needless to say, no phago- cytosis occurred in such preparations. These experiments seem to indicate that opsonins are, at least, not of great importance in the phagocytosis of the oidiomycetes used in this work by the leukocytes of the peritoneal exudates of normal and immune guinea-pigs. SUMMARY AND CONCLUSIONS. Oidiomycosis in the guinea-pig, following intraperitoneal inoculation, is characterized, in fatal cases in male animals, by a gradually developing cachexia accompanied as a rule by a steady loss of weight which is especially marked during the last three or four days of life and by the development, in pigs of 400 gm. or over, of palpable nodules in the testicles in from three to seven days after inoculation. In males weighing 400 gm. or less the nodules in the testicles are inconstant, and, when present, may be difficult to palpate owing to the fact that they occur frequently about the upper pole of the testicle and about the neck of the scrotal sac — points which may be difficult to feel in small animals. Immunological Reactions of Oidiomycosis 227 The appearance of palpable nodules in the skin or anterior abdomi- nal wall has been noted but may be avoided in a great measure by rinsing the outside of the needle before making the injection. In female guinea-pigs symptoms are usually entirely lacking; they survive intraperitoneal doses which kill the males, a point which should be borne in mind when testing the pathogenicity of yeasts for guinea-pigs. The scrotal sac of the male pig seems to afford a locus minoris resistentiae to which the females have no counterpart. The reasons for the lower resistance of this par- ticular corner of the peritoneal-lined cavity are not apparent. Postmortem examination of animals dying of the disease reveals multiple grayish nodules from o.i to 10 mm. in diameter on all peritoneal surfaces. All such nodules contain oidiomycetes, as may be demonstrated in microscopic sections and in cultures. Most nodules have softened centers, especially the large nodules which are found in the testicles, in which oidiomycetes and great numbers of polymorphonuclear leukocytes occur. The lungs and liver may also present small areas of cellular infiltration about oidiomycetes. There are no constant changes in the various organs. Recovery from an infection is accompanied by a low grade of im- munity which manifests itself in a somewhat more speedy recovery from subsequent infections, and by the development of a sUght tem- perature for a few days immediately following reinoculations. Repeated injections of an "extract" of oidiomycetes lead to the development of an immunity in guinea-pigs which is character- ized mainly by the more rapid walling off of organisms injected into the peritoneal cavity, and by a more rapid disappearance of the lesions which appear in the testicles. The sera of guinea-pigs immunized by repeated intraperitoneal injections of living organisms or of an extract of organisms develop precipitating substances against the '"extract" in from 50 to 79 per cent of the cases, which may be manifest in a dilution as high as 1 : 20. Normal sera occasionally precipitate with the "extract" in lower dilutions. Substances which will agglutinate oidiomy- cetes suspended in dilutions of the sera are not formed. Specific amboceptors also do not appear to be developed. Opsonins have not been demonstrated. Prolonged exposure of oidiomycetes to 2 28 Contributions to Medical Science immune or normal serum at 37° C. seems to impair the vitality of the organisms — judging by their diminished power of growth when transferred to suitable media — as compared with that of similarly tested control tubes containing organisms suspended in physiologi- cal salt solution. The reaction of anaphylaxis may, under suit- able conditions, be obtained with the "extract" in guinea-pigs. The "extract" is toxic; prolonged administration by the intra- peritoneal route results in cachexia and death usually by second- ary infection; judiciously used, it may have a favorable effect upon the course of an oidiomycetic infection. Oidiomycetes injected into the peritoneal cavity of guinea-pigs are rapidly taken by leukocytes, macrophages principally, which form plasmodial masses and become attached to the peritoneal surface at first by fibrinous, later by fibrous, adhesions, the mass of cells grouped immediately about the organisms presently assum- ing the structure of typical Langhans giant cells. In cases ending in recovery, the organisms may grow for a time, but eventually degenerate and disappear; the nodule within which they were, being absorbed with them. In fatal cases the nodules increase in size and become soft; many of the organisms degenerate, but others multiply; the inclosing giant cell disintegrates and there ensues an infiltration of polymorphonuclear leukocytes pari passu with the enlargement and softening of the nodule. Sometimes the organisms are surrounded by polymorphonuclear cells from the beginning; such a mass may in turn be inclosed by layers of macro- phages. Just what such a condition may mean as regards the prognosis of the lesion has not been determined. The organisms do not tend to penetrate the peritoneal surface. It seems probable that they leave the peritoneal cavity, only in case there has been actual tearing of the lining tissues. Oidio- mycetic nodules were described in the liver and lungs of a few animals. The belief seems to be justified that such generalization of the infection followed the introduction of the organisms into the circulation by mechanical means, because the condition occurred almost solely (there were two exceptions) in animals in whose peritoneum one could detect tears, as in the animals used for the study of phagocytosis in vivo. Immunological Reactions of Oidiomycosis 229 The mode of defense of guinea-pigs against oidiomycetes in- jected into the peritoneal cavity appears to consist, firstly, in phagocytosis and intracellular digestion; secondly, in a walling off and encapsulation of the phagocyted organisms by connective tissue; and, thirdly, upon a somewhat ill-defined, and possibly questionable, unfavorable influence of the serum upon the vitality of the organism. These agencies act but slowly; oidiomycetes may retain their power to grow on artificial media for days, and their characteristic staining properties for weeks, in the inflam- matory nodules of supposedly immune pigs. Specific antibodies are but poorly developed. REFERENCES. 1. Metschnikoff. Virchow's Archiv, 1884, 96, p. 177. 2. RiBBERT. Der Untergang pathogener Schimmelpilze im Korper, Bonn, 1887. 3. Saxfelice. Cenlralbl. fur BakL, Abt. i, Orig., 1896, 20, p. 219. 4. Charrin and Ostrowsky. Comptes rendus Soc. de Biol., 1896, 48, p. 743. S- Roger. Comptes rendus Soc. de Biol., 1896, 48, p. 728. 6. ScHATTENTROH. Arcli. fHf Hyg., 1896, 27, p. 234. 7. GiLKiNET. Arch, de med. e.vper., 1897, 9, p. 881. 8. JONA. Cenlralbl. fUr Bakt., Abt. i, Orig., 1897, 21, p. 147. 9. Obici. Zeiglers Beitrdge, 1898, 23, p. 197. 10. Skchiwan. Anti. de I'Inst. Pasteur, 1899, 13, p. 770. 11. Ricketts. Jour. Med. Res., 1901, 6, p. 377. 12. Malvoz. Ann. de la soc. med.-chir. de Liege, 1901, quoted from Sanfelice (15). 13. Malvoz. Cenlralbl. fiir Bakt., Abt. i, Orig., 1901, 29, p. 668. 14. Savtchenko. Ann. de I'Insl. Pasteur, 1902, 16, p. 106. 15. Sanfelice. Cenlralbl. fiir Bakt., Abt. i, Orig., 1902, 32, p. 360. 16. Wlaeff. Compt. rendus Soc. de Biol., 1902, 54, p. 412. 17. PoTRON. Blastomycetes dans les tissues. These, Nancy, 1903. 18. Fabozzi. Cenlralbl. fiir Bakt., Abt. i, Ref., 1906, 39 (Abstract), p. 282, 19. Spiethoff. Jahrbiicher der Hamb. Staatskrankenanstalten, 1903-4, 9, Heft II, p. 167. 20. Bowen and Wolbach. Jotir. Med. Res., 1906, 15, p. 167. 21. Christensen and Hektoen. Jour. Amer. Med. Assoc, 1906, 47, p. 247. 22. Hamburger. Jour. Infect. Dis., 1907, 4, p. 201. 23. Marco del Pont. Bull, de I'Inst. Pasteur, 1907, 5, p. 46. 24. Brisco. Jour. Path, and Bad., 1907, 12, p. 66. 25. Brown. Jour. Amer. Med. Assoc, 1907, 48, p. 743. 26. Hektoen. Jour. Amer. Med. Assoc, 1907, 49, p. 1071. 27. Montgomery and Ormsby. Archives Int. Med., 1908, 2, p. i. 28. Rothe. Deut. med. Wchnschr., 1910, 36, p. 30. 29. Widal, Abrami, Joltrain, Brissaud, et Weill. Ann. de I'Inst. Pasteur, 1910, 24, p. I. 230 Contributions to Medical Science EXPLANATION OF PLATE i. Fig. I (at top). — Inclusion of oidiomycetes in uninuclear and multinuclear cells in liver (guinea-pig 53 — see p. 195). Fig. 2 (in center). — Leukocytic rosette from peritoneal exudate (guinea-pig 86 — see p. 212). Fig. 3 (below). — Nodule from omentum of guinea-pig showing transformation of rosettes into Langhans giant cells; degenerative change in oidiomycetes. PLATE 1. m \^ ^ 'I a * e "^ » "^ * ^ c . 9 ■%2-V'^.';,N^ LYMPHATOTOXIC SERUM: NOTES ON ITS CONSTI- TUTION; PRELIMINARY EXPERIMENTS BEARING ON ITS INFLUENCE ON EXPERIMENTAL INFECTIONS/ H. T. RrcKETTS. (From the Pathological Laboratory of the University of Chicago.) In this communication no attempt will be made to summarize completely the literature of "leuko toxic" or " lymphato toxic " serum. This was done in certain aspects in a recent paper by Flexner.^ A few important data may be given, however. A leukotoxic serum was first described by Metschnikofif in 1899. Since that time he and his school have pubHshed many papers on the subject, and the phenomenon has been used to support the phagocytic theory of natural immunity. Besredka in particular has contributed important work. His findings, which have, in the main, been supported by others, are briefly as follows: A leukotoxic serum may be obtained by the immunization of one animal with the lymphatic tissues of another, using lymph glands, spleen, bone-marrow, or pure preparations of leukocytes. The toxicity of the serum depends on the number of injections made and on the animals used; the guinea-pig yields a stronger serum when treated with rabbit tissue than vice versa. Leukotoxic sera are for the most part specific for the cells of the animal used in the immunization, but this is not without excep- tions. All leukotoxic sera are more or less hemolytic. As to toxicity, it is possible to obtain a serum from a rabbit which will kill a guinea-pig in a dose of 0.5 c.c. in a few hours. He also observed that intraperitoneal injections cause a marked desquama- tion of the endothelium, and micro-organisms may then the more easily wander out from the intestines into the peritoneal cavity. ' This work was begun in the laboratory of M. Metschnikofif, who suggested the line of work. The writer's thanks are due to him and his associates for very numerous kindnesses; also to Mr. Dick at the University for assistance in leukocyte counts. From Trans. Chic. Path. Soc, 1902, 5, p. 178. » "The Pathology of Lymphotoxic and Myelotoxic Intoxication," Univ. of Penn. Bulletin, Novem- ber, 1902. 231 232 Contributions to Medical Science When an intraperitoneal injection of the serum is made there follows rapidly a great decrease in the number of leukocytes in the exudate; however, an extreme hyperleukocytosis supervenes, which may be ten times in excess of the normal, and which lasts for a number of days. Further injections cause a further increase of leukocytes, but without the primary leukopenia. Besredka states that normal rabbit serum produces no primary hypoleu- kocytosis. This I have not been able to substantiate. He also found that small doses of the serum given subcutaneously will produce a mild general hyperleukocytosis. Normal rabbit serum had, however, the same effect. Antileukotoxic serum prevented the solution of the leukocytes by the leuko toxic serum. The agglutinating value of the serum for the homologous leukocytes is also a recognized phenomenon. Both the fresh and the heated serum will agglutinate. Leukotoxic sera are said to be less stable than hemolytic sera. Isaeff found that with the resultant hyperleukocytosis the resist- ance of animals to experimental infections was increased. Other works concerning the relation of "phagolysis" to immu- nity will not be entered on here. The serum used in the experiments to be considered was pre- pared by injecting suspensions of the mesenteric lymph glands of guinea-pigs into rabbits. The glands of two animals were used in the first injection, and of three each for the second and third injections. The suspensions, made by grinding the tissue with sterile sand, were injected seven days apart, and either subcu- taneously or into the peritoneal cavity; the latter method gives a somewhat stronger serum. Blood was taken from the rabbits to obtain the serum from seven to ten days after the last injection. The serum was allowed to collect spontaneously at a somewhat lowered temperature. Following the laws of the formation of antibodies, one would expect to obtain in such a serum antibodies for each type of cell or tissue constituent injected. The large mass of cells in lymph glands are, of course, the lymphoid cells, the lymphocytes. There are in addition many endotheUal cells and a small amount of circulating blood. In accordance with these expectations, a Lymphatotoxic Serum 233 lymphotoxic (or, more properly, a leukotoxic), an endotheliotoxic, and a hemotoxic or hemolytic antibody are formed. In addition to these, an agglutinating and a precipitating substance seem to be formed. THE LEUKOTOXIN. The leukolytic property is observed when the fresh or reactivated serum is brought in contact with the leukocytes existing normally in the peritoneal or pleural exudate of the guinea-pig, or when mixed with a suspension of triturated guinea-pig lymph gland. (Regardless of the embryologic origin of the "endothelium" of these surfaces it seems probable that they have certain recep- tors identical with those of the lymphatic endothelium.) So far I have obtained only a moderately powerful leukolytic serum, judging from experiments made in vitro. It is possible that the actual toxicity of the serum may be greater than such experiments indicate. One part of serum to ten parts of exudate would often cause complete solution of the bodies of the leukocytes, the nuclei and granules showing much greater resistance; commonly one to three or four gave more positive results. The leukocytes when first drawn are actively motile and the surfaces are rough with many small pseudopodia. Kept at the proper temperature, this may keep up for several hours. Almost immediately, however, when mixed with leukotoxic serum pseudopodia cease to form, and the surface of the cell becomes perfectly smooth in a few minutes. The cell next enlarges, the protoplasm becomes clear, the nucleus shows more distinctly, and the cell appears to be in- closed by a thin clear membrane. The swelling continues, one portion of the surface may balloon out, the nucleus often becomes eccentric, and the protoplasmic granules group themselves about the nucleus. Solution may not extend beyond this point, or if there be a sufficient amount of antibody and complement the ceil membrane becomes dissolved and the granules are set free. While the granules are not unaffected they have about the same resist- ance to solution as the nucleus. All leukocytes are not attacked equally; the first to show changes are the large finely granular cells, the macrophages; the polymorphonuclears are attacked at about the same time. The most resistant are the small mononuclear 234 Contributions to Medical Science cells, the microphages, which contain coarser granules than the macrophages. With only a moderately toxic serum the micro- phages may become swollen as much as thirty minutes to an hour later than the macrophages or the polynuclears. A particular experiment showed that freezing for one and one- half to two hours did not destroy the lytic property of the serum for leukocytes; a temperature of 55° C. for thirty minutes does, and the serum can be only partially reactivated by the use of fresh serum. THE ENDOTHELIOTOXIN. This action is manifested in intraperitoneal and intrapleural injections of the fresh serum by the production of a marked des- quamation of the endothelium. Normal rabbit serum produced no such desquamation or a trifling amount. The cells are seen singly and in shreds. Furthermore, changes in shape occur after the desquamation of the cells. A third point in evidence is the formation of a marked hematoma at the point of a subcutaneous injection. The peritoneal desquamation occurs very rapidly, within fifteen minutes following the injection, and progresses for a half to one hour. From this time on the number found decreases steadily. The reason is not clear, but possibly depends on the agglutination of the cells and their adhering to the parietes. After desquamation the changes that occur in the cells are somewhat insignificant. They do not become spherical, but are somewhat swollen laterally, and irregularities in the surface are smoothed out. In the unaltered cells, as they appear when first desquamated, the borders are irregularly rectangular and the edges refract the light brilliantly; this is probably due in part to adhering cement substance. After being subjected, in vitro, for about fifteen minutes to the action of the immune serum these refracting edges become less distinct and eventually few of them can be seen. This is seemingly due to a solution of the adhering cement sub- stance as well as to alteration in the shape of the cell. The effect on the cement substance may also play a part in the desquama- tion. It is thus seen that the visible effects on the endothelial cells are much less than on the leukocytes. The hematoma which follows subcutaneous injection develops Lymphatotoxic Serum 235 in the course of one to two hours, i.e., the fluid at that time is found to be hemorrhagic. The amount of blood increases during the first twenty-four hours, and it is diluted with a large amount of lymph. Many leukocytes are found in the fluid in incomplete agglutination; a few of these show moderate lytic changes. The tumor subsides in the course of a week or ten days. Histologically (tissue removed on the second day) there is an enormous edema and vast fields of leukocytes, particularly polymorphonuclear. The fibrous tissue cells are in active proliferation and show mitoses, and the collagenous and muscular tissues are degenerated. The endothehal cells of blood-vessels and lymph spaces have increased, and are often greatly swollen, projecting prominently into the lumen. It was not possible to demonstrate vascular rhexis histo- logically. The overlying skin becomes necrotic and can easily be wiped off. THE HEMOLYSIN. The hemolytic value of the immune serum is not greatly above that of normal rabbit serum for guinea-pig corpuscles; in a care- fully conducted experiment, 0.05 c.c. of the former completely dissolved a given quantity (0.4 c.c.) of a 5 per cent suspension of guinea-pig corpuscles in 0.85 per cent salt solution, while o.i c.c. of the latter was required. THE LEUKO-AGGLUTININ. The agglutination appears in a few minutes after the mixture is made in vitro, and also occurs with equal rapidity in intraperi- toneal injections. That the agglutinating property is distinct from the lytic is shown by the fact that when deprived of the complement by heating, or as a result of standing, agglutination will still occur. Normal rabbit serum has slight agglutinating and lytic properties for guinea-pig leukocytes, but much less than that obtained by immunization. THE PRECIPITIN. Precipitation is observed when one makes the mixture used for studying leukolysis. A control of normal rabbit serum and guinea- pig peritoneal exudate shows a barely perceptible amount of the 236 Contributions to Medical Science precipitate. Bouillon and 0.85 per cent salt solution with the exudate show no precipitate. It is then evident that we have to deal with a complex serum, one not purely leukotoxic or lymphotoxic, and it is probable that this must be taken into consideration in drawing conclusions concerning the effect of the serum on experimental infections. The course of peritoneal leukocytosis following intraperitoneal injections of the serum has been observed and certain findings of pre- vious workers corroborated (especially Metschnikojff and his school). Controls were made with normal rabbit serum and bouillon. It is a generally known fact that many substances when introduced into the peritoneal cavity cause an influx of leukocytes. With these substances there is usually a primary leukopenia, hyper- leukocytosis supervening in from one to two hours and increasing for about twenty-four hours. It then gradually decreases during the course of several days. The estimation of the primary leuko- penia is subject to no small error, owing to the dilution of the exudate by the fluid injected. Since there appears to be normally not more than 0.5 c.c. of fluid in the abdominal cavity of a guinea- pig, it is clear that the injection of 3 c.c. of fluid would produce in itself an apparent leukopenia, and it is even possible that this may constitute the entire leukopenia described. Another error is caused by the agglutination of the leukocytes on the peritoneal walls when a foreign substance is injected. Concerning the even- tual hyperleukocytosis there can, however, be no question. Often it was not possible to follow out the counts more than twenty-four hours when the leukotoxic serum was injected, on account of a marked viscosity which developed in the exudate. It is to be noted that with all three solutions there followed a primary leukopenia which was in part, at least, apparent for the reason mentioned. The hyperleukocytosis was usually most marked following the leukotoxic serum, but both bouillon and normal rabbit serum often gave a vast increase. In the stage of hyperleukocytosis the polymorphonuclears are in excess of all other cells, while in the natural condition they are very limited in number. It will be remembered that the same cell constitutes the mass of white cells found in the subcutaneous exudate also. It was mentioned that Lymphatotoxic Serum 237 the peritoneal exudate soon becomes thick and viscous. Pieral- lini attributes this condition to fibrin formation due to the disinte- gration of the leukocytes (phagolysis) , basing his conclusion on the positive reaction to Weigert's fibrin stain. The following is an example of the course of leukocytosis in two cases following an intraperitoneal injection of leukotoxic serum, with a control of normal rabbit serum: Leukotoxic serum: A. Before Inj. 14,400 Extreme Leukopenia 266 Extreme Hyperleuk. 282,700 B. 9,000 1,532 662,000 Normal senmi: A. B. 14,400 17,200 1,200 1,266 72,000 68,000 The serum used in these experiments was fairly powerful. This probably accounts for the extreme hyperleukocytosis. More recent sera have provoked lower grades of hyperleukocytosis. That the secondary hyperleukocytosis is not a specific result of the injection of leukotoxic serum is shown in two cases where two c.c. of peptone bouillon were injected. Before Inj. E.xtreme Leukopenia Extreme Hyperleuk A. 59,000 2,400 267,000 B. 34,075 3,200 250,000 Here the leukopenia seems real. The hyperleukocytosis is extreme, and in this connection it is perhaps of significance that Delezenne observed that a solution of Witte's peptone produced a marked degree of leukocytosis in vitro. It seems probable that, within certain limits, the greater the toxicity or leukolytic power of the substance injected the greater the resultant hyperleukocy- tosis; this, however, needs further corroboration. The course of peritoneal and vascular leukocytosis has been followed after the subcutaneous injection of 2 c.c. of leukotoxic serum. There was in neither fluid a primary leukopenia, but almost immediately a moderate rise in each. In twelve hours the number began to decrease almost uniformly in the two fluids; as this decrease is synchronous with the pronounced development of the subcutaneous hema-leukocytoma it may be supposed that a vast number of leukocytes was abstracted from the blood and 238 Contributions to Medical Science lymph to swell those at the point of injection. After twenty-four hours this began to be compensated for, a rise occurring in both the blood and peritoneal exudate. One may consider this an illus- tration of excessive repair after an injury (Weigert). Now follows a brief consideration of the relations of these artificially induced h>'po- and h}^er-leukocytoses to experimental infections. The series was begun with the cholera and typhoid organisms, but owing to the great number of animals required the cholera vibrio was abandoned after some preliminary experiments which showed that it conducted itself similarly to the typhoid bacillus. A constant method of growing the organisms was necessary to obtain uniformity in doses and virulence. For a time uniform agar surfaces were employed, but this was later abandoned for a constant amount of peptone bouillon. Twenty-four-hour cul- tures were used. The work falls naturally into two aspects: First, in the primary peritoneal leukopenia, is the resistance of the animal to peritoneal infections increased? Second, during the stage of hyperleukocy- tosis is the resistance increased? For the first the procedure was as follows: The serum was injected at a given time and at the height of leukopenia a sublethal dose of the bacterium followed. Three controls were made in typical experiments, one to insure that the dose of culture was not fatal, a second to show that the serum would not kill the animal, and a third with the same amount of normal rabbit serum and culture used in the experiment animal. After many repetitions the following result was finally reached: Leukotoxic immune serum (3 c.c. of the heated serum) decreases the natural immunity of the guinea-pig for typhoid and cholera, when the culture is injected one hour after the serum, so that one-half the M.L.D. is fatal, both injections being made into the peritoneal cavity. It developed that normal rabbit serum had a similar but less powerful effect when heated, but when the fresh serum was used the ani- mals would withstand much more than the M.L.D. This is attributable to the marked bactericidal property of fresh rabbit serum for the typhoid bacillus. The exact details of the experi- Lymphatotoxic Serum 239 ments will be given later, when other work in the same direction is completed. In dealing with the second part of the question, animals were injected with a given dose of the serum, and twenty-four hours later the resistance tested as before. This procedure resulted in immunity against about twice the M.L.D. of the typhoid bacillus. (The cholera organism was abandoned at this point.) All requisite controls were likewise used in this series. It was also brought out that normal rabbit serum had the same protective value as the leukotoxic serum used in this series. When one considers the difference in the degrees of leukocytosis produced by the two it becomes apparent that the normal serum had even greater pro- tective value than the leukotoxic serum. That is to say, 75,000 leukocytes in the former case would be able to save the animal from as great a dose of bacteria as 300,000 or even 600,000 in the latter (assuming, of course, that the leukocytes are really the protective agents). This may be explained by the supposition that the leukocytes called out by the immune serum are still in a debilitated state as the result of the lytic effect of the serum, and consequently had a lower phagocytic index. Some preliminary work has shown that peptone bouillon has similar effects to leukolytic and normal rabbit serum in the respects under consideration, although to a less degree. From these experiments it seems probable that many substances may be found which by a deleterious influence on the leukocytes (phagolysis, or otherwise engaging them) are able to decrease the resistance of an animal to certain bacterial infections at definite points of inoculation; or by varying the mode of experimentation, the same substances may be the means of increasing the resistance to the same infections, the result depending in this case on an increase in the number of leukocytes, or of complement (Wasser- mann), or perhaps both, at the point of injection. The participation of the endothelium in these processes is undetermined but seems probable where the resistance is decreased as indicated. PRELIMINARY REPORT ON THE ACTION OF NEURO- TOXIC SERUM/ H. T. RiCKETTS AND T. ROTHSTEIN. (From the Pathological Laboratory of the University of Chicago.) We shall not discuss the literature of neurotoxic serum in this communication further than to mention that Delezenne, Centanni, and Enriquez and Sicard have produced such sera by immunization. Brief re\'iews of these works are made by Hektoen in Progressive Medicine, March, 1902, pp. 283, 284, where the original references are given. We have immunized rabbits with nervous tissue of guinea-pigs, and in three series; using in the first the cerebrum, in the second the cerebellum, and in the third the spinal cord. This has been satisfactory so far as the strength of the resulting sera is concerned, although Delezenne found that injections of guinea-pigs with nervous tissue of rabbits did not yield a strong serum. He attrib- uted this to the close relationship of the two animals. However, the fact that rabbit's serum is normally more toxic to the corpuscles of the guinea-pig than the serum of the latter animal for rabbit's corpuscles may suggest an explanation of our better success in following the opposite procedure of that of Delezenne. Large amounts of tissue were used in immunization, and usually the serum was collected ten days after the third injection. The hemolytic power of the serum for erythrocytes of guinea- pigs was about ten times as great as normal rabbit serum (for the lower limit), in one experiment. This increase in hemolytic amboceptors is, no doubt, due in part to blood included in the injected mass, but it is also possible that proper receptors existed to some extent in the brain cells and cerebral fluid. The quantities of neurotoxic serum injected into guinea-pigs did not cause any marked hemolysis, as distinct hemoglobinuria was not observed in fatal cases. The finding of pigmented cells in the spleens of such animals indicates, however, that hemolysis in vivo does occur. When small amounts of the immune serum (0.05-0.1 c.c.) » Trans. Chic. Path. Soc, 1903, $, p. 207. 240 The Action of Neurotoxic Serum 241 are added to 0.5 c.c. of guinea-pig serum a distinct precipitate is produced. A serum sLx weeks old and not recomplemented gives the reaction. Normal rabbit serum has this same power, but to a very limited degree. Therefore a precipitin is formed. In the injected animals hemorrhages occur commonly in the meninges and lungs, and in one animal they were found in the muscles, hence the active serum is probably also endothelio toxic. Intraperitoneal injections cause a marked desquamation of endothe- lium. It has been impossible, so far, to detect solution of nervous tissue in test-tube experiments, or in microscopic preparations. But it seems probable that the value of the serum in antibodies may be estimated approximately by the precipitation of an emulsion or an extract of nervous tissue. A cerebral hemisphere of a guinea- pig was thoroughly ground with sand, intimately emulsified in 20 c.c. of distilled water, and passed through a fine wire cloth. Half of this was filtered through two layers of ordinary filter paper, and the fluids used in precipitation experiments. The filtered portion, even after being passed through the paper several times, was uniformly cloudy-opalescent ; microscopic examination showed the presence of innumerable fine granules. So it was not purely an extract, but an extract plus minute portions of nervous tissue. From one to ten drops of an old serum were added to small tubes, each containing o . 5 c.c. of one of the fluids. It was found that two to four drops of either the anticerebral or the anticerebellar serum would completely clear the unfiltered emulsion, causing a dense, coherent precipitate. The same result occurred with the filtered portion, naturally with a smaller amount of precipitate, and for some reason the overlying fluid was not so thoroughly cleared as in the unfiltered portion. This experiment indicates that anti- cerebral and anticerebellar sera are not specific for the homolo- gous parts of the nervous system. That this test may be an indication of the amount of antibodies present is shown by the fact that a serum taken after two injections produced a consider- ably smaller precipitate in the same solutions, and this serum was also less toxic when injected. It must be mentioned that normal rabbit serum has similar precipitating power, but to so small a 242 Contributions to Medical Science degree that it is easy to avoid error. A definite statement as to whether this action is purely agglutinating, or mixed agglutinating and precipitating, cannot be made until the test is apphed suc- cessfully to an extract free from morphological elements. The neurotoxic serum has been injected into the vessels, the abdomen, and beneath the skin of guinea-pigs. Although symp- toms are produced by the last two methods, they have not been used extensively, as much less serum is required in the case of intravascular injections. The following brief reports of experiments illustrate the symp- toms which may follow intravascular injections. Cerebral neurotoxin, guinea-pig i. — 2 c.c. of active serum injected in left carotid, toward brain. Retarded and labored respiration, at first deep, then superficial, ensued almost immediately. No convulsions, profuse urination; death in 3 to 5 minutes. 2 c.c. of normal rabbit serum injected similarly produced only slight muscular twitchings, which soon ceased (Cerebr. Nt., Control, guinea-pig 5). Autopsy of guinea-pig i showed only a slightly blood-stained fluid covering the brain and minute intramuscular hemorrhages. Cerebral neurotoxin, guinea-pig 4. — 2 c.c. active serum injected in right jugular toward heart. In 3 to 5 minutes appeared mori- bund, the only sign of life being light and slow respiration. Passive, limp, no convulsions. Urinated. Pain sense present. Eyes staring and slightly bulging, but muscles not paralyzed. In two hours began to recover slowly from the marked prostration. In eight hours was able to move about slowly and with difl&culty. Would lie down almost continuously in the cage. In ten days paralysis of the posterior limbs had developed, the fore legs were weakened, and emaciation was marked. Died in fourteen days with no new symptoms. The patellar reflexes were not lost at any time (3 c.c. of normal rabbit serum injected in the jugular produces no symptoms). At the autopsy of this animal the frontal and sphenoidal lobes were very soft and glistening; no congestion or edema. The erector spinae muscles and the posterior muscles of the thighs were pale and waxy in appearance. The Action of Neurotoxic Serum 243 Spinal neurotoxin, guinea-pig i. — 2 c.c. of this serum injected in the right carotid was followed by prolonged spasms, curvature of the body toward the injected side, exophthalmia, lachr>Tnation, and transient paresis of the left side. The deformity, paresis, and convulsions ceased after four hours. Dyspnea then developed and death occurred in five hours. Smaller amounts of cerebral and spinal sera produce convul- sions which last a short time, and prostration and emaciation of several days' duration. Anticerebellar serum produced convulsions, prostration, lachry- mation, and emaciation. In animals dying a few hours after injection, hemorrhages have been found in the brain and meninges, lungs and muscles. Changes in the central nervous system (Dr. Rothstein). — The anatomical changes in the central nervous system may be summed up in two words: hemorrhage and degeneration, i.e., degeneration of the ganglion cells. To facilitate description the animals may be divided into three groups : first, those which died one-half to one hour after the injec- tion; second, those which died three and one-half to four and one- half hours after; and third, the one animal which died fourteen days after the injection of the neurotoxic serum. In the animals of the first group were found multiple pial and subpial hemorrhages in the cerebrum, cerebellum, and spinal cord. Blood extravasates also existed in the gray matter, especially around the central canal and in the anterior horns of the spinal cord. In one animal a hemorrhage had completely destroyed the anterior horn on one side in the cervical region. Not a trace of the central canal could be detected, and the hemorrhage extended in the central part of the cord up into the fourth ventricle. Puncti- form, as well as more extensive hemorrhages existed in the tissue beneath the ependyma, and the ependyma was in places entirely destroyed. The ganglion cells, on the other hand, were only slightly altered. The great majority of them might be used to demonstrate the normal structure of the nervous cells. In a few places, however, can be observed a sUght but distinct chromatolysis. The preparations from the second group of animals present a 244 Contributions to Medical Science vastly different picture. The hemorrhages are few, small in size, and only found in the pia mater. The ganglion cells, on the other hand, are all more or less changed, and, as a general rule, it may be said that the larger the cells the greater are the changes they present. The cells in the spinal root ganglia are an exception to this rule, being markedly less changed than the other kinds of nervous cells. In the spinal cord only a few cells are found in which the Nissl bodies can be distinguished, but the tigroid sub- stance, even in these cells, is less in amount and without the distinct arrangement of the normal cell. In some cells are found only traces of the Nissl bodies and in still others they have entirely disappeared, the cells being lightly and diffusely stained. But not only the tigroid substance is altered. Many of the cells, especially the great motor cells, show vacuolization, and sometimes the nucleus has lost its outlines. Still greater changes are found in a considerable number of cells, the protoplasm being in a state of dissolution, presenting a picture that reminds one of coagula- tion necrosis. The ganglion cells of the cerebellum and cerebrum also show alterations to a greater or less degree. In the preparations from the third group were found still fewer hemorrhages than in the second group. The chroma- tolysis may be said, as a rule, to be more general. In nearly all the cells the Nissl bodies have entirely disappeared; but other changes of the protoplasm are not so great as in the second group. The cells look homogeneous, staining slightly and dift'usely. Vacuo- lization is not found frequently, and seldom is a cell seen that might be said to be in a state of necrosis. Usually the nucleus appears normal. The chromatolysis extends over the whole central nervous system. THE REDUCTION OF METHYLENE BLUE BY NERVOUS TISSUE/ H. T. RiCKETTS. (From the Pathological Laboratory of the University of Chicago.) A number of years ago Ehrlich pointed out the ability of living cells to decolorize (i.e., reduce) solutions of methylene blue. Later, Neisser and Wechsberg^ devised an application of this phenomenon to determine the cytocidal effect of some cellular poisons upon leukocytes. They found that living leukocytes were able to reduce the dye quantitatively, but when they were killed by heat, alcohol, an immune leuko toxic serum, or staphylococcus "leukocidin," they lost this power. Other cells, when fresh (sper- matozoa, pancreas, kidney), also reduced the dye, but not after treatment with an injurious agent (alcohol). Living micro- organisms had a similar effect, which was lost on treating them with a bactericidal substance. Ferments and bacterial toxins had no reducing power. I attempted to use this method in determining the toxic effect of an immune neurotoxic serum upon nervous tissue. The serum was prepared by immunizing rabbits and geese with the central nervous system of guinea-pigs, and was generally of such strength that 2.0 c.c. intravenously, 0.5-1 c.c. in the carotid, or 0.01-0.5 c.c. intracerebrally caused the death of a guinea-pig in two to twelve hours. The histological examination of tissues as reported by Ricketts and Rothstein^ shows that this serum produces, intra vitam, profound changes in the nerve cells, while the effects may extend to other tissues as well. In view of the results of Neisser and Wechsberg, it was expected that this "bioscopic" method might be applied here to advantage. Neisser and Wechsberg placed their mixture of cells, toxin, and methylene blue in small test-tubes, and covered the fluid with liquid paraffin to exclude the air, this being necessary for the accomplishment of reduction. I have found more convenient a ■ From Jour. Infect. Dis., 1904, i, p. 590. ' Miinch. med. Wchnschr., 1900, 47, p. 1261; Zlschr. f. Hyg., 1901, 36, p. 330. J Trans. Chic. Path. Soc, 1903, $, p. 207. 245 246 Contributions to Medical Science tube, suggested by Professor Hektoen, of 4-5 mm. in diameter, drawn to a fine point at either end, which can be sealed in the flame after the tube is filled. For the first experiments an emulsion of a guinea-pig brain in 30 c.c. of 0.85 per cent NaCl was used, the tissue being ground fine with quartz sand, from which it was decanted; later a 5 per cent emulsion was used uniformly. Aseptic precautions were used throughout and cultures made from the pipettes after the termination of the experiment. That contamination was infre- quent may be due in part to bactericidal properties of the emulsion or of the methylene blue, or of both. Experiments which showed contamination were discarded. Reduction was allowed to take place at room temperature. There is a quantitative relationship in the reduction of methyl- ene blue by nervous tissue. In one case were used 0.5 c.c. of emulsion and varying quantities of methylene blue (o . i drop to 16 drops of a methylene blue solution which contained 5 c.c. of a saturated alcoholic solution of the dye and 95 c.c. sterile NaCl solution). One- tenth to one-half of a drop was completely re- duced; in tubes with 0.5-5 drops partial reduction occurred, the color in succeeding tubes increasing gradually. In the remaining tubes the overlying fluid was densely blue, while the sediment and a very small zone above were decolorized after twenty-four hours. The fact that such an emulsion will reduce methylene blue at all probably indicates that intact cells are not essential. An emul- sion kept in the ice-chest for a week will reduce, although less vigorously than when fresh. It occurred to Neisser and Wechs- berg that this might be true, for they say: "It is entirely con- ceivable that certain cells may contain substances which in themselves have a reducing power, without regard to whether the cells as such are living or not." The reduction is in no case permanent. Tubes after being completely reduced for three days regain the original color when shaken with air. A drop of formalin added to 0.5 c.c. of the emulsion inhibited reduction, so that at the end of three days there was only a trace, the control reducing completely in two hours. Reduction of Methylene Blue by Nervous Tissue 247 Exposing the emulsion of nervous tissue to a temperature of 70° C. for thirty minutes destroys its reducing power (Table i). Tubes of the emulsion were treated as indicated in the table; then I drop of the methylene blue solution was added. TABLE I. Effect of Heat on the Reducing Power. So° C. for 30 tain No retardation; complete in 2 hrs. 56° C. for 30 min Retarded; not complete in 10 hrs. 60° C. for 30 min Retarded; not complete in 10 hrs. 65° C. for 30 min Retarded; partial reduction in 18 hrs.; J complete in 48 his. 70-95° C. for 30 min No reduction in 24 hrs. Other experiments show that the emulsion also loses its reducing power when heated to 98° C. for five to ten minutes. Subjecting the nervous tissue to the action of fresh (or acti- vated) neurotoxic serum did not destroy, or even decrease, the reducing power of the emulsion. In Table 2 the mixtures were sealed at once without allowing time for a preUminary digestive ( ?) action of the serum upon the nervous tissue. TABLE 2. The Influence of NEtTROxoxic Serum on the Reducing Po\ver of Nervous Tissue. Serum Obtained from Goose after Immunization with Nervous Tissue of Guinea-Pig. Emulsion Serum Meth. Blue Solution o.s c.c. O.S c.c. o.s c.c. o.s c.c. o.s c.c. o.s c.c. o.s c.c. 0.4 c.c. 0.3 c.c. 0.2 c.c. 0.1 c.c. o.s c.c. S drops Complete reduction in i8 hrs. No reduction Similar results were obtained when fresh rabbit serum was used in order to add complement, previous experiments having indicated that rabbit serum contains complement for the neurotoxic amboceptor of the goose serum. Other experiments showed that the addition of neurotoxic serum to the nervous-tissue emulsion intensified the reducing power of the latter, rather than inhibiting or destroying it. A tube containing 0.5 c.c. of serum reduced much more quickly than one containing 0.05 c.c. The serum alone is non-reducing. Similar results were obtained by treating the nervous tissue 248 Contributions to Medical Science with an immune rabbit serum. Digestion of the tissue for six hours with such a serum resulted in an increase in the reducing power, provided the serum was not removed by washing. If, however, the serum had been removed in this manner, the reducing power was decreased. For example: a) Five c.c. of an emulsion two days old was centrifugated and the overlying fluid drawn off. One and a half c.c. of an active neurotoxin was then added to the sediment and the mi.xture kept at 37° C. for two hours, after which it was diluted to the original volume with salt solution. Test I. — One c.c. of a) + i drop methylene blue solution: complete reduction in twelve to twenty-four hours. Test 2. — a) was again washed, resulting in the removal of a large part of the serum : no reduction in three days. b) Same as a), except that normal rabbit serum was substituted for the neurotoxin. Test I. — Complete reduction. Test 2. — No reduction in three days. c) Same as a) and b), except 0.85 per cent sodium chloride solution was substi- tuted for serum. Test I. — Partial reduction. Test 2. — No reduction in three days. d) Controls: 1. One c.c. of untreated emulsion reduced. 2. Two c.c. of pure rabbit serum gave no reduction. It seems, then, that normal serum as well as the immune serum intensifies the reducing power. It is also evident that repeated washing of the emulsion — c), i and 2 — eliminates its reducing power, the presence of which in the original emulsion is indicated by Control i. The conditions suggest that in the reduction of methylene blue by the nervous tissue and serum, as well as by the fresh emulsion without serum, there may be a co-operation of two substances, neither of which alone is reducing — c), 2 and Control 2. Inasmuch as repeated washing of an emulsion inactivated it, it seemed probable that separation of the two suspected sub- stances might be accomplished by extraction. Attempts at sepa- ration by centrifugating and washing a perfectly fresh emulsion were not successful ; but after it had been allowed to extract for one to several days, one of the substances passed into the liquid portion in a state of solution or suspension, the other remaining incorporated with the solid tissue. Reduction of Methylene Blue by Nervous Tissue 249 In Table 3 the emulsion had been kept in the ice-chest for eight days. TABLE 3- Separation of the Reddcing Substances in Nervods Tissue by Extraction and Washing. 24 hrs. 48 hrs. 72 hrs. I. Unwashed emulsion Partial reduc. Reduced Reduced 2. Sediment washed three times 4. Washed sediment +3 drops of guinea-pig senun * In the course of several days pure serum causes partial reduction of the dye. It appears, then, that the serum of guinea-pig is able to replace the substance which was extracted from the tissue. Normal rabbit serum answers equally well for reactivation. The extract' when added to washed sediment caused reduc- tion, although either alone did not reduce. Reduction is not so vigorous, however, as when the original emulsion is used, probably because a portion of one or both substances is wasted in the inter- mediate washings. Filtering the extract through a Berkefeld filter removes the reactivating substance from the fluid. This would seem to indicate either that the substance is in large molecu- lar form, or that it is associated with tissue particles in sus- pension which are held back by the filter. The effect of successive washings (which means the progressive removal of the soluble substance) is seen in Table 4. Extraction was allowed to proceed for twenty-four hours in the refrigerator. TABLE 4- The Effect of Repeated Washings on the Reducing Power of the Emxtlsion. Unwashed emulsion . Washed once — no serum Washed once — serum, 2 drops. Washed twice — no serum Washed twice — serum, 2 drops. Washed 3 times — no serum Washed 3 times — serum, 2 drop>s. Two drops of serum alone 12 hours Complete o Complete o Partial reduction o Partial reduction 24 hours Complete Trace reduction Complete Slight trace Complete Almost complete 72 hours Complete Trace reduction Complete Trace reduction Complete Complete It is seen that as washing progresses there is a slight tardiness in the reactivation of the sediment by serum. " By the "extract" is meant the fluid of a one to several days old emulsion after the sediment has been removed by centrifugation. 2^0 Contributions to Medical Science A fresh emulsion may be inactivated by heating to 98° for five to ten minutes, or to 70° for thirty minutes. The soluble substance is hereby destroyed, the one resident in the tissue sedi- ment still being capable of reactivation by serum. The rapidity and completeness of reduction are proportionate to the quantity of serum used for reactivation. In Table 5 the emulsion was three and one-half days old, and for the experiment was heated to 98° for ten minutes in order to inactivate its reducing power. Guinea- pig serum four days old was used to reactivate. The unheated emulsion reduced completely in twenty-four hours. Three drops of the serum alone did not reduce. TABLE 5- The Ability of Normal Serum of Guinea-Pig to Reactivate the Reducing Power. After the Latter Had Been Destroyed by Heat. Emulsion Serum 6 hours 24 hours 48 hours 72 hours I.O c.c I.OC.C 3 drops Almost complete Complete Complete Complete I.OC.C 2 drops Partial '* " 1.0 c.c I drop Trace Almost complete '* " 1.0 c.c i drop Marked reduction " ** I.O c.c i drop Partial Marked I.OC.C i^a drop Trace (?) Trace After five to eight days both substances may exist to some extent in the extract, since it is possible to render it reducing by the addition of serum. In order to determine whether or not the reactivating power of serum depends on complement, guinea-pig serum was heated to different temperatures, and its reactivating power then tested (Table 6). In this experiment a 5 per cent emulsion of nervous tissue was extracted for two days, then inactivated by three washings. The serum was from freshly drawn blood. In this experiment it is seen that the heating of the serum, rather than destroying the reactivating power, has increased it, in which respect there is no resemblance to ordinary complement. It is also different from the "soluble" substance of the extract as a reactivator, since the latter is destroyed by a temperature of 70° for thirty minutes. A difference from ordinary complement is also shown by the fact that an old serum will serve to reactivate. Temperatures higher than 70° were not tried. Reduction of Methylene Blue by Nervous Tissue 251 TABLE 6. The Effect of Heat upon the Reacting Power of Normal Serum of Guinea-Pig. Washed sediment (i .o c.c. in each tube) to which was added i, 2, and 3 drops of serum, the latter having been heated to the different temperatures for thirty minutes. The washed sediment in each case was suspended in the original volume of 0.85 per cent NaCl. Temp. Observation Time Serum, i Drop Serum, 2 Drop>s Serum, 3 Drops 40" 12 hours 20 " 33 " Almost complete Complete Complete o(?) Complete 45° 12 " 20 " 3i " Almost complete Complete Cornplete Slight trace Complete so" la " 20 " 33 " Complete Trace Complete Trace Coniplete 5S* 12 " 20 " 33 " Complete Partial Complete Almost complete Complete 6o' 12 " 20 " 33 " Partial (I) Partial (!) Marked Complete Almost complete Complete 6S* 12 " 20 " 33 " Trace Complete Almost complete Complete I 70* 12 " 20 " 33 " Almost complete Complete Almost complete Complete " Controls. — (i) Unaltered emulsion: complete reduction in twelve hours (2) Extract: no reduction in twenty-four hours. (3) Washed sediment: no reduction in thirty-sLx hours. (4) Washed sediment -|-fresh, unhealed guinea-pig serum: I Drop 2 Drops 3 Drops Twelve hours Twenty hours Thirty-three hours . Partial Complete Trace Complete Marked Complete (s) 3 drops of secum heated to 40-70° C. for thirty minutes showed no reduction in thirty-three I. hours. Influence of acids and alkalies. — Hydrochloric acid, except when an exceedingly minute amount is added, destroys the redu- cing powers of the fresh emulsion. It will not reactivate an inactive sediment. Quantities of o. i-o. 2 c.c. of normal potassium hydrate solution will reactivate the boiled emulsion or the sedi- ment which has been made inactive by washing. Smaller quan- tities of the alkali have a noticeable, but weaker, effect. Potassium hydrate alone does not reduce methylene blue, although it affects the dye chemically, as shown by the appearance of a lavender color. 252 Contributions to Medical Science Influence of tetanus toxin. — Tetanus toxin in any quantity does not affect the reducing power of nervous tissue for methylene blue. SUMMARY AND CONCLUSIONS. The ''bioscopic" method proved unsuitable as a means of determining the toxic action of neurotoxic serum, owing possibly to the fact that this serum, in common with normal serums, con- tains properties adjuvant to the reduction of methylene blue. In the reduction of methylene blue by nervous tissue the living cell is not essential. Reduction by the tissue emulsion is accomplished by the com- bined action of two substances, one of which, a thermolabile substance, may be extracted from the tissue by a 0.85 per cent NaCl solution; the other, a thermostabile substance, is closely associated with the solid tissue. Serum, whether it is old, fresh, or heated to 70° for thirty minutes, may be substituted for the thermolabile substance, as also may potassium hydrate. Chemists have determined that the reduction of methylene blue may be accomplished only by the action of nascent hydrogen on the dye. It becomes leuko-methylene-blue when its affinities for hydrogen have been satisfied.' In the original emulsion this hydrogen may have its source in fermentation processes (glycolytic or proteolytic), since certain ferments and the substances which they specifically affect are known to exist normally in tissues and body fluids. When serum is used as a reactivator, it is questionable if the reduction can be referred to the action of ordinary ferments in view of the heat resistance of the substance contained in the serum (70° for thirty minutes). The serum may contain, in addition to ferments, some obscure catalyzing agent which acts chemically upon substances contained in the tissue. It would seem that the action of potassium hydrate must be referred to its catalyzing properties. ' Boiling a solution of glucose and sodium carbonate which contains methylene blue, or the hydrogen evolved by the action of a mineral acid on tin in a methylene blue solution, accomplishes the reduction of the dye. Potassium ferricyanide reduces, the ferrocyanide oxidizes, the dye. Reduction of Methylene Blue by Nervous Tissue 253 It has not been possible to determine the amboceptor and complement nature of the reaction under discussion, since the binding phenomenon could not be demonstrated; however, a degree of analogy is not lacking. A few experiments performed with 5 per cent emulsions of liver and kidney tissues indicate that similar principles prevail in the reduction of methylene blue by these tissues. Although a most interesting chemical problem is concerned in the phenomenon presented, one is confronted with so many unknown substances and reactions that the writer has not thought it profit- able at present to follow this work further. It is not unhkely that methylene blue may be made to play the part of an indicator of reduction reactions where sterile substances can be dealt with. CONCERNING THE POSSIBILITY OF AN ANTIBODY FOR THE TETANOPHILE RECEPTOR OF ERY- THROCYTES: A RECEPTOR STUDY.' H. T. RiCKETTS. {Prom the Pathological Laboratory of the University of Chicago.) In the Transactions of the Chicago Pathological Society, 1904, 6, p. 237,^ I gave a brief summary of some experiments, the object of which was to determine the abihty or inability of various recep- tors to cause the formation of specific antibodies. We are accus- tomed to say that those receptors of erythrocytes with which the hemolytic amboceptors unite in the process of hemolysis or in the binding experiment are the particular constituents of the cor- puscles which cause the formation of the amboceptors when the cells are injected into the body of an animal. The amboceptors are in this sense antibodies for the corresponding receptors of the erythrocytes. Similarly, the receptors of corpuscles with which agglutinin unites, and those of bacteria with which the bactericidal amboceptors and the agglutinating bodies unite, have as their specific antibodies the hemagglutinin, the bactericidal ambocep- tors, and the bacterium-agglutinin respectively. The question arises: Is this a general law? Are all bodies which function as receptors capable of giving rise to the forma- tion of specific antibodies when they are injected into a foreign species ? The question has appeared to the writer to be one of funda- mental importance, especially as it would seem to have a direct bearing on the ability of certain pathogenic bacteria to give rise to the formation of antitoxins. The body of the tj'phoid bacillus contains a toxic substance which we reasonably suppose to be the pathogenic constituent of the organism. Yet immunization with this toxin-containing bacterium does not cause the formation of an indisputable typhoid antitoxin. One may assume, on the one ^ Jour. Exp. Med., 1905, 7, p. 351., ■"Receptor Studies Suggested by the Side-Chain Theory of Immunity"; see also "The Effect of Tetanolysin on Sensitized Erythrocytes," Trans. Chic. Path. Soc, 1905, 6, p. 288. 254 Antibody for Tetanophile Receptor of Erythrocytes 255 hand, that the specific toxin substance of the typhoid bacillus is a true toxin, with a haptophorous and a toxophorous group, but that the conditions in the body are such that a specific antitoxin cannot be formed. The same argument naturally would apply to several other pathogenic organisms. Food-stuffs and various constituents of cells are present in the body continuously, which, according to the theory of Ehrlich, may be taken up by the recep- tors of the cells, but for which it is thought antibodies do not develop {horror autotoxicus) . The particular receptors with which these substances unite may not be over-produced. It is with receptors of this nature that a typhoid toxin, for example, may imite and yet not stimulate to the formation of typhoid anti- toxin. On the other hand, it is possible that the essential patho- genic substance of the bacillus is not a toxin in the sense of Ehrlich. It is the purpose in this paper to present the results obtained in an attempt to demonstrate the presence or absence, in an immune hemolytic serum, of an antibody capable of uniting with those receptors of erythrocytes for which tetanolysin has an affinity. We may speak of these receptors as the tetanophile receptors of the erythrocytes; of those with which the hemolytic amboceptors unite as the serum-lysophile receptors, and of those with which serum agglutinin unites as the serum- agglutinophile receptors. It must be frankly stated at the outset that the results obtained throw no light on the question which was proposed, yet it is thought that the conditions chosen for the experiments and the difficulties which have been encountered should be published. If an antibody for the tetanophile receptor is formed when red blood corpuscles are injected into an animal, the process, according to the theory of Ehrlich, would be the following: When the erythrocytes are injected they disintegrate, are dissolved or digested, the receptors being thus hberated. The tetanophile molecules or receptors would eventually reach a group of cells which contain molecules or receptors with which the former could unite. In accordance with the general law the latter would then be stimulated to over-production and many of them thrown into the general circulation. When the serum of the immunized animal is collected and mixed with erythrocytes of the type injected, 256 Contributions to Medical Science the antibodies for the tetanophile receptors should unite with the latter as they exist in the red corpuscles. What might follow this union would depend on the character of the antibody. Inas- much as hemolysis and agglutination are the only reactions observ- able when corpuscles and serums are mixed, it could not be stated definitely that an antibody for the tetanophile receptor would not be able to cause solution of the corpuscles or their agglutina- tion. In this case the antibody would be a receptor of the second or third order, since it would be possessed of a toxophorous con- stituent. If, however, it were a receptor of the first order it would be passive in its nature, since such receptors possess no ferment- like or toxic action. For the purpose of the experiments it was assumed that an antibody for the tetanophile receptor would unite with the latter and thus make impossible the anchoring of tetanolysin, just as inactive agglutinoids unite with the agglutinophile receptors of bacteria and stand in the way of subsequent union of the latter with active agglutinin. Under the assumption, the antibody in question and the tetanolysin molecule would have identical haptophore groups. It is appreciated that even if the supposed antibody is formed, difficulties might stand in the way of its demonstration by the method chosen. In the first place, the heating of the serum, which was found to be necessary in order to destroy its hemolytic power, might destroy the antibody. Again, special temperature conditions, other than those utilized, might be necessary for union. Or union having occurred, it might be an easily dissociable one; the tetanolysin might have a greater affinity for the tetanophile receptor than the antibody has, and on this account be able to displace the latter, or the union might be dissociated by subsequent washing with salt solution. The method of experimentation consisted of exposing a definite quantity of washed erythrocytes to the action of varying quantities of the heated immune hemolytic serum for different lengths of time and at different temperatures, after which the serum was partially or totally removed by centrifugation, and the effect of tetanolysin then observed on the corpuscles. The heating of the immune serum was found to be necessary inasmuch as I did not Antibody for Tetanophile Receptor of Erythrocytes 257 succeed in preventing some subsequent hemolysis in the binding experiment with unheated serum at low temperatures. When such hemolysis occurred it naturally rendered an estimation of the action of the tetanolysin impossible. Normal serums used as con- trols were treated in the same way. Inasmuch as the tetanophile and the serum-lysophile receptors are not identical, it was assumed that the absorption of the ambocep- tors did not prevent subsequent fixation of the tetanolysin by the tetanophile receptors. If the two receptors were identical, anti- tetanolysin should have a haptophore capable of uniting with the cytophilous haptophore of the amboceptor. Experiments showed that heated normal goat serum, rich in antitetanolysin, when added to fresh immune hemolytic serum from the goat, did not reduce the dissolving power of the latter serum. For a similar reason it was concluded that the tetanophile receptor is not identi- cal with the agglutinophile receptor. TABLE I. The Effect of Saponin on Corpuscles Which Were Agglutinated by Ricin. Ricia and saponin, each, o. 2 per cent solution; 5 per cent emulsion of washed guinea-pig erythrocytes in 0.85 per cent* salt solution; corpuscles and ricin at 37° for thirty minutes; ricin removed by repeated centrifugation; o.i c.c. saponin solution added. Ricin Agglutination Saponin Hemolysis i.o c.c. Complete 0. 1 c.c. Trace 0-5 0.3 0.1 0.0s 0.03 O.OI Partial " Slight Moderate Marked Complete • Throughout the experiments o . 85 per cent salt solution and i c.c. of a 5 per cent emulsion of erythro- cytes were used. Total volume in all tubes was 2 c.c. The first important complication which developed consisted of agglutination of the corpuscles by the serum. Although the agglutinated mass could be broken up into very fine particles by pipetting the solution back and forth, nevertheless such corpuscles often sank to the bottom readily and became reagglutinated. The mere sedimentation of the corpuscles renders them physically less accessible to the action of the tetanolysin. It is possible also that the alteration of the envelope of the corpuscle which seems to occur in agglutination renders them less susceptible 258 Contributions to Medical Science to the action of a hemolytic agent. Corpuscles (guinea-pig) which have been agglutinated by ricin or abrin are to a large degree protected from the hemolytic action of saponin, although they are readily hemolyzed when distilled water is added. The smaller the amount of ricin used the greater the hemolysis which occurs when saponin, not a true toxin, is added subsequently. Here also there may be chemical or physical changes in the envelope, or the union of hemoglobin with the stroma may be made more firm by the action of the ricin, but the physical pro- tection of agglutination cannot be disregarded. Guinea-pig erythrocytes which are treated with a heated immune hemolytic serum (rabbit) are also made resistant to the action of saponin, the corpuscles having been strongly agglutinated by the serum. Experiments with Saponin and Tetanolysin. 1. Control: corpuscles were completely hemolyzed by o.i c.c. saponin solution in a few minutes. 2. Erythrocytes were treated with 0.3 c.c. heated immune serum for thirty minutes; washed free of serum; o.i c.c. saponin solution: no hemolysis. These corpuscles may subsequently be hemolyzed by distilled water. 3. Corpuscles without saponin or serum show no hemolysis. 4. Non-agglutinating normal serum gave no protection. The following is an analogous experiment performed with the heated immune serum (rabbit), and tetanolysin: 1. Control: 0.6 c.c. of 0.2% tetanolysin solution hemolyzed the corpuscles of guinea-pig completely in thirty minutes. 2. Corpuscles without tetanolysin or serum show no hemolysis. 3. Corpuscles treated with 0.3 c.c. serum for thirty minutes; serum removed by washing; 0.6 c.c. tetanolysin solution: no hemolysis in twenty-four hours. The corpuscles were shaken and the clot broken up repeatedly during this time. These corpuscles were subsequently hemolyzed by distilled water. 4. Non-agglutinating normal serum afforded no protection. In view of the evident influence of agglutination as a protective condition against the action of hemolyzing substances, the pro- tection obtained in the experiment last cited cannot logically be assigned to a preoccupation of the tetanophile receptors. More recently experiments have been resumed using a heated hemolytic serum obtained from the goat by immunization with the erythrocytes of the sheep. A serum rich in hemolytic ambo- ceptors, but poor in agglutinins, is obtained, as pointed out by Antibody for Tetanophile Receptor of Erythrocytes 259 Ehrlich and Morgenroth. It was thought that agglutination as a complicating factor might be eliminated by the use of this serum. This proved not to be the case. Apparent protection was obtained in many experiments, even when the agglutination caused by the serum was slight, as shown in Table 2. The serum (goat i) and blood, thoroughly mixed and diluted uniformly to 2 c.c, were placed at 37° for thirty minutes. Tubes were centrifugated, and the overlying fluid drawn of!, leaving a total residue of about one-twentieth of a cubic centimeter. The tetanolysin (i . o c.c. of a o . 2 per cent solution) was then added, each tube diluted to 2 c.c. and the corpuscles thoroughly mixed by pipetting back and forth. That the agglutinated corpuscles were largely separated in this way is shown by microscopic exami- nation, in which no more than three to six corpuscles were grouped together. Tubes were examined after twenty-four hours. The following experiment seems to lose significance when a duplicate, in which the serum was still further removed by an additional washing, showed that the immune serum had no, or little, more protective action than the normal serums. Yet inasmuch as in this experiment two normal serums were used as controls, and the latter showed a lower protection than the immune serum, there must have been present some property in the immune serum, which in the normal serums was absent, or present in lower concentration. There are two possibilities by way of explaining the loss of protection occasioned by the washing, aside from the influence of agglutination. In the first place, an antibody for the tetano- lysin receptor may unite with the latter in the more concentrated serum mixture, but the dilution involved in the washing may cause dissociation of the union, so that after the second centrifuga- tion the tetanophile receptors may again be available for union with tetanolysin. In the second place, some other protective factor may be present in both the immune and normal serums, and in a particular experiment might be present in greater concen- tration in the immune serum than in the two normal serums. Antitetanolysin, or some substance of equivalent power, as choles- terin, may be present In the minute residuum of serum left in the 26o Contributions to Medical Science original tubes, and it might in a given instance be present in greater quantities in the immune serum than in the normal. The cor- puscles treated with the immune serum would in this instance be protected to a greater extent than those treated with the normal serum. If the serum left in the sediment after the first centrifuga- tion is 0.008 c.c. (estimated), a subsequent washing in 2 c.c. of salt solution reduces the residual serum to a quantity which has no antitoxic effect on the amount of tetanolysin used, as shown by experiments. TABLE 2. The Action oy Tetanolysin on Sheep's Erythrocytes Which Had Been Treated with Heated Immune Hemolytic SERim from the Goat. Immune Serum Nonn. Ser. No. I Norm No Ser. 2 Tetano- lysin Hemolysis I 2 3 0.8 0.8 8 I.O No hemolysis SUght 4 S 6 0.6 0^6 6 I No hemolysis Slight 7 8 0.4 0.4 " Trace SUght 9 4 10 II 12 0.2 0.2 2 I Trace Moderate Marked 13 14 0.1 0.1 « Moderate Marked IS I 16 17 Complete No hemolysis Inasmuch as various quantities of serum were used in the different tubes, a higher degree of protection, which the larger amounts of serum may have afforded, might be referred to the larger amount of antitoxin present in those tubes. To eliminate this factor the amounts of residual serum in the various tubes were equalized by washing with proportionate amounts of salt solution (Table 3). Since a large proportion of the protection in A was removed by the washing, there must be present a soluble antitoxic substance in the serum. If we assume that the inhibition of tetanolysin hemolysis in the above experiment was due to two factors, agglutination and an Antibody for Tetanophile Receptor of Erythrocytes 261 antitetanolytic property in the serum, and that in the washing the antitetanolysin was equaHzed, we have the following values for the two factors in those tubes in which 0.6 c.c. of serum was used: Total inhibition 90% 80% 40% 30% Antitoxin inhibition 70% 70% 40% 30% Agglutination inhibition 20% 10% o o These percentages are only of comparative value, inasmuch as somewhat more than the simple dissolving dose of tetanolysin was used, and at the same time the antitetanolysin in the different tubes was equalized but not entirely removed. TABLE 3. A. — Corpuscles of the sheep exposed to action of immune serum from the goat, tubes centrifugated, overlying fluid drawn off; corpuscles not washed; tetanolysin is i .0 c.c. of a o. 2 per cent solution. Seruh Amount Subsequent Hemolysis by Tetanolysin Immune Ser. No. 1 Immune Ser. No. 2 Normal Ser. No. I Normal Ser. No. 2 .6 c.c. ■4 10% 10 25 7S Control: 100 per cent 20% 30 60 100 60% 80 100 100 70% 100 100 100 B. — Duplicate of the above except that the quantities of residual serum in the different tubes were equalized by washing with proportionate amounts of salt solution. Serum Salt Sol. Used in Washing Subsequent Hemolysis by Tetanolysin Immune Ser. No. I Immune Ser. No. 2 Normal Ser. No. r Normal Ser. No. 2 I 3 3 4 S 0.6 c.c. 0.4 0.2 0.1 2.4 c.c. 1.6 .8 • 4 80% 80 100 1 00 Control : i 90 % 80 90 100 30 per cent 100 % 1 00 100 100 100% 100 100 100 The corpuscles treated with the immune serums were rather firmly agglutinated. Nothing is deducible concerning the presence or absence of an antibody for the tetanophile receptor. Any possible pro- tection afforded by the presence of this antibody cannot be sepa- rated from the protection due to the agglutination. Something might be learned concerning the nature of the pro- tection by fractional absorption of the immune bodies from the serum, and testing the action of tetanolysin on the erythrocytes which were used for the removal of the successive fractions of 262 Contributions to Medical Science immune bodies. Quantities of 0.6, 0.4, 0.2, o.i c.c. of the dif- ferent serums were diluted to 2 c.c. each, and the contents of each tube added to the sediment of i c.c. of the 5 per cent blood emulsion. After thirty minutes at room temperature the tubes were centrifu- gated, the overlying fluid drawn off and put into fresh tubes, while to the sediment was added the usual amount of tetanolysin in a total volume of 2 c.c. The serum dilutions were then sub- jected to the absorbing powers of a second mass of corpuscles of equal amount and the process repeated over four absorptions, the hemolytic power of the tetanolysin being tested on the red blood corpuscle sediments of each series. At the same time a series of tubes was made to determine the relative antitoxin con- tent of each of the serums (Table 4, A and B). There are three sources of error which may be responsible for some variations in the percentages of hemolysis in the different tubes and in the different series in A. First,. after the corpuscles were exposed to the action of the serum and the tubes were centrifu- gated, the overlying serum mixtures were drawn off as completely as possible by means of a glass pipette which was drawn to a fine point; in spite of care the amounts of residual serum in the different tubes were undoubtedly subject to variations, so that Tube i, for example, in the column headed "Immune Serum No. i," might contain in the different series variable amounts of residual serum — that is, variable amounts of antitoxin along with other serum constituents. Second, the completeness with which the agglutinated corpuscles in the tubes treated with the immune serums were broken up, probably was not constant in spite of caution. Third, the experiment occupied a whole day and the tetanolysin lost a small percentage of its strength during that time, although it was kept on ice when not in use. Analysis of the columns headed "Normal Serum," in which agglutination was absent, shows that although a moderate loss in the protection of the corpuscles takes place as absorption con- tinues, the degree of protection corresponds to the relative amounts of antitoxin in the two serums as shown in the antitoxin table. All, or nearly all, the protection may reasonably be assigned to the antitoxin. Inasmuch as the total volume of the sediment and Antibody for Tetanophile Receptor of Erythrocytes 263 TABLE 4- A. — The effect of tetaaolysin on corpuscles which have been used for the fractional absorption of antibodies from the serums. First Series. Immune Serum No. i Immune Serum No. 2 Normal Ser. No. 1 Normal Ser. No. 2 Agglut. by Serum Hemol. by Toxin Agglut. by Serum Hemol. by Toxin Agglut. by Serum Hemol. by Toxin Agglut. by Serum Hemol. by Toxin I 3 3 4 Firm Less Marked 10 10 25 75 Very firm Less Firm 20 30 60 1 00 60 80 100 too 70 100 100 100 S No serum, 100 per cent hemolysis: — Control. Second Series. Third Series. I Gradually IS Gradually 35 70 go 2 less from I to 4 20 decreased; more than 65 85 90 3 20 by Im- 100 100 100 4 75 mune Se- rum No. I 100 100 100 I Gradually 10 Gradually 75 ,0 80 2 less from 25 less from 90 90 100 ^ I to 3 6S I to 4 100 100 100 4 No agglut. 80 100 100 100 Fourth Series. I No agglut. 10 Moderate 95 70 80 2 No agglut. 15 Agglut. 95 100 3 No agglut. 55 Slight 100 lOO 100 4 No agglut. 65 100 100 100 B. — Antitetanolysin in the four serums. One cubic centimeter of a 0.2 per cent solution of tetanolysin is mixed with the quantities of the serums stated in the table, in a total volume of 2.0 c.c. After two hours' contact at room temperature the con- tents of each tube is added to the red corpuscle sediment of i .0 c.c. of the 5 per cent blood suspension. Serum Immune Ser. Immune Ser. Normal Ser. Normal Ser. No. I No. 2 No. I No. 2 .005 20 50 45 60 .0045 20 60 55 60 .004 25 70 70 60 .0035 30 70 75 65 .003 35 75 75 70 .0025 45 90 90 75 .002 6S 100 100 90 .0015 8S 100 100 90 .001 90 100 100 95 .0005 90 100 100 100 Control : 1 00 per cent 264 Contributions to Medical Science residual fluid in each tube was about 0.03 c.c, the antitoxin was diluted by that much each time the serum mixture was added to a fresh lot of corpuscles. This, however, amounts to a loss of only about 0.25 per cent from the first to the third series and is a negligible quantity. The corpuscles also may have absorbed some of the antitoxin, a possibility which was not determined. Analysis of the columns headed "Immune Serum," in which agglutination was present, shows a great loss of protection between the first series and the third, and the loss is more uniform and greater with the serum which had the higher agglutinating power; hand in hand with the loss of protection is the progressive decrease in the agglutinating power of the serum as absorption continues. The greater protection with Immune Serum No. i , which was pres- ent in spite of its lower agglutinating power, is explained when we note its greater antitoxic power (Table 4, B). This experiment, then, and duplicates which have been made afford no clue as to the presence of an antibody for the tetano- phile receptor; the protection which was encountered is explain- able on the basis either of agglutination or antitoxin action. Other experiments were performed with the view of deter- mining more exactly the antitoxic value of the residual serum in the different tubes. The results showed that in the case of normal serums the antitoxin of the residual serum accounted for practi- cally all of the protection, while in the case of the immune serums in which some agglutination had occurred, the calculated amount of antitoxin did not account for all the protection. The difference in protection by the two immune serums used corresponded to the differences in the agglutinating and antitoxic powers of the two serums. Here also no means developed of distinguishing an antibody for the tetanophile receptor from the agglutinating and antitoxic powers of the serums. Attempts to eliminate agglutinin by heat failed as the serums produced some aggluti- nation even after exposure to the temperature of coagulation. SUMMARY AND CONCLUSIONS. The treatment of washed erythrocytes of the guinea-pig by heated immune hemolytic serum derived from the rabbit, and Antibody for Tetanophile Receptor of Erythrocytes 265 of washed erythrocytes of the sheep by a similar serum derived from the goat, renders the erythrocytes more or less resistant to the subsequent action of tetanolysin. If all the serum is removed from corpuscles treated in this manner, it can be determined that some of the protection is due to the agglutination of the cells. A certain amount of the protec- tion afforded by agglutination is referable to the physical barrier which the agglutinated mass of cells offers to the uniform distribu- tion and diffusion of the tetanolysin. It is probable that none of the protection obtained is due to the mere union of agglutinin or of hemolytic amboceptors with their respective cell receptors; such union would appear to leave the tetanophile receptors still unoccupied. If a residuum of serum is left with the corpuscles which have been treated as indicated, the added protection which is acquired against the subsequent action of tetanolysin may reasonably be referred to antitoxin which is present in the residual serum. The possibility of the dissociation of the union between the tetano- phile receptor and its antibody, occasioned by the dilution inci- dent to washing away the serum, may not be entirely ignored, but has not been susceptible to determination. The same may be said of the possibility that tetanolysin, having a stronger affinity for the tetanophile receptor than has the antibody, is able to displace the latter from its union with the receptor. Hence treating corpuscles with the immune serums does not allow one to determine the presence or absence of an antibody for the tetanophile receptor, the experiments being formulated on the supposition that such an antibody would unite with the tetanophile receptors and thereby prevent subsequent binding of tetanolysin. THE ADJUVANT ACTION OF SERUM, EGG-ALBUMIN, AND BROTH ON TETANUS INTOXICATION.' H. T. RiCKETTS AND E. J. KiRK. (From the Pathological Laboratory of the University of Chicago.) In the course of other work it became necessary to assure our- selves concerning the content in tetanus antitoxin of the sera of several goats. To our surprise normal goat serum (i .o c.c), rather than exerting an antitoxic action, increased the toxicity of the toxin when suitable doses of the latter were used. The investiga- tion of this phenomenon is the subject of the present paper. TECHNIC AND MATERIALS. The toxin used throughout the experiments was prepared in January, 1903, as follows: Ten liters of beef broth, containing i per cent of glucose and having a reaction of -f-i, were placed in two bottles and inoculated with a large quantity of tetanus bacilli which had grown for 10 days in a similar broth. Washed hydrogen was then passed through the inoculated broth for an hour, and the bottles sealed and placed in the thermostat, the usual provisions being made to permit of the escape of gas from the bottles, and to prevent the access of air. After a growth of nine days the culture was passed through Pukal filters, placed in large moisture dishes, and an excess of ammonium sulphate added ;^ the dishes were then placed in the thermostat over night. The brownish scum which had formed by this time was skimmed oflf, placed between hardened filter papers, and the excess of moisture pressed out. Still more fluid and ammonium sulphate were removed by sub- jecting the precipitate to very high pressure in a pressure machine. The precipitate, now in the form of solid cakes, was dried over sulphuric acid and eventually pulverized. It is preserved over sulphuric acid in the ice-chest and in the dark. For use a o . 2 per cent solution of the precipitate was made in 0.85 per cent sodium chloride solution, and the doses used are ■ Jour. Infect. Dis., 1906, 3, p. 116. ' One-half more than the quantity the broth would dissolve at room temperature. 266 Adjxa'^ant Action of Serum on Tetanus Intoxication 267 expressed in cubic centimeters of this solution. The original fatal dose for white mice of about 15 grams weight was 0.000,007 c.c. per gram of mouse; death occurred in four to five days. For guinea-pigs of 250 to 300 grams, 0.000,001 c.c. per gram killed in four to five days. The toxin has undergone little or no deteriora- tion since it was prepared. When dealing with a substance of such high toxicity it is difii- cult to weigh out small quantities with desirable accuracy and at the same time observe economy of material. Our toxin solutions were usually made by dissohdng o . 2 gram of the precipitate in 100 c.c. of the salt solution, but, in spite of great care in weighing, variations in the toxicity of different solutions frequently came to light. Some of these variations may have been due to an irregular distribution of the toxin in the precipitate. A new toxin solution was prepared for each experiment. White mice were used as the test animals throughout the work, and the dosage of toxin was based on the gram-weight of the animals — i.e., so much toxin per gram. It was found, however, that this method could be used safely only when the various mice of a given experiment were approximately of the same weight, since, for example, 0.000,007 c.c. of toxin per gram is not equally toxic for a 20 gram and a 10 gram mouse. Susceptibility is not directly proportional to weight. The injections were made into the loose subcutaneous tissue of the back. We soon learned that the ability of normal serum to increase the toxicity of the toxin disclosed itself only when suitable doses of the toxin were used. If a dose of toxin which killed all controls in 48 to 72 hours was used, the increased toxicity caused by the addition of serum did not become manifest. Also if the dose was so small that very Uttle tetanus resulted in any of the animals, the effect of the serum was not always a decisive one. The clearest results were obtained when a dose of toxin was used which either caused moderate tetanus in the controls with eventual recovery, or which caused their death in from eight to twelve days. Because of the possibility of variations in the toxicity of toxin solutions, one could not feel sure that a single dose selected for an 268 Contributions to Medical Science experiment would be the optimum dose for the manifestation of the phenomenon. Hence it was decided to use a number of mice for each experiment, and to vary the dosage in such a way that an optimum dose would be administered to two or more of the ani- mals. It was necessary to use an abundant number of controls. The sera of a number of normal goats were used, also of normal rabbits, and for specific purposes which will be explained later, the influence of broth and of egg-albumin was tested. Cus- tomarily I c.c. of the serum (or broth, or egg-albumin) was mixed with a dilution of the toxin of which 0.5 c.c. contained the desired dose. The mixture was allowed to stand for varying lengths of time and the total quantity (i . 5 c.c.) then injected into the sub- cutaneous space in the back of the white mouse by means of a Luer or Roux syringe. After the mixture had been injected an additional 0.5 c.c. of salt solution was drawn into the syringe, and, after rinsing, was injected. In order to prevent any escape of fluid after its injection, the needle puncture was clamped for 15 to 30 minutes, the clamp being applied before the needle was withdrawn. In all experiments daily observations were made of the degree of tetanus present, as indicated by the amount of deformity and the general appearance of the animals. The degree of tetanus is expressed in the tables by figures: thus o = no tetanus; 1= per- ceptible rigidity; 2 = distinct but not pronounced; 3 = marked rigidity; 4 = severe; 5 = very severe to moribund; t = death. The observations of one of us served as a check on those of the other. EXPERIMENTS. Tables i, 2, and 3, which follow, are typical experiments show- ing the influence of normal goat and rabbit sera. Tables 1,2, and 3 require little or no comment. Particularly in Tables i and 2, there can be no question concerning the ability of the serum to increase the intensity of the tetanus intoxication from which death occurred. Table 3 is somewhat less decisive, but here also all animals receiving serum died from one to three days in advance of the controls. Adjuvant Action of Serum on Tetanus Intoxication 269 TABLE 1. The Intluence of Normal Goat Serum ok Tetanus Intoxication. Normal serum, Goat I. — Serum three days old, unhealed, preserved in the ice-chest. Serum and toxin mixed and allowed to stand 30 minutes at 37° C. before injection. Mouse Wt. in Grams Toxin PER Gram C.C. OF Serum Result by Days 1 2 3 4 S 6 7 8 9 lO 11 IS 1 a 14 14 16 16 14 14 14 14 13 13 0.000,001 0.000,001 0.000,003 . 000,003 0.000,006 0.000,006 . 000,008 . 000,008 0.000,01 0.000,01 1 .0 0.0 1 .0 0.0 1 .0 0.0 1 .0 0.0 1 .0 0.0 I 1 2 I I 2 3 1 2 I 2 2 3 I 4 2 5 2 I I 2 2 4 I t 3 t 3 I I 2 I 4 I 3 3 1 I 2 1 4 I 3 3 I 3 I I I I 3 I I I 3 I I I 2 I I I I 4 5 6 7 8 9 TABLE 2. Same as Table i, except that the serum of Goat II was used. Mouse Wt. in Grams Toxin PER Gram C.C. OF Serum Resttlt by Days I 2 3 4 S 6 7 8 9 10 II 12 IS I 14 14 15 15 14 14 .14 14 14 14 0.000,001 0.000,001 0.000,003 . 000,003 . 000,006 0.000,006 0.000,008 . 000,008 0.000,01 0.000,01 1.0 0.0 1.0 0.0 1.0 0.0 1 .0 0.0 1 .0 0.0 I 2 I I 1 2 I 3 2 I I 3 I 3 3 3 3 5 2 1 I 4 I 3 3 S 3 I I 4 2 4 2 t 4 I I 4 3 4 2 S 1 1 1 4 3 4 I t 2 I I S 3 4 1 2 t 2 4 I 2 2 t 2 2 2 I I |:;:::::;:;:;: Gost) TABLE 3- The Influence of Normal Rabbit Serum on Tetanus Intoxication. Serum, unheated, two days old. Mixtures stood for one and one-half hours at room temperature before injection. Mouse Wt. IN Grams Toxin per Gram C.C. OF Serum Result by Days 1 2 3 4 S 6 7 I 2 3 13 11 16 IS 19 16 0.000,003 0.000,003 0.000,006 . 000,006 . 000,008 . 000,008 1.0 0.0 1 .0 0.0 1 .0 0.0 2 2 3 3 4 2 3 3 4 4 t 2 3 5 4 S 4 S t t S t t 6:::::::::::::::::: In accordance with Table 4, o. i c.c. of serum is a more power- ful adjuvant than i c.c, the latter being the dose used uniformly. This point was determined only after the major part of the experi- ments had been completed. Its explanation is by no means clear. 270 Contributions to Medical Science The serum may contain a minute amount of antitoxin, which declares itself when larger quantities are used. TABLE 4. To Determine the Minimum Qoantity of Goat Serum Which Intensifies the Intoxication. Fresh serum, 18 hovirs old, from a normal Angora goat. Mouse Wt. 10 Grams Toxin per Gram C.C. OF Serum Result BY Days I 2 3 4 S 6 . 000,008 0.000,008 0.000,008 . 000,008 . 000,008 0.000,008 . 000,008 0.000,008 0.0 0.0 0.01 0.0s O.IO 0.40 0.70 1. 00 2 I 2 2 I I I 2 3 3 3 I I 4 4 t t S 3 3 S S t 5 t t t 6::::::::::::::;::::::::::::::::: 7 8 t Experiments, of which Table 5 is an example, were performed to determine whether the efifect of the serum could be attributed to its toxic action on the mice. If such a toxic action were present, the early death of the animals might be referred to a summation of the intoxication by the serum and that by the toxin. We looked on the subsequent weight of the animals as a clue to the presence or absence of serum intoxication. TABLE 5. To Determine the Toxicity or Non-Toxicity of Normal Goat Serum for White Mice. Serum three days old, unheated. Mouse Wt. IN Grams C.C. OF Serum Weight by Days I 2 4 S 6 17 14 17 I.O i-S 2.0 18 IS 18 I9S 17.0 20.0 18 16 20 18 IS 18 18 2 IS 18 There was no loss of weight, but even a gain on the second day, which may have been due to the drinking of more water. Hence we conclude that the phenomenon cannot be explained by any manifest toxicity of the serum for the mice. Other experiments were performed with the hope that some Hght might be thrown on the nature of the phenomenon. It occurred to us that tetanus toxin might be an amboceptor, and that the normal serum of the goat might contain suitable com- Adjuvant Action of Serum on Tetanus Intoxication 271 plement which, when injected, would increase the amount of suitable complement in the body of the mouse, and thus activate the toxin more quickly and more completely. To throw possible light on this point the comparative influence of fresh, old, and heated sera was determined (Table 6). TABLE 6. The Comparative Influence of Fresh, Heated, and Old Sera on Tetanus Intoxication. The sera were from Goat IV (female); the "heated" serum had been placed at 56° C. for 30 min- utes and was freshly drawn; the "old" serum had been drawn 11 days previously and kept at about 10° C. Mouse Wt. in Grams Toxin per Gram I c.c. OF Serum Result by Days I 2 3 4 ! S 6 7 8 1 9 |IO II 14 19 I 2 18 18 18 18 18 18 18 18 0.000,006 0.000,006 0.000.006 0.000,006 0.000,008 0.000,008 . 000,008 0.000,008 Fresh Heated Old None Fresh Heated Old None I I I I I I I 2 2 2 2 3 3 3 2 3 3 3 3 3 4 t 3 4 3 1 3 4 3 t 3 4 3 >' S S 4 4 t t 3 3 2 2 3 I 3 5 6 7 8 The experiment shows distinctly that the effect of the serum does not depend on an activation of hypothetical tetanus ambo- ceptors by ordinary thermolabile complement. The heated and old sera were rather more efficient than the fresh serum. Serum heated even to the coagulating point does not lose its adjuvant property (Table 7). TABLE 7. Heat Resistance of the Adjuvant Substance. Mouse Wt. 12 Grams Toxin per Gram I c.c. Goat Serum Heated TO — °C FOR 30 MIN. Result by Days 3 4 S 6 7 8 9 10 000,007 . 000,007 .000,007 .000,007 .000,007 .000,007 .000,007 . 000,007 .000,007 . 000,007 67.5-68* 67.S-68 64 . 5-65 64 5-65 59 ■ 5-60 59 ■ 5-60 Unheated Unhealed No serum No serum • The'serum at 67 . s°-68'' was gelatinous. Temperatures higher than 68° C. were not tried. Other experiments seem to show that the influence of the serum does not depend on any action on the toxin itself. It is not neces- 272 Contributions to Medical Science sary to mix the serum and toxin before injection in order to get the effect of the former. The two may be injected in different parts of the body, or the serum may be injected shortly in advance of the toxin. It is well known, on the other hand, that a substance which acts directly on the toxin (tetanus antitoxin) has a stronger effect when the two are mixed before injection. In order to get the maximum effect of the serum, however, it is necessary that the two substances be injected in fairly close sequence although they may be placed in different parts of the body (Table 8). TABLE 8. The Effect of the Serum When Injected at Different Periods in Advance of the Toxm. I CO. Serum Hours in Mice 10 Grams Weight Toxin per Gram Result by Days Advance of Toxin I 2 3 4 S 6 7 8 9 10 It 12 1 2 3 1 3 1 3 2 4 3 4 4 2 3 4 4 4 t 3 4 4 S 4 3 4 S t S 4 3 5 t 4 3 5 4 4 t S t 21 " " 2 " 3 " 4 " 5 " 6 . 000,008 0.000,008 000,008 . 000,008 0.000,008 No serum 2 t The results of this experiment give us the impression that the action of the serum depends on some temporary influence which it exerts on the tissues of the animal, whereby the latter is made more susceptible to the toxin. After the injection of the serum its influence gradually becomes less, and is not demonstrable after the lapse of about 46 hours. It is equally important from the standpoint of interpretation to know whether the serum intensifies intoxication when given sub- sequent to the injection of the toxin. The result of a single experi- ment indicates that the serum does not hasten the death of the animals when it is given subsequent to the binding of the toxin by the tissues. The controls which received serum and toxin simulta- neously died in five days, while animals in which the injection of the serum was given from one to fifty hours later than that of the toxin either recovered or died in from six to nine days. We had come to believe at this time that there was nothing of a specific nature in the phenomenon, and that many other proteid- containing substances might have a similar influence. We found Adjuvant Action of Serum on Tetanus Intoxication 273 this to be strikingly true in the cases of egg-albumin and broth (Tables 9 and 10). We have before us then the following facts upon which we may attempt to base conclusions: 1 . The normal sera of the goat and rabbit intensify and hasten the course of tetanus in white mice, when suitable doses of our precipitated toxin are used (Tables 1,2, and 3). 2. It is immaterial whether the serum is fresh, old, or heated to the coagulating point (Tables 6 and 7). 3. The effect is most pronounced when the serum is injected TABLE 9. The Influence of Egg-Albumin on Tetanus Intoxication. One per cent of egg-albumin in 0.85 per cent NaCl solution. Result by Days Mouse WT. IN Grams Toxin pee Gram C.C. OF Albumin I 2 3 4 5 6 7 8 9 I IS 0.000,00s 1 o.s I 2 3 4 t 2 15 . 000,005 1 0.0 I I 3 IS 0.000,006 1 O-S I 2 S t 4 IS o.ooo,co6 0.0 I I I I I s IS 0.000,007 o.s I 3 t 6 IS 0.000,007 0.0 I I I I I TABLE 10. The Influence of Broth on Tetanus Intoxication. Mouse Wt. in Grams Toxin per Gram C.C. OF Broth Result by Days 10 II, 12, 13 a IS 3 18 4 18 S ' IS 6 , IS 0.000,006 0.000,006 0.000,007 0.000,007 0.000,008 0.000,008 Recovery simultaneously with, or shortly preceding, the injection of the toxin. If the serum is given 46 hours in advance of the toxin it is in some way disposed of so that it no longer intensifies the intoxica- tion. It is not necessary that the serum be mixed with the toxin, nor injected into the same part of the body (Table 8). 4. A small quantity of serum (o.i c.c.) seems to have a more pronounced influence than a larger quantity (i c.c.) (Table 6). 5. Normal goat serum in a quantity of 2 c.c. produces no perceptible deleterious effect on the mouse (Table 5). 274 Contributions to Medical Science 6. One-half c.c. of a i per cent solution of egg-albumin in physiologic salt solution; and also i c.c. of broth have an influence like that of serum. The possibility which we at one time took under consideration that tetanus toxin might be an amboceptor, and that the serum increases toxicity because it provides an additional quantity of complement cannot be entertained: first, because old and heated sera, in which there is no complement (thermolabile), produce the phenomenon; and, second, because egg-albumin or broth may be substituted for the serum with the same result. We cannot accuse the broth, in particular, of containing complement. Inasmuch as we have not been able to refer the phenomenon to any influence which the adjuvant substances may exert on the toxin, we have been obliged to assume that it is due to some efifect on the tissues of the mice. Although the serum exerted no per- ceptible toxic action on the mice, which if it had occurred might have lessened resistance in some manner, we were bound to con- sider as a possibility that the serum, albumin, and broth might exert some particular influence on the nervous tissue whereby it either absorbed more toxin or was made less resistant to the toxin which it bound. In spite of this possibility, however, we have been unable to conceive of any manner in which these substances could produce such an effect on the nervous tissue. If the serum, etc., were to some extent bound as indifferent food substances by the nervous cells, we believe the latter might be preoccupied, so to say, in digesting them; but since such a process would engage the cells in a general way we believe their affinity for some second sub- stance, such as tetanus toxin, would be decreased rather than increased during this period. Concerning the alternative pos- sibility that the serum, etc., may injure the nervous cells so that they are less resistant to the toxin, we have only the argument that no injury detectable by gross means was produced. We find what appears as a more reasonable explanation of the phenomenon in an influence which the serum, etc., may exert on the remaining tissues of the body other than the nervous tissue. We learn from certain investigations by Metschnikoff,' and by ■ L'lmmuniti, Paris, 1901, p. 343. Adjuvant Action of Serum on Tetanus Intoxication 275 Roux and Borrel/ that other tissues than the nervous are able to bind tetanus toxin in some instances. The rabbit and also the chicken are much more susceptible to tetanus toxin when it is injected into the nervous tissue than into the subcutaneous tissue; in the latter instance the toxin comes in contact with various tissues some of which bind a certain amount. Metschnikoff found that the liver in some of the invertebrates absorbs a great deal of tetanus toxin. We are not aware of any experiments bearing on the antitoxic powers of the organs of the white mouse, nor have we performed such experiments. Wassermann and Takaki found that small quantities of the Hver, kidney, spleen, and bone-marrow of the guinea-pig exerted no antitoxic action, a condition which corresponds well with the exquisite susceptibility of this animal to tetanus. In comparing the susceptibility of the rabbit with that of the guinea-pig, we learn, first, that, gram for gram, the rabbit is about a thousand times more resistant than the guinea-pig (Knorr, cited by Dieudonne);^ and second, from Roux and Borrel, that a large part of this resistance resides in other than nervous tissues. The tetanus sine tetano of Donitz suggested to him that non-nervous organs in the rabbit bind tetanus toxin. ^ Such animals after receiving a dose of toxin which causes little or no tetanus gradually become emaciated and die. We cannot of course hold, without definite experimental proof, that the same conditions exist in the mouse. We know, however, that the mouse, gram for gram, is from six to ten times more resistant to tetanus toxin than the guinea-pig (Knorr);'' in our own experiments it is seven times more resistant, and it is quite possible that some of this resistance depends on the abihty of tissues other than the nervous to bind a certain amount of the toxin. If this condition exists, and if in some way union of the toxin with other tissues could be prevented, for example, with the liver, connective tissues, or lymphoid organs, just so much more toxin would be available for the more susceptible nervous tissue. We conceive that such a result may be caused by the serum, ■ Ann. de I' Inst. Pasteur, 1898, 12, p. 229. ' Immunildt, Schutzimp/ung, etc., Leipzig, 1903, p. 13. J Deutsche med. Wchnschr., 1897, 23, p. 428. < Cited by Dieudonn^, loc. cil. 276 Contributions to Medical Science etc., in one of two ways: First, certain receptors or substances in the serum, egg-albumin, and broth may possess cytophilous haptophores, identical with that of tetanus toxin, and by uniting with the tetanophile receptors of indiflferent organs (liver, etc.) may thereby render impossible the binding of the toxin; this extra toxin would then be available for the nervous tissue, which presumably has a higher affinity than other tissues for the toxin. Such a process would consist of a preoccupation of tetanophile receptors by heterologous substances. Second, the cells of indiflferent organs (liver, etc.) may bind the serum, etc., as they would bind food substances, after which their activities (ferments) may be directed toward the digestion or oxidation of the new substances, and being thus engaged, the affinities of their receptors for other substances (toxin) may be decreased. This condition would also render more toxin available for the nervous tissue. Such a process would consist of a non- specific engagement of the activities of the cells, without a direct occupation of tetanophile receptors. As an example of such a process we may mention the well- known experiments of Besredka,' in which granules of carmin injected into the peritoneal cavity decreased the ability of the leukocytes to take up granules of arsenic, with the result that the animals died the more readily of arsenic intoxication. Von Dungern also observed a clear example of the ability of one substance to interfere with the absorption of a second.^ By injecting the plasm of Octopus vulgaris or egg-albumin into the circulation of a rabbit the ability of the animal to absorb a sub- sequent injection of the plasm of Majo squinado (spider crab) was largely or entirely inhibited. Von Dungern's explanation of this phenomenon is somewhat incomplete, but we understood it to be the following: In reference to the particular substances injected, the cells of the rabbit have two types of receptors. One type is that which is able to take up various materials as food substances and is not a specific receptor. It takes up egg-albumin, the plasm of the spider crab and of the octopus, and doubtless many other • Ann. de t'Insl. Pasteur, iSgg, 12, pp. 49 and 209. ' Die Antikorper, Jena, 1903, p. 97. Adjuvant Action of Serum on Tetanus Intoxication 277 albumins. It is not concerned in the formation of antibodies. The other type is specific, can take up only one particular sub- stance, and, when it proliferates, is cast into the circulation as an antibody. Now when the plasm of the octopus or egg-albumin was injected into the rabbit, the first type of receptor, the non-specific type, was occupied largely or completely, and hence was not able to fix the plasm of the crab which was injected two and one-half hours later. Only the second type, the specific receptor, remained to fix the plasm of the crab. As a consequence the last-named substance disappeared from the circulation much more slowly than when it was injected into an untreated rabbit. We interpret the phenomenon which we have described in a somewhat similar way, although it is necessary for us to introduce the consideration that, the tissues of the mouse having united with the serum, egg-albumin, or broth, they thereby lose in their afl&nity for tetanus toxin. After the foreign substances have been dis- posed of by the cells, the latter again reach a state of equilibrium, and their tetanophile receptors regain their former affinity for the toxin (Table 8). It is not possible to assume, on the basis of our experiments, that an antitoxic or bactericidal serum can in any way intensify the corresponding disease, provided the serum is sufficiently rich in antibodies. In diphtheria, for example, the quantity of antitoxin introduced is able to neutrahze all the toxin which happens to be in the body, hence there is no possibility of intensifying the diph- theritic intoxication. THE STUDY OF "ROCKY MOUNTAIN SPOTTED FEVER" (TICK FEVER ?) BY MEANS OF ANIMAL INOCULATIONS.^ A PRELIMINARY COMMUNICATION. H. T. RiCKETTS. I arrived in Missoula, Montana, April 21, 1906, equipped for the bacteriologic and hematologic study of the so-called Rocky Mountain spotted fever and for the study of the infectious agent by means of animal inoculations. Although the period during which the disease occurs is very limited and the cases very few, it was possible to carry the investiga- tions to a point which affords encouragement for the ultimate solution of some ot the problems involved. At this time I wish to give a brief and preliminary presentation of the most important results obtained this year, reserving for a future date a more detailed interpretation of the experiments, when the bacteriologic and histologic studies will have been completed. It will be remembered that the confidence which was at first manifested in the report of Wilson and Chowning concerning the presence of a piroplasma in the blood and erythrocytes of infected patients was greatly shaken by the observations of Stiles, who, after prolonged study, was utterly unable to confirm the piroplas- matic nature of the disease. From these conflicting reports it seemed very probable that the specific organism, if present in the blood, must be one which is difficult of determination, and for the demonstration of which special experimental methods must be used. The first essential point to be studied concerned the situation of the specific agent in the body. Considering the subject from an unprejudiced point of view, it was necessary to recognize the possibility ot the microbe having a local situation, the general disturbance being caused by toxins which are absorbed from the point of primary infection. This possibility was considered some- ' This work was done in part under a grant made by the American Medical Association through the Committee on Scientific Research. From Jour. Am. Med. Assn., 1906, 47, p. a. 278 The Study of "Rocky Mountain Spotted Fever" 279 what improbable, however, the characteristic and generalized eruption speaking for a systemic rather than a local infection. On this account, my attention was directed chiefly to the blood from the start. After a fairly exhaustive search of the blood in stained and unstained preparations in two cases, I was unable to identify a parasite positively either within or without the erythrocytes. Similarly, aerobic and anaerobic cultures of the blood, made in bouillon and on agar plates of differing reactions, gave negative results with one exception. In the one exception mentioned, a staphylococcus grew in a flask of bouillon which had been inocu- lated with from 3 to 5 c.c. of blood drawn from the patient, Landon. In view of these negative results, it was decided to proceed at once with animal inoculations, although the microscopic examina- tion of the blood and culture experiments were continued as new cases developed. experiments on rabbits. On April 27, two intravenous inoculations were made into rabbits. In one instance, i c.c. of defibrinated blood from the patient, Stevenson, was used, and in the other 4 c.c. of defibrinated blood from the patient, Landon. The Stevenson blood was 17 hours old and the Landon blood 24 hours old at the time of injec- tion, the blood having been kept in the ice-chest except for the first few hours occupied in transporting it to the laboratory. The animals appeared not to be disturbed by the injections in so far as weight, temperature, and demeanor indicated, and nothing of a positive nature was learned from the experiments. On May 6, 5 c.c. of defibrinated blood, 22 hours old, from the patient, Cortsen, were injected intravenously into a rabbit weigh- ing about two kilos. No disturbance was noted for four or five days, when the temperature rose to 104.8°, but .remained above normal for no more than two or three days. The experiment was not considered satisfactory. experiments with guinea-pigs. Turning to the guinea-pig, a supply of which had been received in time for inoculations with the Cortsen blood, astonishing and 28o Contributions to Medical Science strikingly positive results were obtained. On May 5, when the Cortsen blood was six and one-half hours old, one guinea-pig was given 3 c.c. and another 5 c.c. intraperitoneally. One of the animals died in 7 days and the other in 1 1 days. It was necessary for me to be absent from Missoula for two or three days following the inoculations, but on returning I found in both a temperature which ranged between 104° and 106° F. After the fifth day both animals became emaciated rapidly, and for from 36 to 48 hours before death the temperature was subnormal. In guinea-pig i, the scrotum and testicles became enormously swollen, and on the sixth day the skin of the scrotum was occupied by large dark- colored hemorrhages of irregular outline. Subsequent experience showed that this sweUing of the testicles and scrotum and the hemorrhage into the skin of the scrotum is the most characteristic sign of the disease as produced in the male guinea-pig. Cultures of the Cortsen blood, and also of the heart's blood of the guinea- pigs after their death, remained free from discoverable growth. A second generation of three guinea-pigs was inoculated with the heart blood and splenic emulsion from the two animals which had died. Of the second generation one died, having the character- istic scrotal hemorrhage and swelling, and the other two recovered after a course of high fever and emaciation. From the fatal case inoculations were made into a third generation, the members of which did not become noticeably sick. In all fatal cases cultures of the blood on plates gave no visible growth. Although these experiments can leave no reasonable doubt as to the actual transference of the infection, further work seemed to be threatened by the inability to keep the infection going indefinitely in the guinea-pig. On May 20, blood from the patient, Porter, was inoculated into two male guinea-pigs, one of which died in seven and the other in eight days, both cases running t}^ical febrile courses, and showing the changes in the scrotum already mentioned. Cultures of the Porter blood and of the heart's blood of the fatal inoculations remained sterile. A second generation of three male guinea-pigs was injected from each of the fatal inoculations of the first generation. The two The Study of "Rocky Mountain Spotted Fever" 281 members of the second generation became very sick, running typical courses of fever which reached 105.7° and 105. 8°, but re- covery appeared to be complete on the tenth day in each case, at which time the temperature had returned to the normal. Here, too, attempts to perpetuate the infection by direct transference failed. On June 11 blood from the patient, Bradley, whom I discovered accidently near Stevensville, was utilized partly in repetition of previous experiments, but chiefly to determine the filterability of the virus and its distribution among the constituents of the blood. For this purpose a sufiicient quantity of the blood was centrifugated for 45 minutes at an estimated speed of 1,500 revolutions. The serum was drawn off and 6 c.c. were injected intraperitoneally into a guinea-pig. An additional amount of 6 c.c. was passed through a small Berkefeld filter under low pres- sure, the total amount after the filter was washed out with salt solution being between 10 and 12 c.c. This amount was con- sidered too large for a single injection, hence it was given to an- other guinea-pig in two injections separated by an interval of two hours. The erythrocytes, and, of course, leukocytes, which remained after centrifugation, were now washed three times with sterile salt solution, raised to the original volume by means of the same solution, and 6 or 7 c.c. injected intraperitoneally into another guinea-pig. To complete this experiment, the unaltered defi- brinated blood was injected intraperitoneally into another guinea- pig. Without giving details as to the course of the disease in these animals, the results were, briefly, as follows: Fresh defibrinated blood, washed corpuscles, and unfiltered serum produced typical and fatal infections in the guinea-pigs inoculated with them, whereas the filtered serum had no discoverable effect, the animal at no time showing a temperature, even up to the present time. RESULTS IN EXPERIMENTS. The results of these experiments were somewhat surprising, since the failure to recognize the organism microscopically had suggested the probability that it was of exceedingly minute size and might be readily filterable. I would not assume that filtration is impossible, but simply state that it did not pass through the 282 Contributions to Medical Science filter under the condition of experiment. These experiments prove, then, that the condition produced in the guinea-pig is an infection and not an intoxication, for otherwise the filtered serum undoubtedly would be as toxic as the unfiltered. In view of the infectiousness of the serum they also prove that the infecting agent is far from being an exclusive corpuscular parasite; and they suggest very strongly that the organism is of such size that its ultimate and positive recognition with the microscope may be hoped for reasonably. It is recognized that these results do not agree positively or decisively against the piroplasma theory, since piroplasmata may well be in the serum as in the erythrocytes. But they do seem to throw some doubt on this theory, since the serum was almost as virulent as the erythrocytes. On the other hand, it cannot be argued that because the washed corpuscles were highly infectious that the erythrocytes actually are infected. One could as readily assume that the organisms were in the leuko- cytes, or that they were entirely extracellular, a certain proportion of them having been thrown down in the centrifuge together with the corpuscles. Another important point brought out by these inoculations with the Bradley blood is the fact that the guinea-pig, at least in some instances, suffers from a more or less extensive macular and confluent eruption which is in addition to the scrotal hemor- rhage. This had been missed in the earlier experiments, partly because of its unsuspected distribution. After observing that such an eruption occurred chiefly on the dorsal aspect of a monkey inoculated by Dr. King, being entirely absent from the abdomen and chest, I shaved the entire skin of a guinea-pig which had died of the disease, and found the dorsal and lateral aspects of the body, face, and extremities marked by many reddish macules, whereas over the face and buttocks there were confluent hemor- rhages. Subsequent observation on other guinea-pigs shows that this eruption is of frequent, though perhaps not constant, occurrence. experiments on monkeys. I ordered monkeys as soon as I had decided to begin animal inoculations. The season of the disease being short and the The Study of "Rocky Mountain Spotted Fever" 283 situation more or less critical, it was thought best not to jeopardize progress by neglecting the use of an animal so closely related to man. After unfortunate delays and misunderstandings these animals reached me in time for inoculation with the blood of the patient, Porter. On May 20 a healthy and fairly large Rhesus, weighing about three kilos, received intravenously 10 c.c. of defibrinated blood from the patient, Porter. The blood at this time was about 28 hours old, this period representing the time required to transport it from Hamilton, 50 miles away. During this time the blood had been kept at outdoor temperature, about 15° C. On the evening of the second day the temperature of the animal rose to 104.6° by rectum, and from this time until the tenth day it ranged from 103.1° to 105.3°. Ori the tenth day it fell rather suddenly to loi . 1°, and the animal gradually regained his former healthy condition. From the fourth to the eighth day he appeared very sick, was extremely weak, ate very little, the face was cyanotic and the conjunctivae were infected. The cyanosis and injected conjunctivae are recognized as characteristic symptoms of spotted fever as the disease occurs in man. The duration of the sickness also approximated that seen in man. The incubation period was shorter, however, as in the guinea-pig. When the disease is fatal in man, death usually occurs on from the seventh to the ninth day; and, although patients occasionally die after having weathered 14 or 15 days, one who survives the tenth is given a guarded favorable prognosis. The most character- istic sign of the disease as it is seen in man, i.e., the generalized reddish macular eruption, was not observed in this animal. Dur- ing his sickness the eruption was sought for principally on the chest, abdomen, arms, and legs, where there was the least hair, and the skin in some places was of a transparent pink color. The absence of the eruption does not, under the circumstances, render the diagnosis of spotted fever untenable. Those who have had the greatest experience with the disease in man recognize a mild type in which the characteristic eruption is absent. In such cases, the diagnosis rests chiefly on the sluggish circulation, general cyanosis, reddened conjunctivae, drowsiness, and lack of complaint on the 284 Contributions to Medical Science part of the patient in spite of retained intelligence, the course and duration of the fever, slow convalescence, and the season of the year in which the disease occurs. Furthermore, those who are familiar with experimental inoculations in animals have long since recognized the fact that in many diseases, desirable as the result is, one may not be able to obtain in an animal a complete duplication of all the symptoms seen in man, in spite of exquisite susceptibility on the part of the animal. The receptor theory of Ehrlich has rendered such variations altogether intelligible. My second inoculation of the monkey was made on June 11 with the defibrinated blood of the patient, Bradley. This patient was discovered by the merest accident, on June 9, near Stevensville, where I had gone for the purpose of exploring an infected district with Dr. Fessler. On the following day from 60 to 70 c.c. of blood were drawn from a vein of the arm by means of sterile appara- tus kindly loaned by Dr. Brice, the attending physician. The patient was in the tenth day of the disease. After defibrination the blood was placed on ice until the following day, when it was possible to return to Missoula. The animal inoculated was a small female Rhesus, weighing less than two kilos and apparently perfectly healthy. Eight c.c. of the blood were injected intraperitoneally. The animal ran a course of fever similar to that of the first monkey, but died on the ninth day, after a short period of subnormal temperature. In addition to the cyanosis of the face and ears, noted in the first monkey, the hairless skin of the perineum became brilliantly red, although at no time were hemorrhages detectable. The skin of the abdomen, chest, arms, and legs remained free from spots or hemorrhages. No cause of death, other than spotted fever, was found at the autopsy, no peritonitis had followed the injection, and a plate culture of the heart's blood showed no visible growth after five days of incubation. A second monkey and three guinea-pigs were inoculated with the spleen and blood of the dead animal as soon as possible after death. One guinea-pig died of peritonitis following perforation of the bowel with the needle. The other two are at present running temperatures which correspond to that seen in previous inocula- The Study of "Rocky Mountain Spotted Fever" 285 tions of the guinea-pig, and in one the scrotum shows the character- istic enlargement. The monkey of the second generation has a temperature which has gone as high as 106 . 7°, the perineum shows the redness seen in the monkey of the first generation, and the hemorrhagic eruption has appeared on the scrotum, buttocks, legs, back, etc. The appearance of these phenomena in the second generation of animals leaves no doubt as to the success of the inoculation with the blood of the patient, Bradley. I may refer here to a most gratifying result obtained by Passed Assistant-Surgeon W. W. King, of the Public Health and Marine Hospital Service. Up to the time of the inoculations with the Bradley blood. Dr. King had confined himself to the microscopic study of the blood. He had been no more successful, however, in this search than I, and, having obtained a monkey and some guinea-pigs, I readily assented to his proposal that he repeat some of my experiments with the hope of confirming them. For this purpose I shared with him the blood of the patient, Bradley. We drew lots for the route of inoculation, the subcutaneous falling to Dr. King and the intraperitoneal to me, as stated above. The monkey inoculated by Dr. King died two days later than mine and showed a marked hemorrhagic condition of the skin of the scrotum and penis and the flaming erythema of the perineum seen in my monkeys, together with a mixed macular and confluent eruption, hemorrhagic in character, which was distributed chiefly over the external aspects of the arms, legs, buttocks, and back. The face and ears were extremely cyanotic. No exanthem could be discovered on the chest and abdomen. A guinea-pig also inoculated by Dr. King with the Bradley blood ran a typical course, presenting the enlarged and hemorrhagic scrotum, the exanthem already observed by me, and in other respects confirm- ing the results of my inoculations. White rats and mice arrived so late that they could be used in but one case; the disease was not reproduced in them. SUMMARY. The essential anatomic findings in the guinea-pig at autopsy consist of the cutaneous phenomena described, the enlarged and 286 Contributions to Medical Science hemorrhagic scrotum in the males and some swelling of the testicles with pronounced congestion of the epididymis, retention of urine, distension of the seminal vesicles, congestion of the kidneys and suprarenals, swelling and congestion of the spleen and liver, and a right heart and venous system which are enormously engorged with blood. No meningitis nor localized inflammations have been observed, except in two instances in which the skin of the swollen scrotum had become gangrenous, with a consequent staphylococcus infection of the underlying cellular tissue. In the one monkey which has gone to autopsy so far, nothing was found macroscopically, in addition to the cutaneous phenomena, except a congestion of the parenchymatous organs. That spotted fever has been transmitted to the guinea-pig in my experiments is shown by the following data: The fever, dura- tion, and cutaneous phenomena resemble very closely these condi- tions as seen in man. It can be transmitted into the third genera- tion by direct inoculation from one animal to another, but has not been carried beyond this point up to the present. The condition produced has been an actual infection rather than a transferred intoxication, because it can be passed through the second genera- tion of animals, and because the filtered serum which would cer- tainly contain any soluble toxins which might be present causes no disturbance. In further confirmation of the statements just made are the facts that no other cause of death has been found, and that it has not been possible to cultivate any micro-organism from the heart's blood in fatal cases. For similar reasons it is believed that the transmission to the monkey cannot be disputed. I realize that the material on which these conclusions are based is rather scant, but the experiments have been rigorously con- trolled, and I feel that they are safe. The hope is still entertained that it may be possible to keep the infection alive in experiment animals by suitable manipula- tions in order that many other problems bearing on the disease may be taken up properly. I am at present trying to do this by alternating the infection between the monkey and the guinea-pig, with the hope that the character of the \-irus may be so altered The Study of "Rocky Mountain Spotted Fever" 287 that the infection may be passed directly from one guinea-pig to the other indefinitely. With this result in hand attempts toward the production of a vaccine may reasonably be made. The discussion of other features of the disease will be reserved for the present. I shall only refer to the theory of transmission by means of the wood-tick, by expressing my disagreement with the conclusion of Stiles that the tick theory falls with the piro- plasma theory. The tropical "tick fever" goes to show that ticks may harbor and transmit pathogenic parasites other than piroplasmata. For the privileges of the laboratory of the Northern Pacific Hospital and of the hospital itself, I am under the greatest obliga- tion to Dr. W. E. Spottswood, surgeon-in-chief, without whose active interest much of the work of this season could not have been undertaken. I also express my thanks to Dr. Chowning, since I profited greatly by his previous experience in the field, and to Dr. King, who, after the departure of Dr. Chowning, gave me the most generous assistance in the inoculation of the animals. The contributions of the Montana State Board of Health and of Missoula and Ravalli counties, through their commissioners, are gratefully acknowledged. THE TRANSMISSION OF ROCKY MOUNTAIN SPOTTED FEVER BY THE BITE OF THE WOOD-TICK (DERMACENTOR OCCIDENTALIS).^ H. T. RiCKETTS. In The Journal, July 7 (see p. 278 of this volume), I reported briefly my success in transmitting Rocky Mountain spotted fever to the guinea-pig and monkey. With these results in hand, it at once became apparent that a means had been provided by which to test the theory of transmission by the wood-tick, a theory which was primarily advanced by Wilson and Chowning. It was my first aim to establish, beyond the opportunity of doubt, the susceptibility of the animals mentioned, and it was only after this susceptibiUty had been determined by inocula- tion from two different patients that the study of the tick as an intermediate host for the parasite was taken up. On June 19 a small female tick was placed at the base of the ear of "Bradley" guinea-pig i. This guinea-pig had been inocu- lated intraperitoneally on June 11 with three cubic cm. of defi- brinated blood from the patient, Bradley, and died with character- istic symptoms on June 23. After having fed on the infected guinea-pig for two days, the tick was removed, placed in a venti- lated pill-box for two days more, and on June 23 was attached to the base of the ear of a female guinea-pig weighing 300 gms. After an incubation period of three and one-half days, the temperature of the animal rose to 104.2, gradually ascended to 106.4, near which point it remained for seven days, when it gradually returned to normal. The bite of the tick in this instance seemed to be very virulent, the whole ear becoming very much swollen, indurated and cyanotic, and a slough almost one-half inch in diameter, which developed in the vicinity of the bite, eventually separated. ' This work has been done in part under a grant made by the American Medical Association through the Committee on Scientific Research. From Jour. Am. Med. Assn., 1906, 47, p. 358. 288 July I, A.M. 104.6. July 2, A.M. 105.2. July 3, A.M., 104.7; P.M., lOS- July 4, A.M., 104.7; P.M., 103.1 July 5, A.M., 103.8; P.M., 103.9 July 6, A.M., 103 4; P.M., 103 S July 7, A.M., 102.5. Recovery. Transmission of Spotted Fever by Wood-Tick 289 The animal exhibited the following course of fever: June 23, A.M., 102.3; P-^-, 102.2. June 24, A.M., loi .Q. June 25, A.M., 102.8; P.M., 103.9 June 26, A.M., 102.4; P.M., 103.6 June 27, A.M., 102.5; P-^-, 104.2 June 28, A.M., 106.4; P-M., 105.2 June 29, A.M., 105.6; P.M., 105.8 June 30, A.M., 104.8; P.M., 106. 1 Seven days after inoculation, the external genitalia became swollen and congested, and this condition became more marked until the twelfth day, when it began to subside. At no time were distinct hemorrhages in the genitalia or other parts of the skin to be seen. The course of the disease was exactly similar to that seen in a number of other guinea-pigs which ran a severe course, with eventual recovery. The absence of a hemorrhagic condition in the skin, or of a discoverable roseolar eruption, throws no doubt on the success of the transmission, since, as shown by other experi- ments, a perceptible eruption does not appear invariably in spotted fever as it is produced in the guinea-pig experimentally. It was not possible to attempt the transmission of the disease from this animal to others, because of the lack of guinea-pigs at the time. As controls I have two experiments in which uninfected ticks fed on normal guinea-pigs without causing a rise in temperature; furthermore, two guinea-pigs which lived for two weeks in the box occupied by the infected animal showed no abnormal tem- perature, a fact which argues against the transmission of the infec- tion by mere association with excretions of infected animals. The result of this experiment brought very forcibly to my mind the probable part which the tick plays in the infection of man and shows the necessity of repetition of the work with a more abundant material. In view of the result which I had obtained I was not surprised to note the recent report of Dr. King, of the U.S. Public Health and Marine Hospital Service, who, starting with material which I had given him, accomplished transmission in the same manner. Hasty conclusions as to the question of tick transmission in 290 Contributions to Medical Science relation to the infection of man are, by all means, to be avoided until such a time as the experiments can be repeated and the life- history of the infection worked out more thoroughly. This phase of the subject, in common with others, is being studied by me, the infection still being maintained in animals for these purposes. It is hoped that some questions may be settled satisfactorily before the advent of the disease next year. At all events the knowledge so far gained may well be taken into account in institut- ing prophylactic measures against the disease. FURTHER OBSERVATIONS ON ROCKY MOUNTAIN SPOTTED FEVER AND DERMACENTOR OCCIDENTALIS.^ H. T. RlCKETTS. (From the Pathological Laboratory oj the University of Chicago.) In view of the brief season during which Rocky Mountain spotted fever prevails, uninterrupted study of the disease is con- ditioned on the maintenance of the infection in experiment animals. In a previous communication^ (p. 278) my inability to preserve the disease by inoculation from guinea-pig to guinea-pig was referred to, and the hope was expressed that alternation of the infection between the monkey and the guinea-pig might result in the preser- vation of the virus. This experiment has been entirely successful and Table i illus- trates the life history of the virus since it was obtained from the patient, Bradley. TABLE I. Number of Animal Date Source, and Material Inoculated Monkey I June 11 Patient, Bradley; defibrinated blood Monkey II June 20 Monkey I; defibrinated blood Guinea-pig 12 June 27 Monkey II; fresh heart's blood Monkey IV July 7 Guinea-pig 16; blood and organs Guinea-pig 16 July 14 Monkey IV; defibrinated blood Monkey V July 22 Guinea-pig 16; blood and organs Guinea-pig 17 July 28 Monkey V; defibrinated blood Monkey VI August 3 Guinea-pig 17; blood and spleen Guinea-pig 2 s August 16 Monkey VI; blood and serum Monkey VIII August 23 Guinea-pig 25; blood and organs Guinea-pig 36 August 30 Monkey VIII; spleen Monkey IX September 6 Guinea-pig 36; blood and organs All infected animals have run identical courses of fever, which are typical in this respect, that an incubation period of from two to four days intervenes between the time of inoculation and the onset of the regular febrile course. Commonly a moderate rise ■ This work has been done under a grant made by the American Medical Association through the Committee on Scientific Research, and with the aic of Missoula and Ravalli counties, Montana, and of the Montana State Board of Health. From Jour. Am. Med. Assn., IQ06, 47, p. 1067. ' "The Study of Rocky Mountain Spotted Fever (Tick Fever?) by Means of Animal Inoculations," ibid., 1906, 47. P- a- 291 292 Contributions to Medical Science in temperature follows immediately on the inoculation and persists for one or two days, then falls near to normal for one or two days more, when the customary course sets in. This initial or primary fever probably is caused by intoxication from the large amount of material which is frequently injected. When a smaller amount of blood or emulsion of tissue is used this initial fever has not appeared. Several facts are relied on in concluding that the disease trans- ferred by these inoculations is spotted fever rather than some other infection which may have crept in during repeated passage. In harmony with the experience which was met in making cultures from man, no growth is obtained from the blood of infected ani- mals, provided the cultures are made at a time when agonal or postmortem invasion by common bacteria has not taken place. If plates are inoculated with the heart's blood at the instant of death, it is customary to obtain no visible growth on ordinary media, yet this same blood proves infectious for other animals. Again, the rather frequent development of a hemorrhagic eruption in the monkey or guinea-pig, a manifestation which is identical in appearance, if not in distribution, with that seen in man, speaks for spotted fever rather than some other infection. This eruption is much more constant in the monkey than in the guinea-pig. In both animals it begins on the scrotum and foreskin. In the guinea- pig it may be found on no other part of the body, although it can be seen as a more or less generalized eruption occasionally. The external genitals of the female guinea-pig commonly become swollen and congested; recently, however, extensive hemorrhages have occurred in the skin of the labia in a number of females. But one female monkey has been used, and in her no cutaneous hemorrhages were discovered. It will readily be understood that the detection of a pink roseolar eruption in either the monkey or the guinea-pig is difficult. In the guinea-pig it can be recognized with certainty only in white-skinned animals. In both animals the hemorrhagic eruption, when it appears, is a late manifestation and it would seem to be related in some way to the virulence and duration of the infection. The disease may be so virulent that the animal dies before the eruption would appear, or so mild that it Spotted Fever and Dermacentor Occidentalis 293 does not occur. The appearance of the hemorrhages does not necessarily presage a fatal issue ; two monkeys and several guinea- pigs which had extensive hemorrhages into the skin have recovered. A third fact which stamps this experimental disease as spotted fever consists of the identity of the course of fever with that pro- duced in the animals by direct inoculation with human blood. The essential anatomic changes — enlarged spleen and lymph glands, congested kidneys, etc. — are identical with those seen in the animals first inoculated, and correspond with the changes found in man. There seems to be no particular difficulty involved in perpetu- ating the infection, except in the matter of expense. Duplicate inoculations should be made and the blood or organs used for inoculation may be taken at any time after the fever has reached its high point. To await the death of the animal before making the transfer is to risk the loss of the culture ; the animal may recover. THE DISTRIBUTION OF THE VIRUS IN THE BODY OF THE INFECTED ANIMAL. In my first experiments the infectiousness of the blood, in spotted fever, was demonstrated. It would naturally be inferred that the various organs are infectious at least to the degree in which they contain blood. On this account one would reasonably expect the spleen to be infectious and this has proven true in a number of inoculations. The testicles have been used for inocu- lation a number of times. It was thought that the testicles might be a point of predilection for the virus, in view of the severe hemor- rhages which often occur in close proximity to them, i.e., in the scrotum. They were shown to be infectious in several instances, yet their virulence was not so great as to indicate any special con- centration of the virus at this point. The scrotal hemorrhages may have no relation to the amount of infection of the testicles. The distribution of the virus among the organs of an infected monkey (VIII) was recently determined. The animal was exsan- guinated from the carotid to deprive the organs of as much blood as possible. Each organ was then ground with sterile sand and suspended in physiologic salt solution. Identical weights of the 294 Contributions to Medical Science different organs were not used, but approximately the same mass was employed in each case. A further attempt at uniformity was made by rendering the different emulsions of approximately the same density. Five cubic centimeters of each emulsion were injected intraperitoneally into guinea-pigs, which varied in weight from 350 to 550 grams. Table 2 shows the results. TABLE 2. The Infectiousness of Different Organs. Number Matter Inoculated Result 35 36 37 38 39 40 41 42 Mesenteric, inguinal and axillary lymph glands Spleen Bone-marrow Testicles Liver Kidney Medulla oblongata Cerebral cortex and adjacent white mat- ter Death in i8 days Death in 7 days Death in 6 days Death in 15 days Death in 8 days Death in 7J days Recovered Death in i2i days In the fatal cases cultures of the heart's blood were sterile except in the case of Guinea-pig 37, in which a few colonies of a staphylococcus developed. The technical difficulties in preparing the bone-marrow increased the opportunities for infection. It is thought very probable, however, that the animal died of spotted fever because of the duration of the course. Those organs seemed to be the most infectious which contained, naturally, the most blood, i.e., the liver, spleen, and bone-mar- row (?). In these organs exsanguination still leaves rather large quantities of blood. The kidney, however, was fully as infectious as the organs which were much richer in blood. It is possible that the kidney becomes the repository of a larger mass of micro-organ- isms, by virtue of its excretory function. It is realized that this experiment does not give an exact measurement of the degree of infection of the different organs, and that the results, at the best, can be considered no more than approximate. In another experiment an attempt was made to determine the minimum fatal dose of the blood of the infected monkey, the blood being taken six days after inoculation. The injections were intra- peritoneal (Table 3). Spotted Fever and Dermacentor Occidentalis 295 Plate cultures of the heart's blood were sterile in all cases. In this experiment the minimum fatal dose of the blood of the infected monkey on the sixth day of the disease was in the vicinity of o . 5 c.c. Guinea-pig 30, which received this dose, died on the four- TABLE 3. Virulence of the Blood of the Infected MoNKj;y. Result Death in 14 days Death in 7i days Death in 7j days Death in i3i days Death in 6^ days teenth day, whereas animals which live so long ordinarily recover. Guinea-pig 32, which received 2.0 c.c, showed unusual resistance. The guinea-pigs weighed about 350 grams each. Other experi- ments indicate that subcutaneous and intravascular inoculations have a similar degree of virulence. ACTIVE IMMUNITY. I have not learned of an authenic example of a second attack of spotted fever in man, hence it has seemed somewhat probable that one attack renders the body immune. This has been found to be true in a recent experiment on the monkey. On June 20, 1906, Monkey II was inoculated with the blood of Monkey I, the latter having just died as the result of inoculation with the blood of the patient, Bradley. Monkey II ran a typical course of fever, showed a macular eruption of the perineal skin on the eighth day, and extensive hemorrhages of the scrotum, perineum, back, and extremities on the eleventh day. The tem- perature returned to normal on the twelfth or thirteenth day, hence the eruption became hemorrhagic as convalescence was about to begin. The animal regained its former health and activity. Recently this animal was again inoculated with the blood and tissues of an infected guinea-pig (No. 36). The total mass of fluid injected was 8 c.c. With the exception of some indisposition on the following day, the injection produced no apparent effect. 296 Contributions to Medical Science For a few days the temperature was slightly higher than before, but during this time and since the monkey has shown no indis- position and has been as active as before inoculation. By way of control another monkey of the same size (No. IX) was inoculated with the same amount of material, the same quan- tity, and by the same route. Between the third and fourth days the temperature began to rise, and, until it was killed (7 days), it ranged between 105. i and 106.3. The conjunctivae became reddened, the animal shivered at frequent intervals and sat, inat- tentive and with head drooping, continuously. On the fifth day the supra-orbital skin showed reddened patches, and on the sixth day a roseolar eruption was visible on the scrotum. The blood and tissues were used for further experiments. From this experiment it is evident that one attack in the monkey confers an active immunity, and on the basis of reasonable analogies one may conclude that man probably acquires a similar immunity. A similar acquired immunity has also been demonstrated in the guinea-pig. Its duration, beyond the period of two months for the monkey and three months for the guinea-pig, has not been determined. THE LIFE history OF DERMACENTOR OCCIDENTALIS. On account of the suspicion which attaches to Dermacentor occidentalis as the carrier, or a carrier, of Rocky Mountain spotted fever, its Hfe history is important. The following brief notes concerning this question are given by way of recording the changes which were observed under artificial conditions. No description of minute morphologic changes will be given at this time. May 30, 1906, a fully developed female was placed in a cotton- stoppered test-tube. June 21 (about) the tick began laying eggs, of which several hundred (estimated) were eventually deposited. June 28 or 29 hatching began and continued for a week or ten days, resulting in six-legged larvae of minute size. July 13, the tube with its contents was placed in a tick-proof cage with a guinea-pig. The larvae took hold readily for the most part, but did not feed continuously. They increased slightly Spotted Fever and Dermacentor Occidentalis 297 in size. Eventually all remained attached to the skin for several successive days, grew to a length of i to ij mm., and became swollen and of a uniform dark slate color. On August — they began to drop from the guinea-pig, moved to a safe place, and went into a quiescent state, appearing as if dead. August 5 they began to molt, throwing off a snow-white pellicle, and came out as eight-legged nymphs. After 7 to 10 days the molting of the brood was almost complete. When they first appeared they approximated the size of the well-developed larvae. When again placed with a guinea-pig or monkey they seemed to feed intermittently for a time, but eventually attached them- selves firmly and underwent a great increase in size, some reaching a length of about 3/16 of an inch. With the increase in size, the color again changed to that of a uniform gray-brown or slate color. It required from 6 to 10 days for this development to take place after the ticks were placed with an animal. On August 30 prac- tically all the ticks which remained alive had undergone this change. As they reached the condition mentioned they were removed and placed in ventilated boxes where they went almost immediately into a second quiescent stage. In about 10 days molting again began, a white mantle being thrown off, and the mites emerged as adult ticks, with the marking which one finds under natural conditions in the early months in Montana. The experiment was carried on entirely at room temperature, and the various stages may not correspond in point of time with those taking place under natural conditions. Under natural conditions, in the Bitter Root Valley, the adult ticks begin to disappear during the latter part of June. This probably means that the males, and those females which have not found a host, die, whereas the fully developed females deposit their eggs. Locally, knowledge of the further history of the tick was limited, or nil. It is certain, however, that during the warm summer days, the conditions are favorable for the hatching of the eggs, and for the development of the larvae, provided a host is reached. Inasmuch as a crop of adult ticks is not observed in the fall, it is possible that the second molting stage, mentioned above, 298 Contributions to Medical Science represents the hibernating stage of the mite. It is well known, locally, that the first few warm days of the spring bring out the ticks. These few warm days apparently would be sufficient to bring to completion the second molting stage. On the other hand, it must be admitted as possible that the second molting stage is completed in the late fall, and that hibernation begins at once with the animal fully developed, the first few warm days of spring being sufficient to arouse him. Observations of a more complete nature will be made at a future date. OBSERVATIONS ON THE VIRUS AND MEANS OF TRANS- MISSION OF ROCKY MOUNTAIN SPOTTED FEVER.' H. T. RiCKETTS. {From the Pathological Laboratory of the University of Chicago.) Investigations into the nature of Rocky Mountain spotted fever, its etiology, and method of transmission, have been carried on by the writer since April, 1906. Previous to these investigations the disease had been studied extensively by Wilson and Chowning^ and to some degree also by members of the Pubhc Health and Marine Hospital Service. I shall refer to these investigations only to say that Wilson and Chowning described the disease as a pyroplasmosis, and advanced the important theory that man is infected by the bite of the ''wood- tick" which infests the mountainous regions of Montana and adjacent Rocky Mountain states. They furnished no experimental proof of the correctness of the theory. Following the report of Wilson and Chowning, Stiles,^ of the Public Health and Marine Hospital Service, studied the disease and failed utterly to find the pyroplasma of the former investigators. Stiles discredited also the theory of transmission by the tick, but without experimental evidence to refute the theory. This was the status in relation to the etiology and means of transmission of the disease when my studies were undertaken. The results which I have obtained have been described briefly in three communications to the Journal of the American Medical Association (see pp. 278, 288, and 291 of this volume)."* The essential points presented in these articles are the following : The disease was transmitted to guinea-pigs from three different ■ Jour. Infect. Dis., 1907, 4, p. 141. This work has received special aid from the Research Fund of the American Medical Association, from Missoula and Ravalli counties, Montana, from the Montana State Board of Health, from the University of Chicago, and the Memorial Institute for Infectious Diseases. 'Jour. Am. Med. Assn., 1902, 39, p. 131; also Jour. Infect. Dis., 1904, r, p. 31. > Pub. Health and Mar. IIosp. Bull., No. 20, 1903. * (i) "The Study of 'Rocky Mountain Spotted Fever' (Tick Fever) by Means of Animal Inocula- tions," /o«r. /I w. A/eJ. Assn., 1906, 47, p. 33; (2) "The Transmission of Rocky Mountain Spotted Fever by the Bite of the Wood-Tick (Dermacentor occidentalis)," ibid., 1906, 47, p. 358; (3) "Further Observations on Rocky Mountain Spotted Fever and Dermacentor occidentalis," ibid., 1906, 47, p. 1067. 299 300 Contributions to Medical Science cases, and to monkeys from two different cases by inoculation with defibrinated blood from the patients (i, 3). From the last of the three cases the disease has been maintained in the laboratory by alternate passage through the monkey and the guinea-pig, by means of inoculations with blood or emulsions of organs (3). The first attempts to maintain the infection by direct inoculation from guinea-pig to guinea-pig failed, possibly for reasons which will be referred to below. Rabbits were found not to be susceptible to an appreciable degree, and the same was true of white mice and white rats. In a preliminary experiment the virus did not pass through a Berkefeld filter, although the unfiltered serum was infectious. The anatomical similarity of the disease produced in the guinea-pig and monkey to the conditions observed in man and the incubation period and course of fever induced by inocula- tion were considered sufficient evidence of the genuine transmis- sion of the disease to these animals (1,3). In a single experiment a tick, female, was found to be the carrier of the disease from one guinea-pig to another. This result was considered as tenta- tive (2).' The virus was found to be distributed in all of the visceral organs, the most vascular organs (liver, spleen, and bone-marrow) and the kidney apparently being the most infectious (3). It was found that an active immunity, which possibly is relative in degree, is established in the monkey and guinea-pig by one attack (3). The life-history of Dermacentor occidentalis was followed under laboratory conditions. This tick passes through a larval stage and moults twice subsequently before reaching the adult form, the cycle from egg to adult requiring about three months under the modified conditions, which prevailed (3). The tick left the host in order to moult. It is the purpose of this paper to present further observations which have been made in relation to the distribution of the virus in the body fluids, its viability and resistance under different conditions, its filterability and certain other properties, the possi- bility of transmission by means of the bite of the male tick, and ■ King also has reported transmission by means of the female tick (Pub. Health Reports, July 37, 1006). Means of Transmission of Spotted Fever 301 in addition, the preservation of the virus by uninterrupted passage through the guinea-pig. the distribution of the virus in the blood. Rocky Mountain spotted fever is unquestionably a systemic in- fection, since the disease may be transmitted by the inoculation with blood and with emulsions of the solid organs. A study of the distribution of the virus in the blood was under- taken with the hope that the question of an exclusive or predomi- nant erythrocytic infection, such as a pyroplasmosis, might be affirmatively or negatively determined. Although the experi- ments reported are not quantitative in character, it is thought that the sum total of results renders the pyroplasma theory open to suspicion. Quantitative work is very difficult from the fact that the concentration of the virus in different animals is subject to considerable variation; this may be as great as a tenfold varia- tion, the minimum fatal dose of defibrinated blood in one instance being 0.05 c.c. and in another 0.5 c.c. The following results have been obtained in regard to this phase of the work: 1. It is impossible to free the blood cells of defibrinated blood from the virus by 10 to 12 washings with physiologic salt solution (Table 3). Repeated washing, however, seems to decrease the virulence of the corpuscles of infected blood (Table 7). 2. As washing advances a point is reached at which the virus is not separated from the corpuscles in infective quantities by shaking with salt solution. An experiment showed that a fluid of the loth washing was not infectious in quantities of 5 and 10 c.c, whereas the cellular sediment caused the disease in doses of 0.7 and 1.5 c.c. (Tables 3 and 4). 3. The serum obtained from defibrinated blood or from spon- taneous clotting is infectious in doses of 0.5 c.c. or even less (Table 5). 4. Prolonged centrifugation of the serum (6 hours) does not free the overlying portion from the virus (Table 8). 5. An exudate of leukocytes in the pleural or peritoneal cavity of the infected monkey or guinea-pig, caused by the injection of 302 Contributions to Medical Science aleuronat or bouillon, is infectious for the guinea-pig. The sedi- ment of such an exudate is not deprived of its infectiousness by repeated washings, and the overlying fluid remains infectious in spite of prolonged centrifugation (Table 6). The following selected experiments are given as illustrative of the preceding statements: TABLE I. Control Experiment. Inocuiations with Fresh Undefibrinated Blood of Monkey XIV. Guinea-Pig No. Dose Incubation Period Duration from Date of Inoculation Result Autopsy ISO 0.05 c.c. 0.1 0.3 0.7 3-4 days 3-4 3-4 3-4 12 days loj II 9 Death Typical* " " • The following conditions at autopsy are recognized as typical for spotted fever in the guinea-pig: Generalized enlargement of the lymph glands including those of the mesentery and mediastinum. Either extreme congestion of, or hemorrhage into, the lymph glands. Enlargement of the spleen, which may reach the size of two or four times the mass of the normal spleen. The spleen is much congested and cyanotic, fairly homogeneous and of moderate consistency, i.e., neither soft nor hard. The kidneys, suprarenal glands, and liver are somewhat enlarged and congested. The right heart and veins are heavily engorged. The lungs show no changes, The meninges are moderately reddened. The bone-marrow is rich in rather dark red blood. \"ery frequent, but not constant, changes are hemorrhages into the skin of the external genitalia, and, in males, into the testicles and their coverings. Hemorrhage also occurs with some frequency into the suprarenal gland, and less often, into the liver. If visible colonies appear on a plate of the heart's blood which is made soon after death, or if peritonitis or other severe local inflammation is found, the animal has been disregarded. TABLE 2. Control Experiment. Inoculations with Fresh Defibrinated Blood of Monkey XIV. Guinea-Pig No. Dose Incubation Period Duration from Date of Inoculation Result Autopsy 154 0.05 c.c. 0.1 0.3 0.7 S-6 days 5-6 3-4 3-4 13 days II 12 II Death Typical " 156 " " TABLE 3. Inoculations with Washed Defibrinated Blood of Monkey XIV: Ten Washings, the Original Volume Being Restored. Guinea-Pig No. Dose Incubation Period Duration from Date of Inoculation Result Autopsy 162* 0.3 c.c. 0.7 I.O IS 5 days S-6 8 days 13 Not sick KiUed Not sick Death 163 Typical 164 165 :.. " Not sick. Died later, following rectal prolapse. Means of Transmission of Spotted Fever 303 TABLE 4- Inoculation witb Overlying Salt Solution of the Tenth Washing of Blood Used in Table 3. No. Dose Result 167 lOC.C. Not Sick TABLE 5- Inoculations with Serum from Blood of Monkey XIV, the Serum Being Obtained by Defibrina- tion AND CeNTRIFUGATION FOR OnE HoUR. Guinea-Pig No. Dose Incubation Period Duration from Date of Inoculation Result Autopsy IS8 159 i6o* 161 1 .0 2 .0 30 4.0 S-4 days 3-4 3-4 11 days 13 14 Death Typical * Accidental death in 4 days. TABLE 6. Infectious Properties of Leukocytic Exudate from Monkey XIV, Caused by an Injection of Aleuronat. Guinea- Pig No. Dose and Material Inoculated Incubation Period Duration from Date of Inocula- tion Result Autopsy t47 148 149 2 c.c. of unaltered exudate 5 c.c. of leukocj^e suspen- sion, washed 3 times 4J c.c. of fluid from first centrifugation 6 days 6 &-7 13 days 14 21 Death Typical Recovery TABLE 7. The Effect of Repeated Washing of Blood in Decreasing Its Virulence. Mixed Bloods of Monkeys VII and VIII, 12 c.c. Injected into Each Guinea-Pig. Guinea-Pig No. Washings Incubation Period Duration from Date of Inoculation Result Autopsy 44 I 2 S 10 2 days 4 5 5 7 days 10 14 16 Death Typical ta::::::::::: 47 '• TABLE 8. Attempt to Remove Infectious Properties of Serum by Centrifugation for Six Hours at the Rate of 2,000 Revolutions per Minute, s c.c. of Serum from Monkeys VII and VIII. Guinea-Pig No. Injection 50 Centrifugation serum SI Final o. 5 c.c. of above Incubation Period 7-8 days S Duration 20 days 16 Result Recovery 304 Contributions to Medical Science The tenacity with which the virus associates itself with the blood corpuscles suggests that a certain proportion of the parasites may be within either the leukocytes or the erythrocytes. This suggestion seems the stronger since, as shown in guinea-pig 167 (Table 4), a certain proportion of the virus is not readily separated from the corpuscles by moderate agitation with salt solution, but remains rather firmly associated with the blood cells (Table 3). In relation to this fact, however, the following points are to be observed: First, the density and viscosity of serum is consider- ably greater than that of 0.85 per cent salt solution, and although the virus cannot be separated from serum readily by centrifuga- tion, such separation could be more readily accomplished in salt solution. Hence it is still possible that the absence of the virus in pathogenic quantity from the salt solution of the loth washing may be caused in large part by the sedimentation necessary for the separation of the corpuscles. Not only must the difference in the density of serum and salt solution cause a wide difference in the ease with which microbes may be sedimented from their solutions, but the coarse physical structure of serum must interfere greatly with the sedimentation of minute particles. In the second place it is readily conceivable that minute organisms may adhere to the external surface of erythrocytes and leukocytes so firmly that moderate a'gitation in salt solution does not cause their separation in pathogenic quantities. In an experiment, the protocol of which is not given, o. 5 c.c. of serum obtained after spontaneous clotting caused fatal infection in the guinea-pig. This would suggest a general plasmatic infec- tion rather than one which is essentially cellular. Since a leukocytic exudate practically devoid of erythrocytes is infectious, it seems probable that the condition could not well be considered as an essential involvement of the erythrocytes. Furthermore, from the fact that the virxis exists in the fluid portion of the leukocytic exudate in pathogenic quantities, the infection could hardly be considered as essentially leukocytic in character. Although it is desirable to wait until certain quantitative experi- ments are completed before drawing positive conclusions as to the Means of Transmission of Spotted Fever 305 situation of the virus, I believe the qualitative results described above indicate a general plasmatic infection rather than an essen- tial invasion of blood cells. FILTRATION AND INTOXICATION. Filtration experiments have been continued with variations in conditions. Small Berkefeld candles have been used exclu- sively and for the most part those which were fresh from the factory. In case a filter had been previously used, it was washed with distilled water in both directions, subjected to prolonged boiling in a solution of sodium carbonate, and washed first with salt solution and then with distilled water in both directions, before being used again. It has seemed useless to employ porce- lain filters so long as the virus has not been passed through the more porous Berkefeld filters. No greater pressure has been used than that obtained from a vacuum water pump, with a pressure of from 30 to 40 pounds in the mains. In all experiments, except one, the filters have been kept covered with the serum by drawing the latter into a pipette, then letting it fall over the surface of the filter. In the exception mentioned, the filter was covered with a rubber tube which extended one-half to three-quarters of an inch above the height of the filter, the serum being fed into the cup made by the projection of the tube. In this particular instance the attempt was made to filter the serum undiluted. After a time, however, the filtration pro- ceeded so slowly that salt solution was added. In the remaining experiments an equal quantity of salt solution was added to the serum before filtration was begun, and in all cases the filter has been washed out with several cubic centimeters of salt solution after the serum had passed through. The serum of infected monkeys has been used throughout for filtration experiments, the serum being obtained by defibrination and centrifugation. In all instances the infectiousness of the unfiltered serum has been determined by control experiments and the serum has been used as soon as possible after obtaining it. In no instance has infected serum even in quantities of 6 to 12 c.c. been infectious for the guinea-pig after being filtered in the 3o6 Contributions to Medical Science manner described, although 0.5 c.c. of fresh defibrinated blood has not failed to cause the disease (see Table 9, as an illustration). TABLE g. Filtration Experiment. Serum of Monkey XII. 5 c.c. of the Diluted Serum Were Injected Intraperitoneally and the Remaining Portion Subcutaneously. Guinea-Pig No. V'olume of Serum ''tZ:Jr i Incubation Period Result Autopsy Filtered Unfiltered 7 c.c. 7 g-io c.c. 9-10 7 days Not sick Death TjT)ical In view of the possibility that minute forms of the organism, capable of passing through the Berkefeld filter, might exist within the erythrocytes or leukocytes, whereas only larger forms might be extracellular, an attempt was made to free such hypothetical minute forms by crushing the blood cells in a porcelain ball-mill. Inasmuch as the result of this experiment points to the existence of a toxic substance in infected blood, the details may be given. Defibrinated blood from monkey No. XIV was washed lo times with sterile physiologic salt solution in order to get rid of the serum. This washed blood proved to be infectious for guinea-pigs in doses of o. 7 and i . 5 c.c. (see Table 3). The washed cells from 25.2 c.c. of blood were ground in the ball-mill for six hours. The mass was removed from the mill by fractional washing with salt solution, centrifugated to get rid of porcelain sand, the latter then being washed fractionally, and the total volume of fluid being made up to 50.4 c.c. by addi- tional salt solution. The fluid was dark red in color, cloudy, and no cells could be recognized by microscopic examination. Some of this fluid, representing 50 per cent of blood in volume, was injected into guinea-pigs in doses of 1.4, 2.0, 5.0, and 10. o c.c, the effects of which are shown in Table 10. The remaining portion was passed through a fresh Berkefeld filter about three hours being occupied in filtration. Filtration proceeded rapidly at first, but more slowly later, as the fluid became more concentrated in insoluble albuminous particles. The filtrate was dark red and perfectly clear. The entire amount was injected intraperitoneally into two guinea-pigs at interrupted periods (Table 11). Means of Transmission of Spotted Fever 307 In comparing Tables 3, 10, and 11, which represent experiments performed with the same blood, one gains the impression, first, that grinding the blood in the mill had almost entirely destroyed its infectiousness, and second, that with the destruction of the infectiousness of the blood the latter manifested a rather pro- nounced toxicity. TABLE 10. Test of the Infectiousness of the Washed and Ground-up Corpuscles of Monkey XiV. Guinea- Pig No. Volume Injected Equivalent in Nor- mal Blood Volume Less Estimated 10% Loss Result 170 171 >72 173 I .4 c.c. 2.0 50 10. 0.63 c.c. 0.9 2.2s 4S Slight febrile reaction beginning on the second day after inoculation. Recovery Moderate febrile reaction beginning the first day after inoculation. Recovery Moderate febrile reaction beginning the first day after inoculation. The primary temperature subsided, but in 6 days a course of fever developed which persisted for 9 days. Recovery Similar to that of 172, but died on the 27th day after inoculation. Death due to extraneous infection, as shown by culture In support of the first point, it may be noted in Table 3 that the blood before it was ground up produced typical infections in doses of 0.7 c.c. and 1.5 c.c, whereas after it was ground up, its infec- tiousness had been largely destroyed, as seen in Table 10. This effect, possibly, is due to an actual crushing of the organism in the mill. TABLE II. Showing the Toxic Effects of the Filtrate of Washed and Ground-up Corpuscles from the Blood of Monkey XIV. Guinea- Pig No. Volume Injected Fractionally Equivalent in Nor- mal Blood Volume Less Estimated 15% Loss Result 174 J7S IS c.c. 20 6.4 c.c. 8.5 Death in 4 days following febrile reaction. Blood sterile Death in 6 days following febrile reaction. Blood sterile In support of the second point, we have to note first the primary fever which developed in the guinea-pigs of Table lo, and second, the toxicity of the filtrate as shown in Table ii. In order to determine whether the intoxication seen in the animals of Table II was due to microbic poison or to the action of the proteids of 3o8 Contributions to Medical Science monkey corpuscles on the guinea-pig, 12 c.c. of normal monkey blood were ground up and a filtrate prepared analogous to that injected into guinea-pigs Nos. 174 and 175. The total filtrate, 20 c.c, was injected into a single guinea-pig which had the weight of the former animals. A shght rise of temperature which occurred on the fifth day after injection, lasted two days; otherwise there was no disturbance. It may accordingly be concluded that the intoxication of animals Nos. 174 and 175 was not due to the normal proteids of the corpuscles of monkey's blood. One could scarcely consider the condition an infection, since the same blood before filtration had shown practically no infectious properties (Table 10). It could hardly be due to soluble toxins in the monkey's blood, since such toxins probably would have been removed by the washings to which the blood had been subjected. Although further experiments are indicated before positive conclusions are drawn, it seems probable that the intoxication of the animals was due to the liberation of poisons by the crushing of the virus in the ball-mill. resistance to heat. Heating experiments have been performed at temperatures of 45° and 50° C. It was necessary to use rather large quantities of blood in order to be certain that a fatal quantity of organisms was being dealt with. On this account it was preferable to use rela- tively low temperatures over rather long exposures in order to insure as complete diffusion of the heat as possible. Results. — In an experiment in which 3 . 5 c.c. of blood were heated at 45° C. for 5, 10, 15, 20, 25, and 30 minutes, all animals died of spotted fever. In two experiments in which the blood was heated at 50° C, the infectiousness of the virus was destroyed in 25 minutes in one series and in 30 minutes in another. RESISTANCE TO DESICCATION. The following is the technic used in desiccation experiments: Uniform quantities of blood are distributed into open Petri dishes which are placed in a desiccator over sulphuric acid and dried as quickly as possible under vacuum at room temperature. Desicca- Means of Transmission of Spotted Fever 309 tion requires from 18 to 24 hours, depending on the degree of exhaustion of the bell-jar. When desiccation is complete the plates are placed in an ordinary sulphuric-acid desiccator in the ice-chest. Results. — Two series of experiments showed the loss of patho- genicity at some time between 24 and 48 hours after complete desiccation. VIABILITY IN THE ICE-CHEST. The M.L.D. of the blood of monkey No. XV, which when drawn was about o.i c.c, had increased noticeably in five days, reached 2.0 c.c. in 11 days, and in 15 days 3.0 c.c. failed to produce infection. The blood was kept in the ice compartment of the ice- chest. In another instance 5.0 c.c. retained infectiousness for 16 days. transmission by means of the MALE TICK. The possibility of transmission by means of the male tick (Dermacentor occidentalis) has been demonstrated conclusively in a recent experiment. The tick was one which had been raised from the egg in the laboratory, the life-history of the brood having been published previously {loc. cit.). Infection of the tick was accomplished by feeding on two sick guinea-pigs in the following way: On October 16 it was placed on the ear of guinea-pig No. 107, where it remained for about 12 hours, at the end of which time the guinea-pig died. Two days later it was placed on the ear of guinea-pig No. 121, where it remained for about 20 hours, or until the guinea-pig died. After an intermission of three days the tick was placed on the ear of the healthy guinea- pig (No. 169), and the latter died in 13 days, showing changes which have been recognized as characteristic of spotted fever. However, since an adventitious epidemic had developed among the guinea-pigs and since areas of focal necrosis found in the spleen of guinea-pig No. 169 were not entirely typical for spotted fever in the guinea- pig, the experiment was not considered conclusive and the animal was discarded. On November 7, 17 days after the tick had been removed from the infected guinea-pig (No. 121), it was again placed on a healthy guinea-pig (No. 182). It was allowed to remain attached for 3^ days, after which it was removed. Ten days 3IO Contributions to Medical Science after the tick was placed on the guinea-pig the latter suddenly developed high fever and died in five days, showing those anatomi- cal changes which have proved to be diagnostic of experimental spotted fever. Autopsy. — Scrotum moderately hemorrhagic and very cyanotic. Hemorrhagic condition is seen best by the naked eye in cutting through the skin. Testicular coverings are moderately congested and the anterior pole of the testicles deeply infiltrated with blood. The axillary, inguinal, and mesenteric lymph glands are enlarged and hemorrhagic. The spleen is several times the mass of the normal spleen and cyanotic in color. The kidneys are congested and cyanotic; suprarenal glands are enlarged; liver enlarged, congested, and cyanotic. The lungs and heart show no appreciable changes. The meninges are slightly reddened. At the point of the tick bite is a necrotic crusted wound about one-fourth inch in diameter. Cultures from the heart and peritoneum yielded no growth. From the organs of this guinea-pig inoculations were made into two other guinea-pigs, one of which died in seven and the other in eight days, both showing typical anatomical and clinical phenomena of spotted fever. Monkey No. XVII which was also inoculated from guinea-pig No. 182 ran a typical course and presented extensive scrotal hemorrhage. From the second generation in guinea-pigs and also from the monkey, inoculations were made into a third generation, the members of which ran typical courses. This is being continued by successive inoculations of the guinea-pig. The experiment is regarded as conclusive. continuous passage through the guinea-pig. In a previous article {loc. cit.) my failure to keep spotted fever alive by the successive inoculation of guinea-pigs was referred to. In the earlier attempts fresh inoculations were made only as the guinea-pigs were about to die or after they had died. The possi- bility was recognized that the quantity of living virus in an infected animal may be greater early in the course of the disease than at the time of death, hence at a convenient time the attempt was made to perpetuate the infection in the guinea-pig alone by inocula- tion with blood or organs taken on the third to the fifth day after fever had begun. This method has proved entirely successful through five and into six generations of guinea-pigs. Hence it Means of Transmission of Spotted Fever 311 seems probable that the monkey can be dispensed with for the purpose of maintaining the disease in the laboratory. SUMMARY. Rocky Mountain spotted fever is transmissible to the guinea- pig and monkey by the inoculation of defibrinated blood of patients suffering from the disease. The virus may be kept ahve in the laboratory either by alternate inoculation of monkey and guinea-pig, or by continuous passage through the guinea-pig by observing the method described above. The disease is transmissible from one animal to another by means of the bite of either the male or female tick (Dermacentor occidentalis) . One attack of the disease establishes a rather high degree of immunity to subsequent inoculation. Attempts to pass the virus through Berkefeld filters have failed. The parasites are not located essentially in either erythrocytes or leukocytes but are present in the body fluids generally. By grinding infected blood in the ball-mill infectiousness is largely destroyed ; in this process there is some reason to think that the organisms are crushed and that toxic substances are thereby liberated. THE ROLE OF THE WOOD-TICK (DERMACENTOR OCCIDENTALIS) IN ROCKY MOUNTAIN SPOTTED FEVER, AND THE SUSCEPTIBILITY OF LOCAL ANI- MALS TO THIS DISEASE/ A PRELIMINARY REPORT. H. T. RiCKETTS. (From the Pathological Laboratory of the University of Chicago.) THE TECHNIC OF FEEDING TICKS ON THE GUINEA-PIG. Many embarrassments have been encountered in experiments in which the tick is allowed to feed on the guinea-pig. There are few spots on the body of the latter which he cannot reach with his teeth or with the toes of his hind legs, and, since the bite of the tick is very irritating, ticks are frequently dislodged or killed, provided the guinea-pig is not restrained properly. The tick is prone to attach itself to the delicate hairless skin behind the ear or to the ventral surface of the body in the vicinity of the mammary glands. When in the former position it is readily dislodged by the nails of the pig's hind feet, and when in the latter position may be killed by biting. In order to obviate some of these acci- dents and to prevent the loss of valuable ticks, the following fairly satisfactory technic has been devised: When testing or feeding an individual tick, the guinea-pig is placed in a pillory for a short time and the tick is allowed to attach itself to the ear or to the back of the head or neck. The pillory prevents the pig from scratching the head, and because of the temporary confinement he seems not to notice the presence of the tick until the latter is more or less firmly attached. Even under these conditions, however, the process is often tedious, since the tick, before it is firmly attached, may be thrown off by a violent shake of the guinea-pig's head. After the tick appears to be attached, usually within ten to twenty minutes, the pillory with the pig in place is tied up in a sack having the bottom and sides of canvas and the top of cheesecloth, this condition being maintained for one or two hours. At the end of this time the animal is removed from the pillory and placed in a skeleton cage which is surrounded by a sack like that described above, a collar of light linoleum first having been placed about the neck of the pig. The collar is of such width (i to i^ inches) that it precludes scratching the head parts, and also prevents the biting of ticks which have become attached to the ventral skin. This width also permits the collar to rest on the bottom of the cage, thus removing the weight, which is sHght, from the neck of the guinea-pig. 'From Jour. Am. Med. Assn., 1907, 49, p. 24. 312 Role of the Wood-Tick in Spotted Fever 313 The cage is made of a frame of wood slats, and has a floor of ^-inch mesh wire, which is attached to the frame an inch above its bottom. This wire floor permits the escape of water and grain to the dead space below, and also gives free ticks refuge from the teeth of the guinea-pig, the mesh being suflSciently large to permit ticks of ordinary size to pass below. When it is desired to feed a large number of ticks, the guinea-pig, with its collar in place, is confined to the cage with the ticks at large, the latter being allowed to feed at will. Under this method large numbers of ticks can be fed with a minimal loss and a small expenditure of time. Infection of the Nymph. Among a group of ticks received from Mr. F. D. Nichols of Hamilton, Mont., January i, 1907, were four nymphs. They had been taken from horses a few days previously, south of Hamilton, in a district not known to be infected with spotted fever. One of the nymphs was moribund at the time they were received. January 3, at 5 p.m., the three remaining nymphs were placed on an infected guinea-pig (290) from which they were removed three days later. Guinea-pig 290 represented the eleventh consecutive guinea-pig passage of the strain which was obtained in the spring of 1906, the custom of infecting the monkey and guinea-pig alternately having been abandoned some time previously. Guinea-pig 290 ran a perfectly tjTpical course clinically, the anatomic changes were characteristic, and cultures from the heart and organs remained free from bacterial growth. It was the intention at this time to place the nymphs individually on healthy guinea-pigs in order to test their infectivity. They entered the moulting stage, however, almost at once, and it was impossible to make this test. Nymph i came out of its moult January 27, twenty-one days having been occu- pied in the process; whereas the moulting period of nymph 2 lasted thirty-four days. Nymph 3 died during the moulting period. Nymphs i and 2 appeared as females. The further history of tick i is as follows: January 29 it would not feed. On February 2 it attached itself to the ear of guinea-pig 332, from which it was removed after fifty-seven hours. It had increased rather markedly in size during the feeding. Temperature of Guinea-Pig 332. February 4 102 . 6 F. February 7 104 F. February 5 103 F. February 8 106 F. February 6 102.2 F. February 9 105 F. February 10 Had been dead for some hours. Autopsy. — Enlarged and hemorrhagic axillary, inguinal, and mesenteric lymph glands. Spleen was several times the normal size and of a deep bluish-red color, and rather coarsely granular. The suprarenal glands were much enlarged and con- gested. The kidneys and the liver were congested, but did not appear degenerated. The abdominal viscera showed postmortem discoloration. The heart, lungs, and intestines were normal. The skin of the external genitals was not congested nor hemorrhagic, and no eruption was discernible on the skin. A plate culture from the heart's blood gave no bacterial growth. Diagnosis. — Spotted fever. The short incubation period, the sudden and high rise in the temperature of the animal and its early death are features which indicated a highly virulent infection. Under these conditions experience has shown that the skin manifestations do not always develop. 314 Contributions to Medical Science On account of postmortem changes in the viscera, transfer of the infection from the dead animal was not attempted. Second transmission. — On February g, after an interval of five days, the tick became attached to guinea-pig 363, from which it was removed after feeding for twenty-two and one-half hours. Temperatitre of Guinea-Pig 363. February 11 103 . 4 F. February 13 105 . 4 F. February 12 103.8 F. February 14 105.5 F. February 15 105 . 2 F. Blood taken from the heart. February 16 104 F. February 17 103.9 F. February 18 103.8 F. Scrotum swollen and congested. February 19 Dead. Autopsy (immediately after death). — Lymph glands were moderately enlarged and congested. No changes were found in the liver, lungs, heart, or intestines. There were no cutaneous hemorrhages, but the skin of the face and external genitals was extremely cyanotic. A plate culture of the heart's blood remained sterile. Diagnosis. — Spotted fever. The 3 c.c. of blood drawn from the heart February 15 were injected intraperi- toneally into guinea-pig 376 immediately. The following course of fever was recorded: February 16 103 . 4 F. February 19 105 F. February 17 103. i F. February 20 106.2 F. February 18 105 F. February 21 105 F. February 22 96. 2 F. Killed by chloroform when moribund. The findings at autopsy were identical with those recorded above and need not be repeated. Plate cultures of the heart's blood, spleen, and liver remained sterile. In making the plate cultures of the organs fragments of the tissues were triturated in bouillon and the emulsion thus obtained was mixed with the liquefied agar before plating. Third transmission. — On February 26 the tick attached itself to the ear of guinea- pig 406, where it was allowed to remain for thirty hours. On March 3 the temperature rose to 104. 1 F., in the vicinity of which mark it remained until March 6, when the animal died. The anatomic findings were as described above, and cultures from the heart, liver, and spleen were likewise sterile. The skin of the scrotum was deeply congested. The second nymph underwent a positive test on guinea-pig 364 on February 10, after a feeding period of twenty-two and one-half hours. Blood was drawn from the heart of this animal on the first day of fever (February 15) and 3 c.c. were inoculated into guinea-pig 377. The latter animal ran a typical course and was killed on the sixth day after infection in order to obtain material for the inoculation of monkey 26. The anatomic changes were characteristic and the bacteriologic result was, as usual, negative. Ten cubic centimeters of defibrinated blood from guinea-pig 377 were given to monkey 26 intraperitoneally (February 21). The following course of fever developed: February 22 103 F. February 25 105 . 3 F. February 23 105 . 6 F. February 26 105 F. February 24 105 F. February 27 104.4 F. Killed February 27, the blood being utilized for various experiments. The anatomic changes were tj^pical and cultures from the blood and viscera gave no growth. Role of the Wood-Tick in Spotted Fever 315 In order to test the validity of the experiment still further three guinea-pigs were inoculated with the blood of monkey 26. All three developed spotted fever, exhibit- ing typical symptoms and anatomic changes. These results are of especial importance in two particulars: In the first place it must be considered that the tick in its role as the carrier of spotted fever is itself the victim of an infection; that the transmission is not purely mechanical. The following facts support this conclusion: First, the tick called nymph i retained its infectivity for fifty-four or fifty-five days, having in the mean- time passed through the moulting stage, whereas outside the living body the viabiUty of the virus under the most favorable conditions, i.e., in the ice-chest and in the dark, is less than half this period and may be only one-third as long. If we assume that the virus could have remained in a moist condition on the mouth parts of the tick, infectiousness would have been retained no longer than seven or eight days, since under the conditions surrounding the tick, i.e., in diffuse light and at room temperature, the virus in one experiment Hved approxi- mately a week (from unpublished experiments). It is quite certain, however, that the virus would not remain in a moist condition on the mouth parts of the tick, but, on the contrary, that it would become dried speedily. Extreme desiccation over sulphuric acid usually destroys the virus in from twenty-four to forty-eight hours. It is probable that desiccation as it occurs under natural conditions of temperature and atmospheric mois- ture would not destroy the virus so soon, but it would be difficult to believe that its life could extend for fifty-four days under such circumstances. The experiments quoted afford, in addition, some positive evidence that the virus proliferates in the tick. I have never found the blood of animals suffering from spotted fever to be infective in doses under 0.05 c.c, and usually the minimal patho- genic dose is nearer o.i c.c. The tick, nymph i, infected three guinea-pigs at successive intervals, and if we assume that the blood by which the tick was infected possessed the highest virulence, and that the virus underwent no proliferation in the tick, it would have been necessary for the tick to ingest 0.15 c.c. of blood, at least, in 3i6 Contributions to Medical Science order that it might later infect three guinea-pigs. This volume of 0.15 c.c. is greatly in excess of that of the tick, since in this instance the adult was of unusually small size. Hence it seems necessary to conclude that the tick becomes infected, that the virus proHferates in its body, and that, on biting the guinea-pig, the virus, equivalent in quantity or virulence to that contained in 0.05 to 0.1 c.c. of blood from an infected animal, is in some way injected. In the second place these results have an important relation to the disease as it occurs in the Bitter Root Valley and other localities. They indicate that the adult tick which infects man in the spring does not of necessity acquire its own infection in the spring, but may have acquired it by biting any one of several possible hosts during the autumn or winter when in the nymphal stage. It may with some reason be suggested that the experiments cited above do not prove that these ticks acquired their infection when in the nymphal stage, but, on the contrary, that they had become infected under natural conditions, that perhaps they were infected as larvae, or that they derived their infection from the parent tick through inheritance. The possibihty of these events cannot be denied, although the fact that the ticks came from a vicinity not known to be infected makes it rather improbable. Experiments are now being conducted to determine the possi- bilities of infecting the larvae and of hereditary transmission, and they will be reported at a future time. other experiments with the "wood-tick." Transmission of Rocky Mountain spotted fever by means of the adult female has already been reported by King' and by me (p. 278),* and I have also reported (p. 299)^ transmission by means of the adult male tick. By means of ticks obtained from Montana in the early part of 1907 I have repeated these experiments many times. At present I shall only relate the main facts concerning the positive results, adding the statement that many irregularities have been met with ' Public Ueallh Reports, July 26, 1906. 'Jour. Am. Med. Assn., 1906, 47, p. 358. > Jour. Inject. Dis., i, 1907, 4, p. 141. Role of the Wood-Tick in Spotted Fever 317 and that the ease with which individual adults become infected is subject to considerable variation. Male tick 7 produced a course of fever in guinea-pig 339 which lasted for seven days. The animal recovered and one month later died of a bacterial infection. Fe- male tick 4 fed on an infected monkey (22) for six hours and forty minutes, January 12, 1907. Subsequent tests of a thorough character on guinea-pig 315 during the next month demonstrated that the tick had not become infected from the monkey. Feb- ruary 9 the tick fed on an infected guinea-pig (333) for twenty-two hours. Nine days later it was allowed to feed on a normal guinea-pig (381) for twenty-two hours. This animal died in two weeks with anatomic changes which were not characteristic of spotted fever and the test was thrown out. A second test was made February 26 (guinea-pig 404), the animal showing a course of fever, and at its death in nine days, anatomic changes which are characteristic of spotted fever in the guinea-pig. Other successful tests were made on guinea-pig 417, March 7, and later in conjunction with other ticks. Guinea-pig 415 was bitten by female ticks 4 and 7. After an incubation period of forty-eight hours fever developed, and the animal died in five days after the ticks were first attached. A similar result has been obtained several times by per- mitting more than one infected tick to bite a guinea-pig, hence it is very probable that the virulence of an infection has a relation to the number of infected ticks which feed on the animal. Female tick 7 did not become infected after feeding on monkey 22 for fourteen hours, as shown by a test on guinea-pig 320, and on other pigs during the course of the next month. February 10 it was allowed to feed on an infected guinea-pig, the duration of the feeding not being noted. Ten days later the disease was transmitted to guinea-pig 386, on which the tick fed for thirty hours. Female tick 8 did not become infected after feeding on monkey 22 for eight and three-fourths hours. Two subsequent attempts to infect this tick failed, one feeding having lasted twenty-four hours and the other fifteen hours. Similar experiences have been encountered with several females and with many males. Female 8 subse- quently died in a blackened state, when filled with blood. Females 12, 14, 15, 17, 18, 19, and 20 were infected by single feedings which lasted from twenty to thirty hours. This was, of course, proved by tests on normal guinea-pigs. The experience of the past winter has shown that it is much more difficult to infect the adult male than the female tick. This may have a direct relation to the fact that the male is less voracious in its feeding than the female, that it ingests a smaller quantity of blood, and hence has less chance of becoming infected. Concerning the mechanism by which the tick produces infection, it is possible to produce only hypotheses at this time. That the tick injects a toxic secretion into the skin is manifest from the erythema and softening of the epidermis which accompany the bite. Not infrequently a small area of tissue sloughs after the tick 31 8 Contributions to Medical Science is removed. In all probability the secretion of the salivary glands is responsible for these local manifestations. It is certain that the latter have no relation to spotted fever, since they are produced by normal as well as by infected ticks. In all the experiments so far, both by the artificial inoculation of blood and by the bite of the tick, there is no reason to believe that the disposition of the virus produces a visible local reaction in the subcutaneous tissue. That the virus of spotted fever is introduced by the tick with the salivary secretion is the most plausible assumption, although the possibility of infection by regurgitation from the alimentary tract remains for consideration. ADDITIONAL NOTES ON THE LIFE-HISTORY OF THE WOOD-TICK. Any observations throwing light on the life-cycle of the wood- tick (Dermacentor occidentalis) are important, because of the relation which this arachnid bears to the spotted fever. In a previous article (p. 291)^ I gave a short account of the life- history of the wood-tick as it was observed under artificial conditions in a single instance. The life-cycle was observed from the time the female began to lay her eggs, June 21, until the young had emerged from their second moult as young adult ticks about September 10, a period of somewhat less than three months. Through accidents it was not possible to preserve the new brood until they had ripened sexually, hence to this extent there exists a temporary gap in my observations. Inasmuch as adult ticks are rarely found in the autumn in the Bitter Root Valley, it seems evident that the com- paratively rapid evolution of the cycle in the experiment cited was due to the presence of a host constantly and to the rather warm temperature under which the experiment was conducted. In order to determine with some exactness the form in which the tick survives the winter and under what conditions, the Bitter Root Valley was visited in December, 1906. Various theories exist, locally, in reference to this point, the two more prevalent being, first, that the arachnid secretes itself under stones, beneath the bark of dead trees, or that it burrows into rotten stumps to ' "Further Observations on Rocky Mountain Spotted Fever and Dermacentor occidentalis," Jour. Am. Med. Assn., 1906, 47, p. 1067. Role of the Wood-Tick in Spotted Fever 319 remain there until spring; second, that ticks may be found on horses and cattle in the hills during almost any month of the winter. Both possibihties were investigated, the former resulting largely in failure. One individual who felt perfectly confident concerning the first theory failed to discover ticks in the suspected locahties after several days' search, and my own attempts were as fruitless. That the tick may occasionally be found apart from living animals in midwinter was shown by a miner who, the day before I saw him, had found one in the tunnel of his mine. The second theory soon found ample verification, and through the kindness of Mr. F. D. Nichols, of Hamilton, and Dr. E. F. Dodds and Professor M. J. Elrod, of Missoula, I received during January and February several hundred ticks which had been picked from horses brought in from the hills for this purpose. Among those received early in January nearly all were in the second moulting stage and after varying periods emerged as adult ticks. With them, however, were four nymphs which had fed moderately and were more or less distended, and a few which had cast off the second skin. The nymphs were those used in the experiments cited above. During February all the ticks received were either in the second moulting stage or had emerged from it. Early in March I obtained ffom Mr. Nichols about thirty adult females which were greatly enlarged, and some of which began the deposition of eggs within a few days. In the laboratory experiment cited above it was noted that the ticks left the guinea-pig when they had developed sufficiently to moult. This appears not to be the rule, however, since a very large percentage of the ticks which were in the second moult when received still grasped lamellae of dried epidermis with their man- dibles, the softened epidermis having been torn loose at the time the ticks were removed. It seems probable that the appearance of the "wood-tick" of Montana in an overwhelming "crop" during March, April, and May, and its apparent absence during other periods are the results of chmatic conditions, and that it is prevented from being a perennial pest only on account of these conditions. It seems to be only 320 Contributions to Medical Science ignorance of the nature of the larval, nymphal, and moulting forms which has created the impression in the Bitter Root Valley that the tick disappears entirely in July to reappear only when spring comes. Description of these forms to residents has frequently shown that they have been observed, although not identified. Examples of exceptional development have been reported to me a number of times, such as the precocious appearance of a few adult ticks in the late summer or fall. I was shown a larval tick which had attached itself to the ear of a child in December, and in April a nymph obtained from a similar source. It may be assumed that the larva seen in December represented eggs which were laid late in the summer or in the fall rather than in the early sum- mer, and that the nymph seen in April came from an egg which had been deposited unusually early in the spring or unusually late in the fall. These examples of precocious or tardy appearance of the tick may well have a relation to the rare occurrence of spotted fever so late as August, or so early as February, rather than during the months of spring at which time the disease prevails habitually. It is difficult for the residents of the Bitter Root Valley to believe that the tick lives exclusively as a parasite on animals, on account of the widespread and more or less uniform distribution of the arachnid in nature. It is thought it must live to some extent at least on vegetable or on dead insect material. There is as yet no direct evidence to support this suspicion. It is true that among the foothills the tick seems to be omni- present, yet the uniformity of its distribution even in these locali- ties is open to doubt. It is not uncommon to learn from a ranch- man that more ticks may be found on one hill than on another which is in close proximity. Horses and cattle roam practically all the foothills to some extent, especially during the fall and winter. During their residences in such places they collect innumerable ticks, probably in the form of larvae chiefly. The young ticks grow and moult on the host and when they reach adult form one would be safe in assuming that large numbers drop off, or are rubbed of! against bushes and small pines by the animals. A high wind may be an important factor in disseminating ticks, for a ranchman related his experience in driving home one evening in Role of the Wood-Tick in Spotted Fever 321 the spring during a violent windstorm. He felt particles striking his face and clothing, and on reaching home found a large number of ticks on his clothing. From my own experience I can state that ticks climb small pines to at least the full height of a man. It is probable that they ascend higher than this, and they are so light that without doubt they could be easily carried by a strong wind for some distance. It is another fact of importance that the tick has increased greatly in numbers since the valley has been well settled and the quantity of the stock has increased. This is agreed to by all who have lived long in the valley. CONDITIONS IN THE LO LO VALLEY. The conditions in the Lo Lo Valley during the past two years are very important. Lo Lo Creek is tributary to the Bitter Root River, and its valley for some years has been considered as badly infected. In 1906 this locality had no infections, and there was hope that the disease had died out there. If the disease is kept alive by ticks which have bitten infected men, there would be some ground to hope that a year without an example of infection in man might result in the disappearance of the disease. If after a year of no infections cases did appear in the following year, three possibilities present themselves in the way of explanations: (i) Certain ticks may live for two years instead of one year, the latter being the usual period of life; (2) the virus may be transmit- ted through two generations of the tick without its renewal from an infected host; (3) each generation of ticks may acquire the disease afresh each year from some susceptible host other than man (see below). There is not sufficient ground at present for considering a fourth possibility, that the tick becomes infected from decaying organic matter. As stated, no cases occurred in the Lo Lo Valley during 1906, and but one has appeared in 1907. It is hoped that the first two possibilities can be decided experimentally. SUSCEPTIBILITY OF THE GOPHER AND OTHER ANIMALS. The third possibility mentioned above, that spotted fever may be kept alive in a susceptible host other than man, has received 322 Contributions to Medical Science some support this year. Assisted by Mr. P. G. Heinemann, I have found that the indigenous gopher or ground squirreP may be infected by inoculation and that one attack confers immunity to second inoculations. Its susceptibility is much less than that of the guinea-pig and is not perfectly uniform. The disease was again conveyed to the guinea-pig from the inoculated gophers. I do not maintain at this time that the gopher exclusively is respon- sible for the maintenance of the disease, although this is possible. The susceptibility of other animals found in the valley is being investigated. Dr. William M. Chowning co-operated with us in the inoculation of a horse, which received about 60 c.c. of blood subcutaneously from infected guinea-pigs. After a brief incuba- tion period, the animal ran a short but rather high course of fever. The horse resisted entirely a second inoculation of 60 c.c. of infected blood from man and monkey, the experiment being conducted by Mr. Heinemann and myself. Assisted by Mr. Heinemann, a second horse was inoculated with 10 c.c. of infected blood from man. No distinct reaction followed. A calf, 2 weeks old, failed to react following the inoculation of over 30 c.c. of blood from infected guinea-pigs. infected ticks in nature. Although the experimental transmission of Rocky Mountain spotted fever by means of the "wood-tick" attracts the gravest suspicion to this arachnid as the agent of infection under natural conditions, the theory of tick-transmission cannot be considered as proved circumstantially until infected ticks are discovered in nature. Assisted by Mr. Heinemann, it has been possible to demonstrate that infected ticks do exist under natural conditions on the west side of the Bitter Root Valley. The ticks were col- lected from vicinities in which the disease appeared this year, partly from horses and cows and partly from the woods. A group of thirty-six males taken from the vicinity of Mr. Hickey's ranch produced the disease in guinea-pig 596, which presented a typical clinical course with scrotal hemorrhages and died in twelve days after the ticks were first attached. The ■ The following experiments, with others, will be reported in detail, in conjunction with Mr. Heine- mann in the near future. Role of the Wood-Tick in Spotted Fever 323 anatomic changes were characteristic of spotted fever in every detail, as it occurs in the guinea-pig, and cultures from the heart, liver, and spleen remained free from discoverable bacterial growth. I take the position thai these experiments connect the tick unmistakably with the occurrence of spotted or tick fever of man in the Bitter Root Valley, and that without them the "tick theory" would lack essential support. FURTHER EXPERIMENTS WITH THE WOOD-TICK IN RELATION TO ROCKY MOUNTAIN SPOTTED FEVER/ H. T. RiCKETTS. {From the Pathological Laboratory of the University of Chicago.) Experiments which I previously reported show that the adult male and female and the nymph of the Rocky Mountain wood- tick (Dermacentor occidentalis) are able to acquire and transmit spotted fever. In this paper experiments will be reported which show: (i) that the larva may acquire the disease and remain infective during the nymphal stage; (2) that the virus may be transmitted from an infected female to her young through the eggs; (3) that the virus exists in both the gut and the salivary glands of the infected tick. I. INFECTION OF THE LARVA. In these experiments the possibiHty of the lar\'ae having acquired the disease from the female parent through the egg was excluded by testing the females on healthy guinea-pigs before they deposited their eggs, and it is to be understood that all the larvae used in these experiments came from females which had been tested in this way and found uninfected. When from one-third to three-quarters of the eggs of one or more females had hatched, the bottle which contained them was placed in a tick-proof cage with a guinea- pig. As the eggs hatched, the living larvae were removed from the bottle and placed on the guinea-pig, where they were left to feed or to drop off until ready to feed. Within a few days the guinea-pig was inoculated with spotted fever and the course of the fever and the condition of the animal were observed daily. When the larvae had fed sufficiently they fell from the animal and crawled up on the canvas which covered the cage. From this location thej' were removed and placed in boxes to await moulting and the nymphal stage, and when a sufficient number had reached this stage they were placed in a fresh cage with a healthy guinea-pig on which they were allowed to feed. It was considered preferable to carry on the experiments in this way, i.e., to infect as larvae and to test as nymphs, rather than to attempt both steps during the larval stage. The latter course would have involved serious difficulties, and, more- over, the results have a more practical bearing when it is shown that the tick, having acquired the disease as a larva, remains infective after reaching the nymphal stage. I From Jour. Am. Med. Assn., 1907, 49, p. 1278. This work has been supported by an appropria- tion made by the legislative assembly of the state of Montana, at the solicitation of the Montana State Board of Health, and aid has also been rendered by the University of Chicago, and by the Memorial Institute for Infectious Diseases, Chicago. 324 Experiments with Wood-Tick in Spotted Fever 325 Larval infection 2. — On July 8 several thousand normal larvae were placed with an infected guinea-pig (670). A large number fed, dropped off, and were isolated, and when they began to moult were placed with a normal guinea-pig (717). After a period of six days the temperature of the latter rose to 105.7 F., and on subsequent days it registered 105, 106, 105.4, and 105 F. The animal was found dead on July 20, after six days of fever and twelve days after it was placed with the ticks. The condition at autopsy was typical for spotted fever in the guinea-pig. The spleen was enormously enlarged, fairly firm, cyanotic, and granular. The inguinal, axillary, and mesenteric lymph glands were deeply congested or hemorrhagic, and the surrounding areolar tissue was congested. The kidneys and suprarenal glands were enlarged and congested. The liver was somewhat enlarged. The heart and lungs appeared normal. No hemorrhages had occurred in the external genital organs. The guinea-pig was a female, and hemorrhages are less constant in the vulva than in the scrotum. Cultures on agar slants made from the heart, spleen, and kidney remained free from growth. Twelve enlarged nymphs were removed from the pig and cage, and this may be taken as a fair index of the number which had fed. Guinea-pig 732 likewise was infected by the remaining nymphs, the animal making an eventual recovery. Of the several thousand larvae placed with the infected guinea- pig (670) only a small proportion fed during the course of the fever. Under the title of larval infection 7 the remaining larvae were exposed to infection on guinea-pig 637, which was inoculated on the fourth day after being placed with the larvae. This animal had been used previously in testing ticks found in nature, but had not been infected, a fact which was shown by the typical course of fever which followed inoculation. The larvae which fed on guinea-pig 637 were collected and after they had reached the nymphal stage were placed with a normal guinea-pig (725). The temperature of the latter rose to 104.7 F. on the sixth day and death occurred three days later. Although this is a rapid course for spotted fever, the anatomic changes were characteristic. There were enlarged and intensely 326 Contributions to Medical Science congested lymph glands; enlarged, cyanotic, and rather firm spleen, and incipient scrotal hemorrhages. No gross changes were discoverable in the other organs, and inoculations of agar slants from the liver, spleen, and heart's blood showed no growth. At least seven nymphs had bitten the guinea-pig. In order to corroborate this result further, a monkey was inoculated from an emulsion of the Hver, spleen, kidney, and suprarenal gland. The temperature of this animal on the second day rose to 104 F., and on subsequent days registered 105.8, 104.7, 105.4, 104. i F., and death occurred on the seventh day after inoculation. The fol- lowing were the essential changes seen at autopsy: beginning hemorrhages in the skin of the eyebrows and cheeks; the whole perineum was cyanotic and showed a roseolar eruption; the lymph glands everywhere were enlarged and congested but were not hemorrhagic; the spleen was greatly enlarged, rather firm and cyanotic. No changes could be noted in other organs. The virus was again passed through another guinea-pig which reacted typically and the strain was then dropped. Still another test was made with the nymphs which remained. After an incubation period of five days the guinea-pig (739) showed a temperature of 105.3 F. and died six days later with char- acteristic anatomic changes. No ordinary microbes could be cultivated from the tissues. At least six nymphs had fed on the guinea-pig. Larval infection 4. — On July 10 about 2,000 normal larvae were placed with guinea-pig 659 and a few days later the animal was infected by inoculation. The guinea-pig died on July 21 and was replaced by another (688) which was infected similarly and died on August 2. A third guinea-pig (714) was introduced on this date, was inoculated on August 5, and died August 14. All three animals presented the course and anatomic changes of spotted fever. The larvae which had fed were collected and, after hatching as nymphs, 360 were placed with a normal guinea-pig (726), the nymphs being added as they moulted. None of the nymphs became attached for seven to ten days, and about ten days after feeding was discovered a course of fever began in the guinea-pig Experiments with Wood-Tick in Spotted Fever 327 and the latter died in six days, showing the characteristic anatomic changes of spotted fever. In a similar manner the nymphs of larval infection 6 killed two guinea-pigs successively (731 and 751), and the combined n>-mphs of experiments 2 and 7 produced spotted fever in guinea- pig 674, which eventually recovered. II. transmission of the virus from the infected female to her young, through the egg. On hj-pothetical grounds these so-called heredity experiments were performed in two groups: First, certain females were exposed to infection before they were impregnated, by permitting them to feed on guinea-pigs which had been inoculated with spotted fever. This was done in experiments 3 and 5. Second, the ticks of experiments 4 and 6 had been impregnated before their infection was attempted. This course was pursued, since it has been suggested in the literature that failure in hereditary transmission may depend on the time of infection with relation to the impregnation of the female. After impregnation many of the ova begin to enlarge, and, if the female is provided with food, they become surrounded by a denser pellicle. Presumably a microbe could penetrate such eggs less readily than the more minute and more delicate, immatiure ova or germ cells of the unimpregnated female. In this instance, however, the point just discussed is not an essential one, since the most satisfactory results which I obtained involved females which were impregnated before their infection was attempted. Casual observation of the manner in which the eggs mature makes it clear why the conception mentioned is to a certain extent irrelevant in so far as the probability of infection of the eggs is concerned. If one exposes the viscera of a fully developed female which is on the point of laying its eggs it is seen that only a small proportion of the ova have matured, i.e., have reached the limit of size and the appearance which they habitually obtain before extrusion. Hundreds of smaller and still smaller ova are seen in the sac as one follows it back, a fact which makes it apparent that the eggs are matured seriatim, the food which is found in the greatly distended aUmentary sac being gradually consumed in the mean- time. Hence, even in the thoroughly ripened female one finds many minute and delicate ova which apparently would be susceptible to penetration. None of the females used in these experiments was infected in the first instance. They were among ticks which had been collected from animals and from the woods during the search for infected ticks in nature, and each had fed on a normal guinea- pig for a period much longer than that required for the transmission of the disease. If the female failed to enlarge after prolonged feeding and after the passage of abundant feces the result was considered as proof that she had not been impregnated. It is not my intention to analyze all the experiments in detail at this time. In addition to reporting two positive transmissions through the egg, I may mention the fact that short courses of fever were produced by the larvae in several instances, without the development of cutaneous phenomena. Immunity tests are still to be made with such animals, in order to determine whether these occurrences represented mild attacks of spotted fever. The technic was similar to that used in the experiments termed larval infections, 328 Contributions to Medical Science i.e., the lan-ae, as soon as hatched, were placed with normal guinea-pigs in tick-proof cages. Both experiments concern heredity experiment 4, in which infection of the ticks took place subsequent to their impregnation; that is to say, the ticks were not allowed to feed on infected guinea-pigs until the character of their enlargement after feeding on normal guinea-pigs indicated that they had been impregnated. The rather scant supply of guinea-pigs rendered it impossible to prove the infectivity of all females before the eggs were laid. This was done in several instances, however, and all which were so tested proved to be infected. Concerning the two e.xperiments to be reported, the infectivity of the larvae is ample proof that the females had been previously infected. The precaution was taken to perform the experiments in entirely new cages, so that the larvae had no opportunity to come in contact with spotted fever virus other than that contained in the female parent. Females 7 and 9 of heredity experiment 4 fed on four infected guinea-pigs from May 22 until June 14, and at this time they were ready for o\nposition. Experiment with female 7. — This female was removed from an infected guinea-pig June 14; it began laying eggs June 27; hatching began on or about July 27. It was estimated that about 2,000 eggs were laid. Beginning on July 30, the larvae, as they hatched, were placed with a normal guinea-pig (707). Inasmuch as they do not feed for some days after hatching, immediate results could not be anticipated. On August 22 or 23 the temperature of the guinea-pig rose to 104.5 F., the following day to 105.5 F., in the vicinity of which point it remained until the animal died August 28. The following notes were made at autopsy: The lymph glands are deeply congested and considerably enlarged and the surrounding areolar tissue is congested. The spleen is about twice its normal size, is fairly cyanotic and moderately firm, resembling the spleen of spotted fever when death occurs at an early stage. The kidneys are congested and degenerated, and the suprarenals are congested and enlarged. The appendages of the testicles show a good deal of congestion. There is no enlargement of the scrotum and no hemorrhages have occurred into it. (I have stated hitherto that these scrotal changes are sometimes missed when death occurs early.) Other organs appear to be unchanged. Agar slants inoculated from the heart's blood, from the liver and the spleen, remained free from discoverable growth. Experiments with Wood-Tick in Spotted Fever 329 Two hundred and eleven enlarged larvae were removed from the cage, hence this number at least had bitten the guinea-pig. In order to verify the suspicion that this animal had died of spotted fever, an emulsion of its spleen and liver was injected into guinea-pig 746. The temperature of this animal was recorded as follows on successive days: 103.7, 102.7, 104I) io5-5j 105 -6, 105.4, 105.8, 105.3, 104.8, 104, and 102.2 F. Death occurred on the fourteenth day following inoculation. AUTOPSY notes. Lymph glands are enlarged and deeply congested or hemor- rhagic; spleen is greatly enlarged, cyanotic, and fairly firm; kidneys are swollen and congested; the scrotum is greatly enlarged and intensely hemorrhagic, and the tunica vaginalis is deeply congested. No gross changes are seen in other organs. The diagnosis of spotted fever was made positively. Experiment with female g. — This female was removed from an infected guinea-pig June 14; it began laying eggs June 28; hatch- ing began about July 28. Several hundred eggs were laid. Beginning on July 31 the larvae were placed with a normal guinea-pig (710) as rapidly as they hatched. Ticks were first seen attached after thirteen days and the first enlarged larvae were removed on the fifteenth day. Ten or eleven days later the temperature of this animal rose and continued within the limits of 104.3 ^^J^d 105 -3 F- fo^ seven or eight days. Following this the fever subsided and the animal recovered. Slight enlargement of the scrotum after several days of fever was the only condition, aside from the temperature, which suggested spotted fever. Inasmuch as in some other animals in these experiments the disease had run a similar course, with eventual recovery, it was suspected that the condition represented a mild attack of spotted fever. In order to decide this point if possible, 3 c.c. of blood were drawn from the heart of guinea-pig 710 during the height of fever and injected into a normal guinea-pig (749). The following course of the temperature in this animal was observed on succes- sive days: 102.4, 103.4, 103.4, 105.6, 104.8, 106.7, 106.4, 106, 104.6, 105.4, 104.4, I04; 104.2, and 103.5 F. The guinea-pig 330 Contributions to Medical Science recovered. On the fourth day the scrotum began to swell, and on the tenth day the swelling was extreme and extensive hemorrhages had appeared. With recovery the hemorrhagic areas in the scrotum became gangrenous, and following separation of the sloughs a deformed and cicatricial scrotum remained. A positive diagnosis of spotted fever in this animal was made, and the diagnosis was confirmed later by a severe immunity test, in which no reaction occurred. Approximately two hundred larvae had bitten guinea-pig 710. The results of these two experiments prove, without question, that the virus of spotted fever may pass from an infected female to her young through the egg. The practical importance of this fact will, to a certain extent, depend on the viability of the virus in larvae which have been infected in this manner; this is to say: Will such ticks be infective in their subsequent n^nnphal and adult stages? Although one probably should not try to forecast results, the positive results which I have obtained in infecting normal larvae and nymphs suggest that larvae which have acquired their disease through the egg will also prove infective when they have reached the adult stage. This suspicion is particularly well founded since the process of moulting does not destroy the infec- tion in either the larva or the nymph. II. the infectivity of the salivary glands and the alimen- tary SAC OF THE diseased TICK, I wish to report a single experiment in which the salivary glands and the alimentary sac of an infected male tick (15) were dissected out and injected separately into normal guinea-pigs. Male tick 15 had been infected by feeding on a diseased guinea-pig and its infectivity was proved on normal guinea-pig 677. One cannot well remove the alimentary sac without contaminat- ing all other viscera with its contents. On the other hand, one can grasp a salivary gland with forceps near the excretory orifice and, after a little careful dissection, remove the entire gland with- out rupturing it. Hence in this experiment the salivary glands were removed first and without rupturing the alimentary sac, after which the Experiments with Wood-Tick in Spotted Fever 331 latter was removed without regard to remaining structures. The glands and the sac were then triturated separately in sterile salt solution and injected, the salivary glands being inoculated into guinea-pig 683 and the alimentary sac into 694. From the date of inoculation, July 15, the following record of the temperature of guinea-pig 683 was made on successive days: 104, 102.4, 102.3, io5-7> 105.7, 105.8, 105.9, 105. I, 105.4, 105.4, 104. 1, 103.2, 102. 1, 102.6, 102.7 F., etc. There was eventual recovery. Four days after inoculation the scrotum began to swell and on the seventh day it was extremely hemor- rhagic and enlarged. On recovery the hemorrhagic areas sloughed. The course of fever and scrotal changes in guinea-pig 684 were similar, though not so extreme. On August 8 both animals were given immunity tests with three cubic centimeters of infected blood, a quantity which con- tained from thirty to sixty minimum pathogenic doses. Neither animal showed any reaction to the second inoculation. In a repetition of this experiment two ticks should be used, the salivary glands being taken from one and the alimentary sac from the other. It seems probable that the virus passes from the alimentary sac of the tick to the salivary glands; possibly this can be deter- mined experimentally. Indeed, since the oviducts and the eggs of the female are invaded, at least in certain instances, it is not unlikely that spotted fever exists as a generalized infection in the tick for a greater or less period. Whether it eventually becomes localized in certain organs is a question for future investigation. However this may be, the disease is often, if not always, a com- paratively harmless incident for the tick. SUMMARY. I have established the following points concerning the relation- ship of the Rocky Mountain wood-tick to the spotted fever of western Montana: 1 . Infected ticks exist in the so-called infected districts in nature. 2. Both the adult male and the adult female may acquire the disease by feeding on an infected animal, and may transmit it 332 Contributions to Medical Science to a normal susceptible animal for a period of several weeks there- after. 3. During either of its intermediate active stages, larval or nymphal, the tick may acquire the disease in the same manner, retain it during moulting, and prove infective when it reaches the subsequent active stage. 4. The infected female may transfer the disease to her young through the egg. It is possible that this does not happen in all instances, and it is quite certain that the brood of an infected female may include many uninfected larvae. This seems to have been proved in three of my experiments in which the infectivity of the females had been proved before oviposition; the larvae in these experiments failed to infect normal guinea-pigs. 5. The virus exists in both the saHvary glands and the gut of infected ticks at a certain time, and since it also invades the genera- tive organs of the female the condition is probably one of a general- ized infection, at least for a period. The disease is not highly destructive for the tick. At least two important steps may now be taken in an aggressive fight against the disease: first, a thorough dissemination of the knowledge that the tick is the agent of infection; second, a massive reduction of the number of ticks in infected districts by means now used in the destruction of the cattle-tick of the southern states. Extermination of the tick in the mountains will not be possible so long as native wild animals inhabit the soil and roam the hills. On the other hand, it is known locally that the number of ticks in the Bitter Root Valley has increased enormously as greater num- bers of domesticated animals have been introduced, and the latter now seem to be the chief hosts for the tick. A SUMMARY OF INVESTIGATIONS OF THE NATURE AND MEANS OF TRANSMISSION OF ROCKY MOUNTAIN SPOTTED FEVER.' H. T. RiCKETTS. (From the Pathological Laboratory of the University of Chicago.) In the Medical Sentinel (Portland, Ore.) for October, 1899, Dr. Edward E. Maxey, of Boise City, Idaho, called attention to *'the so-called spotted fever of Idaho." He described it as ''an acute, endemic, non-contagious, but probably infectious, febrile disease, characterized clinically by a continuous moderately high fever, severe arthritic and muscular pains, and a profuse petechial or purpuric eruption in the skin, appearing first on the ankles, wrists, and forehead, but rapidly spreading to all parts of the body." The disease is not to be confused with typhus fever nor epidemic cerebrospinal meningitis, to both of which the term spotted fever is sometimes applied. Rocky Mountain spotted fever is not limited to Idaho, however, but occurs also in western Montana, Utah, Nevada, Oregon, Colorado, Wyoming, and perhaps in other neighboring states. The conditions in the Bitter Root Valley in western Montana aroused such concern that Dr. Louis B. Wilson and Dr. William M. Chowning, then of the University of Minnesota, undertook an investigation in 1902.^ They learned that the disease occurs only during the spring and early summer, and that the eastern side of the Bitter Root Valley is free from infection. They also announced the discovery of an erythrocytic parasite, Piroplasma hominis, as the cause of the disease and promulgated the tick-gopher h}po th- esis. According to this hypothesis the gopher is the habitual host for Piroplasma hominis, and spotted fever is conveyed from the gopher to man by means of the bite of the wood-tick (Derma- centor occidentalis) , which feeds on both man and animals. It ' From Trans. Chic. Path. Soc, 1907. This work was supported chiefly by an appropriation made by the legislative assembly of Montana, and also by Missoula and Ravalli counties, Montana, and by the University of Chicago. ' Wilson and Chowning, Jour. Infect. Dis., 1904, i, p. 31. 333 334 Contributions to Medical Science was stated that Piroplasma hominis exists in about 20 per cent of the gophers from the infected side of the valley but was not found in those from the east side. They furnished no experimental evidence in favor of the tick theory. Anderson/ and later Stiles,^ of the Hygienic Laboratory, also studied the disease, the former advocating the tick theory, whereas Stiles discredited it and failed to identify Piroplasma hominis. Neither performed experiments bearing on the tick theory. Hence when I began my investigations in the Bitter Root Valley in April, 1906, the tick theory was little more than a h>^othe- sis, supported only by the coincidence of the season of the adult tick with that of spotted fever, and the almost constant history of tick wounds shortly preceding the onset of the disease. Further- more, the vaHdity of the latter point was considered questionable in some quarters, inasmuch as a great many people are tick-bitten every spring without suffering from spotted fever as a consequence. Although this circumstance would not render the tick theory untenable it has caused a great deal of local prejudice against it. Locally there is a strong disposition to attribute to the water of certain streams the power of carrying the poison to man, presum- ably from foci of decaying vegetation. It is not necessary in this place to go into detail regarding the conditions which a priori lent to the water theory a degree of plausibility. Since the study of spotted fever had hitherto been limited to the season during which man is attacked it appeared of great importance to discover susceptible laboratory animals, if possible, in order that experimental work might proceed without serious interruption. As the result of investigation it was soon deter- mined that the guinea-pig and monkey (Macacus rhesus) are highly susceptible, that rabbits are almost non-susceptible, and that rats and mice exhibit no outward sign of illness when infection is attempted.^ The following points are taken as proof that the disease repro- duced in these animals is spotted fever: (i) The incubation period approximates that in man in so far as it has been ascertained in ■ .\nderson, Bull. No. 14, Byg. Lab., U.S. Pub. Health and Mar. Hasp. Sen., Washington. ■ Stiles, Bull. No. 20, ibid. > Ricketts, Jour. Am. Med. Assn., 1906, 47, p. 33. (See p. 278 of this volume.) Transmission of Spotted Fever 335 the case of man; (2) the course of fever, the eruption and anatomic changes found at autopsy correspond closely with these features in man; (3) as in the case of man it has not been possible to culti- vate a micro-organism from the blood or organs of typical cases; (4) indefinite transmission from animal to animal appears to be possible (100 generations). Inasmuch as the disease was conveyed to the latter by inoculat- ing them with the blood of human patients, and since it can be transmitted indefinitely from animal to animal in this way by means of the blood, two very important points regarding the nature of the disease are thereby estabHshed: First, that the disease is caused by a virus which is capable of propagation, a Hving organism, for otherwise the consecutive inoculation of 100 generations of animals would not have been possible; second, spotted fever is a systemic or generalized infection, since the circulating blood is invaded by the micro-organism. Inoculation experiments, per- formed later, showed that all the important organs of diseased animals are heavily infected.^ Further experiments relating to the distribution of the virus among the blood elements showed the following:^ 1. The cell-free serum of infected blood contains large amounts of the virus. 2. The virus is present in considerable quantities in an artifi- cially induced leukocytic exudate, free from erythrocytes, and also in the overlying fluid from such an exudate after the cells have been sedimented by centrifugation. 3. Prolonged centrifugation of serum does not free the over- Ijdng portion from the virus. 4. It is impossible to free the blood cells in defibrinated blood from the virus by ten to twelve washings with physiologic salt solution, although this process decreases the virulence of the blood. During the washing a point is reached at which the fluid used in washing no longer contains the virus in infective quantities. It was hoped that experiments of this nature might bring out evidence in favor of or against the piroplasma theory of Wilson and Chowning. I beheve, however, that the results do not 'Ibid., 1906, 47, p. 1067. (See p. 291 of this volume.) ' Ricketts, Jour. Infect. Dis., 1907, 4, p. 141. (See p. 299 of this volume.) 336 Contributions to Medical Science justify positive conclusions for or against this theory. In the experiments cited above the cell-free serum proved rather highly infective; this event, however, does not preclude the possibility of a piroplasmatic infection, for, although piroplasmas invade the erythrocytes extensively in all known piroplasmic diseases, it is self-evident that the plasma must be the medium through which the organisms reach the erythrocytes. Hence in all piroplasmoses there must be many extracellular organisms at some stage of the disease. Similarly the result quoted above in paragraph 2 does not of necessity disqualify the piroplasma theory, nor does it support strongly the possibility of a predominant leukocytic invasion by the parasite. The results stated in paragraph 4 would seem to point rather significantly to an intimate relation between the virus and the red or white cells. However, this does not of necessity imply an intracellular location of the parasite, for it is conceivable that the microbes may adhere to the surface of the blood cells either through mutual viscosity or some other attractive or adhesive force. I obtained no positive result in attempting to liberate the organism from its hypothetical position within the blood cells, by crushing the latter in a porcelain ball-mill. This procedure decreased the virulence of the blood, which probably was due to the crushing of many of the organisms in the mill. In brief, this method of experimentation does not give results which speak conclusively for or against the intracellular location of the parasite, and for this reason, if for no other, the experiments may have some worth. That the microbe is either very small or of low specific gravity is suggested by the result stated above in paragraph 3. I have made many attempts to pass the virus as it exists in serum through the porous Berkefeld candle filters, the serum being diluted as much as ten-fold with physiologic salt solution. In no instance has the virus passed through the filters, at any rate in infective quantities, although nine cubic centimeters of serum were used in one experiment. The failure of such a large quantity of filtered serum to produce infection, when it is known Transmission of Spotted Fever 337 that the unfiltered serum contains the virus and that fresh diseased blood is pathogenic in doses of 0.05 to o. i c.c, is significant. The result suggests that the organism, though undoubtedly minute, is of such size that it should be recognized by the use of high magni- fications, or that it is of peculiar form or possesses such adhesive properties that it is not readily filterable. Experiments on the resistance of the virus to cold, heat, desic- cation, dialysis, and other agents have as yet furnished no dif- ferential clue as to the protozoan or bacterial character of the microbe. That recovery from one attack of spotted fever renders an animal resistant to further attempts at inoculation is one of the important practical facts brought out by my experiments. Studies in im- munity have shown that a disease which will incite immunity in one species is very likely to have a similar efifect on other susceptible species, hence it is extremely probable that man generates an immunity similar to that of the guinea-pig which has recovered. This supposition is strengthened from the fact that no authentic instance of two attacks in man has been observed in so far as one can learn. It is, then, somewhat reasonable to hope that an efficient vaccine or curative serum may be prepared provided the micro- organism lends itself to suitable manipulations. I have certain guinea-pigs which are called hyperimmune. Since their recovery from spotted fever they have received many injections of infected blood. Assisted by Mr. Gomez, it has been possible to show recently that o.i to 0.3 c.c. of hyperimmune blood will protect a guinea-pig against twenty or more minimum pathogenic doses of the diseased blood. These investigations are being continued.* It is most interesting that the offspring of an immune female guinea-pig also are immune for a period, the limits of which have not yet been determined.' By showing the transmissibility of spotted fever to the guinea- pig and monkey, a means of putting the tick theory to a decisive test had been provided. In attacking this problem the first logical step was to determine « Not heretofore published. 338 Contributions to Medical Science whether the tick after feeding on a diseased animal is able to infect a healthy animal by biting it. If this should prove to be true, and if, in addition, it could be shown that the relation of the virus to the tick is a very intimate one, so that the transmission would appear to be "biological" rather than purely mechanical, the results would place the tick under the gravest suspicion as the means of infecting man. In the summer of 1906 I showed that the female tick can acquire and transmit spotted fever by means of biting (see p. 278 of this volimie),^ and King^ of the Hygienic Laboratory obtained a similar result at the same time and with the same strain (my strain) of the virus. ^ Later in the year I found that the male tick also may acquire and transmit the disease, which could well be anticipated since the male as well as the female is a blood-sucker. Further experiments have shown the existence of a very intimate relationship of the virus to the tick and that the transmission must be regarded as biological in character rather than mechanical throughout. In no instance has a tick ceased to be infective so long as it lived and would feed, a period of several months. It may acquire the disease as a larva or as a nymph, retain it during moulting, and prove infective when it reaches the subsequent active stage (see pp. 312 and 324 of this volume).'' Of great importance is the hereditary transmission of the disease by an infected female. This occurred in two of my "hered- ity experiments," and this may have an important bearing on the vitality of the disease in nature from year to year. The term "hereditary" is used rather broadly and only in the sense that ' Ricketts, Jour. Amer. Med. Assn., 1906, 47, p. 358. ' King, Pub. Health Rep., 1906. ' At the time this work was done I had no knowledge of previous experiments which demonstrated the ability of the "wood-tick" to transfer spotted fever. Recently I have had authoritative information that Dr. L. P. McCalla and Dr. H. A. Brereton, of Boise City, Idaho, transmitted the disease from man to man in two experiments. The tick was obtained "from the chest of a man very ill with spotted fever," and "applied to the arm of a man who had been in the hospital for two months and a half, and had lost both feet from gangrene due to freezing." On the eighth day the patient became ill and passed through a mild course of spotted fever, leaving a characteristic eruption. The experiment was repeated by placing the tick on a woman's leg, and she likew^ise was infected with spotted fever. Although these results re- ceived no publicity other than that given in reports to local societies, I take pleasure in according to Dr. McCalla and to his colleague, Dr. Brereton, the credit of having first shown that the tick may act as a carrier of spotted fever. Their experiments preceded by more than a year those of Dr. King and of myself, and they were pyerformed with the full consent of the patients concerned. < Ricketts, Jour. Am. Med. Assn., 1907, 49, p. 24; ibid., p. 1278. Transmission of Spotted Fever 339 the infection was derived from the female. It may be argued that during the process of oviposition, in which the head parts play a function, virus may have been deposited on the surface of the eggs from the saUvary glands of the female and that the microbes reached the embryo or its food supply by penetrating the egg membrane, or that the virus having been deposited on the surface of the egg, the embryo later became infected by con- suming a portion of the membrane. The pertinence of the first argument must be recognized, although the virus has shown no penetrating power for the skin of man or for the mucous membranes of man and animals, membranes which would seem to be much more susceptible to penetration than the dense covering of the tick's egg. The second argument could not be considered as valid because the life of the virus when exposed to the light is known to be much shorter than the time required for the development of the embryo. Such considerations, however, do not aflfect the practical fact that the larvae of infected females may be born with the virus of spotted fever in their saHvary glands, and that they are sometimes able to transmit the disease to a susceptible animal. "Hereditary" transmission was attained in only two out of twenty-six experiments, a fact which fits well with the scarcity of infected ticks in nature (see below) and in the rarity of the disease in man. I am inclined to the view that the tick during the time it harbors the virus undergoes a generalized infection. I take it that the successful "heredity experiments" signify uterine, hence general- ized infection. I have shown by inoculation experiments that the virus is present in the alimentary sac and in the salivary glands of the infected adult tick. Undoubtedly the ahmentary sac becomes infected primarily through the diseased blood ingested by the tick, and it is probable that the localization of the virus in the salivary glands is a part of, or residuum of, a generalized infection. The second, and one may say the final, step required to prove that the tick is the agent through which man is infected, involved the discovery of infected ticks in nature. I say the final step, because it would complete a chain of the strongest possible circum- stantial evidence tending to incriminate the tick. 340 Contributions to Medical Science On account of the disproportion between the number of tick- bitten persons and the number of cases of spotted fever occurring any particular season, it was anticipated that the search would be a di£5cult task and, with negative results, might require the testing of many thousand ticks before one would be justified in acknowl- edging failure. In order to increase the probability of finding infected ticks in nature those selected for the tests were taken from localities known to be infected, some being obtained from horses and cattle and others from the clothing of different individuals who had collected them in this way from the woods. They were allowed to feed in groups on normal guinea-pigs, and for a period which is longer than that required for transmission. Five hundred and thirteen ticks were tested in this way and of this number two hundred and ninety-six were found attached to the guinea-pigs. It is quite certain that this does not represent the full number which fed, since in searching the skin of the guinea- pig for ticks a few may unavoidably be overlooked. Eventually one or more infected ticks were found among a group of thirty-six males which had been collected from apparently healthy horses. The guinea-pig on which they fed developed fever on the seventh day after being placed with the ticks and died seven days later, exhibiting the clinical and anatomic changes which render spotted fever easily recognizable in this animal. No micro-organism could be cultivated from the blood or organs of the animal, which, as stated above, is the customary result of cultivation experiments. Seventeen ticks were seen attached to the guinea-pig, the first being discovered on the second day. The experiment was carried further by producing the disease in another guinea-pig through the injection of blood from the first and in a monkey by a transfer from the second guinea-pig. In addition, twenty-four of the ticks which had been placed with the first guinea-pig two days before the onset of fever were given a second test in two groups of twelve each. One of the two guinea- pigs died of peritonitis after four days. The other began a febrile course on the sixth day and was found dead five days later with anatomic changes and cultural results which rendered the diagnosis Transmission of Spotted Fever 341 of spotted fever positive. Six ticks were seen attached to this animal. The experiment was not carried further, since the ticks were still feeding when fever developed and all had had the oppor- tunity of becoming infected. No precaution was neglected to place these experiments above criticism. Each animal with its ticks was isolated in a tick-proof cage in all instances, and the cages either had never been used before or, if used, had been sterilized before being subjected to use again. No tick of uncertain identity was utilized. The finding of infected ticks on the horse does not necessarily implicate the horse as a factor in the maintenance of the disease in nature. In so far as I can judge by preliminary experiments, this animal is not very susceptible to spotted fever. I have accordingly shown the following points concerning the relationship of the wood-tick to Rocky Mountain spotted fever : 1. The disease may be acquired and transmitted by the larva, the nymph, and the adult male and female, that is to say, by the tick during any of its active stages. 2. The larvae of an infected female are in some instances infective. 3. The association of the virus with the tick is very intimate, as shown by the long duration of infectivity of the tick, the retention of infectivity during moulting, and the "hereditary" trans- mission of the virus; hence there is some reason to believe that the tick suffers from a relatively harmless generalized infection, and that the virus proliferates in its body. The disease probably is transferred through the salivary secretion of the tick, since the salivary glands of the infected adult contain the virus. There seems to be no doubt but that the arachnid, when biting, injects some of the salivary secretion into the skin; this is indicated by the marked local reaction which follows the bite of even the normal tick. 4. Infected ticks exist in nature in small numbers in the so- called infected districts. When one adds to such data the generally known facts that the disease in man is limited to the season of prevalence of the adult 342 Contributions to Medical Science tick, the larvae and nymphs rarely feeding on man, and that a history of recent tick bite may be obtained in nearly every case of spotted fever, and cannot be positively excluded in the event that such history is not given, it seems necessary to conclude that the chain of circumstantial evidence is now complete, and that the acceptance of the theory of transmission by the bite of the tick is now justified. I have failed, and Mr. P. G. Heinemann who assisted in the work this year Hkewise failed, to cultivate or identify positively the micro-organism. It has not been possible so far to identify or accept the piroplasma of Wilson and Chowning as the cause of the disease. Concerning the gopher hypothesis, occasional susceptible animals are found among the local species, but it is not certain that the gopher plays the part in the perpetuation of the disease which Wilson and Chowning assumed that it plays. The way is now open to carry on an intelligent campaign against the disease by destroying massive numbers of the ticks, through the use of methods which have been so successful in the fight against Texas fever of cattle in the southern states.* ' Since this article was written the search for a vaccine has progressed so favorably that there is reason to believe that vaccination will prove an efEective means of preventing the disease in man. This work will be reported in detail in the near future. STUDIES ON IMMUNITY IN ROCKY MOUNTAIN SPOTTED FEVER.^ H. T. RiCKETTS AND L. G O M E Z. {From the Departments of Pathology and Experimental Therapy, The University of Chicago.) In these studies it has been our aim to ascertain with certainty- some of the fundamental facts concerning immunity to this disease and to pass as rapidly as possible to the practical aspects of the question, namely, the study of serum prevention, serum therapy, and specific vaccination. Concerning these three important phases of the work, we report preliminary experiments which indicate the scope of our studies, the results of which will be reported more exhaustively in the future. Diagnosis of the Experimental Disease. Before considering the experiments it seems desirable to point out the essential clinical and anatomical features of experimental spotted fever in order to show that the criteria for the recognition and differentiation of the disease are of a definite and convincing character. While diagnosis is usually easy concerning infections caused by organisms that may be cultivated and that show definite biologic characteristics, or that have a definite morphology, as in the case of certain protozoa, it would appear to be more difficult and liable to error wnen the organism is unrecognized and uncul- tivated. The incubation period is definite and is never absent in animals that were fever free and that suffered an accident at the time of inoculation such as might be caused by puncture of the intestines or some accidental infection. Following intraperitoneal inoculation in the guinea-pig, monkey, or rabbit,' two to four or five days elapse before a distinct rise in the temperature occurs. The incubation period is from two to four days longer when the inoculation is subcutaneous. Its length also has a cer- tain relation to the quantity of virus inoculated; when the minimum pathogenic dose is used it is often from one to three days longer than when several multiples of this dose are injected. At the close of the incubation period the temperature rises to a maximum rapidly. Not infrequently it is normal (about 102 . 5) on one day, and on the next day, 105 or » From Jour. Infect. Dis., 1908, 5, p. 221. The essential points of this paper were presented before the Section in Pathology and Physiology of the American Association for the Advancement of Science, January 2, 1908. This work has been supported largely by the Department of Experimental Therapy, The University of Chicago, through the kindness of Professor S. A. Matthews, although the appropriation of the Department of Pathology and funds provided by the state of Montana have been drawn on to a certain extent. ' The rabbit has recently been found to be susceptible to the strains now cultivated in laboratory animab. (See Gomez, " Rocky Mountain Spotted Fever in the Rabbit," Jour. Infect. Dis., 1909, 6, p. 383.) 343 344 Contributions to Medical Science even io6.' More often, however, the first day of the febrile period is marked by a temperature of 103 to 104 with a more pronounced rise on the second and following days. The course is that of a continuous fever throughout, as in man. For working purposes it is satisfactory to take the temperature but once a day and it is immaterial whether this is done in the morning or afternoon. Sometimes the maximum tem- perature is not reached until the third or fourth day of fever. Usually the high point is about 106, although a temperature of 107 . 2 has been observed. In fatal cases the duration of the fever is subject to variations, some dying after four or five days and others after eight or ten days of fever. Similarly animals which recover may exhibit fever for from six to ten days. In recovery subsidence of the temperature is usually gradual. Occasionally, however, a drop to normal is noted within 24 hours, and it may be shghtly below normal for one or two days. In fatal cases the reduction is sudden, and a drop to 97 or 99 is an unfaiUng sign of impending death. The most characteristic external signs of the disease are emaciation, a roseolar eruption, hemorrhages into, and gangrene of, the external genitalia. The roseolar eruption begins on the third to the fifth day of fever and is best seen on the external genitalia of guinea-pigs with white skins. If such animals are shaved it may also be detected on other parts of the body, especially the thighs, back, and face. It is not always recognizable and escapes observation entirely on animals having dark skins. In males the scrotum begins to swell after from two to four days of fever and it is characteristic for the skin at this point to become densely infiltrated with blood; in the event that the animal lives long enough, i.e., in the event of recovery, the hem- orrhagic areas become gangrenous and separate, leaving greater or less deformity of the scrotum. These phenomena are not absolutely constant but occur in a high percentage of the males. Apparently the swelling of the scrotum is due chiefly to congestion and edema of the underlying connective tissue, the tunica vaginalis and the epididymis. In females, the vulva becomes swollen and may show the roseolar eruption as stated; hemorrhage and gangrene of the vulva, however, are rather un- common. Frequently the ears suffer from gangrene to a greater or less extent and it is the impression that this phenomenon now occurs more frequently than when the virus was first cultivated from man. Occasionally both ears fall away to their bases. Less frequently the soles of the feet undergo gangrene. In fatal cases the g\iinea-pig loses approximately one-third of its weight during the disease. The most characteristic findings at autopsy, aside from the changes in the geni- talia, are a greatly enlarged spleen and enlarged and hemorrhagic lymph glands with extreme congestion of the adjacent areolar tissue. Enlargement of the suprarenal glands is of frequent occurrence and they sometimes contain small hemorrhages. A valuable point for the corroboration of the diagnosis is the failure to cultivate any micro-organism from the blood or organs of animals that have been killed during ' It has been the rule in taking the temperature of guinea-pigs to insert the thermometer as far as possible into the colon. If this distance is not more than two inches, it is usually because the colon is occluded by feces, and this condition is recorded at such times. In animals of 300 to 400 grams weight it is usually possible to reach a depth of about three inches, and in full grown guinea-pigs the thermome- ter can be inserted for almost its entire length. (This is the ordinary clinical thermometer.) A large number of observations indicates that the normal temperature of the guinea-pig, as taken in this way, lies between 102.2 and 102.6. A temperature above 103 would seem to be abnormal, although occasional groups of animals are encountered in which the temperatures may vary between 103 and 104.5, without any evident cause. This may persist for several days and, naturally, causes confusion in the recognition of spotted fever. Immunity in Spotted Fever 345 the course of the fever, or which have died as a natural result of the infection. This statement refers to typical cases, in which the clinical course indicates that there was no mixed infection, and in the case of autopsies it applies to those that are performed immediately or soon after death. This negative bacteriologic finding has been constant through two years of work and although it is of the greatest importance as a diagnostic point, the clinical course of the disease is so characteristic that cultures are now resorted to only occasionally to prove the purity of the virus used for passage, and in experiments to serve as a check for animals in which there has been some departure from the normal course of the disease or in which the autopsy reveals unusual conditions. If the animals used for an experiment are free from fever when the experiment is begun, and if fol- lowing injection they or their controls exhibit the usual incubation period followed by a typical course of fever and the development of the cutaneous phenomena, and if the findings at autopsy are as described above the experiments are sufficiently controlled for working purposes. This method has been found perfectly satisfactory and in two instances was sufficient for the recognition of adventitious epidemics, the presence of which was subsequently verified by bacteriologic methods. In view of the fact, as stated later, that an attack of spotted fever, however mild, confers strong immunity, an "immunity test" is of value in determining whether animals which recover have or have not suffered from spotted fever. Our experience indicates that any used animal which develops spotted fever after the injection of 20 to 50 multiples of the minimum pathogenic dose of virus could not have had the disease in the first instance. The Virus. For the major part of the work two strains of the virus have been used, one obtained in the spring of 1906 and the other in 1907. The first is known as the Bradley strain, the second as the Eddy strain. During the summer of 1906 the Bradley strain was grown alternately in monkeys and guinea-pigs and subsequently in guinea- pigs alone. The Eddy strain has been cultivated in guinea-pigs exclusively. The method of cultivation is that of the ordinary passage, the blood of an infected animal being injected intraperitoneally or subcutaneously into a healthy one. The Bradley strain has undergone approximately 125 passages. The two strains show about the same degree of virulence judging from the quantity of virulent blood required to produce infection. On this account they have been used more or less interchange- ably, a record being kept of the strain used for each experiment. It is important to know whether the virus used in these experiments is the same specifically, as when first obtained. We have the following evidence that it is: Ani- mals that had recovered from the infection with the virus after it had been culti- vated in the guinea-pig for one year were immune to the inoculation of fresh virus obtained from man in the spring of 1907. Of two strains, one obtained in 1906 and one in 1907, each confers immunity against the other. This, of course, is independent of the possibility that the virus has undergone variations in the character of its patho- genicity as a result of prolonged cultivation in the guinea-pig. The defibrinated blood of infected animals, the blood being drawn from the carotid artery or the heart, is the source of the virus, and when the term "virus" is used it is understood that it refers to such blood. The virus has been standardized as accurately as possible in terms of the minimum pathogenic dose. It is impossible 346 Contributions to Medical Science to deal with the minimum lethal dose because of the fact that minimum infective doses are about as likely to prove fatal as several multiples of such doses. In most instances the minimum pathogenic dose lies between o.oi c.c. and 0.03 c.c. The former frequently fails to produce infection whereas the latter is nearly always infect- ive. For quantitative work it is important to draw the virus with respect to the duration of the infection, because of the possibility that antibodies which may appear later in the course may lessen infectivity. As illustrating the importance of this consideration, two. strains were lost in the spring of 1906 by waiting until the animals had died before attempting to perpetuate the disease in succeeding generations. We have gained the impression that virus which is taken rather late in the disease is more likely to produce infection, if it is diluted several fold with salt solution. It has happened a number of times that small doses were infective whereas larger doses of the same virus were not. Such an experience occurred recently when infected blood from a horse, drawn rather late in the disease, was injected into two guinea- pigs in doses of i c.c. and 5 c.c; the former amount caused spotted fever whereas the latter did not. This was not due to lack of susceptibility on the part of the latter animal, inasmuch as it contracted the disease later when inoculated with virus from the guinea-pig. We are inclined to the view that the failure of large doses to infect under these circumstances is due to the comparatively large quantity of antibodies in the larger quantities of blood. It is true that the proportionate quantity of anti- bodies to the number of micro-organisms in 5 c.c. is the same as in i c.c. of blood, yet when the smaller quantity is used the antibodies are subject to greater dilution and it is in accordance with the known action of other antibodies to assume that their effect on the micro-organisms would be somewhat attenuated in this instance by their dilution. Also there are good general grounds for believing that antibodies may be fixed or destroyed by the tissues to a certain extent, and if this is true the residuum of effective antibodies, in proportion to the number of micro-organisms, would be lower when small quantities of virus are introduced than when larger quantities are given. For most purposes the injections have been intraperitoneal, although the subcu- taneous route has been employed for certain e.xperiments. CONCERNING IMMUNITY IN MAN. In SO far as we have been able to learn, there are no data regard- ing the immunity of man to Rocky Mountain spotted fever, although our knowledge of the matter is limited to the conditions in the Bitter Root Valley in Montana. There is no authoritative example of two attacks of the disease in the same person in this locahty. It is true that one or more individuals state their behef that they have had more than one attack, but such statements lack the confirmation of the local physicians who are experienced in the recognition of the disease. The results obtained in the monkey and the guinea-pig suggest that man probably acquires a strong immunity as a result of an attack. This is a question, however, which can be approached Immunity in Spotted Fever 347 from the experimental side by a study of the serum of those who have recovered. ACTIVE IMMUNITY IN THE MONKEY. The fact that one attack of spotted fever renders the monkey immune to a second inoculation, performed some months later, has aheady been reported by one of us (see p. 299 of this volume).'' An additional experiment illustrating the same point may be cited briefly. On May 20, 1906, the monkey, a large M. rhesus, received intraperitoneally 10 c.c. of defibrinated blood from the patient. Porter. After an incubation period of about two days its tempera- ture rose rapidly to 105.3 ^^ the vicinity of which it remained for three days, and then gradually returned to normal during the course of five days more. The animal showed no eruption, although his eyes became much reddened. His blood was not inoculated into other animals in order to confirm the diagnosis of spotted fever and on this account there naturally existed a good deal of doubt as to whether the monkey had been infected with spotted fever by the inoculation. On February i, 1907, the monkey was given an immunity test, a control monkey (No. 24) being inoculated with the same quantity of virus. The inoculation caused no rise in the tempera- ture of the animal, whereas the temperature of the control, after a brief incubation period, rapidly rose to 106. i, near which it remained until the animal was killed, on the sixth day after inocula- tion.^ On the second day of high fever the control presented a roseolar eruption of the skin of the perineum, which was hemor- rhagic at the time it was killed. The blood of this animal pro- duced spotted fever in guinea-pigs. The experiment reported previously and the one just cited indicate that an attack of spotted fever confers immunity in the monkey, which in the last instance was present in a marked degree for nine months. The quantity of virus injected was very large, consisting of defibrinated blood and a dense emulsion of the liver, ■ Ricketts, "Further Observations on Rocky Mountain Spotted Fever, etc.," Jour. Am. Med. Assn., 1906, 47, p. 1067. ' It was the custom at that time to kill the monkeys during the height of the fever in order to obtain quantities of virus for further experiments. 348 Contributions to Medical Science spleen, and kidneys of an infected guinea-pig, a total quantity of 15 c.c. ACTIVE IMMUNITY IN THE GUINEA-PIG. In a previous article it was stated that the guinea-pig that has recovered from spotted fever is thereafter immune to the disease, although the details of experiments were not given at that time.' This result has proved to be a rule to which no exception has been found among a great many tests. An illustrative experiment will be cited in this place and other instances are found through the remaining part of this report. Guinea-pig 710 was inoculated with spotted fever by larvae of an infected tick. Twelve days after the first larvae were seen to be attached, the temperature of the guinea-pig rose suddenly to 105.2, and on successive days was read at 105.3, 104 -8, 104.3, 103.8, 102.3, ^^^ 102.2. This was a short course of fever and some doubt existed concerning the correctness of the diagnosis until the result of a transfer inoculation was known. On the day when the temperature of guinea-pig 710 registered 104.3, 3 c.c. of blood were drawn from the heart and injected intraperitoneally into guinea-pig 749. The latter passed through a typical course of spotted fever, the temperature being as follows on succeeding days after inoculation: 103.4, 103.4, , 105.6, 104.8, 106.7, 106.4, 106, 104.6, 105.4, 104.4, io4j I04-3) io3> 10355 recovery. The scrotum of 749 became hemorrhagic and sloughed in a typical manner, and the animal, furthermore, resisted a subsequent immunity test. After these experiments had proved beyond doubt that guinea-pig 710 had been infected by the larval ticks, the animal was given an immunity test by the intraperitoneal injection of 2 c.c. of virus from guinea-pig 781, which had been inoculated with the Eddy strain of the virus. On successive days the temperature of guinea-pig 710 was as follows: 103.4, 103, 102.6, 102.4, 102.3, 102.5, 102.7, 103. 1, 102.3, J 103.6. A good control for the experiment cited is found in the history of guinea-pig 705 which also was used to determine whether or not the infected female tick transmits the disease to her progeny. Twenty-two days after the larvae of an infected tick were placed ■ Ricketts, loc. cil. Immunity in Spotted Fever 349 with the animal, the latter had four days of moderate fever: 104. 3, 104. 5, 103 . 1, 104. 1, 102 . 1. This event suggested that the guinea- pig may have been given a mild infection by the larvae, but this was by no means certain because the animals, confined and often exhausted as they are during such experiments, sometimes exhibit abnormal temperatures for one or several days which is entirely independent of infection with spotted fever. About three months later this animal received an injection of 2.5- c.c. of Bradley virus with the following result: 102, 102.5, 102.8, 105, 105. i, 104.7, 104.8, 103.9, 99.6; death. The figures represent the temperature of the animal on successive days. On the seventh day after inoculation the vulva became hemorrhagic; the autopsy showed typical changes in the spleen and lymph glands. Cultures were sterile. It is a matter of first importance that a mild attack of spotted fever results in the formation of strong immunity. Usually, when small doses of virus are given, the animals either react with a typical and severe attack of spotted fever or they show no sign of infection whatever and in the latter instance it has been the custom to use them as duplicates in the course of routine passage. Occasionally, however, such animals have been found to be resistant to infection. For example, guinea-pig 1064, which in a particular experiment had been injected with o.oi c.c. of infected serum, showed no disturbance other than one day of distinct fever, the following being the record of the temperature: 103, 8, 102.8, 102.4, 102.8, 102.6, 104.8, 102.6, 103.2, 102.8, 103.2. Nine days after the first injection, it received intraperitoneally i . 25 c.c. of virus, diluted with salt solution, which had been kept in the ice-chest for five days. Although a control developed the disease, the animal was undisturbed by the inoculation, exhibiting normal temperature. A second immunity test was given 15 days later with the same result. One might be inclined to explain such occurrences by the assumption that occasional guinea-pigs are immune to the disease naturally. Two conditions, especially, argue against the correct- ness of this assumption: First, we have found no guinea-pigs which are immune to the infection when inoculated with a moderate 350 Contributions to Medical Science quantity of the virus taken from an animal early in the disease, the minimimi quantity necessary for infection being in the neigh- borhood of 0.02 or 0.03 c.c. Second, it is not difficult to repeat the result just quoted by injecting suitable quantities of virus and immune serum, the existence of active immunity being proved by an immunity test which is given after the passive immunity estabHshed by the injection of the immune serum has disappeared. This will be referred to again under the subject of "protective inoculation." The active immunity is probably of long duration. Animals have been found to resist infection with excessive doses of virus more than a year after they suffered from the disease. The occurrence of hereditary immunity, to be referred to below, also indicates the profound change which the disease produced in the guinea-pig as well as the permanent character of the immunity. There can be Httle doubt that the cause of the active immunity lies in the anti-infectious properties which may be demonstrated in the blood and serum of animals which recover, and perhaps also in the acquired power of the tissues of the immune animal to produce additional antibodies readily when fresh virus is introduced at a subsequent time. The properties of the serum are considered below under "Passive Immunity." At present we have no means of deciding whether the antibodies are antitoxic, germicidal, or opsonic. Theoretically all three may be represented. HEREDITARY IMMUNITY. The offspring of a female guinea-pig that has recovered from spotted fever are endowed with a strong and protracted immunity. This has been demonstrated many times and the resistance is trans- mitted regardless of the degree of infection of the female parent, a fact which is shown in the case of certain "vaccinated" female parents which exhibited a minimum febrile reaction. The influence of the immune male in the transmission of the immunity has not been investigated. Illustrative experiments will be described. The female guinea-pig 663 was infected by male tick No. 5 on June 30, 1907. The following was the course of the temperature: 101.6, 102.4, 102.2, 103.7, 104, 105.4, 105.4, 105.2, 104.9, Immunity in Spotted Fever 351 104.6, 104.7, 103-8, 102.7; recovery. She gave birth to one young on October 14. On November 27, when the young animal was about six weeks old, it was injected intraperitoneally with I c.c. of third-day virus (Bradley strain). It had been with the parent continuously until injected. The following daily tempera- tures were recorded: 103, 103.6, 103.9, I03-2, 103, 102.5, 103.4, 103. 1, 103, 103.8, 103, 103.8, 102.6, 103. 1, 102.7.* A control guinea-pig of the same age, from a normal female, inoculated with the same dose of virus, showed the following daily tempera- tures: 103, 102.8, 102.8, 105.6, 107, 106.3, 106, 104.4, 104.5, 103.2, 100.2; death. The anatomical changes at autopsy were typical of spotted fever. This experiment is important as showing that a female may transmit the immunity to her young although her infection ante- dated the period of her pregnancy by several months; that is, the cells of the embryo were not stimulated to form protective sub- stances by the presence of virus; only the germ cell could have been subjected to such an influence.^ The following "exchange experiment" indicates that the inherited immunity does not depend entirely on the milk which the young derive from the immune parent. Immune young which sucked immune parent. — Guinea-pig 938, 30 days old. Sucked for 14 days, was then removed, and 16 days later was inoculated. Had no distinct rise in temperature. Guinea-pig 939, 34 days old. Remained with parent until inocu- lated. No distinct fever. Immune young which sucked normal parent. — Guinea-pig 935, 29 days old. Remained with normal parent until inoculated. No distinct fever followed. Guinea-pig 936, 34 days old. Was with normal parent for 26 days. No fever followed inoculation. Normal young which sucked immune parent. — Guinea-pig 937, 30 days old. Course of temperature following inoculation: 102 .8, 102.4, 104.2, 105.3, 105.5, 105.6; death. Autopsy: typical of spotted fever. ' In oiir experience the young guinea-pig frequently has a higher average temperature than the adult and it is subject to greater fluctuations. ' Spotted fever in the guinea-pig is strictly an acute infection. The blood of the animal which recovers is never infective for other animals, but on the contrary is protective, as stated later. 352 Contributions to Medical Science Normal young -which sucked normal parent. — Guinea-pig 940, 29 days old. Daily temperatures following inoculation: 102.4, 102.5, 103-8, 104.5, 105, 105. 1, 98.7; death. Autopsy: typical of spotted fever. The duration of the inherited immunity has not been definitely determined. In one instance it was present two and one-half months and in another three months after birth. We have not yet studied the character of the inherited immunity. In the event that the infection of the parent occurred before her impregnation, there is reason to think that the immunity of the young is passive in character. If infection occurs during preg- nancy, there may be opportunity for the establishment of active immunity on the part of the embryo. However, infection occur- ring at this time usually results in abortion. In the event that the condition is one of passive immunity, it differs from the passive immunity established by the injection of immune serum by its greater duration. The immunity conferred by the injection of I c.c. of immune blood into a guinea-pig has disappeared largely after the lapse of 30 days. It is to be noted, however, that the oflfspring of an immune female may contain in their body fluids a much larger quantity of protective substances than is introduced in I c.c. of immune blood, and that the elimination of the larger quantity may require a much longer time. This subject will be investigated further. passive immunity. The whole blood, defibrinated blood, and serum of animals that have recovered from spotted fever possess strong protective powers when injected with virus into healthy guinea-pigs. For the sake of convenience immune defibrinated blood, rather than serum, has been used for most of the experiments, although it has been deter- mined that the protective powers lie in the serum. When the term "blood" is used, it refers to defibrinated blood. For some of the experiments on passive immunity the blood of animals which have recovered and which have been subjected to no further treatment has been used; in other cases so-called "hyperimmune" blood was employed. This is the blood of Immunity in Spotted Fever 353 guinea-pigs that after recovery received a series of injections of infected blood with the hope of increasing the quantity of pro- tective substances. We doubt, however, whether such a practice is of much service in producing an increase in the amount of anti- bodies. In two instances the blood of guinea-pigs which were treated in this way showed an actual decrease in their protective power which fell below that of other immune animals that had not received fresh injections of virus. Both of these animals were bled repeatedly during the immunizing process, however, and the decrease may have been due to the dilution accompanying the restitution of the blood. As a rule, the blood of guinea-pigs that have recovered recently has shown as great protective powers as that of "hyperimmune" animals. It is not unlikely that the concentration of antibodies in the blood of the immune animal decreases with time. There is a possible source of error in deter- mining this point, however, in that a number of successive bleed- ings of the same guinea-pig brings about a dilution of the protective substances in his body simulating a decrease by elimination or destruction within the body. Experiments bearing on this point have not been brought to completion. The blood of guinea-pigs that have recovered from the disease recently protects in doses of from o.i to 0.3 or 0.4 c.c. against infection with i c.c. of third-day virus, a quantity which represents at least from 30 to 60, and in some instances 100, minimum patho- genic doses. This result is obtained when the immune blood and virus are mixed before injection, the inoculations being intraperi- toneal. Preliminary experiments indicate that the protection is not so pronounced when the two are injected into different parts of the body at the same time, although the difference is not great. The following experiment gives an approximate idea of the duration of the passive immunity when i c.c. of immune blood is injected subcutaneously (Table i). The immune blood in this instance was a mixture, obtained from three immune guinea-pigs (862, 878, 778), two of which had recovered from spotted fever one month previously and the third about three months previously. One-tenth cubic centimeter protected against i c.c. of third-day virus in controls, the mixtures being injected intraperitoneally. 354 Contributions to Medical Science The experiment animals received the virus subcutaneously at different periods following the injection of the immune blood. The relation of the duration of the passive immunity to the quantity of immune serum injected has not been determined. TABLE I. Duration of Passive Immunity. Guinea-Pig Interval between Injec- tion of Serum and Virus Quantity of Virus Injected Result 20 days 33 38 45 49 55 I.O c.c. I.O l.O I.O o-S 0.03 Slight fever for 5 days Moderate fever for s days Severe fever for 9 days ' I002 period of one week The converse of this experiment was performed also; a number of guinea-pigs were inoculated subcutaneously with i c.c. of virus each, and the subsequent period determined at which i c.c. of immune blood would prevent the development of spotted fever (Table 2). In this case the immune bloods were drawn imme- diately before injection and they came from different immune animals; hence they probably were not of uniform value, though certainly not differing greatly. TABLE 2. Prophylactic Effect of Immune Blood when lN.fEcrED Subsequent to Virus. Guinea-Pig Interval between Injec- tion of Virus and Immune Blood Result 851 1 day 2 days 3 4 852 other causes 8s3 the same day 8s4 three days of low fever (about 104) The experiment shows protection against 30 to 60 pathogenic doses at the least, when i c.c. of immune blood is given three days after the injection of the virus. From this time the injections of immune blood could be considered to have only a curative effect, and this will be mentioned later. Immunity in Spotted Fever 355 The results of these experiments on protection by means of immune blood or serum, of which we have abundant confirmation, would seem to have an important practical bearing. There is good reason to believe, on the basis of the results obtained, that serum prophylaxis of man is feasible, provided sufficient quantities of a serum of reasonable strength can be prepared. Manifestly it would operate successfully only in case the inhabitants of infected districts who are bitten by ticks would report for prophylactic injections within two or three days after receiving the bite. If the results obtained with the guinea-pig apply also to man, the serum when given in sufficient quantity would ward off the danger pertaining to the recent exposure, but probably could not be considered protective for a period longer than three or four weeks. A subsequent exposure would require another injection of serum. We have been able to infect the horse with spotted fever recently and are now engaged in a study of the protective power of the serum obtained after the recovery of the animal. Preliminary experiments indicate that its value differs Httle from that of the guinea-pig, hence there is reason to believe that a protective serum in desirable quantities will be available. AN EXPERIMENT IN SERUM THERAPY. We have made preliminary observations on the curative value of the immune blood from the guinea-pig and shall describe an experiment which seems to indicate that its therapeutic power is not of high degree (Table 3). This is a continuation of the passive immunity experiment described on p. 354, in which equal quantities (i c.c.) of virus were inoculated into a number of guinea-pigs and the effect of the immune blood noted when given at subsequent periods. Inasmuch as the remaining animals began to show fever on the fourth day after inoculation, it became thereafter an experiment to determine the curative value of the immune blood. It had been determined in other experiments that the normal blood of the guinea-pig has no protective power against spotted fever. Thus guinea-pig 884 received intraperitoneally a mixture of I c.c. of normal blood and 0.3 c.c. of third-day virus and died 356 Contributions to Medical Science of spotted fever after nine days. Also guinea-pig 886 received I c.c. of normal blood and 0.05 c.c. of virus and exhibited the course characteristic of spotted fever, including hemorrhage into, and sloughing of, the scrotum. It resisted an immunity test given two months after recovery. TABLE 3. The Curative Power of Immuke Blood. Virus injected December 4, 1907, subcutaneously. Injections of immune blood subcutaneously on dates indicated. Guinea-pigs designated by numbers, 855, etc. Dec. Temp. Immune Temp. Immune Temp. Immune Temp. 8SS Blood 856 Blood 8S7 Blood 858 103.8 103.4 104. 103 -9 103.4 104. 103.7 103.4 103.5 104.2 104.9 104.8 103.6 103. 103.7 103.1 105.6 2 c.c. 105.8 sec. 105. 1 c.c. 105.8 105.5 los- 105.7 106. 105.7 105. 106.3 105.7 104. s 103.4 104.8 104.6 104.2 103.9 104.2 104. 104.2 104.8 102.2 102.5 102.9 103.9 102.3 ioa.3 103.1 103.6 102. 2 102. 102. 5 103. S 102.6 102.6 103.7 103.3 103. 102.3 Recovery Recovery Recovery Recovery Immune Blood 5 6 7 8 9 10 II 12 13 14 IS 16 17 18 19 Dec. Temp. 859 Immune Blood Temp. 860 Immune Blood Controls* 103.3 104. 104.4 103.7 105.6 106.4 106.8 106. 105.8 105.2 103.6 103.2 102.7 102.8 Recovery 2 c.c. 2 2 2 103.4 103.7 104.2 103.6 104.3 106. 106.2 105.2 10S.3 105.8 103.9 104. 103.1 102.7 103.3 Recovery 3 c.c. 3 3 3 103.9 102.7 104. 103. S 105.2 106.3 106. 105.4 105.6 103.2 103.4 103. 102.7 Recovery 103.6 6:::::::::::::::::::::::::: 104. 104. 8': :::::::::::::::::: 103.7 106.4 106.2 105.7 105.1 105.8 103. 16" ;■ ::::::::::::::::::::: 103. S 103. 18 103. 1 Recovery * Inoculated with i c.c. of virus alone. Inasmuch as both controls in this experiment recovered, the influence of the immune blood can be determined only by its effect on the duration and severity of the fever, the normal duration of which in this instance was approximately six days. With this Immunity in Spotted Fever 357 as a basis of comparison, it is seen that 2 c.c. of immune blood given on the first day of fever exercised but slight if any influence on the course. Five cubic centimeters given on the first day quite positively shortened the course, since there were only three days of severe fever. One cubic centimeter given daily had no pronounced influence, although the temperature for the last three days was lower than in the controls. Two cubic centimeters given daily beginning with the second day of fever may have been a factor in the lower temperature during the last three days. The same amount given daily, beginning on the third day of fever, had no effect, and a similar result is noted when the amount was increased to 3 c.c. daily. The experiment cited seems to show quite positively that the immune blood exercises little influence on the disease unless it can be given early and in large quantities. An important practical difficulty arises here, for the disease in man is rarely recognized until the eruption appears, which may be at any time from the second to the fifth day of fever. Physicians who are experienced in the diagnosis of the infection often make a probable diagnosis of spotted fever on the first or second day of sickness, in the spring of the year, in the case of patients that reside in infected districts. The probable diagnosis should be facilitated if a history of tick bite within one to two weeks preceding the onset of symptoms is given. In such cases, a large quantity of the serum given imme- diately may exercise a modifying influence on the course of the infection; in the event that the probable diagnosis was incorrect, there would be no Hkelihood of harm as a consequence of the injection of the serum. The curative value of the immune serum from the horse is being studied carefully; inasmuch, however, as the preliminary tests have shown that it possesses about the same power as the blood of immune guinea-pigs, it is not anticipated that the results will be very different. It is not probable that a serum of greater power than that afforded by the animal which has recovered from spotted fever will be obtained until the micro-organism can be grown in desirable quantities in artificial cultures. During the course of the infec- 358 Contributions to Medical Science tion there is already in the animal's body such a quantity of the organism that 0.02 or 0.03 c.c. of his blood is infective, and at present there is no means of increasing this amount for subsequent injections. protective inoculations. For more than a year occasional attempts have been made to attenuate the virus of spotted fever by desiccation, by the addition of glycerin, and by means of heat so that it would be suitable for vaccination. Desiccation for this purpose has proved to be utterly unreliable and has been abandoned. The experiments will not be described further than to say that the principle followed was to begin the injections with comparatively large quantities of virus that had been dried in a vacuum over sulphuric acid for a longer period than was required to kill all the organisms, and for succeeding injections to use smaller quantities of virus which had been dried for shorter periods, passing finally to minute amounts of fresh virus. Infection eventually resulted in practically all the animals, and although it may in the end be possible to produce immunity in this way, the method is so tedious, long, and uncertain, that it appears to have only theoretical interest. The value of glycerin and low degrees of heat as attenuating agents is still being studied. No systematic attempt has been made to determine the vac- cinating properties of minute doses of the virus. As a rule a minute quantity either produces frank infection or causes no disturbance whatever, and in the latter case the animals are, in nearly all instances, susceptible to reinoculation. Occasional exceptions have been encountered, however. Such an animal (guinea-pig 1064) was referred to under the discussion of active immunity (p. 349). In this instance marked active immunity developed, although the animal had but one day of fever. On account of the uncertainty as to what the virus will do when injected in quantities which approximate the minimum patho- genic dose, it is manifest that minute doses cannot be utilized for practical vaccination. On account of the protective power which immune blood Immunity in Spotted Fever 359 possesses, the possibility of mixed immunization or ''sero-vac- cination" seemed promising and we have devoted some time to the study of this subject. At the beginning of the work three plans of inoculation were adopted: First, the immunizing effect of a single mixture of immune blood and virus. Second, the effect of repeated injec- tions given at short intervals, the ratio of the quantity of the im- mune blood to the virus being gradually decreased until pure virus was injected. Third, the same as the second plan, except that longer intervals were allowed to intervene between the injec- tions. The ultimate object in all three methods was to establish active immunity without causing severe infection. This was found to be readily possible with regard to the second and third methods, although the development of immunity in most instances was characterized by a mild and brief febrile reaction some time during the course of the injections. These methods, however, are protracted and tedious and are more of theoretical than of practical interest, particularly since a high degree of immunity can also be established by a single injection of virus and immune blood when the two are used in proper proportions. We shall therefore limit our description to certain experiments in which immunity was established in one instance by two injec- tions (Table 4), and in another by one injection. Following the second injection all the animals showed a certain amount of febrile reaction, although in the first three the rise in temperature was no greater than that often encountered from other causes, such as minor injuries to the colon by the thermometer. It was attempted to learn the degree of immunity which had been established in 789, which showed a minimum reaction, and in 791, which showed a more pronounced though mild reaction, by testing the protective power of their defibrinated bloods, in two consecutive experiments (Tables 5 and 6) . The first test, shown in Table 5, is not very satisfactory, particu- larly regarding the animals injected with the blood of guinea-pig 791. Guinea-pig 960 which received the largest dose of this blood (i c.c.) ran a distinct course of fever, the cause of which may not be open to determination. When sufficient time has elapsed for 360 Contributions to Medical Science TABLE 4. The Immunizing Effect of Two Injections. First injection, consisting of virus mixed with different quantities of immune blood. Injections intrapertJoneally. Immune blood from two animals (765, 766) that had recovered spontaneously. No. of Experiment Animal Virus Immune Blood Result 1 c.c. I I 1 I.S c.c. 1.0 0-7 0-3 No fever " " " " " " Second injection, consisting of i c.c. of virus, given 25 days after the first. Date Guinea-Pig 789 Guinea-Pig" 790 Guinea-Pig 791 Guinea-Pig 792 Dec. 2 102.0 102.3 102.4 102.1 101.8 103.6 102.8 102.8 103.8 102.7 102.2 '°^/; 102.0 102.6 102.6 103.4 103.8 103.2 103.4 103.7 102.7 102.2 103. 102.7 102.3 102.4 103. 103.4 104.2 103.9 104. 102.2 102.4 103. 102.8 103. i.::.'.'.:'.'.'.'.'.'.'. 104.7 105.5 8 104. 104.8 103.3 94.4 Death TABLE 5. The Protective Power of the Bloods of the Vaccinated Guinea-Pigs 789 and 791. Test 1, performed 25 days after the second injection. Virus, 1 c.c.-1-varying quantities of immune blood. Date Immune Blood from Guinea-Pig 789 Immune Blood from Guinea-Pig 791 Controls, Re- ceiving I c.c. OF Virus EAcnf I c.c. No. 957 0.7 c.c. No. 958 0.3 c.c. No. 959 1 c.c. No. 960 0.7 c.c. No. 961 0.3 c.c. No. 962 No. 965 No. 966 Dec. 27 29 30 T 31 Jan. I 2 3 4 s 6 7 8 9 10 II 102.8 102.9 102.6 102.6 102.4 102.4 103.2 102.8 102.6 102.4 101.6 103.2 Recovery 102.8 103.1 103. 103.6 103. S 103. 104.4 104.3 104. 103.4 103.5 103. 1 103. 5 Recovery 103.0 103.2 104.3 105.2 103.9 104.8 104. 102.8 103.6 104.3 103.2 104.3 104. Recovery* 103.0 103. S 103.8 104. 103.5 104.7 104.3 104.5 104. 104.7 104.8 103.9 102.8 Recovery 102.8 103.3 104. 104. 103.4 103.2 103.2 102.5 102.3 102.4 102.3 103. Recovery 102.7 102.8 102.8 103. 103.1 102.7 102! 8 102.4 102.2 102.4 102.6 103.4 Recovery 102.4 102.6 102.4 103. 104.4 105.2 105.4 KiUed 103.6 103.8 102.8 103.2 104. 105.4 • Guinea-pig 959 died two weeks later, showing pneumonia and cheesy abscesses, t Guinea-pigs 965 and 966 were "passage" animals and they were killed on January 2 in order to perpetuate the strain in other guinea-pigs. Immunity in Spotted Fever 361 TABLE 6. The Protective Power of the Bloods of the Vaccinated Guinea-Pigs 789 and 791. Test 2, performed 29 days later than the first test, and 54 days after the second immunizing injec- tion was given. Virus, I c.c. -(-varying amounts of the immune bloods. Date luvuNE Blood from GniNEA-PiG 789 luuxTNE Blood from GuiNEA-PiG 791 Controls = Virus Alone 0.6 c.c. No. 1074 0.4 c.c. No. 1073 0.2 c.c. No. 1072 0.5 c.c. No. 1077 0.3 c.c. No. 1076 0.1 c.c. No. 1075 o.oi c.c. No. 1078 • 0.05 c.c. No. 1079 Jan. 25 26 '°3„ 102.8 103. 102.6 .g 103.4 00 103.4 c 104. « 104.4 « 103.6 ^ 105.2 ° 104.4 104.4 '^ 102.4 ^ 102.2 Q 102.4 Recovery 103.8 103.2 103.2 103.2 ?, 102.4 ^ 103.2 0. 104.4 c 104.2 « 105.3 ^ 104.8 .„ ^°5- ° 104. 103.4 '^ 103.6 5 102.8 Q 102. Recovery i«3-6 102.6 102.6 g, 102. « 105. 0, 104 . 8 ^- 105.6 « 104.4 " 104. i^ 104.4 ° 104.6 § 104.4 "5 103.8 2 102.8 (5 102.8 Recovery 102.8 102.6 102.6 102.8 102.8 102.8 ^ 103.6 ^ 102.7 ^ 102.8 102. "= 103. 102.8 J 102.8 ■= 102.2 -^ 102.3 Recovery 102.6 101.8 102.2 102.8 5. 102.6 .g 103 . 2 ^ 102.8 « 103.8 t 104. .*^ 104.6 ° 104. s 103.4 "5 102.8 2 102.6 Recovery 103.2 102. g. 102. ^ 103.4 ■.» 102.4 g 'O 103.2 . 105. " 104.8 Z 103.8 :n 104. ° 104. § 102.8 'S X02. 2 102. p 102.3 Recovery 102.8 102.4 102.2 102.4 100.8 102. 102.4 103.2 102.4 102.4 103.2 103.4 103.6 102.2 102.4 102.4 Recovery 103.2 102.2 a8 104.6 105.2 Feb. I 105.8 104. 99.8 Death t 6 8 • Guinea-pig 1078 became infected when an immunity test was given on February 28, hence in this experiment the minimum pathogenic dose was greater than o.oi c.c. of virus, but equal to or less than 0.05 c.c. as shown by the fatal course in guinea-pig 1079. t The genitalia were hemorrhagic, and the spleen was greatly enlarged. the passive immunity, which was conferred by the immune blood, to be eliminated, the protective power of its blood will be ascer- tained, and if this proves to be of sufficient strength it will have to be concluded that the animal suffered from a moderately severe attack of spotted fever, in spite of the large dose of protective blood which it received, and in spite of the fact that one-third of this dose protected guinea-pig 962. Having in mind the animals which received 0.7 and 0.3 c.c. of the two immune bloods, the impression is given that the blood of 791 was somewhat more protective than that of 789. The difference in the protective power of the two bloods is manifest in the second test. Thus, 0.5 c.c. of the blood of 791 prevented fever entirely, whereas 0.6 c.c. of 789 did not prevent the development of a distinct and rather high course of fever; 0.3 c.c. of the former permitted moderate fever for five days, while 0.4 c.c. of the latter allowed a marked febrile course of nine days; and the conditions are similar in regard to the third animal \62 Contributions to Medical Science of each group (1075 and 1072). One probably could not come nearer to a comparison than to say that 0.6 c.c. of the blood of guinea-pig 789 was equal to o.i c.c. of that from 791, since the severity and duration of the fever of the corresponding animals were about the same. Comparison of the first test with the second indicates that the protective substances underwent a diminution in quantity in the interval of 29 days. In the first test the blood of guinea-pig 791 in a dose of 0.3 c.c. prevented fever entirely, with the possible exception of two days when the temperature was 103 and 103 . i respectively. In the second test the same quantity of blood permitted a distinct though moderate fever for five days. Simi- larly, 0.3 c.c. of the blood of guinea-pig 789 in the first test afforded greater protection than 0.6 c.c. in the second test. The conclu- sion that such a diminution does occur has been borne out in other experiments. Such comparative experiments, performed at different times, are subject to possible error in that the minimum pathogenic dose of two different lots of virus may not be identical. In our experi- ence, however, this variation has been slight and certainly would not cause greater error than would result from slight diff'erences in the susceptibility of different animals. An experiment similar to the one just described was carried on with guinea-pigs 835, 836, and 837, only the essential points of which will be given. On November 25 all three were injected with i c.c. of virus to which had been added different quantities of immune blood; for 835, 0.3 c.c, 836, o. I c.c, and 837, 0.05 c.c. Guinea-pig 835 had no fever following the injection. Guinea- pig 836 showed the following course of temperature on successive days: 103.4, 103.7, 103, 103, 102.9, 103 -8, 103.4, 103.8, 104.4, 104.5, 103 -3' 102.8, 103.2, 102.6. The fever traceable to the virus was brief and mild in character. Guinea-pig 837 suffered severe infection with the following course of the temperature: 102.9, 102.8, 102.9, 103. I, 102.8, 105, 105.6, 105.8, 104.9, 104.6, 105. I, 104.2, 103.4, 103, 103.3, 102.6, 102.6. Immunity in Spotted Fever 363 They received no further inoculations of virus, hence the experi- ment represents an attempt to immunize by one injection. The first comparative test of the protective powers of the bloods of the three animals was made on January 7, about two months after the injections were made. This was ample time for all the passive immunity to have disappeared, as shown by an experiment quoted earlier in this paper. Suitable quantities of the different bloods were not used to afford intelHgent comparison of their properties. The second test, performed February 5, showed that 2 c.c. of the blood of 835 had no protective effect, whereas 0.8 c.c. from 836, and 0.5 c.c. from 837 showed moderate protection. In the meantime an immunity test was given to 835, which resulted in its infection with spotted fever, hence it may be concluded that the initial injection, the immunizing ( ?) injection, which was followed by no fever, had produced little or no immunity in this animal. A third test of the blood of 836 and 837 showed that 0.9 c.c. from 836 did not afford complete protection, whereas 0.5 c.c. from 837 entirely prevented the development of fever with the dosage of the virus used, namely, i c.c. of third-day virus. This test was made three months after the immunizing injection was given. Both of the experiments described indicate that there is a parallel between the severity of the reaction following the immunizing injection and the subsequent protective power of the blood, and it is reasonable to suppose that the degree of immunity conferred by the vaccination corresponds to a certain extent with the concen- tration of the protective substances which appear in the blood. This does not mean, however, that a severe or even moderate febrile reaction is necessary in order that distinct immunity be conferred. It has happened frequently that a barely perceptible reaction is followed by the development of pronounced resistance. Two instances may be cited in which the immune blood was given a few days in advance of the virus. On December i, 1907, guinea-pigs 844 and 845 received each I c.c. of immune blood subcutaneously. Three or four days later, respectively, each was given i c.c. of third-day virus (Eddy strain) 364 Contributions to Medical Science subcutaneously. The temperature of guinea-pig 844 was as fol- lows: 102, 102.8, 103.4, 103.3, 103-2, 101.8, 103.3, 103-2, 102, 102.7. That of guinea-pig 845 was: 102.9, 103.4, 102.7, 103.3, 101.7, 102.7, 102.9, 102.5, 102.9. Sixty-seven days later each received intraperitoneally i c.c. of third-day virus, one being given the Eddy and the other the Bradley strain. Neither showed the sHghtest febrile disturbance. Controls developed the disease typically. hereditary immunity in vaccinated animals. Inasmuch as strong immunity may be produced in guinea-pigs by the method of "sero-vaccination," in which only a minimum febrile reaction is necessary, it would be expected that the immunity conferred on females in this way would be conveyed to their oJBfspring. This would be anticipated from the fact that such a transfer occurs in the case of females which have suffered from a severe attack of the disease, as described on a previous page. This phenomenon was noted in relation to guinea-pig 791, whose history has been given already, and to guinea-pig 794, which had been immunized by a series of graded injections in which the quantity of immune serum was reduced until pure virus was admin- istered. Two young were born to guinea-pig 791 on December 30, 1907. This was approximately one month after she received the second injection consisting of virus alone and about two months after the mixed injection was given. One of the young (1080) when 25 days old received 0.5 c.c. of virus intraperitoneally. Its tem- perature and appearance were unchanged by the injection, whereas a control of the same age (1087) which received the same dose died of spotted fever 10 days later. The remaining animal (1081), when 52 days old, was given an intraperitoneal injection of i c.c. of virus. No fever resulted and the guinea-pig is still living; the control acquired a severe infection. The same result was obtained with one of the young of guinea-pig 794, 24 days after its birth. The experiments quoted, and others which it seems unnecessary to describe, show that successful sero- vaccination of the guinea- pig is possible. We are studying this problem further to determine Immunity in Spotted Fever 365 the applicability of the method to man, paying particular atten- tion to methods of standardization of the virus and immune serum, the durability of the protective substances under different condi- tions of preservation and the constancy with which a desired reaction can be obtained. It is our intention at the same time to investigate the behavior of the monkey to this method of vac- cination. It does not follow that one can pass directly from the guinea- pig to man or from the monkey to man in sero- vaccination, using corresponding proportions of virus and immune serum. The unknown susceptibility of man in comparison with that of the monkey and guinea-pig is a serious stumbHng-block in this con- nection. A mixture which is neutral for the guinea-pig or which produces only a slight reaction in it may produce a severe reaction in man. The converse may also be true: that a dosage or a proportion of constituents which would excite an immunizing reaction in the guinea-pig or monkey would be without effect in man. There is also a further possibility to face, namely, that the virus as a result of cultivation in the guinea-pig may have under- gone modifications in its virulence whereby it may have become less virulent or more virulent for man. Only one method could possibly be advocated at the outset: namely, to use such proportions of virus and immune serum as would leave no question as to the safety of the procedure, assum- ing for the time that the virus has the greatest possible virulence for man. An index of the effect of such injections could be obtained by studying the properties of the resulting serum. In the event that no antibodies were formed it would then appear justifiable to decrease the proportion of immune serum to that of virus, again studying the effect of the injection on the properties of the serum. This process could be continued until a mixture is obtained which causes the appearance of antibodies without exciting a severe reaction. SUMMARY. An attack of spotted fever in the guinea-pig and monkey pro- duces a strong active immunity of long duration. This immunity is characterized by the presence of protective antibodies in the 366 Contributions to Medical Science serum which may be demonstrated by injecting mixtures of virus and immune serum. The concentration of the antibodies in the blood of the immune animal undergoes a decrease in the course of several weeks. The female that has recovered from spotted fever transmits immunity to her young. The young are immune even when the female acquired her immunity several months before impregnation. The immunity of the young does not depend on the ingestion of milk from the immune mother. The character of the inherited immunity has not yet been determined, although it is presump- tively a passive immunity that differs from the passive immunity conferred by the injection of immune serum by its longer duration. The long duration of the inherited immunity may depend on the longer time required for the elimination of large quantities of protective substances. Passive immunity may be established in the healthy guinea- pig by the injection of blood or serum from the immune guinea-pig. The immune defibrinated blood contains antibodies in such con- centration that O.I c.c. often protects against i c.c. of third-day virus, representing anywhere from 30 to 100 minimum pathogenic doses. In other instances 0.3 or 0.4 c.c. of immune blood are required for this degree of protection. When i c.c. of strong immune blood is injected subcutaneously into healthy guinea-pigs the passive immunity is still present in marked degree after 20 days; after 38 days it is present only in such degree that a mild course of spotted fever results when virus is injected; after 45 days it is no longer manifest. It is possible that passive immunity would not last so long if the immune blood were injected into a foreign species. The guinea-pig may be protected against spotted fever following its inoculation with infected blood, provided the immune blood is administered on the second or third day after inoculation. The curative power of the immune blood or serum is low and in order to produce a distinct effect it is necessary to begin its administration early in the disease and to give relatively large quantities. It exerts a modifying effect on the severity of the infection without bringing about rapid subsidence of the symptoms. Immunity in Spotted Fever 367 By the method of mixed immunization or sero- vaccination, in which virus and immune blood or serum are mixed in suitable proportions, it is possible to immunize the guinea-pig by one or several injections, with the result that he is thereafter immune to infection. The blood of animals immunized in this way con- tains protective antibodies in fairly high concentration and in the case of females the immunity is transferred to the offspring. The quantity of antibodies produced by this method of immunization probably is in proportion to the severity of the febrile reaction which follows the administration of the immunizing dose. How- ever, strong immunity has resulted in some instances in which the immunizing injection caused only a barely perceptible febrile reaction. It may be possible to use the immune serum from the horse, which we have shown to be susceptible to inoculation, for the prevention of the disease in man. For this purpose the serum should be injected in sufficient quantity within two or three days following the bite of the tick. Such an injection should not be considered protective for a longer period than three weeks. The method of sero-vaccination is not yet sufficiently perfected to warrant its application to man, but the subject is being studied further in order to determine its safety and efficiency. There is no hope of obtaining a stronger serum for curative purposes than that yielded by an animal which has recently recov- ered from the disease, until the micro-organism can be cultivated artificially, thus making available a desirable quantity of antigens for immunization. Even with this condition reaUzed the results of further immunization cannot be anticipated but must await experimental determination. A MICRO-ORGANISM WHICH APPARENTLY HAS A SPECIFIC RELATIONSHIP TO ROCKY MOUNTAIN SPOTTED FEVER. A PRELIMINARY REPORT.' H. T. RiCKETTS. (From the Pathological Laboratory of the University of Chicago.) Since the spring and summer of 1906, bodies which I have referred to in my notes as " diplococcoid bodies," and sometimes short bacillary forms, have been found with considerable constancy in the blood of guinea-pigs and monkeys which were infected with Rocky Mountain spotted fever. They have also been seen in the blood of man but not so frequently. Much more time has been spent on the blood of the experimental animals than on that of man in view of the fact that it could always be obtained in fresh con- dition. The form most commonly found is that of two somewhat lanceolate chromatin-staining bodies, separated by a slight amount of eosin-staining substance. The preparation of Giemsa, as fur- nished by Griibler, has been used almost exclusively, and with variations in the technic the intermediate substance may stain faintly blue. In spite of the constancy with which these bodies were found, it did not seem justifiable to claim that they represent the micro- parasite of the disease, for two reasons: (i) the very complex morphology of the blood, especially in febrile states, when various cells and probably their nuclei are subject to unusual disintegra- tion; (2) because of my inability to cultivate a micro-organism of this character from infected blood by the use of ordinary and some unusual culture media, under various conditions of cultivation, or by other means to obtain it in satisfactory concentration.* ' From Jour. Am. Med. Assn., 1909, 52, p. 379. See also " DemoQstration of a Micro-organism Which Apparently Has a Specific Relationship to Rocky Mountain Spotted Fever," Trans. Chic. Path. Soc, 1907-9, 7, p. 2S4- ' Mr. P. G. Heinemann assisted in an extensive series of culture experiments in the spring of 1907. 368 A Micro-organism in Spotted Fever 369 THE BACILLUS IN THE TICK. Although infected ticks had been examined previously in a more or less cursory manner, their systematic study was not undertaken until recently. In the pursuit of this work, advantage was taken of the fact that the disease is transmitted by the infected female to her young through the eggs, as described in a previous report. A repetition of these experiments in the winter of 1907-8, with the help of Dr. Maria B. Maver, resulted in such transmission in 50 per cent of the ticks used, the fact being determined by allowing the larvae to feed on normal guinea-pigs. This second series has not been published heretofore. Female tick 40, a dermacentor, from Montana, had produced fatal infections of spotted fever in guinea-pigs 1740 and 1764. A number of eggs from the first day's laying were crushed in- dividually on cover glasses, fixed in absolute alcohol, and stained with Giemsa's stain. Each egg was found to be laden with aston- ishing numbers of an organism which appears typically as a bipolar staining bacillus of minute size, approximating that of the influenza bacillus, although definite measurements have not yet been made. Various forms are seen depending on the stage of development and the arrangement in which two or more may be found. It is very common to find two organisms end to end, with their poles stained deeply and the intermediate substance a faint blue, resembling a chain of four cocci. When the chromatin is not yet sharply limited to the poles, the somewhat lanceolate forms so often recognized in the blood are seen. Not infrequently delicate bacilli with a uniform distribution of the chromatin are found. These are all interpreted as stages in the evolution of a bipolar organism. They are present in varying numbers in different eggs, but as a rule they are surprisingly numerous, and in some instances they would certainly count into the thousands. Many faintly staining, apparently degenerate, forms are encountered. Examination of the eggs of three dermacentors from Idaho (different specifically from the Montana dermacentor), which were infected from the guinea-pig, showed the presence of the same forms (ticks 5, 7, and 9). 370 Contributions to Medical Science Conference with zoologic scientists who have made a particular study of the structure of eggs brought out the fact that such bodies are not known as a constituent of the egg of any species of animal. Although it has not yet been possible to examine the eggs of ticks which are known to be free from spotted fever, the equivalent control has been made through a comparison of the visceral organs of infected and uninfected ticks. The salivary glands, alimentary sac, and ovaries of infected females are Hterally swarming with exactly similar micro-organisms. On the contrary, they appear to be entirely absent from the viscera of the uninfected tick, both male and female. agglutination reactions. The most striking evidence of the probable etiologic relationship of this organism to spotted fever is found in the positive outcome of agglutination tests. Fortunately the organism is so numerous in the eggs that a bacterial emulsion of reasonable concentration for agglutination tests can be made by crushing forty to fifty eggs in about 0.05 c.c. of salt solution. The material is so scant that only the microscopic method could be used. The preparations were made as hanging drops, incubated for two hours, dried, fixed with absolute alcohol, and stained. The serum of the normal guinea-pig either causes no agglutina- tion at all, or at the most produces only sHght agglutination in proportions of i to i and i to 20. Dilutions higher than this cause no agglutination. In testing the agglutinating powers of immune serums, three animals which had been infected from different sources and had recovered were used. One (1751) had been infected with a dermacentor from Idaho; another (1692), with a strain handed down direct from guinea-pig to guinea-pig for nearly three years without the intervention of ticks, the original infection having been obtained from the blood of man; the third (1757), with a strain kept in the same way since last spring. Graded dilutions, beginning with i to i and going as high as i to 400, were used in the different series, with the striking result that a complete agglutinating power was present in the three immune serums in dilutions up to i to 320. It was somewhat less in a dilution of i A Micro-organism in Spotted Fever 371 to 400. The highest dilution which will cause clumping has not been ascertained. No fresh immune serum from man is at hand, but tests were made with three specimens which are about five, seven, and nine months old, respectively. They have been preserved in the ice- chest with the addition of 0.3 per cent of chloroform. The pecul- iar phenomenon of failure to agglutinate in concentrated solution was noted with all three. With the oldest serum no agglutination occurred until the dilutions of i to 320 and i to 400 were reached, when incomplete clumping was produced. With the second there was no agglutination in the dilution of i to i, distinct clumping in I to 20, 1 to 40, and i to 80, with httle or none in higher dilutions. In the serum of five months standing the reaction was absent in the dilutions of i to i, i to 20, and i to 40, positive but not com- plete in I to 80, I to 160, and i to 240, with very little clumping in I to 320 and i to 400. Normal human serum caused clumping in a dilution of i to i, a very slight amount in i to 20, and none at all in the higher dilutions.^ The failure of the immune serums from man to agglutinate in concentrated solutions, whereas they did so in higher dilutions, is taken as an example of the action of the so-called proagglutinoids. As explained by Ehrlich's theory, this consists in the occupation of the bacterial receptors by inactive agglutinin which exceeds the active agglutinin in its affinity for the bacteria. With higher dilutions of the serum the agglutinoids are so diluted that they do not completely occupy the bacterial receptors, thus affording a point of attack for the active agglutinin, unless the latter has been eliminated by extreme dilution. This is well known as a property of old agglutinating serums. bacilli in infected serum. As a means of concentrating the organisms in the serum of the infected guinea-pig the following experiment was performed: Three cubic centimeters of fresh infected serum were diluted with an equal amount of salt solution, and to this was added 0.3 c.c. of an agglutinating serum from the guinea-pig. The mixture was ' Some of the dilutions were wrongly stated in The Journal. They have been corrected in this reprint. 372 Contributions to Medical Science placed in the incubator for two hours and then centrifugated for about ten hours at a speed of i,8oo revolutions. All but the merest drop was then pipetted off and stained preparations were made of the sediment. Examination showed the presence of a moderate number of forms which are identical in appearance and size with those often recognized in ordinary smears of infected blood, and also with the " diplococcoid " forms seen in the egg of the infected tick. No such bodies were found in a control tube of normal serum. The evidence pointing to this organism as the causative agent in spotted fever, though not complete, is of a striking character. In so far as I know it would be an unheard-of circumstance to obtain such strong agglutination with an immune serum, in the presence of negative controls, unless there were a specific relationship between the organism and the disease. In favor of the specific relationship in this case are also the presence of the organism in large numbers in infected ticks and in their eggs, its absence from uninfected ticks, and the presence of similar forms in the blood and serum of the infected guinea-pig. Morphologically the organism is a bacillus and somewhat pleomorphic as described. Its resemblance to the bacilli of the hemorrhagic septicemias is striking, and in this connection it is important to note that spotted fever is a hemorrhagic septicemia. It has not been cultivated, although work with this end in view is in progress. I have devised no formal name for the organism discussed, but it may be referred to tentatively as the bacillus of Rocky Mountain spotted fever. A further study of its characteristics may suggest a suitable name. That a bacillus may be the causative agent of a disease in which an insect carrier plays an obhgate role under natural conditions may be looked at with suspicion in some quarters. Yet, even with- out the evidence in this case, it would seem unscientific to be tied to the more or less prevailing belief that all such diseases must, on the basis of several analogies, be caused by parasites which are protozoan in character. Further study of the relationship of the bacillus to the disease is being carried on and will be reported at a future date, together with illustrations and a more detailed account of its characteristics. SOME ASPECTS OF ROCKY MOUNTAIN SPOTTED FEVER AS SHOWN BY RECENT INVESTIGATIONS. THE WESLEY M. CARPENTER LECTURE OF THE NEW YORK ACADEMY OF MEDICINE, 1909.* H. T. RiCKETTS. As your lecturer on this occasion it will be my privilege, a most honorable one, to present a more or less succinct account of an investigation of Rocky Mountain spotted (or tick) fever, which has been under way for more than three years, and which, it is believed, has contributed results of value for a better understanding of the disease in certain important aspects. Many of the essentials have been described in scattered articles in two different journals, and in two reports which have been rendered to the State Board of Health of Montana. However, such a presentation as I wish to give, involving some new data, a consideration of the natural his- tory of the disease, and certain observations regarding its microbic etiology, may perhaps free me from a possible charge of too much repetition. History and occurrence. — It would be interesting if one were able to give a satisfactory history of the disease, bearing on its first appearance among the native Indians and among the white men of the different "infected districts," its early geographic distri- bution, and the possibility of its having extended into new terri- tories. Knowledge on these points is very incomplete, however, and regarding its antiquity it can only be stated that it was a well- known disease when certain of the older physicians and residents first came into the valley of the Snake River, in southern Idaho, thirty or forty years ago. In the Bitter Root Valley in western Montana it is difficult or impossible to learn definitely of cases which appeared prior to the early '8o's. Hence it is possible, though by no means to be taken for granted, that spotted fever appeared more recently in western Montana than in Idaho. The Bitter Root Valley is separated from Idaho only by the Bitter Root range of mountains, * From Med. Record, 1909, 76, p. 842. 373 374 Contributions to Medical Science and since spotted fever occurs only on the west slope of the Bitter Root Valley, it is conceivable that the disease extended into the latter region from Idaho through the various mountain passes and that it has as yet failed to reach the eastern slope of the valley. It is known as particularly hazardous to visit certain canyons of the Bitter Root Mountains during the months of spring. Its distribution in any one state appears to be very irregular, although it is found here and there over a wide range of territory; in southern Idaho along the course of the Snake River, and in west- ern Montana as stated, at another point in eastern Montana, near Bridger, in the northern and northwestern parts of Wyoming south of the Bridger district, in northern Utah, eastern Oregon, and in ill-defined districts of Colorado and Nevada. In none of these states, however, except Idaho and Montana, has any system- atic attempt been made to learn the exact occurrence of the disease. The irregular geographic distribution is greatly accentuated in certain instances. Thus, in Idaho, for many years the disease was known to occur only on the north slope of the Snake River Valley, and if my informants are correct, it has existed on the south side of the valley, in the eastern part of the state, only during the past twelve or fifteen years. Its limitation to the west slope of the Bitter Root Valley was mentioned. The various possibihties which may be adduced in the explanation of this peculiarity and the factors which operate for extension, on the one hand, and for its limitation on the other, cannot well be discussed within the limits of this lecture. No satisfactory estimate can be given of the number of cases which appear annually over this whole territory. In the two infected districts of Montana the number probably varies from fifteen to thirty. Twelve occurred in the Bitter Root Valley in 1908 and twenty-six or twenty-eight in 1909. The increased num- ber during the past spring was due to railroad construction, which resulted in the exposure of an unusual number of persons. In 1908, Dr. Edward E. Maxey, of Boise, Idaho, did an admirable piece of work, in which he obtained all possible data regarding spotted fever from the physicians of the state, locating the disease Some Aspects of Rocky Mountain Spotted Fever 375 exclusively in the basin of the Snake River, chiefly on the north slope of the valley; 380 cases occurred in Idaho during that year.' The work of Dr. Maxey urgently demands repetition in other states. There is no more striking feature of the disease than its sharp limitation to the months of the spring. In Montana it makes its first appearance in the latter part of March or April, and the last cases are seen about the middle of June. The spring is earlier in southern Idaho and, corresponding with this, the "spotted fever season" begins and closes from two to four weeks earlier than in Montana. The relation of the tick to this peculiar seasonal occurrence will be referred to again. General features. — Spotted fever has a history in literature since 1899, when Maxey defined it as follows: " an acute endemic, non-contagious, but probably infectious, febrile disease, character- ized clinically by a continuous moderately high fever, severe ar- thritic and muscular pains, and a profuse petechial or purpural eruption in the skin, appearing first on the ankles, wrists, and fore- head, but rapidly spreading to all parts of the body."^ I shall not at this time discuss in detail the various clinical aspects of the disease, but rather refer you to the reports of Maxey ,^ Wilson and Chowning,-5 Anderson,-* and Stiles^ for information in this regard. I may only state that, in my experience, variations from the typical type of eruption are encountered rather frequently.^ The anatomical changes are not highly characteristic with the exception of the cutaneous phenomena which have been mentioned. Gangrene of the scrotum, prepuce, faucial pillars, and soft palate sometimes occur. Strangely this is met with more frequently in Idaho than in Montana, and the explanation may lie in the differ- ence in the duration of the infection in the two localities. Cus- tomarily patients in Montana die before the period when gangrene is likely to occur, whereas in Idaho the large percentage of recov- eries gives sufficient opportunity for the development of this phenomenon. Gangrene in both man and animals is a compara- tively late symptom. The most striking and constant visceral change is found in the spleen, which is always enlarged and may. weigh two or three times as much as the normal organ. Its con- sistence is rather firm and gives the appearance of a very cellular 376 Contributions to Medical Science organ, such as might be caused by a massive proliferation of pulp and endothelial cells. It has none of the soft semi-gelatinous appearance of the typhoid spleen. The lymphatic glands are uniformly enlarged and moderately congested, but in man I have not found them hemorrhagic. The kidneys are swollen and degenerated and the liver is moderately enlarged, congested, and may be fatty. No striking or constant changes are found in the suprarenals, heart, and lungs. The central nervous system shows no alterations except slight meningeal congestion and edema, but these are of a very minor order. In one case I found a lobular pneumonia and the appearance of multiple infarctions of the spleen. The conditions were such that the bacteriology of the case could not be worked out. The microscopic anatomy is in the hands of Professor E. R. Le Coimt. Severe and mild types. — ^As regards severity, two different types are to be recognized, the mild and the severe, and in this point lies another of the pecuHarities of the disease. In western Mon- tana it is almost uniformly of a very severe character, the mor- tality ranging from about 65 per cent to about 90 per cent. Children recover with some frequency, adults very rarely. In Idaho, on the other hand, the death rate rarely rises above 5 per cent, and in the 380 cases which Maxey collected in 1908 it was 4.86 per cent and 3 . 8 per cent for the preceding year. This difference would not be so remarkable were it not constant; the spotted fever of Montana has always been severe, while that of Idaho has always been mild. Occasional cases in the latter state go through the entire course of fever without being confined steadily in bed. This feature suggests that the disease in the two localities may not be absolutely identical, but may rather represent varieties, just as we accept paratyphoid fever as a variety of typhoid. The difference in virulence is also borne out by animal inocula- tions. Guinea-pigs may be infected with the Idaho disease by the injection of diseased human blood, just as in the case of the Montana disease, and the cHnical and anatomical phenomena are the same in both cases. The Montana disease may be kept going indefi- nitely in the guinea-pig by consecutive passage with a high rate Some Aspects of Rocky Mountain Spotted Fever 377 of mortality (about 90 per cent). On three occasions, however, and with material taken from different patients, I have been unable to maintain the Idaho disease in the same way. In one instance it was carried into the third passage ; in another, into the fourth ; and in the third, into the tenth. In spite of this constant difference, their practical identity is shown not only by the identity of the clinical manifestations, but also by specific immunity tests. Guinea-pigs which have recovered from the Idaho disease are immune to that of Montana, and the converse is also true. It may be that quantitative determinations of the protective and agglutinating power of the two immune serums will show such differences that the two diseases may be considered to stand in a sort of "group relationship" to each other. Although a satisfactory explanation of this difference may not be possible, it is of interest to note that the spotted fever tick of Idaho differs specifically from that of Montana. Banks has identified the latter as Dermacentor venustus. The Idaho tick is also a dermacentor, but its markings are much less distinct than those of D. venustus, and it is constantly smaller in size in all its stages, as shown by numerous comparative measurements of the eggs, larvae, nymphs, and adults. From specimens sent to Banks, he concluded that the species had not been described previously, and he gave it the name of Dermacentor modestus. Our experi- ments have shown that both venustus and modestus will transmit the Montana disease to the guinea-pig, and venustus will also transmit the Idaho disease. So far, we have been unable to obtain transmission of the Idaho disease to the guinea-pig by the use of the Idaho tick, i.e., D. modestus, except in one instance in which a large number of ticks was involved. It will be for future experi- ments to determine whether this is due to the comparatively low virulence of the Idaho virus for the guinea-pig, as it exists in D. modestus; in other words, whether a larger amount of this strain is required for the infection of the guinea-pig than a single tick of this species is able to inject during its feeding. It is, of course, possible that the tick plays only a partial role in the maintenance of virulence. As indicated later, the disease appears to be maintained in nature by alternating between the 378 Contributions to Medical Science tick and some one or more of the native small animals. It is possible that the animals which play this part in Idaho are not identical with those enacting a similar role in Montana, and if this proves to be the case, the difference in virulence may depend in whole or in part on the difference in the animals which serve as hosts in the two localities. This may or may not be open to experimental determination. Theories regarding the source of injection. — In the past various local theories have prevailed, and do still prevail to a certain extent, regarding the source of infection for man. The sharp limitation of spotted fever to the months of spring, suggested to the residents and to many of the physicians that the virus may be obtained from the water which flows down from the mountains dur- ing the melting of the snow. It was assumed that this resulted in the flushing out of certain low places which contained decaying vegetation, in which the virus or poison might breed or have its source. In the Bitter Root Valley in Montana so many of the patients were employees of sawmills, or lived in the vicinity of such mills, that the disease also became associated with this industry. Old and decaying accumulations of sawdust were supposed to con- taminate the water, and many patients give a history of having drunk water from a stream on which a sawmill previously had its site. It is pointed out that there was virtually no spotted fever until the lumbering industry entered the valley. In spite of a fair degree of plausibility which attached to the water theory, careful examination of the distribution of cases during any one season shows that this does not correspond to any type of water supply. Indeed, one of the characteristic features of the disease, though it is an accidental one, is the rare occurrence of more than one case in a given family during a season, and the number of cases which occur on any one stream during a season is too small to admit of the water theory. Also the rough coin- cidence of the spotted fever season with that of the melting snow and high water is a misleading one. A few consecutive warm days are commonly followed by a crop of cases of spotted fever in the spring. This, of course, does melt more or less snow, and does cause a rise in the streams, but it is equally efficient in rousing Some Aspects of Rocky Mountain Spotted Fever 379 from lethargy millions of ticks which in cooler weather are too somnolent to respond to the presence of a host. The lumbering industry has indeed been a factor, but only in this, that it took its employees into the timbered country which harbors infected ticks, and caused a general increase in the popu- lation of the valley, with the consequent exposure of a larger num- ber of persons. The industry of railroad construction did the same thing in 1909. Furthermore, there is no lumbering in southern Idaho, where spotted fever is much more common than in Montana. The theory that the tick may have a relationship to spotted fever had a fluctuating existence in both Montana and Idaho for some time, but it was not generally credited, and really had its birth with the publication of Wilson and Chowning in 1902. Previous investigations. — A laboratory study of the disease was first undertaken in 1902-3 by Dr. Louis B. Wilson and Dr. William M. Chowning, then of the University of Minnesota.'' In addition to giving an account of the clinical and anatomical phases of the malady, they advanced the hypothesis that the tick is the means of infecting man; described as the cause a protozoan parasite, to which they gave the name of Piroplasma hominis, and which they found to have a specific location within the erythrocytes; and suggested that the ground squirrel may be the source of infection for the ticks, basing this hypothesis on the supposed presence of the same parasite in the erythrocytes. Their work was followed by that of several members of the Public Health and Marine Hospital Service during succeeding years: Anderson, Stiles, and King. Anderson agreed with the essential claim of Wilson and Chowning, namely, the existence of a piroplasma in the erythrocytes of the patients, and also took a stand favoring the hypothesis of transmission by the tick. Stiles, on the other hand, was unable to find the piroplasma, and expressed his disbelief in the tick theory. Wilson and Chowning had no experimental proof of the correctness of this theory, just as Stiles had none to show its incorrectness. Personal studies. — The investigations of which it is the purpose to speak were inaugurated in the spring of 1906 with the discovery that the guinea-pig and monkey may be infected with spotted fever 380 Contributions to Medical Science through the subcutaneous or intraperitoneal injection of the blood of human patients.* In view of the fact that the disease has been without a known bacteriology during the course of the work, it has been necessary to rely on the cHnical course and anatomical changes in the guinea-pig, together with negative bacteriological findings, as criteria for the recognition of the disease in this animal. Since the reliability of the different phases of the work really depends on this point, it is desirable to emphasize that spotted fever in the guinea-pig is a characteristic condition, and that con- fusion can occur but rarely. Spotted fever in the guinea-pig. — When the inoculation is accom- plished by the injection of virus (diseased blood, and the organs or eggs of infected ticks), an incubation period of from two to five days is followed by a sudden rise in temperature, which may reach 105° or 106° F. (in rare instances 107°) within two or three days, and which persists without remission for a period of from six to twelve days, or until the death of the animal, four to ten or twelve days after onset. A swelling of the scrotum, which begins in the males on the third or fourth day of fever, progresses rapidly, and the overlying skin is soon infiltrated with blood. In white-skinned animals, which are shaved at this time or later, a generalized roseo- lar eruption, most marked on the back, extremities, and face, can be made out readily. In females the vulva becomes swollen, but exhibits hemorrhages only in occasional cases. Swelling, conges- tion, and hemorrhage of the lymphatic glands, a more or less massive enlargement of the spleen, which is cyanotic in color, and, in males, marked congestion of the tunica vaginalis and often of the testicles are the essential anatomical changes, which are encountered with little variation. Commonly the kidneys are swollen and degen- erated, the Hver swollen, and the suprarenals may show cortical hemorrhages. The serous surfaces, lungs, heart, and central nervous system are never involved, in the absence of mixed infec- tions. As indicated by ordinary culture media, the blood and vis- cera are sterile. * The monkey (Macacus rhesus), rabbit, horse, and at least five species of small wild animals have a greater or less degree of susceptibility. In aU of these except the monkey confirmatory transfers must be made into the guinea-pig in order to prove the character of the infection; hence, for this reason among others, they are of no service for routine experimentation. Some Aspects of Rocky Mountain Spotted Fever 381 The "immunity test" is an important and sometimes an essen- tial aid to diagnosis in animals which recover. In the event of mild infection, which has occurred accidentally, or which has been produced deliberately, by modified virus, the fever may be short in duration and moderate in height, and the external hemorrhages may not appear.* In such cases the immunity test, which consists of the injection of from 500 to 1,000 pathogenic doses of infected blood, determines whether or not spotted fever was present in the first instance. This active immunity is strong and of long dura- tion, and the offspring of immune females possess a passive im- munity which persists for a period of from three to four months. These data have proved sufficient for the positive recognition of spotted fever in the guinea-pig, and on three occasions they have enabled us to recognize mixed infections which crept into experi- ment animals in some unknown way. The transfer of spotted fever to the guinea-pig from human patients in Montana and Idaho has now been repeated not less than fifteen times from as many different patients, and with practically unvarying results. The oldest strain now on hand, representing the Montana disease, has been kept alive by passage in the guinea-pig for three and one-half years, with symptoms and anatomical changes which do not vary from those produced in the first animals injected, and with no decrease in the virulence of the diseased blood. Two other " human " strains have been maintained in the same way, each for approximately a year, when they were dropped. Also a strain from naturally infected ticks (D. venustus) has now been maintained for six months. Inasmuch as the work has been done with these various strains more or less indiscrimi- nately, it is important to be certain of their unity, and this is shown not only by the identity of the cHnical and anatomical phenomena, but also by the fact that each establishes immunity against the others. The integrity of the infection in the guinea-pig was also confirmed by attempts at reinoculation with fresh human virus; animals which had recovered from inoculations with a strain grown in the guinea-pig for a year proved immune to such inoculations, • In some successful vaccination experiments a moderate rise in temperature for only one day was all that marked the course of infection. Such animals were thereafter immune. 382 Contributions to Medical Science and from the further correspondence of the clinical and anatomical manifestations in man and the guinea-pig, we may be certain that the infection which has been used in the experimentation is identi- cal with human spotted fever. That the disease is an infection rather than an intoxication by non-living material is abundantly demonstrated by the statements just made; the causative agent must be one which is capable of proliferation, in order to survive two hundred consecutive passages in the guinea-pig with undi- minished virulence. Experiments with the tick. — With the demonstration of the exquisite susceptibility of the guinea-pig it was manifestly possible to proceed to an investigation of the hypothesis of transmission by the tick.* The technic of this work was very crude in the beginning, but satisfactory methods were soon evolved, as described in a prior communication.^ The first problem was one of simple transmis- sion, namely, is the tick (D. andersoni of Stiles, D. venustus of Banks) able to transmit spotted fever from the sick to the healthy animals by means of its bite ? The initial successful results, obtained with a single female and later with a single male (1906), have been repeated, I should estimate, not less than two hundred times in the course of various experiments conducted in the field and home laboratories. There is, indeed, no difficulty in obtain- ing it. It is only necessary to permit a tick to feed on a diseased guinea-pig for several hours to a day, and then to transfer the arachnid to a healthy animal as soon as it can again be induced to bite. The exact duration of feeding required for infection of the tick has not been determined. This must vary a good deal. The tick * The infective dose of diseased blood drawn on the third day of fever in the guinea-pig is frequently as small as o.ooi c.c, when injected intraperitoneally into a fresh animal. It is interesting to note that the first experiments which were performed with the intention of deter- mining a possible relationship of the tick to spotted fever were done on the human subject, and have not been formally reported by the authors. I have referred to this in a former article, and here may only repeat that Drs. McCalla and Brereton, of Boise, Idaho, infected two individuals successively by the bite of a tick which they had removed from one of their patients. Naturally, the consent of the subjects was obtained, which in this case was not difficult on account of the mild nature of the disease in Idaho. The exf)eriments of McCalla and Brereton followed the publication of the tick hypothesis by Wilson and Chowning, and they were unknown to me until after the larger part of my work with the tick had been done. The first transmissions of spotted fever in the guinea-pig by means of the tick were undertaken and reported almost simultaneously by Dr. W. W. King, of the Public Health and Marine Hospital Service, and by the writer. Some Aspects of Rocky Mountain Spotted Fever ^8 O'-'O is a slow feeder. From fifteen to twenty minutes are required for the hungriest and most vigorous to attach itself firmly, and without firm attachment by means of the palpi, feeding is impossible. In other instances the process is much slower, and after two or three days of attachment there is little evidence of the ingestion of blood. In so far as we have determined, the incubation period in the tick is brief, and, perhaps, is represented only by the time required for the distribution of the virulent organisms throughout its body and eventually into its salivary glands. In a number of instances the tick proved virulent immediately following its removal from the infected animal. In one exceptional experiment, transmission could not be obtained until nearly two weeks following the infected feed. Likewise the minimum duration of the bite required for inocu- lation probably varies a good deal, depending on the vigor of feed- ing. Dealing with ticks whose infection had already been proved by longer feedings, transmission resulted several times from ten- hour exposures, and recently from a feed of less than two hours.* The mechanism of transfer would seem to consist of the injec- tion of the saHvary secretion, laden with virus, into the cutis of the host. By injecting into the guinea-pig separate triturations of the salivary glands and aUmentary sac of the infected tick, the presence of the virus in these tissues was shown in 1907. It seems that the tick must undergo a generalized invasion by the virulent organisms, from the intestine as a point of departure, with their eventual localization in the salivary glands. The hereditary transmission of the disease, involving the presence of the virus in the oviduct and germ cells, is a further indication of this generalized infection. That a toxic secretion is injected at the time of feeding is indicated by the intense local reaction which follows the bite of even the normal tick. There is, indeed, no difference between the bite of the normal and the infected tick. The spotted fever virus does not produce a local cutaneous or subcutaneous reaction. Inheritance in the tick.— In the summer of 1907 three additional points were determined regarding the relationship of the virus to the tick, namely, the occurrence of hereditary transmission, the • The problem of time relationship is now the subject of further study by Mr. J. J. Moore. 384 Contributions to Medical Science possibility of "stage-to-stage infection," and the existence of natu- rally infected ticks. The first point was demonstrated by permit- ting the offspring of infected females to feed on normal guinea-pigs, and the development of spotted fever in the latter signified the transmission of the disease from the female to the larvae through the medium of the germ cell. The frequency with which hereditary transmission occurs has an important bearing on the natural history of the disease, particularly on the means by which it is maintained from year to year, and for more accurate information in this regard, the subject was again taken up in the winter of 1907-8 with the assistance of Dr. Maria B. Maver. In the neighborhood of thirty females were infected and the virulence of each was proved by means of a feed on a normal guinea-pig, after which they were put aside for the laying of eggs and the hatching of the latter. Broods were obtained from only eighteen of this number, and of the eighteen only nine proved capable of infecting the normal guinea-pig, giving us 50 per cent of hereditary transmission. This percentage is greater than that obtained in either earlier or later experiments, and it seems to be a safe conclusion that under natural conditions hereditary transmission does not take place from more than 50 per cent of the infected females. The practical significance of this result is seen when one attempts to answer the query: How is spotted fever maintained in nature? The occurrence of hereditary transmission explains perfectly how the disease is kept alive from one spring to the next, but does not explain its permanent maintenance, since not more than 50 per cent of the infected females transmit the disease to their young. Assuming that the total number of ticks in an infected region remains the same over a period of years, and this condition appears to be approxi- mated in the Bitter Root Valley, a 50 per cent reduction in the number of hereditary transmissions for each succeeding year would, in the course of time, spell the extermination of the disease among the ticks. There is no evidence, however, to show that spotted fever is decreasing. It would seem that its very existence at this time signifies a prolonged previous history, and that there exists in nature a mechanism for the annual replenishment of the virus among the ticks, to offset the reduction (50 per cent or greater) Some Aspects of Rocky Mountain Spotted Fever 385 which takes place every year. I shall have occasion to speak of this further when considering the relationship of native animals to spotted fever and the source of infection for the ticks. ^' Stage-to-stage infection.^' — Still another phase of the relation- ship of the virus of spotted fever to the tick would seem to be of importance for the maintenance of the disease; this relates to the occurrence of "stage-to-stage infection." In the summer of 1907 larvae from uninfected females were permitted to feed on guinea- pigs which had been inoculated with spotted fever. After they had moulted into the next active stage (nymphal) they were placed with normal guinea-pigs ; the development of spotted fever in the latter, after being bitten by the nymphs, proved that the organisms, when taken in by the larvae, retain their life and virulence until the succeeding active stage is reached. In a similar way it was demon- strated that the nymph which feeds on diseased blood is still able to transmit spotted fever when it reaches the adult stage, some weeks later. "Stage-to-stage infection" would be of little significance for the maintenance and spread of spotted fever were it not for the habit which Dermacentor venustus and D. modestus possess, of dropping from the host for the process of moulting.* To assume a case, larvae which have inherited their spotted fever may infect a ground squirrel; dropping from this animal in order to moult, certain of them, after reaching the nymphal stage, may reach a hitherto uninfected ground squirrel, or a rock squirrel, and convey the disease to the latter; and this event may again be repeated after the adult stage is reached. Thus it is possible for a single tick to infect two or three different animals in the different stages of its development. The habit of dropping also affords opportunity for extension of the disease among the ticks. Non- virulent larvae and nymphs, feeding on a susceptible animal in company with infected ticks, may well derive the disease from the animal after its infection is established, and they in turn after moulting and seeking further hosts may cause additional extension in the same way. • This is the custom of most species, but some, as Margaropus annulatus, which carries Texas fever, remain on the host for both moults. The latter, after reaching a host as larvae, remain with it until the adult stage is reached, and, indeed, until the female is ready to drop and lay her eggs. Hence, even if stage-to- stage infection occurs in this instance, it can have no great significance for the extension of Texas fever. 386 Contributions to Medical Science Hence, it is manifest that the faciHties for extension are much greater, in view of the habit of "dropping," than they would be if the tick remained on its original host during its entire develop- ment. From what has been said it may be inferred that the ticks in question have a multiplicity of hosts. They are known to occur naturally on at least seventeen of the domestic and wild animals of the Rocky Mountains (horse, cow, dog, swine, sheep, elk, deer, mountain goat and sheep, bear, ground squirrel, pine squirrel, rock squirrel, ground hog, chipmunk, mountain rat, and the so- called rock rabbit), and experimentally they feed readily on the guinea-pig, rabbit, and monkey. Naturally infected ticks. — As proof of the etiological relationship of the tick to spotted fever, I have cited the experiments of McCalla and Brereton, in transmitting the disease from man to man, and m}r own work demonstrating the carrying power of the tick, and the intimate relation of the virus to the tick, as illustrated by heredi- tary transmission and stage-to-stage infection. To these may be added the almost unfailing history of recent tick bites in the cases of human infection.* With all these data, however, it may still be argued, with some show of reason, that these conditions are accidental, that the mere fact that the tick has the carrying power does not constitute proof that man acquires his infection from the tick. Furthermore, to remove an infected tick from a human patient and prove its virulence does not improve the character of the proof, since the critic may say that the tick derived its infec- tion from the patient, rather than the converse. Appreciating both the pertinence and the improbability of this conception, it seemed that the situation demanded the discovery of naturally infected ticks. This was undertaken successfully in 1907. Several hundred ticks from an "infected district" were allowed to feed on guinea-pigs in groups of from thirty to fifty. One of these groups produced spotted fever in its guinea-pig, and the * As stated in reports to the state of Montana, the failure to find a tick on the patient, or evidence of a recent tick wound, provides no evidence of the absence of a tick bite. A few hours' feed only is required for infection, and a tick may feed and escape or be brushed off accidentally and the individual himself have no knowledge of the bite. Tick wounds situated on the scalp or perineum may well escape very care- ful search. Some Aspects of Rocky Mountain Spotted Fever 387 diagnosis was verified by repeated passage and immunity tests. During the past spring it seemed desirable to repeat this observa- tion. Ticks were collected from Lo Lo Valley, which is tributary to the Bitter Root Valley, and sent to Chicago, where Dr. Maver conducted the tests, placing the ticks with guinea-pigs in groups as in the previous experiments. In this instance infected ticks were found in two of the groups, and a strain of spotted fever derived in this way is still being carried on by the method of passage in the guinea-pig. With the discovery of naturally infected ticks, it would seem that the requirements for proof of the relationship of the tick to the infection of man have been met as fully as it is possible to meet them. Seasonal relations. — As I have stated elsewhere, the peculiar limitation of spotted fever to the months of spring depends on two facts in relation to the development and habits of the ticks involved: First, that the cycle of development is annual, and second, that only the adult ticks are found as parasites of man* and adult ticks are to be found only in the spring, or roughly so. One may, indeed, find occasional adults in February or the latter part of January, or, on the other hand, as late as August or September. High in the mountains they are found later in the summer than in the lower val- leys. Presumably, this is on account of the lower temperature in the former localities, which favors both a late development, and a prolon- gation of life after the adult stage is once reached. Such observations are exceptional, however, and it remains true that the appear- ance and activity of the adult begins some time during the latter part of March or the first part of April, the exact time depending on the occurrence of warm days, and that they are not commonly seen after the middle of June. We may be quite sure that the temperature of the region is the most potent factor in determining this rather sharp annual cycle. From experience gained in labora- tory cultivation, and by observations on the stock and wild animals, we may fix the egg-laying period in nature in the latter part of the spring and the early part of the summer. During the summer * Only in two instances have I learned of larvae and nymphs feeding on man; children were concerned in both instances. 388 Contributions to Medical Science larvae and nymphs are found on the small wild animals in large numbers, and it is probable that many of the nymphs have had their feed and dropped from their hosts before the onset of cold weather. In several instances, however, large numbers of feeding nymphs have been obtained from horses in December and January, hence it is probable that the nymphal development is in progress during the most of the winter. Those which drop from the hosts late in the autumn probably spend the entire winter in the resting or moulting stage, which they complete on the advent of warmer weather. Little progress is made in moulting during the winter. Enlarged nymphs have been kept in the laboratory at a tempera- ture of about 10° C. for nine months without moulting, but when removed to the temperature of the room, the process would go on to completion in about the usual time. The enlarged nymphs which drop from their hosts at any time during the winter must wait for the warm weather of spring before they can cast their skins and appear as adults. Source of virus for the tick. — The observations in relation to hereditary transmission in the tick indicate, as stated previously, that it is necessary to assume some extraneous source of infection for the tick in order to explain the maintenance of the disease. At the outset it would be natural to inquire whether man can play the same role in the maintenance of spotted fever as he does in that of malaria and yellow fever, and as he possibly does also in the maintenance of human tick fever of South Africa. Without going into a full discussion of all the conditions which would seem to render this impossible, the essential reasons for this conclusion may be given as follows: Ticks which are naturally infected have no relation to the habita- tion of man; they are wood- ticks, not house pests, and are not found in the houses except as they are brought in accidentally on the clothing. The conditions are very different in the relation of Ornithodorus moubata to South African tick, or recurrent, fever. The latter have the habits of the bedbug to a large degree, secreting themselves in the house during the day and coming out at night to feed on the inmates. In this case man may well be the common or only source for the fresh infection of the carriers. For man to Some Aspects of Rocky Mountain Spotted Fever 389 be concerned in the maintenance of spotted fever it would be necessary to assume that ticks which were hitherto normal must feed on him during the course of his illness, and that the females must either mature on him directly in preparation for the laying of eggs, or, after leaving him, must subsequently find some other host on which it may mature. One who has seen the enormous size which the matured female attains, and who knows that a continu- ous feed of not less than a week or ten days is required for this degree of development, can hardly conceive of its occurrence on a patient who has been stripped of clothing and is confined in bed. Spotted fever patients are invariably looked over carefully for ticks, and it is a safe conclusion that those which are found are destroyed with little ceremony. Regarding the possibihty that a tick after once feeding on a patient may escape and subsequently find another host on which it may mature, this must be admitted as a rare possibility, or better as an extreme improbability. The tick travels slowly, covers very short distances, and does not readily find a second host. On account of the virtual impossibility of a female maturing on man, we should have to assume that the second host in this case would be an animal rather than another human. The dog, and perhaps the cat (?), exist as the only reasonable possibilities in the way of such a second host, and they alone render the occurrence under consideration a rare possibility. Given even the matured and infected female, the likehhood of the off- spring being infected is not more than 50 per cent, and, further- more, the chances which operate against the development of the progeny of any one female to the succeeding adult stage are very heavy. Hence, for these various reasons the process mentioned virtually drops out of consideration as a means for the maintenance of spotted fever. The study of small wild animals. — Holding to the correctness of the conclusion just stated, it seemed necessary, in a search for the source of infection for the tick, to turn to its more general food supply, namely, to one or more of the domestic or wild animals on which the arachnid feeds more or less extensively. The question may be raised as to whether it is justifiable to exclude the vege- table kingdom from an investigation of this character. In answer 390 Contributions to Medical Science one can only say that the probabilities of the case seemed to justify such an exclusion. It is the general and firm belief among ento- mologists that such ticks live exclusively on the blood of animals, and that the anatomy of their mouth parts permits of no other type or mechanism of nourishment. I have undertaken no work to determine whether or not the life of ticks may be prolonged by keeping them in the presence of various types of living, dead, or decaying vegetables, or, for that matter, with dead or decaying animal tissues. In a fairly extensive experience, however, they have never been observed in the position of feeding on such vege- tables as are kept with the animals in the tick-proof cages, namely, carrots, cabbage, dandelions, and clover. Concerning the possi- bility that they may feed on dead or decaying animal matter, it can only be said that ticks, as a rule, make haste to leave the body of an animal on which they have been feeding after the death of the latter. As a determining influence, moreover, we had before us the demonstrated fact that the tick readily acquires spotted fever by feeding on the blood of an infected animal, and it seemed rational to assume that it would acquire the disease in a similar manner under natural conditions. Hence, a campaign was outhned, which had as its object a study of the various animals which serve as natural hosts for the tick — a campaign which has not yet been brought to a conclusion, but which has yielded results so suggestive in their character as to fur- nish an ample reward for the labor and difiiculties involved. Certain types of animals were eliminated from the study for what appeared to be justifiable reasons. Because of the limita- tion of spotted fever to the Bitter Root Valley, in so far as western Montana is concerned, it was decided to postpone the study of animals which wander widely. Hence deer, elk, bear, birds, and animals (mountain goat and sheep) which continuously remain remote from human habitation were omitted from the list, at least for the time being. Also, because of the demonstrated resistance of the horse (relative resistance), ox, sheep, and the fowl, the domes- ticated animals were not subjected to further study for the purpose under discussion. Just one definite clue served to indicate the type of animal Some Aspects of Rocky Mountain Spotted Fever 391 which, it appeared, should be selected for the first experimentation. This concerns the segregation of spotted fever not only in the Bitter Root Valley as contrasted with adjacent districts in which the ticks and animal life are identical, but also its localization to the west slope of the valley, and even to particular foci on the west slope. On the basis of this consideration it would seem that an animal to be concerned in maintenance must be one which has a tendency to a more or less permanent abode, one which does not wander widely from its habitation from year to year, hence one which would not be likely to disseminate infected ticks over a large area. The ground squirrel (a spermophile) , ground hog or woodchuck, the rock squirrel, the chipmunk, the pine squirrel, rock rabbit, mountain rat, and perhaps the three species of rabbit found in the district (cotton-tail, snow-shoe, and jack rabbit) were considered as meeting this requirement more closely than other species. Briefly, the plan was: first, to determine the susceptibiHty of these species as it might be indicated by the injection of virulent blood; second, with those which were found to be susceptible by the method of injection, to attempt infection by means of the bites of ticks of known virulence; third, to perform "tick cycle" experi- ments, by which is meant the production of the disease by means of the tick, while at the same time permitting normal ticks to acquire the disease by feeding on the animal during the course of fever; fourth, and last, to attempt to discover individual animals in nature, which were either suffering from spotted fever at the time of obser- vation, or which had survived an infection at some previous time. The last point was to be determined by means of immunity tests, provided, of course, that the previous experimentation should show that one attack in the various species confers lasting im- munity. It has not yet been attacked experimentally. It is to be understood that in the study of the first three points, the animals used in the experiments were taken from localities in which spotted fever had not been known to occur. The first experiments with the ground squirrel were perplexing. It had been anticipated that a rise in the temperature of this animal would serve as a guide to the occurrence of infection, as it does in 392 Contributions to Medical Science the case of the guinea-pig, rabbit, and monkey. The results, however, were very inconstant in this regard. In four out of thirteen animals inoculated, four had absolutely no fever. The shortest febrile period observed was one day, and it persisted no longer than four consecutive days in any of them; in four animals it lasted for three days, and in another for two. "The incubation period varied from one to five days. It was five days in one animal, four days in two, three days in four, two days in one, and one day in one animal. During the incubation period the temperature varied between ioo° and 103°, being for the most part below 102 . 5°. One observation was as low as 97 . 8°. After the cessation of fever the temperature returned to about what it was during the incubation period."' But one of the animals inoculated died, and at autopsy the only discoverable change was a pronounced enlargement of the spleen, as compared with controls which were killed for the purpose of comparison. From this animal, and also from one other which exhibited distinct fever, transfers of blood were made into the guinea-pig, and in both instances the latter developed typical attacks of spotted fever, as shown by the fever, hemorrhages into, and gangrene of, the scrotum, and the changes in the spleen and lymph glands seen at autopsy. The series showed the possibility of infecting the ground squirrel with spotted fever, but it seemed necessary to conclude either that the susceptibility of the different individuals is not uniform, or that the temperature which follows injection cannot be taken as an index of infection. With the hope of obtaining information regarding the last point, "the temperatures of a number of normal gophers were taken for several consecutive days in order to learn its natural variations. It was found to be very inconstant and the irregularities were not always to be accounted for. When the weather was cool, as in the mornings, it was common to observe a low temperature, and for several days it might remain in the vicinity of 97° or 98° F. At other times it was low when the outside temperature was too high to account for the condition. That of the normal animal rarely exceeded 102.5° or 103° F., and it was concluded that this would represent approximately its temperature when in a state of normal Some Aspects of Rocky Mountain Spotted Fever 393 activity, and with metabolism at par. The fact that the gopher is a hibernating and a semi-cold-blooded animal, no doubt is responsible for many of these variations. When in actual hiberna- tion the temperature in three instances was so low that it did not register on the clinical thermometer. It has been observed, also, that there are individual variations in the tendency to go into the somnolent state, that even at the warm temperature of the labora- tory one may be so somnolent that he can be taken out and handled without awakening, while others will be quite active. It seemed more probable, therefore, that the temperature of the gopher is not a good index of his infection rather than that 30 per cent of the species should be naturally resistant to the disease." During the following spring and summer (1908) this was proved to be true. Of a series of squirrels which received quantities of virus from the guinea-pig varying from 0.02 c.c. to 3.0 c.c, and which showed the greatest irregularities in temperature following the injections, every one acquired spotted fever, the diagnosis being established by the injection of blood from each of the animals, at the proper time, into normal guinea-pigs. It may be mentioned as a point of incidental interest that a relative fever developed in a large percentage of the ground squirrels. If the temperature of a particular animal had been found in the neighborhood of 98° or 99° for several days, it would commonly rise to a height of 101° to 103° during the customary febrile period, the latter being scarcely above that of the normal active animal. The experience taught a new point in technic, namely, that for the positive recognition of spotted fever in the ground squirrel it is necessary to carry the inoculation back again from the latter to the normal guinea-pig. Eventually this step was adopted as a uniform procedure in the work with other small wild animals. Following this plan of experimentation, the ground squirrel, ground hog, rock squirrel, chipmunk, and mountain rat were found to be susceptible to the inoculation of the virus of spotted fever, as represented in the blood of the infected guinea-pig on the third day of fever. The first four species were also found to be suscep- tible to the bite of single virulent adult ticks, the infectivity of which had been proved previously on normal guinea-pigs. It was 394 Contributions to Medical Science not possible at the time to study the susceptibiUty of the mountain rat to the bite of the infected tick. It is of theoretical and practical importance to know whether spotted fever is an acute self-Kmited infection in the animals under consideration, or whether it may exist in a chronic form. If it were sufficiently chronic, lasting for a year or more, as in the case of certain protozoan diseases, maintenance could be accomplished without the aid of inheritance in the tick. Inasmuch as the latter does occur the only evident influence of chronicity in the animal would be a more massive infection of the ticks of the locaHty. Investigations have shown, however, that the number of infected ticks, in comparison with the whole, is very small — so small indeed that this fact alone would seem to be a strong argument against chronicity in the four-footed hosts. Two general arguments also speak for acuteness in the small wild hosts: first, that in the other animals which have been studied carefully, and in man, the disease is acute and self-limited; second, that after recovery protracted (probably permanent) immunity is estabUshed against reinfection, and the serums of such animals possess protective antibodies in considerable concentration — much more than is found in any known chronic disease. In only two of the wild animals has the subject been studied specifically, the ground squirrel and the rock squirrel. After they have recovered, the blood is free from virus, as shown by transfers into the guinea-pig, and they resist attempts at reinoculation. One-half c.c. of the blood of the immune ground squirrels was found to protect the guinea-pig against approximately i,ooo pathogenic doses of virus, the blood of the normal squirrel having no protec- tive effect. The possible presence of the virus in some of the internal organs of these animals after recovery has not been investi- gated. The concentration of antibodies in the blood would seem to preclude this. It is not found in the spleen of the guinea-pig and monkey after recovery. It is manifest from previous statements that spotted fever in the tick is very chronic. The relation of the disease to the tick, indeed, does not differ from that of various microparasites to their insect carriers in this respect. Some Aspects of Rocky Mountain Spotted Fever 395 ^^ Tick-cycle'^ experiments. — The "tick-cycle" experiments repre- sented an attempt to duplicate what was assumed to be a step in the maintenance of the disease in nature. Individual ticks, rather than groups of ticks, were dealt with, and the following steps were involved : First, the individual testing of virulent ticks on separate guinea- pigs in order to prove their infectivity before beginning the experiment; second, the testing of the supposedly normal ticks on guinea-pigs to insure their freedom from the virus ; third, the attach- ment of a single infected tick to the ear or head of the small wild animal, permitting the former to feed for a period which was known to be sufficient for the production of the disease; fourth, after the lapse of an interval corresponding to the established incubation period, to draw blood from the heart of the experiment animal, and inject it into the guinea-pig for the purpose of determining the ex- istence or non-existence of spotted fever in the wild animal; fifth, the attachment of one or more normal ticks to the latter as soon as the incubation period has passed; and sixth, the testing of the latter on normal guinea-pigs in order to determine whether or not they had acquired the virus during their feed. There were many failures and embarrassments due to the acci- dental death of valuable ticks, or of animals, or to failure to do the right thing at the right time in all cases; but eventually positive "tick-cycle" experiments were obtained with the ground squirrel, ground hog, and rock squirrel. With the delicate Httle chipmunk the cycles were broken in character. Several times this animal was infected by the bite of virulent ticks, and we were successful in transferring the disease from the chipmunk to the tick, following the inoculation of the former by means of the syringe. In the "cycle" experiments, however, the intermediate step of bleeding from the heart proved fatal in nearly all cases, which, of course, prevented the completion of the experiment. Nevertheless, the broken character of the experiments with the chipmunk does not materially impair their significance. These experiments would have little justification were it not for the fact that the animals mentioned are utilized as hosts by the tick at one stage or another of its development. Particularly on the 396 Contributions to Medical Science ground squirrel, rock squirrel, and pine squirrel (the last not yet studied), larvae and nymphs have been found in great abundance, and on the other animals mentioned, to a less extent. That adult ticks are found less commonly, indeed rarely in my experience, on these animals than the earher stages, in no way invalidates the experiments, which were performed with the adults, since an abun- dance of experience shows that the larvae and nymphs do not differ from the adults in their ability to acquire and transmit spotted fever. We may only bear in mind the possibility that more than one larva may be required for the production of the disease,* whereas a single nymph or adult is sufficient. Maintenance. — In accordance with the results and deductions which have been outlined, it is conceived that spotted fever is maintained as follows: A certain percentage of the female ticks which have acquired the disease as a consequence of feeding on animals, the latter having been infected by other ticks, transmit the disease to their offspring through the egg. The new genera- tion during the process of feeding transfer the virus to certain of the susceptible small wild animals (ground squirrels, rock squirrels, chipmunks, ground hogs, and perhaps others), and this may take place during either the larval, nymphal, or adult stage, hence at various times of the year. During the infection of the wild animal it is required that hitherto normal ticks, either as larvae, nymphs, or adults, acquire the disease by feeding simultaneously with, or shortly after, the feeding of the infected ticks. Regardless of the tick's stage of development at the time it acquired the disease, the virus is retained into the adult period, and in certain of the females reaches the germ cells and again appears in the next generation. The infection of man is an unessential incident for maintenance, and depends on the occasional and accidental bite of the infected adult tick. I shall not discuss the bearing which these results have on the control and the eradication of the disease further than to say that this work naturally should take two directions, one having as its object the destruction of the ticks in proximity to human habitation, * In testing the virulence of eggs, five is the smallest number which so far has proved capable of infect- ing the guinea-pig. Some Aspects of Rocky Mountain Spotted Fever 397 following methods which have proved successful in the cam- paign against Texas fever, and the other leading to the extermina- tion of the susceptible wild species of animals, a procedure which also has as a consequence the curtailment of the food supply for the ticks. Detailed recommendations embodying these points have been made to the State Board of Health of Montana. Microbic etiology. — Inasmuch as rational methods for the control and eradication of spotted fever depend on an accurate acquaint- ance with the means by which man is infected, and the agencies through which the disease is maintained in nature, it is fortunate that these questions could be investigated without any knowledge of the micro-organism which causes the disease. Yet so much theoretical, and perhaps practical, importance pertains to the recognition, and particularly to the cultivation, of the causative agent, that this part of the problem has never entirely dropped out of mind. Following many failures, further study was often deferred until other lines of work which promised solution could be disposed of. With the knowledge that the tick is the agent through which man is infected, one turns naturally, and first of all, to some other diseases of known etiology which are also carried by one or another species of tick, at the same time keeping in mind the dangers of generalizations and analogies. When spotted fever was first studied ticks were known as the carriers of piroplasmas only. On the other hand, since 1903, it has been definitely shown that ticks are the medium of conveyance in at least three diseases which are caused by spirilla, the spirillosis of fowls (Marchoux and Salembini), South African tick fever of man (Button and Todd, and Koch), and a disease of cattle in South Africa (Theiler) . With these analogies in mind, and animated also by the con- flicting opinions of Wilson and Chowning on the one hand, and Stiles on the other, a careful search has been carried on for both piroplasmas and spirilla, but with absolutely negative results in both cases. If it is fair to judge from the known infections with piroplasmas and spirilla, it would seem that the recognition of these organisms should not be attended with great difficulties, particu- larly since the introduction of the Romanowsky method of staining, 398 Contributions to Medical Science with the various convenient modifications. In the case of the known piroplasmoses the blood is often heavily infected, with many parasites in a single field, and even the most minute species, as Piroplasma parvum in the cattle of South Africa, are of such size in relation to the erythrocytes that they could not escape detection in prolonged microscopic studies which were carried on, during all stages of the disease, with both the blood and the various internal organs. Such studies were entirely negative with both fresh and stained material, and after many months of study in which fresh material was always at hand the conclusion was reached that spotted fever either is not caused by a piroplasma at all, or if it is, that the organism is extremely minute (ultra- microscopic) and not susceptible to microscopic demonstration. There are also some general considerations which suggest strongly that spotted fever is not a piroplasmosis. In the first place there is no massive destruction of the erythrocytes in either man or the guinea-pig, the number of which does not fall below a level commonly seen in various febrile infections. There is no hemoglobinuria, such as is seen in many piroplasmoses, and the degree of icterus is not more than frequently occurs in different fevers of bacterial cause, nor is the blood pale and watery. Further, the facts that spotted fever in man is invariably an acute self- limited disease and that the serum of man and animals after recovery possesses protective antibodies in rather high concentra- tion are not in harmony with a piroplasmatic etiology. The results of the search for spirilla were similar. In relapsing fever, in the South African tick fever of man, and in the spirilloses of animals the number of organisms in the circulation is often very conspicuous. In spotted fever, on the other hand, exhaustive search of the blood and tissues in all stages of the disease and painstaking examination of the tissues and eggs of infected ticks have failed to reveal the presence of spirilla. Here again it must be concluded that if spirilla are present they are so small as to justify the assumption that they would be ultramicroscopic or filterable. Similarly the possibility of a leukocytic parasite has always been kept in mind, but such forms have not been observed. From time to time various experiments were performed with Some Aspects of Rocky Mountain Spotted Fever 399 the hope that some light might be thrown on the character of the organism. The virus was found to have a fairly uniform distribu- tion in the various body fluids and organs. It proved not to be an exclusive parasite of either the red or white cells of the blood, but was found in approximately the same concentration in the lymph and serum as in the full blood. Its resistance to heat and light and plasmolysis* was such that no conclusion could be drawn regarding its bacterial or protozoan nature. None of these indirect methods furnished evidence of particular value relating to the char- acter of the organism. A generalization frequently made in former years emphasized the hypothesis that all diseases which are transmitted habitually by insects are probably due to protozoan parasites, the assump- tion resting on facts which were known in relation to many insect- borne diseases, as malaria, piroplasmosis, the malaria of the birds, and more recently, trypanosomiasis. Plainly, however, the newer developments regarding the relation of the flea to plague, and the tick to the relapsing fever of South Africa, both bacterial diseases, remove the ground for the generalization, and in facing a new disease of this type today, the possibiUty of a bacterial rather than a protozoon etiology demands unprejudiced consideration. If, indeed, one were to resort to generalization in the case of spotted fever, it may well be in favor of a bacterial instead of a protozoan parasite, because of the chronicity and low grade of immunity which pertain to the known protozoan infections. Even this, however, cannot be ventured with justifiable assurance because of the num- ber of diseases whose etiology is still unknown. Interest in the micro-organism has been sustained by the results of filtration experiments. Repeatedly, virulent serum, both when concentrated and greatly diluted, lost its infective power after being passed through Berkefeld candles. It is, of course, con- ceivable that the filter may have been more or less occluded by the large number of albuminous particles in the serum, a condition which would interfere with the passage of the organism through the filter. Later, when dealing with virulent eggs from the tick, the filtration experiment was repeated under more favorable con- * Experiments conducted in association with Dr. P. G. Heinemann. 400 Contributions to Medical Science ditions. In one instance, 200 eggs were crushed and thoroughly shaken in 5 c.c. of salt solution. The albumin was so scant that even a faint degree of opalescence could not be detected in the salt solution. This was filtered and the candle washed with 5 additional c.c. of salt solution. Of the filtrate a quantity representing 150 eggs was injected into one guinea-pig, and another, representing 50 eggs, into a second. Neither of the animals developed spotted fever although 5, 10, and 15 of the unfiltered eggs produced the disease in controls. This was accepted as proving finally the non- filterability of the virus through the ordinary, unmodified Berkefeld candle, and is further evidence that the organism is of such size that it should be seen with the microscope. Following this brief account of negative findings, reference may now be made to a bacillus which has been found in association with the disease under interesting circumstances. In the early part of the studies a diplococcus-like body with an eosin staining intermediate substance was encountered, first in the blood of an infected monkey. Further examination disclosed the presence of similar bodies in the blood of infected guinea-pigs, and later in man, and it came to be the common experience that they could be found in any preparation of infected blood during a search of not more than a half-hour's duration, the specimens being stained with Giemsa's solution after fixation in absolute alcohol. The morphology varies somewhat regarding the size and proximity of the two staining bodies, and even small bacillary forms are encountered. In spite of this constant observation the conditions were such that a positive interpretation was impossible merely on the basis of the microscopic findings. The forms are very minute and often appear as a pair of granules which might arise from a disintegrated leukocyte or from a fragmented nucleus. As ren- dering the latter interpretation plausible, disintegrating leukocytes frequently occur in the blood, particularly of human patients, and in such examples small globules of chromatin may sometimes be found in proximity to the injured cell. Further, my own attempts, and the attempts of Mr. P. G. Heinemann, to cultivate an organ- ism of this or any other character from the blood of patients and animals had been fruitless. Some Aspects of Rocky Mountain Spotted Fever 401 Some months ago it was possible to take up seriously a long- deferred line of investigation, namely, a study of the organs and eggs of the tick with the hope that the micro-organism might be present in such concentration or form that its identification would be possible. The disease being hereditary in the tick, it seemed that the freshly laid eggs might be a most favorable medium for search as in South African tick fever, one in which there would be comparatively few opportunities for confusion. The first preparations from a batch of virulent eggs showed the presence of an astounding number of organisms which, for the most part, have the form of minute polar staining bacilli. Frequently, two or three are attached end to end, and the staining poles give the appearance of a chain of four or six cocci. In the eggs, as a rule, the amount of intermediate substance interposed between the poles is compara- tively large, the stained masses being rather widely separated. Frequently, however, the poles are more closely approximated, such forms resembling closely the " diplococcoid " bodies seen in infected blood. Still other bacilli are found which stain solidly, and they too are often in pairs (diplobacilli) . In the latter it appears that the staining substance has not yet become segregated into the poles of the organism. Intermediate forms in which the chromatin of the two ends communicates at one or both borders of the cell substantiate this belief. When specimens from a batch of eggs are examined from day to day a change takes place in their predominat- ing morphology, as the cells of the embtyo tick multiply. The sharply bipolar forms with the wide intermediate substance gradu- ally become fewer and fewer, and the presence of some which have become abnormally large and swollen indicates that many are in process of dissolution. On the other hand the number of smaller "diplococcoid" forms undergoes an increase, and many of them eventually appear in groups within the cytoplasm of the embryonic cells of the tick. The growth of the latter, however, hardly com- pensates for the disappearance of the larger bipolar forms, and the impression is gained that the total number of organisms has under- gone a decrease. In spite of this change the virulence of the eggs does not decrease. In one batch in which five eggs produced infec- 402 Contributions to Medical Science tion when freshly layed, the same number was equally virulent at a later period, when the total number of organisms had appar- ently undergone a great decrease and none remained but the smaller **diplococcoid" bodies, which for the most part were intracellular. This series of observations, which has been repeated many times, leaves the impression that the cells with the large amount of inter- polar substance are involution forms, whereas the shorter "diplo- coccoid" bodies and the solid staining cells represent the more active proliferating organism. Examination of the alimentary sac, salivary glands, and oviduct disclosed the presence of large numbers of similar organisms, but here the "proliferating" forms are much more numerous than the so-called involution forms. These organs, in a certain sense, represent free surfaces and render possible the removal of the products of bacterial metabolism and a renewal of food substances for the micro-organisms. The eggs, on the other hand, are closed sacs, covered with a mucilaginous substance, which renders impos- sible the separation of deleterious and inhibiting material; when the bacterial proHferation has reached a certain degree, further growth may also be limited because of exhaustion of the food supply by the growing embryo. Thus it is conceived that only the younger and more resistant bacilli survive, and appear in the embryonic cells and later in various tissues of the active tick. That a specific agglutination reaction can be obtained with the organisms as they exist in the eggs has been reported previously." From fifty to one hundred eggs, depending on the number of organisms they contain, are crushed in about 1/20 c.c. of salt solu- tion, and the suspension shaken until the distribution of the bacilli is uniform. On account of the small quantity of material avail- able it is necessary to use the hanging-drop method, and since the emulsion is so dense in refractile elements from the eggs, determi- nation of the result cannot be made with the moist preparation. After two hours' residence in the thermostat, the cover glasses are removed, the drops permitted to dry, after which they are fixed in absolute alcohol and stained with the solution of Giemsa. I may quote the results from the article mentioned above: "The serum of the normal guinea-pig either causes no aggluti- Some Aspects of Rocky Mountain Spotted Fever 403 nation at all, or, at the most, produces only slight agglutination in proportions of i to i and i to 20. Dilutions higher than this cause no agglutination. In testing the agglutinating powers of immune serums, three animals which had been infected from dif- ferent sources, and had recovered, were used. One (1751) had been infected with a dermacentor from Idaho; another (1692) with a strain handed down direct from guinea-pig to guinea-pig for nearly three years without the intervention of ticks, the original infection having been obtained from the blood of man; the third (1757) with a strain kept in the same way since last spring. Graded dilutions, beginning with i to i and going as high as i to 400, were used in the different series, with the striking result that a complete agglutinating power was present in the three immune serums in dilutions up to i to 320. It was somewhat less in a dilution of I to 400. The highest dilution which will cause clumping has not been ascertained. "No fresh immune serum from man is at hand, but tests were made with three specimens which are about five, seven, and nine months old, respectively. They have been preserved in the ice- chest with the addition of 0.3 per cent of chloroform. The peculiar phenomenon of failure to agglutinate in concentrated solution was noted with all three.* With the oldest serum no agglu- tination occurred until the dilutions of i to 320 and i to 400 were reached, when incomplete clumping was produced. With the second there was no agglutination in the dilution of i to i , distinct clumping in i to 20, i to 40, and i to 80, with little or none in higher dilutions. In the serum of five months' standing the reaction was absent in the dilutions of i to i , i to 20, and i to 40, positive but not complete in i to 80, i to 160, and i to 240, with very little clumping in i to 320, and i to 400. Normal human serum caused clumping in a dilution of i to i, a very slight amount in i to 20, and none at all in the higher dilutions." These experiments have now been repeated many times with invarying results, in one instance distinct agglutination being obtained with the immune serum in a dilution of i in 1,000. The result of the agglutination test, then, is a distinct one. *The " agglutinoid " phenomenon. 404 Contributions to Medical Science The blood of the immune animal contains agglutinins in high concentration for the bacillus, whereas these bodies are almost entirely absent from the serum of the normal animal. Hence, in the light of experience we must either conclude that the organism is present in the body of the animal during his attack, or suffer our faith in the value of the agglutination test for diagnostic purposes to be shaken. Having outlined the observations which favor the position of the bacillus in question, I may refer to other findings, which apparently throw some discredit on it. Scientific proof, among other things, would seem to demand that the bacillus be found constantly in all virulent ticks and eggs, and that it be absent constantly from all non-virulent material. In the early stage of this study eggs from ticks which had been definitely proved to be non-virulent were not at hand, but in view of the presence of the organism in the internal organs of the viru- lent tick, it was considered that the tissues of non-virulent ticks would serve as satisfactory controls in the place of normal eggs. It so happened that in the material then available for study no bacilli could be discovered in the salivary glands, genital organs, and alimentary sac of normal ticks, and the results were considered as satisfactory and confirmatory. Later, when more material was available, and particularly when the eggs from non-virulent ticks were obtained, it was, to say the least, a cause for astonishment, that bacilli of exactly similar morphology were found, often in great abundance, in both eggs and tissues which experiments showed to be absolutely without viru- lence. Continued observations, indeed, showed that an organism of this character is present with almost perfect constancy in the eggs and tissues of Dermacentor venustus and D. modestus, regardless of their infectivity. In only occasional individuals has it been impossible to find them. Not only do they resemble exactly the bacilli found in virulent eggs and tissues in their morphology, but they respond to the specific agglutination test in the same manner, i.e., they are agglutinated by immune spotted fever serums in high dilutions and not by normal serums, except in dilution of i-io or 1-20. Therefore on the basis Some Aspects of Rocky Mountain Spotted Fever 405 of morphology and the agglutination test there seems to be no question but that the bacilli found in avirulent eggs are identical with those found in the virulent.* The situation, therefore, is a perplexing one. In favor of the causative relationship of the bacillus are the following two points: first, the presence in infected blood of forms which correspond in their morphology with those seen in the eggs and tissues of the tick; and, second, the positive agglutination reaction with immune serums in contrast with a negative result with normal serums. Against it is the fact that organisms of similar morphology, which respond to the agglutination reaction in the same manner, are found in the tissues and eggs of the non-virulent ticks. One may entertain one of three hypotheses as affording an interpretation of the results. It may, in the first place, be assumed that the organism described is merely a constant associate of the true virus in the infected animal, and, consequently, although not having a necessary relationship to the disease, it may cause the formation of agglutinating substances. An example of this con- dition apparently exists in hog cholera, in which the true virus is very minute and filterable, whereas the hog cholera bacillus is only a more or less constant associate in the general infection. Against such an assumption in this case, however, are the facts that the virus of spotted fever is not filterable, and in the perfectly fresh eggs, in which there is the least possible morphological con- fusion, the bacillus described is the only organism which can be recognized. Or, second, one may suppose that for some unknown reason the agglutination reaction is not specific in this case, and that the clumping of the bacilli by the immune serum, in contrast to the negative action of normal serum, is brought about by some pecuhar and unusual conditions in the experiments as they were performed. The possibility, however, does not seem sufl&ciently strong to shake our faith in the time-honored specificity of the agglu- tination reaction. Or, finally, it may be assumed that the bacillus in question is the true cause of spotted fever, and that it has a tendency to undergo a loss of virulence as a consequence of long * Such bacilli could not be found in the eggs and tissues of specimens of Ornithodorus meenini, Amblyoma americanum, and an eastern Dermacentor [variabilis (?)]. 4o6 Contributions to Medical Science residence in the tissues of the tick, thus explaining the non-infectiv- ity of the eggs and tissues of "normal" ticks which are rich in similar micro-organisms. Such an assumption carries with it the idea that the virus of spotted fever in an attenuated form has an exceedingly wide distribution in the ticks of the Rocky Mountain states, and that it assumes virulence only under certain conditions, such as might be attained by sufficiently frequent passage through some suitable animal. If a virulent tick and its progeny do not, within a certain period, feed on a susceptible animal and convey spotted fever to the latter, it is possible that the strain of micro- organisms which they contain becomes hopelessly avirulent and incapable of producing infection in any animal save the tick itself. If, on the other hand, they do convey the disease to susceptible animals before the strain has lost its virulence, the organisms may again be acquired by other ticks, and the period of pathogenicity thus prolonged. Indeed, scarcely any fact is better known among bacteriologists than the loss of virulence which a micro-organism may undergo for one species of animal as a consequence of repeated passage (long residence) in another species, and the frequency with which various pathogenic organisms lose virulence for a host in the absence of occasional passage through the host is known to all. Further, I may call your attention to evidence already cited that strains of the spotted fever virus showing differences in viru- lence actually do exist in nature, as illustrated by the low virulence of the Idaho strains for both man and the guinea-pig in contrast with the high virulence of the virus of Montana. Reasonably, therefore, strains of still less or no virulence for these animals may exist under natural conditions. As a working basis, then, I have chosen the last assumption as being the most reasonable and in best accord with the known facts. On the one hand it is hoped that the bacillus may yet be cultivated, in spite of many failures up to the present time, after which proof of its position may be more easily obtained. At the same time another type of experiments is being carried on, which, there is reason to believe, will be equally decisive as to the rela- tionship of the bacillus. The possibihty of vaccination against spotted fever has already been determined, and the method of Some Aspects of Rocky Mountain Spotted Fever 407 sero-vaccination has been described." Successful vaccination, not hitherto reported, has been accompUshed by the injection of tissues of eggs of virulent ticks (rich in bacilli) after the material has been sterilized either by desiccation or by chloroform. If the micro-organism under discussion is the cause of spotted fever, and if the bacilli found in a virulent material are merely attenuated strains of the virulent organism, equal success may be attained in vaccination with the latter. Although this has, indeed, been accomplished quite recently, it seems desirable to await the study of a more abundant material before submitting further conclusions. Prevention and serum therapy. — ^To go into detail regarding methods of prophylaxis and control of spotted fever, which have been evolved on the basis of the results cited, would be to trespass too greatly on your time. An account of the procedures recom- mended is published in a Report to the State Board of Health of Montana, previously referred to. Serum therapy, as practiced with an immune serum derived from the horse, has not afforded decisive results, and perhaps for several theoretical reasons. In the first place, a degree of immunity in the horse greater than that established by simple recovery from the infection cannot be obtained until the organism can be cultivated and injected in excessive quantities. The repeated injection of virulent blood into an immune animal is of little or no value in increasing the quantity of protective antibodies because of the comparatively small quantity of virus which such blood contains. The immune serum is of such strength that from 0.3 to 0.5 c.c. protects against i.o c.c. of virus (approximately 1,000 pathogenic doses). Second, experiments with animals have shown that the serum cannot be expected to exert a marked cura- tive effect when it is given later than the second or third day after inoculation, and even ^in this case such an amount is required as to reach an almost prohibitive quantity when translated into terms of the human body weight. Further, it is very exceptional to obtain cases of spotted fever earlier than the third to the fifth day of sickness, i.e., until the eruption renders the diagnosis reason- ably certain. Third, it is probable, though not a demonstrated fact, that the serum is bactericidal rather than antitoxic in its 4o8 Contributions to Medical Science nature. If this proves to be correct, experience with other similar serums augurs a low curative value regardless of the concentra- tion of germicidal antibodies. No unfavorable effects, so far as can be determined, have attended its use. The advisability of modifying the technic of administration so that large quantities are given intravenously early in the course of infection is still a subject for consideration. REFERENCES. 1. Med. Sent., 1908, 16, p. 704. 2. Ibid., 1899, 7, p. 433. 3. Jour. Infect. Dis., 1904, i, p. 31. 4. Bull. No. 14, Hyg. Lab. U.S. Pub. Health and Mar. Hosp. Serv. 5. Bull. No. 20, ibid. 6. Fourth Biennial Report, State Board of Health of Montana, 1907-8, p. 137. 7. Jour. Infect. Dis., 1904, i, p. 31. 8. Jour. Am. Med. Assn., 1907, 49, p. 84. 9. Fourth Biennial Report of State Board of Health of Montana. 10. Jour. Am. Med. Assn., 1909, 52, p. 379. 11. Jour. Infect. Dis., 1908, 5, p. 220. UNFINISHED EXPERIMENTS OF DR. HOWARD T. RICKETTS ON ROCKY MOUNTAIN SPOTTED FEVER. Benjamin Franklin Davis. (From the Pathological Laboratory of the University of Chicago.) Following the death of Dr. Ricketts it became necessary to discontinue the work on Rocky Mountain spotted fever pursued in this laboratory under his direction. A number of problems were being studied but only in the case of a few were the results obtained such that conclusions might be drawn from them. It is the pur- pose here to mention the various paths along which the research was progressing and to indicate the conditions obtaining in each when the work was stopped. This paper does not include, however, investigations which had been brought to a reasonable degree of conclusion. I. RESULT OF ATTEMPTS TO CULTIVATE THE VIRUS OF ROCKY MOUNTAIN SPOTTED FEVER. It seems quite definitely to have been shown that the virus of Rocky Mountain spotted fever cannot be maintained on any of the ordinary culture media at room or incubator temperature, aerobically or anaerobically or under a partial oxygen pressure such as may be obtained by the method used by Novak in his study of the bacillus of Bang.'' Virus added to uncoagulated egg-white (hen's egg) or to yolk of egg, or to a mixture of white and yolk, and left at room temperature, gradually loses its infec- tivity for guinea-pigs in the course of four or five days. Various serum and blood media have failed to give positive results as have also media containing a filtrate of various bacterial cultures. It was thought that possibly the virus might be destroyed in vitro by the action of immune bodies contained in the serum and the attempt was made to reduce this action to a minimum by dilu- tion of the serum with salt solution, with broth, and by the addition ' Ann. de I'Inst. Pasteur, 1908, 22, p. 541- 409 4IO Contributions to Medical Science of serum in which an anticomplement specific for guinea-pig complement had been developed, but apparently without success. The results of preliminary experiments with the intraperitoneal collodion sac method of cultivation seem to indicate that the virus whether undiluted, whether suspended in salt solution (0.85 per cent), in ordinary nutrient broth or in anti-complement serum, loses its infectivity very rapidly. That the virus does not pass through the collodion membrane seems certain because (i) guinea-pigs in which sacs containing virus have been placed do not contract spotted fever, and (2) they develop the disease at subsequent immunity tests. Clegg's^ reported success in cultivat- ing B. leprae in symbiosis with amebas suggested a similar experi- ment with the spotted fever virus. Up to this time, however, no substantial progress had been made. II. DISAPPEARANCE OF ORGANISMS FROM THE EGGS AND TISSUES OF TICKS MATURED ON IMMUNE GUINEA-PIGS. Dr. Ricketts has described^ (p. 36) a minute, bipolar body which is found almost constantly in the eggs and tissues of normal and infective Idaho and Montana ticks (D. modes tus and D. venustus, respectively), and which he was inclined to believe bears a causal relationship to Rocky Mountain spotted fever. Granting that this bipolar body is the cause of the disease, the assumption was made that by feeding ticks in whose tissue such bodies were present on immune guinea-pigs, antibodies would be absorbed by the ticks from the pigs and would destroy the supposed bacteria, with the result that the latter would disappear from the tissues of the ticks and that the eggs of such ticks would be free from bacilli. In accordance with this idea two batches of bacilli-rich larvae were selected. One batch was matured on immune guinea-pigs; the other batch was matured on normal guinea-pigs as a control experiment. Although this experiment is unfinished, the indica- tions seem to be that there is a progressive diminution in the num- ber of organisms in the tissue and eggs of the ticks of both batches. Actual counts have not been made; the above is simply the opinion, ' Phil. Jour. Sc, B. Med., igog, 4, p. 77. 'Jour. Am. Med. Assn., 1909, 32, p. 379- Unfinished Experiments of Dr. Howard T. Ricketts 411 based on examination of the material, of the one (Dr. Maria B. Maver) who was carrying out the experiment. An interpretation of such a result is hardly possible at this time. If such a diminu- tion does occur, however, several explanations suggest themselves, viz.: (i) the so-called bacillus may bear no relation to spotted fever but may be some form which is picked up by the ticks in nature and which gradually disappears from the ticks when the latter are removed from their natural surroundings and hence from the source of supply; (2) the bacillus may be the cause of spotted fever; it disappears from the ticks matured on immune pigs in accordance with the considerations outlined above. To explain its disappearance from ticks matured on normal pigs we may reason as follows: The ticks used were non-infective, which means presumably that the bacilli which they contained were avirulent, that is, were not capable of proliferation in the body of a guinea-pig. If this were the case, then the blood of a normal guinea-pig might be supposed to exert an influence on the organ- isms similar to that of the blood of an immune pig with the result that the bacilli would tend to disappear from the body of the tick. This explanation seems to gain some support from the results of vaccination experiments, to be described later, in which suc- cessful vaccination against the active virus was obtained by the use of eggs and tissues from non-infective but bacilli-rich ticks. III. THE PRESENCE OF INFECTED OR IMMUNE ANIMALS IN NATURE. This problem was assigned to Mr. J. J. Moore for the spring and summer of 19 10 in the hope that the results would test the hypothesis of Dr. Ricketts that Rocky Mountain spotted fever is maintained in nature by a continuous cycle of passage from infected ticks to susceptible animals and from animals thus infected back to tick larvae, nymphs, and adults. Infective ticks have been found in nature; it has been shown that certain of the smaller wild animals are susceptible to the disease and that ticks may not only acquire the virus from infected wild animals, but may in turn transmit the virus to other animals. The finding of infected or immune animals in nature would, therefore, complete the evidence. The method to be followed was, briefly, to capture wild animals, 412 Contributions to Medical Science to inject some of their blood into normal guinea-pigs and to watch the pigs for symptoms of spotted fever. In case spotted fever failed to develop, each wild animal was to be given an injection of virulent guinea-pig blood and, after a suitable interval, again bled for injection into a normal guinea-pig. The production of spotted fever in the first guinea-pig would prove that the wild animal under observation was suffering with the disease. Failure to produce spotted fever in the second guinea-pig would prove that the wild animal was immune to the disease and had therefore probably suffered from it at some previous time. It would be necessary to make the passage into the guinea-pig because in none of the susceptible wild animals thus far tested are the symptoms produced by the spotted fever infection sufl&cient for diagnosis. Unfortunately the experiments outlined could not be carried out. IV. INFECTIVITY OF THE EGGS OF INFECTIVE AND NON- INFECTIVE TICKS The technic of these experiments was as follows: normal guinea-pigs were given intraperitoneal injections of an emulsion in salt solution (0.85 per cent) of the eggs to be tested and their temperatures taken for the subsequent ten days. A week or two later, an immunity test, consisting of the injection of i c.c. of fresh virus, i.e., blood from a sick guinea-pig, was given; if the animal failed to develop a fever after the egg injection and became infected in his immunity test or if he did develop a fever follow- ing the injection of the eggs and still was not immune to the active virus, the result of the experiment was declared to mean that the eggs were non-infective. The outcome of the immunity test then furnished the basis for the interpretation of results. At first a series of guinea-pigs was used in each experiment, dilutions of egg emulsion equivalent to 5, 10, 20, 40, 60, and 80 eggs being injected into susceptible animals. Later, and this applies to the majority of the experiments, a uniform dose, equivalent to 50 eggs, was employed. The eggs were, on the average, less than one week old when the experiments were performed and had been left in the dark, at the temperature of the ice-chest, since they were laid. Unfinished Experiments of Dr. Howard T. Ricketts 413 The results are: Of 12 infective Montana females (D. venustus) the eggs of 3 produced infection — 25 per cent; of 16 non-infective Montana females none of the eggs were virulent; of 24 infective Idaho females (D. modes tus) the eggs of but 3 were virulent — 12| per cent; none of the eggs of the 23 non- infective Idaho ticks which were tested produced the disease. Eggs from 63 of the batches used were examined microscopically for bacilli and in all cases typical bipolar organisms were found; 12 batches were not examined. In five of the six experiments which resulted positively the eggs were injected in quantities vary- ing from 5 to 80 according to the method mentioned above. In all such cases 5 eggs proved sufficient to produce infection. Work on the infectivity of emulsions in salt solution of larvae, nymphs, and the tissues (5) of infected adult ticks was begun and some positive results were obtained with all the materials which had been prepared. The experiments are too few in number to justify more extensive recapitulation. The results of these experiments seem to indicate that but a small proportion of infective ticks transmit the virus through the egg, thus supporting the contention that the virus is kept alive in nature by passage from tick to wild animal and from wild animal to tick. It may be that in the egg the virus passes through a developmental cycle of some kind in certain phases of which it is non-virulent, or possibly the technic was not suitable to the object of the experiment, all of which may be decided only by further investigation. In a measure, the results seem to support the idea that the Idaho virus is less virulent than the Montana virus, in that there were but half as many positive results using the eggs of the Idaho ticks as there were when the eggs of Montana ticks were injected. Such a conclusion on the evidence at hand, however, must be regarded as tentative only. V. VACCINATION WITH TICK EGGS AND TISSUES. In one of his last communications' Dr. Ricketts said: "Suc- cessful vaccination, not hitherto reported, has been accomplished by the injection of the tissues and eggs of virulent ticks (rich in ' Med. Rec., igog, 76, p. i. Date 6/30 7/1 7/2 Temperature 103 103.2 Date 7/8 7/9 7/10 Temperature 102 101.6 414 Contributions to Medical Science bacilli) after the material has been sterilized either by desiccation or by chloroform." It seems desirable to supplement this state- ment by the publication of tj^^ical experiments and the tabulation of the results obtained up to the time when the work was discon- tinued. This work was begun by Dr. Ricketts and subsequently carried on at various times by Mr. R. M. Wilder, Dr. Maria B. Maver, and Mr. W. F. Petersen. Experiment i. — June 29, 1909. Dried eggs of Montana female No. 49. They were virulent when fresh and contained bacilli. They had been dried for about 4 months. None of the eggs hatched, although the embrj'os had begun to form. 354 eggs were injected intraperitoneally into guinea-pig 2244 after they had been ground up in salt solution in an agate mortar. 7/3 7/4 7/5 7/6 7/7 103 102.4 101.6 102.6 102.2 7/11 7/12 7/13 7/14 7/1S 102.2 101.8 102.7 103.7 102 First immunity test, j-ij-'og — i c.c. virus intraperitoneally. Date 7/18 7/19 7/20 7/21 7/22 7/23 7/24 7/25 7/26 7/27 Temperature loi 102.4 103.2 102 102.2 102.6 102.4 102.2 102.8 102.6 Date 7/28 7/29 7/30 7/31 8/1 8/2 8/3 8/4 8/s Temperature 102.4 103.8 102.6 102 101.4 102.4 103 102.8 101.8 Second immunity test, ii-6-'o9 — i c.c. of virus was injected intraperitoneally. The result paralleled that of the first immunity test. Therefore vaccination was successful, the virus used for the tests being potent as shown by the production of spotted fever in the control animal. Experiment 2. — September 26, 1909. D. venustus, female 190. Infected with "natural" tick strain on about 6-29-'o9; tested on guinea-pig 7-26-'o9 and found infective; killed and dissected 9-20- '09: Many bacilli found in ovary, gut, and saUvary glands. Nearly all of the salivary gland removed for staining. Fair portion of ovary removed, and a small part of gut. Organs triturated in salt solution, and a drop of chloroform added to sterilize. After two or three daj's placed in desiccator and allowed to remain there in a partial vacuum for three days. On 9-26 the material was re-emulsified in salt solution and injected intraperitoneally into guinea-pig. Date 9/27 9/28 9/29 9/30 lo/i 10/2 10/3 10/4 lo/s Teinp)erature 103.4 102.6 102.8 102 102.2 103.2 102.6 102.8 103 Immunity test — i c.c. virus intraperitoneally io-io-'o9. Date lo/ii 10/12 10/13 10/14 10/15 10/16 10/17 10/18 10/19 10/20 Temperature 102. i 102.9 102.3 103.2 103.2 103.2 102.4 102.9 102.8 Date 10/21 10/22 10/23 10/24 10/25 10/26 10/27 10/28 10/29 Temperature 103.4 103 103 102.4 102.4 103 102.8 102.6 102.8 A second immunity test, given 40 days later in order to rule out a possible passive immunization derived from the injected tick tissues, resulted similarly to the first; therefore the injection of sterilized tissues from infected ticks appears to have induced a high degree of immunity in guinea-pigs. The duration of this immunity was not tested. Unfinished Experiments of Dr. Howard T. Ricketts 415 The serum of this animal appeared to possess but little, if any, power of protecting other animals when injected simultaneously with large doses of virus. An accurate immunological analysis was not attempted. In the same communication Dr. Ricketts said further: "If the micro-organism under discussion is the cause of spotted fever, and if the bacilH found in avirulent material are merely attenuated strains of the virulent organism, equal success may be attained in a vaccination with the latter. Although this has, indeed, been accomplished quite recently, it seems desirable to await the study of a more abundant material before submitting conclusions." The technic of these experiments was the same as that used in the work with the material from ticks whose infectivity had been proven. The results are summed up in the accompanying tables. These tables show: (i) that practically all the tick material used contained bacilli; (2) of the experiments with the eggs of Montana ticks (D. venustus) 33 per cent yielded positive results, while with the tissues, 50 per cent were positive; of the experiments with the Idaho ticks (D. modestus), 11 per cent were positive, while with the tissues in 66 per cent vaccination was obtained ; of the positive results with the eggs of the D. venustus 33 per cent were obtained with the eggs of non-infective ticks (avirulent material) while in the case of the tissues 50 per cent of the positive results were obtained with virulent material and 50 per cent with non- virulent material; of the positive results with the eggs of the D. modestus, all (100 per cent) were obtained with the eggs of infect- ive ticks, as were also the positive results obtained through the injection of tissues. Although these experiments must be regarded as being purely of a preliminary nature, they seem to indicate quite strongly that vaccination with tick eggs and tissues is pos- sible and that this may be accomplished by the use of material from non-infective ticks (of the Montana strain at least) about as readily as with material from ticks whose infectivity has been proven. This would seem to bear out the suggestion quoted above as to the possible relationship of the bacillary forms found in certain normal and in apparently all infective tick eggs and tissues to spot- ted fever and to each other. The ideal control for such experiments 4i6 Contributions to Medical Science — that is, the use of eggs and tissues which are free from bacilli — has not been attempted. TABLE I. Showing the Result of Experiments in Vaccinating Guinea-Pigs against Spotted Fever by the Injection of Tick Eggs and Tissues. Ticks No. Tests of Eggs (Av. No. Eggs per Test =400) No. Infec- tions from Eggs No. Suc- cessful Vaccina- tions No. Unsuc- cessful Vaccina- tions .0£ "> S No. Suc- cessful Vaccina- tions No. Unsuc- cessful Vaccina- tions D. venustus D. modestus 9 19 I 3 (33%) 2 (11%) 6 16 4 3 2 (so%) 2 (66%) 2 I Total 28 I S (i8.S%) 22 7 4 (S7%) TABLE 2. Showtng the Relation of the Bacill,ary Content of Eggs and Tissues to the Infectivity of THE Corresponding Ticks — the Ticks Being Tested on Normal Guinea-PigS" before the Eggs Were Laid and before Being Dissected for the Purpose of Obtaining the Tissues. The Tissues Used Consisted Usually of Ovary, Salivary Gland, and Gut. Ticks No. Sets of Eggs Which Contain Bacilli No. Sets of Eggs not Ex- amined No. Sets of Eggs from Infective Ticks No. Sets of Eggs from Non- infective Ticks No. Sets of Tissues Which Contain Bacilli No. Sets of Tissues not Ex- amined No. Sets of Tissues from Infective Ticks No. SeU of Tissues from Non- infective Ticks D. venustus D. modestus. . . . 8 i8 I I 5 4 7 12 4 3 2 3 2 ToUl 26 2 12 16 7 S 2 TABLE 3- Showing the Relation of the Vaccinating Power of Eggs and Tissues to the iNFEcnviTY op THE Corresponding Ticks. Ticks No. Positive Results FoUowing Injection of Eggs of In- fective Ticks No. Positive F^rng Tot^^No. Injection of ^°^^^J^ Eggs of Non-! JthEek infective ^^^° ^^gs Ticks Total No. Positive Results with Tissues No. Positive Results Following Injection of Tissues of Infective Ticks No. Positive Results Following Injection of Tissues of Non-infect- ive Ticks D. venustus D. modestus 1 (33i%) 2 (100%) 2 (66?%) 3 2 2 2 1 (so%) 2 (100%) I (So%) Total 3 2 5 4 3 VI. ALTERATIONS IN THE VIRULENCE OF SPOTTED FEVER VIRUS IN THE BODIES OF TICKS. From the much lower death rate among the patients infected with the spotted fever of Idaho, as compared with that of those suffering from the Montana disease; from the much milder clini- cal course of the former; and from the fact that in 1908 it had Unfinished Experiments of Dr. Howard T. Ricketts 417 been found impossible to keep the Idaho disease going in the guinea-pig by successive passages, the conclusion seemed to be warranted that the Idaho virus was less virulent for man and for the guinea-pig than the Montana virus — the probable identity of the two having been established by the similarity of clinical mani- festations and by the results of immunity tests. The question now arose, assuming that the viruses are identical, why this dif- ference in virulence ? It had been shown that the Idaho tick could acquire and trans- mit the spotted fever of Montana and that the Montana tick could acquire and transmit the spotted fever of Idaho. As these ticks differ slightly from each other morphologically it was assumed that other differences might be present of such a nature as to alter the spotted fever virus which happened to become resident in the tissues, thus accounting for the difference in the virulence of the two strains. Experimental demonstration was attempted as follows: Alteration of Virulence of Spotted Fever Virus in the Idaho Tick. The plan was to infect Idaho ticks (D. modes tus) by allowing them to feed on guinea-pigs which were infected with the Montana spotted fever, and to keep the disease going in them through inheritance for as long a period as need be. As normal guinea-pigs were infected by these ticks, consecutive passage was to be made from such guinea-pigs through a series of 7 to 10 generations in order to determine whether or not the disease would "run out" in the guinea-pig, this being taken as a test of virulence, based on the experience of 1908. Idaho Disease in Montana Ticks (D. venustus). The primary object of this experiment was to determine whether residence of the Idaho spotted fever in the Montana ticks would result in an increase of the virulence of the Idaho disease. The test of increase of virulence would be the possibility of keeping the disease alive by continuous passage in the guinea-pig after it had been in the Montana ticks. Regarding the first of these, the Montana fever in the Idaho 41 8 Contributions to Medical Science tick, but three experiments were carried through, of which a brief statement follows: 1. Duration of residence of virus in the tick, approximately 19 days; infected guinea-pig on September 12, 1909. Successive passages were made into guinea-pigs in each case on the third day of fever; all, excepting one, died. There was no decrease in viru- lence manifest. 2. Duration of residence of virus in tissues of the tick was much shorter, since the transfer from infecting pig to test pig was immediate. The second pig ran a typical course of spotted fever and successive passages were made from him into the seventh generation of pigs. All of these pigs died except the last which was killed for other purposes. The infection appeared to be as virulent as the Montana disease ever had been. 3. The eggs of this tick were injected into guinea-pig on March 3, 1909. Residence of virus in tissues of the tick about two months. The guinea-pig had a typical attack of spotted fever; consecutive passages were made into the fifth generation of guinea-pigs with no sign of a decrease in virulence. In so far as the experiments were carried, then, no alteration in the virulence of the Montana virus in the Idaho tick was demon- strated. Concerning the second line of experiments — the Idaho disease in the Montana ticks — no work has been done, because the Idaho strain obtained from the patient "Spaniard," which was the only one available at the time, proved to be of such virulence, con- trary to previous experience with the Idaho strains, that it could in the first instance be kept alive by passage, thus destroying the criterion for determining the increase of virulence. COMPLEMENT DEVIATION IN ROCKY MOUNTAIN SPOTTED FEVER.^ Benjamin F. Davis and William F. Petersen. {From the Pathological Laboratory of the University of Chicago.) The experiments here described have been made at various times in this laboratory, the original object being to provide a basis for an early diagnostic test for Rocky Mountain spotted fever. The first experiments were carried out by Liborio Gomez at the suggestion of Dr. Ricketts and we shall give a brief outline of his unpublished results. Gomez used an ox-rabbit system, using the serum from infected guinea-pigs as antigen (drawn during second to fourth day of fever) and immune serum from guinea-pigs and rabbits as antibody. He first determined that the serum contain- ing the antigen and the serum containing the antibody possessed no hemolytic amboceptors for ox corpuscles. His experiments then followed the method of Bordet-Gengou. A positive devia- tion was obtained in the case of one guinea-pig which had recovered from the disease and which had received an immunity test some ten weeks before being killed. In a recovered rabbit he also obtained a positive though weak fixation of the complement. In two guinea- pigs killed four or five months after recovery from a single attack of spotted fever Gomez found a retardation of hemolysis of from two to five hours, as compared with controls of immune serum and virus alone. In his experiments with the immune guinea-pig serum Gomez left the tubes containing the complement-antibody-antigen mixture in the incubator for over 14 hours before adding the hemolytic system. This period of time is long and leaves some doubt as to the value of the result obtained. Gomez notes that the smaller doses of antigen-antibody recommended by Bordet and Gengou failed to give results. In his experiments with the blood of a re- covered rabbit he established a binding with 0.6 c.c. of the serum. Gomez notes, however, that large doses of antibody (1-2 c.c.) will, • From Jour. Infect. Dis., igix, 8, p. 330. 419 420 Contributions to Medical Science when mixed with complement, bind or destroy its action without the presence of antigen. With the idea of further testing this reaction, Dr. Ricketts and Mr. R. M. Wilder made several experiments, using the eggs of infected ticks as antigen. It has already been shown that the eggs i of infected ticks are rich in an organism which is supposedly the cause of Rocky Mountain spotted fever and it was supposed that these eggs would therefore afford a suitable source of antigen. In these experiments a goat-rabbit hemolytic serum was used and the exact dosage of fresh, normal guinea-pig complement was fixed for each experiment. In the first three experiments of the series eggs were used in quantities varying from i to loo, the antibody being varied from 0.05 to 0.1 c.c. As an example of these experiments we insert Table i. From these experiments it will be noted in Group A (Table i) that there is a shght decrease in hemolysis as the number of eggs is increased. In Group B with a larger amount of antibody the same result is noted, although the total amount of hemolysis is greater in each corresponding tube than in Group A. The degree of hemol- ysis is not dependent on the number of eggs, since Group C shows a constant amount of hemolysis in each tube. Evidently there was some reactivation^ of the heated serum by the unheated, and this reactivation was greater when the amount of heated complement in the antibody (guinea-pig serum) was greater. This was borne out by the results in tubes 17 and 18 in which a large amount of antibody increased the hemolytic action of o.oi c.c. of complement. In this experiment the total amount of complement used was not quite sufiicient to cause complete lysis, as shown in tube 13. In the second experiment of Ricketts' series larger numbers of eggs were used (75-100); the results were again inconclusive. It was found that the eggs alone, without the addition of antibody, were capable of preventing lysis. It was again noted that antibody and complement, when together, were more hemolytic than com- ■ That such an activation takes place was shown in an experiment in which we first established the minimum dosage of normal complement and then added varying amounts of antibody to a subminimal dose of complement. Doses of inactivated antibody from o.oi c.c. up caused complete laking in the tubes, although alone the antibody had absolutely no effect. Complement Deviation in Spotted Fever 421 plement alone, and it was assumed that the antibody contained hemolytic amboceptors which were activated by the complement. An effort was made in the third experiment to absorb the hemolytic amboceptor supposed to be present in the antibody, but the results TABLE I. CoMPi£MENT Deviation in Spotted Fever. Groups Group A I 2 3 4 Group B S 6 7 8 Group C 9 10 II 12 Group D 13 14 Group E IS 16 Group F 17 18 Group G 19 Sensitized Corpuscles Antibody Antigen (Eggs) 0.0s I 0.0s 10 0.0s 20 0.05 5° 0.1 1 0.1 10 0.1 20 0.1 SO I 10 20 50 0.0s 0.1 0.0s 0.1 Complement Results o-S o.s o-S o.s o.s 0.5 o.s o.s o.s o.s o.s o.s o.s o.s o.s o.s o.s o.s Unsensitized o.oi o.oi 0.01 O.OI O.OI O.OI O.OI O.OI O.OI O.OI Marked hemolysis (7s per cent) Marked hemolysis (less than tube i) Marked hemolysis (less than tube i) Moderate (less than 2 or 3) Complete Complete Complete Almost complete Moderate Moderate Moderate Moderate Marked hemolysis (75 per cent) Almost complete Complete o (solution clear) The following materials were used: Sensitized erythrocytes from goat, washed in 5 per cent suspension, i day old; complement — fresh, normal guinea-pig serum; antibody — inactivated serum of guinea-pig 2,537, recently recovered from spotted fever; antigen — «ggs of D. modestus (56 [57], 11, F. 4), 10 days old, rich in bacilli. The goat corpuscles were sensitized by adding o . 8 c.c. of immune rabbit serum to 20 c.c. of a s per cent suspension, incubating 20 minutes and washing once. This was done in order to avoid the necessity of making separate measurements of corpuscles and of the hemolytic serum. Antigen-|-antibody-|-complement was allowed to incubate 20 minutes (i hr. in Experiment 3) in a total volume of 0.3 c.c; 5 c.c. suspension of corpuscles was then added and the total volume raised to 2 c.c. by the addition of normal salt solution. The mixture thus prepared was incubated 2 hours and then placed on ice over night. were negative. In a fourth experiment dried serum of an immune horse was dissolved and used as an antibody, gi\dng a fairly strong deviation with o. 2 c.c, but it was found on further tests that 0.4 c.c. of this serum alone would bind all the complement. This was possibly due to the presence of an anti-complement in the serum of 422 Contributions to Medical Science the horse, it ha\dng been actively immunized by the injection of virulent guinea-pig blood. We have repeated these experiments with tick eggs, crushing a given number of them (loo) in sterile salt solution and then using decreasing dilutions of this emulsion. Three series of eggs were used : (i) eggs from infected ticks (showing large numbers of organisms); (2) eggs from normal ticks; and (3) eggs from ticks raised on im- mune pigs. The three series, however, showed complete lysis. It is true that no lysis was obtained in the 100 per cent emulsion, but it was found that tubes without antibodies would give the same result. In these experiments the antigen-antibody-complement mixture was permitted to incubate as long as 3-3^ hours. All the results so far obtained, while showing some deviation, were not as clear-cut as might be expected in a disease like spotted fever in which the serum of recovered animals possesses such a marked protective power.' We therefore undertook the testing of various tissue extracts and of sera taken from animals during fever, in the hope of finding an efficient antigen. The inactivated serum of recovered guinea-pigs served as antibody. An ox-rabbit hemo- lytic system was used, the corpuscles being first sensitized, and the exact dosage of complement established for each experiment. In the case of various bacterial suspensions used as antigens, a stand- ardized loop of a 24-hour agar culture was always made up in i c.c. of normal salt solution. In one experiment varying amounts of a watery extract of macerated lymph glands, spleen, and testicle of a guinea- pig which had died of spotted fever were used as antigen, with varying amounts of antibody (0.025 to o. i c.c), but there was no deviation.^ This negative result led us to investigate whether the antibody was thermostable or whether it had been destroyed by the heating at 56° C. We found that both fresh and heated immune sera protected equally well. Evidently the diflSiculty remained in the antigens. We therefore made separate extracts ' Ricketts and Gomez, Jour. Infect. Dis., 1908, 5, p. 221. See page 343 of this book. » It may be stated here that the most marked internal changes during a course of spotted fever in the guinea-pig occur in the lymph glands, testicle, and spleen. Large numbers of bodies similar in appear- ance to the bacillus found in the tick eggs can be seen in the spleen, lymph glands, and liver when sections are stained with Giemsa (Dr. Maria B. Maver). The organism must also be present, of course, in the blood, which is highly virulent, but it has not been seen equally well here in stained preparations. Complement Deviation in Spotted Fever 423 from the macerated material of infected pigs, using serum, lungs, liver, spleen and lymph glands, kidneys, muscles, brain, and intes- tine. Varying quantities were used (o.i c.c. to 0.5 c.c), with a constant amount of immune guinea-pig serum. The organ ex- tracts were not hemolytic. The results of the experiment were again negative. In connection with these experiments a series was made with a suspension of the culture of the "Spaniard" or- ganism^ as antigen and the serum of guinea-pigs recovered from infection with it and from Rocky Mountain spotted fever as anti- body. As will be seen from Table 2 complete binding of the comple- ment was obtained with both antibodies. While this was to be ex- ' The "Spaniard" organism, culturally, morphologically, and immunologically, seems to be practically identical with a stock culture of B. cholera suis obtained from the Department of Bacteriology of the University of Chicago. It crept into the spotted fever passage virus derived from the patient "Spaniard" and gradually replaced the spotted fever virus. Whether it actually outgrew and "smothered" the virus is not known. The "Spaniard" patient was suffering from the Idaho spotted fever — a strain which on three previous occasions it had been impossible to keep going by passage through the guinea-pig — so that the adventitious infection may not have been responsible for the "dying out" of the spotted fever virus. By proper attention to dosage, the infection produced in the guinea-pig by the intraperitoneal injection of the "Spaniard" organism can be made to parallel very closely in some respects a case of spotted fever. There is the same three days' incubation period, the subsequent high fever, and death in about lo days. Post mortem the enlarged, hemorrhagic lymph glands and swollen spleen are conspicuous. The gangrenous ears and scrotum, the cutaneous eruptions, and the hemorrhages in the region of the pampiniform plexus so common in spotted fever are wanting, while a fibrinous exudate in the peritoneal cavity, never found in the latter disease, is always present. In non-fatal cases the temperature of spotted fever pigs returns to normal in about lo days and there remains; in non-fatal infections with the "Spaniard" organism, a low fever may persist for 20 days to a month with occasional exacerbations. Cultures from all spotted fever cases remain sterile; the specific bacilli may always be cultivated from the heart's blood, peritoneal exudate, and spleen of pigs dying from infection with the "Spaniard" organism. Finally, recovery from spotted fever does not confer immunity to the "Spaniard" organism, though a high grade of specific im- munity is produced. Both cause a "hemorrhagic septicemia" but the "Spaniard" organism is much less prone to produce hemorrhages than is the spotted fever virus. It is a striking coincidence that in three diseases produced by micro-organisms which apparently can- not be cultivated, namely yellow fever. Rocky Mountain spotted fever, and hog cholera, and the first two of which at least may be transmitted by insects, cholera-suis-like organisms should have intruded themselves. This fact, coupled with those detailed in this paper on the comparative fixation tests, sug- gests that there may be something more concerned here than mere coincidence. There may be an actual relationship of some kind between the various causal organisms. With this idea in mind the following preliminary experiment with precipitins is of interest: The precipitogens consisted of filtered, 24-hour broth cultures of B. cholera suis, "Spaniard," Staph, aureus, and B. typhosus; the precipitins were contained in the unhealed sera of spotted fever immune guinea-pigs Nos. 2,658 and 2,381 and of "Spaniard" immune guinea-pigs Nos. 2,342 and 2,669. Precipi- tates formed in those tubes containing the cholera suis and "Spaniard" filtrates with the spotted fever and "Spaniard" immune sera alike in a serum dilution of 1-20 and not at all in the staphylococcus and typhoid tubes. Control tubes remained clear. '' Sample Protocols. Precipitogen Precipitin NaCl Sol. o.s c.c. or c.c. 1.49 c.c. o.s c.c. I c.c. 1.4 c.c. o.s c.c. i.S c.c. I c.c. 1.9 c.c. 424 Contributions to Medical Science pected in the "Spaniard-Spaniard" combination, the ''Spaniard"- spotted fever result was rather unexpected. While these controls had been intended principally as a check on our technic by the ad- dition of known factors, and showed that our method was correct, the result indicated either that the "Spaniard" culture was a specific antigen for Rocky Mountain spotted fever, or stood in a group rela- tionship to the same (we knew that the first supposition was not true) or that the guinea-pigs had at some previous time been infected with the "Spaniard" bacillus and were therefore immune. TABLE 2. CoMPLEstENT De\tation wtth "Spaniard" Bacillus and Spotted Fever and "Spaniard" Immune Serum. Antee oores Sensitized Corpuscles Antigen Culture "Span." VI Complement Sp. Fever "Spaniard" Results 2,517 2,343 o.S 0.025 0.05 0.01 No hemolysis 5 0.025 0.1 O.OI No hemolysis o S 0.02s 0.25 O.OI No hemolysis o S 0.025 o.S O.OI No hemolysis o S 0.05 0.2s 0.01 No hemolysis o 5 0.1 0.25 O.OI No hemolysis o S 0.0 0.5 O.OI Complete o 5 0.1 O.OI Complete o 5 0.02S 0.05 0.01 No hemolysis S 0.02S 0.1 O.OI No hemolysis o S 0.02S 0.2s O.OI No hemolysis o S 0.025 o.S O.OI No hemolysis o S 0.0s 0.2s O.OI No hemolysis O.S O.I 0.2S O.OI No hemolysis Tubes incubated for 30" and corpuscles then added and tubes made up to 2 c.c. with normal salt solution. Incubated i hr. 30" and then placed at room temperature over night. This experiment was repeated using the same antibodies and fresh suspensions of "Spaniard," typhoid bacilli, and Staphylococcus aureus as antigens. The results with the typhoid bacilli and staphylococci were absolutely negative but there was deviation as in the previous experiment with the "Spaniard" organism. The effect of large doses of antibody causing an increase of he- molysis, as noted in the Ricketts experiments already quoted, was observed. Certain preliminary experiments indicated that the bacillus of hog cholera, when used as an antigen with Rocky Mountain spotted fever and "Spaniard" immune sera as the antibodies, would give Complement Deviation in Spotted Fever 425 evidence of binding the complement. A quantitative experiment was performed in order to compare this binding with the results obtained using the "Spaniard" culture as antigen. These results are shown in Table 3. As will be seen in Series A (spotted fever- "Spaniard"), there is first a progressive decrease in the amount of TABLE 3. Groups Experi- ment 12 Sensitized Corpus- cles Antibodies Sp. Fever 2,381 "Spaniard' 2,342 Antigens Span. VI B. cholera suis Comple- ment Results Group A I 2 3 4 S 6 7 8 Group B 9 10 II 12 13 14 IS i5 Group C 17 18 19 20 21 22 23 24 Group D 2S 26 27 28 29 30 31 32 o-S o-S 0-5 o-S o-S o.S o.S o-S o.S o.S o.S o.S 0.5 o.S o.S 0.5 o.S o.S o.S o.S o.S 0.005 0.0075 o.oi 0.02 0.03 0.04 0.0s 0.1 0.005 0.007s 0.01 0.02 0.03 0.04 0.05 0.1 0.005 0.0075 0.01 0.02 0.03 0.04 0.05 0.1 0.005 0.0075 0.01 0.02 0.03 0.04 0.0s 0.1 0.0s 0.05 0.0s 0.05 0.05 0.05 0.05 0.05 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.05 0.0s 0.05 0.05 0.05 0.05 0.05 0.05 Incomplete laking Incomplete laking Trace of laking Trace of laking Trace of laking o Slight laking Moderate laking Incomplete laking Incomplete laking Incomplete laking Trace of laking Trace of laking Least laking Slightly more laking Slightly more laking Incomplete laking Slight laking Slight laking o Complete laking Complete laking Complete laking Complete laking Complete Complete Complete Complete Complete Complete Complete Complete laking laking laking laking laking laking laking laking laking reaching a maximum in tube 6 in which we find complete binding and there follows increased laking in tubes 7 and 8. Series B (spotted fever+hog cholera) shows a similar reaction, though the best tube still shows a trace of laking. In Series C ("Spaniard"+ "Spaniard") we find a marked deviation in the first four tubes (90 per cent to 100 per cent) and complete laking in the last four tubes. Series D, on the other hand ("Spamard"-hog cholera), gave com- plete laking. This experiment was repeated twice, using sera from different immune pigs, and similar results were obtained with the 426 Contributions to Medical Science exception that now the "Spamard"-hog cholera combination gave the reverse of the result in Experiment 4, in which we found a nega- tive reaction. Further experiments have confirmed these results, and also seem to show (i) that the serum of normal guinea-pigs does not contain the specific antibody; (2) that the spotted fever immune serum causes no deviation in the presence of typhoid, anthrax, dysentery (Shiga), and Friedlander bacilli, staphylococci, and cholera germs, although it has a specific afl^ty for certain strains of bacteria of the hog cholera group. The spotted fever an- tibody, in certain proportions, will bind complement with the hog cholera and "Spaniard" bacilH, and the "Spaniard" immune serum will bind with the same bacilli. We determined, however, that pigs immune to Rocky Mountain spotted fever are not protected against the "Spaniard" organism. We have failed to find a parallelism between the bacteriolytic power of these immune sera on hog cholera bacilH and the "Span- iard" bacillus, as determined by the Neisser-Wechsberg technic, and their power to deviate complement. That is, the bactericidal power of normal sera and of "Spaniard" immune sera was almost absolutely identical, while only the immune sera supplied the am- boceptor necessary for the deviation of the complement. Appar- ently the presence of substances in sera which cause a fixation or deviation of complement need not imply the presence of bacterio- lytic or protective antibodies. A somewhat similar phenomenon has been observed by Torrey and others.^ conclusions. Positive deviation of complement, using spotted fever antigen and antibody, has been obtained in but one instance, and the results of experiments undertaken to confirm this positive finding have been negative. The negative results are probably to be explained by the small quantities of organisms present in any one preparation of antigen and by the admixture of various substances in the antigens which in themselves absorb or neutralize complement, this in turn pro- ■ Torrey, Jour. Med. Res., 1910, 22, p. 95. Complement Deviation in Spotted Fever 427 hibiting the use of the large doses of antigen which might otherwise be effective. The protective power of spotted fever immune serum is not destroyed by heating to 56° C. for 30 minutes. An apparent reactivation of heated serum (antibody) was met with in many cases which, by causing hemolysis, tended to mask results. This might become a source of error in work of this kind. An organism probably identical with the hog cholera bacillus replaced the spotted fever virus in one of the passages. The serum of guinea-pigs immune to spotted fever and of guinea-pigs immune to this organism deviated the complement alike when a suspension of the above bacterium was used as antigen. This fact, together with other considerations mentioned in this paper, suggests that there may be a definite relationship of some kind between the virus of Rocky Mountain spotted fever and the hog cholera bacillus. Our results suggest, further, that the presence of substances in sera which cause a complement-fixation need not imply the presence of bactericidal or protective antibodies. TIME RELATIONSHIPS OF THE WOOD-TICK IN THE TRANSMISSION OF ROCKY MOUNTAIN SPOTTED FEVER/ JosiAH J. Moore. {From the Pathological Laboratory of the University oj Chicago.) The chief object of the following experiments, undertaken at the suggestion of the late Dr. Howard T. Ricketts, was to determine (i) the minimum duration of feeding by infected tick necessary to infect the guinea-pig; (2) the minimum duration required for the infection of the tick from the infected guinea-pig; and (3) the length of the incubation period in the tick, that is, the period of time required for the etiological factor to establish itself in the tick and render it infective. Duration of feeding by infected tick necessary to infect the guinea- pig. — All the ticks used in this experiment were bred in the labora- tory in tick-proof cages. They were of two species, Dermacentor venustus and Dermacentor modestus. The parents of the former were the second generation of that species raised in the laboratory, the original ancestors having been procured from the Bitter Root Valley in Montana in the spring of 1907. These had been infected in the usual manner with the Montana strain of fever and had trans- mitted this from generation to generation up to the present time. The parents of the latter species (Dermacentor modestus) were collected in Idaho in the spring of 1909 and these also were infected with the Montana strain. The method was as follows: In the group of experiments in Table i , the tick was first permitted to feed for five or ten days on a normal guinea-pig to test its infectivity. After making sure that it was infected, the tick was removed and kept at room temperature for a variable period ; then fed on a normal guinea-pig for a certain number of hours; again removed and kept at room temperature. The feedings were repeated at variable intervals on new guinea- pigs until the tick died. 'From Jour. Inject. Dis., 1911, 8, p. 339. 428 The Wood-Tick in Transmission of Spotted Fever 429 In the second group of experiments (Table 2) the ticks were not given a preliminary test, but the first feed was regarded as a time test; in all other respects the experiments were similar. When testing or feeding an individual tick, the guinea-pig was placed in a pillory and the tick allowed to attach itself to the ear. The time of feeding was taken from the minute the tick became so firmly attached that when the tick body was flipped back it could not be detached from the pig. In the beginning of the work it was thought that a feeding of one hour would be sufficient to infect a guinea-pig, but it was soon found necessary to extend this period to five hours. If the tick proved infective in the five-hour test the time could be reduced in the second test, or if the infection were not transmitted to the guinea-pig in the first test, the time could be lengthened. The results of the experiments as shown by the tabular summary of Tables i and 2 are not constant. However, when it is consid- ered that the tick is a living factor, this inconstancy is partially explained. The tick is a slow feeder. The hungriest and most vigorous require from 15 to 20 minutes to attach themselves firmly by means of the palpi, and without this firm attachment feeding is impossible. In other ticks the process is much slower, and after two or three days of attachment there is Httle evidence of the ingestion of blood. Table i shows that the minimum duration of tick feeding neces- sary to infect a guinea-pig was found to be one hour and forty- five minutes (Experiment I, with Dermacentor modestus. Female 90). In Experiment XV, D. modestus, Female no, was infected in a two-hours' feed and again in a three-hours' feed. A five-hours' feed inoculated in Experiments IV, V, XIII, XV, XVI, and XVIII; an eight-hours' in Experiment XII; a ten-hours' in Experiments I, XI, and XIV; a fifteen-hours' in Experiments XV (twice), XVIII, and XIX; a twenty-hours' in Experiments III, X, XIV, XVII, and XX. Twenty-seven five-hour tests were tried, and six, or 22 per cent, infected; ten ten-hour tests, of which three, or 30 per cent, infected; eight fifteen-hour tests, in which four, or 50 per cent, were positive; six twenty-hour tests, with five, or 83 per cent, positive; nine one- to three-hour tests and three, or 33 per cent, infected. To 430 Contributions to Medical Science TABLE I. DxjaATiON OF Tick-Feeding Required for the Infection of the Gcinea-Pig. Ticks previously tested and proved infective. Experiment Tick Species and Number End of Infectivity Test Interval Length of First Test and Date 1 Pi > C Length of Second Test and Date ■3 (2 1 V a I D. modestus F. 90 D. venustus M. 196 D. modestus M. 100 D. modestus F. 87 D. modestus F. 88 D. venustus F. 198 D. venustus F. 199 D. venustus F. 202 D. venustus F. 203 D. venustus M. 204 D. modestus F. 103 D. modestus F. 108 D. modestus F. 106 D. modestus F. 107 D. modestus F. no D. modestus F. los 9/1 S 9/1 S 10/5 9/21 9/20 10/2S 10/25 11/2 11/16 11/19 10/18 11/28 11/22 11/28 11/29 ri/22 3 days 7 days 17 days 43 days 44 days 9 days 9-10 days 10 days 14 days 13 days 37 days 10 days 25 days 19 days 21 days 31 days 18 hr. 9/18 1 hr. 9/22 10 hr. 10/22 Shr. 11/3 shr. 11/3 Shr. 11/4 Shr. 11/4 shr. 11/12 shr. 11/30 , shr. 12/2 shr. 12/8 shr. 12/8 shr. 12/17 IS hr. 12/17 IS hr. 12/20 3-S hr. 12/23 + + + + + 4 days 4 days 4 days 8 days 8 days 26 days 18 days 20 days 14 days 21 days 17 days tick dead 12/12 23 days 20 days 21 days 15 days 10 br. 9/22 Shr. 9/25 shr. 10/26 2jhr. ii/ii 1 hr. 11/11 Shr. 1 2/1 shr. ir/22 shr. 12/2 IS hr. 12/14 20 hr. 12/23 10 hr. 12/19 8hr. 1/9 20 hr. disc ed i/io io(?) hr.di ded i/ii Shr. 1/7 + + + ard- scar- 16 days n tick dead rrr 9/27 13 days IV 19 days V 19 days VI 12 days vn 20 days yrrr tick dead rx 12/S tick dead X 12/26 ; IS days XI tick dead xn 2/23 xm 19 days XIV 16 days XV 20 days XVI 23 days TABLE 2. Duration of Tick-Feeding Required for the Infection of the Guinea-Pig. Ticks not previously tested for infectivity. Experiment Tick Species and Number Date of Moulting Interval Duration of First Test and Date Result Interval xvn xvm D. modestus F. 128 D. venustus M. 222 D. modestus M. 129 D. modestus M. 130 Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Shr. 1/17 iShr. 1/18 shr. 1/19, shr. 1/19 - 18 days 10 days XIX 21 days XX 23 days The Wood-Tick in Transmission of Spotted Fever 431 TABLE I. Duration of Tick-Feeding Required for the Infection of the Guinea-Pig. Ticks previously tested and proved infective. ^ Length Length Length Length of of of Third Fourth ■3 Fifth Sixth •5 oQ p, *> "^ Test and •3 t Test 3 > Test 3 > Test •g r, 3 Date S 4> and 8 and and ? 2 ^ ►s Date P< a Date ^ A Date P< 5 ^a>ca ^ I hr. 45 m. + 11 days 6 days + tick dead 10/9 10/19- 10/25 10/30 10 hr. _ 22 days II days + 28 days 20 hr. + 10 days 10 hr. _ 11/9 12/1- 12/12 1/9 10/19 2ihr. — tick dead "/30 1 2/5 2jhr. — tick dead "/30 12/s 10 hr. — tick dead 12/13 12/16 10 hr. — tick dead 12/13 12/18 Shr. 20 days 20 hr. + II days 10 hr. 14 days 10 hr. 1/7 1/27 2/7 2/21 Shr. li days Shr. tick dead 1/28 2/9 2/20 20 hr. + II days 10 hr. + tick dead 1/27 2/7 2/1S IS hr. + 8 days Shr. + 14 days 3hr. + 9 days 2 hr. + 3 day s I hr. — 2/1 2/9 2/23 3/4 3/7 Shr. + tick dead 2/1 2/10 TABLE 2. Duration of Tick-Feeding Required for the Infection of the Guinea-Pig. Ticks not previously tested for infectivity. Duration of Second Test and Date Result Interval Duration of Third Test and Date Result Interval Duration of Fourth Test and Date Result Shr. ^/'* I. IS hr. 1/28 IS hr. 2/9 + + + 12 days 15 days 16 days tick dead 15 hr. 2/16 5 hr. 2/1 1 10 (?)hr. 2/25 + 20 days 12 days tick lost 20 hr. 3hr. 2/23 + 2/11 432 Contributions to Medical Science account for the higher percentage of infections in the one- to three- hour than in the five- and ten-hour tests we find that in the former cases two highly infected and voracious ticks caused the three inoculations, while in the latter the ticks were apparently not so heavily infected, and were only average feeders. From the data given it is observed that the average time for infecting the guinea-pig by the wood-tick is about ten hours, and that twenty hours is almost surely infective. Only once did this fail, but, on repeating (Experiment X), a positive result was obtained. D. modestus. Female 90 and Female no, gave the best results, infecting the animals upon which they were tested at practically each feed. In Experiment XIII the first five-hours' feed was positive, while two others repeated later at varying intervals did not cause infection. The results obtained in this experiment, in Experiment XVI, and also in Experiment XVII, where, after a negative fifteen-hours' test, there followed positive twenty- and five- hour tests, may have been due to disincHnation of the tick to feed vigorously. In Experiments I and III the weakened condition of the ticks necessitated long feeds to restore their vitaHty. From the tabulated summary it appears that both males and females of the modestus species gave more positive results than those of the venustus. This was probably due to the fact that the former were younger and more vigorous and therefore fed more rapidly. Duratioft of feeding required for the infection of the tick and length of incubation period in the tick. — The technic in this series of experi- ments was as follows: A number of normal ticks were allowed to feed on an infected guinea-pig for a definite number of hours; they were then removed and kept at room temperature for several days, when feeding was again permitted, this time, however, on a normal guinea-pig, for a definite period. Both species, D. venustus and D. modestus, were tested. The infected guinea-pigs on which the ticks were fed were regular passage animals, and were used between their second and fifth days of fever. In only two of the 27 was the spotted fever virus demonstrated in the tick, so that the results cannot be considered conclusive. It seems highly probable that a much shorter feed is required for The Wood-Tick in Transmission of Spotted Fever 433 TABLE 3. Duration of Feeding Necessary to Infect Tick and Incctbation Period in the Tick . Exp. Tick Species Date of Length of Feeding Inter- Date Length Re- Inter- Date of Second Test Length Re- and Number Feeding val of Test of Test sult val of Test sult I D. venustus M. 204 iVs S hrs. 7 days 12/12 10 hrs. — M. 205 12/5 5 hrs. 14 days 12/19 10 hrs. — F. 206 12/s 5 hrs. tick died before b eing test ed M. 207 12/3 10 hrs. 7 days 12/12 5 hrs. — M. 208 12/S 10 hrs. 13 days 12/18 20 hrs. — F. 209 12/s 10 hrs. 39 days 1/13 IS hrs. — M. 210 12/S-6 20 hrs. 6 days 12/12 20 hrs. — 29 days 2/2 IS hrs. — M. 211 12/S-6 20 hrs. 13 days 12/18 20 hrs. — F. 212 12/5-6 20 hrs. 39 days 1/13 15 hrs. — II D. venustus M. 214 1/9 10 hrs. tick died before b eing test ed M. 215 1/9 10 hrs. 4j days 1/14 20 hrs. — M. 216 1/9 10 hrs. 25 days 2/4 20 hrs. — F. 217 1/9-IO 20 hrs. 4 days 1/14 20 hrs. — F. 218 1/9-10 20 hrs. 23 days 1/31 15 hrs. — M. 219 1/9-10 20 hrs. 41 days 2/18 15 hrs. — F. 220 1/9-10 25 hrs. 4 days 1/14 20 hrs. — 40 days 2/18-3/4 14 days + M. 221 1/9-10 25 hrs. 24 days 2/2 IS hrs. — Ill D. venustus M. 22s 1/24 25 hrs. 37 days 3/2 20 hrs. — M. 226 1/24 25 hrs. tick died before b eing test ed IV D. modestus M. 168 3/1-2 25 hrs. 20 days 3/22 20 hrs. — M. 173 3/1-2 40 hrs. 10 days 3/12 20 hrs. — M. 174 3/1-2 40 hrs. 20 days 3/22 20 hrs. — V D. venustus F. 232 3/24-25 25 hrs. 35 days 4/28 25 hrs. — VI D. modestus F. 178 3/24-30 25 hrs. 30 days 4/28 30 hrs. + F. 179 3/29-30 25 hrs. 38 days 5/6 25 hrs. infection and that the incubation period in the tick is shorter than is indicated by these experiments. The complete protocols of the two tests in which complete results, as indicated by the trans- mission of the disease by the tick to the normal guinea-pig, were obtained, follow. D. venustus, Female 220, was attached to infected guinea-pig 2,547 for 25 hours on January 9-10. No hyperemic area was observed around the bite and no feces were passed. On January 14, after an interval of four days, it attached itself to normal guinea-pig 2,713 and was removed after 20 hours. The tick did not appear to feed much, not increasing any in size, and passing no feces. The temperature of guinea-pig 2,713 was taken for two weeks but did not rise above 103.4° F. A month later, Feb- ruary 27, the pig was given an immunity test of i c.c. of blood from an infected pig. It became infected and died, the organs showing changes t>T)ical of spotted fever. This proved conclusively that the guinea-pig had not been infected by the tick, since Ricketts and Gomez" have shown that one attack of spotted fever renders the guinea- pig immune to a second inoculation. On February 18 Female 220 attached itself to normal guinea-pig 2,779. It was found ofif March 5, having been attached for 15 days. It had increased markedly in size during the feeding. A hyperemic area about 2 cm. in diameter was observed around the bite. *Jour. Infect. Dis., 1908, s, p. 221. 434 Contributions to Medical Science TEMPERATURE OF GUINEA-PIG 2,779- February 18 103. 6° F. February 25 102. 6° F. March 4 105. 2* F. February 19 104° F. February 26 102 . 4° F. March s 105° F. February 20 103. 6° F. February 27 March 6 F. February 21 103. 8° F. February 28 102. 6° F. March 7 ios.6°F. February 22 103. 6°F. March i 102. 4°F. March 8 105. 2°F. February 23 102 . 9° F. March 2 March 9 104° F. February 24 102. 6°F. March 3 105° F. March 10 ioi.2"'F.* March 11 Dead. • Blood taken from heart. Autopsy. — Axillary glands enlarged and congested. Spleen twice the normal size and of a deep, bluish-red color. The heart, lungs, and intestines normal. The suprarenal glands and liver slightly enlarged and congested. The external genitals swollen and the skin slightly hemorrhagic. Diagnosis. — Spotted Fever. Blood drawn from the heart on March 10 was in- oculated into guinea-pig 2,831, which ran a typical spotted-fever temperature. The scrotum became swollen, the ears became necrotic and finally sloughed off. The guinea-pig recovered and later was given an immunity test during which its temperature remained normal. All these facts prove beyond doubt that guinea-pig 2,779 had spotted fever. In most instances it has been found that an infected tick will produce the fever in a guinea-pig after an incubation period in the pig of five or six days. In the present case this is doubled. Two assumptions may explain the results: (i) either the tick did not feed for several days after becoming attached to guinea-pig 2,779; or (2) the amount of virus ingested by the tick in its feed upon in- fected guinea-pig 2,547 may have been exceedingly small, since the tick did not feed actively, as is shown by the absence of a hyperemic area around the bite, and this small quantity of virus may not have increased in the body of the tick sufficiently to infect a guinea-pig until the tick had enjoyed for some days what we may regard as a very favorable condition of attachment to a guinea-pig. On March 29 D. modestus, Female 178, attached itself to the ear of infected guinea-pig 2,771, where it was allowed to remain for 25 hours. Considerable feces were passed by the tick, and a slight hemorrhagic area surrounded the bite. After an intermission of 30 days, during which the tick was transferred from Chicago to Montana, it attached itself on April 28 to normal guinea-pig 3,004, being removed after 30 hours. On May 5 the temperature of the guinea-pig rose to 104° F. TEMPERATURE OF GUINEA-PIG 3,004. April 28 103° F. May 5 104° F. April 29 102. 6° F. May 6 105. 1'F. April 30 102. 8° F. May 7 106° F. May 1 103° F. May 8 106. 2° F. May 3 103 . 2° F. May 9 105° F. May 3 103. 1'F. May 10 104.1° F. scrotum swollen. May 4 103 . 6° F. May 11 103° F. scrotum hemorrhagic. May 12 103 . 4° F. Recovered. The Wood-Tick in Transmission of Spotted Fever 435 An immunity test was given May 16, but no fever resulted, showing that guinea- pig 3,004 had been infected the first time. The incubation period in the tick in this case was 30 days with a minimum duration for infecting the tick of 25 hours. With D. modestus. Female 179, which was attached the same length of time as Female 178 on guinea-pig 2,771, but was given a 25-hour test after an interval of 38 days, a nega- tive result was obtained. In previous experiments the duration of feeding required for the infection of the tick was shown to be less than was found here, 20 hours being the shortest period.^ The incubation period is, perhaps, "represented only by the time required for the distribution of the virulent organisms throughout the tick's body and eventu- ally into its salivary glands. In a number of instances the tick proved virulent immediately following its removal from the in- fected animal" (Ricketts).^ In the latter cases, however, the tick was permitted to feed for several days on the infected guinea-pig and thus the incubation period may have been reduced by the tick's remaining attached to the guinea-pig, or it may be that the chances that the tick will infect the susceptible guinea-pig are greatest immediately after its proboscis has been contaminated by feeding on an infected animal. More experiments will have to be performed before any definite statements can be made relative to the length of the incubation period in the tick. The great number of negative results may have been due to the fact that the ticks were raised under artificial con- ditions and did not feed as rapidly as those obtained from nature. It will probably be found that ticks taken directly from the woods will give better results. Similar experiments with fresh ticks were to be undertaken in the field during the spring of 1910, but the unfortunate death of Dr. Ricketts prevented the carrying-out of the work. It will, however, probably be completed at some future date. SUMMARY. The minimum duration of feeding necessary for a tick to infect a guinea-pig was found to be one hour and forty-five minutes. The average time necessary seems to be about ten hours, while twenty 'Ricketts, H. T., Jour. Am. Med. Assn., 1907, 49, p. 24. 'Med. Record, 1909, 76, p. 843; page 383 of this book. 436 Contributions to Medical Science hours were almost constantly infective. The duration of feeding necessary to infect a tick is approximately twenty-five hours, while the minimum incubation period in the tick was not definitely determined. With ticks obtained from nature it is possible that the duration of feeding necessary to infect the tick and the incu- bation period in the tick will be found to be much less than is indi- cated here. TRANSMISSION OF SPOTTED FEVER BY THE TICK IN NATURE.^' ^ Maria B. Maver. (From the Pathological Laboratory of the University of Chicago.) The discovery of ticks in nature carrying the virus of spotted fever is published by Dr. Ricketts^ (page 312), who points out the importance of this fact in connection with the theory of transmission of spotted fever by the tick. The following experiments were made in order to throw further light on the natural infectivity of the tick. Six hundred and fifty-six ticks were collected during the spring months of 1909: 254 of the species D. modes tus and 402 D. venus- tus, for the purpose of studying the infectivity of the tick in nature. These ticks were found on cows, bushes, and vegetation in the Lo Lo Valley and Owl Canon, Montana, and in the vicinity of Poca- tello, Idaho. Fourteen groups were arranged, consisting of 25 male and 25 female ticks each. Each group was placed in a new tick-proof sack, with a selected guinea-pig of medium size; nine sacks of Mon- tana ticks (D. venustus) and five sacks of Idaho ticks (D. modestus) were arranged in this way. The daily temperature of each pig was taken and the number of ticks attached noted. In Groups IV and V the guinea-pig became infected. The ticks in both groups were gathered in the Lo Lo Valley from cows. Experiment IV. — Normal guinea-pig 2,155 was the host of Group IV. For two weeks this pig kept up a vigorous scratching. The ticks were gathered daily and placed among the hair. Many were attached on the fourth day. On the sixth day two engorged females were found. On the 14th day the i&rst indication of a disease was noted in a temperature of 104°, followed on the next two days by a rise to 106°, with hemorrhagic vulva; i| c.c. of blood was with- ' From Jour. Infect. Dis., 1911, 8, p. 322. 'This work was supported by a grant from the Memorial Institute for Infectious Diseases, Chicago. J Jour. Am. Med. Assn., 1907, 49, p. 24. 437 438 Contributions to Medical Science drawn from the heart, and the pig died two days later, with lesions characteristic of spotted fever. The heart's blood taken from the pig was immediately injected into normal guinea-pig 2,203, producing a typical fever curve with extensive vulvar changes, and death on the 13th day after injection. This pig was also bled on the fourth day of fever and the blood passed into two normal animals. Of these guinea-pigs, one died of the disease and the other survived and proved immune to i c.c. of the virus carried in the "Bradley" strain. It was not possible to determine how many of these ticks transmitted the infection. Experiment V. — This group of ticks was also gathered in the Lo Lo Valley from cows. These ticks immediately attacked nor- mal guinea-pig 2,162; 30 males and females were found attached to the abdomen inside of 72 hours. On the fourth and succeeding days the temperature range was 104.8°, 105.6°, 105.4°, 97°, death resulting on the eighth day. No external lesions were observed. The spleen and inguinal glands were greatly enlarged. Blood was transferred to normal guinea-pigs 2,181 and 2,185 ^^^ both died of spotted fever. Blood from guinea-pig 2,185 was injected into two normal pigs. One of these pigs died of spotted fever, the other survived a typical and severe attack of the disease. A subsequent immunity test with "Bradley" passage blood proved that this pig had become immune. A "passage strain," using two pigs for each passage, was started with virus drawn from this guinea-pig at the height of the disease. Fifteen pigs in the first eight passages died with lesions typical of spotted fever. A survivor of the ninth passage proved immune to a known active virus. During the following months 49 passages of this strain were made. Of the 98 guinea-pigs infected in this manner, in 49 the disease was allowed to run its course without interference; of this latter group, 8 pigs recovered from the disease, and were thereafter immune to the active virus which had been obtained originally from a patient sick with spotted fever. In more than one-half of all these pigs hemor- rhagic lesions were noted. To render this series of experiments somewhat more complete, normal Montana ticks 190, 191, 192, 193, and 195 were allowed to feed on guinea-pigs infected with this so-called "natural tick Transmission of Spotted Fever by the Tick in Nature 439 strain." Ticks 191 and 192 became infected and transferred the disease to normal guinea-pigs 2,394 and 2,398, thus completing a tick cycle of this strain. Five experiments were attempted with 254 ticks gathered in the vicinity of Pocatello, Idaho. Groups of 50 ticks were associ- ated with normal guinea-pigs. In each case a short period of low temperature was noted. In Experiments III and IV passage was made to normal guinea-pigs on the third day of the fever. No disease resulted and these pigs were not immune to the "Bradley" strain. In the remaining three experiments of the group no conclusive evidence of a specific infection by the Idaho ticks was obtained. TRANSMISSION OF SPOTTED FEVER BY OTHER THAN MONTANA AND IDAHO TICKS.'-^ Maria B. Maver. (From the Pathological Laboratory of the University of Chicago.) Experiments having shown that the mild Idaho type of spotted fever may be transmitted by D. venustus of Montana and that the more virulent northern type of fever may be carried by D. modestus of Idaho, it became a matter of interest to determine whether ticks of other species, indigenous in widely separated areas, might also be capable of acting as agents of transfer of the spotted fever virus. In order to carry out experiments on this point arrangements were made with several entomologists for supplies of wood ticks from their localities. Ticks were received from Maine, California, Texas, Utah, and Missouri. The study of these ticks as intermediate hosts for the parasite of spotted fever was taken up according to the same plan as that used in studying the transmission of the virus by the Idaho and Montana ticks. Two strains of the disease were maintained in guinea-pigs at the Chicago laboratory at this time — the "Bradley" strain, which had arrived at the 179th consecutive passage from the human virus, and a so-called "natural tick strain," also in guinea-pigs in its 8th passage, produced by the bite of D. venustus (Montana) from nature. Both strains were used in these experiments. I . Transmission by Dermacentor variabilis (Say) . — Specimens of this tick have been found from Labrador to Florida. West of the Mississippi it is not common. It has been found on a great va- riety of small animals, more frequently on cattle and dogs, and, occasionally, on man. During the month of July, 1909, two collections of tliis species were received from Dr. Ricketts, having been collected by him in the vicinity of Woods Hole, Mass. Fifty adult ticks, the females in various stages of engorgement, comprised these groups. 'From Jour. Infect. Dis., igil, 8, p. 327. ' This work was supported by a grant from the Memorial Institute for Infectious Diseases, Chicago. 440 Transmission of Spotted Fever 441 Guinea-pig 2,266, 4th passage "natural tick strain," after an incubation period of four days, was placed in the stock, with a temperature of 104° F. On the fifth day after inoculation, one male — No. i — and one female — No. 5 — tick were attached to this guinea-pig for 48 hours during the period of fever. This animal died on the eighth day of the disease, the autopsy showing specific spotted fever lesions with the hemorrhagic vulva. The impregnated female tick became much enlarged and dropped off semi-mature after two days' feeding. Any further test of this tick was impossible owing to the fact that semi-mature females rarely become reattached. The female was placed in a pill box for oviposition and a later test of her larvae for inherited infection was made. The male tick which had fed on guinea-pig 2,266 was placed on normal guinea-pig 2,284 and remained attached for 72 hours; the temperature of this guinea-pig remained normal during a period of 18 days following attachment. The tick was found off in the sack, reattached, and removed four days later, in all a period of seven days' attachment. An immunity test given this pig with I c.c. of "Bradley" virus developed a characteristic case of the disease. This male tick did not transmit the disease. Guinea-pig 1,840 of the 197th "Bradley" strain passage was taken on the third day of fever with a temperature of 105.6° F. Two female ticks — Nos. 8 and 9 — attached themselves behind the ears of the pig after a short period. This guinea-pig died four days later with the engorged ticks still attached. The female tick 8 was immediately transferred to normal guinea- pig 2,321, and remained in the same location on this pig for six days. No disease developed in this guinea-pig, and on being given I c.c. of "Bradley" virus it died of spotted fever in eight days. For further maturation and impregnation of female No. 8, male No. I of this group was placed with her. Oviposition began eight days later. Examinations of the stained specimens of these fresh eggs were made for bacilli, but no organisms were found. Exami- nations of eggs from normally developed ticks were also negative. Fifty fresh eggs from female No. 8 were crushed in physiological salt solution, and injected into a guinea-pig intraperitoneally, but with no production of symptoms. About 600 larvae were hatched 442 Contributions to Medical Science from the eggs. These larvae matured on a normal guinea-pig without infecting it, 26 enlarged larvae being collected from this pig. After their normal period of moulting, the nymphs were again tested on a normal pig. In no period of their development did they transfer the disease. Seven of the eleven females matured and deposited eggs, hatching large numbers of larvae. From female 5 about 300 larvae were placed with infected guinea-pig 2,375 of the 189th "Bradley" passage; 35 larvae fed on this pig. Five infected guinea-pigs were placed in this sack and died of the disease during the feeding of these larvae. Fourteen nymphs and 19 enlarged larvae were collected. Normal guinea- pig 2,536 was infected by this batch of nymphs. The temperature of the pig ran 103°, 105.6°, 105.6°, 104.5°, and it died on the fifth day. On the third day 3 c.c. of its blood was withdrawn and inoc- ulated into a normal and an immune guinea-pig of the natural tick strain. The immune pig did not respond to the virus. The normal died on the sixth day of fever with specific spotted fever lesions. Again blood was transferred from this guinea-pig to immune and normal pigs with the same result; the normal guinea-pig died on the eighth day, the immune showed no indication of disease. The larvae from five females were associated with infected guinea-pigs 2,371, 2,436, and 2,482 and 2,311 of the "Bradley" strain. These pigs died of the disease, one after the other, while the larvae were feeding. A second test of these nymphs was made on normal guinea-pig 2,563. Two of the above larvae fed on this pig. On the seventh day after attachment a fever temperature began. The pig died on the eighth day. Two passages of blood were made to normal pigs and to immune pigs. The normal pigs developed spotted fever; the immune pigs were unaffected. Two adults ticks from this group produced infection in normal guinea-pig 2,873. This pig died on the eighth day after tick at- tachment; transfer of blood to normal pig 2,915 produced a less severe course of fever, which conferred immunity to the "Bradley" virus. 2. Dermacentor marginatus Utah (Banks). — Note received with these specimens: Dallas, Sub. ace. No. 654, seed ticks hatched from eggs deposited August 14, and days following. Hatching began Transmission of Spotted Fever 443 September 3, adult females taken from jack rabbit, August 9, 1909, at Milford, Utah, by W. V. King. The larvae were placed with normal and infected guinea-pigs in equal numbers. Eight nymphs matured on the normal animal. As a routine this pig was given an immunity test with "Bradley" blood. This resulted in the usual course of fever. Five infected guinea-pigs were used to mature and infect the other half of these larvae. After a period of moulting four of these nymphs were allowed to feed on normal guinea-pig 2,567. On the seventh day after attachment a fever began, resulting in death on the fifth day. Passage of the heart's blood of this pig into an immune and a normal pig gave no results in the immune animal but produced typical spotted fever in the normal. A second passage was made with the same result. 3. Amblyomma Americanum Linnaeus. — This species is com- monly known as the "lone star tick" because of the single yellow spot on the shield of the female. These specimens were received from Missouri in all stages of adult engorgement. The many attempts made to attach these ticks to guinea-pigs were unsuccess- ful. Nine of the number were selected for oviposition, which was evidently premature in every case. The ticks were small and com- paratively few eggs were deposited. Hatching took place from the eggs of females i, 2, 8, and 9, with about 1,200 larvae in all. Two groups were made of these larvae, about 600 in each, and placed with normal and infected guinea-pigs. From two pigs, 2,410, 2,426, infected with the "natural tick strain" 100 larvae fed and were collected later from the debris in the sack. These enlarged larvae developed in two weeks into nymphs, passing through a moulting stage. These nymphs were associated with normal guinea-pig 2,544 for a period of two weeks and eleven attachments noted before fever temperature with the characteristic hemorrhagic scrotum developed. The pig was bled from the heart and died on the fifth day. Two passages of blood from this infected pig to normal and immune pigs were made, both resulting as before in infection of the normals and no disease in the immunes. From the 14 infected nymphs, four females and two males were attached to normal guinea-pig 2,874. Three days 444 Contributions to Medical Science later the disease was indicated in this pig by a temperature of 104°, reaching 106.8°, and the development of hemorrhagic genitaHa. Death resulted after a sHght decline in temperature on 13th day. The ticks were removed after the death of the pig and immediately transferred to another pig. This pig died on the tenth day, after a severe run of fever, with necrosis of the ears and scrotum. The autopsy in each case showed typical spotted fever lesions. 4 and 5. — A large number of the larvae of the species D. albipictus were dead when they reached the laboratory. About 50 of the Kving ones were placed with infected guinea-pig 2,391. Only one of these was found mature and died before further test could be made. The larvae of the D. variabilis (Utah) were in poor condition when received. An effort made to keep them alive on a normal pig failed. 6. Ornithodoros megnini Duges (CaHfornia) was a very interest- ing specimen. Much time was given to the work of attaching these ticks to rabbits, without success. After a latent period of five months, oviposition took place, and a great number of larvae hatched out. When placed with an infected guinea-pig they im- mediately filled the ears and hung in bunches about the face of the pig. The pig died in this condition. The ears were cut off in the hope that these larvae would detach themselves later. This did not prove the case, and the whole mass died attached. SUMMARY. Ticks from six different localities were received, and of these three groups developed so that complete tests could be made, and by means of each of these three, spotted fever was transmitted from infected to normal guinea-pigs. In the case of Dermacentor marginatus (Utah) and Amblyomma Americanum Linnaeus (Missouri) the transfer was made with nymphs ; no adult ticks were tested. In the case of Dermacentor variabilis (Mass.) transmis- sion of the virus was effected by nymphs and later by adult ticks. From these experiments it appears that in so far as abiHty of ticks from various regions to transmit the virus of spotted fever is concerned, the disease might find favorable conditions for its existence in localities other than those to which it now is Hmited. A CONTRIBUTION TO THE PATHOLOGICAL ANATOMY OF ROCKY MOUNTAIN SPOTTED FEVER.'- ^ E. R. Le Count. (.From the Pathological Laboratory of Rusk Medical College, Chicago.) Six postmortem examinations were made by Dr. Ricketts on the bodies of persons dying of the disease. The cutaneous hemor- rhages were general in all and varied from petechiae no more than pin-head sized to large confluent and grotesquely shaped, bluish- black hemorrhages covering all parts of the body except the palms and soles. In a few of the bodies hemorrhages were observed in the pleurae or omentum, in one instance in the substance of the lungs. The superficial and visceral lymph glands were moderately enlarged. An icterus was always present. Enlargement of the spleen is quite characteristic of the disease. It was from two to three times the normal size in all the bodies examined. In other respects it was but little altered, fairly firm, and moderately hy- peremic. A large infarct was present in one instance. The large veins and right chamber of the heart were distended with blood. The liver and kidneys were enlarged and hyperemic and in both organs varying degrees of cloudy swelling, and fatty degeneration were met with. The changes in the other organs were neither marked nor constant enough to deserve mention, with the excep- tion of the bone marrow. In two bodies where this was examined the marrow of the long bones was red to bluish red. In monkeys and guinea-pigs the changes are like those in man with the exception of the absence of generaHzed cutaneous hemor- rhages; in white-skinned guinea-pigs these are sometimes seen. Localized hemorrhages with necrosis of the scrotum, vulva, or ' From Jour. Infect. Dis., igii, 8, p. 421. ' Several years before the death of Dr. Ricketts he and I planned a joint work on Rocky Mountain spotted fever to which he was to contribute accounts of its etiology, clinical aspects, and gross morbid anatomy, and I a study of the minute changes. The plan dealt with a monographic consideration of the disease. Some of the phases of this work were to be reported by me at the last meeting of the Association of Pathologists and Bacteriologists in Washington, May, 1910, and for this purpose he sent me notes of the gross changes he had observed in human and animal bodies. No report was made, news of his death arriving during the meeting. His observations are included in the following account. 445 446 Contributions to Medical Science prepuce are a more constant result of the disease in these animals; occasionally the ears become necrotic.^ Hemorrhages are also met with in the lymph glands in guinea-pigs. In other respects the lesions of both guinea-pigs and monkeys are quite like those in man. It will not be necessary except in a few instances to compare the microscopic changes in human tissues with those in animals; they may be considered as a whole. The animal tissues examined were from six monkeys and 32 guinea-pigs; the human tissues from the six bodies mentioned. The changes are of two sorts, those connected with the occlusion of vessels and the more diffuse lesions affecting entire groups of organs. The diffuse changes are hyperplasia of lymphoid tissues and cloudy swelling and acute fatty changes in organs commonly the seat of such lesions in acute infectious diseases. The focal lesions are more varied in their nature since they include not only the pro- cesses leading up to the occlusion of vessels, but the results of such obstructions, necrosis in different degrees and the hemorrhages re- sponsible for so many of the clinical and gross anatomic features of the disease as well as for the name "spotted fever." In sections of the skin, liver, kidney, spleen, and adrenal both vascular occlu- sion and the necroses resulting from obstruction were present. In the lung and heart the capillaries and small veins were found practically occluded with leukocytes, but there were no serious consequences of these conditions with exception of minute hem- orrhages beneath the endocardium. In the least marked lesions of this sort the vessels contain simply an excess of small and large lymphocytes and some of the latter cells have as inclusions red cells or other leukocytes. In the more obvious occlusions agglutinated red cells, fibrin, and polymor- phonuclear leukocytes are also present. The agglutinated red cells often are stained a deeper red with eosin than adjacent single red cells. Fibrin is not demonstrable in all the sections of each occluded vessel, but is intermingled here and there with the other constituents. In the human spleen, for example, and in the monkey ■ In the spotted fever of Idaho gangrene occurs rather frequently in the h»iman fauces, tonsils, and palate, also in the skin of the penis and scrotum; such changes have not been met with in Montana. Pathological Anatomy of Spotted Fever 447 spleen, small sinuses occur crowded to distension with polymorpho- nuclear leukocytes, and these obstructed vessels form the centers for minute necroses scattered throughout the spleen so abundantly that in each field of lenses of an amplification of 100 diameters or thereabouts, at least one such lesion is present. The necroses re- ferred to of the ears and skin of the scrotum in guinea-pigs are due apparently to anaemia from the plugging of small blood vessels. In serial sections of ears, the seat of such necrosis, and made from base to the peripheries, the entire gamut of early and late changes which terminate as necrosis are present as well as concomitant results of the altered circulation — edema, hemorrhage, and depo- sitions of blood pigment. In the propagation of the disease the blood serum of an infected animal was sometimes injected into the fresh animal, but the lesions described have nothing to do with the use of ahen or homologous serum in the manner indicated, for they are present in the human tissues and organs as well as in those of the animals infected experimentally by tick bites. A possible exception is the presence of rather compact masses of polymor- phonuclear leukocytes in the perivascular sheaths of lymphoid tissue in the spleens of monkeys, for in this animal the disease was brought about by the inoculation of blood from an infected guinea- pig or monkey ; the disease was not produced by tick bites in any of the monkeys. The excess of leukocytes in the vessels, which is such a conspic- uous feature in different organs, is in marked contrast to the esti- mations of the leukocytes made during Hfe; in the sections they seem too numerous to be consistent with estimates of 10 to 14 thousand per cubic millimeter.^ But we are acquainted with a similar condition in typhoid fever and in acute interstitial nephritis^ — that is, an apparent disparity between the leukocyte content of the blood during life and the great number in the vessels of certain localities or organs. It would seem, therefore, as though the action of the toxin of this disease was manifest in certain places, •John A. Anderson {Hygienic Laboratory Bulletin No. 14, igo3, p. 22) noted an increase in the large mononuclears in the blood of two patients as the most interesting change from the normal. ' Mallory, "The Histology of Typhoid Fever," Jour. Exp. Med., 1898, 3, p. 661; Councilman, "Acute Interstitial Nephritis," ibid., p. 393; Le Count and Batty, "Purpura Hemorrhagica with General- ized Infection with Bac. paratyphosus," Jour. Infect. Dis., 1907, 4, p. 175. 448 Contributions to Medical Science and at such points, even though the vessel occlusions were not present, there was likely to be an excess of leukocytes in the blood, mainly of the mononuclear type. No extensive search for the bacillus described by Dr. Ricketts* (see page 368) was made. In the examination of the focal lesions it was difficult to dismiss the idea that the cause of the disease was in all likelihood within the field of vision. This was naturally more true of the minuter lesions. The minuteness of the bacillus, and the shrinkage which is probably greater in tissues than in cover-glass preparations, as well as its refractory staining peculiarities, are some of the obstacles to such a search. It is reasonable to believe that the focal lesions are much more widely spread than has been demonstrated. The bone marrow of a few animals was examined without finding any focal lesions; no human marrow was obtained. Similar necessity for further study concerns the central nervous system. In the few instances in which sections were examined of the cord or brain (some of these had to do with human tissues) no focal lesions were seen. In dismissing this phase of the subject it is proper to liken the focal necroses and the prehminary vessel changes to the alterations caused by the so-called "endothelial toxins"; furthermore, to recall that some such toxins, it is believed, are liberated from the bodies of bacteria. As for the diffuse changes, the cloudy swelling and fatty changes are not marked. In the lymph glands an hyperplasia occurs and in the spleen a similar process is in part responsible for its consid- erable enlargement. Hemorrhages are not met with in microscopic preparations of the lymph glands as frequently as in the gross ex- aminations. In two instances, both guinea-pigs killed on the sixth and seven days of the disease respectively, the microscopic prepa- rations of the lymph glands look not very unlike fiver at first glance, due to the great amount of cytoplasm visible with low powers of the microscope and its affinity for the cell-body stains. These large cells crowd and distend the sinuses and are markedly phago- cytic. One of the interesting features of the enlargement of the spleen 'Jour. Am. Med. Assn., igog, 52, p. 379. Pathological Anatomy of Spotted Fever 449 in guinea-pigs is the appearance of large numbers of multinucleated cells and cells with nuclei aggregated into rings, cells corresponding in some respects to the giant cells of bone marrow. So far as indicated by these megacaryocyte-Uke cells, this assumption by the spleen of the characteristics of bone marrow was met with in only a few of the spleens of guinea-pigs and then early in the disease. The guinea-pig spleen normally contains such cells in small numbers, but the number present in the spleens of some of the animals is very great. In the human spleens, too, large multinucleated cells possessing some points of resemblance to megacaryocytes are occasionally found, most of them with several small nuclei in the central part of the cell, a few with ring-shaped nuclei. Cells similar to either the megacaryocytes or to multinu- clear endothelial cells were absent from the spleens of monkeys. To what extent these large cells in the spleen are to be attributed to a myeloid transformation of the spleen is difl&cult to state. We are accustomed to attributing similar metaplasias to severe anaemias of some standing. In some instances, as for example about the vessels of the guinea- pig's seminal vesicles, in and between tubules of the testes and epididymites of monkeys, some of which had not attained puberty, and in other structures, the accumulations of cells and evidence of their multipHcation in situ suggest the formation of new depots for the production of leukocytes or other cells which presumably are in some way designed to play some role in the defensive processes. These are usually in perivascular situations and so Hmited to the regions of the lymph channels that it seems unreasonable to ascribe them to the focal processes of blood vessel obstruction and their sequences. The evidences of obstruction to the outflow of bile are present in the human livers, more marked in some than in others. I have not observed them in the livers of other animals. These changes do not differ in any noteworthy way from those encountered in the livers of persons dying from typhoid fever or lobar pneumonia when an icterus develops during the course of these diseases from retrogressive processes in the liver cells and the resulting interfer- ence with the biliary current. 450 Contributions to Medical Science EXPLANATION OF PLATES. (Figs. lo [Plate 5], 12 [Plate 6], 15 and 16 [Plate 8] pertain to the more diffuse lesions, the remainder to localized or focal processes.) Plate i. Fig. I. — Human Rocky Mountain spotted fever. JMinute necrosis in the spleen. X400 diam. Fig. 2. — Larger focal necrosis in the human spleen. X 200 diam. Pl.\te 2. Fig. 3. — Human Rocky Mountain spotted fever. In the peripheral capillaries of the liver lobules leukocytes are very numerous and at various points (a) eosin stain- ing rounded masses occur. In some degree they resemble those met with in typhoid fever. X120 diam. Fig. 4. — One of the eosin-staining ball-like aggregations, with cell inclusions referred to in Fig. 3. Human. X400 diam. Plate 3. Fig. 5. — Experimental Rocky Mountain spotted fever. Large subcapsular necrosis in the liver of a guinea-pig. X 200 diam. Fig. 6. — Experimental Rocky Mountain spotted fever. Small region of necrosis in the liver of a guinea-pig. X400 diam. Plate 4. Fig. 7. — Thrombosed vein in the skin. Monkey. X250 diam. Fig. 8. — Aggregations of leukocytes in the perivascular sheaths in the spleen of a monkey. Experimental Rocky Mountain spotted fever. X 250 diam. Plate 5. Fig. 9. — Illustrating the necrosis of the guinea-pig's ear. X30 diam. Fig. 10. — Multinucleated cells in the human spleen. X 700 diam. Plate 6. Fig. II. — Illustrating the amounts of tissue necrotic in some of the livers of guinea-pigs. X 18 diam. Fig. 12. — Numerous megacaryocytes in the guinea-pig's spleen. X80 diam. Plate 7. Fig. 13. — Aggregations of polymorphonuclear leukocytes in perivascular sheaths of lymphoid tissue. Monkey. X400 diam. Compare with Fig. 8, Plate 4. Fig. 14. — Collections of fibrin, agglutinated red-blood corpuscles, and leukocytes in the vessels of the renal pyramid. Human. X4S diam. Plate 8. Fig. 15. — Subintimal proliferation of cells. Human spleen, (a), mitotic nucleus. X400 diam. Fig. 16. — Perivascular accumulations of cells in the wall of the seminal vesicle. Guinea-pig. X 200 diam. PLATE 1. '•••» -^ %'-:*•. ••?'**.* "•::.; Fui. 2. PLATE 2. 1» » ■ • • .• "■'o •. Fig. 3. # 5# §> •' 6 bar •* ft *t Fig. 4. PLATE 3. ^K^ I'Ki. r,. A '^^■^v. (/ > \i^^ 3 'd^°- ^- * I. J-'- *•' «i^ f * 1 ^1 ''^ ^ ^^'^ /^i ^' '■■ rfi M ^ ■¥ ».. _^ *«> 4 9 8 (^ ^S y# ## U-* ^ -^ ,^t.«K KJ (). PLATE 4. Fig. 7. Fig. 8. PLATE 5. my^ IV r^'-. Fig. 9. Fk;. 10. PLATE 6. -v^.W. W . •'.'V^r^' •*/.'>•••''. •••••'1 '.'"•/* '-^ Fig. 11. ^:^/;5-// »** 'i,*« '^^\!* *^ • Ik.. 12. ♦.? PLATE 7. ^mm^^s^^ ^'^ Tft:^^ ^^^ ^ <«^ *-.^. ^0 Fig. 13 iM.i. 14. PLATE 8. e.> . ^- -n-* * \ C ^ IS; «^ "^ ••^«#* a- ^.^•^••^•' Fig. 15. \' I'Ki. 1(3. s THE TYPHUS FEVER OF MEXICO (TABARDH^LO). PRELIMINARY OBSERVATIONS.' H. T. RiCKETTS AND RuSSELL M. WiLDEE. (Prom the University of Chicago [Department of Pathology] and the Memorial Institute for Infectious Diseases, Chicago.) . It will be recalled that NicoUe^ and his associates recently have greatly advanced our knowledge of typhus fever by two achievements. First, it was shown that the disease could be transmitted to the chimpanzee by the injection of the blood of human patients, and in a similar way from the chimpanzee to the macacus monkey (Macacus sinicus). Strangely they were not able to infect the macacus directly with virulent human blood. Their second important result consists in the transmission of the disease from macacus to macacus by means of the ordinary body louse (Pediculus vestimenti). From the epidemiologic conditions which prevail in Tunis they were able to rule out the flea and the bedbug as carrying agents. When the monkey had once been infected with the louse, they attempted to transfer the disease from animal to animal by means of injection from the first monkey, but the experiments resulted only in what were considered abortive attacks of t)^hus. Hence they concluded that the virus underwent a loss of virulence in the monkey. They obtained apparently typical eruptions in both the chim- panzee and the monkey, but not uniformly in the latter. Recently also, Anderson and Goldberger^ apparently have been successful in transmitting the typhus fever of Mexico to a macacus and a capuchin by the direct inoculation of virulent human blood. Their success in this regard and NicoUe's failure (i.e., to infect the monkey by direct injections of human blood) would seem to bring into question the identity of European typhus with that of Mexico. Prominent Mexican physicians, indeed, for some years have contended that the two are not identical, the chief difference ' From Jour. Am. Med. Assn., igio, S4, p. 463. ' Compt. rend. Acad. Sc, 1909, 149. 3 Anderson and Goldberger, Pub. Health Rep., December 24, 1909. 451 452 Contributions to Medical Science being in the slower onset and defervescence of the Mexican disease. This is particularly true in adults. In children the onset and defervescence of the Mexican disease resemble the European more closely. It seems that the relationship of the two cannot be decided definitely without a more detailed experimental comparison. They may be very closely related, though perhaps not absolutely identi- cal, particularly in view of the fact that both are probably carried by the body louse. We may now describe some of our own experiments bearing on the transmission of the disease to the monkey and the non-filtera- bility of the virus. I. transmission to the macacus by injection. We have been able to confirm the probable susceptibility of the monkey to inoculations with the blood of patients suffering from the tophus fever of Mexico, as reported briefly by Anderson and Goldberger, and to obtain additional information regarding the degree of susceptibility of this animal. In the two monkeys which apparently were infected by Ander- son and Goldberger, multiple inoculations were used. Their macacus received a total of 20 c.c, divided into three injections, and the capuchin a total of 8 c.c, in two injections. In our inoculations single injections were given, the animal being Macacus rhesus. On January ii a quantity of blood was drawn from the median basilic vein of Jose Hernandez, a patient in the Hospital General (Mexico City), on the eighth day of his sickness. The fever and condition of the patient were typical for an attack of typhus of this duration, the attack being one of moderate severity. Although the skin was quite dark, the spots, not yet petechial, could be seen over the abdomen, chest, axillary skin, arms, legs, and back, and the conjunctivae were reddened charac- teristically. The spleen showed little or no enlargement on percussion, and it could not be palpated. Since the blood was drawn the patient has passed through a ty^pical "crisis," which occupied about three days, and is now convalescent. Two cubic centimeters of the blood of the patient planted in 50 c.c. of broth have remained free from discoverable micro-organisms, which corresponds with the usual results of cultivation experiments with typhus blood. Blood inoculation experiment. — Inoculations of i, 5, and 10 c.c. of defibrinated blood were made respectively into monkeys 5, 6, and 7. No. 5 died five days after its inoculation, showing a consolidation of the lungs, and since it had had no fever. The Typhus Fever of Mexico 453 it was discarded from the experiment. In addition, monkey 4 received 8 c.c. of the serum from the same blood. The blood after defibrination stood at room temperature (15 to 20 C.) and in diffuse light, for from six to seven hours before injection. The 5 c.c. of blood which morLkey 6, weighing 2,010 gms., received was diluted to IS c.c. with sterile physiologic salt solution, and the entire amount introduced intra- peritoneally. Its temperature on successive days was as follows: 9:00-10:00 3:00-4:00 9:00-10:00 3:00-4:00 A.M. P.M. A.M. P.M. January 11 100.9 January 17 too. 2 104.9 January 12 99 January 18 97.6 104.3 January 13 99-3 IOI.4 January 19 103.3 101.8 January 14 100. 1 102.4 January 20 97 101. January 13 100.4 103.6 January 21 99 I 93 o* January 16 97-3 102.7 • Death. On January 17, six days after inoculation, the animal ate little, and sat "huddled- up," with hairs more or less erect. This condition continued, and on the 19th there was increased secretion from the conjunctivae, and the animal coughed moderately. The illness appeared more severe on the 20th; there was no resistance to manipulation; emaciation; moderate diarrhea. On the 21st, when the temperature became sub- normal, the animal was still somewhat responsive until about the middle of the after- noon, when its condition grew rapidly worse, and at 9 o'clock it was moribund. The autopsy, which was performed at once, showed nothing distinctive, and very little that appeared abnormal. The lymph glands generally were moderately enlarged, but were not congested or hemorrhagic. Those of the axilla and groin were the seat of old pigmentation. The lungs were pink, and showed no inflammation or other alteration except for a slight amount of atelectasis at the upper border of the left lower lobe. The pleurae were free from signs of inflammation. Moderate swelling of both the kidneys and liver were present, but they were not degenerated. The spleen was rather firm but not distinctly enlarged. No evidence of infection was found in the peritoneal cavity. The meninges and cerebral cortex were free from congestion, edema, or other signs of inflammation. A bouillon flask culture from the heart's blood, and agar slants from the viscera, remained sterile. Monkey 7, a male weighing 2,150 gms., received 10 c.c. of the same defibrinated blood, made up to 20 c.c. with salt solution. One-half the quantity was injected intraperitoneally, the other half subcutaneouslj'. Its temperature on successive days is given below. 9:00-10:00 3:00-4:00 9:00-10:00 3:00-400 A.M. P.M. A.M. P.M. January 11 104.0* January 19 102.2 10S.2 January 12 101.8 102.6 January 20 102.7 IOS.6 January 13 100.8 102.9 January 21 102.4 103.6 January 14 100.2 102.7 January 22 103.2 106.6 January 15 100. s 104.1 January 23 100.6 104.8 January 16 101.4 104.3 January 24 105.1 10S.4 January 17 98.0 105. 1 January 25 103.2 104.6 January 18 100.4 104.7 January 26 103.1 107.2 • Before injection. 454 Contributions to Medical Science The animal first began to appear sick on the i6th, the second day of fever. On the 17th it made little resistance to manipulation and from this on it appeared dis- tinctly ill and ate little. It developed no marked diarrhea, although the stools became rather soft. While the temperature became high the animal at no time lost its respon- siveness. The conjunctivae were not noticeably reddened, and a distinct eruption could not be identified. After a rather long and severe course, the animal recovered. Monkey 4, weighing 1,800 gms., received 8 c.c. of serum of the same blood; the serum was obtained by defibrination and centrifugation. This quantity was diluted to 25 c.c. with salt solution, one-half being injected intraperitoneally, the other half subcutaneously. The temperature on successive days was as follows: 9:00-10:00 3:00-4:00 1 ! 9:00-10:00 3:00-4:00 A.M. P.M. A.M. P.M. January 11 103. 6» January n . . . . 104.2 104.2 January 12 100. January 18 . . . 101 5 104.2 January 13 99-4 I03-3 I January 19 ... . •; 103.2 103. s January 14 99-7 102.1 January 20 ... . .1 101. I 103.0 January 15 100.4 103.4 January 21 ... . . 1 100.6 103.7 January 16 101.3 102.9 January 22 ... . .1 t 1 I • Before injection, t Too low to register. % Death at 6 p.m. On the 17th, the first day of distinct fever, the animal, which had hitherto appeared vigorous and healthy, looked sick, unkempt, its hairs stood up, and it "huddled up" even in the sun. This condition continued, food was refused, and emaciation developed. On the 20th a moderate diarrhea appeared and continued until death. On the afternoon of the 2 2d it became soporose, and died at about 6 o'clock. An eruption which could be referred positivelj- to the infection, or bearing a resemblance to that of t>'phus fever, could not be identified. An "eruption" which did appear on the skin of the lower chest and the upper portion of the abdominal skin probably was accidental. On the i8th, the second day of fever, the monkey had been bled from the heart, for cultivation and other experiments. A small amount of blood escaped through the skin when the needle was withdrawn, and at autopsy it was found that some subcutaneous hemorrhage had occurred and that the blood had "diffused" posteriorly in the form of a rather narrow band, following the median line. The "eruption" was roughly median although it extended about an inch beyond the visible limit of the subcutaneous extravasation. In character, it was at first pink, that of the early spot, and seemed to disappear on pressure. It appeared two days after the heart had been punctured and on the fourth day of fever. On the following day and subsequently it became darker, more or less cyanotic in color, and could not be effaced by pressure. The spots were rather ill-defined, and appeared to consist of collections of minute punctiform hemorrhages. A similar condition could not be identified on other parts of the bodj', and in view of the subcutaneous hemorrhage which compUcated the situation it seems probable that the eruption had its source in the latter rather than as a manifestation of tjiihus fever. At the autopsy the lungs were found of a normal pink color; there were no signs of inflammation. On the visceral pleura, particularly of the left lung, were a number of small, circular, dark red hemorrhages from 0.5 to 1.5 mm. in diameter. The pleural cavities were normal; the heart normal; no inflammation of the valves or The Typhus Fever of Mexico 455 pericardium. The liver was apparently somewhat enlarged, pale, as if fatty, but showed little or no congestion; the lobules were well marked; anterior border dis- tinctly rounded. The spleen was about i . 5 cm. longer than that of monkey 6 and perhaps a few millimeters broader; was distinctly enlarged, bluish-red in color, and of rather firm consistence ; contained no hemorrhages. The kidneys were perhaps a little enlarged and moderately congested; cortex and pyramids of a homogeneous normal color; striations normal; the cortex had a relation to the medulla of about one to one. The mucous membrane of the colon was much reddened, and perhaps even hemorrhagic; the colon contained a large amount of glairy mucus but no blood or feces. The mucous membrane of the ileum appeared normal, and the ileum contained nothing but a slightly viscous yellowish fluid; there was a short intussusception with no inflammatory or obstructive signs of the parts involved. The duodenum con- tained bile-stained mucus, the mucosa being normal. The stomach contained some undigested banana and mucus. The lymphatic nodes everywhere seemed more or less enlarged but were not congested or hemorrhagic. Those of the groin and axilla were almost black from some previous pigmentation. The meninges showed a good deal of congestion and edema, the fluid being per- ectly clear. The cerebral cortex appeared normal. Other parts of the central nervous system were not examined. Cultures, as in the case of No. 6, remained sterile, including those from the meninges. As a basis for interpreting these experiments, we have for con- sideration : the existence of an incubation period which was approxi- mately the same in all three animals, and during which they remained healthy; the occurrence of illness and fever followed by the death of two of the animals; the sudden onset of fever and illness, and the rapid defervescence in the animal which recovered ; the negative outcome of cultures; and the more or less negative findings at autopsy, corresponding with the condition in typhus in man. The temperature curves of monkeys 4 and 6 suggest that the interpretation of infection purely on the basis of the fever mani- fested may be attended by diflSculties in some instances. The variations between the morning and afternoon temperatures of uninfected and apparently healthy monkeys is often considerable, and in our experience has not been constant in the same animal from day to day; marked variations have been encountered when different animals, apparently healthy, were compared. The morning temperatures of healthy controls have habitually been from 0.8 to 1.8 or 2.0 degrees lower than those of the afternoon. 456 Contributions to Medical Science The former is commonly below 102.0, and may lie between 99.0 and loi .0 for several days in succession; whereas in the afternoon (from 3 to 4 o'clock) it commonly is foimd at some point between 102.0 and 103.0. The comparatively cold nights and the warmer days may have some influence on the more extreme variations. We have attempted to eliminate this factor as much as possible by warming the room from the hours of 5 p.m. to 9 a.m., the tempera- ture at night being about 15 C. During the warmer portion of the day the animals have been placed where they could avail themselves of the sun. It is noteworthy that No. 6 on only one day, the 19th, exliibited a morning temperature (103.3) which was distinctly above the normal for that hour, and that on only two days after illness began (17th and 1 8th) was it distinctly above the normal for the after- noon, yet from the 17th on it was manifestly a very sick animal. The temperature was subnormal for two days preceding death. In No. 4 the morning temperature was above the normal Kmit on two days (17 th and 19 th) and in the afternoon on five successive days, the period of subnormal temperature preceding death being much shorter than in the case of No. 6. In No. 7 the morning temperature was somewhat above the expected normal (being from 102.2 to 103.2) for four successive days, and in the afternoon there was rather marked elevation for ten successive days. The short course of the fever and the fatal termination in Nos. 4 and 6 may perhaps be accounted for by a relatively low resistance on their part. In drawing the line between the end of the incubation period and the onset of disease, the general appearance and behavior of the animal appear to be of equal or greater importance than the temperature alone; taken together they are a sufiiciently satis- factory index. There is some difference between the incubation period of Nos. 4 and 6, on the one hand, and No. 7, on the other. In the two former it lay between five and six days, whereas in the latter it was reduced to four days or less. The difference in dosage may have accounted for this. The Typhus Fever of Mexico 457 It is worthy of comment that the smaller dose (5 c.c.) which No. 6 received seemed more virulent than the 10 c.c. administered to No. 7. Differences in the resistance of individuals may, of course, be called on to explain this. No. 6 received its entire injection intraperitoneally, whereas in No. 7 the amount was divided equally between the peritoneal cavity and the subcutaneous tissue. Whether there is a difference in \drulence depending on the route of injection has not been determined. No. 4 is hardly comparable with 6 and 7 in this regard inasmuch as it received serum rather than defibrinated blood. We may further call attention to the fact, as mentioned at the outset, that a single injection of virulent blood, as well as the multiple injection used by Anderson and Goldberger, is capable of producing infection, and this with a short incubation period. It is unfortunate that the animal which received i c.c. of the virus died of other causes, since it was hoped that this injection might furnish some index regarding the degree of susceptibility of the monkey. This is important from the point of view of insect trans- mission, which would seem to demand an animal of rather high susceptibility. The more exact susceptibility of the species will be studied further. It is of some significance, too, that the serum appears to be infective as well as the defibrinated blood. As yet we have no data concerning the relative infectivity of the two, further than that 8 c.c. of serum appeared to be as virulent for No. 4 as 5 c.c. of blood was for No. 6. The material injected was diluted with salt solution on theo- retical grounds, and because experience with Rocky Mountain spotted fever suggested that dilution of the virus might favor infection. Dilution is known to render less effective specific antibodies which may be destructive to the virus, and there can be little doubt that blood drawn from typhus patients on the eighth day of the disease contains germicidal antibodies, particu- larly since the disease is one which, clinically, is known to cause the development of distinct immunity. Dilution may also favor infection through the peritoneal route by affording better condi- tions for rapid absorption or dissemination of the micro-organisms. 458 Contributions to Medical Science In experiments with spotted fever we have noted occasionally that smaller doses of virus would cause infection when larger doses of the same material would not, the injections being intra- peritoneal. Small doses, as o.oi c.c, are usually diluted with 2 or 3 c.c. of salt solution, whereas larger doses, as 5 ex., are diluted to a much less degree or not at all. The moderate diarrhea which appeared in No. 6 was probably of no significance, inasmuch as certain of the uninoculated animals were similarly affected. We may, however, call attention to the fact that some Mexican observers insist that there is a certain degree of intestinal disturbance in typhus in man, manifested frequently by diarrhea, and the right ihac region is habitually examined cHnically for signs of such disturbance. We have observed such disturbance in a number of instances, but have no personal knowledge of its relationship to typhus fever. We are hardly in position as yet to make a thorough comparison of the infection in the monkey with that in man. It may be stated, however, that the incubation period as obtained by this method of infection was shorter than that which is accepted for man, the latter being in the neighborhood of ten to twelve days. Manifestly this is subject to considerable variation in experiments, since in the macacus infected by Anderson and Goldberger it was twelve days, and in our No. 7 only four days. The duration of typhus in man varies, according to Dr. Genaro, Escalona, from about twelve days in children, in whom it is com- paratively mild, to twenty-one or twenty-four days in adults. In monkey 7 the course lasted for approximately twelve days. The onset in all three animals was sudden and the defervescence in No. 7 moderately rapid, comparable to the conditions in human patients. In man the temperature is said to rise gradually for from three to four days, when the fastigium is reached, the morn- ing remissions being marked, as in the case of the experiments described. In man also defervescence occupies from three to four days, followed by from two to four days of subnormal tem- perature, corresponding to the conditions in monkey 7. Bacteriologically the conditions are similar, in that attempts at cultivation by ordinary methods have given negative results. The Typhus Fever of Mexico 459 Likewise the findings at autopsy are similar, nothing distinctive being found in the animals, as in the case in human beings. The cerebral congestion and edema of No. 4, the slight or moderate enlargement of the spleen, congestion of the abdominal viscera, doubtful or moderate enlargement of the lymph glands, and the absence of localized inflammations are conditions which are commonly found in man. The negative findings appear to be of more diagnostic significance, when considered with the clinical course, than positive anatomic changes. Degeneration of the liver and kidneys is frequently found in man, but it is not a necessary accompaniment of fatal infections. In the experiments performed so far, there is a lack of corre- spondence with the human infection, in that no eruption resembling that of typhus appeared in the animals. Although it would be gratifying to reproduce this phenomenon, failure to do so can hardly constitute a valid argument against the character of the infection in the monkeys. We recall that not all monkeys and guinea-pigs infected with spotted fever exhibit the eruption, and the conditions may be similar in typhus. Eventually it seems probable that the nature of the infection in all cases of recovery can be checked up by means of immunity tests. The disease appears to be one which, in man, confers distinct immunity, and animals which have once suffered from infection should be immune to further inoculations. This will be investigated and reported on at a future date. II. FILTRATION EXPERIMENT. With the desire of obtaining information about the virus of the disease, especially as having a possible bearing on the approxi- mate size of the micro-organism, we have performed a filtration experiment which it is desirable to report. Of the blood referred to in the previous experiments, 34 c.c, were centrifugated until the corpuscles occupied approximately the lower three-fifths of the column. The overlying serum was drawn off and replaced by an equal amount of sterile salt solution. The corpuscles were thoroughly mixed or washed with the latter and again centrifugated moderately, after which the overlying 460 Contributions to Medical Science fluid was added to the first portion. This was repeated again and the three fluid portions then combined. It seemed probable that through this procedure one might obtain a larger quantity of micro- organisms in the fluid than by resorting to a single more vigorous centrifugation. This seemed better also than to attempt to filter the uncentrifugated blood, which indeed is an almost impossible task with moderate pressures. It seems sufficiently accurate to consider that the defibrinated blood consisted of about equal parts of serum and corpuscles, and that we obtained the equivalent of about 17 c.c. of serum by this washing process. For the purpose of filtration the serum was diluted to 51 c.c. by means of salt solution, and this quantity was divided into two equal portions, one to be filtered and the other to be injected without filtration. The first portion was passed through a small Berkefeld candle, and with the threefold dilution it filtered readily. The filter was washed by passing through it an additional 5 c.c. of salt solution, this filtrate being added to the first. The interval between the drawing of the blood from the patient and the injections was from six to seven hours, as in the preceding experiments. The total quantity of each portion was approximately 25 c.c, of which one-half was injected intraperitoneally, the other half subcutaneously, after suitable preparation of the skin. The result of the injection of the unfiltered serum into monkey 4 was given in the first part of this paper, it being our conclusion that the animal became infected with and died of tj^phus fever. No. 3, which received the filtered serum, exhibited the following temperatures on successive days: A.M. P.M. A.M. P.M. January 1 1 103.2 January 18 102.6 102. s January 12 102. s 102.6 January 19 103.6 101.6 January 13 102.8 103.3 January 20 102.6 103.6 January 14 102.7 103.6 January 21 101.9 103.9 January 15 103.6 103.8 January 22 102.6 104.4 January 16 103 102.6 January 23 101 .6 102.7 January 17 103 102.8 January 24 loi .7 103.6 The Typhus Fever of Mexico 461 In spite of rather high morning temperatures, the animal has been in apparently perfect health since the date of inoculation, which is in distinct contrast with the course shown by No. 4. As stated previously, the latter, after an incubation period of about six days, developed fever, grew sick, and died eleven days after inoculation, with findings which are in harmony with those of typhus fever. We may therefore conclude that the virus did not pass through the filter employed, or if it did, that it did not pass through in a quantity sufficient to produce recognizable infection. It seems probable, therefore, that the virus of the typhus fever of Mexico may be classed with the unfilterable. It is, of course, to be under- stood that the micro-organisms may find their way through filters of greater porosity which may be encountered, or through filters with thinner walls. It is possible, too, that they may even pass through filters of the type and thickness employed in this experi- ment, in quantities which are sufficient to vaccinate, but not sufficient to produce virulent infection. As affording an analogy for this possibility, it may be mentioned that we have on several occasions produced extremely mild attacks of spotted fever in the guinea-pig, resulting in permanent immunity, by the injection of very small doses of virus. This possibihty will be investigated. m. THE QUESTION OF CONTAGIOUSNESS AND TRANSMISSION BY INSECTS. Although the old idea that typhus is a contagious disease still has strong supporters in Mexico City, the conception that insect transmission is in better accord with the epidemiologic conditions is prominent in the minds of those who are in daily contact with the disease. Without giving a detailed presentation of the conditions at this time, it may be stated that of the three insects which are most open to suspicion, i.e., the louse (Pediculus vestimenti), the bedbug, and the flea, only the first would seem to merit serious considera- tion, because of the epidemiology of the disease. This conclu- sion is, of course, strongly supported by the results already obtained by Nicolle. 462 Contributions to Medical Science The season of greatest predominance of typhus in Mexico City (the spring) does not coincide with the period of greatest prevalence of the flea (summer). And, regarding the bedbug, we have knowledge of individual cases which, it would seem, could not possibly have been carried by this insect. Although our own experiments with the louse, which have been under way for some time, have not yet resulted in transmission they will be carried further and reported in detail at a future date. We take pleasure in acknowledging our great indebtedness to the local authori- ties and physicians of Mexico City who have co-operated with us in various ways: more particularly to Dr. Gavino, Director of the Bacteriologic Institute, where we enjoyed the advantage of a well-equipped laboratory, and to his assistant, Dr. Girard; as well as to Drs. Fernando Lopez and Genaro Escalona, of the General Hospital. 1 THE TRANSMISSION OF THE TYPHUS FEVER OF MEXICO (TABARDILLO) BY MEANS OF THE LOUSE (PEDICULUS VESTIMENTI).^ H. T. RiCKETTS AND R. M. WILDER. {From the Department of Pathology of the University of Chicago and the Memorial Institute for Infectious Diseases, Chicago.) In a previous article^ we referred to Nicolle's experiment, in which he apparently succeeded in transmitting the typhus fever of Tunis from the chimpanzee to the macacus monkey, through the bite of the louse. Nicolle used twenty-nine lice in his experi- ment; after feeding on the chimpanzee they were allowed to feed on the monkey for several successive days. Also, Anderson and Goldberger^ report two attempts to transmit the typhus fever of Mexico City to the macacus by means of the louse. Their first experiment seemed to be without result, and although the second showed a slight rise in temperature, their inability to give an immunity test to the animal makes it impossible to interpret the result. This test is essential, as will be manifest from consid- erations which follow. Since January i we have performed a series of experiments bearing on the relation of Pediculus vestimenti to the typhus of Mexico, some of which may be recorded at this time. Certain conditions pertaining to the experiments should be detailed before the protocols are given. The first relates to the susceptibiHty of the monkey to typhus, and the positiveness with which the disease may be recognized in this animal. As shown by the experiments of Anderson and Goldberger, and by our own, the injection of virulent blood from man, under suitable conditions, gives rise to a well-marked course of fever, following an incubation period from five to twelve days. A further constant factor is the absence of micro-organisms, which are sus- ■ From Jour. Am. Med. Assn., igio, 54, p. 1304. 'Ibid., p. 463. See p. 451 of this book. ' Anderson and Goldberger, Pub. Health Rep., February 18, 1910. 463 464 Contributions to Medical Science ceptible to cultivation by ordinal methods, from the blood of the animal during his course of fever. This sequence of events has now resulted a sufficient number of times without any failures, so that it may be expected constantly when the established technic is observed. In the course of our experiments so far, six monkeys have received virulent blood from man in doses of from i to 7 c.c; two have received virulent unfiltered serum in doses of 7 . 5 and 8 . 5 c.c; and one received 5 c.c. of defibrinated blood from another monkey as an experiment in passage. All of these animals exhib- ited an incubation period followed by a distinct course of fever, accompanied by signs of severe illness, and with negative attempts to obtain cultures from the blood. Anderson and Goldberger obtained similar results in the four animals inoculated by them. It would seem, therefore, that even in the absence of an eruption resembhng that of typhus (and indeed, no such eruption has appeared), the constancy of the phenomena mentioned is a satisfactory index of the transmissibility of the disease to the monkey. Constantly the temperatures of seven or eight normal monkeys were taken twice daily as a means of detecting possible adventitious infections which might be contagious in character; but none was observed. A second, and, as it seems, a conclusive method of determining whether a course of fever was really due to an infection with typhus, consists of an immunity test, i.e., a second injection of virulent blood after the subsidence of the fever. The immunity test must, of course, be accompanied by the injection of a similar quantity of the same blood into a normal monkey, as a control to the virulence of the material. Thus, six monkeys (Nos. 3, 7, 9, II, 24, and 25) which exhibited a distinct course of fever following their first inoculation, and all of which were seriously ill, showed no febrile response, or other signs of illness, when the second inoculations, or the immunity tests, were given. They received their immunity tests in three groups at different times and in all five normal controls were used for each group. Five c.c. of human virus were injected into each member of one of the groups, and 4 c.c. into each of the other two. The controls in each case developed a course of fever similar to that already described. Transmission of Typhus Fever of Mexico by the Louse 465 The degree of susceptibility of the monkey is another important consideration, and, if it is low, experiments on insect transmission might be rendered difficult or impossible. It will be recalled that NicoUe failed entirely in his attempts to infect the macacus with the Old World typhus by the direct injection of human blood; he succeeded only by first infecting the chimpanzee and then trans- ferring blood from the latter to the macacus. After the macacus was once infected, he attempted to maintain the disease in this species by passage, but the virus apparently underwent a rapid loss in virulence, and the infection "died out" after two or three passages. It seems also that Anderson and Goldberger had a similar experience in trying to maintain Mexican typhus in the macacus. We, also, have made two attempts to keep the disease alive by passage. In one instance, the second monkey (No. 11) became infected, whereas the third (No. 22) showed no fever. In the second instance the second animal (No. 9) exhibited only an in- definite fever, and further passage was not attempted. There was, therefore, sufficient reason to suspect that the macacus might be sufficiently susceptible for satisfactory work on insect transmission. Information of at least suggestive value regarding this question might be obtained by determining the minimum infective dose of human blood for the monkey. Although it appeared inadvisable, on account of the cost, to investigate this point extensively, one experiment was performed in which one animal (No. 25) received i c.c. of virulent human blood, and another 0.2 c.c. of the same material. After the lapse of an incubation period of ten days, the former developed a severe course of fever of ten days' duration, and showed no reaction to a second injection which was given a month later. The animal which received o . 2 c.c. had no fever and showed no signs of illness. Although the immunity test of the latter remains to be given, it is probable that the minimum infective dose in this instance lay between o. 2 c.c. and i c.c. The question arises, however, as to whether the monkey may undergo a mild type of infection, which might be accompanied by little or no fever, and which would be recognizable only from 466 Contributions to Medical Science the fact that the mild or ''abortive" attack had established immu- nity to a subsequent injection of virulent blood. The possibility of such an occurrence is well known. We have observed it a number of times in experiments with spotted fever on the guinea-pig. We have, furthermore, one instance of a very mild and scarcely perceptible infection with typhus in the second attempt at passage referred to above. Monkey 4, which was infected with virulent hximan serum, and whose history was given in a previous paper, was bled from the heart on its second day of fever and 5 c.c. of the blood were injected intraperitoneally into No. 9, which showed on succes- sive days the temperature given in Table i : TABLE I. Temperature of Monkey 9 atter Intraperitoneal Injection of Infected Blood. AM. P.M. F. c. F. c. 103.6 102. 5 101.8 103.2 102.3 102.0 101.6 102.0 102.0 loi .0 102.7 102.9 101.8 102.9 102.7 104.0 104.0 J03O 103. 1 101.8 lOI.O 39.8 39-2 38.8 39 S 391 38.9 38.6 38.9 38.9 38.3 39-3 39-4 38.8 39-4 39-3 40.0 40.0 39-9 39-5 38.8 38:3 102.6 103.4 103.1 103.9 102.8 102. s 102.6 103.6 103.2 102.4 104 -3 102.0 103.1 103.4 103.2 104.8 104-4 103.9 103.6 103 -3 103.0 102.8 January 20 39.6 January 23 39-3 39. 2 January 26 January 27 39-8 39 S 40.2 38.9 January 31 February i 39S 39-7 40.4 February 4 40.2 39-9 February 6 3Q.8 February 7 39.6 February 9 39-3 Below 103 for two weeks thereafter. Not until the sixteenth day after inoculation did the temperature rise above a possible normal (for this individual), remaining somewhat elevated for only three or four days. A positive diagnosis of infection with tjqphus could not be made under these conditions; yet, when an immunity test was given (February 18) the tempera- ture did not rise higher than 103 during the twenty-six days it was under observation. The control (No. 18) exhibited, on the other hand, a severe course of fever. Hence, as a consequence of this experiment, and from the experience with spotted fever, it seemed not unreasonable to anticipate that the monkey might suffer from mild infections with typhus, not manifested by marked elevation of the tempera- ture or by distinct signs of illness, but the occurrence of which Transmission of Typhus Fever of Mexico by the Louse 467 might be determined later by means of immunity tests. This, we believe, is what happened in our attempts to transmit the disease by means of the louse, as will appear below. The point may also be raised as to whether some of our monkeys may have acquired typhus by contagion, or by means of their own lice from other monkeys which had been infected deliberately, thus vitiating later experiments. Although the evidence is entirely against t^'phus being contagious, we protected our experiments (by first killing the lice on the monkeys by means of insect powder, petroleum, and scrubbing with green soap; and furthermore, by the practice of keeping a healthy monkey in the same cage with an infected animal in order to test the question of contagiousness). Two monkeys which were thus exposed (Nos. 18 and 20) proved susceptible to later inoculation, the result tending to disprove contagiousness. Comparatively large numbers of lice were used in each experi- ment, because of the supposed low susceptibility of the monkey. This condition lays the experiments open to the objection that the transmission might be purely mechanical, and may not, on this account, reproduce the natural conditions. The natural method, however, is still unknown, and may indeed be of a mechanical nature rather than the so-called "biologic." If sixty lice are able to infect a monkey mechanically, a smaller number may infect man, provided the latter is more susceptible. Protocols follow. Transmission of Typhus from Man to the Monkey by Means of Pediculus vestimenti. Group 3, Pediculus vestimenti, was infected as follows: January 5, they fed for thirty minutes on E. S., on the tenth or eleventh day of his sickness; January 6, on J. v., tenth day of sickness; January 7, on C. A., ninth day of sickness; January 8, on the same patient. On January 9, 45 lice were alive and 40 fed on monkey i for about one hour: January 10, 334 were alive and 23 fed on the monkey; January 11 they were not fed; January 12, 19 fed; January 13, 13 fed; January 14, 11 fed; January 15, 9 fed. The feedings were not carried farther. The temperature of the monkey on successive days was as shown in Table 2 . It will be noted that in the afternoons of January 17, 20, 21, and 22, the tem- perature lay between 103.2 and 103.6, the first rise being eight days after the lice began to feed, or, having the 20th in mind, the interval was ten days. Inasmuch as the animal's temperature was constantly below 102 . 5 for a period of thirty-one days 468 Contributions to Medical Science later, it is not improbable that the slight elevation referred to represented a mild infection, although no other signs were apparent. TABLE 2. Temperature of Monkey i apter Being Bitten by Infected Lice. 4 A.M. F.U. F. c. F. C. I00.4 I02.8 98.6 99.2 99-4 101.2 100.8 99.0 100.6 100.6 90-1 98.9 101 .2 99-7 100.9 99-4 98.5 38.0 39-3 37.0 37.3 37.4 38.4 38.2 37.2 38.1 37.3 37-2 3S.4 37.6 38.3 37.4 37-5 104.6 101.2 102.9 99-9 loi .7 100.2 103 -6 102.9 102.0 103.2 103.2 103.2 102.6 102. 1 102.6 103.3 40.3 38.4 39-4 37-7 38.7 37 -9 39-8 39-4 38.9 30 S 395 395 39.3 39.0 39.2 39.6 Did not rise above 102 . 4 F. thereafter. The belief that the animal had suffered from typhus was corroborated by an im- munity test, consisting of 7 c.c. of defibrinated blood from patient 27, given February 8. For nineteen days thereafter, the morning temperature lay between 99.0 and 102.0, and in the afternoon between 101.5 and 102.8. This immvmity test was controlled by inoculations of 4 c.c. of the same blood' into No. 3, and i c.c. into No. 25. No. 3, after an incubation period of seven days, and No. 25 of nine days, developed fever, which in both animals lasted for ten days, going as high as 106 in the former, and 106.8 in the latter. The immunity of No. i was again tested a month later, March 11, with the same result, whereas control 20, after an incubation period of five or six days, passed through the usual course of high fever. The first immunity test, however, is the one which indicates, and it would seem conclusively, that No. i was infected by the lice. Transmission from Monkey to Monkey by Means of the Louse. Monkey 7 was infected by the injection of blood from man, as described in our first paper. Group 5, Pediculus vestimenti, was infected by three feedings on monkey 7, on the sixth, seventh, and eighth days of its fever. Thereafter, the lice were fed for eight successive days on monkey 12, their number gradually decreasing from eighty- one on the first day to nine on the last day of feeding. The temperature of No. 1 2 was irregular and rather high prior to the experiment, although the animal was active and well nourished. ' No. 3 had been injected p^e^'iously with filtered serum, without becoming infected; hence this injection constitutes an immunity test for No. 3. Transmission of Typhus Fever of Mexico by the Louse 469 Its temperature continued as in Table 3: TABLE 3. Temperatukje of Monkey 12 after Being Bitten by Infected Lice. F. C. January 23 January 24 January 25 January 26 January 27 January 28 January 29 January 30 January 31 February i February 2 February 3 February 4 February s February 6 February 7 February 8 103.4 103.6 102.7 102.6 102.9 104.2 103.3 103.0 102.7 103.7 103.7 103.6 103-4 102.1 103.2 103.6 103.4 103.9 104.3 103.2 loi .9 103.0 104.2 104.0 103.6 103.4 103.3 104.0 103. S 103.7 103.7 A definite course of fever cannot be made out positively. If present, it would appear to lie between January 30 and February 5 or 6, a period of seven or eight days, and the incubation period would be seven or eight days. On February 2, the animal was manifestly ill, in contrast to its former active condition, and this condition continued for four or five days. On February 9, however, it was again active and appeared well. At this time an immunity test was given; the temperature is given in Table 4: TABLE 4. Temperature of Monkey 12 after Immxtnity Test. A.M. P.M. F. C. F. c. February 9 103.9 102.5 103.3 103.1 102.9 102.9 102.7 102.3 102.9 103.2 103.2 IOI.9 102.9 102.6 102.4 lOI.I 102.9 102.4 102.4 103.9 103.9 103.9 103.7 103.4 103.4 104-3 104.4 103. S 103. S 104. 1 103.7 103.6 102.8 102.9 103.2 102.4 103.7 102.6 39.9 39.9 39.9 39.8 36.7 39.7 40. a 40.3 39-7 39.7 40.1 39-7 39-7 39-3 39-4 39.5 39-1 39-7 39-2 39 39 39 39 39 39 39 39 39 39 38 39 39 39 38 39 39 39 2 6 5 4 4 3 I 4 S S 8 4 2 Z 3 3 I February 12 February 15 February 25 Continued similarly for twelve days more. The controls were Nos. 3, 24, and 25, already cited. A second immunity test resulted in the same way, No. 20 (see above) being the control. During neither of the immunity tests did the animal show any sign of illness. 470 Contributions to Medical Science No. 225 is a particularly good control for No. 12, since, like the latter, its tempera- ture was naturally high and irregular. The course of fever was as shown in Table 5 : TABLE s- Temperature of Monkey 25, Control of No. 12. A.M. P.M. F. c. F. C. 103.9 104.0 103.3 103 -9 103.5 102. s 100.9 103.4 102.3 102. s 104.6 103.7 104.4 104.2 106.3 106.2 104.6 105.8 103.6 101.6 40. 40. 39-6 40. 39-7 III 39-7 39- 1 39-2 40.3 39.8 40.2 40.1 41-3 41.2 40-3 41 .0 39-8 38.7 104.4 104.0 104.1 103.2 103.6 103-3 103.9 103.9 103.2 105.2 105.5 104.4 104.4 104.7 106.8 105.8 104. 5 104.2 103.2 102.9 40.2 40.0 40.1 39-6 39-8 39 6 40.0 40.0 39-6 40.7 40.8 40.2 40.2 40.4 41.5 41 .0 40.3 40.1 39. 5 39-4 Continuing as before the fever appeared. In our opinion, the result justifies the conclusion that No. 12 was infected by the lice of Group 5. Infection of the Monkey by Introducing Intestinal Contents of Lice into Scarifications. A first experiment, which consisted of the subcutaneous injection of the intestinal contents of infected lice, resulted in death in less than twenty-four hours from sep- ticemia. A second experiment was performed as follows: As the lice of Group 5 were feeding on monkey 12, a small quantity of feces was collected from a number of the lice as it was extruded, and placed in a sterile test-glass. To this material the abdomi- nal contents of three lice were added and the mass was triturated in sterile salt solu- tion. This was done three days after the last feeding of the Hce on infected monkey 7. Twelve small incisions, each less than one-eighth inch in length, and extending through the entire depth of the skin, were made in the abdominal skin on monkey 13. The emulsion of feces and abdominal contents was then instilled into these incisions, which thereafter were massaged by means of a sterile probe. The incisions healed promptly and without suppuration. The temperature of No. 13 was as shown in Table 6 . As appears in the table, the temperature rose on the fifth day after inoculation and remained above the normal for this animal for five or six days. During this period the animal became passive and was not inclined to run about, although it was not seriously ill at any time. As a consequence of an immunity test given on February 9, the animal showed no febrile reaction whatever and appeared perfectly well, whereas the controls (Nos. Transmission of Typhus Fever of Mexico by the Louse 471 3, 24, and 25) reacted with severe fever, as stated above. A second immunity test, given a month later, gave the same results, the control in this instance being No. 20 (see above). TABLE 6. Temperature of Monkey 13 after Injection with Abdominal Contents of Infected Lice. A.U. P.U. F. C. F. C. 102.7 102.4 100.3 102.2 102.8 103.3 103.0 102.2 104.4 103.7 102.4 101.8 102.0 103-4 103.6 102. 1 IOI.9 104.0 103.9 103.8 103.8 102.4 103.0 103. 1 102. I 103.0 39.6 39.7 390 38.8 40.0 40.0 39-8 39-8 39.1 39-4 39. S 390 39-4 39 38 39 39 39 39 39 40 39 39 3§ 38 I 3 6 4 2 8 1 7 8 February 3 February s February 6 February 7 In our judgment, this experiment proves the existence of the virus of typhus fever in the abdominal contents of the louse for at least three days after feeding on infected blood. SUMMARY AND CONCLUSIONS. 1 . It seems that Macacus rhesus can be infected with tabardillo invariably by the injection of virulent blood from man taken on the eighth to tenth day of fever. The blood should be diluted with salt solution, as stated previously. 2. Attempts to maintain typhus in the monkey by passage through other monkeys were not successful. 3. The monkey may pass through an attack of typhus so mild that it cannot be recognized clinically. Vaccination results. 4. The immunity test is a reliable proof of the previous occur- rence or non-occurrence of typhus at least within a period of one month. 5. Typhus was transmitted to the monkey by the bite of the louse in two experiments, the lice in one instance deriving their infection from man and in another from the monkey. 6. Another monkey was infected by typhus through the intro- duction of the feces and abdominal contents of infected lice into small incisions. Other experiments, in which the immunity tests have not yet been given, corroborate the carrying power of the louse. 472 Contributions to Medical Science As to whether these experiments on transmission are to be considered merely as substantiating that of Nicolle, or whether they should be regarded as new observations, will depend on the identity or non-identity of tabardillo with the Old World typhus. If Macacus rhesus is susceptible to the direct injection of blood from patients suffering from tabardillo, and not to the injection of the blood of the Old World typhus, the two diseases in our judgment cannot be regarded as identical, although they may be closely related. As stated in a previous paper, Nicolle was unable to infect the macacus directly with human blood, but succeeded first in infecting the chimpanzee and then the macacus by the injection of blood from the chimpanzee. Nicolle's result, if it is to be regarded as final, when compared with ours and with those of Anderson and Goldberger, seems to indicate non-identity of the disease in the two locaKties. It would seem that Nicolle's work, at least in this regard, demands repetition. Possibly, also, the point may be elucidated further by determining whether an attack of Old World typhus confers immunity to tabardillo. We are greatly indebted to Director Gavino of the Bacteriologic Institute, to his assistant, Dr. Girard, to their superiors of the Department of Public Instruction, and to the authorities of the General Hospital of the City of Mexico, in particular to Dr. Escalona, for their numerous courtesies and co-operation. THE ETIOLOGY OF THE TYPHUS FEVER (TABARDILLO) OF MEXICO CITY. A FURTHER PRELIMINARY REPORT.' H. T. RiCKETTS AND R XJ S S E L L M. WiLDER. {From the University of Chicago [Department of Pathology] and the Memorial Institute for Infectious Diseases, Chicago.) Some months ago^ we reported an experiment in which the virus of typhus, as it exists in the diluted serum of the patient, failed to pass through a Berkefeld filter. The monkey which received the unfiltered serum exhibited a severe course of fever after an incubation period of five days, whereas the animal which received the filtered serum developed no fever and remained perfectly well. The immunity test, which has not been reported heretofore, confirmed the conclusion that the virus did not pass through the filter. This test, which was given about one month following the "filtration experiment," consisted of the intraperitoneal injection of 7 c.c. of diluted defibrinated blood drawn from a human patient on the tenth or eleventh day of his fever. No. 3, which had tolerated the filtered serum without visible disturbance, showed a course of high fever lasting eleven days, and preceded by an incubation period of seven days, as the result of the immunity test. The animal which had received the unfiltered serum died as a consequence, and of course, no immunity test could be given. This experiment has been repeated with a similar result. No. 19, which received filtered serum, showed no signs of infection, whereas No. 18, which was given unfiltered serum, passed through the customary incubation period followed by a severe course of fever. Anderson and Goldberger^ obtained a similar result. It seems, therefore, that the evidence is sufficiently strong that the virus of tabardillo does not pass through a filter of the type mentioned. This being the case, the result may be taken as oflfer- ' From Jour. Am. Med. Assn., 1910, S4, P- 1373- ' Ibid., p. 463; p. 45 1 this book. 'Anderson and Goldberger, Pub. Health Rep., February 18, ipio. 473 474 Contributions to Medical Science ing a suggestion in regard to the probable or approximate size of the virus. It seems not unhkely in view of its non-filterability that the micro-organism is of such size that it should be suscep- tible to observation microscopically, provided its tinctorial affinities and density are favorable. Prompted by this probability, we have been engaged in a careful search of the blood of patients for the presence of micro-organisms, correlating our findings with the microbic content of the organs of the body louse, which has the power of carrying the disease. It will be apparent that the results do not prove that the organism to be described is the cause of typhus, but to our minds they have a suggestive value which is sufficient to render their preHminary presentation justifiable. They may be stated briefly as follows: 1. In the stained (Giemsa) preparation of the blood of patients, taken on from the seventh to the twelfth day of the disease, we invariably have found a short bacillus which has roughly the morphology of those which belong to the "hemorrhagic septicemia group." Usually it appears to stain solidly, but on minute exami- nation an unstained or faintly stained bar is seen to extend across the middle. Occasionally two organisms are seen end to end. Exact measurements have not been made, but when compared with the size of the erythrocyte, their length is estimated at hardly more than two micromillimeters, and their diameter at about one-third this figure. Certain other bodies, the identity of which is not so clear, may represent degeneration or involution forms of the above. They consist of two stained granules, connected by an "intermediate substance," which is stained faintly blue or not at all. Frequently one of these granules, or "poles," is larger than the other and stained a deep purple, whereas the smaller takes a faint blue color. 2 . In moist preparations of the blood of patients, bacillary bodies, with a structure Hke that mentioned above, have been encountered in all cases. The differentiation of the forms into two halves, separated by a line or narrow zone of a substance of different refractive power, may be observed. They possess no active motihty, but vibrate more or less rapidly. 3. The dejecta and various organs of a large series of lice have Etiology of the Typhus Fever of Mexico City 475 been stained in a similar way and examined for the presence of micro-organisms. Certain groups had been deUberately infected by permitting the Hce to feed on patients, while others were sup- posedly normal, having been collected from healthy individuals. Streptococci, staphylococci, an oval bacillus occurring in clusters, and certain solid staining bacilli are encountered irregularly and indifferently in the feces and intestinal contents of both "normal" and "infected" lice. Polar-staining organisms have been found occasionally in the feces, and intestinal contents of "normal" lice, whereas they are present almost constantly, and often in large numbers, in similar material from "infected" individuals. Protozoa have not been recognized. Micro-organisms are encountered occasionally in preparations of the salivary glands, ovary, eggs, and testes of the louse. Further study is needed to determine whether they are present in these tissues as a result of faulty technic, or not. TECHNIC. Probably no field of microscopic research is more exposed to errors and misinterpretation than the search of the blood for micro-organisms of an unknown character. We shall not at this time consider the many sources of confusion which are inherent in the blood itself, and in methods of its preparation. Naturally we have attempted to avoid being led astray by these conditions. Accidental contaminations from outside have been guarded against carefully. Slides and cover-glasses were cleaned in sulphuric acid and potassium bichromate, washed in freshly distilled water, and preserved in absolute alcohol. The latter was removed by freshly laundered linen just prior to making the preparations. Forceps previously flamed were used in manipulating the glass. The water employed in diluting the Giemsa stain was distilled, then filtered, autoclaved, and preserved in small flasks, which had been cleaned and rinsed repeatedly with filtered water. Preparations of the blood were covered at once to avoid contamination from the air. The blood was taken from the ear after the latter had been scrubbed with green soap and absolute alcohol by means of ster- ilized cotton. 476 Contributions to Medical Science frequency of the bacilli in the blood. The organisms described have been found to be more numerous in the blood within three or four days prior to crisis than before this period. In comparison with the frequency with which the or- ganism of malaria and trypanosomiasis are encountered in the blood of patients, the baciUi would be considered quite rare. It is rare to recognize a bacillus short of one trip across a three-fourths-inch cover-glass, and frequently it is necessary to cover this distance two or three times. This is true of both the stained and the fresh preparations. A rough, and naturally inaccurate, idea of their number may be obtained from the consideration that o.oi c.c. of blood made from fourteen to twenty-four smears on the three- quarter-inch cover-glass, as determined in two observations. With the 2-miUimeter apochromatic objective of Zeiss and with the No. 6 compensating ocular, and with the tube drawn out to i6 milli- meters it is required to traverse the preparation from 80 to 90 times in order to observe the entire surface of one cover-glass. Hence one might expect, roughly, to find from 300 to 2,000 bacilH in 0.01 c.c. of blood. In the intestinal contents of the "infected" louse, it is occasion- ally necessary to search for three or four minutes before bipolar organisms are found, but in most instances organisms of this type are much more numerous in the intestine of the "infected" louse, and fifteen or twenty may be found in a single field. It has occurred to us that the segmented bacilH observed in the blood may be identical with the bipolar forms, which are found in the intestinal contents of the louse. The differences are not greater than those which are encountered in the known organisms of this type, such as the plague bacilli and that found in association with Rocky Mountain spotted fever. THEORETICAL CONSIDERATIONS. Considering that typhus is an insect-borne disease, the first thought in regard to its microbic etiology would naturally involve an organism of protozoan character, on the basis of analogies which are known to all. Yet our knowledge of the role of the flea in the Etiology of the Typhus Fever of Mexico City 477 transmission of the bacillus of plague, of the tick in South Africa in carrying the spirillum of tick fever, and of the same insect in carrying the bacilli which are associated with Rocky Mountain spotted fever, should leave us without prejudice regarding the etiology of some other insect-borne disease. It seems, therefore, that our minds should be open to conviction in relation to a bac- terial cause for typhus as well as protozoan. There are, in addition, at least two features of typhus which would suggest a bacterial rather than a protozoan cause. The first consists of the fact that typhus is an acute self-limited disease, in so far as clinical evidence can show. This condition is more in harmony with a bacterial rather than a protozoan cause, inasmuch as the protozoan diseases, in so far as their etiology has been determined, are chronic in character, provided the animal survives the initial acute attack. A second condition which suggests bacterial etiology consists of the fact that one attack of typhus, at least in the monkey, confers immunity to further inoculations. It is true that not all bacterial diseases confer such immunity, but it is equally true that immunity is not recognized as a distinctive feature of protozoan infections. If we may be permitted to resort to another analogy, some ground may be found for the theory that typhus is caused by an organism of the "hemorrhagic septicemia group" of bacteria. This relates to the fact that typhus actually is a hemorrhagic septicemia from the clinical and experimental standpoints, and that it has certain points which are distinctly common to plague and Rocky Mountain spotted fever, both of which may be classed as hemorrhagic septicemias. We are inclined to consider, there- fore, that the three diseases mentioned constitute a group of human "hemorrhagic septicemias." We are aware that others have described organisms more or less similar to the one we have considered in relation to typhus, but in so far as we know they are organisms which are susceptible to cultivation by ordinary means. Gavino and his assistants in Mexico City and Anderson and Goldberger have failed to cultivate an organism of any type from typhus, and this has been our expe- 478 Contributions to Medical Science rience also. We consequently conclude that the organism we described is not susceptible to cultivation under ordinary condi- tions. This is corroborated by the fact that two experiments failed to yield an organism of this type from the intestinal con- tents of the louse. It is the purpose of this paper to present the observations and the theoretical considerations which have been mentioned solely because of their suggestive value. It is clear in our minds that the grounds are not sufficient for claiming an etiologic role on the part of the organism described, yet the conditions under which they are found, together with the theoretical argument presented above, appear to demand that they be taken somewhat seriously and subjected to further study in their relationship to typhus. We desire to express our obligations to the Department of Public Instruction of Mexico, to Director Gavino of the Bacteriologic Institute, to the authorities of the General Hospital, and particularly to Dr. Escalona, for their numerous courtesies. THE RELATION OF TYPHUS FEVER (TABARDILLO) TO ROCKY MOUNTAIN SPOTTED FEVER.^ H. T. RiCKETTS AND RuSSELL M. WILDER. (From the Department of Pathology, University of Chicago, and the Memorial Institute for Infectious Diseases, Chicago.) One who has seen Rocky Mountain spotted fever cannot fail to be impressed with certain points of similarity which the disease shows to typhus fever (Typhus exanthematicus) , basing the com- parison on the descriptions of typhus which are given in standard treatises. These descriptions refer to typhus as it occurs in certain European and Asiatic countries. It seemed desirable, therefore, to study their relationship, along chnical, anatomical, and immuno- logical Hues, at least in certain essential respects. Our observations concern tabardillo, the typhus fever of the great Mexican plateau, which differs in some important respects from European typhus, according to the opinion of those who have studied the disease minutely (e.g., Jose Terres, in "Etiologia del Tabardillo"). It is stated that the typhus of Mexico has a more gradual onset and defervescence than that of the Old World. These are both said to be very sudden in the latter, whereas in the former the fever rises gradually for three or four days during the onset, and defervescence occupies a similar period. The two diseases should again be subjected to a close comparative study in these respects. A peculiar topographic distinction exists between the two, in that the typhus of Mexico is Hmited to the plateau and is said not to occur at or near sea-level; that is to say, it does not occur in the so-called "hot country" of Mexico. European typhus, on the other hand, finds its home to a large extent at sea-level. In view of the fact, however, that typhus, the world over, is a disease of temperate and cool climates, the discrepancy mentioned is only an apparent one; it loses significance when we consider that the climate of the great Mexican plateau is a temperate one, while " From Arch. Int. Med., 1910, s, p. 361. 479 480 Contributions to Medical Science that of the sea coast towns is warm or torrid, in which general experience indicates that typhus is not able to prevail. comparison of spotted fever and typhus fever. Eruption. — In both spotted fever and typhus fever a macular or slightly elevated roseolar eruption occurs, which commonly becomes petechial, and which appears at about the same time in both diseases, but perhaps a little earlier in spotted fever. In typhus it is first seen at about five days after the beginning of the fever, and in spotted fever on from the second to the fifth day. In both diseases small hemorrhages (petechiae) may occur either in the pre-existing rose-colored spots or at points in the skin which were hitherto uninvolved. In the former case a petechia with a congested zone is formed, and in the latter the areola is absent. The hemorrhages appear to occur earlier and with greater regu- larity in tabardillo than in spotted fever, although this phenome- non is subject to great variations in the latter disease.^ In some instances of spotted fever sharply marked petechiae do not occur at all, while the opposite extreme occasionally is encountered, the petechiae appearing before a roseolar eruption is definitely recognizable. Although petechiae do not occur invariably in typhus, they would seem to be more constant than in spotted fever. In both diseases the "spots" show a certain slight degree of induration; this is quite marked in some cases of spotted fever. There seems to be a characteristic difference in the regions of the body first involved. In spotted fever the spots first appear on the forearm and lower leg in a large percentage of the cases, whereas in typhus they are first seen on the abdomen and sides of the chest. There are variations, however, in the sequence of distribution in spotted fever, so that it is doubtful if this point is of great distinc- tive value. In both infections the distribution in the end is a very general one, including the face, palms, and soles: it would seem that the involvement of the palms and soles is more prominent in spotted fever than in typhus. As regards profuseness there is no essential difference. ' See the Fourth Biennial Report of the Slate Board of Health of Montana, 1907-8, p. 137- Relation of Typhus Fever to Spotted Fever 481 Gangrene. — In the spotted fever of Idaho, gangrene of the foreskin and scrotum, the tonsils and faucial pillars occurs not infrequently; this is not seen so frequently in the more severe type which prevails in Montana. In typhus, the toes, feet, and lower leg occasionally become gangrenous, and extensive bedsores are rather frequent. Changes in internal organs. — In spotted fever the spleen is habitually enlarged. This can be determined clinically in practi- cally all cases; and at autopsy the mass of the spleen is sometimes three or four times that of the normal organ. In Mexican typhus it shows Httle enlargement and this can rarely be detected cHnically. In one autopsy it was of normal size; in another it was slightly enlarged, but cirrhosis of the liver was also present. In both diseases it is of rather firm consistence, in no way resembling the spleen in typhoid fever. In spotted fever the lymph glands are distinctly but moderately enlarged; in typhus they are smaller, but probably a little larger than normal. They show more congestion in spotted fever. In typhus the meninges almost constantly show a great deal of congestion and edema at autopsy, so that it has sometimes been spoken of as meningeal typhus. This condition also is stated as being present in European typhus, but perhaps not as constantly as in that of Mexico. It is a minor finding in spotted fever. At autopsy the right heart and venous system show more engorgement in spotted fever than in typhus, and this corresponds also with the clinical appearances of the two diseases. Other organs appear uninvolved in typhus and spotted fever, except for the presence of occasional complicating infections, particularly pneumonia. There appears to be nothing unique, therefore, as to anatomical changes in either disease. Fe^er. — In typhus the fever begins and ends with a good deal of abruptness. Two or three days may be required after onset before it reaches its high point, and an equal period is occupied in defer- vescence. On the other hand, the temperature in spotted fever may not reach its maximum until a week or more after onset, and 482 Contributions to Medical Science defervescence may occupy a week or ten days. This is one of the marked differences between the two diseases. Pulse. — In the early part of the course, and in mild cases throughout, the pulse in spotted fever (90-110) is slower than in typhus (i 10-120) under similar conditions. In both it rises to 140 or more preceding a fatal issue. Metital effects. — Stupefaction, or a low nervous delirium, is common to both. Convalescence. — This varies with the severity of the infection in both diseases, but on the whole it is much slower in spotted fever than in typhus. This may be due in part to the longer duration of spotted fever. Duration. — The "crisis" in typhus usually occurs on from the tenth to the fourteenth day, although some cases may cover a period of three w^eeks. Patients suffering from spotted fever are rarely convalescent until the end of the third week, and they commonly remain bedfast for from four to six weeks. Character of the infections. — In both, the condition is that of a systemic blood infection (and presumably lymph infection), without the critical involvement of particular organs. These points are brought out by the findings at autopsy, and by the result of inoculations with the blood of patients. Blood from spotted fever patients is always infective for the guinea-pig, mon- key, and certain other animals. In a number of instances reported in the literature, the blood of tophus patients apparently has produced the disease when injected into other human beings, and NicoUe's infection of the chimpanzee was done by the same means ; this concerns European typhus. The experiments with Mexican typhus, reported recently by Anderson and Goldberger and by ourselves, also show that the latter is a generalized infection. Transmission. — Spotted fever is not contagious, and the evi- dence indicates that the same is true of typhus. The former is transmitted by the bites of certain species of ticks, while presumably the latter is carried by the body louse (Pediculus vestimenti). It is probable, therefore, that they have the feature of insect transmission in common, but two altogether different types of insects are concerned. Relation of Typhus Fever to Spotted Fever 483 It is a peculiar fact that the conception of contagiousness has adhered to typhus up to, and including, the present time. Yet, in view of the facts that typhus, when endemic in a city, remains rather strictly segregated in the poor quarters, and that more or less intimate contact is required for transmission, it is manifest that contagiousness, if present at all, must be of a peculiar character and of a low grade. Typhus has never overwhelmed a whole city as smallpox did again and again in former times. In recent years, however, behef in the theory of insect transmission of typhus has extended widely, as affording a better explanation of the epidemio- logic features of the disease. Thus the flea and bedbug have repeatedly been mentioned in relation to European typhus, and Gavino and others have called attention to the possibihty of insect transmission in Mexico, without indicating the probable species, however. The recent experiments of Nicolle in transmitting European (rather Asiatic) typhus from monkey to monkey by means of the body louse affords a good working basis for clearing up the natural means of transmission. Our own experiments with the louse, which have been successful in a measure, will be reported at a future date. Susceptibility of animals. — Aside from the discrepancies between spotted fever and typhus which were mentioned above, a striking difference is found in the susceptibility of animals to the two diseases. As regards European typhus the literature contains numerous references to attempts to infect the guinea-pig and other ordinary animals of experimentation by the injection of blood from patients, the results being uniformly negative. Likewise, in extensive experimentation with the typhus of Mexico, Director Gavino of the Bacteriologic Institute of Mexico City, and his assistants, failed to produce any evidence of infection in guinea- pigs, rabbits, white rats, and mice by the subcutaneous and intravenous injection of blood from human patients, the blood being taken at different periods of the disease.^ Similarly, and in substantiation of Dr. Gavino's results, Anderson and Goldberger ■ General reference to these experiments is made in an article by Dr. Gavino in Gaceta Midica de Mexico, 1906, 1, p. 218. They are also cited elsewhere in the same publication, the exact references not being before us at this moment. 484 Contributions to Medical Science reported their failure to infect the guinea-pig by the injection of virulent typhus blood/ These results seem so conclusive that we decided not to repeat the experiments. In contrast to this condition, a fairly large experience has shown that spotted fever may be transmitted to the guinea-pig invariably by the subcutaneous or intraperitoneal injection of virulent blood, provided no serious error in technic has been made.^ The difference in the susceptibility of the guinea-pig to the two diseases must be taken as showing definitely that typhus and spotted fever are not identical. It might also stand as sufficient reason for concluding that they could not be even related infec- tions, were it not for the fact that there are two types of spotted fever, one of which appears to be less virulent for the guinea-pig than the other. The spotted fever of western Montana, which represents the more virulent type, can be maintained indefinitely in the guinea-pig by passage from one animal to the other. On the other hand, it has been impossible, on three occasions, by the use of the same method, to keep alive in the guinea-pig the mild type of the disease which prevails in southern Idaho. It "died out" after from two to ten passages, presumably because of a loss of virulence for the guinea-pig. Yet, other experiments, particu- larly that of agglutination, as performed with the bacilH found in the eggs of the tick, indicate that the two types which occur in Montana and Idaho, respectively, are identical or closely related. As bearing on typhus fever, this condition raises the question as to whether there may be a third type of infection (typhus fever) which is related to spotted fever, but which differs from it not only in certain important clinical respects, but also in possessing even a less degree of virulence for the guinea-pig than the mild spotted fever referred to above. In other words, may typhus fever have a group relationship to the spotted fever of the Rocky Mountains ? 'Anderson and Goldberger, "The Relation of Rocky Mountain Spotted Fever to the Typhus Fever of Mexico — A Preliminary Note," Pub. Health Rep., 1909, 24, 1861. 'It is important that the blood injected should be drawn rather early in the disease, and that the quantity injected should not be too large; dilution with salt solution favors infection. This has been referred to in previous articles by one of us. Relation of Typhus Fever to Spotted Fever 485 immunization experiments. We have resorted to protective and agglutination experiments in order to obtain more conclusive data regarding this point. The experiments may be reported briefly. One attack of spotted fever, or of typhus, renders the individual immime to further attacks of the same disease. This acquired immunity in spotted fever is characterized by the formation of protective antibodies, which appear in the blood, and which can be demonstrated by experiments on the guinea-pig. From o.i to o . 2 c.c. of serum from the immune guinea-pig protects against i.o c.c. of virulent blood, representing from 200 to 1,000 patho- genic doses. From 0.3 to 0.5 c.c. of serum from the convalescent human patient were required to exert the same protective effect in two experiments. If typhus were identical with spotted fever, the serum from typhus convalescents should exert a similar pro- tective effect against spotted fever; or, if typhus occupied a "group relationship" to spotted fever, this might be manifested by a certain (perhaps low) degree of protective effect against spotted fever, on the part of serum from typhus convalescents. As will appear, the experiments to be reported do not disclose with certainty the relationship which has been suggested, and possibly they are of such value that a close relationship is actually disproved. The immune typhus serums were all taken from patients whose course we had observed in the General Hospital (Mexico City), and in whom the diagnosis of typhus seemed to be with- out doubt. Their histories will not be recited. The blood was drawn on from the seventh to the tenth day after the sub- sidence of fever, after the patients had left their beds, and the serum obtained by defibrination and centrifugation or by spon- taneous clotting. Three sets of controls were utilized: first, the protective power of normal human serum as compared with that from the typhus convalescents; second, the tests of the toxicity of human serum alone for the guinea-pig; third, inoculations to determine approxi- mately the strength (quantity or virulence) of the spotted fever virus used in each experiment. The virus was that represented in 486 Contributions to Medical Science cT TS- J3 rt 1 rt a «-• ■S a>. 1 .s 0- a E iJ ■S S '^ " 2 ** -« "^ tf) ^ rt -"^ *j— 8SS0 § II ^ 1hi lie' ••r ?s w 14 fc^ b4 iM C il C h U h^ .w >£.ii t- u^ u a e V 0) > > ;> > > > > > > > > > > i"l-- o V O V V V V u u jj iSi>S^ OS T3-C-3-3 -3 V v-a Tr-T3 -3 •0 -B-a-o ♦J CI. ovou o>>a> 2'" V ■J «-*«^.MW Wq^^-4^ 3 U rt ■s w >, J I.sg.sS "a 0« CI C Q. 0.^ "3 C^ II 1 1 III 2-- 3 »; ^ . ..>•>'.. .a >.>.>.>. >.^^ >. '3 i tf) t/) W (/) •** >, >>>■>. i O O O O Ow'i o Its fi 3 3 -Ji 3*° a c.ac. 2 2 "5 ." -S 3333 3-2. Sa 3 33 3 i: 3 S >.« >, > . 4) . . . J3 >j=j:j3 4) .'u m< >.r: >J3 ? . 3 g ^»j»j^ O'S^^'rt •" ■" a> -J.- •^ ti -y ri S-S^s «lf.§ll u rt rt— u C3 5 *J rt u rt rt cj B4 u a> V ?T::7S a> £ £ 1) iu:3 (u 41 1) 41 « 0) 0) QQQQO^QUJ^Q ^ p. » ja-r- c3 fOfOfO'^^'O'^'^fO'^ ir,~t ^ ^^ ■"t-O ■ ^vO VO •* -.J • H 1'^ '^ 10 u H 2 w £ Tt »i->\0 t^oo o> O M « "> r f "o a " N fo •* >«o t- X X K X X X M MX X «4 -*'*TfTl-'Tj-' vO f^r^r^r-t^r^r^t- ■O OO'O^^OO^'O^ OO-OsOO^OvOsC C3 '5 C«MNCI-.««r. « r. IH M f^ Tl- u- Ov « 00 « >- •< O o •V in 3 r^io OOOOOOmOO i« 10 in M » « » M E 00 8 a^^" ^^HiMWMddd d d t^ 66660 d d 00 d g o PLr O 3 S U . . . o O m • ■ • C 10 . . • ■ • ■ ... g : ■. « i-i d • < d . . co« ■ • • • ...> M 1 1 a 3 •^ W i "^ Id £ Id 1 S! '> > >> sr '> S"- '> pi J3 ■5 0) a 1 •2 u > «i «w J3 ""2 2 - 1 1 £ S! c c - !'£ 4 1 1 III -0 " c c Si c c rt ' canvas jacket in order to prevent inter- ference with the lice. Thus on April 28, fifty young lice were placed on the monkey. On April 30, thirty more were added. At this time it was noted that the Hce of April 28 had fed, their bodies being gorged with blood. On May 2, twenty-five additional lice were collected from the stocking and placed on the monkey. Thus in all 105 lice, the offspring of infected lice, but themselves never di- rectly infected, were given an opportunity of feeding on monkey 42. Unfortunately the temperature of this animal could not be taken regularly during the following three or four weeks, nor was he under very careful observation during this period. On May 26, however, he appeared to be in good health. But more impor- Investigations of Mexican Typhus Fever 493 tant is the fact that he proved resistant to a subsequent immunity test described below. In our work on transmission with Hce we have, with one excep- tion, never succeeded in provoking a very characteristic febrile reaction in the macacus monkey by means of the bite of Hce, the relative insusceptibility of the monkey to typhus probably account- ing for his resistance to infection by lice. It has been found, however, that monkeys who have been subjected to the bites of infected lice are thereby rendered immune to subsequent inocula- tions of typhus blood which constantly provokes in normal monkeys a high febrile reaction. This immunity has been interpreted as indicating that the animal has been infected by the Hce and has suffered a mild attack of typhus, and the immunity test thus con- stitutes the chief criterion for determining whether or not an animal has been infected by insects. Such an immunity test, consisting of the intraperitoneal inocu- lation of 3 . 5 c.c. of virulent typhus blood from a typhus patient (No. 58), was given to monkey 42 on May 27. For the following three weeks the animal remained in perfect health, although con- trols inoculated with the same quantity of the same material all contracted a typhus of moderate severity, their temperatures maintaining an elevation of 104 to 105 . 7 F. for a period of ten or eleven days. TABLE I. Immunity Test of Monkey 42. May 28 May 29 May 30 May 31 June I June 2 June 3 June 4 June 5 June 6 June 7 June 8 June 9 A.M. 102.6 101.6 ICI.2 101.6 101.4 101.6 102.0 101.6 104. 1 103.2 103.0 I01.6 100.4 P.M. 102.2 101.9 101.9 102.0 101.8 102.2 102.4 102.4 103.7 103.0 102.8 103.0 102.2 June 10 June II June 12 June 13 June 14 June IS June 16 June 17 June 18 June 19 June 20 June 21 June 22 A.M. 102.3 102.0 102.0 100.8 99.6 100.3 101.2 100.9 100.4 101 .0 100.8 100.2 101.3 P.M. 102.1 102. S 102.8 102.0 100.6 101. 5 101.4 101.9 102.0 102.2 101 .7 102.4 101.8 Table i shows the temperatures of monkey 42 on the successive days foHowing the inoculation. Table 2 those of the control monkey (No. 44)- 494 Contributions to Medical Science TABLE 2. Temperature of Monkey 44 Following Inoculation with Typhus Blood. A.M. P.M. A.M. P.M. May a8 102.3 102.0 loi .9 loi.s 100.6 101.6 102.6 105.6 104.0 103 -3 103.5 102.5 102. s 102.0 102.3 loi .6 103.7 103-3 105.7 1 04 -3 103.3 104.9 June 8 June 9 June 10 June II June 12 June 13 June 14 June IS June 16 June 17 104.0 104.1 105.6 104.6 103.2 loi .6 loi .0 lOI.O 100.6 100.3 105.5 104.7 103.9 105.2 103.1 102. 1 102.2 101.6 101.6 lOI.O It has been our experience to find a great variation in the daily- temperatures of normal macacus monkeys. These animals are prone to slight intestinal disorders and a sporadic elevation of one, or even two degrees, has been frequently noticed. The normal afternoon temperatures of many of our monkeys lay constantly between 103° and 104° F. Hence there arises a certain difficulty in interpreting the sHght elevation of temperature shown by monkey 42 on June 5 and June 6. This may or may not be a mild efifect of the virus, but in any case the effect was far less than that obtained before in all normal animals, including the control monkey (No. 44) of this particular experiment, and it is quite probable that the elevation observed was purely accidental. We appreciate that the result of one experiment does not con- stitute decisive proof, but the definiteness of the result justifies us, in our opinion, in concluding that monkey 42 owed his immunity to his previous infection by the yoimg lice of group 17 and that hereditary transmission of the infectivity of the louse is established to the extent of reasonable probability. infectiousness of the flea and of the bedbug. Theoretical considerations make it seem extremely probable that neither the bedbug nor the flea plays any role in the trans- mission of typhus. We have called attention to this in a previous article,' and Anderson and Goldberger' have expressed the same opinion. The bedbug is only rarely carried about in the clothing and yet it is well known that people who handle the clothing ■ Jour. Am. Med. Assn., February s, 1910, 34, p. 463. See p. 451 of this book. " Pub. Health Rep., February 18, 1910, No. 7. Investigations of Mexican Typhus Fever 495 of typhus patients, such as laundresses and servants, are fre- quently stricken with typhus. A number of cases of typhus have occurred during the last year among the nurses and servants of the typhus pavilion of the General Hospital of Mexico City. A careful search failed to reveal any bedbugs in this building. Furthermore, if the bedbug transmitted typhus we should expect the disease to be a "house disease"; but cases are constantly occurring which it is impossible to trace to any "typhus house." The flea also seems innocent. Did it play any role, the dis- tribution of typhus in Mexico City would be much more general than is the case, for the flea is widely distributed. Typhus, how- ever, is confined almost exclusively to those living under less satis- factory hygienic conditions. Finally, as previously mentioned, the season of typhus is the winter and early spring, whereas the period of greatest prevalence of the flea is the summer. In order to throw further light on this question the following experiments were undertaken. EXPERIMENTS ON THE INFECTIOUSNESS OF THE BEDBUG. A group of about fifty bedbugs was fed on three successive days on patients at the General Hospital. The bugs were confined beneath a small wide-mouthed glass bottle inverted over the skin of the patient. In this way ten or twelve could be fed at one time. All were given the opportunity of gorging themselves with blood. On April 2, two days after their last feeding on a patient, they were placed on the shaved abdomen of a monkey (No. 35) and allowed to feed. On April 3 they were again placed on the monkey, only eight feeding. On April 4 they were not fed. On April 5, twenty-eight bugs fed well on the animal; April 6, eighteen bugs fed Hghtly; April 7, seventeen bugs fed lightly; April 12, 29 bugs fed well, gorging themselves with blood. The animal showed absolutely no rise in temperature, and continued in excellent health during the following thirty-two days. Unfortunately he died from an accident before an immunity test could be given him, and hence the experiment cannot be taken as proof of the non-infectiousness of the bedbug. It does seem to eliminate the possibiUty that the bedbug can transmit tophus more 496 Contributions to Medical Science readily than can the louse, in so much as both the period of feedings on the infected host and the period of feeding on the monkey were considerably in excess of the feedings of the louse in certain of our experiments. EXPERIMENTS ON THE INFECTIVITY OF THE FLEA. Human fleas, of flea group 2, were infected by repeated feedings on typhus patients at the General Hospital in the following manner : One or two fleas were confined in each of several long and narrow tubes. These tubes were made of 4 mm. glass tubing, were sealed at one end and cut sufl&ciently long (20 cm.) to prevent the escape of the flea by jumping. For this purpose also they were slightly bent in the middle. The fleas were fed by inverting this tube on the patient's skin. Subsequent to their last feeding the fleas were allowed to rest for about sixty hours. The entire bodies of ten of the group were then emulsified in physiologic salt solution and rubbed into scari- fications of the abdominal skin of a normal monkey (No. 41), the technic employed in this procedure being the same as that used in the scarification experiment performed with the intestinal con- tents of lice and previously reported. The wounds healed with but little suppuration. The animal's temperature was taken twice daily for the following thirty-four days and during this time he remained in perfect health. On May 27 this monkey was given an immunity test, receiving an inoculation of 3 . 5 c.c. of typhus blood from patient 58. After an incubation period of seven days he began to run a fever which lasted for twelve days, his temperature being recorded in table 3. On June 4 a leukocyte count was made^ which showed 30,350. The animal was very irritable and his coat dry and ruffled. On the 5 th he had diarrhea and until the 14th of the month seemed very sick. His subsequent recovery was rapid and complete. In brief. No. 41 had not been infected by the fleas, as is shown » Leukocytosis is observed in nearly all cases of typhus in man. In conjunction with Dr. Francis E. Prestley of Mexico City, we have made a series of leukocyte counts of most of the monkeys in our possession. It was found that monkeys sick with typhus fever usually showed a marked increase in the number of white blood corpuscles over the normal which approximates 12,000. Differential counts based on the same material will be reported in a later communication. Investigations of Mexican Typhus Fever 497 by the fact that he was not immune to a subsequent inoculation of typhus blood. TABLE 3. Temperature of Monkey 41 after Receiving an Immunity Test. May 28 May 29 May 30 May 31 June I June 2 June 3 June 4 June s June 6 June 7 June 8 June 9 A.M. 102.6 101.8 102.3 loi .6 loi .9 101.9 101.4 103. 1 104.5 103. 1 105.0 103.2 roi.4 P.M. 102.6 102.0 102.8 102.6 IOI.9 102.7 102.3 104. 1 104-5 104. 1 105.3 103.6 103. 1 June 10 June II June 12 June 13 June 14 June IS June 16 June 17 June 18 June 19 June 20 June 21 June 22 A.M. 104.4 104.4 103.6 104.6 103.5 103.9 102.9 102. s 102.5 loi.s 101 .6 loi.s loi .9 P.M. 105.0 105.6 105.5 104.8 104.8 103.5 102.8 102.8 102.0 101.8 102. s 102.6 These results seem to strengthen our previous position as to the unimportance of the flea and the bedbug in the transmission of Mexican typhus fever. We take pleasure in expressing our obligations to Dr. Liceaga and his assistants of the Superior Board of Health of Mexico, to the authorities of the General Hospital, and to Dr. Gavino of the Bacteriologic Institute for numerous courtesies. In partic- ular we wish to acknowledge our indebtedness to Dr. Genaro Escalona, Dr. Francis E. Prestley, and Sr. D. Perez Garga of Mexico City, for co-operation and assistance. UNIVERSITY OF CALIFORNIA LIBRARY Los Angeles This book is DUE on the last date stamped below. BiowiDpfS 1 4 REC'D c^.RtCC tSL NOV ?l^^ 2 WK from Receipt JUL 4 1974 JI1.>SI»MLS JUM^2"Cr^Er&- MAR 2 4 REC'[^ BioMEo MAR 2076 DEC ^®f Form L9-100m-9,'52(A3105)444 R4:i^C 3 1158 OCpS 462P '^■'^;'''^ -.